SPECIAL REPORT No 2/95 concerning the Stabex fund in the context of the first financial Protocol of the fourth Lomé Convention together with the Commission's replies (95/C 167/01)

(Observations pursuant to Article 188C (4), subparagraph 2, of the EC Treaty)

TABLE OF CONTENTS

Page 1. INTRODUCTION 2

2. OUTLINE DESCRIPTION OF THE STABEX SYSTEM 3

3. THE MANAGEMENT OF THE LOMÉ IV STABEX ALLOCATION 4

Financial situation of Lomé IV Stabex 4

Insufficiency of resources 4

'All Destinations' Status 5

Measures taken to combat shortages of Stabex resources 6

Allocation of Resources by Country 7

4. THE CALCULATION OF STABEX TRANSFERS 7

Timetable and reliability of statistics 7

Reductions in transfer entitlements 8

Errors noted in calculation of Stabex transfers 9

The impact of quality on price 9

5. THE PAYMENT OF STABEX TRANSFERS 9

6. THE PREPARATION OF THE FRAMEWORKS OF MUTUAL OBLIGATIONS 10

Provisions of the Convention concerning the utilization of funds 10

Procedures for preparing frameworks of mutual obligations (FMOs) 11

Delays in preparation of FMOs 12

Identification of operations to be financed 12

Conditionality 13

Commission procedures 13

7. THE UTILIZATION OF STABEX TRANSFERS 14

Financial implementation of the frameworks of mutual obligations 14

Analysis of allocation of Stabex funds 14

Import programmes 14

Page Internal debt relief measures 18

Price stabilization 20

Projects 21

8. IMPLEMENTATION OF ACP STATES' COMMITMENTS 21

The principle of conditionality in relation to Stabex 21

Types of conditions 22

Respect of conditionality 22

9. FINANCIAL AND ACCOUNTING PROCEDURES 25

Legal framework for implementation of Stabex operation 25

Functioning of the double signature accounts 25

Counterpart funds 25

Financial monitoring of Stabex by the Commission 26

10. MONITORING AND EVALUATION OF LOME IV STABEX 26

11. CONCLUSION 27

General context of the implementation of the Stabex system 27

Economics of the system laid down in the Convention 27

Financial management 28

Implementation of the Stabex transfers 28

Overall assessment 28

The Commission's replies 38

1. INTRODUCTION

1.1. Stabex is one of the main financial instruments created by the Lomé Conventions with the aim of assisting African, Caribbean and Pacific (ACP) States. The Stabex system is designed to guarantee the stabilization of ACP States' export earnings from the Community for certain agricultural products, laid down by the Convention (1). As such it has a unique role in development cooperation.

1.2. According to Article 186.1 of the Lomé IV Convention, the aim of stabilizing ACP States' export earnings is that of: 'remedying the harmful effects of the instability of export earnings and to help the ACP States to overcome one of the main obstacles to the stability, profitability and sustained growth of their economies, to support their development efforts and to enable them, in this way, to ensure economic and social progress for their peoples by helping to safeguard their purchasing power'.

1.3. While preserving the same basic objective as previous Conventions, the Lomé IV Convention has introduced a number of significant changes in the Stabex system (see 2.8). Its continuing importance is underlined by the fact that the first Financial Protocol of Lomé IV allocates 1 500 Mio ECU to Stabex (for the implementation years 1990-1994), which represents 12,5 % of the total resources available under the Protocol.

1.4. The Court of Auditors has carried out an audit of Stabex operations under Lomé IV for the period 1991-1993 (2). The audit paid particular attention to the effects of the changes to its functioning introduced by the new Convention. However, as the new system has only been in place for a short time, some of the conclusions can only be considered provisional. The Court's observations are based on the results of on-the-spot audits of five Stabex beneficiaries (3) and detailed documentary audits for all the other beneficiary countries.

1.5. On various occasions the Court noted contradictions between the objectives assigned by the Convention to the Stabex system. As is shown in paragraphs 6.1 to 6.8 below, we are not, pace the Commission, talking about a hierarchy of objectives but of opposed objectives, some of which tend to counteract the effects of falls in revenue whilst others are supposed to deal with the causes. These contradictions often lead to delays in implementation and render the procedures more cumbersome. This situation cannot be remedied by mere amendment of the Commission's management practices. These ambiguities and contradictions must be removed from the Convention itself so as to arrive at a consistent policy in respect of the various financial instruments.

2. OUTLINE DESCRIPTION OF THE STABEX SYSTEM

2.1. The purpose of this chapter is to provide a basic overview of how the Stabex system operates. A diagram of the mobilization procedures for the Stabex instrument is provided in Annex 1.

2.2. The overall Stabex allocation is divided into five annual instalments (4), each of 300 Mio ECU, covering the lifetime of the first Financial Protocol of Lomé IV.

2.3. At the end of each calendar year, the Commission determines for each ACP State whether it has suffered a loss of export earnings for any of the 49 (largely agricultural) products listed in the Convention (5) where earnings from that product represent more than a certain percentage of the country's total export earnings (6).

2.4. The export earnings for the year under consideration (known as the 'application year') are compared with a reference level for export earnings established over a defined period (7).

2.5. If it is established that there has been a fall in export earnings for a particular product for the application year compared with the reference level, the ACP State concerned is entitled to a transfer for that product from the annual Stabex instalment. The amount of the transfer is calculated on the basis of the amount of the fall in export earnings (Article 197).

2.6. Prior to the transfer(s) being paid over, the ACP State informs the Commission of how it proposes to use the funds to which it is entitled (Article 209). In principle, the funds are to be used to assist the sector in which the loss of earnings has occurred (Article 186).

2.7. Within twelve months of the mobilization of the funds, the ACP State reports to the Commission on how the funds have been used (Article 212).

2.8. The main innovations of Lomé IV are as follows:

(a) while previous Conventions required ACP States to repay Stabex transfers under certain conditions, this is no longer the case under Lomé IV;

(b) Lomé IV provides for much closer control over how the ACP State uses Stabex transfers:

(i) for each transfer the ACP State and the Community are to agree on a detailed programme for using Stabex funds to be set down in a document entitled 'Framework of Mutual Obligations' (FMO) which is signed by the Commission and the National Authorizing Officer (NAO) of the ACP State (Article 210);

(ii) Stabex transfers are first paid into a dual-signature bank account in Europe from which funds are only released for disbursement as and when operations specified in the framework of mutual obligations are implemented (7.1), and only after the signature of the Commission delegate in the ACP State concerned, as well as that of the NAO, has been obtained. The Convention also requires counterpart funds generated by Stabex operations to be held in joint accounts (Article 211).

This closer control over the utilization of Stabex transfers was introduced in return for the Community agreeing that future transfers should be in the form of grants rather than loans;

(c) while not explicitly stated in the Convention itself, the term 'Framework of Mutual Obligations' implies the existence of conditions concerning the utilization of Stabex transfers which the ACP States undertake to fulfil;

(d) Lomé IV Article 186.2. states that the transfers are to be used in sectors recording a loss of export earnings 'for the benefit of economic operators adversely affected by the loss'. The reference to Stabex being used to benefit economic operators is in contrast to the previous two Conventions where funds were to be used for 'maintaining financial flows in the sector in question' (8). This change, which complements the aim of safeguarding the population's purchasing power (see paragraph 1.2. above), was intended to ensure that a greater proportion of the transfers went to producers;

(e) as far as the calculation technique is concerned, the main innovation has been that the statistics for exports to the Community used for calculations should in principle be those produced by the Statistical Office of the European Communities (Article 200.2). Under previous Conventions, calculations were based on close cross-checks of the export statistics supplied by the ACP States and the Community.

3. THE MANAGEMENT OF THE LOMÉ IV STABEX ALLOCATION

Financial situation of Lomé IV Stabex

3.1. As at 31-12-93, Stabex primary commitments (9) made under Lomé IV amounted to 876,5 Mio ECU. Secondary commitments (10) and payments from the EDF accounts both amounted to 789,7 Mio ECU. The situation by application year is set out in the table below:

>TABLE>

3.2. The difference between primary commitments and payments was very largely due to the fact that the Commission had blocked the signing of transfer agreements, amounting to 85,1 Mio ECU, to three countries because of their political situation (11). In fact the fourth Lomé Convention includes respect for human rights as a general condition for the granting of aid to all EDF operations (12) and the Commission and has not only blocked three Stabex transfers but has also provisionally withheld its signature to FMOs with two other countries (Equatorial Guinea and Togo).

3.3. On 31 December 1993, no definitive transfer decisions had yet been carried out for the 1992 application year although normally these decisions should have been taken by 31 July 1993 (Article 207 (3) of the Convention). This situation arose because of the shortage of Stabex resources to meet eligible transfers.

3.4. The payment figures for the 1990 and 1991 application years should be treated with caution, inasmuch as they do not reflect the actual implementation of Stabex-financed operations but simply transfers from the EDF accounts into ACP States' bank accounts in Europe (see 2.8 (b) (ii) below). By presenting only payments made from the EDF accounts, the Commission's annual financial statements thus give a misleading impression of Stabex operations. An analysis of final payments is given in Chapter 7 of this report.

Insufficiency of resources

3.5. For each of the 1990, 1991 and 1992 application years, the 300 Mio ECU annual instalment for Stabex was insufficient to cover losses in eligible export earnings. This was despite the fact that the Stabex resources for the First Financial Protocol of Lomé IV had been increased by 62 % in nominal terms compared with Lomé III and that the provisions of the Convention leading to a reduction in the basis for calculating entitlements to transfers had been applied in the prescribed manner (13).

3.6. Thus, as Table 3.2. below shows, for the first three application years of Lomé IV, total losses of earnings compared to the reference level amounted to 4 142 Mio ECU, resulting in gross Stabex transfer bases of 3 289 Mio ECU, compared with total Stabex allocations of 900 Mio ECU. This left a total financing gap of 2 389 Mio ECU. By various procedures examined in paragraphs 3.11. to 3.15., it was possible to raise the allocation to 1 205,5 Mio ECU. However, this still left a coverage rate of only 40,7 % for the three years.

>TABLE>

3.7 The excessive recent demands on the Stabex system in fact represent a continuation of the problems already faced by the system under Lomé III for the 1987, 1988 and 1989 application years.

3.8. These difficulties reflect above all the exceptional decline in commodity prices in recent years, which can essentially be traced to continually expanding production in the face of stagnant or only slowly growing demand. In the context of the ACP States, coffee and cocoa, in particular, have been affected by this decline in prices following the collapse of the agreements between producer and consumer countries covering these products. Thus, real coffee prices in 1991 were just 38 % of their 1980 level, while cocoa prices were only 32 % of the 1980 level.

>TABLE>

3.9. It is to be noted that the insufficiency of resources has been exacerbated by the ACP-EEC Council decision taken in November 1991 to waive the requirement for the 13 ACP States that were not classified among the least developed to repay, in certain circumstances, transfers received under the first three Lomé Conventions. The amount of transfers potentially repayable in this way could have amounted to 900 Mio ECU.

'All Destinations' Status

3.10. A further feature in the operation of the Stabex system itself which increased the problem of the insufficiency of resources was compensation for certain losses on exports to non-Community countries. The provisions of the Lomé IV Convention (Article 189 (3)) contain a new clause whereby falls in export earnings may be automatically compensated on the basis of exports of a Stabex product to all destinations, and not just the Community, but only in cases where at least 70 % of an ACP State's earnings from products covered by the system come from exports to non-Community countries. However, 13 countries which were granted an 'all destinations' status on the basis of a special derogation under previous Conventions continued to enjoy this status under Lomé IV, even though the proportion of non-Community export earnings of five of them (Burundi, Cape Verde, the Comoros, Ethiopia, and Rwanda) was less than 70 % of total export earnings. While the derogation for the 13 countries was to continue only for an interim period of three years under Lomé IV, it has since been extended for another year, but only for 1993. In view of the shortage of resources which has constantly plagued the system since 1987, the maintenance of this derogation, which cost a total of 96,1 Mio ECU for the application years 1990-92, is open to question.

Measures taken to combat shortages of Stabex resources

3.11. In order to increase the amount of resources available for Stabex, Article 192 of the Lomé Convention provides for interest earned on Member States' contributions to the Stabex system to be credited to the system's resources. Whereas they could have been placed in a special account with clearly indicated interest rates, Stabex contributions are in fact held in the same accounts as other EDF contributions and a notional interest rate is calculated. This lack of transparency in the system enabled the Commission to state, for the 1992 application year, that Stabex interest available to finance transfers amounted to approximately 30 Mio ECU whereas, according to the accounts of the EDF accounting officer, there was only 1 Mio ECU available. The resources for the subsequent years were thus reduced by 29 Mio ECU.

3.12. The Lomé IV Convention provides for a series of steps to be taken in application years where annual Stabex instalments and interest earned are insufficient to cover transfer entitlements. Thus, in the first place, Article 194 (1) states that advance use may be made of up to 25 % of the following year's instalment. If resources available are still less than the total amount of the transfer bases, Article 194 (2) provides for a 10 % reduction of each transfer basis. If the situation is still not resolved after this procedure, Article 194 (4) states that: 'the Council of Ministers shall evaluate the situation on the basis of a Commission report on the probable development of the system and shall examine the steps to remedy that situation, within the terms of this Convention'.

3.13. For the 1990 application year it was decided by the ACP-EEC Committee of Ambassadors (14), on the basis of Article 194 (4) to increase the 7th EDF Stabex allocation by 100 Mio ECU, principally by transferring to Stabex resources allocated to the Sysmin (support for ACP mining) instrument under previous Conventions (15). As a result of this operation, together with the advance use of 25 % of the 1991 application year instalment and the utilization of interest earned on Stabex funds, a total of 483,7 Mio ECU was made available for the 1990 application year. This still only covered 39 % of the financial needs of the system, however (see Table 3.2).

3.14. For the 1991 application year, the ACP-EEC Committee of Ambassadors decided once more to increase the resources available for that year by authorizing advance use of 12,5 % of the annual instalments for both the 1993 and 1994 application years, in addition to the 25 % advance from the 1992 application year. The total of 391,5 Mio ECU thus made available allowed a coverage rate of 40,9 % (see Table 3.2).

3.15. As already noted in recent Annual Reports of the Court of Auditors, the measures taken by the ACP-EEC Committee of Ambassadors do not fall 'within the terms of the Convention' insofar as:

(a) a transfer such as that made from Sysmin to Stabex for the 1990 application year is not provided for within the general terms of the Convention (16);

(b) the advance use of annual instalments other than the annual instalment directly following the application year in question is not provided for in the Convention (17). The European Parliament's Development and Cooperation Committee has expressed its concern over this procedure in its Opinion on the discharge for the 1992 European Development Fund accounts.

3.16. For the 1992 application year, the Community did not agree to make any additional funds available to Stabex to cover the shortfall for that year. The ACP States' reaction was to insist that additional resources should be made available. As a result, it was only in February 1994 that the ACP States accepted the situation, while expressing reservations and continuing to claim, as in previous years, a right to compensation for the deficit (434 Mio ECU). In this connection, the ACP/EEC Council of Ministers had formally stated in May 1992 that there could not in future be any compensation to cover those years in which there had been an insufficiency of resources, without however questioning the provisions of Articles 194-4 and 195 of the convention.

3.17. For the forthcoming revision of Lomé IV, the ACP States have proposed that a special mechanism should be established to allow ACP States which have not received their full transfer entitlement to have the amount of the shortfall in the transfer deducted from their debts towards the European Community.

3.18. As a result of the continuing insufficiency of Stabex resources, the ACP-EEC Committee of Ambassadors agreed in October 1991 that a study should be carried out to find possible solutions to this problem. The study's initial findings, issued in November 1993, were not satisfactory and no further action was taken to find a solution.

Allocation of Resources by Country

3.19. For the first three application years of Lomé IV, a total of 40 of the 69 ACP States (58 %) have been allocated Stabex transfers. Under Lomé III, 37 out of 66 ACP States (56 %) had received Stabex transfers (18). Details of transfers to individual countries under Lomé IV are given in Annex 2 while Table 3.4 below highlights the principal Stabex beneficiaries under Lomé III and Lomé IV.

>TABLE>

3.20. Whereas under Lomé III, a quarter of the transfers went to the Côte d'Ivoire, the latter's share under Lomé IV has fallen to 18 %. On the other hand, the proportion of Stabex transfers allocated to the three main beneficiary countries remains almost as high under Lomé IV (46,2 %) as under Lomé III (46,8 %).

3.21. Whereas under Lomé III, the 43 ACPs classified as 'least-developed' received 40,1 % of Stabex transfers, under Lomé IV the 44 ACP States in this category received 48,2 %. Nevertheless, the richest third of ACP States continue to be allocated more than half of Stabex resources. This situation is clearly related to the nature of Stabex, the size of transfers being linked to the volume of agricultural trade with the European Community.

4. THE CALCULATION OF STABEX TRANSFERS

4.1. The assessment of transfer entitlements is the responsibility of the Commission. A small division (VIII/B/3) exists for this purpose (19).

Timetable and reliability of statistics

4.2. The Lomé IV Convention (article 207-3) imposes a tight timetable for the completion of assessments, all transfers being required to be decided by 31 July of the year following the application year (see 3.3. and 3.16.).

4.3. The new procedure (under Lomé IV), whereby Stabex transfers are assessed primarily on the basis of the Community's import statistics, has led to a considerable improvement in the way the Stabex system operates. However, it is not possible to complete an assessment without reference to certain statistics which only the ACP States are able to provide (20). These statistics are particularly important for making the reductions provided for in the Article 203 procedure (see paragraphs 4.4.-4.6.) and in cases where ACP States have the benefit of 'all destinations' status (see paragraph 3.10. above). It was noted in relation to statistics supplied by the ACP States that:

(a) the reliability of some statistics is open to question, essentially due to the limited capacities of the ACP States' statistical services. Although the Convention (Article 199 (1)) provides for statistical cooperation to ensure the efficient and rapid functioning of the Stabex system, there has been little cooperation in practice (21);

(b) while the deadline of 31 March set by the Convention (Article 207.1) for submission of statistics by the ACP States was generally respected, the Commission frequently noted anomalies in the statistics which required correction, because ACP countries only submit provisional data for this date;

(c) although the Commission's Delegations are required to certify the correctness of statistical data provided by the ACP countries, they can hardly do so without carrying out detailed investigations, for which their resources are generally insufficient.

Reductions in transfer entitlements

4.4. For the three application years 1990 to 1992, the provisions of Article 203 of the Lomé IV Convention were extensively applied in the calculation of Stabex transfer entitlements. This Article lays down that where the trend of an ACP State's production of the product in question, exports to all destinations, or exports to the Community shows significant changes, consultations shall take place between the Commission and the ACP State concerned to determine whether the transfer basis is to be maintained or reduced, and, if reduced, to what extent (22).

4.5. For these three years, the transfer bases were thus reduced by a total of 580 Mio ECU. These reductions call for the following reservations:

(a) the Stabex Beneficiaries' Handbook (Lomé IV) states that if an ACP State is fully responsible for the drop in production and/or exports, the transfer base shall be reduced by the same percentage as the drop in production and/or exports. On the other hand, if the ACP State is only partially responsible for the drop, the amount of the transfer reduction may be diminished. In cases where the loss of earnings is due to factors outside the control of the ACP State, such as a natural disaster, adverse weather, or a fall in world market prices, no reduction is to be applied (23). However, in practice, it is difficult to quantify how far drops in production are the responsibility of the ACP State and how far they are due to external factors, notably the drop in world prices. While the latter is very frequently a factor, mismanagement of the sector concerned by the ACP State may also often, to varying degrees, affect the situation;

(b) the consultation process in fact gives the Stabex division of the Commission a considerable role in determining the size of the reduction to be made to the final transfer. The Delegations and other Commission departments have had little input in the process. The fact that the consultations have, in practice, to be completed within a short time also limits the scope for a full investigation of the extent to which different factors have caused the drop in production;

(c) there is also a risk that the outcome of the consultations will be influenced by the negotiating capacity and weight of the various ACP States concerned rather than objective technical considerations;

(d) although he may do so, the Financial Controller (DG XX) does not carry out a review of how reductions made under Article 203 were calculated. As the matter is brought to his attention at too late a stage in the procedure, he is not in a position to influence the negotiation process.

4.6. A review of a large sample of the 1990 and 1991 transfers which had been subject to the Article 203 procedure showed that, generally speaking, drops in production had always resulted in proportional reductions in transfers. However, two important exceptions must be pointed out:

(a) in the Côte d'Ivoire, a 41,5 Mio ECU reduction corresponding to a 22,8 % fall in marketed coffee production in 1991 was not applied to the transfer basis by the Commission on the grounds that a revision of the Côte d'Ivoire's statistics had already led to a 34,8 Mio ECU reduction and that the Commission did not believe 'it could have brought about the same level of reduction by conducting its enquiry otherwise' (24). The amount of the reduction was thus more influenced by the negotiations than by the calculation based on the purely technical factors laid down in the Convention;

(b) in Cameroon, for 1991, transfer bases were not reduced, despite a 25,8 % fall in cocoa and a 22 % fall in the case of coffee, which would have led to a 45,6 Mio ECU reduction in the transfer base. The Commission considered that the falls in production were the result of the structural adjustment policies it was supporting and therefore made no reductions. However, by acting in this way it ignored the fact that the reduction in production was an objective of structural adjustment and Cameroon was already receiving payments from the Community to assist it through the adjustment process (25).

Errors noted in calculation of Stabex transfers

4.7. Various cases were noted where transfers had been incorrectly calculated:

(a) an overpayment of 1 298 287 ECU was made to Uganda on its 1991 coffee transfer because of the incorrect treatment of a previous advance payment made in 1981; in its reply to paragraph 15.76 of the Court's annual report on the financial year 1993, the Commission said it had 'initiated procedures, as notified to the budget authorities, to recover the ECU 1 298 330 wrongly paid to Uganda'. The amount was to be deducted from transfers in respect of the 1992 year of application. At the end of the first quarter of 1995, however, the Commission had still not given the Court proof of the actual recovery of the amount concerned;

(b) the 1991 transfer basis for coffee from Burundi was understated by 1,55 Mio ECU because of an arithmetical error in its calculation;

(c) due to a calculation error, partly attributable to the confused data supplied by the ACP State, no transfers for the 1990 application year were made to Lesotho for mohair and wool although the combined transfer bases for these products amounted to at least 1,16 Mio ECU.

These errors point to the need for stricter checks by the Financial Controller (see 4.5.(d)).

The impact of product quality on prices

4.8. In many ACP countries, there has been a deterioration in product quality, often resulting from such factors as a lack of investment in agricultural inputs and inadequate extension facilities and a shift from plantation to smallholder production (26). It is important to note that this drop in quality, which is essentially the responsibility of the producing country itself, has also been a significant cause of the fall in prices. However, the Convention deals exclusively with changes in quantity and price and has no provisions for taking into account the part played by declining quality in falls in prices.

5. THE PAYMENT OF STABEX TRANSFERS

5.1. The Lomé IV Convention requires that, when a transfer agreement is signed, the funds should be paid over in ECUs into an interest-bearing account (see 2.8). Under previous Conventions, the total amount of the transfer was paid over directly to the ACP State after the transfer agreement had been signed. This meant that the Commission had little control over the actual utilization of the funds.

5.2. These new arrangements have resulted in several new practices:

(a) the Lomé Convention refers only to the requirement for the Commission to be one of the joint signatories. In practice, however, the Head of Delegation of the Commission in the ACP State concerned by the transfer has been made the account signatory for the Commission. This creates a second level of management of EDF funds which, though under the control of the Commission, is not recognized by the EDF Financial Regulation, and is not subject to the approval of the authorizing officer, the EDF accounting officer and the Financial Controller (27);

(b) the Lomé Convention does not specify the location of the ECU account for Stabex transfers, although it does in the case of structural adjustment transfers (Article 250.3). The Commission has taken the position, arguing by analogy, that ECU accounts are only to be opened in Europe, on the grounds that funds are more secure in Europe and that ECU interest rates are higher there (28);

(c) for everyday management, another tier of joint signatory accounts has had to be established in each ACP State to enable payments in local currency to be made;

(d) finally, to ensure that operations are clearly accounted for, the Commission has insisted that separate accounts (or sub-accounts) should be opened for each application year.

All of this mechanism makes management by the Commission's Delegations cumbersome, with no related increase in their resources.

5.3. A feature of the new system has been the often considerable delays in the payment of transfers, as can be seen from the statistics in Table 5.1. (29). Originally payment was to be made immediately after the signing of the transfer agreement which, for its part, would have been completed immediately after the transfer decision (30). However, in practice, for the 1990 application year, the median time for payments to European bank accounts (to be carried out after the transfer decision) was five and a half months, whereas for the 1991 application year the time required was four and a half months.

5.4. Although the Lomé IV Convention allows advance transfers of funds prior to the end of the year of application on the basis of estimates of likely losses, no such advances have been made as they were under the previous Conventions. The main reason for this is the difficulty of determining the size of advances in a situation where the insufficiency of resources requires transfer levels to be scaled down.

>TABLE>

5.5. While there has been some improvement in payment times for the 1991 application year, the delays are nevertheless of a serious nature, particularly given that Article 206 of the Lomé IV Convention specifically states that: 'the ACP State concerned and the Commission shall take such steps as are required to ensure rapid advances and transfers . . .'

5.6. The essential reason for the delays has been the length of time taken to open the ECU bank accounts. In certain cases, ACP States, noting that the Convention itself did not require ECU accounts to be opened in Europe, wished to open ECU accounts in their own countries. In fact the Commission, in its first guidance notes, while expressing a preference for a European location for the accounts, was ready to admit exceptions (31). While the principle that accounts were to be opened in Europe was accepted as from the 1991 application year, the requirement for new accounts/sub-accounts to be opened for each application year has led to even longer payment delays.

5.7. In practice, the ACP States have been made responsible for opening the bank accounts. Not only the Convention but also the Stabex beneficiaries' handbook is silent on who is responsible for opening the accounts. Once this question had been settled, however, the ACP States did not always make the necessary arrangements rapidly.

5.8. Despite the delays in paying funds into European bank accounts, ACP States earned significant amounts in interest on funds in these accounts, largely as a result of subsequent delays in implementing Stabex operations. Thus, it is estimated that total interest earned on these accounts amounted to 50 Mio ECU as at 31 December 1993. It is likely that interest earned could have been even higher had ACP States required banks to tender for the placement of Stabex funds. Only one country, Madagascar, chose to do this.

5.9. While in virtually every case interest was paid by European banks on Stabex monies deposited with them, approximately a third of ACP States subsequently transferred these funds to an account with a local bank which did not pay interest, despite the requirements of the Convention (Article 211.3). In Burundi, for example, it was estimated that the non-payment of interest on Stabex led to a loss of income of over 0,7 Mio ECU for the 14 months from July 1992 to September 1993.

6. THE PREPARATION OF THE FRAMEWORKS OF MUTUAL OBLIGATIONS

Provisions of the Convention concerning the utilization of funds

6.1. It would have been possible to conceive of Stabex simply as a system to help ACP States by providing compensation for loss of export earnings. Indeed, Article 186 (1) of the Lomé IV Convention refers to 'the aim of remedying the harmful effects' of export earnings rather than their causes. Transfer payments could thus be made to the ACP State to afford assistance at the macroeconomic level, notably by compensating for the loss of foreign exchange by providing balance of payment support to maintain import levels or service external debt. This stabilizing function of Stabex emerges from the requirements of the Lomé IV Convention for the rapid implementation of Stabex transfers (32).

6.2. However, Article 186.2 restricts the use of transfers by requiring that, in order to attain the objectives laid down in Article 186.1, the 'transfers shall be devoted, in accordance with a framework of mutual obligations to be agreed between the ACP State concerned and the Commission in each case, either to the sector, interpreted in the widest possible sense that recorded the loss of export earnings and be used there for the benefit of economic operators affected by the loss, or where appropriate, to diversification, either for use in other appropriate productive sectors in principle agricultural or for the processing of agricultural products'.

6.3. Stabex is thus required by the terms of the Convention to be used in a sectoral context. However, Article 209.4 notes that the transfer should be used to second any adjustment operation and support reforms taking place in the sector in question.

6.4. The requirement of Article 186 of the Lomé Convention for the transfers to be used in the sectors affected by loss of earnings is based on the following principles:

(a) as Stabex transfers are generated by drops in export earnings in relation to specific products, it is logical to link utilization of the transfer to the sector affected by the loss, rather than using it at the macroeconomic level;

(b) in terms of the aid instruments available under the Lomé Convention, Stabex has a sectoral role which complements the macroeconomic vocation of the structural adjustment funds and the interventions at project level of the National Indicative Programmes;

(c) the causes of losses of export earnings should be tackled in order to reduce the need for further transfers to the sector in question.

6.5. However, the following observations may be made concerning these principles:

(a) in many ACP States the importance of Stabex products to the economy is such that losses in export earnings in the sector concerned have a serious impact at macroeconomic as well as sectoral level;

(b) it should be established whether specific instruments should have, or do have, a unique vocation. Thus, for example, National Indicative Programme funds under Lomé IV have been extensively used to support structural adjustment programmes (33). The element unique to each instrument is in fact not always its vocation but the criteria for determining eligibility for aid from the instrument;

(c) it may be argued that transfers should be directed not to sectors in decline but to those where there is potential for growth. Moreover, present economic orthodoxies being promoted in Africa emphasise the desirability of the market allocating funds, rather than relying on assistance from public funds.

6.6. Independent of the issues noted above, a fundamental contradiction exists between, on the one hand, the aim that Stabex should be a quick-disbursement instrument and, on the other, the policy of investing transfers in the agricultural sector in order to bring about long-term solutions to the problems faced by the sector. It could be argued that the National Indicative Programme (NIP), for which an implementation timeframe of at least five years is specified, would be a more suitable instrument for such an approach. Indeed, the flow of Stabex funds to a country over a period of several years is difficult to predict, whereas the amount of NIP funds available is fixed at the beginning of the five-year period. This makes planning medium to long term investments easier with NIP funds than with Stabex funds.

6.7. The actual application of Stabex funds was debated at length within the Commission during the first two years of the Convention, notably between those seeking to insist on the sectoral vocation of Stabex and those wishing to use it in a broad macroeconomic context to provide balance of payments and budgetary support. The fact that the utilization of Stabex transfers often leads to the generation of counterpart funds has further fuelled the debate. Various ACP States have also formulated their own opinions as to the utilization of transfers by attempting to take account of their own particular circumstances.

6.8. In February 1991, the Commission initially stressed the sectoral role of Stabex. Subsequently, in May 1992, it placed considerably more emphasis on the link and coherence between Stabex and structural adjustment.

Procedures for preparing frameworks of mutual obligations (FMOs)

6.9. The proposals for the utilization of funds submitted by the ACP States to the Commission (2.6) are examined jointly (Article 209 (3)) and, when agreement has been reached, the ACP State and the Commission sign the protocol setting up a framework of mutual obligations (Article 210).

6.10. The substantial analysis document provided for in Article 209 of the Fourth Lomé Convention calls not just for information but 'analysis', which should examine not only the problems encountered by the sector but also the policies being pursued by the ACP State in the sector. This analysis should ensure that the ACP States have put clear policies in place prior to determining specific measures to finance.

6.11. The substantial analysis document was intended to be the basis for drafting the FMO document. With the support of the Stabex beneficiaries' handbook, (4.5(a)), no general guidelines were issued by the Commission to ACP States stating more precisely what was required in the substantial analysis.

6.12. As already noted, the Framework of Mutual Obligations represented a completely new procedure and the Commission developed various guidelines concerning the format and content of this document. Thus, in February 1991, a checklist of the main points to be included in the FMOs was established (34). This was followed by the identification in June 1991 of fixed clauses to be included in each FMO (35). Finally a proposed outline for FMOs was issued in May 1992 (36). However, these various documents ought to be systematically distributed to the ACP States, because they will be required to make proposals.

Delays in preparation of FMOs

6.13. As was the case for the payment of Stabex transfers into ACP States' European bank accounts, the stage of preparing the utilization of Stabex funds has required considerably more time than was originally foreseen. While it had been foreseen that draft FMOs might be available by the time of the transfer decisions, the median time between the 1990 application year transfer decision and the signature of the FMOs for the 1990 application year amounted to 12 months (37). Thus the FMOs for 14 countries were not signed within twelve months of the transfer decision (38).

6.14. For the 1991 FMOs, the median time required increased still further to 19 months, the first FMO (Côte d'Ivoire) not being signed until almost eight months after the transfer decisions had been taken. Only five other countries had signed the FMO by the end of July 1993, 12 months after the transfer decisions (39).

Identification of operations to be financed

6.15. The actual choice of measures to be financed was frequently the subject of lengthy negotiations between the ACP State and the Commission as well as within the ACP administrations. The ACP State and the Commission were supposed to consult with other donors over proposed uses to ensure proper coordination. While such negotiations and consultations inevitably required time, the process was often extended as a result of differing views on the admissibility and desirability of proposed uses of Stabex. This was particularly the case in relation to ACP proposals to utilise funds directly to support price stabilization Funds for their own commodities or to finance numerous individual projects. In general, the Commission did not accept either approach. In the case of the stabilization Funds, it sought to persuade ACP States to adapt domestic commodity prices to world prices; it also encouraged them to pursue an overall sector approach aimed at overcoming the problems of the sector, rather than following the traditional project approach which had limited results.

6.16. The process of preparing the FMOs was also made longer because the substantial analyses carried out by the ACP States often proved to be of limited utility. In many cases, they were overly concise (e.g. the Central African Republic 1990 and 1991; Tanzania 1990; Ethiopia 1991), submitted late (e.g. Ghana 1990 and 1991; Kenya 1990 and 1990; Madagascar 1990) or, in certain cases, not submitted at all (Côte d'Ivoire 1991; Ethiopia 1990; Uganda 1991). The observed shortcomings were due to the failure of individual ACP States, the sectoral policies of which have not always been consistent, to produce adequate proposals for utilizing the Stabex funds. Moreover, the need to redesign sectoral policies to reflect new structural adjustment programmes or re-examine them following changes in government further contributed to delays. For its part, the Commission on several occasions found itself in the position of having little experience of the sector concerned by the transfer and therefore not being able to evaluate proposals made by the ACP State or itself contribute to policy formulation (40).

6.17. The size of some of the transfers meant that a considerable planning effort was required. Apart from the absolute size of certain transfers, the size of the transfer relative to overall EDF aid is an indication of the additional workload involved, particularly for the Commission's Delegations. Thus, for 19 countries, the combined Stabex transfers for the 1990 and 1991 application years amounted to more than 20 % of their National Indicative Programme (NIP) allocations covering the five years of the First Financial Protocol of Lomé IV (see Appendix 4). For three of these countries, the transfers were actually greater than their NIP allocation (Côte d'Ivoire: 181 %; Cameroon: 124 %; Papua New Guinea: 105 %).

6.18. The fact that the 1991 FMOs are taking on average significantly longer to prepare than the 1990 FMOs is due to the time required to implement the 1990 FMOs, it being generally desirable to evaluate existing operations before preparing new FMOs. There is thus a risk that there will be a build-up of Stabex transfers waiting to be disbursed, as transfers are generated more quickly than funds can be disbursed. This is even more the case as a result of the delays in taking the transfer decisions for the 1992 application year caused by the hesitations of the ACP countries.

Conditionality

6.19. Apart from the negotiations regarding the utilization of the funds, the conditionality attached to the release of funds to ACP States was a further important factor slowing down the preparation of the FMOs. The concept of making the utilization of transfers conditional on the ACP State respecting certain commitments was new to Stabex. The Convention itself did not clearly spell out the notion of conditionality, but merely hinted at it in Article 210 in the phrase 'Framework of mutual obligations'. The ACP States thus had to adapt to a new dimension to Stabex which was not always easily accepted.

6.20. The specific conditions which the Commission desired to see attached to the release of Stabex funds in most cases focused on the liberalization of the sectors concerned by the transfers. Reservations on the part of the ACP States, whether for technical reasons connected with the specific proposals put forward for reform, or for political reasons owing to the threat posed by the reform process to certain interest groups, required considerable negotiation to overcome. The Commission had to be careful to identify commitments which were of sufficient import, once implemented, to have an impact on the sector while avoiding insisting on commitments which would not be complied with and would hence cause funds to be blocked.

Commission procedures

6.21. While the process of negotiation between the ACP State and the Commission often took considerable time, within the Commission the period required for the different departments which approve the proposals (Delegation, Stabex service, Structural Adjustment Unit, geographical services, Chief Authorizing Officer (41)), has further slowed down the finalization of the FMOs. The need to reconcile the differing opinions of the departments concerned over certain utilizations has contributed to such delays.

6.22. Two other specific procedures have further delayed the process of approval:

(a) while initially the Commission took the view that the FMOs could be signed by Commission Delegates on behalf of the Commission, from September 1992 onwards a different interpretation of Article 210 of the Convention was applied, so that each FMO had to be signed by the Chief Authorizing Officer after approval by the Commission;

(b) all FMOs, once finalised but before signature, are required to be submitted to Member States for information and a period of one month must elapse before the FMO can be signed. In fact, only one case was identified (42) where a Member State raised a question at this stage. This procedure, which supplements consultations between the Delegation and the Member States represented in the ACP State during the drafting of the FMO, does not appear essential.

6.23. Lastly, it should be noted that while the Convention, in an effort to ensure that Stabex fulfils its vocation as a quick-disbursement instrument, lays down strict deadlines for establishing transfer rights and making transfer decisions, no deadlines exist for the implementation of subsequent stages of the system. On the other hand, however, Annex XLVI to the Convention states that: 'The Contracting Parties agree to adopt the simplest possible procedures for the application of Article 210 and Article 211 (2) and (3) with a view to making transfers available to the ACP States as swiftly as possible.' Accordingly, a broader delegation of powers to the Commission's Delegates, together with a boost to their resources and increased ex post audits, would have been desirable.

7. THE UTILIZATION OF STABEX TRANSFERS

Financial implementation of the frameworks of mutual obligations

7.1. As of 31 December 1993 the Commission's central departments had no information either on overall payments from these ECU accounts or on the disbursement of Stabex counterpart funds. According to the Commission, the reason for this is that, once the money is paid out from the EDF accounts it becomes the property of the ACP States. However valid this view might be, it clearly behoves the Commission to monitor the actual implementation of a system which it is responsible for managing until completion.

7.2. In order to overcome this lack of information, the Court requested each Commission Delegation responsible for an ACP State which had received a Lomé IV Stabex transfer to provide information on final Stabex disbursements as of 31 October 1993, or two years after the transfer decisions had been taken for the 1990 application year.

7.3. From the information supplied by the Delegations, it is estimated that, as far as transfers for the 1990 application year are concerned, 291,6 Mio ECU of the 437,6 Mio ECU paid into the ACP States' European bank accounts (See Table 3.1) had been disbursed as of 31 October 1993, which corresponds to an implementation rate of 67 %. It is important, however, for this rate to be seen in context:

(a) firstly, disbursements were intended to compensate for the effects of losses in revenue suffered in 1990, that is almost three years earlier;

(b) secondly, 53 % of the total funds disbursed only concerned two countries, the Côte d'Ivoire and Cameroon (43), which each had disbursement rates of more than 90 %. Apart from these two countries, only four others had a disbursement rate of more than 90 % (Burundi, Ghana, Papua New Guinea and Tanzania). Eighteen countries had a disbursement rate of less than 50 % and half of these had made no disbursements at all (Benin, Burkina Faso, Kiribati, Mali, Sao Tomé, Sierra Leone, Togo, Tuvalu, Western Samoa). Thus, excluding the Côte d'Ivoire and Cameroon, the average disbursement rate was below 50 % (see Annex 5);

(c) in ten countries, it was intended that ECU payments would create counterpart funds in local currency with a total approximate value of 186 Mio ECU. Of this amount, the local currency equivalent of only approximately 53 Mio ECU had been disbursed as of 31 October 1993, a utilization rate of 28 % (see Annex 6);

(d) while in European bank accounts, the Stabex transfers had generated significant amounts of interest, estimated to be approximately 50 Mio ECU on 31 December 1993. These came on top of the original transfers.

7.4. As far as the 1991 application year was concerned, on 31 October 1993, 15 months after the transfer decisions had been taken and almost two years after the financial year in which the losses in export earnings had occurred, only 39 Mio ECU of the 352,1 Mio ECU paid into ACP States' European bank accounts had been disbursed. This represented an implementation rate of just 11 %. Once again, the majority of these payments were made in the Côte d'Ivoire (36,7 Mio ECU), only two other countries, Cameroon and Uganda, having commenced disbursements. This situation is largely to be explained by the fact that, as of 31 October 1993, most 1991 FMOs had only recently been signed or had still to be signed (see paragraph 6.14).

Analysis of the allocation of Stabex funds

7.5. Table 7.1 summarizes the main uses of Stabex transfers as provided in the FMOs which had been signed as of 31 December 1993:

>TABLE>

7.6. The following paragraphs highlight the particular issues that have arisen during implementation and examine the reasons for slow disbursement in specific countries.

Import programmes

General Import Programmes

7.7. Three East African countries allocated the majority of their 1990 Stabex transfers to general import programmes (GIPs):

>TABLE>

7.8. The choice of an import programme can be considered appropriate in so far as the loss of earnings has led to a reduction in the foreign exchange available for imports. None of these countries having convertible currencies, the loss of foreign exchange was particularly problematic. It was also expected that an import programme would permit the rapid disbursement of funds. However, given that Stabex is intended to be a sectoral instrument, it might have been expected that the transfers would have been used to support a sectoral rather than a general import programme. Nevertheless, general import programmes were preferred by the Commission for the following reasons:

(a) linking the utilization of Stabex funds to general import programmes financed from EDF structural adjustment funds strengthened the Community's support for the structural adjustment process;

(b) general import programmes were in greater harmony with the policy of the liberalization of foreign exchange mechanisms being pursued under the structural adjustment programmes;

(c) the sectors concerned did not appear to be in much of a position to absorb large sectoral import programmes;

(d) each GIP was to give rise to counterpart funds which would themselves be allocated to the agricultural sector.

7.9. In Tanzania, Stabex funds were used to support a GIP to which 30 Mio ECU from Lomé IV structural adjustment support funds were also allocated. This operation can be considered to have been relatively successful, despite initial delays which occurred after the signing of the FMO (12 September 1992). These were due to an agreement between the Government and the Commission to disburse the structural adjustment funds first. Nevertheless, the full allocation of 16 Mio ECU from the Stabex transfer was disbursed between March and September 1993 on goods of a productive nature as required by the FMO.

7.10. By contrast, utilization of funds for the GIP in Uganda was blocked as of August 1992, by which time only 9,3 Mio ECU (33,2 % of allocation) had been disbursed. This was because the country's requirements for such aid in 1992 and 1993 were only half the estimated figure and the country preferred to make use of other support measures that allowed the counterpart funds created to be channelled directly to the Treasury (budgetary support). The utilization of structural adjustment funds also suffered considerable delays (44).

7.11. In Burundi, 9 Mio ECU of the Stabex transfer was added to structural adjustment funds of 12 Mio ECU to finance a GIP. Although it had been planned that the 21 Mio ECU would be disbursed in 10 months, in October 1993, 15 months after signature of the FMO, utilization of the structural adjustment funds had yet to begin, as documentation to justify the utilization of the Stabex funds was still being put together.

7.12. Although a first tranche of the Stabex transfer (4,5 Mio ECU) was released in July 1992, the documentation to justify its utilization for oil imports was not submitted by the Burundi authorities until December 1992. The second tranche was not released until March 1993 and the supporting documentation not provided until September 1993. The delays involved are even less justified in view of the fact that the financing of imports was largely retroactive (45). They are to be largely attributed to administrative inefficiencies and problems in finding import documentation complying with EDF regulations, notably certificates of origin and pro forma invoices. In these circumstances it is clear that Stabex funds were not used for a GIP but only as balance of payments support. Thus, the requirement for imports to comply with EDF regulations may merely be seen as a cosmetic exercise whose only effect is to generate pointless administrative costs.

7.13. In a fourth country, Ghana, the 1990 FMO allowed the Stabex transfer (6,9 Mio ECU) to be made available directly to the Bank of Ghana to increase its exchange reserves without there being a GIP. The FMO thus sought only to attach conditions to the release and use of the resultant counterpart funds on the basis of the fact that the Ghanian currency was internally convertible (46). This approach, while having the clear advantage of accelerating the disbursement of Stabex funds, raises questions concerning the degree of currency convertibility required in a particular country before such a policy is implemented (47). It is also unclear why Ghana still uses GIPs to disburse structural adjustment funds when this is not required for Stabex resources.

7.14. Moreover, the increasing liberalization of foreign exchange markets in many African countries may invalidate the GIP/SIP approach with its strict requirements concerning supporting documents (48).

Sectoral Import Programmes

7.15. Several countries allocated their Stabex transfers entirely or largely to sectoral import programmes. The main countries concerned were Ethiopia, Rwanda, Kenya, Mauritania and Madagascar.

7.16. Ethiopia established SIPs both for the sectors generating the Stabex transfers (coffee and leather) and for other sectors as is shown in Table 7.3:

>TABLE>

7.17. The FMO was drawn up in the context of the Emergency Recovery and Reconstruction Programme (49) which was financed by several donors and launched in July 1992 to help the country, which was just emerging from civil war. Approximately 60 % of the funds were to be devoted to quick-disbursement operations under the ERRP. However, notwithstanding the possibilities for simplification offered by Article 248 of Lomé IV, approximately half of the value of the supplies to be purchased were subject to EDF procurement procedures. The unfamiliarity of beneficiaries with these procedures, the considerable workload involved and numerous delays meant that no supplies had actually been delivered as of 31 October 1993, eight months after the signing of the FMO.

7.18. The large allocation of funds to the transport sector was partly intended to overcome the transportation difficulties faced by all crops. This objective was to be achieved in the short term through the provision of spare parts for vehicles.

7.19. Although only 28,4 % of the Stabex funds were allocated directly to the coffee and leather sectors, applications from these sectors amounted to much less than this and only concerned 20,6 % of the available funds. On the other hand, applications had been received for 87,6 % of the funds allocated for agricultural diversification operations outside the coffee and leather sectors. It would thus seem that a sectoral import programme for the agricultural sector as a whole, without funds being specifically directed to the coffee and leather sectors, might have been more appropriate.

7.20. The 1991 FMO (signed on 26 August 1993) sought to tackle the problem of slow tendering procedures by enabling operations targeted on urgent rehabilitation needs to follow similar procedures to those established for emergency assistance contracts under Article 300 of Lomé IV. At the end of 1993, the implementation of the programme had only just started and the effects of this change in procedures did not appear to be decisive.

7.21. In Rwanda, the national administration, following the example of the three East African countries examined in paragraphs 7.9. to 7.11., had wished to use the funds to finance a GIP but the Commission instead insisted on the implementation of an SIP. The programme sought to provide urgently needed inputs for the sectors affected by the loss of earnings in 1990, principally coffee. However, while it was intended that the inputs should be made available in October 1992, at the beginning of the new season, on 31 October 1993, only 35 % (4,9 Mio ECU) of the 14 Mio ECU reserved for the SIP part of the FMO had been disbursed. The delays were due to the internal tensions already faced by the country, the weakness of its administration, and the unsuitability of lengthy EDF tendering procedures.

7.22. In Kenya, a sectoral import programme for the agricultural sector as a whole was initially adopted. The authorities felt that the absorptive capacity of the coffee sector itself, where the losses occurred, had already been taken up and therefore decided to allow imports within the agricultural sector to be determined by the market. However, in January 1993, to respond to what was considered by the authorities to be an emergency situation, and in the context of a complete revision of the FMO (8.24), 20 Mio ECU was reallocated to an emergency fertilizer import programme designed to meet the needs of the new cereal planting season in the first quarter of 1993. This programme encountered a number of problems:

(a) the Kenyan authorities overestimated the amount of fertilizer required by failing to take account of fertilizer imports by the private sector (50);

(b) despite the Commission's efforts to accelerate tender procedures, the first deliveries only took place in May 1993, partly as a result of difficulties encountered at the tender evaluation stage;

(c) because of the devaluation of the Kenyan shilling and the late delivery of the fertilizer, the surplus consignments had to be sold below import prices.

7.23. In Mauritania, a sectoral import programme for 8 Mio ECU was used to finance the purchasing of goods and services for the fisheries sector, which had suffered a loss of earnings. The programme was launched in August 1992 and by the end of October 1993 only 5 Mio ECU (62,5 %) had been used because private sector firms resented the requirement that supplies costing more than 100 000 ECU had to be subjected to tendering procedures, a practice to which they were not very accustomed. In order to get round this requirement, the imports were split up so as to make purchases correspond to the needs of the vessel for one single tide (51), or less than 100 000 ECU.

7.24. In the case of Madagascar, because the World Bank was already carrying out restructuring programmes in the two main sectors (coffee and vanilla) generating the transfers, it was decided to allocate the transfer to an import programme to support manufacturing firms. After the signature of the FMO in July 1992, the first tranche of the programme for 10,5 Mio ECU was disbursed and justified between October 1992 and February 1993. In all, 757 orders were placed, of which 276 were for less than 5 000 ECU. Inadequate documentation proved a frequent problem, particularly with regard to certificates of origin. As of 31 October 1993, the second tranche had not been released because of the Government's failure to meet certain conditions laid down in the FMO.

Utilization of counterpart funds

7.25. For those countries which used Stabex funds to finance general and sectoral import programmes to provide balance of payments support, the counterpart funds generated by these programmes also represented an important source of assistance (7.3 (c)). Although the possible uses of counterpart funds resulting from Stabex operations were not explicitly laid down in Lomé IV, the Commission, in its various guidelines, notably Article 186.2, insisted that they should be used in accordance with the provisions governing the Stabex system, and that their specific utilization should be agreed in the relevant FMOs (Article 211-3). This approach reflects the significance of the counterpart funds to ACP States' economies and the desire to ensure that they are utilized effectively, but the necessary controls thus created slowed down Stabex implementation still further.

7.26. In several countries, the counterpart funds were used in the same sector as those that had generated the Stabex transfers:

(a) in Ghana, the 6,9 Mio ECU raised in counterpart funds was used to pay end-of-service entitlements to approximately 5 300 staff affected by redundancies at the Ghana Cocoa Board. To ensure the exercise was correctly implemented, the payments were made in two tranches, each one subject to an external audit. No major problems were identified on this occasion;

(b) in Tanzania, counterpart funds generated by the GIP were allocated to supporting the prices paid to coffee producers with the aim of maintaining their confidence in the sector. Approximately 420 000 farmers benefited from the scheme. The payments, made between August and October 1993, generally provided a 20 % increase on domestic prices for the 1990-91 season. The scheme was carefully designed both to ensure that payments actually reached farmers and to reward those that produced better quality coffee. Overall, the scheme can be considered a success. However, in the south of the country where these were the only payments the farmers received in 1993, it is doubtful that coffee production was revitalized;

(c) in Burundi, for the local currency equivalent of 9 Mio ECU which was to be generated from the GIP, the 1990 FMO did not allocate any amounts to specific utilizations but simply stated four priority areas. In fact, by October 1993 when 79,5 % of the funds to be generated had been disbursed, nearly all of the disbursements (the equivalent of 6,9 Mio ECU) involved support to just one of these priority areas: the coffee stabilization Fund, for which the hard currency funds allocated for this purpose had proved insufficient (see paragraph 7.49);

(d) for Uganda, the 1990 FMO set down an indicative allocation of the 14 Mio ECU in counterpart funds to be generated by the first tranche of the GIP. The funds were to be allocated to the cotton sector (9 Mio ECU), in particular in the form of a crop financing revolving fund, and to the coffee sector (5 Mio ECU). While operations in the coffee sector to assist the Coffee Marketing Board proceeded satisfactorily, the cotton crop financing revolving fund experienced difficulties. On 31 October 1993, of the 4 500 Mio Uganda Shillings (Ush) lent since May 1992 (52), only 2 200 Mio Ush had been repaid, mainly as a result of one of the three organizations benefiting from the loans, the Lint Marketing Board, having defaulted on a repayment of 1 200 Mio (Ush) (53). The difficulties encountered seem to have arisen because of the continued fall in world cotton prices and the fact that, at the time, the Commission had a limited knowledge of the Ugandan cotton sector. Contrary to the provision in the FMO regarding the allocation of counterpart funds, in December 1992, 300 Mio Ush was disbursed to support an army retrenchment programme involving the demobilization of 22 000 soldiers. The utilization of Stabex funds in this way is clearly outside the scope of the instrument.

7.27. In four other countries, the FMOs allocated at least part of the counterpart funds outside the sector generating the transfer:

(a) in Ethiopia, the counterpart funds were to be used in the context of the ERRP (7.16-7.20). Priority attention was to be given to the local costs of the ERRP and in particular local expenditure under the ERRP social fund. Despite the emergency nature of the programme, no counterpart funds had been spent as of 31 October 1993;

(b) in Madagascar, counterpart funds generated by the first tranche of the SIP were used to support the 1992 and 1993 national budgets to cover priority areas in health, education and public works. While such a usage was not foreseen for Stabex, it corresponded to a policy agreed in the context of Madagascar's National Indicative Programme, and approved by Council resolutions adopted in May 1991, whereby all counterpart funds, whichever instrument had generated them, would be used in a consistent fashion;

(c) similarly in Rwanda, counterpart funds were to be used to support the national budget, in accordance with an agreement on the utilization of counterpart funds which had been reached between the Rwandan authorities and the principal donors in the country. However, due to the large amount available and the low disbursement rate of the national treasury, no Stabex counterpart funds had been disbursed as of 31 October 1993;

(d) in Kenya, the 1990 FMO provided for the counterpart funds to be allocated outside the coffee sector, as was the case with the transfer itself. Thus, the first tranche of 10 Mio ECU in counterpart funds generated by the SIP was used to contribute to the Community's continued large scale support of the country's cereal sector reform programme (CSRP) (54). However, after difficulties were encountered in the implementation of the CSRP, the remaining 20 Mio ECU in counterpart funds to be generated were reallocated to the coffee sector in spite of the sector's very limited absorptive capacity. Disbursement was nevertheless suspended, pending agreement between the Kenyan government and the Commission on an appropriate reform policy for the coffee sector.

Internal debt relief measures

7.28. Whereas for ACP States with non-convertible currencies Stabex transfers above all provided a means of easing their balance of payments difficulties through financing import programmes, countries in the franc zone primarily used the 1990 transfers to tackle the financial difficulties faced by the coffee and cocoa sectors. In these countries, stabilization funds had not been enough to compensate for falling world prices in the late 1980s. Indeed, as domestic price reductions had been insufficient and the surpluses generated in earlier years had often been used for purposes outside the price stabilization systems, the latter ended up bankrupt, threatening the collapse of the crop marketing system (producer, banks and exporters) and placing a growing burden on State finances. The persistently overvalued CFA franc only exacerbated these problems (55). The authorities of the franc zone prevented governments from reacting by increasing the money supply and Stabex funds were thus employed to try to address the internal debt problem that had arisen through the mismanagement of agricultural price stabilization schemes in the Côte d'Ivoire and Cameroon and, to a lesser extent, in the Central African Republic. In each case, the granting of Community aid was made subject to specific commitments from the governments to reform the sectors concerned (see Chapter 8).

7.29. In Cameroon, the stabilization board, the ONCPB (56), had become bankrupt as a result of the Government's consistent policy of maintaining producer prices for coffee at levels considerably higher than world prices, which had fallen steeply since 1987. Moreover, it had also preserved the system of administratively fixed prices for intermediaries of the system (buyers, factories, exporters), which were set at levels considerably higher than their actual costs. In 1989, the coffee and cocoa sectors had become virtually paralysed as a result of ONCPB's debts towards the various economic operators, estimated at more than 100 000 Mio FCFA (288 Mio ECU).

7.30. In response to this situation, the donors supported the Cameroon government in its efforts to reform the sectors concerned (see paragraphs 8.20. to 8.22.), while providing the majority of the finance to allow the sectors' debts to be paid off. In 1989 and 1990, the Community contributed 48 000 Mio FCFA (138 Mio ECU), almost half of the funding needed, using Lomé III Stabex transfers for the 1988 and 1989 application years.

7.31. At the time, it had been foreseen that the restructuring process would be completed and financial equilibrium restored without the need for further Stabex interventions. However, delays in adjusting domestic prices to world prices resulted in further arrears being generated. In this situation, the Commission decided that part of the 1990 Stabex transfers to Cameroon (57) could also be used to pay off these arrears, while nevertheless insisting that this was the final year in which Stabex funds would be used in this way. Thus a total of 14 984 Mio FCFA (43,0 Mio ECU) was allocated in the 1990 FMO for this purpose.

7.32. As a result, producers received the payment of two sums: one of 2 164 Mio FCFA (6,2 Mio ECU) to cover the debts of the stabilization fund and another of 2 857 Mio FCFA (8,2 Mio ECU) to cover those of exporters. Whereas the 1990 FMO only provided explicitly for the reimbursement of the principal on the new support arrears (7.31) resulting from the delays that had occurred in the process of reform, the sum of 7 776 Mio FCFA (22,3 Mio ECU) was paid over directly to 19 exporters in order to cover both past claims (the 1986-1990 harvests) and the interest owing to the banks by exporters who had not been able to repay their loans (58). More generally, it should be stressed that it is difficult to judge whether the payments made were always justified in the case of old operations because of the considerable scope economic operators had for manipulating the system (e.g. the virtual impossibility of establishing whether sales actually took place, administratively fixed prices bearing no relation to real costs and the problem of some operators being both debtors and creditors of ONCPB).

7.33. Despite the large amounts contributed under the 1990 FMO, as of 31 July 1993 it was still estimated that the debts of the stabilization board amounted to approximately 20 000 Mio FCFA, of which about half was owed to exporters for price support measures. At the same time, however, ONCPB still held significant assets, notably property in Cameroon and France, which, if the government had not been against it, could have been used to help to pay its creditors, thereby proportionately reducing the drawings on Stabex funds.

7.34. The Commission maintained a firm position towards the Cameroon authorities during the negotiation of the 1991 FMO, insisting that the 1991 Stabex transfers (69,2 Mio ECU), should be used to encourage the development of production in the coffee and cocoa sectors and the promotion of the rural sector in general.

7.35. In the Côte d'Ivoire, 82,9 Mio ECU of Stabex transfers were used to pay off internal debt, not only relating to the coffee and cocoa sectors, which had given rise to the 1990 Stabex transfers, but also in the cotton and rubber sectors. These measures were part of a World Bank agricultural adjustment programme and, at a macroeconomic level, the economic recovery programmes, in particular, those seeking to restore balance in public finance and clear the internal and external arrears of the government.

7.36. In relation to the coffee and cocoa sectors, as in Cameroon, the maintenance of a guaranteed price to producers that was considerably higher than the world price had caused the stabilization fund (CSSPPA (59)) to become seriously indebted towards the economic operators. This was despite the fact that, under Lomé III, the Côte d'Ivoire had received 365,3 Mio ECU in Stabex transfers.

7.37. At the same time, in 1992, the indebtedness of the Treasury towards the central bank (BCEAO) (60) had risen to 160 000 Mio FCFA (459 Mio ECU), even though the authorities of the franc zone had set a ceiling of 120 600 Mio FCFA. A proportion of this debt had been incurred after a Treasury decision to buy back stocks of low-quality coffee and cocoa from the coffee/cocoa seasons 1985/86 to 1988/89 (61) which had originally been purchased by exporters who had then been unable to sell them. The exporters thus found themselves unable to repay their banks, who in turn were unable to repay the BCEAO which had refinanced the loans for the season. In June 1990, so as to restore a degree of liquidity to the sector and thereby complete the 1990/91 harvest, the Treasury purchased the stocks at the same price that the exporters had paid for them rather than at their disposal value. Following a series of agreements reached between June 1990 and December 1991 between the State, the BCEAO, the local banks and exporters, a solution to these problems had been found in principle without the need for Stabex support, thus allowing the various parties concerned to take part in the 1990/91 and 1991/92 harvests.

7.38. In spite of these agreements to eliminate overhangs from the past, 12 620 Mio FCFA (37,8 Mio ECU), or approximately 40 % of the funds available under the 1990 FMO, an amount which represented 7,8 % of the total of Stabex transfers made to all ACP States for the 1990 application year, was paid over on 26 November 1992, to reduce the Treasury's debt to the BCEAO. The reason given to justify this utilization was that the operations linked to the coffee and cocoa sectors described above had to some extent contributed to this indebtedness. The allocation in 1992 of a significant proportion of the 1990 Stabex transfers to clearing the deficit of the Treasury with the central bank was not, contrary to what the Commission says, directly related to the functioning of the sector. On the contrary, it constituted a major macro-economic support for the State's financial dealings and a practical way of correcting the consequences of risky past political and financial decisions. It therefore constituted a dubious use of Stabex when compared with its intended role.

7.39. In addition, approximately 5,7 Mio ECU was allocated to paying off the CSSPPA's debts towards four banks (62) and a further 2,9 Mio ECU to two companies and a local insurance firm.

7.40. A further 12 000 Mio FCFA (34,4 Mio ECU) was allocated to the cotton and rubber sectors to reduce the debts of the CSSPPA and of the State towards these sectors which mainly resulted from a failure to pay export subsidies. Thus the CIDT (63) received 9 000 Mio FCFA and a further 3 000 Mio FCFA was paid to three companies involved in rubber production. These payments in fact constituted the main single utilization of Lomé IV Stabex funds for diversification purposes. The use of Stabex funds in this field is problematic in so far as the system of export subsidies which was largely responsible for the debt would not have been necessary had the CFA franc not been over-valued. Nevertheless, the payments made allowed the sectors to survive until a devaluation did eventually take place in January 1994.

7.41. The 1991 FMO for the Côte d'Ivoire provided for further large operations to pay off internal debt. In total, such operations amounted to 11 900 Mio FCFA (35,7 Mio ECU out of total transfers of 71,8 Mio ECU) and consisted mainly of CSSPPA debts concerning VAT and marine insurance (4 200 Mio FCFA), the repayment of export subsidies and other debts to the CIDT (4 000 Mio FCFA) as well as the repayment of export subsidies to the rubber sector (3 000 Mio FCFA). Once again, therefore, funds were used to cover commitments from the past which had not been met by the State.

Price stabilization

7.42. In several countries, Stabex transfers were used to stabilize or support domestic producer prices. Given that stabilization is one of the objectives of Stabex, this might be considered a natural use of Stabex transfers. However, the poor management of stabilization funds in certain countries made this a particularly difficult area when it came to the allocation of funds under Lomé IV.

7.43. The country to make the greatest use of Stabex transfers for price stabilization/support under Lomé IV has been Papua New Guinea (PNG), all its transfers being allocated to this purpose under the 1990 FMO (24,9 Mio ECU) and the 1991 FMO (17,2 Mio ECU). This represented a continuation of the PNG policy under Lomé III. The Commission was ready to support such a utilization on the basis that stabilization funds in PNG had been relatively well managed in the past.

7.44. Thus, all the 1990 transfer had been disbursed by the end of 1992. Apart from representing a means of ensuring rapid disbursement, the mechanism also succeeded in transferring funds to smallholders (64) and plantations. The Government of PNG also provided a significant contribution. As a result, production of the country's main export crops (coffee, cocoa, copra and palm oil) all increased over the 1991-93 period. Basically viable but heavily indebted estates were able to continue in business, thus helping to stem the rural-urban drift from which the country suffers.

7.45. The utilization of Stabex resources in this way was not without its problems. While it was originally intended that prices would be reduced to world market levels over a three to five year period from 1989, a change of government in mid-1992 led to support prices being frozen at their subsidized levels. Although the price stabilization policy was intended to give the agriculture sector time to take steps to become more competitive, only limited progress was made in this area, the government, in fact, transferring funds from agricultural investment to price support for reasons of social stability. The sustainability of the scheme appeared questionable, especially as Stabex transfers are expected to decline in subsequent years.

7.46. While in two other Pacific States, Kiribati and Vanuatu, Stabex transfers were also used for price stabilization, in Africa funds were only used for this purpose in a limited number of cases, the main countries being Cameroon, Côte d'Ivoire and Burundi.

7.47. The 1990 FMO for Cameroon provided for a total of 3 900 Mio FCFA (11,2 Mio ECU) to be made available for stabilization purposes. The great majority of these funds (3 400 Mio FCFA) were provided to support robusta coffee prices. The need for the funds arose because the Government, having already set the 1990-91 prices at too high a level in relation to world prices, maintained the same price levels for the 1991-92 season. This was despite the continued fall in world prices and government undertakings, made as part of the overall restructuring of the coffee and cocoa sectors, to reduce domestic prices to reflect world prices. Moreover, the loss of an estimated 7 000 Mio FCFA (20,1 Mio ECU) from the stabilization fund in 1991, because of unwarranted withdrawals by the Treasury, had reduced the reserves available to the fund and made the use of Stabex funds unavoidable. However, as was the case for repayment of internal debt, the Commission refused to provide any further support to the stabilization fund under the 1991 FMO.

7.48. In the Côte d'Ivoire, while no funds had been provided for stabilization purposes under the 1990 FMO, the 1991 FMO allocated 3 500 Mio FCFA for this purpose. This was despite the fact that one of the conditions of the 1990 FMO had been for the Côte d'Ivoire to release the resources necessary to cover any possible deficit in the coffee and cocoa sectors as from the 1991/92 harvest. In order to preserve its credibility in the eyes of its ACP partners, the Commission ought to desist from such inconsistent actions and should refuse to compromise on questions of principle of this kind.

7.49. In Burundi, 2,5 Mio ECU in hard currency was provided for price stabilization purposes. This amount of 2,5 Mio ECU, which was provided for in the FMO and supplied in hard currency - whereas all the Fund's payments were made in local currency - could not be justified during the Court's on-the-spot audit in September 1993. In October 1993 (see paragraph 7.26 (c)), the equivalent of a further 6,9 Mio ECU in counterpart funds generated by the GIP had also been used for price stabilization. The Commission thus accepted support being provided for the stabilization fund without it having clear information on how the fund operated. In actual fact, the support helped maintain a pricing system which the wider restructuring of the sector sought to reform. Approximately 2,7 Mio ECU of the funds provided in local currency for the stabilization Fund were thus used by the Burundi authority responsible for the stabilization Fund for other interventions in the coffee sector without this being agreed in advance with the Commission.

Projects

7.50. A number of countries used Stabex transfers to finance projects designed to assist either the sector where losses had occurred or other agricultural sectors.

7.51. Certain smaller countries thus allocated the bulk of their Stabex resources to development projects. In such cases, although the amount of the transfer was small compared to other Stabex transfers, in relation to these countries' national indicative programmes it was relatively large (e.g. Grenada 45 %; Vanuatu 24 %; Solomon Islands 24 %; Western Samoa 24 %). While the amounts concerned provided the opportunity for measures which could have a clear impact on the sectors in question, their implementation placed a considerable workload on the small Commission Delegations in these countries. At the same time, in most cases, disbursements tended to be slow.

7.52. Thus, for example, the 1990 FMO provided for the financing of no fewer than 11 projects in the Solomon Islands. On 31 December 1993, 17 months after the signing of the FMO, disbursements amounted to only 29 % of the transfer amount while the main project to be financed, a smallholder development programme, was not scheduled to be completed before 1998.

7.53. Several larger countries (e.g. Uganda and Burundi) also included agricultural development projects in their 1990 FMOs. Implementation rates in these cases were also generally slow and placed an additional burden on delegations.

8. IMPLEMENTATION OF ACP STATES' COMMITMENTS

The principle of conditionality in relation to Stabex

8.1. Although financial assistance to the economies of ACP States suffering from losses of export earnings is an essential part of the role played by Stabex, encouraging the ACP States to reform their economies, particularly those sectors relating to Stabex products, is just as important. Stabex is able to do this thanks to a Lomé IV innovation, consisting of making the utilization of Stabex transfers conditional on the ACP States fulfilling certain commitments laid down in each FMO.

8.2. The subordination of the utilization of Stabex funds to a series of conditions under Lomé IV reflects an increasing recognition by donors that a number of aid projects and programmes have failed in the past because of the excessively large imbalances, both in certain sectors of activity and at macroeconomic level, in the recipient countries. Thus it was considered that unless ACP States implemented reforms, particularly in sectors suffering export losses, the benefits that these sectors stood to gain from Stabex transfers risked being greatly diminished.

8.3. This approach is conceptually sensitive, to the extent that the Commission cannot place conditionality over the eligibility of an ACP State for a Stabex transfer. Once it has been established by the terms of the Convention that an ACP State has suffered a loss of export earnings which entitles it to a transfer, this right cannot be denied and the Commission can only introduce conditions concerning the utilization of the resources in question. This chapter reviews how it was possible to subordinate the releasing of funds to the fulfilment of certain conditions in practice.

Types of conditions

8.4. Generally, the conditions attached to the utilization of Stabex transfers operated at three levels:

(a) general commitments;

(b) specific commitments;

(c) financial and administrative provisions.

These conditions correspond to those found in the structural adjustment support programmes.

8.5. General commitments refer to the ACP States' continued adherence to all the structural and sectoral adjustment measures agreed with the donors. Specific commitments normally involve particular undertakings agreed upon by the ACP States in the context of the FMOs. Financial and administrative provisions concern requirements relating to the implementation of individual Stabex operations, such as the submission of the necessary documentation on import programmes before further funds are released.

8.6. Of particular importance in both the 1990 and 1991 FMOs were general and specific commitments by ACP States to reform their coffee and cocoa sectors. In many ACP States, the mismanagement of these sectors by government through marketing boards or stabilization Funds had contributed significantly to the crisis in which these sectors found themselves. Thus, commitments most often included measures aiming at:

(a) the liberalization of internal and external marketing;

(b) reducing the role of the State in managing the sectors;

(c) establishing a financial equilibrium in the sectors by adjusting domestic prices to world prices;

(d) ensuring producers obtained a fairer share of the export price;

(e) increasing the influence of the private sector, in particular producers, over decisions affecting the sector concerned;

(f) improving the quality of crops, notably by price incentives and better quality control.

8.7. When deciding upon what conditions to attach to the release of Stabex funds, the Commission had to find the right balance so that the conditions chosen would be of sufficient weight to lead to significant reforms but would not be so demanding as to prevent the ACP States from being able to implement them and thus block Stabex disbursements.

8.8. The scope of the conditions laid down in the FMO and the evidence required of their fulfilment do however vary considerably from country to country. Thus, in the Côte d'Ivoire and Cameroon, detailed conditions were laid down for each component to be financed and specific documents were required as proof that the measures had been taken. In other countries, for example Ethiopia, Ghana and Papua New Guinea, the conditions were less wide-ranging and the proof required of their implementation not stated. In general, the existence of structural adjustment programmes in a country facilitated the setting of conditions, since conditions relating to the sector concerned could be taken over from the adjustment programme. This was particularly the case, for example, in the Côte d'Ivoire and Uganda.

Respect of conditionality

Macroeconomic Conditions

8.9. In some countries (65), general commitments included the pursuit of macroeconomic policies established in the context of structural adjustment programmes. However, in practice, the extent to which the Commission linked the respect of these policies to the release of funds varied from one country to another.

8.10. Thus, in Rwanda, although the government committed itself in the 1990 FMO to further implement its structural adjustment programme, the Stabex funds continued to be disbursed even after donors suspended funding under the structural adjustment programme because of the government's failure to comply with it adequately.

8.11. In contrast, in Madagascar, where the general and specific commitments made in the FMO were based on the country's further implementation of structural adjustment measures, the Commission took a firmer position. Thus, disbursement of the second tranche of the Stabex SIP was suspended for approximately 12 months pending the drawing-up of a satisfactory budget, involving, in particular, the cutting off of funds for what were considered to be 'white elephant' projects.

8.12. In two other countries Stabex payments have been continued, although the EDF has suspended disbursement of structural adjustment support:

(a) in Cameroon, the 1991 Stabex transfer of 69,2 Mio ECU was released even though the release of the Lomé IV structural adjustment funds had been suspended since the end of 1992 because of the country's failure to respect the structural adjustment programme (66);

(b) in Kenya, Stabex transfers for the 1990 and 1991 application years amounting to 46,3 Mio ECU were released despite the fact that most of the donors, including the Community itself, had blocked the release of quick-disbursement funds because of the political situation in the country (67).

8.13. The fact that the carrying out of reforms was only included among the FMO commitments in a limited number of countries undergoing structural adjustment constitutes a discriminatory treatment of certain ACP countries. It also points to the need for the Commission to establish a clear policy with regard to the relationship between Stabex and structural adjustment support in the case of countries which benefit from both instruments. In particular, it should decide whether failure to comply with structural adjustment programmes should lead to the suspension of Stabex transfers. In so far as Stabex can be considered a quick-disbursement instrument with similar functions to the balance of payments support associated with structural adjustment aid, the continuation of Stabex disbursements in cases where structural adjustment support has been suspended may be considered to be an inconsistent policy and undermine donor coordination.

Sectoral Level Reforms

8.14. It is at the level of sectoral reform that compliance with the general and specific conditions laid down in the FMOs has most often proved problematic. This situation can be explained by a number of factors, notably the fact that the reforms envisaged are frequently on a considerable scale, affecting key sectors of the economy.

8.15. In these circumstances, when the Commission has had to decide whether conditions have been adequately respected, it has been faced by two contradictory objectives:

(a) seeking to maintain Stabex's traditional role as a quick-disbursement instrument;

(b) trying to ensure conditions are met, which may imply the blocking of disbursements for a period.

In this case, it may be particularly difficult to take a decision where, in overall terms, an ACP State has made progress with the restructuring of the sector concerned but specific conditions forming part of the reform are not adequately fulfilled.

8.16. In the Côte d'Ivoire, the restructuring of the coffee and cocoa sectors, carried out in the wider framework of the country's Agricultural Sector Adjustment Programme, led notably to the effective liberalization of internal marketing, one of the general commitments of the 1990 FMO. But as far as the liberalization of external marketing was concerned, while some progress had been made, the situation concerning several specific commitments had remained unclear. In particular:

(a) serious doubts existed over whether the 'programme de ventes anticipées à la moyenne' (PVAM) had been fully introduced by the CSSPPA (68);

(b) the non-implementation of an electronic system designed to make the CSSPA's awarding of contracts to exporters more transparent;

(c) uncertainty over whether the CSSPPA had reduced its direct sales to 10 % of total exports as required.

8.17. Despite doubts on the part of the Commission over compliance with conditions in these areas, the Stabex funds were released after a period of blockage. However, the problematic nature of their release is best demonstrated by the fact that the implementation of the PVAM and the electronic mail system were once more included as conditions in the 1991 FMO. Reluctance on the part of the CSSPPA to communicate information on the state of the reforms, due both to the commercially sensitive nature of such information and the technical nature of these reforms, made the Commission's task a particularly difficult one. The use of technical assistance provided by the APROMA (69) consultancy to monitor the implementation of reforms performed a valuable role in aiding the Commission in this task.

8.18. In Tanzania, significant progress was also made in the restructuring of the coffee sector. However, the bill to liberalize the coffee sector was not passed until August 1993, after virtually all the hard currency component of the FMO had been disbursed, despite the fact that the liberalization of the sector was a general commitment of the government. As far as the specific commitments of the FMO were concerned, whereas the organization of the Export Division of the Tanzanian Coffee Marketing Board had actually been rationalized, significant problems still remained with export procedures, notably in relation to the limited number of buyers with access to auctions.

8.19. In Cameroon, where the FMO also aimed at the liberalization of the coffee and cocoa sectors, one of the major reforms was the abolition of the ONCPB (70), which was to be replaced by a much smaller organization with no part at all to play in the marketing and exporting of cocoa and coffee, its role being limited to monitoring and checking the quality of the products. The replacement of the ONPB by the ONCC (71) was one of the general conditions of the 1990 FMO while the effective functioning of the ONCC and the satisfactory further implementation of the liquidation of the defunct ONCPB was a condition for the release of the second tranche of the Stabex transfer.

8.20. Although the ONCC was established, there were considerable misgivings on the part of donors concerning its functioning, particularly in relation to the size of its budget, where salary costs, travel expenses and expenditure on the furnishing and rehabilitation of accommodation for its management appeared excessive. However, the Commission did not consider the situation to be so serious as to justify blocking or delaying Stabex disbursements.

8.21. The liquidation of the ONCPB was marked by very slow progress, partly as a result of the government's reluctance to relinquish certain assets as the donor community was financing its liabilities. After the dismissal of the liquidator by the Cameroon authorities and the failure to replace him within a satisfactory period of time, the Commission blocked further utilization of the 1990 FMO. However, the Stabex system being such that the Commission cannot actually withdraw Stabex transfers once an ACP State's entitlement to them has been established, the balances were transferred to the 1991 FMO, their release being conditional once more on the satisfactory liquidation of the ONCPB.

8.22. In Papua New Guinea, the conditions of the FMO were not adequately respected. They provided for the establishment of industry corporations to take over the running of the cocoa, copra and oil palm sectors along the same lines as the new Coffee Industry Corporation. However, as a result of no industry corporation being set up for the cocoa and copra sectors, and the new palm oil industry corporation only being invested with some of the intended functions, the same conditions were repeated in the 1991 FMO. Although a further condition of the 1990 FMO was that the Government would take steps to improve extension, training and research in the three sectors in question, no framework was established for evaluating progress. Thus, this condition was also repeated in the 1991 FMO without any assessment being made of progress to date. In fact, progress made was only limited. The problems encountered can be partly explained by the failure to release funds in tranches subject to the implementation of the conditions of the FMO. They were also caused by the lack of a formal structural adjustment programme from which conditions could be taken.

8.23. In Kenya, the Commission found itself faced by the Government with a clear abandonment of the latter's undertakings concerning the liberalization of the maize sector (72). In this case, the Commission took the view that the suspension of Stabex transfers would not bring the Government to change its position and, in order to maintain Stabex's quick disbursement function, funds should be allocated instead to the coffee sector. However, the amended FMO did not impose any sectoral policy reform conditions on the release of the 20 Mio ECU in hard currency (73).

Financial and administrative conditions

8.24. Each FMO contained a fixed clause to the effect that Stabex funds were only to be allocated by the ACP State for purposes prescribed in the FMO. If this condition was not complied with, the implementation of the FMO was to be halted and only recommenced when the sums in question had been transferred in ecus to the European bank account. This fundamental condition governing the utilization of Stabex funds was satisfactorily respected by the ACP States.

8.25. FMOs which provided for the utilization of Stabex funds for import programmes also contained a series of conditions concerning the implementation of these programmes. In particular, these were designed to ensure competition between suppliers and that supplies originated in the Community or the ACP States.

8.26. Conditions governing the implementation procedures of import programmes generally provided that open procedure tendering would be applied to all applications in excess of 1 Mio ECU and restricted invitations to tender would be issued for all applications exceeding 100 000 ECU. However, as these conditions generally did not correspond to commercial practice in the countries concerned, importers preferred to continue to deal with their traditional suppliers. As a result these conditions were often either not satisfactorily complied with (Kenya, 7.22), or suppliers split up purchases to avoid the constraints (e.g. Mauritania, 7.23) or considerable time was taken up ensuring that documentation was gathered which complied with the competition conditions (e.g. Burundi, 7.12). Though to a lesser extent, the provision of documentation concerning rules of origin also posed problems (e.g. Madagascar 7.24, Burundi 7.11) (74).

9. FINANCIAL AND ACCOUNTING PROCEDURES

Legal framework for implementation of Stabex operation

9.1. The implementation of Stabex under Lomé IV has created a new approach to the management of EDF funds without this being adequately laid down in the relevant regulations and internal procedures. The main ambiguity is the fact that Stabex transfers are considered to belong to the ACP State once the transfer right has been established, although their actual utilization not only has still to be approved by the Commission, but the procedures for their disbursement are also required to be in conformity with EDF rules and the financial provisions of FMOs.

9.2. The fact that certain EDF procedures have been retained in relation to Stabex funds in spite of the provisions of Article 248 of the Convention and even though the Stabex transfers belong to the ACP State, has had the effect of slowing down the disbursement of Stabex funds. This is particularly the case for the supply of goods, both in cases of direct importation of goods under EDF procurement procedures, where the requirement for open international tenders to be held caused serious delays to the implementation of FMOs in Ethiopia, Kenya and Rwanda, and in cases where foreign exchange was made available in order to enable the government (central banks) to meet firms' import demands. This incongruous situation, where Stabex transfers belong to the ACP State but their disbursement is subject to certain EDF procedures, needs to be addressed.

9.3. A revision of the provisions of the Lomé Convention should be considered to make eligibility for Stabex transfers dependent not only on an ACP State having suffered a certain loss in export earnings but also on the satisfactory utilization of the funds. At the same time, however, ways of simplifying the EDF procedures used to implement Stabex should be considered.

Functioning of the double signature accounts

9.4. The most significant new financial procedure for Stabex under Lomé IV is the requirement that transfers are to be paid into accounts which are held by the ACP States but from which disbursements can be made only gradually, as operations are implemented, and subject to the authorization of the Commission's delegate as well as the National Authorizing Officer. This system thus provides a form of decentralised, second-level control over the management of EDF funds, the normal legal requirement for the disbursements to be approved by the financial services of the Commission (see 5.2(a) being bypassed. This, in principle, allows an accelerated form of disbursement, although in many cases disbursement is subject to acceptance by the Commission that the conditions of the FMO have been respected (75).

9.5. Compared with the previous Conventions, the establishment of double signature accounts has undoubtedly contributed to the Commission's much improved control over Stabex utilization under Lomé IV. Nevertheless, on-the-spot audits by the Court revealed certain difficulties in the functioning of the system. Thus, in Papua New Guinea (PNG) it was found that 120 000 ECU in interest on the 1990 Stabex transfers had been transferred out of PNG's ECU account in Europe without the authorization of the Delegation and had not been accounted for in the local Stabex account. Similarly, interest on the 1991 Stabex transfers, amounting to 762 000 ECU, had been accounted for in the Central Bank's profit and loss account instead of the Stabex account. In Burundi it was found that the Delegation had agreed to a system allowing the release of funds from the ECU account without the Delegation's signature.

9.6. It was also noted in certain cases (e.g. Burundi, Cameroon, Papua New Guinea) that Delegations were experiencing difficulties in obtaining regular bank statements for the double-signature ECU accounts. As they represent an important element in allowing Delegations to ensure that the double-signature system is functioning satisfactorily, steps should be taken to ensure that Delegations do indeed receive bank statements from banks in Europe.

Counterpart funds

9.7. Delegations are also responsible with National Authorizing Officers for implementing certain procedures concerning counterpart funds arising from Stabex-financed import programmes. In general, the actual generation of counterpart funds took place satisfactorily, the main exception noted being in Rwanda where, at 31 October 1993, counterpart funds deposited amounted to approximately 0,25 Mio ECU (43 Mio Rwandan francs), whereas it is estimated that a sum of the order of 1,95 Mio ECU (333 Mio Rwandan francs) should have been on deposit at that date.

9.8. Although counterpart fund accounts were required to be interest-bearing, this was often not the case (Burundi, Ethiopia and Rwanda) or only partially the case (Ghana, Kenya, Tanzania (76)). The result was a loss of value on the counterpart funds because of the high inflation rates in the countries concerned.

9.9. In addition, while Stabex counterpart funds were required to be held in separate accounts from other counterpart funds so as to subject them to the system of conditional mobilization provided for in Article 211-3 of the fourth Lomé Convention, in three countries this was not the case (Ghana, Madagascar, Tanzania). The Commission should consider whether keeping Stabex funds on separate accounts is in fact desirable, when its general policy with regard to counterpart funds is to consolidate them on one account.

Financial monitoring of Stabex by the Commission

9.10. The EDF accounting system only shows the implementation of Stabex transfers as far as payment into the ACP bank account in Europe. However, as shown in Chapter 7, the Commission's Delegations then have a considerable task in implementing and monitoring the use of Stabex funds in the ACP State. Despite this, no formal system exists in the Delegations to account for the utilization of Stabex funds. The Commission should ensure both that Delegations are able to account adequately for the utilization of Stabex funds and that the central services establish a system for consolidating the financial implementation of the FMOs.

9.11. In order to enable the discharge authorities to be in a better position to assess the implementation of Stabex, the financial implementation of Stabex in the individual ACP States should be reported in the EDF's annual financial statements (77). This could be produced as supplementary information to the financial statements in the same way as the Commission includes information on counterpart funds. Such information should also be included in the Commission's annual report to the Member States on the operation of Stabex (78). These reports, which set out the intentions stated in the FMOs, are ambiguous as to the actual implementation of the measures provided for.

10. MONITORING AND EVALUATION OF LOME IV STABEX

10.1. The new provisions for Stabex under Lomé IV were aimed principally at enabling a more detailed control on how Stabex resources were actually utilized. An important aspect of this ought to have been the establishment of a monitoring and evaluation framework (7.1) to ensure that FMOs were implemented as foreseen and provided the evidence to assess whether their objectives had been achieved.

10.2. The new arrangements gave Commission Delegations responsibility for monitoring Stabex operations. Some Delegations took on technical assistance, generally paid for out of Stabex funds. This was particularly the case for import programmes, where technical assistants monitored programmes financed out of structural adjustment funds. Given the very limited resources of Delegations and the need for specialized expertise in monitoring the technical details of many aspects of the reforms supported by Stabex transfers, it would seem advisable for Delegations to make wider use of technical assistance and consider the possibility of taking on local technical assistance wherever possible.

10.3. The FMOs frequently provided for the establishment of steering committees, made up of both representatives of the ACP State and the Commission Delegation, to oversee the implementation of the FMO. However, in a number of cases such committees failed to fulfil the role assigned to them (e.g. Burundi, Madagascar, Papua New Guinea, Uganda). In order to improve the contribution made by such committees, their satisfactory functioning should be included as one of the conditions in the FMO.

10.4. The Commission's central services play only a limited role in the day-to-day monitoring of Stabex. It is generally left to Delegations to contact the Commission on an ad hoc basis as particular matters felt to warrant the attention of the central services arise. While this approach is overall considered appropriate, it would nevertheless be desirable for Delegations to formally report to the central services to ensure that the latter have a clear overview of Stabex operations in each particular country. Such a reporting system, operating on an annual basis, already exists for National Indicative Programme projects and could be adapted for Stabex operations carried out in ACP States.

10.5. As under previous Conventions, the Lomé IV Convention requires the ACP State to send a report to the Commission on the use it has made of the funds transferred (79). This is to be done 12 months after the mobilization of resources by the ACP State. It would seem this deadline was chosen on the assumption that operations to be financed would be substantially completed a year after funds were first transferred, but in the majority of countries this is not the case. This may partly explain why some countries had not submitted reports within the timeframe foreseen. Of those that were submitted there was a tendency for reports to be unduly concise, limiting themselves to simply listing or summarizing payments made (e.g. Ivory Coast, Papua New Guinea). The establishment of a standard framework for such reports would be desirable, in order to improve both the presentation and the contents and thus facilitate reporting.

10.6. In order to assess the impact of Stabex in individual countries, each FMO systematically provides for an evaluation of the operations it finances. This potentially represents an important step in improving the effectiveness of Stabex although, at the end of 1993, the slow implementation rate of most FMOs meant that virtually no evaluations had actually been carried out yet. In addition, the Commission also planned an evaluation of the Stabex instrument as a whole in the course of 1994. This is particularly desirable as no global evaluation of Stabex has taken place since 1981 and, moreover, the evaluation of the inadequacy of the system's resources has not produced any definitive results (3.18).

11. CONCLUSION

General context of the implementation of the Stabex system

11.1. Because of the continuing rise in world output of the two main commodities supported by the Stabex system, namely coffee and cocoa, the prices of these commodities had fallen to, respectively, 38 % and 32 % of the 1980 level (3.8) when the fourth Lomé Convention entered into force in 1991. Against this background, the international agreements concluded between producer and consumer countries of these products with a view to stabilizing their prices collapsed. The Stabex resources were thus mobilized in the midst of a generalized crisis and against a background of large-scale shortages of public funds.

11.2. At the same time, most of the ACP countries, at the behest of the International Monetary Fund and the World Bank, had made painful efforts to restore their internal finances (budget), as well as the external ones (balance of payments), to balance, which had obliged them to carry out structural reforms that had had grave socio-political consequences (6.4).

11.3. In the countries of the franc area, the overvaluation of the CFA franc (7.28) increased the cost of factors of production and cancelled out a large part of their efforts to improve productivity. The existence of excessively rigid systems of guaranteed prices for producers created unbearable financial burdens for these countries.

11.4. In this especially difficult general environment, the Community, through the mechanism of Lomé IV, offered a limited number of countries a system of compensation for their losses on agricultural exports from which other countries which had concluded cooperation and development agreements with the Community were excluded. In this way, the Stabex system takes on the appearance of a selective support instrument which, unlike the mechanisms of the international agreements on basic commodities, makes little contribution to the improvement of the overall situation of producer countries, because it does not tackle the structural questions of world overproduction of these commodities on a global basis.

Economics of the system laid down in the new Convention

11.5. Until the third Lomé Convention, the Stabex system was a mainly short-term economic instrument, intended to smooth out irregularities in the cycles of production and marketing of basic agricultural products. It included a mechanism for the reconstitution of resources during boom times which gave it, in principle, a measure of permanence (3.9).

11.6. The fourth Lomé Convention abandoned the notion of reconstitution in favour of a mechanism of non-repayable subsidies combined, for the Community, with a right to oversee the use of funds by means of frameworks of mutual obligations (2.8 and 6.12).

11.7. In spite of the chronic inadequacy of the resources (3.7), the new Convention has retained a system within which the beneficiaries consider they are entitled to compensation in full for their export losses, whence the difficulties encountered each year in attempting to increase the amount of resources to be allocated to the system (3.13 to 3.18).

11.8. The mechanism for calculating transfer entitlements is moderated by a correction procedure which leaves too much room for negotiation and the effect of relative strength, with the result that some countries derive excess benefits from it (4.4 to 4.6). On the other hand, the Convention does not make up revenue loss caused by falls in the quality of products and which ought to continue to be the responsibility of the ACP States (4.8). Similarly, an exceptional and expensive status is granted to some States for mainly historical reasons (all destinations status) even though the Convention had made other provisions in this respect.

11.9. The changes made by the new Convention have introduced contradictions between some of its objectives. Whereas the fundamental aim is to remedy export losses, the frameworks of mutual obligations are also attempting to deal with the causes of these losses (6.1, 6.15, 7.8 and 7.28), which really comes more under long-term measures than under national indicative programmes (6.6). Although the Convention does retain, in principle, the sectoral approach (6.2), the overall economic context cannot be ignored without the risk of inconsistencies with the structural adjustment policies applied elsewhere (6.4 and 6.5). Finally, the introduction of frameworks of mutual obligations brings with it long drafting (6.13 and 6.14) and implementation periods (7.3 and 7.4), which deprive the Stabex mecahnism of its characteristics as a short-term, rapid-disbursement instrument (5.5 and 6.24).

11.10. The Convention did not provide for the allocation of considerable amounts of Stabex funds to general programmes or sectoral import programmes (7.5). As a result, it did not provide a framework of management for the considerable quantities of counterpart funds that these programmes generate (7.3(c) and annex 6). There was therefore nothing to stop the counterpart funds being allocated several times over to operations that were alien to the objectives of the Stabex system (7.25 and 7.27).

Financial management

11.11. The financial management of the system continues to lack transparency in spite of the mechanisms for the presentation of the accounts laid down in the internal financial agreement (9.11). The entering in the accounts of financial interest produced by the system's resources is deficient (3.11), whilst the mechanisms for boosting allocations are in some cases questionable (3.14 and 3.15). Thus, in the EDF accounts, at the end of December 1994, the last annual tranche of Stabex funds (application year: 1994) was no more than 109,8 Mio ECU instead of the 300 Mio ECU specified in the Convention.

11.12. The payments entered in the accounts by the EDF overstate the actual use that is made of the transfers by treating the payments made into intermediate bank accounts opened in the name of the beneficiary ACP States as final payments (3.4). Moreover, the Commission has no consolidated information available as to the use that is actually made of the transfers (7.1) or of the counterpart funds (7.3).

11.13. The implementation of the frameworks of mutual obligations has given rise to a second level of management of EDF appropriations which is not subject to the control procedures prescribed by the Financial Regulation, whereas comparable operations (e.g. import programmes) financed through other instruments are subject to the Financial Regulation (5.2(a) and 9.4).

11.14. The frameworks of mutual obligations have burdened the Commission Delegations with a considerable amount of preparatory and monitoring work for the support and reform programmes although their staff numbers have not been reassessed in the light of this extra work (5.2., 6.17 and 7.53), nor do they always have the appropriate technical assistance (8.17 and 10.2).

Implementation of the Stabex transfers

11.15. As under Lomé III, the division of resources amoungst the beneficiaries continues to be unbalanced, as three countries (see Table 3.4) received 46,2 % of the transfers for the years 1990-1992 (3.19 and 3.21).

11.16. The decentralized mechanism for the preparation and the implementation of the frameworks of mutual obligation has experienced delays caused by the intervention of the central departments during the preparatory stages (6.15, 6.22 and 6.23) and has been affected by the contrasting positions adopted, which varied according to the country or the financial instrument, when questions such as the choice of operations to be financed (7.8, 7.13, 7.31, 7.38, 7.42 and 7.43) or observance of the conditions governing the release of funds (8.9 to 8.23) were concerned.

11.17. Stabex funds have too often been orientated towards operations intended to clear up past errors rather than towards the making of investments in new sectors, instead of in sectors with no real prospects (6.2, 6.5 and 7.5). These debatable allocations of funds have been facilitated by the presence of counterpart funds accompanying the import programmes. These transfers have thus served to make up for the poor management by public bodies (7.26, 7.27(d), 7.29 to 7.33, 7.39 and 7.41), the debatable choice of monetary policy (7.28, 7.37 and 7.38) or faults of viability in the price support systems (7.42 to 7.49), most usually to the detriment of producers (2.8(a) and 6.2).

11.18. Not much has been done under Stabex to remedy the institutional weaknesses in the beneficiary countries whether concerning statistical cooperation (4.3), or substantial analyses of the sectors affected by losses of revenue (6.10 and 6.16).

11.19. The delays in implementing the frameworks of mutual obligations have seriously slowed down the evaluations that were provided in each FMO for the purpose of assessing the impact of the financing and drawing conclusions from them for future operations (10.6).

Overall assessment

11.20. In spite of the questions that one may reasonably ask regarding the capacity of the Stabex system to make up for losses of income which are caused by world-wide factors and which are strongly influenced by the production of countries that are not part of the system (e.g. Brazil), the new provisions of the Lomé Convention have nevertheless made it possible for the EDF to progress in the direction of greater control of the funds that are mobilized.

11.21. Three main questions remain to be settled as a matter of priority: the inadequacy of the resources (3.5), the objectives and their complementarity with those of other EDF financial instruments (6.1) and, finally, the question of the introduction of a genuine mechanism for the monitoring of the frameworks of mutual obligations such as may give the discharge authority a precise and faithful overview of how they have been implemented (7.1) (80).

This report was adopted by the Court of Auditors at the Court meeting of 26 April 1995.

For the Court of Auditors

André J. MIDDDELHOEK

President

(1) See Article 16 of the Lomé I Convention; Article 23 of the Lomé II Convention; Article 147 of the Lomé III Convention; Article 186 of the Lomé IV Convention.

(2) Implementation years 1990, 1991 and 1992.

(3) Burundi, Côte d'Ivoire, Papua New Guinea, the Solomon Islands and Tanzania.

(4) Article 193 of the fourth Lomé Convention.

(5) Article 187 of the Lomé IV Convention. The non-agricultural products are:

(a) wood in the rough and squared wood;

(b) sawn wood;

(c) shrimps and prawns;

(d) squid, octopus and cuttlefish.

(6) The percentage is in principle 5 %, but only 1 % for least-developed, landlocked and island ACP States (see Article 196 (1) and (2) of Lomé IV Convention).

(7) For Lomé IV the reference level is based on average export earnings over the six calendar years prior to the application year less the two years with the highest and lowest figures (see Article 197 (2)).

(8) Article 147.2 of Lomé III Convention; Article 23.2 of Lomé II Convention.

(9) Primary commitments correspond to the decision taken by the Commission as to the transfer amount a country should receive for a loss of earnings relating to a particular product.

(10) Secondary commitments are made when a transfer agreement has been signed by the Commission and the ACP State.

(11) The countries concerned were: Sudan 63,9 Mio ECU; Haiti 17,0 Mio ECU; Malawi: 4,2 Mio ECU.

(12) Article 5 of the fourth Lomé Convention stresses the importance of respect for human rights, whilst Article 130U (2) of the Treaty of Maastricht formally establishes a link between development cooperation and respect for human rights.

(13) Article 197.3 requires that the difference between the reference level and actual earnings should automatically be reduced by 4,5 % of the reference level (1 % in the case of least-developed ACP States) to arrive at the transfer basis.

Under Article 203, reductions may be made to take account of significant changes in trends in marketed production and exports (see 4.4. ff.).

Article 194-2 prescribes an automatic 10 % cut in transfer bases in the event of a shortage of available Stabex resources.

(14) The ACP-EEC Council of Ministers delegated this matter to the ACP-EEC Committee of Ambassadors under Article 345 of the Lomé IV Convention.

(15) 45,6 Mio ECU was transferred from the Lomé II Sysmin allocation and 52,3 Mio ECU from the Lomé III Sysmin allocation.

(16) See the Court of Auditors' 1991 Annual Report, paragraph 17.17. (O.J. 92/C 330/01).

(17) See the Court of Auditors' 1992 Annual Report, paragraph 14.25. (O.J. 93/C 309/01) and similar provisions of the Lomé IV Convention applicable to Sysmin (Article 219.2.a).

(18) In this context it should be noted that under the Lomé II Convention, the Sysmin instrument was established to assist ACP States which rely on mineral rather than agricultural exports. Thus, as at 31-12-93, eight of the countries which had not received Stabex transfers under Lomé III and Lomé IV have benefited from Sysmin allocations since the creation of the instrument: Botswana (21,7 Mio ECU); the Dominican Republic (23,0 Mio ECU); Guinea (35,0 Mio ECU); Guyana (30,6 Mio ECU); Liberia (3,7 Mio ECU); Namibia (40 Mio ECU); Zambia (82,3 Mio ECU); Zaire (73,3 Mio ECU).

(19) In 1993, the Stabex Division comprises 3 'A' grade staff, 3 'B' grade staff, and 2 'C' grade staff.

(20) Article 199 (2) of the Convention requires the ACP States to notify the Commission of the following annual statistical data:

(a) the value of their exports of all goods to all destinations for the year preceding the application year;

(b) the volume of marketed production of the product or products in question during the reference period and in the year of application;

(c) the volume and value of exports of the product or products in question to all destinations in the reference period and in the year of application;

(d) the volume and value of exports of the product or products in question to the Community during the reference period and in the year of application.

(21) In the case of Mozambique, statistical data submitted for the 1991 application year were found by the Commission services to be insufficiently reliable to allow the assessment of the transfer. A technical assistant was therefore appointed to review the statistical data in Mozambique which allowed transfer decisions amounting to 1,27 Mio ECU to be taken in February 1993. Other studies were financed by the Commission in respect of mohair in Lesotho, wood in the Côte d'Ivoire and all the products for Malawi in 1994.

(22) On the basis of the decision of the ACP-EEC Council of Ministers of 21-6-85, 'significant' is defined as where the percentage for the criterion in question is at least 15 points below the weighted average percentage of the reference period.

(23) See Stabex Beneficiaries' Handbook, Fourth ACP-EEC Convention of Lomé, Chapter 4, Paragraph 2.6. The Lomé IV Convention itself does not state in which circumstances a reduction should be applied in full and in which circumstances only a partial or no reduction should be applied.

(24) See the Commission's reply to paragraph 14.28. of the Court of Auditors' 1992 Annual Report.

(25) An amount of 18,5 Mio ECU had been allocated to Cameroon at 31-12-93 from EDF structural adjustment funds.

(26) Products produced by smallholders are usually of inferior quality to those produced by plantations.

(27) This new level of management of EDF funds should be distinguished from the management of counterpart funds. The Lomé IV Stabex procedures cover the management of EDF funds prior to their utilization whereas counterpart funds arise as a result of the utilization of EDF funds.

(28) Of the 32 countries in receipt of Lomé IV Stabex transfers as at 31-12-93, 21 had opened accounts in Belgium, three in France, three in the U.K. and one in each of Germany, Luxembourg, the Netherlands and Portugal.

(29) The table does not include three countries where transfers have been blocked for political reasons: Haiti, Malawi, and Sudan.

(30) See Stabex Beneficiaries' Manual, ACP-EEC Lomé IV Convention, Chapter 2, Paragraph 11.

(31) The analysis does not include the three countries to whom the payment of Stabex transfers were blocked for political reasons (Haiti, Malawi, Sudan).

(32) See Commission guidance notes to Delegations dated 4-5-90, paragraph 5.4., footnote 1, and 4-2-91, paragraph 43.

(33) In particular, Article 206 concerning the rapid transfer of Stabex funds to ACP States and Annex XLVI which also stresses the need for 'making transfers available to the ACP States as swiftly as possible'.

(34) As at 31 December 1994, primary commitments from Lomé IV National Indicative Programme funds to support structural adjustment amounted to 222,4 Mio ECU.

(35) Principles, Guidelines and application Rules For The Use Of Stabex Transfers Under Lomé IV dated 4-2-91. Paragraph 46.

(36) Circular to Delegations dated 28-6-91 'Establishment of the draft for a framework of mutual obligations'.

(37) Framework of Mutual Obligations 18-5-92, Annex II. The main headings to be covered by the FMO were:

(a) background to the transfer;

(b) objectives and expected results;

(c) consistency with the country's economic policy;

(d) uses;

(e) Government's commitments;

(f) implementing procedures, monitoring and checking.

(38) The median is determined by establishing the number of FMOs to be signed, classifying them in order of the date of signature, and identifying the middle value. Thus for the 1990 FMOs, the median time was the time taken to sign the 15th of the 29 FMOs to be signed, while for the 1991 FMOs it was the time taken to sign the 17th of the 33 FMOs to be signed (the FMOs relating to Haiti, Malawi and Sudan, where payment of Stabex transfers had been blocked for political reasons, were excluded from the calculation). The median rather than the arithmetic mean is used in this context to determine the average because the time that will be required to sign FMOs which have not yet been signed is not known.

(39) Benin, Burkina Faso, Dominica, Ethiopia, Equatorial Guinea, Ghana, Kiribati, Mali, Sierra Leone, Togo, Tonga, Tuvalu, Vanuatu, Western Samoa.

(40) These were Dominica, Kiribati, Mauritania, Uganda, Vanuatu. In the case of Dominica and Vanuatu, their 1990 and 1991 FMOs were drawn up as a single document.

(41) This was the case, for example, with the coffee sector in Burundi and Kenya, the hides and skins sector in Burkina Faso, and the cashew-nut sector in Mozambique.

(42) Within the Commission's central departments, the geographical services have the lead role in coordinating the preparations of the FMO, while the Stabex service checks that the prepared utilization is in conformity with the Convention and the structural adjustment unit ensures that the FMO is coherent with structural adjustment policies in the country concerned.

(43) This concerned the 1990 FMO for the Comoros.

(44) For the 1990 application year, 36,2 % of all Stabex transfers paid went to these two countries.

(45) At the end of 1993, 14,25 Mio ECU had been disbursed from the resources of the NIP whereas the 17 Mio ECU allocated from the structural adjustment instrument were unused.

(46) The first tranche covered imports made in the first half of 1992 while the second tranche covered imports made over the period March 1992 to July 1993.

(47) That is the foreign exchange could be freely bought within Ghana using local currency.

(48) Countries in the franc zone have fully convertible currencies. A number of other African countries have established or are seeking to establish the internal convertibility of their currencies.

(49) In the revision to the Lomé IV Convention, it is proposed that structural adjustment support to countries with convertible currencies should no longer be directed through import programmes.

(50) ERRP.

(51) The initial estimate of 130 Mio tonnes of fertilizer was reduced to 65 Mio tonnes.

(52) I.e., for a single fishing trip.

(53) Equivalent to approximately 3,3 Mio ECU at October 1993 exchange rates.

(54) Taking into account interest accrued, the amount owed as of 31 October 1993 was 1 840 Mio Ush, or approximately 1,36 Mio ECU.

(55) Under Lomé III, the counterpart funds generated by the Sectoral Import Programme and Stabex had been used to support the CSRP. See also Special Report No 2/94 of the Court of Auditors, paragraph 4.48, OJ C 97, 6. 4. 1994.

(56) The CFA franc was finally devalued by 50 % in January 1994.

(57) Office National de Commercialisation des Produits de Base (the National Board for the Marketing of Staple Commodities).

(58) 1990 Stabex transfers to Cameroon amounted to a total of 67,4 Mio ECU.

(59) While four exporters, as a result of these payments, had more than 90 % of their outstanding debts paid off, three other exporters had only between 6 % and 30 % of their debt paid off.

(60) Caisse de Stabilisation et de Soutien des Prix des Productions Agricoles (Fund for the Stabilization and Support of the Prices of Agricultural Produce).

(61) Banque Centrale des Etats d'Afrique de l'Ouest (Central Bank of the West African States).

(62) Coffee: 147 823,1 tonnes for 68 494,2 Mio FCFA (197,8 Mio ECU); Cocoa: 29 175,5 tonnes for 12 295,8 Mio FCFA (35,5 Mio ECU).

(63) This amounted to only 5 % of the debts towards the banking sector, the World Bank paying off the other 95 % in the framework of the 'programme d'ajustement structurel du secteur financier' (financial sector structural adjustment programme).

(64) Compagnie Ivoirienne pour le Développement des Textiles (the Ivory Coast Textile Development Company).

(65) 85 % of the PNG population earns its living from agriculture.

(66) Burundi, Madagascar, Ruanda, Tanzania.

(67) This involved part of the second tranche of the first allocation of the seventh EDF.

(68) Kenya had been allocated an amount of 23,5 Mio ECU from the seventh EDF's structural adjustment facility

(69) Caisse de Soutien et de Stabilisation des Prix des Produits Agricoles (Fund for the Support and Stabilization of the Prices of Agricultural Products).

(70) 'Association des Produits à Marché', an ACP-EEC organization partly financed from EDF regional cooperation funds.

(71) Office National de Commercialisation des Produits de Base (National Office for the Marketing of Staple Commodities).

(72) Office National du Café et du Cacao (the National Coffee and Cocoa Office).

(73) The Government's position was in spite of the large financing provided for the Cereal Sector Reform Programme under Lomé III and Lomé IV (CSRP).

(74) The only condition attached was that the Government would sell fertilizers imported under the emergency import programme at full market prices (see above).

(75) Problems related to the financial provisions of import programmes were also noted by the Court in its 'Special Report No 2/94 on the import programmes carried out under the sixth EDF, paragraph 4.16'. See Official Journal 94/C 97/01.

(76) Normally the geographical services and the structural adjustment unit.

(77) A Commission audit of counterpart funds in Ghana in March 1993 found that no interest was paid on funds held on current accounts (although generally such accounts attracted interest at a rate of approximately 8,5 %), while amounts on the deposit account were earning 10 % interest, when local rates were at least 22 %.

In Kenya, counterpart funds began to be moved onto interest-bearing accounts in May 1993.

In Tanzania a portion of funds were on deposit accounts but the rate of 12 % was less than average rates and only approximately half the inflation rate.

(78) The European Parliament's discharge report of 19 April 1994 on the 1992 Financial Year (document A3-0257/94) requested that the Commission should report to the Parliament and the European Court of Auditors on the implementation of FMOs.

(79) This report is drawn up under Article 31.3 of the Internal Agreement on the financing and administration, of aid under Lomé IV.

(80) See Article 212 of the Lomé IV Convention.

(81) See in this connection the annual report of the Court of Auditors for the financial year 1993, paragraph 15.75, OJ No C 327, 24. 11. 1994.

ANNEX 1

PROCEDURE FOR DECISION-TAKING IN RESPECT OF NOTIFICATION AND PAYMENT OF STABEX LOMÉ IV TRANSFERS

>REFERENCE TO A FILM>

ANNEX 2

>TABLE>

ANNEX 3

>TABLE>

ANNEX 4

>TABLE>

ANNEX 5

>TABLE>

ANNEX 6

>TABLE>

COMMISSION REPLY TO THE COURT OF AUDITORS' SPECIAL REPORT ON STABEX UNDER THE FIRST FINANCIAL PROTOCOL OF THE FOURTH LOME CONVENTION

SINCE 1991 the Lomé Convention has provided for stricter controls on the use of Stabex funds, in particular that the use to which they are to be put be set out in a 'framework of mutual obligations' (FMO) and that the funds themselves pass through a double-signature ecu account in the name of the recipient country. The new procedure, coupled with a shortage of funds, did cause some delays: the opening of new foreign-exchange accounts has sometimes made for late payments: the negotiation of FMOs adds to the workload on both sides, sometimes taking some months; compliance by the ACP State with the obligations laid down in the FMO also means that, unless Stabex funds are used in support of structural adjustment, the foreign exchange or counterpart funds are disbursed more slowly.

However, the discipline introduced by these more exacting procedures offsets any loss of speed in Stabex transfers. The new system allocates the transfers more efficiently, thus helping to overhaul radically the productive sectors concerned and making for greater consistency between Stabex and other Lomé instruments. Stabex funds are playing a key role in reorganizing production and export chains in Cameroon and Côte d'Ivoire, for example, and in economic diversification strategies, such as those in Ethiopia and Uganda.

The Commission is pleased that the Court recognizes that the new measures have led to better controls over the funds mobilized.

The Court refers on a number of occasions to the staffing level of Delegations. It should be noted that the Commission tries to allocate the resources available to it as best it can.

1. INTRODUCTION

1.5. In the Commission's view, there is no contradiction between the objectives assigned to Stabex under the Convention. There is merely a hierarchy of possible objectives, which responds to the complexity of the economic problems facing the ACP States, particularly those where a process of structural adjustment is under way.

3. MANAGEMENT OF THE LOME IV STABEX ALLOCATION

Financial situation of Lomé IV Stabex

3.3 The Commission made a transfer decision for the 1992 application year under Article 206 (3) of Lomé IV on 22 July 1993. The decision was taken on the basis of the normal resources available and was conditional on additional resources being found to top them up. It could not be implemented until the ACP-EC Committee of Ambassadors agreed in February 1994 on how to deal with the shortfall in resources.

3.4 The Commission's obligations in respect of transfers to the ACP States, formally speaking, fulfilled once payment is made into an ecu account. As to the implementation of operations financed with Stabex transfers, the Commission has set up a system to show how funds are used that will become operational this year. The Commission is, however, aware of the advantages of tracking the use of funds and is developing a bookkeeping system for FMOs on the lines of what is done for programmes that generate counterpart funds. It is proposed to produce an annual report, the first of which will cover the 1995 financial year.

Insufficiency of resources

3.9 The Commission believes that the decision of the ACP-EEC Council in November 1991 was warranted by the economic and social situation of the ACP States concerned. The Court's calculation of the amount that should have been reimbursed (ECU 900 million) is purely theoretical and economic developments after the decision was taken show that these countries could not have repaid the money.

'All destinations' status

3.10 When the Council decided to prolong the derogation accorded under Annex XLVII to the Convention for another year, the Commission let it be known that it greatly regretted the extension, which, in its view, went against the spirit of the Convention. The shortfall in resources was particularly acute for the 1990 to 1992 application years, coverage rates falling to less than 50 %.

Measures taken to combat shortages of Stabex resources

3.11 The recorded interest accounts for only part of the interest generated for the financial year in question. Interest still to be recorded can be found in the regularization accounts or will be known only in the following financial year.

3.15 (a) & (b) That there is no explicit provision in the Convention for the transfer of resources or an exceptional drawing of funds does not, in the Commission's view, mean that the ACP-EC Council of Ministers does not have the right to make such a decision. Article 194 (4) provides that the Council may, on the basis of a report from the Commission, examine the steps to be taken to remedy the shortfall.

3.18 It appears that the consultant who produced the report referred to by the Court did not comply with the terms of reference.

Allocation of resources by country

3.21 There is a strong positive correlation between the amounts transferred and the export flows of agricultural products to the Community (or to all destinations in the case of ACP countries which enjoy this derogation) when the prices of major export products (coffee and cocoa, for instance) plummet on the world market, as happened in the 1990 to 1993 application years.

Note, however, that a corrective factor in favour of the least-developed, landlocked and island ACP States is due to be introduced in the mid-term review of Lomé IV.

4. THE CALCULATION OF STABEX TRANSFERS

Timetable and reliability of statistics

4.3 (a) The Convention lays down an extremely tight schedule for the appraisal of transfers. Nevertheless, the Commission will try to increase statistical cooperation in future (see reply to point 11.18).

(c) The Delegations check statistics with the means available to them, the workload depending on the number of products in question. However, once the basis for calculating the statistics for products giving rise to transfers has been properly established, it can be used in following years.

Reductions in transfer entitlements

4.5 (a) It is indeed difficult to establish quantitatively to what extent a drop in production is attributable to internal or external factors. This is why the ACP States concerned should take active part in consultations and present their point of view.

(b) Traditionally consultations took place between the diplomatic representatives of the ACP States and the Commission. In 1994 Commission Delegations were asked to play a greater role in the consultations on the 1993 dossiers and they are expected to play a bigger role in future. Other Commission departments are also becoming more systematically involved.

(c) A more active involvement of Commission Delegations in the consultations should ensure that the outcome is less influenced by the diplomatic representations of the ACP States.

(d) The proposal for the Commission's decision on the transfers is sent to Financial Control with the appraisals of the transfers in question, including protocols setting out the calculations of the reductions made under Article 203.

The difficulties inherent in trying to assess to what extent the drop in production has been caused by the ACP State or by external factors (as pointed out in point 4.5(a) of the Court's report) are still greater for Financial Control than for the authorizing officer. Even if Financial Control had been consulted at an earlier stage, it would have been obliged to confine itself to a check on the calculation of the transfers.

4.6 (a) The greater involvement of Delegations (see point 4.5(b)) should ensure that consultations will be focused as far as possible on technical considerations.

(b) It may be remarked that the fall in production, while not an objective of structural adjustment, was an unavoidable consequence of it.

Errors in calculation of Stabex transfers

4.7 (a) The sum in question was recovered before year-end by deducting it from the transfers due for the 1992 and 1993 application years.

(b) The sum will be added to the transfer for the 1994 application years and paid in 1995.

(c) Lesotho did not receive transfers for the 1990 application year because it sent the Commission completely unusable data. The Commission will, however, look into ways of finding a solution to the recurring problem of this country's statistics.

Financial Control will continue to carry out the checks incumbent on it. The issue of the management of, and checks on, counterpart funds generated by the use of EDF funds will be tackled when the EDF Financial Regulation is revised.

The impact of product quality on prices

4.8 A deterioration in product quality is indeed sometimes the fault of the ACP State. Other factors, such as the weather, can also come into play. It is, however, a factor not easy to quantify and the Court's recommendations concerning the technical nature of consultations would not be furthered if it were systematically taken into account. But the Commission will examine how to take account of this factor under the existing provisions in cases where the ACP State bears manifest responsibility.

5. THE PAYMENT OF STABEX TRANSFERS

5.2 (a) These funds are no longer EDF funds but belong to the ACP country concerned, the use to which they are to be put being stipulated in the FMO. The Commission therefore does not think one can talk about a second layer of EDF management.

(b) The decision to open an ecu account in one Member State rather than another is entirely a matter for the ACP State concerned. Moreover, as part of the Lomé IV revision, it is proposed to make explicit provision that the ecu account must be opened in an EU Member State.

(d) The Commission disposes of the resources it has available for the Delegations as best it can. Technical assistance is usually provided if large amounts are involved.

5.3-5.7 The Commission is not convinced that the measuring instrument used by the Court (median time) is the best one. It does not show the size of the amounts in question, for example.

But the Commission does believe that the time taken to open accounts is too long and ways of reducing it should be found. The procedure for opening a double-signature account is now relatively simple and well known to the ACP States.

The instructions given to the Delegations, and thus to the ACP States, were clear enough for rapid payment of the transfers. One of the factors responsible for delays was the opening of a double-signature account, the principle of which was not easily accepted by the ACP State in more than one case.

Article 1 of the transfer agreement concluded with each of the ACP recipients contains the provision that the Commission should effect a transfer to the ACP State for a given application year. It is then up to the ACP States, in the Commission's view, to open an ecu account into which the transfer can be paid.

5.8 The Commission departments concerned have been trying since the entry into force of Lomé IV to convince the ACP States to make banks compete for Stabex funds; Cameroon and Uganda have done so. Such a move can only increase the financial returns and the Commission is encouraging the ACP States to do so.

5.9 These problems are being studied with the Delegations concerned with the aim of ensuring that the provisions of the Convention are fully complied with.

6. THE PREPARATION OF THE FRAMEWORKS OF MUTUAL OBLIGATIONS

Provisions of the Convention concerning the utilization of funds

6.4-6.6 The Lomé Convention lays down that transfer should be used as a matter of priority in the sectors affected by the losses of export earnings. Thus, under Article 186 (2) it is a priority rather than a requirement. This interpretation is supported by the fact that various uses are provided for in the Convention and this answers the questions raised by the Court. Moreover, in implementing the FMOs national authorities working hand in hand with the Commission have made full use of the provisions of the Convention and sought to find the swiftest and most effective ways of using the transfers in the context of the individual country concerned.

Procedures for preparing FMOs

6.11-6.12 The 'Principles, Guidelines and Rules for the Use of Stabex Transfers under Lomé IV' referred to by the Court in points 6.8 and 6.12 provides comprehensive guidance for an analysis of the causes of revenue losses. The Commission will look into the updating of the documents in question and their dissemination to the ACP States.

Delays in preparation of FMOs

6.13 The measure used by the Court does not take account of the amount to be transferred. The FMOs signed after the 'median time', apart from those of Ghana, Ethiopia and Togo, concerned amounts that were among the smallest for the year in question. But, whatever the case, the time taken was too long and the Commission will make every effort to reduce it as much as possible.

6.14 The lengthening delays are attributable to the late date on which the decision on the 1990 application year was taken. When the time came for the following decision (on 1991) most ACP States were still busy negotiating the 1990 FMOs.

Identification of operations to be financed

6.16 The Commission believes that the take-up of resources could be speeded up by systematically providing technical assistance for the preparation and implementation of the FMOs if the sums in question warrant it. For smaller amounts the Commission will look into ways of simplifying procedures in keeping with the principles that led to the introduction of FMOs under Lomé IV.

6.18 The exceptional delay in the adoption of the decision on the 1992 application year by the ACP-EC Committee of Ambassadors is regrettable but not the Commission's fault in any way. Nevertheless, steps have been taken to prevent the accumulation of unused transfers, mainly by drawing up FMOs covering two application years.

Commission procedures

6.21 There has to be consultation between the various departments of the Commission to ensure that the best possible decision is taken on the utilization of the resources transferred.

6.22 (a) The Commission is looking into the possibility of abolishing this procedure.

(b) The Commission has asked the Member States if this procedure could be abandoned.

6.23 The delegation of powers to Delegates is only one aspect of the problem; as the Court itself has noted, another is the paucity of the resources available to the Delegations.

7. THE UTILIZATION OF STABEX TRANSFERS

Financial implementation of the FMOs

7.1-7.3 Before a system for the monitoring of Stabex transfers is set up, there has to be an analysis of the use of funds and of the methods of gathering information and how to deal with this information. The first step has been to track transfers generating counterpart funds and gradually enter them in the existing data base. This was followed in 1994 by an inquiry into the accounts in Europe and the sending of three on-the-spot FMO monitoring missions. At the end of that year a fresh survey of accounts in Europe was undertaken and extended to cover spending in ecus on projects.

The 1995 financial year will undoubtedly be a turning point. Annexed to the 1995 management accounts will be a summary report on implementation of the FMOs, while missions to monitor counterpart funds will also cover Stabex transfers.

Detailed implementation of FMOs

General import programmes

7.10 In this case the forecast of requirements made by the World Bank and the IMF, to which other donors backing the structural adjustment process subscribed, proved to be an overestimate. There were a number of reasons for this, among which:

- a shortage of local currency, the production of which was blocked by the IMF at a time when much of it was tied up in treasury bonds or used as a substitute currency in neighbouring countries;

- abnormally high interest rates in a context of slow inflation, which discouraged investment and thus reduced the need to use the resources set aside to finance imports.

7.12 Article 247 (2) of the Convention provides that balance-of-payments aid under structural adjustment programmes can be provided in the form of general import programmes (GIPs) or sectoral import programmes (SIPs). A GIP in a particular economic situation is justified balance-of-payments aid, while the counterpart funds used for targeted budgetary spending give a sectoral dimension.

7.13 In application of the note of 18 May 1992 on the implementation of FMOs, the Commission believed that once the conditions were satisfied, Ghana could be given direct budgetary aid (i.e. without recourse to a GIP) since the cedi was convertible.

Sectoral import programmes

7.17 It is true that the purchase of supplies was generally subject to normal EDF procurement procedures in view of the large amounts of funds involved. Substantial progress has been made in the meantime.

The Commission did not want to abandon EDF tendering procedures in Ethiopia simply in the interests of sound financial management. In a country that is being rebuilt it is important to have transparent procedures and equal treatment of importers, even though this can prove time-consuming.

7.20 In October 1994 the Commission concluded with the Ethiopian authorities a Memorandum of Understanding on the use of 1990 and 1991 Stabex resources through the foreign exchange auction system established within the ongoing reform progress. At the end of December 1994 34 % of the funds had been committed (100 % of the agricultural input supply and the revised foreign trade promotion budget) and 10 % disbursed.

7.23 While importers did seek to have purchases split up, so delaying the release of the second tranche, they did so in order to maintain their traditional sources of supply.

Utilization of counterpart funds

7.26 (d) While it is true that the Commission did not have an in-depth knowledge of the cotton sector, the funds were allocated with the full agreement of the World Bank, which is the lead agency for the sector. The funds paid off the industry's debts and paved the way for the privatization of the linter office. In view of the difficulty of allocating the counterpart funds as proposed in the FMO either speedily or usefully, the Ugandan authorities and the Commission decided, in accordance with the Convention (Article 209 (4) - consistency with structural adjustment), to allocate some of these funds to the public-sector demobilization programme that was backed by most aid donors.

7.27 (a) As no deliveries had taken place by 30 October 1993, no counterpart funds were generated and so no disbursements could be made. This was because of the conditions of the transition in Ethiopia.

(c) See reply to point 8.10.

(d) The decision to reallocate the counterpart funds generated by the second and third tranches of the 1990 Stabex transfer for the coffee sector was taken as a result of the amendment of the 1990 Stabex FMO decided on 18 February 1993, by which the counterpart funds in question were to be added to the 1991 Stabex funds, which were allocated to the coffee sector. An analysis of the coffee sector (where the loss of earnings had taken place) identified financing requirements that could be covered with Stabex resources.

Internal debt relief measures

7.32 The change as regards the repayment of arrears was in accordance with Article 18 of the 1990 FMO and the result of an agreement between the contracting parties, namely the national authorizing officer and the Commission Delegate. The grounds for the payments to be made, including financing costs, were checked with a view to establishing each exporter's share of the amount owed. As an old debt costs more than a new one for exporters, the financing-cost element led to a share-out that was unequal if you simply compared Stabex payments with what was owed to exporters.

7.33 Even though the ONCPB (National Marketing Board) does have assets, there is no evidence that they would be enough to pay off all debts (CFAF 20 000 million, or ECU 60 million). Failing the government's agreement to the liquidation of these assets, the Commission has sought to do what it could for creditors, in accordance with the Convention.

7.37-7.38 Between September 1990 and June 1991 the Treasury repaid CFAF 15.15 billion of its CFAF 80 billion debt to the BCEAO (Central Bank of the West African States) in order to get the 1990/91 campaign under way even before the signing of the 1990 FMO on 21 December 1991. This repayment had been agreed in talks between donors on the 1990 budget. Stabex funds were regarded as a potential external resource for providing budgetary support and so easing the Treasury's repayment: Stabex aid was clearly seen by the parties concerned (including donors, the World Bank and the IMF) as one of the means for carrying out, starting at the end of 1991, the reform of the sector laid down in the 1990 agricultural sector adjustment loan (ASAL), thanks to conditionality.

For administrative reasons, it took time to mobilize the Stabex resources (transfer decision on 31 July of the year following the FMO application year). The Côte d'Ivoire could not wait for the 1990 Stabex resources to become available to cover the first debt repayments, which were necessary to ensure cash flow in the sector. But in the macroeconomic context the repayment had been posited on the availability of Stabex funds.

The CFAF 12 620 million were given to the Treasury at a later date once the conditions attached, including the transmission of agreements concerning the debt-arrears consolidation operation and half-yearly reports on the sales of non-graded stocks, were deemed to have been met.

The funds were thus used for the ex-post financing of public expenditure that was made at the right time and which helped the recovery of the sector.

The Commission believes that Stabex funds were crucial in finding a solution to these financing problems and that the undertaking made in 1990 to allocate the Stabex transfer in question for the repayment of the Treasury's debts at the end of 1992 helped ensure the resumption of activity in the sector in the 1990/91 marketing year.

7.40 The use of Stabex funds to clear the arrears of the CSSPPA in respect of cotton and rubber producers is justified by the fact that some liquidity had to be restored to key operators in the sectors of diversification to enable them to become operational again (this also reduced the government's domestic arrears).

7.41 The priority at the time was to supply liquidity to key operators in these sectors, so restoring credibility and financial capacity.

Price stabilization

7.48 The 1991 FMO, by providing funds to cover the sector's deficits, enabled the government to fulfil its commitments under the 1990 FMO, namely to release clearly identified resources needed to cover any deficit of the sector in the 1991/92 marketing year with a view to avoiding any increase in arrears. The 1990 FMO did not prohibit the government from using external resources, and the 1991 Stabex transfers were known and clearly identified resources. The aim of restoring balance to the sector in budgetary transparency was thus achieved.

7.49 The funds given to Burundi were for an operation supported by all donors of the coffee sector, namely the Commission, the IBRD, French bilateral aid and the Burundian Finance Ministry. Donors did, however, ask the government to carry out a study to give further details about the statutes and operation of the stabilization fund. The FMO also specified the use to which the funds were to be put.

8. IMPLEMENTATION OF ACP STATES' COMMITMENTS

Respect of conditionality

Macroeconomic conditions

8.10 The Commission merely applied the 1990 FMO, which made disbursements conditional on the setting-up of a structural adjustment programme. This was but a possibility provided for in the FMO. The first project to start up was completed. Later, disbursements were suspended following suspension of the structural adjustment programme. It should also be emphasized that the civil war going on in Rwanda since October 1990 militated against implementation of a continuous policy by the government. The few counterpart funds generated by the first project were not spent since they were subject to the same conditions as those referred to by the Court (1990 Stabex FMO).

8.12 Although the two instruments (structural adjustment aid and Stabex) are often closely linked in practice, Stabex disbursements are not necessarily conditional on the extent to which the country's structural adjustment programme is implemented.

(a) In Cameroon the Commission had made a substantial contribution to the overhaul of the cocoa and coffee sectors through previous Stabex transfers (1988-90) and the 1991 Stabex was to be used to relaunch production. The FMO included accompanying measures designed to promote rural development and mitigate the adverse effects of falling plantation revenues on the local population. While the operations provided for in the 1991 FMO were consistent with the structural adjustment programme, the macroeconomic conditions attached to the programme concerned a wider context than the sectoral adjustment financed with Stabex funds and this justified continuation of Stabex operations by the Commission and the national authorities.

(b) In Kenya the Advisory Group's decision in November 1991 to suspend balance-of-payments aid temporarily affected only new operations, whereas preparation of the 1990 Stabex operation was already under way and the particular nature of this instrument (automatic triggering in certain conditions) placed it outside the scope of the decision. Lastly, the improvement in the political situation following the Advisory Group's meeting led to the decision to sign the FMO. Moreover, note that the 1991 funds have not yet been used because the FMO has not yet been signed (March 1995).

Any changes decided on subsequent to these transfers fully complied with Articles 186 (2) and 209 (4).

8.13 Where transfers are used in combination with structural adjustment, the conditions are usually the same.

Sectoral-level reforms

8.17 Although the World Bank and the IMF briefly suspended their structural adjustment programme, the Commission found that the general conditions it had set for the continuation of its aid were still met, even though some of the specific conditions referred to by the Court were not. This is why disbursements were made. It was therefore logical that these initial conditions should feature in the new FMO.

8.20 As part of the progressive liberalization of these sectors, which started under the 1990 Stabex FMO, the Commission and the government committed themselves in 1994, under the 1992-93 FMO, to wind up the ONCC (National Coffee and Cocoa Office) no later than 30 June 1995.

8.21 Since there was no provision in the 1991 FMO concerning payments to creditors of the ex-ONCPB, the Commission was no longer tied by the condition to wind up the affairs of the ONCPB satisfactorily and focused its attention on relinking production when the unexpended balance of the 1990 FMO was transferred to the 1991 FMO.

8.22 The Commission considers that the release of funds in tranches would not necessarily have led the government to respect its commitments, since only part of the stabilization fund was financed with Stabex funds, the bulk being provided by the government.

It should also be borne in mind that certain political and social factors had to be taken into account by the Commission during the preparation of the 1990 and 1991 FMOs, including the election of a new government in 1992 and the fear of social unrest.

Nevertheless, the observations made by the Court have been taken into account in the 1992-93 FMO, which establishes a clear set of conditions and a precise time frame (justification and release of funds from the ecu account in two tranches).

8.23 The Commission used the guidelines contained in the FMOs to avoid immobilizing for a long time funds that could be used in other sectors.

9. FINANCIAL AND ACCOUNTING PROCEDURES

Legal framework for implementation of Stabex operations

9.1 In addition to the note 'Principles, Guidelines and Rules for the Use of Stabex Transfers under Lomé IV', there are other instruction notes dealing with the problems of managing counterpart funds. Forms of managerial support (provision of technical assistance) complement resource management. More detailed instruction notes and financial execution procedures could now be drawn up on the basis of the experience gained under the first financial protocol and in the light of any new guidelines emerging from the mid-term review of Lomé IV.

9.2 In the Commission's view, the slow-down in Stabex disbursements caused by EDF procedures is unavoidable. These are cases which lend themselves to a use of EDF funds in the framework of programmable aid where the requirements of quick disbursement are not the same as in other cases.

9.3 To subordinate eligibility to satisfactory use of funds accorded would require a fundamental change in the Convention's provisions on Stabex.

Functioning of the double signature accounts

9.4 See reply to point 5.2(a).

9.5 Interest earned on the 1990 transfer in the ecu account has not been transferred to the Stabex kina account. The Delegation will advise the government that, in accordance with the 1990 FMO, the counterpart value of these interest payments should also be credited to the Stabex kina accounts and ask the government to make appropriate corrections in the accounts.

Concerning the interest on the 1991 transfer, the Delegation has asked the government for further clarification.

This operation will not involve additional payments by the government to the stabilization funds as the government's contributions far outstripped Stabex funds in that year; it is rather a matter of proper bookkeeping entries for that year which adjust the Stabex contribution upwards and that of the government downwards.

In the case of Burundi, the Commission insisted that Stabex funds of the following year be placed on an account requiring two signatures, those of the Delegate and the national authorizing officer.

9.6 The Commission takes note of this and will act accordingly for the FMOs currently being negotiated.

Counterpart funds

9.7 The problems arising in connection with the setting up of the counterpart funds in Rwanda can be explained by the difficult wider context in the country, the war which started in October 1990 having disrupted agriculture (the RPF's attack in the North), especially export crops. This meant it was difficult to sell the products and so generate counterpart funds; the coffee section of the OCIR (Rwanda's cash crop marketing board) complied with the deadlines for transferring counterpart funds as and when it was actually able to dispose of the products, and not when it took possession of them.

9.8 In the case of Ghana note that the 'loss of value' of non-interest bearing counterpart funds was very slight. Most of these funds were mobilized immediately and used for 1990 and 1991 Stabex operations. But the accounts of counterpart funds are now interest-bearing.

9.9 Article 211 (3) of Lomé IV lays down that Stabex-generated counterpart funds be put on deposit in interest-bearing accounts for which the two signatures of the NAO and the Commission Delegate are required. The Convention does not stipulate that such accounts be separate. Therefore, accounts may be consolidated whenever it is possible to carry out the accounting and financial controls necessary for their identification and conditional mobilization.

Financial monitoring of Stabex by the Commission

9.10 See reply to points 7.1 to 7.3.

9.11 The annual report to the Member States on the operation of Stabex will carry the additional information referred to by the Court as from the 1993 application year.

10. MONITORING AND EVALUATION OF LOME IV STABEX

10.1 The framework for monitoring the overall implementation of Stabex is being drawn up (see replies to points 7.1 to 7.3).

10.2 A report is drawn up every six months and sent to the Commission (see reply to point 6.16).

10.4 See replies to points 7.1-7.3 and 11.12.

10.5 The Commission believes that, in accordance with Article 212 of Lomé IV, the report drawn up by the ACP State should be confined to what was done with the funds transferred. Their impact can be properly evaluated only by external consultants (see point 10.6).

10.6 Evaluations have taken place, or are under way, for the following countries: Ghana, Côte d'Ivoire, Uganda and Cameroon. Others are planned for 1995 and should be followed by a general evaluation of Stabex as a whole. Furthermore, many one-off evaluations have been, or are being, carried out on specific aspects of the FMOs.

11. CONCLUSION

General context of the implementation of the Stabex system

11.1 Despite these difficulties, Stabex continued to function. Although the coverage rates fell considerably, a comparison with the national indicative programmes shows that the Stabex transfers were in fact substantial sums that did help mitigate the effects of lost export earnings.

11.3 The Commission does not believe that it is the system of guaranteed prices which is mainly at fault. While the guaranteed price must bear some relation to the world price (what relation to be established), evaluations show that the difficulties referred to by the Court are mainly the result of bad management of the surpluses generated by the stabilization funds in the years of high world prices.

11.4 Stabex is an 'acquis' of the Lomé Convention. Its role is not comparable to that of commodity agreements, the aim of which was to manage prices by regulating world supply. If such agreements work (which was not the case in recent years for the main products covered by the system), they can have only a beneficial effect on Stabex, which would once again fulfil its original function of smoothing out cyclical fluctuations.

Economics of the system laid down in the new Convention

11.7 In the Commission's view, the principle of compensation for losses is subject to the availability of resources, as provided for in Article 193, in addition to which are the extraordinary resources the ACP-EC Council of Ministers may make available under Article 194 (4).

11.8 By making greater use of the Delegations the Commission now tries to focus consultations on technical aspects and avoid leaving too much scope for negotiation. The Commission will also see how a decline in the quality of a product for which the ACP State is manifestly responsible could be taken into account under the existing provisions. The derogation referred to by the Court will expire from the 1994 application year.

11.9 Where the transfer is large in relation to other economic factors (the country's budget of balance of payments, for instance) it inevitably has a macroeconomic impact that should not be overlooked. The uses to which the transfer is put should therefore at least be consistent with the country's general economic policy, and preferably actively second and further this policy. What applies at macroeconomic level is all the more true at sectoral level: transfers that are large in comparison with sector variables must be used in a way that is consistent with, and supportive of, sectoral policy.

The importance of macroeconomic and sectoral consistency is not confined to Stabex transfers but is true of all forms of aid. It is particularly true in the case of countries where Stabex transfers are relatively substantial and are used in a fairly short span of time.

This is not a 'contradiction' but a reality that has to be taken into account in implementing Stabex. And while it does take more time to negotiate FMOs and satisfy the conditions for their implementation than it does to use a transfer not subject to the same conditions, this is not a contradiction but a desire to find a balance between quick disbursement and the checks on their use demanded by the Member States. Despite the time taken to do this, Stabex funds are mobilized quite quickly compared with other quick-disbursing instruments of the Commission and other donors (see reply to points 6.4-6.6).

11.10 The 'Principles, Guidelines and Rules for the Use of Stabex Transfers under Lomé IV' of 4 February 1991 referred to by the Court (see point 6.12) has a whole chapter devoted to the subject of generation of counterpart funds by import programmes. The Commission had foreseen that Stabex-funded import programmes under Lomé IV would generate large amounts of counterpart funds and that a degree of flexibility in applying the rules on the allocation of foreign exchange would be called for.

Financial management

11.11 The documents available to the unit at the time concerned the situation of Stabex transfers of previous years and the accounts reflected this situation. The accounts are currently being checked by the Commission (see replies to points 9.11, 3.11 and 3.15).

11.12 Although the payment of Stabex transfers and the entry of these payments in the accounts are the last steps of the financial procedure, the departments in Brussels have established a number of measures to keep track of both the foreign exchange and counterpart funds. It is planned to strengthen and improve these measures still further.

11.13 See reply to point 5.2(a).

11.14 The Commission will ensure technical assistance is used where appropriate; it can easily be financed with Stabex resources, if need be. As to staffing levels, the Commission can only, as we said in the introduction, do the best it can with the resources available.

Implementation of Stabex transfers

11.15 The Commission does not share the Court's view that the division of resources is unbalanced. The division is dictated by the provisions of the Convention and transfers bear a close relation to the value of imports of Stabex products into the Community. It is planned to introduce an amendment to help the less-developed, landlocked and island countries during the mid-term review.

11.16 Delays in drawing up FMOs usually arose because of discussions with the national authorities on the use of transfers and conditionality. Such discussions are only to be expected in the case of transfers of what are often substantial sums of money and they can take some time.

Internal consultations within the Commission on how the FMOs were to be applied (especially as regards macroeconomic and sectoral consistency) were indeed needed at the beginning of the FMO implementation procedure but countries receiving large transfers (Côte d'Ivoire and Cameroon, for instance), where there was intensive debate on the best way of ensuring macroeconomic coherence, were also among the countries that were quickest in approving and implementing their FMOs. This internal debate did not, therefore, slow down the adoption of the FMOs; on the contrary, it paved the way for speedy implementation thanks to proper identification of the recipient country's priority needs.

11.17 The Commission believes that Stabex funds have usually been used as required by the Lomé Convention (see replies to points 6.4-6.6, 7.26, 7.27(d), 7.32, 7.37; 7.38, 7.41, 7.48, 7.49).

11.18 The deadline for appraising transfers certainly leaves little time for any specific statistical cooperation operations (particularly when resources are inadequate and all decisions have to be taken before the end of July of the year following the application year). Subject to constraints, the Commission will strive to improve statistical cooperation with the ACP States.

11.19 The Commission cannot be held responsible for delays in areas within the remit of the ACP States (see reply to point 10.5).

Overall assessment

11.20 It is certainly not the job of Stabex to provide solutions to the international problems referred to by the Court. What it can do, within defined geographical areas, is provide ex-post total or partial compensation for a shortfall in resources because of lost export earnings from products covered by the system.

The Commission is convinced that the system remains valid after 19 years, even when the compensation is only partial. The Commission notes that the new provisions of the Convention have improved control of the funds mobilized. However, further improvements are still possible and the Commission will endeavour to bring them about.

11.21 The rise in coffee and cocoa prices in 1994 should ease the problem of the shortfall in resources in coming years.

The Commission notes that it does not have sole responsibility for the amount allocated to Stabex, having only a right of initiative. It also notes that, if there is a shortfall in resources, any attempt to favour one particular group would reduce the coverage rate for countries outside that group.

The Commission will endeavour to increase the linkage between Stabex and other EDF instruments, especially in ACP States which are implementing a coherent reform policy to which the system can make an important contribution.

On the question of more systematic monitoring of the FMOs, the Commission confirms that steps are being taken. It will describe progress in its report on the functioning of Stabex in the 1994 application year.