Special Report No 15/2000

on the Cohesion Fund, together with the Commission's replies

(submitted pursuant to Article 248(4), second subparagraph, of the EC Treaty)

(2000/C 279/01)

CONTENTS

>TABLE>

SUMMARY

For the period 1993 to 1999 a total budget of EUR 15150 million (1992 prices) was allocated to the Cohesion Fund (CF) in order to co-finance, at the rate of between 80 % and 85 % of the public expenditure involved, projects in the field of the environment and in the area of transport infrastructure in Spain, Greece, Ireland and Portugal. The Court examined the Commission's management of the CF and its implementation in the Member States, both from the point of view of the legality/regularity of the transactions and of the measures taken to evaluate the effectiveness of these transactions (see paragraphs 1 to 5).

An even spread of appropriations between the environmental and transport infrastructure fields was achieved, overall, by the end of the period, mainly in favour of road projects and projects concerning water. The financial implementation tended to concentrate on the clearance of projects approved at the beginning of the period (see paragraphs 6 to 13).

The project approach, devised to enable the Commission to retain tight control over the management of the CF, does not always ensure that the projects proposed are the most productive before the Commission can authorise them. Even though the administration of the CF and the ERDF respectively is carried out within the same ministries, which in practice promotes coordination between the two instruments, priority is given to making the best use of the appropriations allocated (see paragraphs 21 to 24). The selection of projects within the framework of the CF, while generally taking into account how far advanced the projects are, is influenced by the chronological order in which the applications for assistance to projects are submitted with no provision for any real programming in partnership with the national authorities concerned (see paragraphs 29 and 30). Although the examination of project applications by the Commission led, in some cases, to lengthy delays, some files revealed deficiencies as they did not contain essential information or this information was unreliable (see paragraphs 31 to 37). The term "project" was applied in the cohesion countries in different ways. Likewise, in the case of projects which generate revenue, the basis of assessment of the amount eligible for CF financing may vary according to the components included and the latter are neither homogeneous nor exhaustive. Occasionally, projects in the process of completion and finalisation were entered as new projects which qualified for CF assistance (see paragraphs 38 to 48).

In contrast to Structural Fund aid, CF assistance is subject to a conditionality clause. The application of this principle makes CF co-financing dependent on an examination of the budgetary position of those beneficiary countries which have excessive government deficits. The results of this review did not call CF financing into question. In the short term, however, conditionality can make the achievement of investment goals and those of reducing government debt and government deficits difficult (see paragraphs 49 and 50).

The development of a project financed by the CF occurs in several stages and the Commission intervenes mainly in the initial and final phases. The intermediate stages and their financial consequences would merit more rigorous management, in partnership. The Court's checks revealed a number of anomalies in respect of eligibility and accounting for expenditure and also the finalisation of projects. The Commission should devote more resources to on-the-spot checks in order to detect weaknesses in the management systems (see paragraphs 51 to 70).

For the evaluation of the CF, the Commission called in a consultant who developed an approach which used various macroeconomic models. In spite of certain limitations, these produce useful results. The Commission should develop this tool further and apply it more fully so that general conclusions can be drawn. At the level of the cohesion countries only Spain and Portugal attempted to evaluate the national impact of, firstly, the CF and, secondly, certain projects (see paragraphs 71 to 84).

The cost-benefit analyses (CBA) performed before approval of the projects are not always prepared with equal rigour and the calculation methods used may vary greatly depending on the managers involved. The indicators are sometimes too broadly defined to evaluate the real impact of the projects, to draw comparisons and to enable reliable measurements to be taken (see paragraphs 85 to 98).

The improvement of the transport infrastructure and of the environment in the beneficiary countries underlines the impact, in terms of quality, of the CF as an instrument of greater economic and social cohesion in the EU. As regards the quantitative aspect, the models used show a positive impact by projects, in terms of jobs, output, income and investment, on the regions concerned and on other regions and on the economy of the beneficiary Member State. They also show, albeit less clearly, a positive impact on cohesion and convergence (see paragraphs 99 to 108).

In view of these encouraging results, the Commission should attempt to concentrate its management on a medium-term planning perspective which would offer more guidance and continuity, and should endeavour to perfect the tools for the macroeconomic evaluation of the projects and of the CF as a whole (see paragraphs 109 to 117).

INTRODUCTION

Description of the Cohesion Fund

1. The Cohesion Fund (CF), provided for in Article 130d of the Maastricht Treaty and preceded, on a provisional basis, by the Cohesion Financial Instrument, was established in May 1994 by Council Regulation (EC) No 1164/94(1). The objective of this Regulation is to strengthen the economic and social cohesion of the European Union and its beneficiaries are those Member States in which the gross national product (GNP) per inhabitant is less than 90 % of the Community average and which have set up a convergence programme, i.e. Spain, Greece, Ireland and Portugal. The CF contribution is linked, in accordance with Article 6 of the Regulation, to the macroeconomic conditions stipulated in Article 104 of the Treaty and is designed to avoid excessive government deficits.

2. Following the decision by the Edinburgh European Council in December 1992, the CF received financial support totalling EUR 15150 million at 1992 prices for the 1993 to 1999 period, which was to provide financial support to the beneficiary countries for the implementation of two types of individual investment project:

(a) projects in the field of the environment which contribute, on the one hand, to achieving the goals of Community environmental policy as defined in the EC Treaty, in particular in accordance with the Directives now in force concerning the environment and, on the other hand, to meeting the priorities set out in the action programme on the environment and sustainable development;

(b) transport infrastructure projects designed to put in place or extend transport infrastructure as part of the trans-European transport network (TEN) or to provide access to the TEN.

3. An appropriate balance(2) must be found, for the whole of the period 1993 to 1999, between the financing of transport infrastructure projects and the financing of environmental projects. An indicative allocation of the resources available from the CF for each beneficiary Member State is set out in the Annex to the Regulation. The principle of co-financing by the EU and the beneficiary Member State applies, with the CF financing as much as 80 to 85 % of the total public expenditure relating to each of the projects.

4. The CF is characterised by a project approach based on three main principles:

(a) the Commission alone is responsible for the approval of projects and their implementing provisions and also for the evaluation of the potential and real impact of their implementation and, once they have been completed, it is also the Commission which is responsible for approving the final implementation report. Within this framework, the Commission may draw upon the assistance of experts should it not dispose of such expertise;

(b) with regard to monitoring and controls, a monitoring committee has been set up for each beneficiary Member State to allow discussion, once or twice a year, of technical questions relating to any amendments to the projects' financing plans or various other matters concerning, amongst other things, intermediate payments;

(c) eligibility, as defined under Article 3 of the CF Regulation, was also the subject of "principles governing the eligibility of expenditure within the framework of projects supported by the Cohesion Fund". The beneficiary Member States were notified of these principles, which take the distinctive features of this instrument into consideration, in May 1998.

Scope of the audit

5. The overall objective of the Court's audit was to examine the management by the Commission of the CF, on the one hand, and its implementation in the Member States, on the other. Apart from the legality and regularity of the operations co-financed by the CF, the investigation also included an evaluation of the efficiency and effectiveness of the projects involved and also of the complementarity of CF projects with other structural actions such as the European Regional Development Fund (ERDF). Although it was not the goal of the inquiry to make a direct evaluation of the CF's impact, certain findings are reported which have been noted by experts with regard to the results obtained and the impact produced by the CF. The inquiries were carried out in the Commission's offices and in the four beneficiary Member States. They covered a selection of 37 of the 181 projects which had been closed at the beginning of the year 1998 or were in the process of completion.

IMPLEMENTATION OF THE BUDGET

Financial development

6. An initial budget of EUR 1500 million was adopted in 1993. The budget increases every year in step with the rate of inflation so that the CF retains its value in real terms. In 1999 it amounted to EUR 2600 million. Taking the budgetary resources carried over from previous financial years into account, a total budget of EUR 2800 million was available to the CF in 1998 for commitment appropriations.

7. The implementing provisions featured in the CF Regulation emphasise that payments in support of approved projects of a duration of two years or more are to be made in accordance with the following rules:

(a) an advance of up to 50 % of the amount of the first annual instalment committed must be paid following the adoption of the decision granting Community assistance;

(b) intermediate payments may be paid provided that the project is progressing satisfactorily towards completion and that at least two-thirds of the expenditure relating to the previous payment has been made. None of these payments may exceed 50 % of the amount of each annual instalment committed. Furthermore, the total amount of the intermediate payments for all the instalments may not exceed 80 % of the total assistance granted. In the case of large projects and in justified cases this percentage may be increased to 90 %;

(c) the balance of the Community assistance is paid, provided:

- the project, stage of the project or group of projects, has been carried out in accordance with the objectives,

- the final application for payment is submitted to the Commission within six months of the physical completion of the project, stage of project or group of projects,

- the final report is submitted to the Commission,

- the Member State certifies to the Commission that the information given in the application for payment and in the report is correct.

The Commission's payments are made to the authority or body designated by the Member State and are made, as a general rule, not later than two months after receipt of an admissible application for payment.

8. Financial assistance began on 1 April 1993, the date on which the Cohesion Financial Instrument came into effect(3) and remained in force until 26 May 1994, the date on which the Regulation setting up the CF came into effect. The analysis of financial implementation within the framework of the CF does not apply in full until the year 1995 which is actually the first complete year in which the mechanisms provided for in the CF Regulation were applied in their entirety.

9. The expenditure charged by the Commission to the CF budget heading, which is managed by the Commission's Regional Policy Directorate-General, is characterised by a systematic concentration of commitments and payments at the end of the financial year and by a very low level of financial activity during the first quarter. In the course of the period 1995 to 1998 an average of 38,9 % of the amounts committed and 47,7 % of the amounts paid in respect of the financial year were committed or paid in the two last months of the year, while the first quarter shows an average of 18,3 % for commitment appropriations and 4,8 % for payment appropriations.

10. The utilisation of the CF appropriations also shows great variations between the numerous approvals of projects at the beginning of the period and a smaller number of new projects adopted at the end of the period. Thus, in 1998, commitments for new CF projects fell considerably in favour of allocations to projects approved in the course of the preceding years.

11. As Table 1 shows, there are still substantial amounts awaiting settlement and these cause uncertainty about the use of payment appropriations within the framework of the financial year concerned. In fact, as future financial movements in respect of these amounts cannot be estimated with any certainty, the Commission cannot identify sufficiently early in the year the amount of payment appropriations available for new projects. There is a risk that these amounts to be settled could momentarily hinder the payment of advances for new projects. Although, in principle, Annex II to the CF Regulation, as amended by Regulation (EC) No 1265/1999, is designed to allow an improvement in the management of the use of annual appropriations in respect of the period 2000 to 2006, it will not entirely resolve the above problem for the budget years in which the period 1994 to 1999 will be closed.

Table 1

Summary of the implementation of the budget for the Cohesion Fund

>TABLE>

Distribution of finance by sector

12. The provisions of Article 10(2) of the Regulation establishing the CF stipulate an appropriate balance of Community finance between projects relating to the environment and those relating to transport infrastructure over the entire period between 1993 and 1999. Table 2, which covers the entire period 1993 to 1998, shows that, overall, an even distribution of appropriations between sectors has been achieved, even if the distribution of resources between these two fields was very uneven over each financial year and varied according to the beneficiary Member State. This unequal distribution of appropriations between the two fields can be explained by some very large projects in a particular sector, by difficulties encountered in respect of public contracts, by the somewhat slow integration of the environmental dimension in the programming of sustainable development and by the small scale of the environment projects.

Table 2

Balance between transport projects and environmental projects

>TABLE>

13. In addition, Table 3 "Cohesion Fund assistance by sector" - shows that the distribution of investments favours certain types of intervention, such as road projects and those concerning drinking-water supply and the treatment of urban waste water, to the detriment of others such as rail and maritime infrastructures or the conservation of natural resources.

Table 3

Cohesion Fund assistance by sector - Commitments for the period 1993 to 99 in total by field

>TABLE>

OBSERVATIONS ON THE MANAGEMENT OF THE COHESION FUND

Programming the projects

14. Before the final decision to grant assistance from the CF, the Commission must carry out an in-depth evaluation in order to guarantee that the projects selected are of high quality and, in particular, in accordance with the criteria in Article 10(5) of the CF Regulation which sets out the essential elements on which the Commission's decisions are based, namely:

- the medium-term economic and social benefits of the project concerned,

- the priorities established by the beneficiary Member States,

- the contribution which projects can make to the implementation of Community policies on the environment and trans-European networks,

- the compatibility of projects with Community policies and their consistency with other Community structural measures,

- the establishment of an appropriate balance between the field of the environment and that of transport infrastructure.

15. In addition, under a framework agreement signed on 15 December 1994 covering the entire duration of the CF Regulation, the Commission may, if necessary, invite the EIB to assist in the evaluation at the technical and financial level, in order to assist the Commission in the examination of projects and help to achieve the best combination of sources of finance.

Deficiencies in the guidelines relating to applications for assistance

16. Desk audits of the case files in the relevant Commission offices showed that the essential elements stipulated by the Regulation (on which the individual decisions were based) were not always on file and that the cost-effectiveness, cost/performance and impact analyses in respect of the project had not been drawn up systematically since no precise instructions concerning the methods to be followed had been included in the form provided for this purpose. Likewise, alternatives to the project had not always been studied and taken into consideration.

17. For this reason, in the case of some actions within the framework of the four groups of projects audited in Spain, the Commission was not always aware of the details (as far as the locality and the amounts involved were concerned).

18. In Portugal, analysis of the costs and benefits associated with an infrastructure project did not produce any insight into the economic and social returns. In the case of another application for assistance in the field of the environment the potential receipts were not taken into account as the Commission concluded, without conducting a financial analysis, that it was impossible to establish a link between the receipts collected by the beneficiary company, a commercial water company, and its investment, financed by the CF, which mainly consisted of replacing worn and sometimes leaking water pipes in the network. Although there are no provisions on this in the current Regulation (792/93), on the basis of which the project was financed, in certain cases the Commission could usefully have taken potential revenue into account in its analysis.

19. In Greece, in the field of the environment, the consultants draw up files on applications for assistance and, in particular, compile socioeconomic studies (cost-benefit analyses, calculation of the economic and financial internal rate of return (IRR) and estimation of the project's impact in terms of job creation), often without making visits to the final beneficiary concerned. Within the framework of the checks carried out inconsistencies were discovered, in particular:

- the existence of potential revenue generated by a project included in the socioeconomic analysis which was not taken into account when fixing the maximum rate of CF assistance,

- different types of cost are considered to be acceptable by the Commission, in the CBAs, in respect of sectoral projects of the same type. Some allow for handling charges, others do not, without any explanation. Indirect staff costs and operating costs such as electricity may be included in some projects but not in others. For the most part, these data are often overestimated or incorrect,

- for a project in Kalamata (Greece), the socio-economic study used data from a similar project in Rhodes and at no time was this queried by the parties involved,

- job-creation estimates submitted in the project applications had sometimes been overestimated or were unrealistic.

20. Considering the fact that the explanatory and analytical documents included with a project application are not always exhaustive and reliable, the Commission is not always in a position to ensure that the projects proposed are the most effective before authorising them. On the other hand, it was not possible, in the context of the controls carried out, to identify the projects rejected by the Commission or to draw up a list of the main motives for the rejection of projects submitted by the beneficiary Member States.

Coordination between the Cohesion Fund and the ERDF

21. In all the cohesion Member States the national system for the implementation of the CF has been fully integrated with the existing structures for the administration of the ERDF. Often, as in the case of Portugal and Ireland, it is the same national administrators that are responsible both for the sectoral OP of the ERDF and the CF projects in the same field. In practice, this concentration of responsibilities on the same administrators and ministries favours coordination between the CF and the ERDF, even if the two instruments obey different administrative logics, one based on programming and the other on the management of projects.

22. Generally, the management and organisational systems work in such a way that there is no more than a slight risk of non-compliance with Article 9 of the CF Regulation concerning the cumulating and overlapping of sources of Community finance, i.e. no item of expenditure may benefit simultaneously from CF and Structural Fund assistance.

23. It is useful for the Commission to know the possible links and synergies between the projects financed by the CF or the ERDF. For this reason the CF application forms explicitly provide for mention of the potential links with other projects financed by the European Community. It was found, however, that the forms were not always filled out correctly by the national authorities.

24. Even if the management of the CF and of the ERDF is carried out in the same ministries, priority is nevertheless given to optimising the appropriations allocated to the two instruments, rather than to the overall coherence of the structural policies which they support.

Budgetary and sectoral constraints

25. The aim of the project approach was to enable the Commission to apply more rigorous project selection criteria, to gain a better understanding of the impact on economic and social cohesion of the projects approved, to improve, as a consequence, controls and to increase transparency and the quality of CF actions. The Commission is thus able to play a more active role in the construction of trans-European transport networks and in the protection of the environment.

26. However, this project approach was difficult to implement. At its launch projects were sometimes prepared in haste then, subsequently, in the course of implementation, allowance had to be made for an even distribution between the sectors. The fact that projects are submitted individually and at different times makes it difficult to compare them with each other and is not conducive to their selection for approval.

27. Within the framework of a multiannual budget, the Commission notifies the responsible national authorities of the indicative annual budget with a balanced distribution between transport infrastructure and the environment. The actions to be carried out are determined both on the basis of general guidelines as set out, in particular, in the Treaty, and on the basis of a national strategy with its own priorities in the two sectors concerned.

28. The programming for the two sectors concerned was, in general, carried out at different rates. Generally, the transport infrastructure sector can draw on long experience of programming and had sufficient well-designed projects, whereas in the environmental field the experience acquired is of more recent date. In most cohesion countries, therefore, the environmental projects were approved late in comparison with those in the field of transport infrastructure.

29. Generally, at national level, the project selection procedure takes into account how well-developed, in terms of quantity and quality, the project plans are, in particular how far advanced the projects are; it is also influenced by the chronological order in which projects for which assistance has been requested are submitted.

30. Even if the Regulation made no provision for any real programming for the CF with the goal of achieving a coherent overall view, in the medium term, of the actions to be implemented, its implementation on the part of the Commission, in partnership with the national authorities concerned, would have allowed the beneficiary Member States to improve the quality of the management of their applications for assistance in terms of impact and strategy.

Geographical distribution of the projects

31. Community financial assistance within the CF framework is given to the beneficiary Member States and not to the regions. There is no provision for a regional breakdown of the assistance granted and, in accordance with the principle of subsidiarity, it is the responsibility of the national government to select the applications for financing the projects which are to be implemented in the regions concerned and, after their approval by the Commission, to transfer the appropriations to the said regions.

32. Although the centralised administrative structures responsible for the CF files facilitate the implementation of the application, monitoring and control procedures, there may be differences between the treatment of the projects managed at central and those managed at regional level. Thus, in Spain, following a decision by the national authorities, the rate of Community co-financing for projects submitted by central government is set at 85 %, whereas those managed by the regional administrations only receive 80 %.

Slowness of the approval procedures

33. In accordance with Article 10 of the Regulation, the Commission must take a decision concerning the granting of CF assistance within three months of receipt of the application submitted by the Member State. In practice this deadline is not always adhered to by the Commission, as shown by Table 4, which highlights the longest delays found.

Table 4

Time taken for approval of project applications at the Commission

>TABLE>

34. Analysis of the system shows that a project application must be examined by the Ministry responsible for the environment or by the Ministry responsible for the transport infrastructure, as appropriate, and then sent to the national office responsible for the CF, which, in turn, forwards it to the Commission after final verification, so that an individual decision of approval can be taken by the Commission. The various stages of the national application procedure are not subject to any intermediate decisional criteria that could be applied at an intermediate stage and so ensure that all the conditions for CF financing have been fulfilled. At no time during the national procedure, therefore, are the responsible national, sectoral or regional offices in a position to verify the facts submitted in the project application with regard to their possible financing by the Community.

35. At the national level, the duration of the examination of each application is, in general, quite short on account, in particular, of the following aspects:

(a) progress from one level to another, as described in the preceding paragraph, is all the more rapid because no responsibility has been delegated by the Commission to the Member State concerned to enable it to approve, even partially, a project application;

(b) on-the-spot verifications of the constituent parts of the project can be carried out rapidly within the national framework;

(c) since the date on which the project application is forwarded by the Member State concerned to the Commission is taken as the mark for the beginning of the period of eligibility, it is in the Member State's interest to speed up the entry of the application.

36. At Commission level every project application is subjected to a thorough examination which takes into account the essential elements on which the decision is to be based and the evaluation carried out, in the majority of cases, by the EIB (see paragraphs 14 and 15).

37. The period of time found to have elapsed between the date on which the Member State concerned forwarded the project application and the date of its approval by the Commission may be more than three years and requests for additional information are often the explanation. The likely effect of these approval times, the causes of which may also lie with the Member States, is uncertainty on the part of the national administrators both in the project's initial approval stage and in respect of any amendments which may be required.

The definition of project and project financing

38. The CF may contribute to the financing of projects or stages of projects which are technically and financially independent or groups of projects linked to a visible strategy which form a coherent whole. In any event, the total cost of a project or group of projects may not, in principle, be less than EUR 10 million. This definition of a project has met with difficulties with regard to its interpretation and application, especially in the field of the environment where the projects are traditionally on a smaller scale and sometimes more complex in conception than those in the field of transport infrastructure. In the field of the environment, the Commission gave general approval to groups of projects in Spain and Greece but did not always first identify all the components of the subprojects. These components, and sometimes the more specific provisions linked to the project, such as the timetable or the estimated costs as well, subsequently had to be clarified by the national authorities responsible.

39. In Ireland, some projects (Dublin ring Road, N3-N1 section of the Northern Cross, Belview harbour, etc.) had already received Community finance, within the Structural Funds framework, for preliminary studies, work or previously completed parts of projects, before receiving CF assistance in the field of infrastructure, even though the project or stage of the project was not clearly identified. The Irish Belview harbour project was split into four independent projects, each of them below the EUR 10 million threshold stipulated in Article 10(3) of the Regulation and each of them therefore required a separate cost-benefit study whereas an ex ante evaluation covering the four projects would have been appropriate.

40. In Greece and Portugal certain projects first received Community financing from the Structural Funds, especially Envireg, then received CF assistance even though it was not possible to identify the stages of the project concerned clearly on the financial and technical level. Often the expenditure co-financed by Envireg and by the CF was based on the same contract and the same schedule for its physical and financial completion had been laid down in the specifications, which made no distinction between the sources of Community finance.

41. There is no working definition of the terms "project" and "stages of a project which are technically and financially independent" as stipulated in Article A of Annex II to the CF Regulation and, in those cases where there is a definition, it was merely drawn up by the Member States concerned according to their own interpretation, because the Commission had not given precise instructions or guidelines on this point. This lack of a working definition of the term project mentioned in Article 1(3) of the CF Regulation sometimes results in the CF financing whole categories of expenditure, individual parts of projects or a combination of heterogeneous projects, and this makes any subsequent evaluation of the impact difficult.

Calculation of estimates of net receipts

42. The Court's checks on the expenditure included in the basis for assessment of the amount eligible for financing under the CF revealed that no methodological framework existed which would permit a standardised calculation, in particular for similar projects, of the estimates of net receipts which must be deducted from the total expenditure eligible for co-financing by the CF.

43. In Ireland, the calculation of net receipts may depend on the intended use of the project: if it is a project for commercial use, the amount of revenue will be estimated. If, on the other hand, it is destined for use by private households or for public use, the potential receipts relating to it are not taken into consideration. For example, the CF financing for the construction of a motorway should be revised if a toll were subsequently introduced. Likewise, a project concerning the treatment of waste water for the benefit of local populations rarely includes identifiable receipts, while the same project for the benefit of commercial companies or factories will receive less finance from the CF after allowance has been made for the possibility of identifying the receipts generated.

44. In Portugal a revenue-generating project in the field of transport was based on a somewhat limited cost-benefit analysis presented by the promoter. Furthermore, at the time when the project was approved, the Commission concluded, without conducting a detailed financial analysis, that the traffic generated by the extensions of the motorway in question and the revenue arising from it would be only marginal and difficult to quantify.

45. In the case of the "Pathe motorway - sections A and B - Kastro-Martino" project in Greece, net receipts were estimated at 34 % of the cost of the investment, which includes the interest on an EIB loan and is based on a flat amount of GRD 12 million (approximately EUR 36400) per kilometre for running costs, without any justifying documentation.

46. The choice of which elements of the financial analysis are used to calculate the net revenue therefore varies according to the persons in charge. Some will apply a longer estimated depreciation period than others for a project of the same kind, while others will include the estimated banking charges as additional cash costs and will not take Community financing into account, which would have the effect of reducing the expenses of the project. The technical presentation of similar projects also varies according to the managers. In this respect the Commission has not yet drawn up guidelines highlighting the best practice in this area with a view to encouraging a certain degree of harmonisation and standardisation of the technical and financial analysis.

47. Even if a project is in the public sector, it must show a balance between financial inputs and outputs for each year, otherwise it risks finding itself short of cash. Few of the applications for assistance and implementation reports submit a financial plan with estimates of receipts and expenditure per year. This lack of information makes the calculation of the working capital(4) and of the related working capital requirement(5) difficult and makes it impossible to verify the financial viability of the project.

Approval of projects already virtually completed

48. The inspections demonstrated that, in certain cases, at the time of the Commission's approval the project was already virtually finished. The submission and approval of projects in the course of completion, or completed, is not consistent with the principle of partnership. This principle involves consultation between the Commission and the authorities responsible at national, regional and local level, which should cover the entire course of the project, or stage of project, including its planning, implementation, any amendments, the completion of the work and its becoming operational. In addition, the Court has already drawn attention to(6) the various repercussions of the two approaches which consist in submitting either new projects or projects in the course of completion; in the first case the time required to complete the projects is greater whereas the second approach allows Community financing to be put to use more rapidly; the two approaches also have different effects in respect of the impact of Community financing, which is less direct in the case of projects already being implemented or completed.

Conditionality

49. In contrast to Structural Fund aid, financial assistance from the CF is subject to a conditionality clause. Article 6 of the CF Regulation 1164/94 of 16 May 1994 makes the co-financing of new projects or of new stages of projects in countries with excessive deficits dependent on the achievement of targets set by Council recommendations in compliance with Article 104C of the Treaty. Application of this principle of conditionality resulted in an examination by the Commission of the budgetary position of CF beneficiary Member States which had excessive deficits. Each of these examinations (in spring in respect of the preceding year, in autumn for the current year) showed that the annual target recommended by the Council in respect of the budget deficit had been achieved. As a result, co-financing from the Cohesion Fund was continued.

50. In the short term, however, conditionality can make the achievement by the less prosperous Member States of the various macroeconomic objectives difficult. On the one hand, these countries are committed to substantial government investment, as shown in Table 5, in order to reduce the lag in their development and to increase their capacity for growth and prosperity. On the other hand, they must, at the same time, reduce their budget deficit and retain control over their government debt.

Table 5

Public investment expenditure

(Gross fixed capital formation by general government as a percentage of GDP)

>TABLE>

Project development cycle

51. The development of a project can be seen in terms of stages of a cycle and this allows tighter control over its management. In CF projects, the Commission is mainly involved at the beginning and end of the cycle. At the beginning, during the phase leading up to the approval decision, the Commission must evaluate a certain number of factors that are internal (planning, cost estimates, IRR, feasibility, performance forecasts, etc.) or external (cost-benefit analysis, desired global strategy, synergies with other projects, national or regional priorities, location, etc., to the project). At the end of the cycle the Commission must ensure that the project has been closed, approve the final implementation report and release the final balance.

52. Between these two stages a number of events may substantially alter a project's development (unforeseen events, budget overruns, delays, incomplete initial design, unrealistic planning, etc.). These lead to amendments to the initial decision without the possibility of the Commission's intervening on a precautionary basis to correct any change of direction. The Commission finds itself obliged to accept developments that have taken place at the investment level, whereas it could have detected them at an earlier stage if it had had a better understanding of the factors influencing project costs and had subjected them to adequate controls. In addition, apart from their consequences for the extent of the outcome and impact of projects, the budget and timetable overruns ratified by the Commission absorb appropriations to the detriment of approvals of new projects. Furthermore, once the Commission has decided to approve a project, subsequent amendments to its decision are often accepted without the initial rigour and without the necessary in-depth analysis of the causes which triggered them. Examination of the documents justifying the amendments that had been requested and adopted did not always enable the factors that brought them about to be identified, or the set of solutions found.

53. Numerous amendments were found in several cases, and the projects concerned nevertheless achieved the same objectives as those initially approved. For a project at Argos in Greece, for example, the initial Commission decision made provision for a total cost of EUR 751000 for the laying of 20 km of water mains, i.e. an average cost per metre of EUR 38. A subsequent amendment to the initial decision introduced an additional investment of EUR 130000 for the construction of a reservoir and reduced the mains to 10 km, i.e. an average cost per metre of EUR 62; a third amendment to the decision increased the length of the mains to 72 km at a cost of EUR 993000, which took the average cost to EUR 14 and, finally, the cost of the investment made for the mains came to EUR 1377 000, which brought the average cost to EUR 19 per metre.

54. At the monitoring level, the implementation reports on CF projects, which are attached to payment applications, usually provide very little information on the state of progress, the synergies or the aspect of coordination between these projects and other structural operations, even when this information is available in the Member States concerned.

55. In respect of the approval of the final reports, which allows the final balance to be paid and the projects closed, there are also delays in the Commission departments but they are significantly shorter than those occurring in the initial phase of projects. The time between the submission of the final report and the related payment can be as much as approximately one year. The closure procedure sometimes produces several versions of the final implementation reports, as well as several successive applications for final payment, but the reasons for this are not immediately apparent from the case files.

Monitoring committees

56. Provision for the monitoring committees is made in Article F of Annex II to the Regulation establishing the CF, and they were set up as of 1994. Representatives of the Commission, the EIB, the responsible national authorities and, sometimes, of regional or local authorities, attend the meetings. They use physical and financial indicators to examine the state of progress of the current projects, and check whether the financing plans are adequate to ensure the projects' continued physical progress. Where there are significant divergences in relation to the initial estimates they propose amendments to the Commission. In addition, the decisions to grant aid are taken outside the monitoring committees and the latter are not all informed of the situation of the project applications under examination at the Commission.

Controls carried out by the Commission

57. The main objective of the missions carried out by the Regional Policy Directorate-General, which is responsible for management of the CF under Article 13 of, and Article F, Annex II, to Regulation (EC) No 1164/94, was to establish contact with the national and regional authorities responsible for the submission and implementation of projects and to discuss unresolved questions concerning certain projects. These missions, however, do not deal with these matters in sufficient depth to evaluate the implementation of the various projects, their physical execution or whether the observations made previously are actually taken into account. The checks carried out in 1996 on the management and internal control systems currently applied in the Member States, which revealed certain weaknesses, were followed up with only a limited number of checks on relatively large projects and, in these cases, on the financial control systems concerned, the declarations of revenue and physical implementation.

58. The audits carried out by the Financial Control Directorate-General, on the other hand, reveal a certain number of irregularities, as shown in Tables 6 and 7, some of which were subsequently rectified. The classification of the types of irregularities detected (see Table 6) was also confirmed by the Court's audit visits. These audit findings show that there are still weaknesses in the internal control systems. The responsibility for guaranteeing that Community funds for CF projects are used for the purposes intended falls, in the main, to the Commission. The Commission, however, does not devote sufficient resources to audits of CF projects for it to be in a position to meet this obligation in full.

Table 6

Types of irregularity detected by Commission checks in the Cohesion Fund context

>TABLE>

Table 7

Incomplete list of irregularities detected by the Commission during audit visits in connection with the Cohesion Fund

>TABLE>

OBSERVATIONS RELATING TO THE FINANCIAL MANAGEMENT OF THE PROJECTS EXAMINED ON THE SPOT

59. The Commission, albeit belatedly, in its concern for transparency and clarification of the methods used to evaluate statements of expenditure, informed the beneficiary Member States on 19 May 1998 of the principles governing the eligibility of expenditure within the framework of the projects supported by the CF. These are based on the rules set out in the forms on eligibility for Structural Funds assistance, but make allowance for the special features of the CF. Even if they have not led to new rules, these principles have made the eligibility criteria on which expenditure declarations are based more transparent. The late notification of these eligibility criteria may explain some of the anomalies found during the audit visits.

Eligibility of expenditure

60. In Greece, in the case of a project in Argos the expenditure incurred by the beneficiary enterprise for its own account was not documented. The recoverable VAT, to an amount of GRD 145 million (around EUR 433000), was considered to be eligible expenditure in a project at Thessaloniki. The qualifying expenditure included an amount of GRD 87 million (around EUR 260000) for a number of expenditure items not provided for in the decision approving a project at Nafplio. In the case of the project to improve access to Athens airport, expenditure totalling EUR 765000 incurred after the deadline laid down in the initial project decision should be considered ineligible.

61. In Ireland, within the framework of a water-supply project in Limerick, expenditure totalling IEP 395324, 51 (approximately EUR 494155) incurred respectively before and after the period of eligibility from 1981 to 1992, was declared to the Commission as eligible expenditure.

62. In Spain, recoverable VAT expenditure in connection with a water-supply project in Palma de Mallorca, amounting to approximately ESP 101 million (approximately EUR 0,6 million), was included in the total of expenditure eligible for CF financing. For another project in the Balearic Islands supplementary work totalling ESP 13,2 million (approximately EUR 77500) in connection with the inclusion of sand-removing equipment was added to the expenditure eligible for CF financing, although no provision had been made for it in the Commission's decision nor in any subsequent amendment submitted to the Commission by the Member State. The amount involved in this additional work had no repercussions on Community financing.

Finalisation of projects

63. In Greece, the final technical report on a project in Argos stated that 72 km of old water mains had been replaced, whereas the on-the-spot check revealed a smaller quantity, i.e. 60 km. For another project involving a water pumping station the final report mentioned its completion, although on-the-spot checks revealed that it was neither completed nor operational.

64. In Ireland, the results of the technical controls carried out by the units responsible were not checked against the statement of financial implementation. There was no systematic reconciliation to verify that the pattern of eligible expenditure was matched by corresponding progress in physical implementation.

65. In Spain, in the case of a group of projects concerning reafforestation on the Balearic Islands, the eligible costs approved by the Commission in its decision of 28 July 1997(7) amounted to EUR 3,68 million, which is significantly below the EUR 10 million threshold mentioned in Article 10(3) of the CF regulation. In the case of another project concerning sewage treatment plants on the Balearic Islands, the on-the-spot audit revealed that the work was already well advanced at the time of the application for assistance dated 11 April 1995 and that two of the five plants entered had already been completed on the date of the Commission's approval of financing for them.

66. In Portugal, a project completed and closed in the transport infrastructure sector still revealed constructional faults.

Adequate accounting

67. In Ireland, expenditure declared eligible within the framework of the CF is often expenditure for which provision has been made or which has been committed but not yet paid. This situation is then regularised in subsequent declarations, which makes later reconciliations of the accounts difficult. There is not always a perfect match, therefore, between the expenditure really incurred and paid by the final beneficiary and the expenditure declared to the Commission for co-financing. These discrepancies when encountered during audits make identification of the audit trail difficult.

68. In Greece, during checks on a project in the transport sector, not all the eligible expenditure declared could be verified on the basis of the relevant supporting accounting documents.

69. In Portugal, in the case of a project in the environment sector co-financed by Envireg and the CF, it was not possible, during the on-the-spot check, to clearly distinguish between expenditure covered by the one or the other intervention; the was also the case in respect of the physical work.

70. In Spain, for a VTS (vessel traffic system) project in Tenerife, no documentary evidence could be found, on the spot, to support the issue, by the financial beneficiary, of a payment order for an amount of ESP 17698662 (approximately EUR 100000) there in spite of the fact that it was entered in the accounts by the General Directorate for Economic Planning at the Ministry of Public Works.

EVALUATION IN THE COHESION FUND DOMAIN

Macroeconomic models

Models developed at the Commission's instigation

Reference framework and contract management

71. The Regulation establishing the CF provides that the Commission is to report regularly on the economic and social impact of the CF in beneficiary Member States. This provision implies that the project-based approach is to be supplemented by a more general analysis of the impact of CF-financed projects on certain variables, such as growth, employment and trade, in the economies concerned. This analysis should take into account the short-term incidence of the measures in question on demand, which is felt mainly during the execution phase, and the medium and long-term effects on supply, which first become apparent during the operational phase.

72. In order to be able to satisfy these requirements of the Regulation, the Commission signed two contracts with a consultant. The first, dated 3 July 1995, for an amount of EUR 675965, was for the development of methods of evaluating the macroeconomic impact of CF projects. The second, dated 1 April 1997, for an amount of EUR 283500, essentially concerned the collection of new data to allow new project evaluations.

73. The specifications for the first contract were based on a feasibility study carried out in 1994 by the same consultant on the same subject and took over the same proposed structure and the same evaluation criteria. The Commission did not consider that this unusual situation posed any problems. Most of the modelling approaches proposed by the various applicants in connection with the call for tenders were already familiar and were all considered by the Commission to be suitable for evaluating project impact, as the Commission had not adopted any special criteria for appraising these approaches. In the end the Commission took the consultant who submitted one of the highest bids and whose approach excelled by virtue of its originality, but had still not been tested in practice as far as its applicability was concerned. A subsequent modification of the first contract is signed, but not dated, and extends by three months the time allowed for performance, so extending the total duration to 15 months. This extension devalues de facto the initial criteria for the awarding of the contract, given that the tender selection procedure had included consideration of tenderers' ability to complete the study within the time allowed and, in addition, there was no provision for penalties in case of overruns. This should have led to an amendment of the contract.

74. The second contract signed by the Commission was in response to a proposal from the same consultant, suggesting that the use of the models created under the first contract should be extended to all the CF measures, so that the consultant would be able to estimate the short, medium and long-term impact of projects on key economic variables such as income, regional growth, private investment and employment.

75. This extension to the contract ought not to have been necessary, since the first contract had already provided that the Commission was to be supplied with an adequate, operational tool (models and database), so that it could ascertain the impact of the CF for itself. In fact, the contracts specify on each occasion that all the results obtained by the contractor, as well as the software needed for the application and the database created for that purpose, are the property of the Commission.

76. The expenditure associated with the first contract amounted to EUR 674403,98. Examination of the service providers' travel expenses showed that in many cases the Commission refunded the cost of travel by air in business class, whereas Annex IV to the contract stipulated in Article 1(a) that travel expenses would be refunded up to the price of economy-class tickets. In the case of the second contract, which was signed on 1 April 1997 and set a time limit of 31 March 2000 for performance, there were no payments until 31 January 2000.

Conformity with the evaluation objectives

77. In order to establish a coherent framework for analysing the impact of CF expenditure in the four countries concerned the consultant referred to in paragraph 72 adopted an innovative approach, using three complementary methods vector autoregression models (VAR), regional computable general equilibrium models (CGE) and models of explicit distribution dynamics (MEDD). The models are described in more detail in the Annex.

78. The analysis required by the models was, amongst other things, to provide answers to the following three questions. How do the measures assisted under the Fund, together or individually, influence the economies of the four cohesion countries? What effect does this have on adjacent economies and on the Community in general? What is the impact of the measures on the economic and social cohesion of the Union?

79. The MDRF models were not, in fact, used to evaluate the impact of CF projects directly. They are concerned with the dynamics of growth and convergence of all the regions of the four cohesion countries, as well as of other countries like France, the United Kingdom and Italy, and are used to examine questions such as the tendency of regional incomes to converge or diverge over time and the dynamic in the regional distribution of income. The analyses apply to the period 1980 to 1990. They show that the increasing inequality can be explained by the fact that the growth achieved by groups of neighbouring regions is different from that of other adjacent groups and that the performance of one region can be explained by neighbouring regions. Furthermore, the results concerning spillover from public and private investment could be taken into account for an appraisal of the CF projects carried out in the period 1993 to 1999.

80. The aim of the VAR and CGE models was to evaluate the impacts of individual projects from a number of viewpoints, including the effects on the region directly concerned, the spillover onto other regions and the whole economy and the impact on cohesion and convergence. These models can provide additional answers and the results which they give, when they are comparable, are usually consistent.

81. A more complete application of these models might have been useful. For example, VAR2, which is more complete than VAR1 from the modelling viewpoint, was applied only to the Spanish regions, because they were able to provide more, detailed information in terms of historical series. Likewise, the CGE model, which takes a large regional breakdown into account, could be used to analyse the implications of the CF for regional inequality in the cohesion countries. These models have their limitations as regards evaluating the impact of environmental projects and are more suitable for transport infrastructure projects. Furthermore, the selection of projects analysed by means of these models gave preference to transport infrastructure projects.

82. Even though these models (VAR, CGE and MEDD) provide useful and interesting results about the potential impact of CF intervention and thus help towards a better understanding of the related effects, there is no real coordination between the models used. Furthermore, none of the models specifies, for the results obtained, the level of imprecision that is due to the poor quality or availability of the data collected or to the limits of the extrapolation methods used. The Commission did not carry out any reconciliation of the results obtained from these models and did not carry out the necessary harmonisation of them in order to establish overall conclusions in terms of the CF's impact on the economies of the four cohesion countries. The results produced by each model separately are generally positive. It is recommended that the Commission refine this resource for evaluating projects macroeconomically and that it apply it more fully.

Models developed on Member States' initiative

83. Only Spain has used CF financial support to carry out a study of the overall macroeconomic impact of the CF on the national economy. The approach that was adopted for this study, which was presented in November 1995, was to use three models (Input/output, Moisees and Hermin) (see the Annex).

84. Portugal carried out specific impact assessments for some major projects, such as the new bridge over the Tagus, using what is essentially an input/output model. The other cohesion countries have not produced any special studies at national or regional level on the macroeconomic impact of projects, or of the CF itself.

Cost-benefit analysis (CBA)

85. When the projects submitted by the Member States concerned are being examined, cost-benefit analysis is one of the main criteria for the decision on whether to grant Community assistance. The Commission can, if necessary, call on the European Investment Bank to assist in the project evaluation (see paragraph 15). In the field of transport infrastructure the CBAs are usually based on long experience and a well-established methodology which allows, amongst other things, quantification of all the necessary parameters.

86. In the environment field, on the other hand, the difficulty of quantifying the expected benefits has often led Member States to use methods of a more qualitative and less quantitative nature.

87. If the Commission finds the CBA insufficient or unconvincing it may ask the Member States to provide additional information or to extend the analysis, and this may lead to a redefinition of the projects in question or, even, ultimately to a refusal of Community funding for certain projects. For example, projects such as the Gran Porto drinking water supply or the Enxoé (Portugal) dam were rescaled following the extension of the CBA.

88. Examination of the CBAs showed that despite the Commission's efforts in the matter (notably the publication in June 1997 of a guide to the analysis of the costs and benefits of large-scale projects within the Community regional framework), the treatment and calculation of socioeconomic variables can differ widely between promoters within the same sectoral field and the same Member State.

89. Basic CBA concepts (net present value, internal rate of return, discount rate, constant or current prices, tax aspects, sensitivity analysis(8), etc.) are not always understood in the same way by the parties concerned. External factors are often insufficiently evaluated and quantified. Evaluating the feasibility, risks or alternative options is rarely considered. The socioeconomic costs or benefits are sometimes incomplete and sometimes only the financial aspects are taken into account. Moreover the CBAs do not always include precise and detailed information that would make it possible to verify, amongst other things, their reliability.

90. Thus in the case of the Lisbon water-distribution network project it was not possible to establish any link between the revenue received by the final beneficiary over all the networks and the investment financed by the CF. For the drinking-water treatment plants, the Ibiza-desalination plant project, the price of water was calculated in 1994 as 95 ESP/m3 (approximately 0,48 EUR/m3) in the CBA, whereas the real price was, for 1997, 213,84 ESP/m3 (approximately 1,07 EUR/m3). The CBA for the Limerick City Water Supply Project did not take into account private-sector revenue, which amounted to around IEP 1,2 million (approximately EUR 1,6 million) in 1997. For the same project the IRT was calculated on the basis of a 25-year period, whereas the lifetime of the associated buildings was assumed to be 50 years. Only the financial aspects were considered in the CBA of the Tralee landfill site project, which ignored the socioeconomic aspects. The Belview rail crossing project does not contain an IRT calculation and states only that economic benefits are expected in terms of time saving, without quantifying them.

91. The CBA for the Autoroute Pathe-F-Pont project estimated an investment cost of around EUR 4 million, whereas the construction contract signed was for an amount of some EUR 2,3 million.

92. This disparity between design and interpretation in relation to the application of CBAs tends to produce results which differ according to the hypotheses adopted and may distort the appraisal of the benefits or drawbacks of a given project relative to others. The Commission should be careful to draw up more precise guidelines on this point in order to harmonise the methods applied, particularly as regards the essential information that must be submitted with CBAs so that their reliability can be verified and they are comparable.

Indicators

93. The indicators submitted in support of CF projects by national authorities are very often the same ones that are used for the ERDF. Some of these indicators from the ERDF are part of a general system of measures that are programme-based, located in a specific context relative to an overall objective and are determined by a number of specific objectives that give rise to operational objectives within a cohesive set of priorities as part of the implementation of multiannual measures. The indicators were not always improved and made more detailed in order to adapt them to the particular nature of CF projects, especially in the context of the two sectoral fields in question (transport infrastructure and environment).

94. Furthermore, the indicators that are used in the CF framework are sometimes too general, or lack the detail required for proper verification of a project's progress, or the quantification of its impact and performance. Some of the cohesion Member States, Portugal, for example, have, on their own initiative, developed and introduced quarterly or six-monthly monitoring forms setting out the data that is to be collected in order to provide the requisite information on progress, physical implementation and impact, together with the corresponding indicators. These attempts to improve the indicators within the CF framework at national level are the work of a few responsible authorities and are not harmonised, nor are they generally used in the cohesion countries.

95. The Commission should develop more relevant indicators that specifically apply to CF sectoral projects or it should improve them, without making the monitoring method unnecessarily cumbersome, so that it is possible to represent the improvement of the sector concerned quantitatively, to promote sustainable development and to make sector-wide comparisons.

Evaluation of closed projects

Commission

96. When the projects are closed the Commission and beneficiary Member States must, in accordance with Article 13 of the Regulation establishing the CF, evaluate the manner in which the projects have been carried out and must also evaluate the potential and real impact of their implementation in order to assess whether the initial objectives can be, or have been, achieved. The Commission has accordingly selected two external consultants on the basis of calls for tenders and these ex postevaluations of individual closed projects, numbering around 120 projects evenly distributed over the two sectoral domains in question, are to be carried out over a period of three years. The ex post evaluations that have already been completed are proving to be very useful, especially for improving the definition of the performance indicators in the two sectoral domains during implementation of the next CF generation.

Member States

97. No Member State has yet completed the ex post evaluation of closed projects. Portugal has essentially drawn up a first account of jobs created, project by project, in 1997 and has begun an ex post evaluation process, while in Greece an evaluation of the management of projects and the quality of the work took place in 1995 to 1996.

98. Although it is not a question of ex post evaluations, approximately 100 closed projects were examined in the cohesion countries, on the Commission's initiative, with the aim of evaluating their impact on the environment.

IMPACT OF THE COHESION FUND

Qualitative impact

99. Analysis of the trends in the four cohesion countries shows encouraging results during the period (1994 to 1999) in transport infrastructure and in the environment, the fields of CF intervention where there has been notable progress towards reducing disparities relative to the rest of the Union.

100. In Greece the number of towns covered by water-treatment systems doubled between 1993 and 1999, and by the latter year 71 % of the population's requirements were covered. In Ireland the proportion of urban waste water treated in accordance with EU standards rose from 20 % in 1993 to 80 % in 1999. In Portugal the percentage of the population connected to a drinking-water supply system rose from 61 % in 1989 to 95 % in 1999 and the population having access to a sewerage system rose from 55 % in 1990 to 90 % in 1999. There has also been a substantial effort to assist the development of strategic road networks in the cohesion countries, mainly with the aim of reducing the impact of traffic in residential areas, lowering the levels of noise and pollution, reducing user transit times and improving road safety.

101. The numerous projects audited show that the CF is involved in a very wide range of economic activities. For example, the VTS maritime traffic surveillance and control project (CF assistance EUR 8,3 million) in Spain provides, amongst other things, improved accident and marine pollution prevention. In Greece the modernisation of drinking water supply systems and sewers, particularly in Athens (CF assistance EUR 125 million), has improved the quality of life for the inhabitants, as well as encouraging the development of tourist facilities. In Ireland the extension and modernisation of transport infrastructure, e.g. the Dublin ring (CF assistance EUR 20 million), have, as a general rule, brought increased safety, cut journey times and facilitated links between Ireland, France and the United Kingdom. In Portugal the Vasco da Gama bridge over the Tagus, which cost a total of EUR 603 million, received EUR 311 million of CF assistance and EIB loans of PTE 41500 million (approximately EUR 207 million) and DEM 24,8 million (approximately EUR 13 million). It is one of the largest civil engineering projects to be undertaken in Europe in recent years. The improvement of water supplies in the Algarve (CF assistance EUR 45 million) has helped to satisfy the increasing demand for drinking water brought about by tourist development, as well as preventing over-depletion of the water-table.

102. The first report on cohesion in 1996(9) also highlighted the positive role of the CF as a means of fostering increased economic and social cohesion within the EU.

Quantitative evaluation

103. Macroeconomic impact evaluation results were obtained using the models mentioned in paragraph 77 (see Table 8), as regards the impact of the projects on the region concerned, on the economy of the Member State and on cohesion and convergence.

Table 8

Main results according to the VAR1, VAR2 and EGC models

Main results according to VAR1 model

>TABLE>

Impact of transport infrastructure projects in Spain for period 1993 to 1999 according to VAR2

>TABLE>

Short, medium and long-term effects of Cohesion Fund projects (CGE models)

>TABLE>

Impact of projects on the region concerned

104. The positive effects in terms of job creation and unemployment, output and real incomes must be emphasised. The gains in real income that were shown by the CGE model were sometimes substantially lower than those estimated by means of the VAR models (Madrid ring road, Rias Bajas motorway). As regards the investment of private capital, it may be that, depending on the model used, the implementation of the CF project leads to a reduction or an increase in such investment (Rias Bajas motorway in Spain).

Impact of projects on other regions and on the Member State's economy

105. For individual projects the VAR2 model, which was applied only in Spain, gives the same type of effect for other regions and the whole economy as for the region concerned, in terms of job creation and unemployment, output, real income and investment of private capital. The medium and long-term effects of individual projects, which are estimated by means of the CGE models for the whole economy of the Member State, are also found to be positive. The finding made in paragraph 104 about the consistency of the VAR and CGE models as regards gains in real income also applies as regards the whole economy.

106. The following are the main conclusions of the evaluation of the overall impact of the CF on the Spanish economy, using the models mentioned in paragraph 83:

(a) the Leontief input/output model highlights a positive impact of CF investment of around 0,3 % on gross annual added value and states that one job/year is created for ESP 4,23 million (approximately EUR 0,02 million) invested in CF projects, which would be equivalent, on average, to the creation of 40000 jobs per year;

(b) the Moisees model in turn shows that investment within the CF framework could generate an annual increase in GDP of around 0,1 % which would create the conditions for the creation of approximately 20000 jobs per year;

(c) the Hermin model shows an annual increase in GDP of around 1 % and estimates that in the framework of CF investment, job creation could amount to some 130000 jobs per year.

Impact of projects on cohesion and convergence

107. The VAR1 model does not permit direct calculation of the extent to which CF projects reduce or accentuate regional inequalities in the cohesion countries. In particular it shows that in Spain increasing public investment leads to a strong expansion in private investment in the richest and the poorest regions, whilst the intermediate regions experience less growth. The effects on employment, on the other hand, are concentrated in the richest regions and the intermediate regions. In Greece, the gains in private investment are more substantial in cases where the projects concern the poorest regions, while in Portugal government investment in the richest regions gives rise to increased private investment.

108. The VAR2 model contains the information needed for analysing the CF projects' contribution to regional cohesion in Spain. The results in terms of output, employment and unemployment at the level of the regions and the whole economy can be used to estimate regional convergence at the level of GDP and employment. However, that was not done. A simulation for a theoretical project leads to the conclusion that the additional output generated by one euro is higher in the rich regions than in the poor regions, mainly due to inter-regional differences at the level of the capital invested/output ratio.

CONCLUSIONS

109. The CF is a financial instrument designed principally, according to the Treaty, to promote economic and social cohesion. CF assistance has undoubtedly had very positive effects in the beneficiary Member States. It has improved the implementation of the environmental directives and, in these countries, has improved the situation in respect of the trans-European networks and access to them.

110. However, in spite of these positive results, the Commission should improve certain aspects of this framework, with a view to greater efficiency: the financial management should be better adapted to programming which encourages medium-term projects; the Commission's control over the project cycle should be improved; the term "project" requires a more precise definition; the approval procedures should be simplified; programming and monitoring should be more rigorous; impact evaluation more exhaustive, and the partnership with the beneficiary Member States further developed.

111. As regards the financial management of the CF projects, the Commission should exercise stricter control over the build-up of amounts outstanding and awaiting clearance against commitments from previous financial years, thus ensuring that these amounts outstanding do not exceed the amount of payment appropriations entered in the annual budget against the commitments for the financial year, so that advance payments to the new projects are not hindered in the course of the financial year.

112. Management of the project cycle should receive greater attention at the stages where it is being implemented in partnership with the Member States. At present, management is monitored by the Commission mainly at the beginning and the end of the cycle.

113. With regard to the terms "project", "stages of projects" and "group of projects" the Commission should be careful to give an operational definition of them so that they can be applied in a uniform manner, without ambiguity, in all the cohesion countries.

114. In order to speed up the approval procedure for some projects, the Commission should set up simpler procedures, for both the preliminary stages (project application) and the final stages (final report) of the decisional framework. In particular it should define its decisional criteria more precisely, so that they can be applied and verified at the level of each of the administrations in the Member State concerned.

115. Even though the CF has a project-based approach and analyses the advantages and drawbacks of each project separately, the set of projects approved by the Commission should, taken as a whole, be part of a coherent, medium-term concept. An underlying planning philosophy of this kind would give greater transparency to the CF's coordination, synergy and complementarity with the other Community instruments and programmes, especially with the ERDF, and make for more harmonious economic and social development.

116. The Commission should increase its controls in the context of the CF as they have proved to be inadequate in this field.

117. As regards impact, a qualitative evaluation of the projects audited shows that the CF contributes to the strengthening of economic and social cohesion. Some of the cohesion countries, for example, have experienced very high growth rates. At the macroeconomic level, the "project" dimension makes complete global evaluation of the CF difficult, even though it has been possible to identify positive results in the case of certain large-scale projects. The Commission should perfect the instrument for the macroeconomic evaluation of projects and apply it more fully.

This report was adopted by the Court of Auditors in Luxembourg at the Court meeting of 6 July 2000.

For the Court of Auditors

Jan O. Karlsson

President

(1) Council Regulation (EC) No 1164/94 of 16 May 1994 (OJ L 130, 25.5.1994).

(2) In a communication addressed to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions in 1995 (COM(95) 509 final), the Commission clarified its position on this balance. The Commission considers that a 50/50 distribution must be sought, in partnership with the Member States, between environmental and transport projects for the whole of the period. This objective must not, however, be seen in too rigid a light, as a certain flexibility must be maintained in order to make allowance for special situations.

(3) A Special Report on the Cohesion Financial Instrument, published in OJ C 59 of 8 March.1995, has already reported numerous shortcomings in the implementation of this instrument.

(4) Working capital (FR) is used to assess the risk of failure and to tell whether an enterprise can meet its obligations. It can be calculated in two ways:

FR = short-term current assets (less than one year) - short-term current liabilities (less than one year);

or

FR = fixed capital - long-term assets (more than one year).

It serves to show the extent to which the resources that fund the assets remain available to the enterprise over a term at least equal to that of the asset funded.

(5) Working capital requirement (BFR) is usually calculated as follows:

BFR = current assets - current liabilities, where:

current assets include inventories, debtors and other short-term assets

current liabilities include creditors and other short-term liabilities.

If the cash position is positive and the position is favourable, FR covers BFR. If the cash position is negative, BFR exceeds FR.

(6) The Annual Report concerning financial year 1998, paragraphs 3.50 and 3.51.

(7) C(97) 2582 final, project FC 961161042.

(8) Sensitivity analysis: study of the possible effect of a change in crucial estimates relating to costs and benefits on the profitability or actual value of a project.

(9) Commission document. Etudes de développement régional - L'impact des politiques structurelles sur la cohésion économique et sociale de l'Union 1989 - 1999.

Annex

PRESENTATION OF MODELS USED TO EVALUATE THE IMPACT OF THE COHESION FUND

Models developed at the Commission's initiative(1)

The autoregressive vector models (VAR)

1. Whereas traditional econometric models basically express each endogenous variable only as a function of exogenous variables and of all the time-shifted values of the endogenous variables, VAR models comprise only endogenous variables and there is an explanatory equation corresponding to each endogenous variable and linking that variable to the past achievement of all the endogenous variables in the model.

2. In the context of the CF, VAR models are used to evaluate the dynamic impact of CF expenditure on transport infrastructure and to study the spillover from one region to another. The reliability of the results obtained is highly dependent on the availability and quality of the data collected in the four cohesion countries.

3. The VAR1 model provides results for the impact of projects on the region concerned in terms of annual increase (as %) in cumulative private investment and in terms of job creation (see Table 8). It seeks to position a region receiving CF assistance in relation to the mean for the country. For this model the equations used do not always take account of information produced by some in order to consider it in the others, contrary to traditional econometric systems. In consequence the information from the first equation of the VAR1 model, which estimates the effects of public investment on private sector investment, are not included when the second equation is calculated (the second equation evaluates the effects of public investment on employment), even though, from an economic viewpoint, job creation can be indirectly stimulated by an increase in private investment brought about by increased public spending. This model is not really suitable for measuring the spillover onto other regions(2), as it does not contain any equations for the production function, increase in regional GDP or spillover to adjacent regions.

4. The VAR2 model makes it possible to calculate the effects in terms of relative changes concerning output, public investment and unemployment and evaluates the job-creation effect in absolute terms, for both the region directly involved in the CF project and the whole economy, and having regard to the possibility that part of the national funds initially earmarked for investment in transport infrastructure and the environment may be assigned to other areas. This model was used only for the Spanish regions (see Table 8), because they were able to provide far more detailed information in terms of historical series. It was used for both transport infrastructure projects and environment projects.

5. It is based on absolute-value variables and is more complete from the modelling point of view. Nevertheless, given that there was no evidence of any significant statistical differentiation concerning the impact of the different types of project in the public sector, the regression coefficients used are the same, whatever the sectoral domain concerned. However, environmental objectives cannot be measured in terms of increased output, private investment or job creation.

Computable general equilibrium models (CGE) on a regional scale

6. These models are based on the principles of general equilibrium theoretical models, namely, all markets are in equilibrium, every agent acts in a way which maximises a specific function and expectations are assumed to be rational.

7. The CGE models developed were used to apprehend the effects of investment in the domain of transport infrastructure on a particular region or on the regions of a country as a whole, taking as starting point the savings in distance and time and, therefore, in transport costs, resulting from road infrastructure projects.

8. The effects break down as follows:

- short-term effects: cost reductions, increased profits for enterprises in the region, lower prices for customers, increase in consumers' real incomes,

- medium and long-term effects:

- enterprise relocation (installation of new enterprises and delocalisation of enterprises in the face of increased competition),

- adjustments in the pattern of employment (net job supply associated with enterprise relocation, higher wages),

- changes in inter-company relations (increased demand for intermediate products, increased profits for the industries producing intermediate products, lower prices for intermediate products, increased profits for enterprises using such products),

- increased mobility of workers between regions, tending towards wage equalisation.

It should be noted that the short-term effects essentially concern the region that is the direct beneficiary, whereas the medium and long-term effects also affect other regions.

9. Three CGE models on a regional scale were developed: one for Spain and Portugal, one for Greece and one for Ireland. The results of calculating the effects for a number of projects are given in Table 8.

10. The development of these models requires a substantial amount of data and the analysis of policies relating to them, both of which are subject to a high degree of regional fragmentation, to ensure that these policies are able to deal with trade between places in a context of imperfect competition. Information relating to Input/output tables, output disaggregated on the regional and sectoral side, added value, employment, or road, rail or air transport is essential for an analysis of this type.

11. As regards the CGE models that were developed, difficulties appeared at the level of information. Any incomplete or missing information gave rise to estimations, approximations or substitutions. For example, technical coefficients from the Spanish input/output tables were used without adjustment for Ireland and for Greece, even though the output structures are markedly different. The degree of sectoral fragmentation in Greece and in Ireland was limited in the CGE models to agriculture, industry and services. Moreover the transport costs used were usually taken from the initial cost-benefit analyses, whose reliability was not tested.

12. This model shows that the medium and long-term effects are virtually identical (interregional mobility of workers is disregarded). As the medium and long-term effects reflect regional interdependence the model proves adequate for the analysis of spillover onto other regions.

13. The CGE model, which takes into account an extensive regional breakdown, could be used to analyse the CF's implications for regional inequality in the cohesion countries. This analysis is carried out only for the changes which affect the income from employment in the different regions, and not for gains in real income, nor for the regional levels of income per inhabitant and unemployment. It is thus not possible to obtain from it any conclusions about convergence between the richest and the poorest regions in terms of GDP and employment/unemployment.

Models of explicit distribution dynamics (MEDD)

14. These models are very innovative in as far as they rely on stochastic distributions and three-dimensional graphic and statistical representations concerning changes in the regional distribution of income and are concerned with the dynamics of growth and convergence of all the regions of the four cohesion countries, as well as of other countries such as France, the United Kingdom and Italy. They are used to examine questions such as the tendency of regional incomes to converge or diverge over time and the dynamics in the regional distribution of income. The analyses carried out were based on data for the period 1980 to 1990.

15. The results obtained with MEDD models show that some regions tend towards increased wealth, others, on the contrary, remain or become poorer. In the cohesion countries it seems that increased convergence runs parallel to the rate of growth of the economy. However, Portugal experienced both a substantial rate of growth during the period under consideration and a significant disparity between regions. Greece and Spain, with a lesser growth rate, over the same period, attained a higher degree of regional cohesion. As regards Ireland, no information was available on this subject.

16. Modelling with the emphasis on the level of regional income in relation to neighbouring regions, and not on the absolute level of regional income, also shows (where the measurement in absolute terms at the level of incomes highlights increasing inequality but the measurement involving relative values, does not demonstrate the same tendency) that increasing regional inequality can be explained by the fact that the growth achieved by groups of neighbouring regions differs from that of other groups of adjacent regions.

17. This model similarly concludes that a region's performance is explained by neighbouring regions. This appears evident for the Spanish regions, whereas for Portugal there continues to be a dynamic of increasing inequality between the regions.

18. Applying these models to data concerning public and private investment shows that, in Spain, public investment is followed by repercussions for neighbouring regions and plays an important part in the furtherance of regional cohesion. In Portugal, public and private investment both play this role. It is necessary to take into account this result of analysis based on data for the period 1980 to 1990 in order to appraise CF projects.

19. The MEDD models could be further refined and would produce more useful results if they were accompanied by explanations based on confirmed statistical techniques, if they were concerned primarily with the impact of CF projects on the trend of income throughout the period in question and, finally, if the period were not confined to the period 1980 to 1990, before the Cohesion Fund existed.

Models used by the Member States

20. The Input/output model seeks to depict the links between all the resources available to an economy and all final consumption. It thus identifies any interdependence at the level of the different sectors of a country that are interrelated on the basis of intermediate consumption.

Moisees (Modelos de Investigación y Simulación de la Economia Espanola) is an econometric model of 150 equations which provides a simulation of the macroeconomic effects of public policies in the medium term (five years) over the whole economy.

The Hermin econometric model takes account of the specific characteristics of each Member State and aggregates the effects at national level, starting from the hypothesis that economies do not operate at full output and, in consequence, the multiplier and accelerator effects associated with increased investment create an expansion of output and employment. The model appears to give satisfactory results in terms of the effects on demand.

21. Use of these macroeconomic models is based on variables and hypotheses that may be incomplete or partial. The Input/output model, for example, only considers the effects of demand and mainly analyses short-term impacts in relation to Community intervention. Moisees does not consider long-term and sectoral effects. Hermin is based, in particular, on complex assumptions about the pattern of national government expenditure that may vary significantly and, in consequence, may lead to variation in the calculation of the impact of CF investment. In view of the assumptions underlying the models, the results should be treated with caution.

(1) For the study of the models developed at the Commission's initiative the Court called on an outside expert specialising in the field of macroeconomic models and the European economy.

(2) Because of the approach (relative positioning of a region in relation to the mean for the Member State) and the fact that the statistics are pooled without any other disaggregation, with the result that the regression coefficients refer to a "representative" region.

Commission's replies

SUMMARY

Overall, the Cohesion Fund (CF) has had a positive impact. An analysis of trends in the four cohesion countries shows encouraging results over the period 1994 to 1999 considerable progress has been made in narrowing the gap with the rest of the Union.

The Commission agrees with the Court of Auditors that financing of the Cohesion Fund projects has proved effective.

The Commission considers that coordination with the ERDF was ensured during discussions with the Member States of the choice of projects to be submitted to the Cohesion Fund and during the interdepartmental consultation exercise. This coordination is apparent in the concentration of Cohesion Fund operations on specific sectors and corridors.

The Commission assessed each project in the light of the criteria laid down in the Regulation. The Cohesion Fund generally finances large-scale projects playing a pivotal role in the economies of the countries concerned. These projects form part of a planning process and require lengthy preparation.

The Commission has thus ensured that the projects financed by the Cohesion Fund are in accordance with the sectoral priorities (concentration on water and waste management) or the priorities for strategic corridors or nodes (concentration on a limited number of corridors and nodes). The 1994 to 1999 priorities emerged from discussions between the Commission and the Member States. The changes introduced by Regulations (EC) No 1264/1999 and (EC) No 1265/1999 require the Member States to provide the information needed to assess projects for conformity with a sectoral or territorial strategy. For the period 2000 to 2006, the Commission has therefore asked the Member States to draw up a reference framework for CF operations. This also figures in the Community Support Framework (CSF) (2000 to2006).

The Commission also ensured that Member States provided the information required by the Regulation and specified in the application form. The information and detail available to the Commission are of a fairly high standard.

The Regulation requires revenue from projects to be taken into account only if substantial net revenue is generated for the promoters. The Member States were given guidance on the method and applied it according to the particular nature of the projects submitted.

The Cohesion Fund Regulation mentions an eligibility date for expenditure but does not contain the concept of "new projects". It permits funding of projects which are already under way.

The linkage is set out in Article 6 of the Regulation in accordance with Article 104c of the Maastricht Treaty. The Cohesion Fund is specifically intended to reconcile budgetary convergence and the public investment necessary for the catching-up process by making the financing of such investment less of a burden on national budgets. Even on the extreme assumption that national co-financing is additional, its impact on the budgetary deficit would be marginal in the short term and positive in the medium and long term.

At every monitoring committee meeting, the Commission routinely monitored the status of each project and specified remedial action when necessary. This committee allowed a fairly close watch to be kept on the progress of projects, including their financial management. The Commission also carried out audit missions to supplement the committee's work and checked projects' physical and financial progress when processing payment applications. Extending the Commission's monitoring role would require a large increase in the staff assigned to managing and monitoring CF projects.

The Cohesion Fund has had a clearly positive impact, trends in the four cohesion countries in the period 1994 to 1999 are encouraging and there has been appreciable progress in catching up with the rest of the Union.

The Commission will make every effort to refine the assessment tools in the light of experience.

IMPLEMENTATION OF THE BUDGET

Financial development

9. Commitments are concentrated at the end of the financial year largely because of the dates on which the Member States submit their applications - usually towards the end of the first six months of the year. The only commitments in the first six months are those for budgetary instalments under decisions adopted in previous years. Similarly, payments are concentrated at the end of the year because of the dates on which Member States apply for payment.

In order to correct this situation, Article C(2) of Annex II to the new amended Regulation stipulates that the annual instalments for projects should generally be committed no later than 30 April. Article C(4a) requires applications for payment to be submitted three times per year, at the latest on 1 March, 1 July and 1 November.

10. There have been fewer new project approvals in recent years because of the increasing number of commitments under decisions taken in previous years which involve the payment of annual instalments. This reflects the priority given to large infrastructure projects.

11. The amount awaiting settlement arises from the project financing plans, which cover a period extending beyond 31 December 1999. It does not reflect delays in execution but results, as already mentioned in the previous paragraph, from the fact that the projects cover several years.

The uncertainty to which the Court refers is linked to possible variations in these financing plans. The Commission has never found itself short of payment appropriations. The Commission in any case allows for the outstanding amounts in the budgetary planning for each year.

Distribution of finance by sector

13. Article 10(5) requires the Commission to assess projects on the basis of the contribution they can make to Community policies on the environment.

In applying this provision, and being mindful of the need to avoid dissipating its efforts, the Commission decided to give priority to investment in waste-water treatment, drinking-water supply and waste management.

The Commission's interpretation of the Article was already clearly enunciated in the first report on the Cohesion Financial Instrument and the CF produced in 1995. The Communication of 22 November 1995 from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions on cohesion policy and the environment also stated that these were priority sectors.

This priority allowed substantial progress to be made in the beneficiary Member States and was reaffirmed for the period 2000 to 2006 in the Commission Communication of 1 July 1999 on the Structural Funds and their coordination with the Cohesion Fund for programmes in the period 2000 to 2006.

However, the Cohesion Fund has also played a part, for example, in combating erosion. Experience shows that, since such projects make considerable administrative demands on the Community and national authorities, this kind of operation is more appropriate to programmes of the type financed by the ERDF.

Commitments for rail transport showed an upward trend in the last years of the period, when there was a substantial increase in support for projects of this kind.

As regards maritime infrastructures, the Commission considers that maritime transport is a priority since it is less polluting. On the other hand, detailed case studies have to be made for the financing of port development projects, bearing in mind that investment in ports often generates substantial revenue. These studies must also address problems of competition and overcapacity affecting the other means of transport.

OBSERVATIONS ON THE MANAGEMENT OF THE COHESION FUND

Programming the projects

Deficiencies in the guidelines relating to applications for assistance

16. Regulation (EC) No 1164/94 did not require the Commission to consider the alternatives to a project. Moreover, this had been suggested in the Commission's proposal for a regulation but had been rejected by the Council in 1994. However, the recent revision of this Regulation requires the Member States to provide information on the alternatives considered.

The Commission has usually received cost-benefit analyses or other analyses of economic impact, albeit of uneven quality. When the Commission felt that the material supplied did not provide a sufficient basis for determining whether the project satisfied the Regulation's criteria, it made further enquiries, in particular by seeking assistance from the EIB.

The Commission has produced a detailed form stating what information the Member States must supply with applications for assistance. Several workshops were held in the Member States to brief the administrations concerned.

17. Generally, the Commission always has available information on the nature, location and cost of groups of projects. This is contained in the application or is at least supplied to the Commission while the project is being processed and is summarised in the decisions granting assistance. The forms drawn up for each group of projects are placed before the monitoring committees, with information on location, cost and status of expenditure for each project.

The location and amounts for some small subprojects within the four project groups mentioned by the Court were submitted in summary form by the Member State concerned.

18. When Regulation (EC) No 1164/94 was adopted, it included specific provisions for taking account of receipts.

19. For most small environmental projects co-financed in Greece, the Commission generally applied a rate of 80 % and not 85 %, since the operations funded could potentially generate revenue exceeding the operating costs.

Cost-benefit analyses are commissioned from consultants by the project promoters. The methods used may vary (see reply to paragraphs 91 to 94), especially as regards handling of operating costs and financial charges.

20. As the Court mentions in paragraph 31, it is for the national authorities to select applications, in accordance with the principle of subsidiarity. However, the Commission has always asked the Member States to submit more projects than can be funded so that a better selection can be made.

The CF Regulation requires the Commission to check that every project satisfies the eligibility and quality criteria of Article 10(5), but it does not provide for any comparative analysis by the Commission. When the Commission considers that a project does not satisfy the Regulation's eligibility and priority criteria, it informs the Member State of its reasons. However, the Regulation makes no provision for formal rejection decisions.

Coordination between the Cohesion Fund and the ERDF

23. The Commission has endeavoured to ensure that ERDF and CF operations are compatible and mutually reinforcing. For the period 2000 to 2006, the reference framework drawn up for the transport and environment sectors should make it possible to enhance such synergy and ensure coherence.

24. The Commission does not believe that the aim of optimising appropriations allocated to the Financial Instruments and the ERDF conflicts with the overall coherence of structural policies. Its view is that the concentration of Cohesion Fund operations on specific sectors and corridors makes it possible both to optimise the use of funds and to ensure that policies are coherent.

Budgetary and sectoral constraints

26. Comparability is not required by the CF Regulation. The Commission checks that each project conforms to the specified eligibility criteria.

28. The CF rules do not require programming to proceed at the same pace in the transport and environment sectors. The Commission is required to ensure that resources are suitably allocated between the fields, and this has been fully achieved.

29. The Cohesion Fund's resources are to a large extent devoted to big projects which play a pivotal role in the economies of the countries concerned. These form part of a planning process and require lengthy preparation. Other projects are covered by sectoral plans drawn up at the appropriate territorial level (national, regional, by catchment area). The standards of project preparation and project maturity are very important criteria for selection at both national and Community level.

30. The Commission believes that inter-project coherence has been ensured by concentrating operations on specific sectors and transport corridors. This was achieved through partnership with the Member State, covering programming and project preparation. The Commission also draws attention to the functional links between many of the projects funded.

Geographical distribution of the projects

32. The Commission applies the same procedure and criteria in preparing, adopting and monitoring all the projects financed from the Cohesion Fund. At Commission level, there are thus no differences in treatment.

The co-financing rate is determined on the basis of Article 7(1), taking account of the assistance requested by the Member State.

Slowness of the approval procedures

33 - 37. The Regulation stipulates that the Commission should generally decide within three months, provided the Member State has supplied all the information which the Regulation requires. This is conditional on the budget available each year, since some projects will not be approved if the commitment appropriations in a particular year are insufficient.

Delays in approval may be caused by:

- incomplete application forms, leading to requests from the CF for further information,

- insufficient commitment appropriations, with too many projects for the resources available (see also replies to paragraphs 10 and 20-a),

- lack of information on compliance with Community policies, especially as regards the environment and the award of public contracts.

The definition of project and project financing

38. As a general rule, applications for individual projects state the location precisely (see reply to paragraph 17). In the monitoring committee, the Member State provides detailed information on the execution of each project within the group.

39. In the framework of the Structural Funds in Ireland, elements of a number of projects had been funded prior to the introduction of the Cohesion Fund. From 1993 further elements of these projects were assisted from the Cohesion Fund in view of the suitability of these projects for CF assistance. The Commission considers that the elements covered under the Cohesion Fund were clearly identified and constituted project stages (see reply to paragraphs 40 and 41).

The four project applications concerning Belview port, Waterford, involved a new port access road, a new controlled rail crossing, the extension of the port quay and the dredging of the harbour approaches. The applications were submitted separately by the Irish authorities on various dates between June 1993 and February 1994. CF assistance to each project was approved by decisions on dates between November 1993 and July 1996.

The Commission accepted separate project applications in respect of these four improvement projects principally for the following reasons.

- The Commission saw an advantage in the treatment of separate applications on grounds of ensuring transparency of CF assistance between the various transport sectors (road, rail, port, etc.)

- There were three separate implementing authorities responsible for delivering these projects in view of their nature (road, rail and port improvements) and the division of responsibility for such transport issue in the Irish administrative system.

- Separate applications allowed the early adoption of decisions in cases where the Commission could quickly agree to assist the project (e.g. the port access and rail crossing projects). The examination of the Belview quay extension application was significantly delayed while the Commission clarified with the promoter important questions related to the project as initially proposed. In this case the separate applications allowed budget allocations to be made more efficiently.

Finally, despite the relatively small size of the four projects, even on a combined basis, Cohesion Fund support for port projects in Ireland has been economically justified because of the disproportionately large effect of ports investments on the Irish economy due to Ireland's island nature.

40 and 41. The Regulation provides for funding of projects, project stages and groups of projects. During the negotiations on Regulation (EC) No 1164/94, the Commission introduced the concept of a project stage and subsequently made it clear to the Member States that financially independent project stages are:

- stages which correspond to a functional entity (e.g. sewage treatment works) or which are defined in geographical terms (e.g. successive sections of a motorway or clearly delimited railway line), or

- stages corresponding to a given period of project implementation and to certain works which are physically identified by means of suitable quantified indicators and which are carried out in the course of this period.

At the end of 1995, the Commission informed the Member States of its interpretation of the "group of projects" concept, according to which a group of projects must satisfy the following criteria:

(a) they must be located in the same area or situated along the same transport corridor;

(b) they must be carried out under an overall plan for the area or corridor with clearly defined goals;

(c) they must be supervised by a body responsible for coordinating and monitoring the group of projects in cases where the projects are carried out by different competent authorities.

This concept was subsequently incorporated into the revised CF Regulation.

Calculation of estimates of net receipts

42. The Regulation requires receipts to be taken into account in determining the basis for assessment of the amount eligible for financing. The method used by the Commission requires the assistance rate to be determined on the basis of the ratio of the current value of net receipts to the current value of the investment. This method was formalised in February 2000 in a vade-mecum 2000 to 2006 and applied by the Commission to all revenue-generating projects on the basis of information supplied by the Member States. The Commission does not think it essential to lay down binding rules for the beneficiary Member States on the main parameters entering into the calculation (life of the investment, discount rate). Different parameters may be justified in different circumstances.

44. The Regulation requires the receipts actually anticipated by the promoter to be taken into account provided they give rise to substantial net revenue. It is therefore natural for toll charges to lead to a different rate of CF assistance for a motorway or for industrial waste-water treatment projects to attract a lower rate of assistance if the users pay charges which cover part of the investment cost.

45. When taking the decisions on motorway widening in Portugal, the Commission had available the EIB report on the financial and economic study for a broader project encompassing that submitted to the CF.

These projects were adopted under the Regulation governing the Cohesion Financial Instrument, which makes no provision for adapting the rate of assistance to take account of receipts. However, the Commission took financial viability into account in deciding on the rate of assistance and concluded that the traffic generated by widening the motorway and the revenue accruing to the promoter would be only marginal and difficult to quantify at the time when the project was approved.

46. Where there is substantial net revenue, the Commission uses information from the Member States to calculate the rate of assistance to be applied.

The differences observed by the Court in Member States' applications were largely eliminated when the data were recompiled by the Commission. However, different parameters may be justified in different circumstances (discount rate, residual value, etc.)

47. When submitting an application, the Member State undertakes to ensure the project's financial viability. As a minimum, the Commission requires operating costs to be covered by receipts when the Member State can apply a charging system.

Approval of projects already virtually completed

48. The CF Regulation specifies that the eligibility date for expenditure is the application date. It does not require CF assistance to cover the entire project cycle. Moreover, Article 1(3) specifically provides that the CF may finance project stages. Expenditure incurred before the project approval date may thus be eligible.

Conditionality

50. According to the political and economic rationale of the Cohesion Fund and its conditionality, the basic idea of the Cohesion Fund was to reconcile, on the one hand, the achievement of the convergence criteria on the budgetary situation and, on the other hand, the higher need for public investment with a view to the process of catching-up by alleviating the impact of its financing on the national budget through the Cohesion Fund. Conditionality is to reinforce the commitment of Member States eligible to the Cohesion Fund with regard to the public deficit criterion. Most macroeconomic models simulating the impact of Structural Funds and the related national co-financing, whose magnitude is far more important than in the CF, give only a slightly higher and then progressively diminishing public deficit as a consequence of two opposite forces. On the one hand, the extreme assumption of national public co-financing being fully additional increases the public deficit. On the other hand, the positive supply-side effects of investment in infrastructure and human capital on growth and employment, which increase tax revenues and reduce social transfers, tend to reduce the public deficit in the long run.

Project development cycle

52 and 53. The Commission has no way of knowing in advance how costs may increase during the course of a project.

It was decided to change the amount of assistance originally approved when this increase was considered justified, having regard to the views of the monitoring committees, where the progress of each project is discussed and the reasons for any departures from the initial forecasts are examined. All requests for changes were processed by the Commission in accordance with the Regulation (EC) No 1164/94 criteria.

54. The approved form for payment applications does not include information on synergies and coordination of the project with other structural operations. The payment application is not the appropriate stage for providing such information and it is difficult to see how it could be supplied in this context.

However, the Member State is required to provide information on the physical and financial progress of the project and on compliance with any requirements laid down in the decision.

55. The Commission scrutinises the final report and may ask the Member State for further information. Payment is generally made within two months after receipt of the application. However, this may take longer if the report is incomplete.

When several reports or payment applications have to be submitted, this is because of questions raised in the monitoring committees or arising from scrutiny of the case files.

Monitoring committees

56. The CF Regulation does not provide for the Commission and the Member State to discuss approval of new projects in this forum. However, the Commission has occasionally provided information in the monitoring committees on the progress of certain applications which were pending when the projects were of direct interest to the committee members.

Controls carried out by the Commission

57 and 58. CF-financed projects are monitored by the Member States and the Commission in accordance with the CF Regulation. Under the protocols signed with the national authorities, the Commission also coordinates these activities by means of frequent contacts with the national monitoring services.

As far as resources allow, the monitoring services of the Regional Policy Directorate-General carry out on-site and desk audits of CF projects as part of the annual audit plans for the four beneficiary Member States.

Article 3 of Regulation (EC) No 1831/94 requires the beneficiary Member States to report to the Commission any irregularities which have been the subject of initial administrative or judicial investigations.

Since the entry into force of Regulation (EC) No 1831/94 and Regulation (EC) No 1681/94 (the latter applying to the Cohesion Financial Instrument), five cases have been reported by the national authorities: one by the Greek authorities, two by the Spanish authorities and two by the Irish authorities for non-compliance with the rules on public contracts and non-eligible expenditure.

Not all the cases shown in Table 6 were reported by the authorities under the above Regulation.

Irregularities discovered during an audit by the Community institutions are also covered by Regulation (EC) No 1831/94 and the Member States have an obligation to bring them to the Commission's attention and to ensure that they are followed up.

OBSERVATIONS RELATING TO THE FINANCIAL MANAGEMENT OF THE PROJECTS EXAMINED ON THE SPOT

59. The eligibility criteria adopted by the Commission in 1998 reflect and confirm the practice followed since 1994 in defining and analysing eligible expenditure.

Eligibility of expenditure

60. It is sometimes hard to define recoverable VAT when processing a project. The VAT-related expenditure for the Thessaloniki project was regarded as eligible by the Member State. A letter was sent in this connection and the Member State itself is checking whether the expenditure declaration should be changed.

The Member State will deal with the Nafplio case in the same way.

The Commission will look into the Court's findings with regard to the Athens airport project and will take the necessary action.

61. A figure for ineligible expenditure pre-dating the eligibility date was declared by the Irish authorities in the application for the assistance and taken into account by the Commission in its grant decision. The sum of this ineligible expenditure will be re-examined in the light of the information provided by the Court and appropriate action taken where required.

62. The Commission will look into the Court's findings with regard to the water-supply project in Palma de Mallorca and will take the necessary action.

The finalisation of projects

63. The Argos project mentioned in paragraph 53 will be checked before it is closed.

64. The Commission acknowledges that it is sometimes difficult to see coherence between the various financial and physical indicators. This is particularly the case where the cost of completing a project has risen and the Commission has taken the decision not to match these increases with an increase in the Cohesion Fund grant. For monitoring purposes in such cases the Commission has focused on the comparison on the indicators of physical completion and financial completion based on forecast final costs (rather than absorption of grant and declaration of eligible expenditure).

65. The Regulation permits assistance for projects below EUR 10 million when this is justified. In the case of the project on the Balearic Islands mentioned by the Court, their island nature is a factor which justifies approving the project.

No provision in Regulation (EC) No 1164/94 forbids the financing of projects which are at an advanced stage or even completed on the date of the decision, since the eligible expenditure is that incurred by the Member State as from the date of submission of the application. However, the Commission tries to ensure that this situation is truly exceptional.

66. The Commission regards it as very important that works carried out by Member States should be of good quality, but it cannot be responsible for any constructional faults.

Adequate accounting

67. The declaration of contractors' invoices presented for payment as eligible expenditure before the invoice has been paid has been identified as a problem in a number of Irish projects. The Irish authorities have been made aware of this failing and have entered into commitments to address the problem. Where such a practice has been detected in individual projects the problem appears to have been resolved, though the issue has unfortunately arisen subsequently in other projects.

69. The project involves replacing an existing wastewater treatment plant at Tavira (Portugal) whose capacity and treatment standards were no longer adequate. The project was financed from two sources: the Envireg programme and the Cohesion Fund.

The works to be financed from each fund were specified in the grant decisions, the final report and the ex post evaluation report.

The Cohesion Fund was to finance the reconstruction and upgrading of the treatment plant, covering preliminary and primary waste-water treatment, ancillary installations (buildings, pumping stations) and connections between the primary and secondary treatment facilities, while the works to be financed by Envireg, which were physically about 1 km distant from those financed by the CF, covered secondary and tertiary treatment.

70. The amount mentioned by the Court is supported by certificate No 3 of 30 June 1995 in the accounts of the Ministry of Public Works, Directorate-General of Economic Planning.

EVALUATION IN THE COHESION FUND DOMAIN

Macroeconomic models

Models developed at the Commission's instigation

Reference framework and contract management

73. As regards the Court's comment on the applicability of the method, the purpose of the work was to develop new methods of evaluation: it was therefore impossible to have previous experience of their application.

The main aim of the contract was to develop a model or set of models for estimating the socioeconomic impact of projects at regional or national level or on other Community countries. The successful tender was indeed original and it was not known at the outset how far it would prove applicable, but the tenderer's experience gave the Commission a basis for judging the soundness of the bid. The consultant selected is a highly specialised university institute of high standing in the field of macroeconomic analysis.

The Commission does not regard the three-month extension as a change in the award criteria.

74 - 75. The second contract is for quantified simulations of the socioeconomic impact of individual projects using the models developed under the first contract.

The models produced by the consultant under the first contract are the property of the Commission. The Commission's reasons for contracting out application of the models produced to the individual projects to be assessed were the complexity of innovative macroeconomic models (which are not far short of applied research) and its lack of staff with the necessary knowledge and expertise in this field.

The contract is a framework contract with a total allocation for various simulations which are not initially specified. Individual simulations are subsequently called up as required and charged against the overall estimate in the contract.

76. The first contact contains fixed costs (fees) and variable travelling and subsistence costs, which were initially estimated at EUR 56160. Since it proved unnecessary to use all of this estimated amount, there was a cancellation of EUR 1561,02. The estimate was thus very close to actual travel and accommodation expenses.

No payments were made on this contract until 31 January 2000 but the invoices received amount to EUR 60600 for 21 simulations carried out by that date and now being studied by the Commission.

Conformity with the evaluation objectives

82. The Commission considers that each of the models has yielded positive results and it is making every effort to ensure that they are refined and applied in future. As regards coordination between models, the approach chosen was to use different models, each measuring different aspects of the investment.

Cost-benefit analysis (CBA)

86 - 90. The Member States have made great efforts to improve the quality of CBA but more remains to be done. The Commission notes the comments and will take them into account in improving future cost-benefit analyses for the Cohesion Fund.

For the Lisbon project, see the reply to paragraph 18. With regard to the Ibiza project, the water price used by the Spanish authorities in the 1994 CBA was based on previous experience. It is perfectly acceptable and indeed greatly to be welcomed that this indicative price should be revised three years later to take account, inter alia, of updated operating costs. Before finally closing the project, the Commission will in any case require a review of the economic factors which may have affected the assumptions of the initial CBA.

91. Construction costs are one of the elements entering into any infrastructure investment. Investment costs also include other types of expenditure such as the costs of land, studies, etc.

92. The Regulation requires the Commission to ensure that the socioeconomic benefits are proportionate to the resources used. Despite the different methods used, the Commission considers that the reports available would justify drawing this conclusion. The Commission also has at its disposal other means of appraisal, in particular the reports produced by the EIB at its request.

The Commission shares the Court's view that CBA methods can be improved. In the new version of the guide, the Commission will encourage the Member States to align their methods in the interests of greater comparability, giving greater weight to the EMU context. For reasons of subsidiarity, however, the Commission cannot impose a harmonised method for CBA.

Indicators

93. The indicators used by the CF are indicators of projects' physical and financial progress and not indicators of the progress, results and impact of programmes like those produced for the ERDF.

94. The indicators are agreed between the Commission and the Member State. Their purpose is to provide an overall picture of project status. The Commission regards this approach as satisfactory in view of the aims pursued.

95. The monitoring indicators used by the CF do not seek to "represent the improvement of the sector concerned quantitatively" but rather to track physical and financial progress in executing the project.

CONCLUSIONS

110. The Commission believes that, overall, the Cohesion Fund has been effective. The new regulation amended by the Council introduces improvements which should increase this effectiveness.

111. The situation envisaged by the Court has not arisen in practice. The risk was rather that the payment appropriations would not be fully used, since expenditure resulting from the projects approved was lower than the pre-approval estimates in certain years, including 1999. The outstanding amounts are smaller for countries which make greater use of the annual instalment system.

112. Monitoring of projects is a matter of particular concern to the Commission. Intensifying the monitoring of CF projects as the Court recommends would require a large increase in the number of Commission staff.

113. The Commission provided information on this matter at the various monitoring committee meetings. The revised CF Regulation gives more precise definitions of the various concepts.

114. It takes a long time for certain projects to be approved because the information supplied is inadequate, because compliance with Community rules is not demonstrated or because the projects do not appear to have the highest priority given that resources are limited.

With regard to its internal administrative procedures, the Commission has instituted a new delegation procedure for project approval which should speed the process up.

115. The Commission considers that the projects approved in the period 1993 to 1999 fit into a coherent philosophy. However, it hopes to strengthen the programming approach in the Community support framework III (2000 to 2006). It has therefore asked the Member States to draw up a framework of reference setting out the strategy and allowing CF and Structural Fund operations to be coordinated.

117. The Commission will endeavour to improve the evaluation tools in the light of experience.