Special report No 7/99

concerning the development of industrial sites, together with the Commission's replies

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

(2000/C 68/02)

CONTENTS

>TABLE>

Summary

Within the context of its audit of the development of industrial sites (see paragraph 1), in addition to examining the legality and regularity of the co-financed projects the Court also wished to investigate the degree of integration of the development of industrial sites into that of the regions and the impact of the sites. In order to do so, it conducted on-the-spot audits in six regions receiving European Regional Development Fund (ERDF) aid and sent questionnaires to seven others. Investigations were also carried out at Commission departments (see paragraphs 2 to 4).

Although industrial sites are not mentioned in the texts of the regulations governing the Structural Funds, they are nevertheless often co-financed by the ERDF and it is difficult to quantify the total amount of its contributions. An improvement in the quality of the data sent in by the Member States would be necessary in order to obtain more accurate information on the use of ERDF resources (see paragraphs 5 to 7).

As regards legality and regularity, some observations have been sent to several Member States concerning the eligibility of expenditure declared by them and irregularities or legal disputes not notified to the Commission (see paragraphs 10 to 12).

As regards regional development and conversion, it was found that there are often a large number of operators with responsibility for the creation of industrial sites, but no coordination between them (see paragraphs 15 to 18), that there are no land development priorities for industrial sites and that the objectives and selection criteria for the projects are weak (see paragraphs 29 to 39). The initiatives that are occasionally taken to draw up inventories of sites are not always adequately targeted (see paragraphs 19 to 28).

The Commission ought therefore to encourage the Member States to create analytical and monitoring tools (inventory and sales statistics) to enable active management of the creation or conversion of industrial sites within the framework of a genuine regional development plan (see paragraph 70).

In the specific case of "business nurseries", a maximum length of stay should be imposed on companies so that the nurseries can remain focused on their aim of helping in the creation of new operations (see paragraph 28); they should compile statistics on the undertakings' success rates and the jobs created (see paragraphs 33 and 46).

The evaluators of ERDF measures should also investigate whether there is a consistent site development approach and an analysis of needs (see paragraphs 40 and 41).

With regard to the impact of the ERDF aid, it was found that where inventories of the sites exist they enable the number of undertakings and jobs within them to be monitored in a more appropriate way than is possible using occupancy rates alone; little information is available on the jobs created by undertakings which have benefited from the "nurseries" (see paragraphs 43 to 46).

There are few regions where it is possible to verify the application of the rules on competition in connection with the fixing of rent levels and sales prices for developed infrastructure. The market price, which should where necessary be taken as a reference point in order to measure distortion of competition, is not in fact helpful since it is too closely connected to the initiatives of State property-developers alone (see paragraphs 49 to 55). The Commission should for this reason examine the lack of transparency in the fixing of prices for developed plots. In addition, in view of the competition between developers in a region to attract undertakings which are moving onto sites, ERDF aid should a fortiori only be granted within an overall regional development framework laying down the selection criteria for subsidised projects (see paragraph 56).

When ERDF aid enables the developer to make a profit, there is no counterpart to the Community financing and the non-refundable ERDF aid has no observable impact. In some cases, replacing the subsidy with a loan should be considered. Subsidies are justified when there is a desire to create an incentive for localisation and in the redevelopment of derelict or abandoned former sites. Subsidies should anyway always be granted within the framework of a structured regional development plan, in order not to outstrip actual needs (see paragraphs 57 to 62).

Introduction

1. Two of the objectives ascribed to the Structural Funds (SFs) are to further the development and structural adjustment of regions whose development is lagging (Objective 1) and to restructure regions seriously affected by industrial decline (Objective 2). In this connection the ERDF co-finances projects to set up or modernise infrastructures, including industrial or small business areas whose developed plots are generally sold vacant to undertakings. In some cases these areas are equipped with buildings, whether they be "nurseries" intended to house very recently established undertakings on a temporary basis by providing assistance, management support and shared office services, or "bridging buildings" to house undertakings that are not yet in a position, or do not wish to bear installation-investment costs. Such areas and, where appropriate, the buildings within them are termed "industrial sites".

2. Neither the concept of industrial sites nor the types of infrastructure it covers are specifically mentioned in the regulatory texts governing the SFs. The financial resources devoted to them may, however, account for a significant proportion of the total ERDF aid in regions whose development is lagging or which are being restructured (see paragraphs 8 and 9), and they are therefore covered indirectly by the Community regional policy. The Court therefore wished to examine, in addition to the legality and regularity of the operations co-financed by the ERDF (see paragraphs 10 to 12), firstly, how the procedure for developing industrial sites fits into that of the regions receiving ERDF subsidies (see paragraphs 13 to 41) and, secondly, the impact of these operations (see paragraphs 42 to 65).

3. The ERDF contributes to the development of industrial sites primarily through operational programmes made up of "sub-programmes" which are in turn subdivided into "measures", under which the projects are co-financed in the eligible regions. The Commission intervenes chiefly at the level of the main priorities for these programmes. For the measures, its role is primarily to ensure that the Member States and regions submit files describing the planned actions. At the Commission, the investigation was conducted at the Directorates-General for Regional Policy and Cohesion (DG XVI) and Competition (DG IV).

4. Audits in the Member States were carried out in six regions receiving ERDF subsidies, three with Objective 1 status (the Land of Thuringia (D), the Autonomous Community of Valencia (E) and Sicily (I)) and three with Objective 2 status (the eligible regions of Limburg (B), the Nord-Pas-de-Calais region (F) and Wales (UK)). These regions are economically and geographically different, which has an effect on their development or restructuring. They were selected because of their interest value as regards the development of industrial sites and the volume of appropriations they allocate to finance them in the context of ERDF support measures. This sample is not intended to ensure statistical representativeness, but rather to demonstrate the range of differing practices, some of which should be encouraged. Questionnaires were also sent to seven other regions, of which six replied (Hainaut (B, Objective 1), Saarland (D, Objective 2), the Autonomous Community of Extremadura (E, Objective 1), the Loire region (F, Objective 2), Basilicata (I, Objective 1) and Limburg (NL, Objective 2)), and one did not (Ireland (Objective 1)).

Budgetary and physical data

5. The Commission departments have various systems for recording accounting and financial data concerning ERDF support measures:

(a) Sincom is the general system that records the Commission's commitments and payments, including those relating to ERDF support measures; this system is designed for accounting purposes and contains no information concerning the content of the measures;

(b) Garfield is a local system, specific to DG XVI, which details the appropriations(1) under ERDF support measures down to measure level; however, it does not record the expenditure declared by Member States, even though this is broken down in exactly the same manner; it contains a database that is used by DG XVI to list, using a predetermined thematic classification, measures under ERDF assistance programmes dating from the 1994 to 1999 generation; some items of information are not included in this database;

(c) details of expenditure actually incurred and broken down in accordance with the measures provided for in the programmes are only available in the form of hundreds of paper and computer files concerning these programmes; no overview of the situation is drawn up.

6. The Court has already noted on several occasions the difficulty of quantifying the financial resources earmarked by the ERDF for actions in favour of specific fields or target groups(2). Any attempt to identify the appropriations allocated for developing industrial sites also comes up against several difficulties, of which the main ones are:

(a) projects to develop new industrial sites may be charged to measures that include projects of a different nature. Examples from the programmes audited are conference centres (Nord-Pas-de-Calais, 1989 to 1991/A operational programme), water-supply and tourism infrastructures (Thuringia, 1989 to 1993 operational programme) and other local development, tourism or industrial projects (Valencia, 1990 to 1993 operational programme); nurseries and bridging buildings can actually be covered by measures to support SMEs;

(b) In Objective 2 regions new industrial sites were created by redeveloping abandoned former sites and are charged to measures that can also include urban regeneration; furthermore, it is often the case that redeveloping industrial sites does not result in new sites being developed, but rather residential developments or greenfield sites.

7. The appropriations earmarked for developing industrial sites in the financing schedules are therefore only indicative and approximate. The only way of obtaining more accurate information would be to improve the quality of the data sent in by the Member States, and one principal means of achieving this would be for assistance programmes to be more homogeneous and detailed in structure. The information concerning the 1994 to 1999 period extracted from the Commission database (see paragraph 5) shows that the appropriations set aside for developing completely new areas amount to ECU 2493 million and those for redeveloping run-down former sites to ECU 1338 million. Furthermore, appropriations totalling ECU 4363 million were earmarked for structures to support economic activities that may include industrial sites, but the nature of the infrastructures was not further specified beforehand.

8. With regard to the six regions visited, Table 1 shows the initial financial appropriations under ERDF assistance for the 1989 to 1993 and 1994 to 1999 periods for, firstly, these regions' main co-financed programmes, with all measures taken together, and, secondly, measures in these programmes likely to include industrial-site development projects. The corresponding expenditure for the 1989 to 1993 period is also shown. Depending on whether the regions visited have Objective 1 or Objective 2 status, it was observed that the amount of ERDF aid devoted to the industrial sites may account for up to 23 % or 48 % of the total.

9. Physical data relating to these same regions and their industrial sites are set out in Table 2. They mainly indicate that, in comparison with their surface area and population, Objective 2 regions have more extensive infrastructures than those with Objective 1 status. Owing to lack of availability, some data concerning the Objective 2 regions (population and surface areas) are shown only at NUTS II level(3), whereas areas eligible for ERDF Objective 2 support measures are set at NUTS III level.

Legality and regularity

10. The audit of projects in the regions visited gave rise to observations concerning the eligibility of certain items of expenditure declared, and revealed irregularities or legal disputes not notified to the Commission.

11. The issue of ineligibility arose in three of the six projects audited in the Community of Valencia (E), for which VAT and litigation provisions and various sums unconnected with the projects audited had been declared (ineligible expenditure of ECU 1,2 million out of a total of ECU 9,1 million; ERDF aid of 58,3 %). Moreover, one amount of ECU 2,5 million corresponded to excess expenditure declarations compared with what should have been declared under the principle according to which a maximum of 10 % of the total eligible cost is accepted for purchasing land(4). For one of the six projects audited in Nord-Pas-de-Calais (F) the declared expenditure could not be justified (ECU 0,2 million out of a total of ECU 1,7 million; ERDF aid of 19 %).

12. Failure to declare irregularities to the Commission in accordance with Regulation (EC) No 1681/94(5) applied, firstly, to two projects in Thuringia (D), the value of which was deducted from the expenditure declared following an audit conducted by the Land Audit Institution (expenditure of ECU 4,3 million and ECU 2,9 million) and, secondly, to Sicily (I), where two of the four projects audited included in the expenditure declaration were subject to legal disputes before the courts (expenditure of ECU 3,6 million and ECU 1,9 million; ERDF aid of 50 %). In one of the latter cases the building co-financed by the ERDF had collapsed.

Mode of development of industrial sites co-financed by the ERDF

13. Assistance in the form of programmes presupposes action being taken within an institutional organisation specific to each Member State or even to each region (see paragraphs 15 to 18). This action should include:

(a) a determination of requirements (see paragraphs 19 to 28);

(b) the setting of objectives appropriate to a regional development project and indicators for measuring the degree to which the projects are implemented and their impact (see paragraphs 29 to 33);

(c) the setting of criteria for selecting projects (see paragraphs 34 to 39);

(d) an ex post evaluation of the projects implemented and their impact, which should be taken into account when subsequent analyses are carried out (see paragraphs 40 and 41).

14. Where such action is not taken at global level within the region concerned, there is a risk that too many or too few industrial sites will be developed, either in relation to demand on the part of undertakings (output mismatch) or in ignorance of the latter's requirements. Undertaking demand is not, however, the only

Table 1

Financial data on ERDF aid in the regions visited

>TABLE>

Table 2

Physical data on the regions visited and their industrial areas

>TABLE>

criterion that should guide the development of industrial sites. It is the responsibility of the authorities concerned to ensure that economic activity is stimulated in areas eligible for ERDF assistance, in accordance with the specific requirements of these areas, and to prevent industrial sites being spread too widely, as this may have a detrimental impact on the environment and land-use planning.

Organisational background

15. The initiative to develop sites co-financed by the ERDF may be taken at various levels. The main approaches can thus be grouped into three categories: (1) the sites are developed exclusively by the municipalities in accordance with their specific objectives; (2) local authorities whose responsibilities cover a group of municipalities, if not the actual region, also take the initiative in this direction; (3) sites are also developed by bodies set up for this purpose. The feature generally common to these different categories is a lack of coordination between the operators and, where this occurs, failure to examine the expediency of the projects. In the rest of this report, the authorities or bodies which receive ERDF assistance are referred to as "final beneficiaries" and sometimes as "developers"(6). The final beneficiary is thus not the undertaking which moves onto the industrial site.

16. Thuringia (D) and Limburg (B) belong to the first of the above categories. In the latter region, however, cooperation between the communes and the private sector is achieved through a regional development company that is at the centre of various initiatives in favour of undertakings, manages a detailed database on sites and draws up related long-term analyses.

17. The Autonomous Community of Valencia (E), Nord-Pas-de-Calais (F) and Wales (UK) may be placed in the category where industrial sites are also developed by authorities which are not municipalities. In the Autonomous Community of Valencia (E), the industrial and small business areas are created by a State-owned company under private law, which accounts for some 70 % of site supply, by a regional State-owned company under private law and by the provinces, municipalities and the private sector. A lack of coordination between these operators was observed. In Nord-Pas-de-Calais (F) there is fierce competition between the municipalities to attract undertakings, due largely to the volume of budgetary revenue that can be generated by means of "business tax"(7). This competition has led to a sharp increase in the number of relatively small industrial and small business areas (see paragraph 41). Attempts are, however, being made to encourage the municipalities to cooperate with each other. In some cases the private sector is involved in developing industrial sites through State and private sector partnerships. In Wales sites are developed on an individual basis by the regional, county and municipal administrations and a development agency.

18. In Sicily (I) 33 industrial areas have been developed by 11 consortia set up by law under the supervision of a section of the regional administration. Small business areas, on the other hand, are developed by municipalities under the supervision of a different section. There is no coordination between these operators, even though industrial undertakings can set up in small business areas and vice versa.

Determination of requirements

19. Carrying out an analysis of the number of developed industrial sites required presupposes the appointment of a body to be responsible for developing them or for coordinating the initiatives launched by the various parties, particularly in view of the lack of coordination between State property-developers. If global knowledge of a region's requirements is to be acquired, then there is a need for information concerning existing infrastructures and a method and data that will enable long-term demand to be assessed.

Industrial and small business areas

20. In addition to the fact that the occupancy rate of industrial and small business areas(8) may conceal subregional imbalances, as an indicator it fails to ensure that sufficient developed surface areas are available, given that this can only be achieved by taking account of sales. For example, in Limburg (B), whose average site-occupancy rate is 92,8 % (see Table 2), availability is relatively greater than in Nord-Pas-de-Calais (F), whose average rate is 81,5 %, as annual sales in the former in fact amount to 115 hectares and in the latter to 1000 hectares. Limburg therefore has vacant developed sites that will enable it to meet sales for five years, whereas availability in the other region is limited to just over two years.

21. Some regions visited that have a body responsible for such matters, information on existing infrastructures and a method and data for assessing long-term demand took the initiative of drawing up analyses of requirements for industrial and small business areas. In Limburg (B) analyses were made on the basis of historic sales of plots of land over a 10-year period and an optimum stock-level was defined as being one sufficient to enable demand to be met over the average period required to develop a new area, i.e. three years, plus a safety margin. A similar analysis, with a 15-year horizon, is carried out by the Flemish community with the provinces' assistance when the land-planning scheme is defined. In Nord-Pas-de-Calais (F) an initial survey of industrial areas, which detailed, inter alia, the surface areas developed and available to undertakings, was conducted in 1996 and co-financed by the ERDF. On the basis of the average figure for the surface area sold annually, it can be seen that there is a very uneven locational spread of developed-plot reserves and that there are shortfalls in certain districts. This survey remains to be checked by the departments responsible for it, but the fact that they only have to update it every three years reduces its usefulness. It is not used to evaluate plot-development requirements.

22. Other regions visited have information concerning their industrial and small business areas, but do not use it to determine requirements, in particular because no information concerning site sales exists. In Thuringia (D) the inventory, which was drawn up in 1997 and covers areas for which some municipalities have received public subsidies, still contains no information on unredeveloped areas opened up in the former German Democratic Republic. A need for 1500 to 1600 hectares of developed sites is quoted under the 1994 to 1999 SPD for this region, but these figures were not based on an analysis of the existing supply and sales. In the Autonomous Community of Valencia (E) there is an inventory detailing the areas, created at different times, which still have vacant equipped plots. They account for 1917 hectares out of a total of 3198, giving an average occupancy rate of 40 %; 120 hectares were quoted as being required under the 1989 to 1993 operational programme, but no global analysis had been made. In Sicily (I) an inventory of industrial areas was compiled in 1989, with ERDF assistance, and was updated in 1996. However, it does not indicate the developed surface areas that are unoccupied and available to undertakings. There is also a list of small business areas developed by the municipalities, several of which have received ERDF assistance, but there is no information as to their surface area or occupancy rate. In Wales (UK) a partial database has been developed by the development agency, but the information it contains is not used as a forward-planning tool.

23. The information in the abovementioned inventories is forwarded by the site-owners to the authorities responsible for keeping such inventories. Discrepancies between the information in these inventories and that obtained from final beneficiaries were detected in Limburg (B) and Nord-Pas-de-Calais (F), or when situations established on different dates were compared in Thuringia (D) and Sicily (I). They are due to, inter alia, incorrect data being forwarded or the protracted intervals at which these databases are updated.

24. Of the regions that replied to the questionnaire (see paragraph 4), Hainaut (B) has both an inventory of the industrial areas under the five local authority associations that are responsible for their development and statistics on sales of developed sites. Although examination of sales is cited in the questionnaire as being a guiding principle when a decision is taken to develop new areas, the occupancy rate of the developed plots is only 59,5 %. At the average sales rate observed in recent years the time needed to exhaust the stock of equipped plots available may be considered excessive. The redevelopment of fallow land will further increase the available surface area. The Court has already detected overcapacity in this region on past occasions(9). The Loire Region (F) has neither inventories of sites nor sales information. In Extremadura (E) there is a list of completed industrial sites which indicates total surface areas and the surface areas of plots, which makes it impossible to estimate actual needs; the local authorities have, however, said that a detailed inventory is being prepared now. In Basilicata (I) a small amount of basic information has yet to be supplemented. Limburg (NL) has a database of industrial sites and some information on the annual sales of developed plots. A requirements analysis, broken down according to the various types of site, is carried out every five years on the basis of job-creation scenarios and sectoral statistics regarding the number of people employed per hectare. In addition, an analysis of plot availability and sales is also carried out annually. It is considered necessary to maintain a permanent level of stock equal to the average quantities that can be sold over a five-year period.

25. Feasibility studies have been carried out for some large industrial sites and in some cases have been co-financed by the ERDF. This was the case with a logistical and transport area and a services zone in Limburg (B) and a mixed transport (road/rail) site in Nord-Pas-de-Calais (F). In the latter region feasibility studies were also carried out for smaller projects. One of these studies, carried out at a municipality's request, described the highly unfavourable environmental situation with regard to two sites it had planned and the extensive site availability in surrounding municipalities. This was not reported, however, in the document presenting the project to the regional authority which was to decide whether ERDF assistance would be granted. These projects' development was finally suspended, owing to a lack of interest on the part of local undertakings. Cases can also arise where, despite a feasibility study having been carried out, the site developed does not accurately meet demand. In Wales (UK) the plots on offer comprise, inter alia, redeveloped former mining- and steel-industry sites that are very small in size. Undertakings are aware of the lack of large-scale sites and have been forced to set up outside the region. Nevertheless, when an area of several dozen hectares was redeveloped, with ERDF assistance, it was decided that it would be geared towards SMEs operating in high-technology sectors and turned into a showcase area. Some time after it was opened, part of the development co-financed by the ERDF had to be demolished to allow a large undertaking to be installed. Furthermore, some landscaping equipment there was no longer being maintained.

26. In Limburg (B) and Nord-Pas-de-Calais (F) surveys were carried out before measures to redevelop abandoned industrial sites were taken. The scope for assigning these sites to other uses is frequently limited for legal, technical, environmental or, simply, access reasons. The delays that may result are not particularly consistent with the desire to be in a position where new commercial surface areas are promptly available, and the sites are generally reallocated in a piecemeal fashion. In Limburg (B), although additional financial resources were allocated for developing new areas, those initially set aside for the redevelopment of former sites were underutilised. In Thuringia (D), as of 1994, site creation and redevelopment were linked by the requirement that redevelopment of former sites should absorb 30 % of all financial resources allocated to industrial sites, despite the lack of an inventory of former sites.

Business "nurseries" and bridging buildings

27. Public authorities are usually prompted to start developing business "nurseries" or bridging buildings by the results of similar projects (see paragraph 1). In this way a network of centres is gradually established that is more or less coordinated(10).

28. Since these buildings' operating costs have to be covered by rents paid by the undertakings using them, their occupancy rate is of primary importance. It is therefore essential to determine potential demand as accurately as possible, so as to avoid low occupancy rates and the attendant need to increase the rents charged to cover operating costs. This would be particularly detrimental in the case of "nurseries", since their objective is to encourage new undertakings to "hatch out" and to reduce their mortality rate. If there is insufficient information on the level of demand, some "nurseries" may be created and then prove not to have been needed. They may end up being converted into buildings that are simply leased to undertakings, as was the case with two co-financed projects in Limburg (B). Although three of the remaining five centres have been operational since the end of the 1980s, their average occupancy rate is still only approximately 70 %, whereas when they were set up, the rents had been calculated so that their profit and loss accounts would balance on the basis of a 90 % occupancy rate. The centre with the highest average occupancy rate is the one that had the highest number of undertakings staying for more than five years. In Nord-Pas-de-Calais (F), the occupancy rate of one "nursery" set up in 1992 was below 50 %. Following difficulties at the beginning of the 1990s, one group of seven "nurseries" set up at the end of the 1980s and managed by a French département authority had an occupancy rate of over 80 %. As these structures are very sensitive to the economic climate, the occupancy rate increased overall in 1998. In Wales (UK) the failure to impose a maximum length of stay on undertakings resulted in one "nursery" co-financed by the ERDF gradually turning into a building leased under normal conditions.

Objectives and indicators

29. Up until 1994 several programmes made no reference to the specific objectives of developing industrial and small business areas, or merely paraphrased the title of the measures. Implementation indicators existed in certain cases, whilst impact indicators were very rare and could not always be calculated. In Sicily (I) the main objectives were to increase industries' added value, stimulate entrepreneurial initiative and encourage the relocation of activities to areas outside urban centres; physical and impact indicators were proposed.

30. In Wales (UK) the intention was to measure the sites' impact on jobs. Although there are no statistics on this topic, estimated figures were put forward when the projects were submitted for ERDF funding. The ex post evaluation of programmes for the 1989 to 1993 period revealed a considerable difference between the estimates and what was actually achieved, which may be attributed in part to over-optimistic forecasts: 3793 and 633 gross jobs(11) were created respectively under measures relating to industrial sites under the 1989 to 1991 operational programme and the 1989 to 1993 RECHAR Community Initiative (conversion of coal-mining areas), whereas the projects adopted had forecast that 12658 and 1095 gross jobs would be created respectively.

31. As of 1994 the programmes almost systematically included more detailed objectives and some physical and impact indicators (surface area developed and numbers of undertakings set up and jobs), with the exception of those of Thuringia (D). Very few indicators exist that show concern for regional development and, more particularly, no distinction is ever drawn between gross and net employment.

32. The information concerning the undertakings set up and jobs available is not very relevant, since it is drawn up when the project is merely at the start-up stage. In this regard, ex post evaluation is nevertheless carried out too early. However, in some cases the inventories compiled by some regions (Thuringia (D), Sicily (I), Limburg (B) and Nord-Pas-de-Calais (F)) sometimes include information on the undertakings located at the various sites or on employment, which allows the impact of these items of infrastructure to be assessed more accurately (see paragraph 43).

33. From the outset the stated objective of some "nursery"-creation measures has been to reduce the failure rate of new undertakings by lowering their fixed costs and obtaining management subsidies for them. However, little information is available concerning job creation by undertakings that have benefited from this support infrastructure (see paragraph 46).

Selection criteria for projects

34. An active industrial sites development policy would need to use selection criteria suitable for development projects with, for example, priorities as to the siting of projects, their design quality or the type of undertaking they could house. This principle is all the more necessary where State property-developers compete on selling prices or rent levels (see paragraphs 48 to 62). The programmes do not usually contain selection criteria for projects.

35. In Limburg (B), ERDF assistance was supposed to be granted according to urgency and need, and the quality of the land. In fact, the supply of developed sites decreased at the end of the 1980s and all the projects were co-financed as and when applications were submitted. For business "nurseries", location criteria had been set and the success of previous projects was supposed to determine whether or not new ones were set up. In fact, three projects out of six had already been completed before implementation of the programme began, implying that in this case the related ERDF co-financing had no direct effect on regional development. A fourth project was set up a few kilometres from a centre which already had occupancy problems; in the end both these centres had to be converted into bridging buildings, bringing the projects into the sphere of property rather than support for new undertakings.

36. In Thuringia (D), local authorities wishing to receive public assistance to create a zone had to demonstrate interest on the part of undertakings, at least 50 % of which were supposed to fulfil specific conditions on sector of activity and turnover outside a zone defined around the administrative area concerned (this condition was abolished for projects receiving assistance from the Land as of 1995). The concept of eligible undertakings laid down for this purpose resulted, for instance, in the exclusion of the development cost of plots occupied by commercial undertakings with a large surface area. About half of the projects for industrial sites accepted in this way were charged to the ERDF appropriations, without any specific criteria.

37. The Autonomous Community of Valencia (E) programmes do not include any selection criteria for projects.

38. In Nord-Pas-de-Calais (F), although general criteria had been set for redevelopment measures, of which the main one was the size of the site, if one looks at the measures during the 1989 to 1993 period it can be seen that the choice of projects was more often determined by urgency or the less complex nature of the sites and was not based on a regional redevelopment plan. For new industrial and small business areas, contrary to the provisions of the 1997 to 1999 SPD, the 1996 register was not used as a basis for selecting projects, since the data were supposed to be updated and verified in 1999. In addition, in order to regulate competition between municipalities (see paragraph 17), a higher rate of assistance was set for projects with an intermunicipality structure.

39. In Sicily, there were no criteria during the 1989 to 1993 period and the administration was left to select the projects. In the subsequent programme, five industrial areas were chosen, with the aim of reinforcing the effects of assistance, of which four were selected on the basis that they were already well developed. However, because the projects in these five areas advanced too slowly, all the areas are to be given the possibility of receiving Community appropriations.

Evaluation

Interim and ex post evaluation within the context of Community measures

40. Evaluations of ERDF assistance all too often say nothing about the measures for industrial sites or do not discuss whether it would be advisable to create any (existence of a development plan based on an analysis of supply and demand or on priorities). For Limburg (B), it was not until the interim evaluation of the 1994 to 1996 SPD that the evaluator examined the comparative levels of supply and demand for developed land and pointed out the relatively high level of availability at that time. For Thuringia (D), the interim evaluation of the 1994 to 1999 operational programme highlighted the lack of data on the need to develop new sites and the absence of planning which could be taken into consideration when approving aid. For the Autonomous Community of Valencia (E), the interim evaluation of the 1994 to 1999 operational programme is the only one available, and it does not deal with the question of industrial site development needs. In Nord-Pas-de-Calais (F), the ex post evaluations of the 1989 to 1993 period did not draw attention to the lack of information on the stock of developed plots. In Sicily (I), ex post evaluation was only carried out for the "Aree Attrezzate" multiregional programme, and it was mainly concerned with the problems of putting Community appropriations to use and the aim of decongesting urban areas, which was pursued by relocating undertakings in industrial zones. In the absence of data on where undertakings were located, the evaluator took as a basis the figure for the percentage of wage earners in the areas, which had dropped from 35,1 % to 29,9 %, from which he concluded that the objective had not been achieved.

Specific evaluations outside the framework of the Community measures

41. In the two Limburg provinces (B and NL), sales of developed plots were monitored and a "stock management" approach was introduced (see paragraphs 21 and 24). In Nord-Pas-de-Calais (F), the most important points revealed by the survey (see paragraph 21) were the role of business tax in the fragmentation of the supply of developed land and the lack of selection of undertakings in the developed zones. This report pointed out the risk of costly competitive bidding between local authorities in certain districts where there seemed to be an oversupply. It recommended setting priorities, increasing efforts aimed at coordination, image, development quality, service and harmonisation of the business tax. In Thuringia (D), since mid-1997 questionnaires have been sent to the communes with co-financed projects to obtain data on their state of progress and on the occupancy of the completed zones and the corresponding level of employment. This ex post monitoring does not, however, negate the benefits of a long-term analysis.

The impact of industrial sites and the co-financing of them

Undertakings and jobs

42. It is difficult to evaluate the impact of ERDF co-financing of industrial sites for the following reasons:

(a) an industrial or small business area does not in itself create employment; jobs are created, rather, by making optimum use of endogenous potential (research and development assistance, management support, particularly through business "nurseries", etc.) while taking into account exogenous factors which may influence job creation (road infrastructure, attracting outside undertakings, etc.);

(b) co-financed projects often relate to only part of an existing or new site, such as installing a sewage system, wiring, a section of access road or internal road, a lighting system, a water treatment plant, adding plants to improve the visual aspect of a group of zones, etc.; for this reason one cannot attribute the impact made by the sites solely to the projects co-financed by the ERDF;

(c) an industrial or small business area can only be satisfactorily occupied several years after the project has been completed, and the length of time involved depends on the project's quality(12) and also on the phase in the economic cycle;

(d) the contribution made by a new site to regional development will depend in particular on its occupancy rate, and also on relocations by undertakings in the region to the new site, which reduce its impact (diversion effect(13)); this impact can also include beneficial effects which are rather diffuse and are usually not evaluated: increasing the effectiveness of undertakings located in a better environment and reducing congestion in urban centres, which can in turn make redevelopment measures necessary.

43. Table 2 shows, where figures are available, the number of undertakings and jobs in the industrial and small business areas surveyed. Employment varies from eight individuals per occupied hectare in Sicily (I) to 26 individuals in Nord-Pas-de-Calais (F), and average surface area used varies from one hectare per undertaking in Nord-Pas-de-Calais (F) to five hectares in Sicily (I).

44. Employment per hectare is gradually decreasing, mainly because of the increasing concentration of capital in production. It varies significantly according to the location, the features of the site and the activities carried on on the site. In Limburg (B), it fell from 25 individuals per hectare at the beginning of the 1980s to 17 at the end of the decade, and to 11 at present. In Sicily (I), there were 10 jobs per hectare in 1989, in comparison with eight now. Moreover, in this region, which is the only one of those visited where data of this kind are collected, areas which had been equipped and sold but were "inactive"(14) represented, in 1997, 10 % of the surface area equipped and sold. In one area which was examined, the "inactivity" rate was as high as 16 %. Although the rate of job creation that may be expected is not as high as in the case of the setting-up of new businesses, company relocations may help to improve the urban environment and boost productivity.

45. Relocations of companies do not create as many jobs as when new companies are set up but they may, in certain cases, help to improve the urban environment and increase productivity. The regions visited do not collect data on relocations, except for Limburg (B) where, on the basis of the relatively incomplete data concerning the sites visited by the Court, the relocation rate is estimated at between 50 % and 90 % of the undertakings which have moved in. In Sicily (I), where the decongestion of urban centres was one objective, the undertakings located in the industrial zones appeared to result primarily from the relocation of former operations.

46. In Limburg (B), during the 1986 to 1997 period, 122 undertakings out of 152 emerged successfully from the five "nurseries", which is equivalent to a rate of 80 %. However, there is no information available on the jobs created by these undertakings at the end of their period of stay. In Nord-Pas-de-Calais (F), at one "nursery" which was examined only seven undertakings out of 15 were able to leave and open for business on the outside during the 1992 to 1998 period. Finally, between 1993 and 1998, 148 undertakings out of 192 were able to graduate successfully from seven other "nurseries" (rates varied between 54 % and 89 %), creating 627 jobs.

47. It was not possible to find many synergies within the areas. Such synergies are only likely to come about in a limited number of cases, such as technological zones where undertakings have set up either through internally-generated development or precisely because of potential relationships with universities, research centres or neighbouring undertakings. Another case, which is often seen in the automobile sector, is where subcontractors collect around large undertakings which had arranged for neighbouring sites to be reserved for them at the time they moved in.

Selling prices for infrastructure

48. ERDF aid is not granted directly to the companies that set up on co-financed industrial sites (see paragraph 15). It is therefore necessary to find out whether this assistance is passed on to benefit the undertakings, in the form of a reduction in the normal selling price, or if it benefits the developer, who may possibly make a profit on the project which might be reinvested in other operations. In the first case competition aspects must be considered, first concerning the combination of subsidies (see paragraphs 49 to 55) and secondly, within the relevant region, concerning competition between developers who are receiving ERDF assistance and those who are not (see paragraph 56). In the second case, the issue of the purpose of the aid must be borne in mind (see paragraphs 57 to 62).

Competition

49. Regional State aid is designed to compensate for the disadvantages which certain regions may have, and is subject to an aid intensity ceiling, above which it might cause distortions of competition. In order to make the various forms of aid comparable with one another and aid intensities comparable from one Member State to another, aid notified must be converted into "net grant equivalent (NGE)"(15). Any State aid included in the sale of developed sites must be included in this calculation. This State aid should as a rule be evaluated on the basis of the market value.

50. In the Commission communication on the sale of land and buildings by public authorities, two systems are described for establishing sale prices in a way that automatically precludes the existence of State aid(16). The first is sale through an unconditional bidding procedure, comparable to a public auction. The second system is based on establishing a market value and provides for one or more independent experts to carry out, prior to the sale, an evaluation on the basis of generally accepted market indicators and valuation standards.

51. Among the regions visited, sales of plots of land could only be categorised within one of the two systems laid down in the Commission communication in Limburg (B) and in Wales (UK). The selling prices realised correspond to the valuations of State officials in the former or of experts from the Royal Institute of Chartered Surveyors in the latter.

52. In the Autonomous Community of Valencia (E) and in Nord-Pas-de-Calais (F), the price is set by the site developer in accordance with its own aims. More broadly speaking, in France there are no restrictions on the selling price or the levels of rent for developed land and free transfers or sales for the token price of FRF 1 are permitted. In Sicily (I), the regional government sets the selling price for plots each year. It has not been possible to reconstruct either the method used to fix these prices or their relationship with the project's development costs.

53. In Thuringia (D), although the regulations provide that the developed plots should be sold, after being advertised, for the best price offered, in reality the municipalities fix the selling price in a discretionary manner. However, in contrast to the other regions visited, a system has been put in place which is designed to prevent both developers keeping the subsidies they have received as a profit and the plots being sold at excessively low prices. If the selling price exceeds the net cost price (after deduction of the various subsidies), adjusted so that the municipality bears 10 % of the development costs, the latter will have to repay the surplus to the partners involved in financing the project, namely the Federal Government, the Land and the European Union; however, if the plots are sold at a price which is too low to cover the cost price, the difference will be regarded as a subsidy and will be added to any other assistance the undertaking has received. This balance sheet must be drawn up no later than 10 years after the beneficiary's final declaration of expenditure has been submitted. To date, this system has not been put into practice in this Land.

54. It is clear from the procedures observed for establishing selling prices for subsidised infrastructure that in most of the regions neither of the systems which preclude State aid are applied. According to the Commission communication, the sales should therefore be notified, but this is not done (see paragraphs 49 and 50).

55. Moreover, in the majority of cases State property-developing bodies are in an extremely dominant position in supplying this infrastructure, and as a result set the prices, frequently below cost price. Thus, the market price cannot be considered as a true market price, and using it to establish whether there is a distortion of competition is only of limited help.

56. ERDF aid is not neutral in its impact on competition between developers from an eligible region to attract undertakings, in particular through setting selling prices. For this reason, it is all the more necessary to observe the principle that such aid should only be granted within an overall framework of regional development which sets out clear selection criteria for subsidised projects.

The role of ERDF aid

57. ERDF aid for the development of industrial sites is designed to encourage the setting-up of enterprises and the creation of jobs in such places. Although the project's socioeconomic objectives may have been achieved, where the ERDF contribution is not passed on in its entirety to the undertakings it will enable the developer to reduce its loss, to break even or to make a profit. In these latter two cases, where there is no financial contribution by another partner, the ERDF will, ultimately, be the sole financial contributor to the project. Any profit realised by the developer after it has received the subsidies and the sales proceeds, which the ERDF does not receive, will accrue to its own resources.

58. The fact is that developing industrial sites may generate profits. After the first stage, when the developer incurs the expenditure needed to acquire the undeveloped land and to develop it, for which the ERDF grants assistance, there is a second phase during which it alone receives the sales proceeds of developed plots; these receipts do not go towards covering the operating costs. If the developed plots are sold at the net cost price (after deduction of the subsidies), the developer will recover its contribution to the project's financing, and all the non-returnable aid paid, and specifically the ERDF assistance, will be profit which may in some cases contribute to the financing of other operations which may be far removed from the ERDF's objectives.

59. For example, in Limburg (B), the developer's final profit for two industrial or small business area projects where all the land had not yet been sold, out of five projects audited, is estimated at 26 % of the total cost for one of them and 16 % for the other (average ERDF assistance: 10 % of the total cost). In Thuringia (D), three projects were audited where not all the land had yet been sold. Of these, one was expected to show a profit in the region of 21 % (ERDF assistance: 18 % of the total cost). In the Autonomous Community of Valencia (E), the developer of one site, a State-owned company with activities other than developing industrial sites, made a profit corresponding to 30 % of the project's development costs (ERDF assistance: 65 % of the total cost).

60. It is difficult to discern the impact of non-returnable ERDF aid in the field of industrial site development, either with regard to its effect on the price at which developed plots are sold to undertakings or the use to which sales proceeds are put.

61. In addition, since in theory the development of sites generates receipts which constitute a recouping of the investment, this should be considered when granting ERDF subsidies, as loans might in some cases replace subsidies.

62. However, developing new sites in parts of a region which have few facilities or where there is a particular need for development (location effect) and redeveloping former sites which are now abandoned and derelict, and sometimes polluted, are special cases in that they may require expenditure, rendering it impossible to put them on the market at a price which is competitive with other sites not burdened with similar costs. In these cases ERDF aid is justified in order to compensate for the handicap of these costs and, as a result, to make the selling price of the plots competitive in comparison with the other sites. Here, ERDF assistance has an incentive effect which is of general public benefit as regards the location of infrastructure in certain parts of the region and the development of derelict sites (see paragraph 64).

Value added by the Community programmes

63. ERDF measures provide considerable financial resources to the final beneficiaries and are chiefly viewed in this light. However, the nature of industrial sites facilities suggests that they would still have been completed without assistance, although probably in smaller numbers. For this reason in particular, it is important that infrastructure is developed within the framework of a structured regional development plan.

64. With regard to the redevelopment of former sites, ERDF aid seems to be more of a deciding factor. Renovating disused industrial land seems less urgent and is often only considered after several other priorities have been attended to. In this case, the aid makes it possible to lessen significantly the project's financial burden on the final beneficiary. In the same way, some qualitative projects would not have been carried out without ERDF aid: these are environmental or architectural projects which do not fulfil any of the regions' essential needs, but are additional factors in attracting undertakings. In some regions, the ERDF has also supported the setting up of databases on industrial sites.

65. Finally, very few synergies were found between the measures in favour of industrial sites and other Community measures or actions. Although most operators connected with developing industrial sites in the regions involved are at the centre of economic life, they are not aware, with the exception of those responsible for creating business "nurseries", of the possibilities offered, for example, by the European Social Fund or the European Investment Bank (EIB).

Conclusion

Financial data

66. On several occasions, the Court has found it difficult to quantify the sums allocated by the ERDF to specific fields and target groups. The same difficulty has been encountered with respect to the development of industrial sites. An improvement in the quality of the data sent in by the Member States would be necessary in order to gain a better awareness of the use of ERDF resources, and one way of achieving this would be for assistance programmes to be more homogeneous and detailed in structure (see paragraphs 5 to 7).

Development and restructuring of the regions

67. Initiatives for developing industrial sites are often the prerogative of the municipalities and occasionally of bodies that group together several municipalities and State-owned companies or consortia, but there is no coordination between these operators (see paragraphs 15 to 18). However, ERDF co-financing of projects should be dependent on a Community regional development or restructuring policy which supports action programmes conceived at a regional level and financed by State funds. The ERDF should therefore avoid initiatives which emerge at the most decentralised level and should instead encourage the region's development or restructuring on the basis of an overall regional plan (see paragraphs 13 and 14).

68. The regions visited by the Court do not have land management priorities for industrial sites, except for a few large infrastructure items. Measures for developing such sites are weak, both with regard to their objectives (see paragraphs 29 to 33) and the selection criteria for projects. It is vital that an overall regional development framework be put in place, including selection criteria for projects, and this is all the more necessary in view of the competition which can arise between State property developers (see paragraphs 34 to 39).

69. Because of the increasing numbers of these facilities, an increase partly encouraged by the ERDF (see paragraph 63), on some occasions a desire has arisen to ascertain the region's stock (see paragraphs 40 and 41) and some survey initiatives have been taken. However, they have not always been sufficiently well-targeted to be used for forecasting purposes or to provide an overall picture of needs. Sometimes, the programmes still set out needs even where they have not actually been evaluated (see paragraphs 19 to 28).

70. In order to increase the effectiveness of ERDF assistance in this field, the Commission should encourage the Member States to create analytical and monitoring tools, including a regularly updated inventory of industrial sites and statistics on site sales. These tools ought to enable site development to be actively managed, in the following ways:

(a) drawing a distinction between areas with too much availability and those with not enough, in order to redirect State aid co-financed by the ERDF, according either to priorities, or to demand;

(b) identifying those parts of an area where, although the supply of facilities may satisfy existing needs, it is none the less insufficient with respect to their surface area or population and where it is necessary, for this reason, to carry out actions designed to increase economic activity, since these are also part of the area eligible for ERDF measures;

(c) monitoring the satisfactory use of these facilities, in a way that is more relevant than that permitted by current impact indicators, and possibly redirecting supply in quantitative and qualitative terms as a result;

(d) channelling some development efforts towards avoiding a deterioration in the supply, which can have detrimental effects on the environment and on land use planning.

71. These data will only be useful if the regions set and observe selection criteria for projects, drawn up in accordance with a regional development plan.

72. In the specific case of the business "nurseries", there should be a maximum length of stay for undertakings so that these facilities can remain focused on their aim of helping to create businesses (see paragraph 28) and statistics should be kept on the success rates of undertakings which have benefited from nurseries and on jobs created, in order to measure their contribution to regional development (see paragraphs 33 and 46).

73. The evaluators of the ERDF measures should also investigate whether there is a consistent site development approach and an analysis of needs (see paragraphs 40 and 41).

Impact of the ERDF aid

74. Since ERDF aid is not granted to the undertakings which move onto subsidised sites, the setting of the selling prices for this infrastructure is a key element in analysing the impact of the aid (see paragraph 48).

75. As far as competition is concerned, this audit has revealed a lack of transparency in the setting of selling prices for developed plots which the Commission ought to look into, and a lack of a genuine market, even though the market value is supposed to be used as a reference point when setting the selling prices (see paragraphs 49 to 56).

76. No link can be shown between ERDF assistance and selling prices. For this reason, the Commission should clarify the objective of this subsidy. It is possible for a project to end up breaking even or even showing a profit, which may imply the lack of a counterpart to the ERDF co-financing and the use of this profit for other purposes which may diverge from the ERDF's objectives. Since in theory the development of sites generates revenue which constitutes a recouping of the investment, this should be considered when granting ERDF subsidies, since loans might in some cases replace subsidies (see paragraphs 57 to 62).

77. However, ERDF aid in the form of subsidies is of particular additional value where a location effect is sought or in the redevelopment of abandoned former sites, since these projects may initially be perceived as less urgent. In such cases, the costs of development or treatment may prevent sites being put on the market at a competitive price and the subsidy, which in these cases acts as compensation for this handicap, should have an incentive effect as regards location or the development of derelict sites (see paragraphs 62 and 64).

This report was adopted by the Court of Auditors in Luxembourg at the Court meeting of 9 December 1999.

For the Court of Auditors

Jan O. Karlsson

President

(1) These are maximum amounts up to which expenditure may be incurred and are laid down in the programmes' financing schedules.

(2) See paragraph 13 of Special Report No 3/96 on tourist policy and the promotion of tourism (OJ C 17, 16.1.1997) and paragraphs 6.49 to 6.52 of the Annual Report concerning the financial year 1996 (OJ C 348, 18.11.1997; implementation of measures in favour of undertakings, SMEs in particular, in the context of the ERDF).

(3) The territory of the European Community is subdivided according to a nomenclature of territorial units for statistical which comprises 71 regions at NUTS I level, 183 at NUTS II level and 1044 at NUTS III level.

(4) See in this connection the observations already published by the Court in its Annual Reports on the financial years 1994 (paragraph 4.18b) and 1992 (paragraph 6.43) as well as in its Special Report No 1/95 on the financial cohesion instrument (paragraph 4.18).

(5) Commission Regulation (EC) No 1681/94 of 11 July 1994 concerning irregularities and the recovery of sums wrongly paid in connection with the financing of the structural policies and the organisation of an information system in this field (OJ L 178, 12.7.1994, p. 43).

(6) In the regulations the final beneficiaries are the bodies or public or private companies responsible for commissioning the work (the principal); in the case of aid schemes and in the event of the granting of aid by bodies appointed by the Member States, the final beneficiaries are the bodies that distribute the aid.

(7) The rate is set by the local authorities and relates to the rental value of the tangible fixed assets and a proportion of salaries paid. Competition between municipalities to attract undertakings can also have the effect of lowering the level of this tax rate.

(8) The occupancy rate can be calculated by comparing the surface areas occupied by the undertakings with either the total zone surface area (whether equipped or not) or the total equipped surface area (occupied or vacant). The first ratio is useful as regards determining surface areas set aside for undertakings in land-planning schemes and the second, which was used for this report, for deciding on the surface areas that need to be developed.

(9) See paragraph 7.34 of the Annual Report concerning the financial year 1979 (OJ C 342, 31.12.1980) and paragraph 7.26 of the Annual Report concerning the financial year 1980 (OJ C 344, 31.12.1981).

(10) Coordination by a central body enables "nurseries" to be guided towards specialising (e.g. technological or small business "nurseries", etc.), links to be established at group level with the banking and finance sectors, universities and laboratories, increased exchanges of experiences, etc.

(11) Jobs in an undertaking in a certain region which has relocated to a new area in the same region will be counted under the heading of gross employment connected with the creation of the new area, even if the undertaking concerned does not increase its staff. It would not be counted as net employment, however.

(12) On this topic, see "New location factors for mobile investment in Europe", Netherlands Economic Institute in cooperation with Ernst & Young, 1993, a study launched by the European Commission's Directorate-General for Regional Policy and Cohesion. This report speaks of a combination of subjective factors which are not costly (market, labour, transport infrastructure, etc.), and financial factors (labour, land and buildings, assistance).

(13) See paragraphs 6.1 to 6.75 of the Court's Annual Report concerning the financial year 1987 (OJ C 316, 31.12.1988).

(14) That is, areas equipped with buildings which are empty or belong to undertakings which have ceased operating.

(15) See "Guidelines on national regional aid" (OJ C 74, 10.3.1998, p. 9).

(16) Commission communication on State aid elements in sales of land and buildings by public authorities (OJ C 209, 10.7.1997, p. 3).

The Commission's replies

Summary

The Commission does not regard the development of industrial sites as an objective of structural policy in itself. Structural Fund programmes do not always contain specific measures for the development of industrial sites but use such sites as a means of achieving other goals. Consequently, spending in this field may vary considerably from one programme to another (paragraphs 1 and 2).

Implementation of the Structural Funds is based on the principles of programming and partnership; accordingly, the Commission collects data at the level of the measure and Member States are free to design measures according to their needs and choices. In the new programming period, measures will probably be designed in a more detailed way, which will improve monitoring and reporting on activities financed by the Structural Funds (paragraphs 5 to 7).

The Commission will examine the irregularities detected by the Court (paragraphs 11 and 12) and take the necessary action.

One of the main principles of structural policy is systematic planning together with close monitoring of implementation (paragraphs 13 and 14). In accordance with partnership, the regions should decide which action is the most efficient and suitable for them (paragraph 19).

Despite some progress, the Commission is aware that there are shortcomings as regards goals and indicators (paragraph 31). Project selection is the responsibility of the Member State and this selection must always comply with guidelines and priorities agreed with the Commission (paragraph 37).

As regards evaluation, it is difficult to measure and compare the impact of measures concerning industrial sites. This point received particular attention in some interim evaluations (see paragraphs 40 and 41 (e.g. Thuringia)). The new Structural Funds General Regulation explicitly requires mid-term evaluations.

As regards competition (paragraphs 49 and 50), the transfer of Community public resources to a public body does not constitute a State aid within the meaning of Article 87(1) of the Treaty. It is a matter for the Member State alone to decide subsequently whether sales by the beneficiary public authority will take place under market conditions or on favourable terms.

ERDF money must not be used to make profits - where such profits occur, they should be reinvested (paragraphs 57 and 58). The new regulations favour the use of repayable forms of financing.

The development of industrial sites must always form part of a structured regional development project based on a sound definition of priority guidelines in the SPDs and programming complements.

Introduction

1 and 2. The Court of Auditors is right to say that the Structural Funds Regulations make no explicit reference to the development of industrial sites. (In more general terms Article 1(b) of Regulation (EEC) No 4254/88 as amended defines as one field of assistance for the ERDF "[infrastructure] investment relating to the regeneration of areas suffering from industrial decline, including inner cities"). The Commission does not regard the development of industrial sites as an objective of structural policy in itself. However, developing such sites has played and, to a lesser degree, may continue to play a role of some significance within a wider policy to support and develop business activities in assisted regions, particularly those of SMEs. For this reason, Structural Funds programmes do not always contain specific measures for the development of industrial sites but use industrial site development as a means of implementing other measures designed to assist business development. Spending in this field may vary considerably from one programme to another.

Budgetary and physical data

5 to 7. The Commission collects data from the Member States about the implementation of Structural Funds programmes using a list of categories of assistance. However, one consequence of the principle of programming and partnership (introduced in 1989), as opposed to the management of individual projects by the Commission, is that the most detailed level at which data may normally be collected is that of the measure. Gathering data about activities below measure level (such as the development of industrial sites) has proved very difficult.

Title II (Programming) of the new Structural Funds General Regulation confirms this approach which gives the Member States - within their overall strategy for implementing Structural Funds programmes - the freedom to design measures according to their needs and choices rather than following a standard model and nomenclature. Over a seven-year programming period, the Member States need a degree of flexibility in programme implementation in order to be able to respond to developments. Otherwise there would be the risk of returning to project approval at Commission level.

The experience gained since the introduction of programming has led the Commission to draw up a revised and more detailed categorisation of areas of assistance for the new programming period. It is likely that measures will in future be designed in a more detailed way as the need to get Commission approval even for minor changes disappears under the new Regulation. Member States will have to demonstrate the link between categories of assistance and the proposed measure. In conjunction with the new information system replacing Garfield, this should improve monitoring and reporting on the activities of the Structural Funds, facilitate follow-up and provide a sound basis for evaluations. In this context rehabilitation of old sites will be a category distinct from other activities to support SMEs.

8. The share of total ERDF aid actually going into the development of industrial sites may in practice be lower than indicated by the Court.

Legality and correctness

11. The Commission will look in detail at the cases raised by the Court so that it can take appropriate steps.

12. After checking the notifications of irregularities sent by the Member States under Regulation (EC) No 1681/94, the Commission can confirm that no irregularities were notified under that Regulation for projects either in Thuringia (D) or Sicily (I). It will approach the relevant departments in those two Member States to obtain details of the situation before deciding what, if any, action to take.

Since 1994, it has noted a considerable increase in the number of cases notified by the Member States under Regulation (EC) No 1681/94, although the situation is still not ideal. The Commission is taking steps to improve the situation.

Process of developing industrial sites part-financed by the ERDF

13 and 14. The Court rightly calls for a systematic approach in programme establishment and implementation, including criteria for project selection as stipulated by the regulations. For measures where Community aid consists essentially of the part-financing of an existing aid scheme (previously approved by the Commission), the criteria for project selection form part of such an aid scheme (see paragraph 34).

Systematic planning for future needs and developments is one of the main principles of structural policy. Economic reality will, however, often move faster than public administrative planning (of which the developments in the new German Länder over the past 10 years are a significant example). Close monitoring during the whole cycle of programme implementation helps to ensure that strategies developed at the start of programmes keep up with developments. None the less there will always remain a possibility of error, easily criticised in hindsight, especially when economic framework conditions change rapidly or adversely.

Ex post evaluations will yield useful results only if they are done sufficiently long after the end of a programme lifespan to take account of longer-term effects. (In paragraph 32 the Court itself states that ex post evaluations come too early). The programming cycle, where in effect new programmes are designed while the current ones are still being implemented, implies that the findings of ex post evaluations may often be taken into consideration only in the next or even later programming phase. Nevertheless, Articles 14(2) and 42 of the new Structural Funds General Regulation explicitly require mid-term evaluations. These will be useful not only for adjusting ongoing programmes but also for designing programmes for the following programming period.

Organisational context

15. It is correct that, under Structural Funds rules, businesses moving into developed sites are not regarded as final beneficiaries. Under these rules, except in the case of aid schemes, these are considered to be the promoters (project developers).

Analysis of needs

19. In order to improve analysis of regional needs for industrial sites, the Court of Auditors proposes designating a body to undertake this work. The regions should have the right to implement the approach which is the most efficient and suitable for them.

It is essentially the role of the partnership (monitoring committees and in the future managing authorities) to determine the needs of an assisted area, to ensure the coordination of public project promoters and, if necessary, adjust planning to unexpected developments during the lifetime of a programme. In certain cases, a degree of competition between public promoters can even have positive effects on the quality of industrial sites offered to businesses. Articles 35(3), 36 and 42 of the new General Regulation reinforce the role of partnership whilst facilitating its implementation.

Industrial and craft areas

24. Figures on the availability of industrial sites and the average annual rate of sales are important indicators when determining further needs for site development. A purely statistical interpretation of such figures, however, would neglect the quality aspects of sites, such as the geographical location within the assisted region, access to transport, availability of other resources (raw materials, suppliers) and infrastructure, possible synergies with related businesses, training or research institutions present. In some countries even local business taxation may be a factor. Not all sites will therefore be equally suitable or attractive for all kinds of business activity.

Incubators and advance factories

28. Even now many incubator centres gradually increase the rent to the market level over the time that an enterprise stays in them. This creates an incentive for viable businesses to leave. It also provides essential income to complement the public subsidies to the incubator centres and help them finance the services they offer to new businesses.

The difficulties of identifying economic needs by means of the administrative procedures described above are greater in the case of incubator centres, which are normally for the exclusive use of new businesses, often specialising in high-tech and/or services. By contrast, normal industrial sites will regularly accommodate a large share of existing businesses, whose needs are more easily established. In this context, the fact that demand may also be created by attractive supply should not be neglected.

Goals and indicators

31. The Commission is aware of shortcomings as regards the quantification of goals and indicators, despite the considerable progress made during the current programming period. For the new period, in accordance with Article 36 of the new General Regulation of the Structural Funds, the Commission has drawn up a methodological document (Working Paper No 3) on the design and utilisation of the monitoring indicators.

32. As stated above, the findings of ex post evaluations and reliable data on the long term effects of Structural Funds measures are often available only several years after the end of a programme. This leads to the general dilemma that they can be taken into consideration only in the next or even later programming phase. The results of mid-term evaluations, however, often provide good indications for the adaptation of ongoing programmes and the preparation of the next series of programmes.

Project selection criteria

34. The Commission shares the Court's view that project selection criteria should play a strategic role in the implementation of Structural Funds measures. Programmes did not always reflect this in the past. This does not, however, necessarily mean that implementing authorities did not apply such criteria, as project selection is the responsibility of the Member State. In the new programming period, this division of responsibility will be further strengthened by the fact that detailed information on the implementation of measures will be contained only in the programme complement to be established by the Member States after the Commission decision on their programme (and not requiring Commission approval).

37. The text of these operational programmes lists the projects planned and the goals of the measures to be taken. In any event, the choice of these projects must reflect the guidelines and priorities agreed with the Commission.

Evaluations

Interim and ex post evaluations as part of Community assistance

40 and 41. In general, the evaluations carried out, particularly of Objective 2 programmes, look at the relevance of the strategy adopted and its economic impact on the areas concerned. The question of industrial sites was looked at more specifically in some interim evaluations - such as the programme in Thuringia which includes quantified indications in physical terms (hectares prepared, jobs per hectare, etc.). However, their specific features (size, type of activity, etc.) make it difficult to assess and compare the impact of measures for industrial sites. Securing a better understanding of how effective these measures really are would require a series of in-depth thematic studies based on a sample of projects.

As regards the next programming period, the question of needs for industrial sites should be raised when the development and conversion plans for the areas concerned are being drawn up and negotiated.

Impact of industrial sites and their part-financing

Competition

49 and 50. Sales to businesses of lots developed by public authorities on terms which do not reflect market conditions constitute State aid and must be notified and approved under Article 88(3) of the Treaty, except for cases covered by the de minimis rule (Commission communication on de minimis aids (OJ C 68, 6.3.1996)) which do not constitute State aids within the meaning of Article 87(1) of the EC Treaty. The Commission communication on sales of land and buildings by public authorities (OJ C 209, 10.7.1997) enables market conditions to be ascertained.

The possible existence of State aids for the purposes of this Report may be considered at two levels. The first concerns the transfer of funds from the Community to the regional or local authority (ERDF assistance) while the second, at a subsequent stage, concerns the sale of land developed by that regional or local authority to purchasers, normally businesses.

At the first level, the transfer of Community public resources to a public body, even on the condition that they are used to develop land for subsequent sale to businesses, does not constitute State aid within the meaning of Article 87(1) of the Treaty, since the transfer of financial resources and the development does not confer any advantage on the businesses.

It is up to the Member State alone to decide whether sales by a public authority which has received Community aid are to be made on market conditions (so that the assistance simply increases local supply), or on more favourable terms (discount on the market price). In the second case, the Member State must comply with the above obligations stemming from Article 88(3) of the Treaty.

As long as ERDF assistance does not explicitly provide for the sale of land on favourable terms as an integral part of the measure, it does not include an aid element; this appears to be the case for most of the programmes concerned.

Nevertheless, irrespective of the question of whether the measure itself includes a sale on favourable terms or not, the Commission has in some cases insisted on the inclusion in the texts of operational programmes of certain clauses concerning the conditions of sale, either to remind the Member State of its obligation to notify, to cover cases where a sale on favourable terms forms part of the assistance from the ERDF as an integral part of the measure, or to note the commitments given by the Member States with regard to the conditions of sale. These clauses are:

- "sale on market conditions"; this fairly general clause occurs without further detail in a large number of Objective 1 operational programmes for 1994 to 1999; the 1997 communication sets out what is meant by "sale on market conditions". This clause is still binding on the Member States outside the measure in question and so is binding even if the sale of the land does not form part of the measure itself,

- "sale in accordance with the de minimis provisions"; the granting of aid under the de minimis rules is the responsibility of the Member States alone (see the 1996 communication); failure to comply with the de minimis provisions suggests that the aid was granted illegally. This clause too commits the Member States beyond the measure itself,

- "reference to an approved State-aid scheme"; this clause is included where the measure in the operational programme part-finances a State-aid scheme approved by the Commission which itself includes conditions for the sale of land, which may themselves be favourable. This clause too commits the Member States beyond the measure itself.

If aid is granted after the end of ERDF assistance, i.e. when the lots are sold to businesses, on the basis of an "aid" approval by the Commission, the aids in question must comply with the ceilings on aid intensity and, if they are combined with other aids, the ceilings on aggregation laid down by the Commission.

Role of ERDF aid

57 and 58. The Commission considers that ERDF resources (with the exception of productive investment) must not be used to make profits. Where such profits occur they should be reinvested for similar eligible purposes in the same area, or if that is not possible, ultimately be repaid(1). At project level, only the Member State can ensure this. When in the past the Commission has learnt of such cases, it has insisted on proper reinvestment.

59. As the regional authorities informed the Court in the Valencia case, revenue from the sale of this industrial site was reinvested in other industrial sites meeting the same criteria which, for that reason, were not part-financed by the ERDF.

60 and 61. The Commission shares the view of the Court that for this kind of investment alternative forms of financing should take the place of non-repayable grants. This means that the impact of the ERDF will be more clearly focused on assisting business start-ups and would avoid the risk of undesirable repercussions when the assets are resold. The new regulations favour the use of repayable forms of financing. Furthermore, Article 28(3) now stipulates clearly that "assistance repaid... shall be reallocated to the same purpose".

62. The Commission generally gives preference to the refurbishment and reutilisation of derelict sites to the development of greenfield sites, not least for environmental reasons. The high costs involved in such operations often require public financial assistance to make these projects competitive.

Value added by Community programmes

63. The implementation of projects as part of a clear strategy of which they form part is central to the concept of programming. Measures within programmes should be designed so that this is evident. The Commission is careful to ensure that the development of industrial sites forms part of a structured regional development project, based on a sound definition of the priority guidelines in the SDPs and programming complements.

65. Monitoring Committees, especially when representing a wide partnership, are excellent forums where synergies between the different actors can be developed. It is there that detailed information about the possibilities offered by the ESF or the EIB can be given by representatives of the Commission or the Bank.

Conclusion

Financial information

66. As stated above the Commission has taken steps under the new Regulations to develop more refined monitoring and reporting by Member States on individual categories of assistance. The additional administrative tasks imposed on Member States by the collection of such data should, however, remain proportional to the value added in terms of better programme implementation.

Development and conversion of regions

67 to 71. The development of industrial sites is not an objective of regional policy in itself. It may nevertheless constitute an important tool for the development of business activities and form part of the wider strategies in regional plans and programmes to achieve this objective. A systematic approach to project implementation (including the development of industrial sites) should be an important part of programming and the development of measures. The implementation of Community structural policy in the regions must be based on the specific analysis of regional needs and integrated into an overall strategy for the development of the region concerned. It should include criteria for project selection related to the needs analysed. The SPDs and programming complements set out the goals and priorities of the programmes. The projects financed by these programmes must meet the criteria defined by these goals and priorities.

The new Regulations underline the need for improved regional analysis contained in programming documents and monitoring. Development of new industrial sites will in future be included in programmes only where a clear need for them and a strategy can be demonstrated. For environmental reasons and to support urban regeneration, the refurbishment of existing derelict sites is to be favoured wherever possible.

72 and 73. The Commission remains convinced that industrial site development should always be seen as one part of a wider policy to assist business development, in particular of SMEs. As regards the maximum duration for staying in business incubator centres, the approach described in the Commission's reply to paragraph 28 would be a more appropriate recommendation than setting a cut-off point.

Impact of ERDF aid

76. As reflected in the new Regulations, the Commission favours the use of repayable forms of Structural Funds assistance for this kind of activity. Such an approach would also prevent ERDF aid flowing back to project promoters as a potential benefit once the sites are resold.

(1) Article 2 of the Financial Regulation; Article 17(3) of the Coordination Regulation (for the current period); Article 30(4) of the General Regulation (for the period 2000 to 2006); Sheet 9 (bookkeeping entries of other receipts) of the SEM 2000 rules on the eligibility of expenditure under the Structural Funds.