52013SA0005

GLOSSARY

AADT : Annual average daily traffic — most commonly used traffic counting method. The figure indicates the average number of vehicles crossing the measuring point during a 24-hour period. The figure therefore takes into account traffic for peak and low seasons of 1 year.

+++++ TIFF +++++

Express road :

A road reserved for motor traffic, accessible from interchanges or controlled junctions only, and which:

(i) prohibits stopping and parking on the running carriageway(s);

(ii) does not cross at level with any other road, railway or tramway track or footpath;

(iii) in most cases has at least four traffic lanes. The most common speed limit on express roads is 110 km/h.

Greenfield projects : Projects that are built on undeveloped land in a city or rural area either used for agriculture, landscape design or left to nature. Renovations sometimes have greenfield elements, e.g. when a carriageway is added to an existing road.

ERDF Major Project : A project co-financed from the European Regional Development Fund for which the total eligible cost is more than 50 million euro.

+++++ TIFF +++++

Motorway :

A road specially designed and built for motor traffic, which does not serve properties bordering it, and which:

(i) is provided, except at special points or temporarily, with separate carriageways for the two directions of traffic, separated from each other by a dividing strip not intended for traffic or, exceptionally, by other means;

(ii) does not cross at level with any road, railway or tramway track, or footpath;

(iii) is specially signposted as a motorway.

OP : Operational programme.

Public Procurement by mathematical formula : Tender process that is designed to exclude unsustainably low offers. The mathematical formula calculates the price limits below which offers would be excluded from further evaluation.

Road alignment : The route of the road, defined as a series of horizontal tangents and curves.

EXECUTIVE SUMMARY

I. For the period from 2000 to 2013 the EU has allocated around 65 billion euro from the ERDF and Cohesion Fund to co-financing the construction and renovation of roads.

II. The Court audited 24 European Regional Development Fund (ERDF) and Cohesion Fund road investment projects in Germany, Greece, Poland and Spain to assess whether they achieved their objectives at a reasonable cost. The total cost of audited projects exceeded 3 billion euros.

III. The Court concluded that the road projects partly delivered results and fulfilled their purpose. In particular, all the road projects audited provided travelling time savings and improved road safety. However, several projects did not deliver the planned returns on investment as forecasted in cost-benefit analyses (CBAs). Moreover the assessment of the additional economic development deriving from the realisation of the projects was impossible due to lack of measurable indicators.

IV. The Court also concluded that insufficient attention was paid to ensuring cost-effectiveness of the projects. Most of the audited projects were affected by inaccurate traffic forecasts. The type of road chosen was not best suited to the traffic it carried. Motorways were the preferred option by beneficiaries, even for sections where express roads could have solved the traffic needs. Compared to the initial plan, the average cost increase was 23 % and time overruns averaged 9 months or 41 % when compared to the initial deadlines agreed in the construction contracts.

V. The lowest roadway construction costs were found in the projects audited in Germany, followed by Greek, Spanish and Polish projects. Cost per use was also cheapest in Germany while the most expensive was in Spain. Some procurement practices did not deliver optimal costs.

VI. The Court recommends that:

(a) The Commission should analyse the differences in roadway construction costs between the Member States in order to identify the causes of significant price differences thereof and assure that the best practices are applied in future.

(b) The EU co-financing of road projects should be made subject to the existence of clear objectives that are accompanied by indicators for travel time savings, gains in road safety, capacity improvements and effects on the economy. The co-financing should depend upon the use of reasonable and cost-effective technical solutions that are in line with the best practices identified and the steps taken by the Member States to ensure international competition on national and/or regional construction markets and focusing procurement systems on delivering the most economical offers.

(c) The Commission should promote the exchange of best practices between national road administrations concerning optimum technical solutions for road projects based upon reliable traffic forecasts.

(d) The Commission should consider the establishment of a European Union-wide unit cost information database for engineers preparing estimates for new projects, in order to help beneficiaries to lower the procurement prices.

INTRODUCTION

COHESION POLICY FUNDS AND INVESTMENTS IN ROAD PROJECTS

1. Enhancing accessibility is of key importance to strengthening national and/or regional economies, cohesion and competitiveness. Improving the road network is an important element to increase mobility for people and goods and therefore create the conditions for economic growth.

2. As the road investments are financed from the Cohesion Policy funds, they fall under the Shared Management set-up. Within Cohesion Policy, the Commission issues guidelines for drawing up Operational Programmes (OPs). At the beginning of the programming period, the Commission negotiates and approves the individual OPs proposed by the Member States. Its task is also to supervise the setting up and operation of management and control systems in the Member States. Later, its main role is to monitor the implementation of the OP, but it is not involved in the day-to-day management of individual projects. The Commission receives annual implementation reports from the Member States and participates in monitoring committees [1]. The Commission is ultimately responsible for the implementation of the budget [2].

3. Each Member State decides which managing authorities are involved in the management of EU co-financing for roads, as roads expenditure can be handled at all levels of government, including ministries of transport, national or regional road agencies, regional administrations, county administrations and local government.

4. Infrastructure investments in the Member States are mainly financed from their national funds. The Organisation for Economic Cooperation and Development (OECD) has calculated that the EU Member States spent some 66 billion euro on road investments in 2010 alone [3]. In the course of the last two programming periods, the EU has provided on average some 4,9 billion euro per year for roads investments through Cohesion Policy funds and TEN-T [4] projects. In the Member States with the highest share of Cohesion Policy funds’ allocation, a large majority of road projects are EU co-financed [5].

5. Planned Cohesion Policy funds’ spending on roads totals around 65 billion euro for the two programming periods between 2000 and 2013 [6]. According to an evaluation of the 2000–06 programming period, these funds co-financed almost 8000 projects, which constructed around 75000 km [7] of roads [8].

TABLE 1

PLANNED ROADS EXPENDITURE THROUGH COHESION POLICY FUNDS (MILLION EURO)

Source: European Commission.

2000–06 |

Cohesion Fund | 4629 |

ERDF | 19288 |

of which: | |

Motorways | 4416 |

National roads | 2405 |

Regional/local roads | 2908 |

Roads (not specified) | 9558 |

TOTAL | 23917 |

2007–13 |

Cohesion Policy Funds | |

| |

| |

Motorways | 4577 |

Motorways (for TEN-T network) | 18565 |

National roads | 7041 |

Regional/local roads | 10581 |

TOTAL | 40764 |

AUDIT SCOPE

6. The Court carried out an audit addressing the main question: "Have the EU Cohesion Policy funds’ road infrastructure projects achieved their objectives at a reasonable cost?"The scope was further developed in two sub questions: (i) "Did the road projects audited achieve their objectives?" and (ii) "Were the projects managed to ensure reasonable costs?".

7. The audit involved co-financed road projects in Germany (DE), Greece (EL), Poland (PL) and Spain (ES) [9]. These four Member States have the highest allocations of Cohesion Policy funding for roads in the period 2000–13 (see Figure 1), representing approximately 62 % of all EU road co-financing.

FIGURE 1

BREAKDOWN BY COUNTRY OF PLANNED COHESION POLICY FUNDS’ CO-FINANCING FOR ROADS BETWEEN 2000 AND 2013 IN MILLION EURO [1010]

+++++ TIFF +++++

Source: Court calculations based on European Commission data.

8. A sample of three projects from the two operational programmes (OPs) with the highest spending on roads from each four selected Member States were included in the audit. The sample of 24 projects was drawn at random from a preselected population [10]. The audit covered motorways (10 projects), express roads (10 projects) and ordinary two-lane trunk roads (4 projects). The full list of the projects audited with a short description of their content and details of their cost is to be found in Annex I.

9. The audit procedures included documentary review, interviews with the auditees, analysis of project management systems, experts’ assessments and on-the-spot verification of project outputs and outcomes.

OBSERVATIONS

THE ROAD PROJECTS AUDITED PARTLY ACHIEVED THE INTENDED RESULTS

10. In order to analyse the performance of the road projects, the following criteria were considered by the Court:

(i) establishment of transport infrastructure development objectives;

(ii) measurable impact on economic development, based on appropriate indicators;

(iii) economic viability of the investments made;

(iv) achievement of results regarding quantitative and/or qualitative improvements.

THE ROAD PROJECTS HAD SET OBJECTIVES …

11. The defined objectives of road projects usually relate to national or regional economic development goals and interregional transport needs. At operational level the most common objectives are shorter transit times, increased road safety, reduced congestion and less pollution. Ways to achieve these goals include adding capacity and quality to the existing road network through building a new alternative connection (eight audited projects involved building a bypass), or in-depth renovation of the roads (seven audited projects).

12. For 21 of the 24 projects audited the objectives had been set in terms of expected outcomes. For three projects, no needs analyses were conducted so they had been implemented based on common understanding among the decision-makers about the poor state of existing connections.

… BUT THEIR IMPACT ON ECONOMIC DEVELOPMENT COULD NOT BE ASSESSED …

13. In all ERDF Major or Cohesion Fund project applications for assistance [11] (11 projects audited) and in a further eight project applications the expected impact on economic development (such as the number of new jobs in the region, number of jobs due to the construction activity and maintenance, potential tourism activity) was described in detail. However, due to the lack of appropriate indicators (such as actual employment created, share of new transit traffic, number of new enterprises in the region), the reporting on the achievements of the projects focused on the amount of funding used and physical output delivered, with almost no information about the effects of the projects on the local economy or actual gains for road users. As a result, it is not possible to assess whether the funded projects actually generated the expected economic impact.

… AND MOST OF THEM DELIVERED LESS THAN THE PLANNED RETURN ON INVESTMENT …

14. For 19 of the 24 [12] projects, economic viability analyses have been undertaken to assess whether the projects would deliver sufficient returns on investments. The indicators used included economic net present value (ENPV), economic rate of return (ERR) and the benefit–cost (B/C) ratio, the last of which was available for all 19 projects. This ratio depends on the estimated cost of the projects and on the estimated future economic benefits. For the other five projects, that were not major projects, the managing authorities did not consider the cost-benefit analyses necessary (for example Jaraicejo project (ES)) or there was no traffic forecasting available (for example Keratea–Lavrio project (EL)).

15. There is no EU requirement for a minimum B/C ratio for EU co-financed road investments. Nevertheless the Commission has provided all managing authorities with "A Guide to Cost-Benefit Analyses for Major Projects".

16. Out of 19 projects for which information was available, 14 recorded less than expected usage [13] (see Figure 3). Also, on average the cost of the ERDF Major and Cohesion Fund projects audited turned out to be 26 % higher than the costs taken into account in feasibility studies [14], making the investments much less economically viable than planned.

17. Two ERDF Major Projects (A17 and A20) audited in Germany were classified as "urgent" in the federal road investment plan. Applying the actual figures, known after the projects were completed, to calculate the B/C of the projects, they do not classify as "urgently needed", as the building costs increased by 50 % (A17) or the traffic volume was 50 % less than initially forecasted (A20).

18. Similarly, in Greece, according to the application for assistance, the B/C of the E75/PATHE corridor including the audited section was 3,1. The total cost of the project in the application for assistance was forecast to be 278 million euro. The actual cost of the project was 378 million euro. If all other factors had remained unchanged, the additional cost would have reduced the B/C to 2,3.

… WHILE CLEARLY IMPROVING ROAD SAFETY AND HELPING TO SAVE TRAVELLING TIME

19. The improvements in road safety were set as an objective in all project applications. For two Spanish and one Greek motorway project the safety gains were especially noticeable. For the two projects audited on the motorway A66 (ES) the aim was to have the fatality rate not higher than 0,03 (number of dead persons per km per year) and the rate of injured persons not higher than 1,14. In 2010, the actual statistics showed the fatality rate to be zero and the injury rate only 0,1. The E75/PATHE (EL) project reduced the number of deaths from an average of six during the 2001 to 2010 period to zero. Further good examples for safety improvements are projects in the City of Częstochowa (PL) and Loja (ES) (see Photo 1) that solved persistent accident problems at railway level crossings.

20. The Court also tested the reduction in travelling times achieved by those 11 projects which had such an objective and/or for which comparative testing was possible. All tests proved that when compared to alternative connections, the road sections built delivered time gains of up to 60 %. All projects also delivered the time gains foreseen in the applications for assistance.

+++++ TIFF +++++

© European Court of Auditors.

21. Although all audited projects achieved travelling time savings, seven projects [15] were not able to realise their full time saving potential as other parts of these transit routes did not provide the same speed, capacity and safety standards as the sections audited. Two of those (the road A1 (PL) and Thessaloniki–Kilkis (EL)) will be connected to roads of a similar type in the near future (the necessary projects were already ongoing at the time of the audit), but for the others issues remain (see the Box for details).

BOX

THREE CASES OF POTENTIAL FURTHER IMPROVEMENTS FOR ADDITIONAL TRAVELLING TIME GAINS

The road S7 (PL) will be completely renovated in the coming years but at the time of the audit was linked to a section with road-level crossings that slows down the traffic flow.

The road S8 (PL) was a four-lane express road that at one end led into regular two-lane trunk roads, making the time savings on the express road section insubstantial. The full length of the road S8 from Bialystok to Wroclaw is being upgraded.

The road B104 (DE) is the second section of a northern bypass of Schwerin. The further parts of this bypass are not yet built.

ENHANCED TRANSPORT CAPACITY COULD HAVE BEEN ACHIEVED AT LOWER COST

22. In order to evaluate the costs of the projects, comparative analyses were made by the Court based on input unit prices for the most comparable construction materials and works. When analysing the reasons for these costs, special attention was given to project management practices (planning, procurement and implementation) that are critical for delivering the projects with the most advantageous costs. As criteria the Court expected that the projects were properly delivered within the contracted time schedules and costs.

DIFFERENT WAYS OF COMPARING THE COST OF ROADS

23. For all audited projects three cost categories were calculated by the Court:

(a) Total costs of the project — the cost of a road per kilometre is easy to calculate but individual projects cannot be fairly compared using this measure as the geological and natural conditions and type of works can differ from one project to another. The figures for the total costs of projects can, however, be used to analyse different road types, for example the cost of building a motorway when compared to an express road.

(b) Total construction cost — total construction costs cover all materials and works needed to deliver the projects. When compared to total costs, such elements as supervision and land purchasing have been eliminated. Total construction cost indicates the price level in a country at which roads can be built.

(c) Roadway construction cost — this category only covers the costs of building the road surface and excludes planning and supervision, engineered objects and road accessories such as barriers and traffic signs. This category is the best to compare the costs of the projects with each other as all geological and nature-related cost elements have been eliminated.

24. In order to exclude the cost differences caused by the varying width of roads, all the costs have also been calculated per 1000 m2 of road surface. Figure 2 presents the total cost, total construction and roadway construction costs per 1000 m2 of road surface. This figure gives a comparable overview of the cost levels of the projects included in the audit. Detailed information concerning the different type of costs on the project level can be found in Annex II.

25. Figure 2 shows that:

(a) the projects audited in Germany had the lowest cost per 1000 m2 in all three categories. There is no evidence that this can be explained by labour costs;

(b) for the projects audited in Spain, there is considerable difference between the total construction and roadway construction costs. This indicates a heavy use of engineering objects such as bridges or tunnels;

(c) the average roadway construction costs are the highest for the projects audited in Poland closely followed by projects in Spain.

FIGURE 2

AVERAGE TOTAL COST, TOTAL CONSTRUCTION COST AND ROADWAY CONSTRUCTION COST FOR 1000 m2 OF THE ROAD PROJECTS AUDITED PER MEMBER STATE IN EURO [16]

+++++ TIFF +++++

Source: Calculation by the Court of Auditors.

26. The cost of a road project can be influenced by (i) road alignment and need for land acquisition, (ii) technical specifications based on traffic forecasts and regulatory construction standards, (iii) procurement systems and the competitiveness of the market and (iv) project management during actual construction. In the following sections these factors are analysed on the basis of the audit results.

THE ROAD PROJECTS MAINLY FOLLOWED THE MOST ECONOMICAL ROAD ALIGNMENT

27. For all 24 projects audited the Court verified whether the most economical road alignment had been chosen. For all projects alignment alternatives were checked and where possible the conditions were verified during the on-site visits.

28. For 19 projects the reasons for the selected road alignments were sufficient to confirm their economic advantage. Five [17] projects were not built on the most economical alignment due to environmental constraints or opposition of local inhabitants/authorities. For example project E75/PATHE (EL) was built through difficult mountainous terrain in order not to block the access to the seaside in the towns of Kamena Vourla and Agios Konstantinos.

29. Eight alignments followed previously existing roads. The remaining 16 greenfield projects were on average twice as expensive as the roads built on existing alignments as they involved the additional costs of land acquisitions and necessary earth works. The average total cost of the fully greenfield projects audited was 9,4 million euro per km while for roads using an existing alignment the average total cost was 4,7 million euro. Thus it is less costly to build on existing road alignments.

30. In one case, choosing the existing road alignment increased the cost. The S7 express road project (PL) turned out to be more expensive than the similar road S8 (PL), which was built as a greenfield project. In order to upgrade the road S7 to express road standards while maintaining roadside inhabitants’ access, collecting roads [18] were needed along the entire length of the new express road.

TRAFFIC FORECASTS WERE IN THE MAJORITY OF CASES NOT IN LINE WITH ACTUAL ROAD USE

31. The design of the roads and the definition of their technical specifications are based on estimates of future traffic. The Court analysed the accuracy of traffic estimates for 19 projects for which both the planned and current traffic figures were available. For five projects the difference between actual traffic volumes and estimates was less than 20 %. In 11 cases it was between 21 % and 50 % and in three cases the difference was 51 % or more (four were underestimations and 10 overestimations) (see Figure 3, which compares the traffic forecasts with the actual traffic counting figures).

FIGURE 3

MEASURED AADT COMPARED TO THE FORECAST

+++++ TIFF +++++

Note: Value 1 means that actual and planned traffic are the same.

Source: Calculation by the Court of Auditors.

COSTS PER USER VARY SIGNIFICANTLY

32. The Court calculated the total cost of the projects by the measured AADT. This calculation that is presented in Figure 4 is based on the information from traffic counting and presents the cost of one kilometre of road divided by the number of average daily users.

33. The average cost in the projects audited of 1 km of road per AADT was more than four times higher in Spain than in Germany with Poland and Greece being in the middle between the two extremes.

FIGURE 4

TOTAL COST OF 1 KM OF AUDITED ROAD PROJECTS PER COUNTED TRAFFIC (AADT) IN EURO

+++++ TIFF +++++

Source: Calculation by the Court of Auditors.

34. La Herradura project (ES) had the highest cost per user. The project was technically complex and needed a lot of earthworks in order to complete the tunnels and the actual traffic was less than 50 % of the planned AADT. Other projects that have a significantly higher cost per daily user are either technically complex (including a high number of engineered objects such as bridges and tunnels [19]) or do not have much traffic [20].

EXPRESS ROADS WERE CLEARLY LESS COSTLY TO BUILD THAN MOTORWAYS

35. The key technical decision that determines the future cost of the road results from the type of road to be built. In the Court’s sample, the total costs of types of roads per kilometre is indicated in Figure 5. The full size motorways can be more than twice as costly to build as the two other road types.

FIGURE 5

AVERAGE TOTAL COST PER KILOMETRE FOR VARIOUS TYPES OF AUDITED ROADS [21] IN EURO

+++++ TIFF +++++

Source: Calculation by the Court of Auditors.

36. The Court calculated the average total costs, total construction costs and roadway construction costs for the two most comparable road types – motorways and express roads. Figure 6 confirms that in all cost categories, express roads required less investment than motorways. The reasons for the difference in costs can be attributed to two main categories: (i) express roads need less width for the road corridors, so the cost for land purchase and construction is lower and (ii) existing alignments of ordinary trunk roads can be upgraded to express road standard because they are designed for lower speeds, while motorways usually require totally new road alignments [22].

FIGURE 6

AVERAGE TOTAL COST, CONSTRUCTION COST AND ROADWAY CONSTRUCTION COST PER 1000 m2 OF AUDITED MOTORWAYS AND EXPRESS ROADS IN EURO

+++++ TIFF +++++

Source: Calculation by the Court of Auditors.

37. As the express roads audited provide two carriageways with at least four traffic lanes, they are able to deal with a comparable number of vehicles providing similar safety standards as motorways. The audited express road projects had an average recorded AADT of 16950 while eight comparable four-lane motorway projects recorded an average AADT of 13398. Only in two cases did four-lane motorways record higher AADT than was the average AADT for express roads. This indicates that traffic on at least six motorways could be served also following express roads standards.

+++++ TIFF +++++

© European Court of Auditors.

THE BEST POSSIBLE PRICE WAS NOT OBTAINED FOR ALL AUDITED PROJECTS

38. The Member States in the Court’s sample had different procurement systems. In Greece and Spain the prices of construction input units were set by the administration and bidders offered discounts on the "official" prices [23] and all future unit cost increases were covered by the "cost plus" contracts. In Poland and Germany the offers were itemised and bidders were free to offer their prices for each input unit. In Germany, the road corridors were divided into shorter procurement units, of 1,5 to 5 km. In other audited countries, the projects were procured in one section [24].

39. In Greece and Spain the access to procurement processes was subject to prior registration in qualification registries [25] kept by the ministries of economy [26] while in Poland and Germany all companies can immediately take part in published procurement procedures.

40. For eight projects, the authorities awarded the contracts to bidders which had not offered the lowest price among the technically compliant bids. Three projects in Spain and four in Greece were procured using a mathematical formula that excluded companies making low offers (below 10 % of the average price offered). The practice of using mathematical formulas in Greece and Spain has been phased out. The Commission applied a 10 % penalty on all projects using this type of procurement in Greece. Some of the companies excluded from the procurement later formed joint ventures with the winning bidders [27], provided the construction services and received payments based on the higher unit prices offered in the winning bid.

SIGNIFICANT COST DIFFERENCES EXIST FOR ROAD ACCESSORIES AND BRIDGE DECKS

41. Roads in Europe are built according to broad norms concerning therequirements for each road type but the specific regulations concerning the road accessories and markings are still fragmented. For example, the norms for road accessories such as road signs and some safety barriers are not standardised and, as a result, there are entrance barriers to the national markets. As a sign of progress a new harmonised European performance standard (EN01317-5) has been in force since 1.1.2011. The standard allows road authorities and manufacturers to express the product performance in a common European way. The designs of the barriers remain different from country to country so economies of scale are difficult to obtain. Figure 7 presents the differences in price in different markets for a similar road safety barrier, showing that the difference between the lowest and the highest average price is 75 %.

42. As a specific cost item which can easily be compared, the Court also examined the cost of bridge decks in all projects audited. The results (see Figure 8) show that in Spain bridges cost significantly less than in the other countries audited.

FIGURE 7

AVERAGE COST OF 1 m OF SAFETY BARRIER OF AUDITED PROJECTS IN EURO

+++++ TIFF +++++

Source: Calculation by the Court of Auditors.

FIGURE 8

AVERAGE COST OF A BRIDGE DECK PER m2 OF AUDITED PROJECTS IN EURO

+++++ TIFF +++++

Source: Calculation by the Court of Auditors.

43. In order to explain these cost differences, the Court compared the price of the core material for bridge construction, i.e. the price of concrete used. The comparative analysis (see Figure 9) shows that concrete is cheapest in the projects audited in Spain and rather expensive in Germany and Poland. The comparison of the cost of concrete indicates that the average price offered can be significantly lower in one Member State than in another [28].

44. Six out of 10 motorway projects audited [29] included noise reduction elements. The required level of protection differs from one country to another. For example, the highest acceptable noise levels in Poland are lower than the corresponding values in neighbouring countries by up to 7 dB. The stricter rules make it necessary for a considerable part of nearly every new road project to be shielded by noise barriers (see Photo 3), which has an influence on the costs.

FIGURE 9

AVERAGE COST OF m3 OF CONCRETE FOR THE BRIDGE DECK OF AUDITED PROJECTS IN EURO

+++++ TIFF +++++

Source: Calculation by the Court of Auditors.

+++++ TIFF +++++

© European Court of Auditors.

COST OVERRUNS OF MORE THAN 20 % FOR 11 AUDITED PROJECTS

45. Different project management systems used in Member States have an impact on additional costs and timely delivery of the works. While in Germany and Poland the contractors have to bear the full responsibility for possible changes in the prices of inputs, in Greece and Spain indexing systems were used, whereby the input prices were periodically reviewed and revalued by the ministries of public works.

46. In Greece, Germany and Spain remeasurements of works and contract updates were common leading to tens of modifications and additional payments. In contrast, the Polish project management system is designed to reduce any risk of additional claims and very few additional payments are made after the procurement.

47. Out of the 24 projects audited, only seven were delivered at the original contract price (one in Germany, five in Poland and one in Spain). Moreover, the costs of 11 projects increased by more than 20 % during construction, mainly because of archaeological finds, planning mistakes, erroneous estimates in the documents concerning quantities and price indexing. A comparison of contract prices and the actual payments made is presented in Figure 10.

48. The average additional payment for all projects after contracts had been signed was 23 %. The extremes were an increase of 36 % for the Greek projects and 1 % for the Polish ones.

FIGURE 10

ADDITIONAL COSTS OF THE AUDITED PROJECTS IN RELATION TO THE CONTRACT PRICE

+++++ TIFF +++++

Note: The initial contract price equals 1.

Source: Calculation by the Court of Auditors.

49. Delays in road construction are commonplace. In countries where indexing systems were applied (Spain and Greece), the longer the extension of the construction period, the more advantageous it is for the contractors. Also long processes for land expropriations and problems with availability of working sites increased up to 1 year the duration of the project execution.

50. The average delay among the projects audited was 9 months or 41 % from the planned to the actual date of opening to traffic. In Poland, the average delay was 3 months or 16,5 % and in Germany 7 months or 59,5 %; for the projects in Spain, the average delay was 11 months or 31 % and in Greece roads were opened to traffic on average 16 months or 57 % later than expected (see Figure 11).

FIGURE 11

DELAYS IN MONTHS OF THE AUDITED PROJECTS IN OPENING THE ROADS TO TRAFFIC

+++++ TIFF +++++

Source: Calculation by the Court of Auditors.

CONCLUSIONS AND RECOMMENDATIONS

DID THE ROAD PROJECTS AUDITED ACHIEVE THEIR OBJECTIVES?

51. The EU spent some 65 billion euro on road investments between 2000 and 2013. All the projects audited added capacity and quality to the road networks and resulted in travelling time savings and improved road safety. However, the economic development objectives were not measurable and there is no information available on the impact of the projects on the local or national economy.

52. The actual economic viability calculated using a Benefit/Cost ratio based on the actual cost and use was significantly lower for half of the projects when compared to the assumptions made in the planning phase with significantly lower costs and/or higher traffic.

WERE THE PROJECTS MANAGED TO ENSURE REASONABLE COSTS?

53. Insufficient attention was paid to ensuring cost-effectiveness of the projects:

(a) Most of the audited projects were affected by inaccurate traffic forecasts. Only four out of 19 projects for which sufficient information was available carried volumes of traffic which were in line with forecasts.

(b) The type of road chosen was not best suited to the traffic it carried: motorways, costing much more than express roads (the average total cost per km is around 11 million euro for motorways and 6,2 million euro, i.e. 43 % cheaper, for express roads), were the preferred option of construction, even for sections where express roads could have solved the traffic needs.

(c) During the project implementation the projects audited became more expensive and were affected by delays as the average increase to the original contract price was 23 % and time overruns averaged 9 months or 41 % of the planned time.

54. Furthermore the Court notes that projects in two Member States were procured using a mathematical formula that excluded bidders making offers below 10 % of the average offered price. The Commission applied a 10 % penalty on payments to the abovementioned projects where such a formula was used. The Court notes that this kind of procurement practice has been phased out.

55. Among the projects audited, the cheapest roadway construction costs were found in the projects audited in Germany, followed by Greece, Spain and Poland. There is no evidence that this can be explained by labour costs. The lowest average cost per actual use was recorded in Germany, followed by Greek, Polish and Spanish projects. The cost per user is on a scale of 1 to 28 times more between the road with the best and that with the worst ratio (or a scale of 1 to 12 times if two extreme cases are taken out).

RECOMMENDATION 1

The Commission should analyse the differences in roadway construction costs between the Member States in order to identify the causes of significant price differences thereof and assure that the best practices are applied in future.

RECOMMENDATION 2

The future co-financing of road projects should be made conditional upon:

(a) the existence of clear objectives that are accompanied by indicators such as for travel time savings, gains in road safety, capacity improvements and concrete effects on the economy;

(b) the use of reasonable and cost-effective technical solutions that are in line with the best practices identified, including for the accessories required for road construction.

(c) steps taken by the Member States to ensure competition in construction markets, focusing procurement systems on delivering the best value for money while avoiding entrance barriers such as for example broader international publicity for the tender processes, recognition of quality categories of companies provided in other Member States and easing SMEs’ access to procurement procedures.

RECOMMENDATION 3

The Commission should promote the exchange of best practices between national road administrations concerning reliable traffic forecasts.

RECOMMENDATION 4

The Commission should consider: the establishment of European Union-wide unit cost information for engineers preparing estimates for new projects, in order to help the beneficiaries lower the procurement prices.

This report was adopted by Chamber II, headed by Mr Harald NOACK, Member of the Court of Auditors, in Luxembourg at its meeting of 29 May 2013.

For the Court of Auditors

+++++ TIFF +++++

Vítor Manuel da Silva Caldeira

President

[1] Council Regulation (EC) No 1083/2006 of 11 July 2006 laying down general provisions on the European Regional Development Fund, the European Social Fund and the Cohesion Fund and repealing Regulation (EC) No 1260/1999 (OJ L 210, 31.7.2006, p. 25).

[2] Article 17(1) of the Treaty on European Union (OJ C 326, 26.10.2012, p. 13) and Article 317 of the Treaty on the Functioning of the European Union (OJ C 326, 26.10.2012, p. 47).

[3] OECD International Transport Forum statistics, "Road infrastructure gross investment spending", http://www.internationaltransportforum.org/statistics/investment/data.html

[4] TEN-T — Trans European Networks facility, managed by TEN-T Executive Agency.

[5] Other non-domestic sources for financing the road projects are EIB and World Bank loans and support from TEN-T financing instruments.

[6] The average EU co-financing for the total costs of the projects audited was 44 %.

[7] Out of 75000 km, 12744 km were newly built and 62256 km renovated roads.

[8] Source: "Ex post evaluation of Cohesion Policy programmes 2000–06 co-financed by the European Fund for Regional Development", January 2010, Steer Davies Gleave, London (http://ec.europa.eu/regional_policy/sources/docgener/evaluation/pdf/expost2006/wp5a_final_report_summary.pdf).

[9] In this field, the European Court of Auditors has already reported on EU co-financing of transport infrastructure (see Special Report No 1/93 on the financing of transport infrastructure (OJ C 69, 11.3.1993), Special Report No 8/2010 Improving transport performance on Trans-European rail axes: Have EU rail infrastructure investments been effective? and Special Report No 4/2012 Using Structural and Cohesion Funds to co-finance transport infrastructures in seaports: an effective investment? (htpp://eca.europa.eu)).

[1010] The "others" in the table are all other EU Member States except Luxembourg and Denmark that had no recorded roads-related spending.

[10] The selection criteria were: the total project cost had to be more than 5 million euro, the projects had to be finalised and open to traffic and the projects should have at least one bridge. Half of the projects from each MS had to be ERDF Major Projects or Cohesion Fund projects.

[11] These applications are subject to an ad hoc decision for co-financing by the European Commission.

[12] The exceptions were: road DK43 in the city of Częstochowa (PL), Keratea–Lavrio and Thessaloniki–Kilkis (EL), Jaraicejo (ES), road L132 (DE).

[13] Comparative analyses of available traffic information indicate reduction of traffic demand of up to 20 % in some of the audited countries in 2011 compared to 2010 but no such effect in earlier years. No significant changes in traffic demand were noted in Germany and Poland.

[14] The figure relates to the expected total costs of projects audited as presented in the applications for assistance.

[15] A1, S8 and S7 (PL); Thessaloniki–Kilkis and Keratea–Lavrio (EL); S177 Pirna and B104 (DE).

[16] For comparison of Polish projects long-term average exchange rate of 4 zloty = 1 euro was used for all calculations.

[17] S177 Radeberg (DE), A1 (PL), E75/PATHE and A2 Nymphopetra–Asprovalta (EL) A7 La Herradura (ES).

[18] A collecting road is a normal two-lane trunk road that collects the traffic to/from the properties that are by roads with entry barriers such as motorways and expressways.

[19] Loja (ES), E75 Pathe (EL), DK50 (PL)).

[20] Jaraicejo (ES), S17 Radeberg (DE), A1 (PL).

[21] Calculations based on total cost of the 24 projects audited.

[22] The two green field Express Roads audited S8 (PL) and Thessaloniki – Kilkis (EL) had an average total cost of 4,2 million euro per kilometre. That is below the average cost per km for all express roads.

[23] In Spain the discounts were offered on the total price while in Greece the discounts were offered on the groups of works and materials.

[24] With the exception of E75/PATHE (EL) project, that was procured in two sections. In Poland and Greece major project management companies were awarded the large tenders due to high capital requirements set at the procurement phase.

[25] Qualification registries keep the list of enterprises that are certified to undertake projects with a certain size or technical complexity. For each tendering process, a minimum qualification requirement is set.

[26] Registered companies can create joint ventures with other companies to carry out projects.

[27] This was the case in the Keratea–Lavrio and Kymis Avenue projects (both EL).

[28] The ingredients of concrete (cement, sand and steel) are internationally tradable, concrete itself not.

[29] A1 (PL); A17 and A20 (DE); E75/PATHE and A2 (EL); A7 (ES).

--------------------------------------------------

ANNEX I

OVERVIEW OF AUDITED PROJECTS

Country and region | Name of the project | Total cost (euro) | EU co-financing (euro) | Traffic volume as AADT [1] | Short description of the project |

GERMANY Saxony | Motorway A 17 from Dresden to the Czech border (40,8 km) | 655510000 | 277600000 (42 %) | 51100 | Major road artery that connects Germany to Czech Republic. The thoroughly audited section was 3,88 km Construction Segment 1.2a. |

Express road S 177 Western Bypass at Pirna (3,7 km) | 20700000 | 15600000 (75 %) | 12000 | Bypass of the town of Pirna. Part of a future link between motorways A 17 and A 4. The thoroughly audited section was 1,5 km Construction Segment 5.2. |

Express road S 177 Bypass at Radeberg (6,2 km) | 38000000 | 28500000 (75 %) | 12000 | Bypass of Grosserkmannsdorf–Radeberg. Part of a future link between motorways A 17 and A 4. The thoroughly audited section was 3,2 km Southern Construction Segment. |

GERMANY Mecklenburg and Western Pomerania | Motorway A 20 Grimmen-East to Strasburg (91,2 km) | 367600000 | 25100000 (6,8 %) | 12662 | Part of the Baltic Motorway, from Lübeck to Szczecin in Poland. The thoroughly audited section was 12,4 km from Jarmen junction to Anklam junction. |

Express road B 104 Northern Schwerin Bypass (7,5 km) | 32000000 | 19000000 (65 %) | No information available | Important element in the renovation of the European road E 22. The thoroughly audited section was 5,5 km Schwerin urban bypass, Construction Segment 5. |

State road L 132 Niendorf — Rostock (1,5 km) | 5178000 | 3874000 (75 %) | 10000 | Connection between the A 20 motorway and the southern part of the town of Rostock. Construction Segment 3. |

GREECE Attica | Motorway E75/PATHE section Agios Konstantinos–Kamena Vourla (20 km) | 378806726 | 146062798 (39 %) | 15283 | Part of Greece’s most important north–south transit route. The thoroughly audited section was the 9 km motorway section Kamena Vourla. The works were delayed due to difficult mountainous building conditions. |

Express road Kymis Avenue from Kifissias junction to Olympic village (6,6 km) | 75822106 | 22877923 (30 %) | No information available | Connection between the 2004 Athens Olympic village and the motorway E75/PATHE. The road continues to the municipality of Tharkomakedones. In future the road should be directly connected to Athens’ ring road. |

Express road S89 Stavros–Lavrio, section Keratea–Lavrio (12,9 km) | 18395555 | 7994920 (43 %) | No information available | Upgrade of two lane road into four-lane express way that connects Athens ring road to the port of Lavrio. The project saw delays due to the bankruptcy of the main contractor. |

GREECE Central Macedonia | Motorway A2 Egnatia Odos, section Kouloura–Kleidi (25,14 km) | 59309257 | 24904493 (53 %) | 21600 | Upgrade of two-lane road into four- lane motorway. Part of the east– west transit route Egnatia Odos. |

Motorway A2 Egnatia Odos, section Asprovalta– Nymphopetra (31 km) | 184095172 | 77095200 (42 %) | 15500 | New motorway section on Egnatia Odos. Missing section, as both ends were already constructed. |

Express road Thessalonica– Doirani, section Nea Santa to Mavroneri (11,3 km) | 31327495 | 25200000 (80 %) | 8000 | Middle section of express road linking Thessalonica area to Kilkis. The potential continuation of the connection to FYROM remains uncertain. |

POLAND Masovia | Arynow bypass on a road DK50 (1,942 km) | 12481450 | 7921600 (63 %) | 11800 | The project was a continuation of the renovation of the road DK50, as the first part was co-financed by the World Bank. The project created crucial connection between the DK 50 and National Road No 2. |

Express Road S8, section Radzymin– Wyszkow, including Wyszkow bypass (28,336 km) | 216444000 | 140197790 (83 %) | 19701 [2] | A new greenfield express road that leads from Warsaw towards north-eastern Poland. The road continues as an ordinary two-lane trunk road. Thoroughly adited section12,8 km bypass of Wyszkow. |

Express Road S7 section Bialobrzegi– Jedlinsk (13,652 ikm) | 118022250 | 85257005 (72 %) | 24837 [3] | Upgrade of national road to express road standards. Continuation of previous projects on the same road. Includes collecting roads for local traffic. |

POLAND Silesia | Motorway A1, section Sosnica–Belk (15,411 km) | 307887511 | 193803465 (63 %) | 9374 [2] | First section of a new motorway that connects Central Poland to Czech Republic. Out of the audited sample of 24 projects the A1 was the only one awarded to a non-domestic bidder. |

National Road nr 43 in the city of Częstochowa (1,4 km) | 14028195 | 9401244 (67 %) | 31525 | New bridge over the railway tracks and large upgrade of the national oad in the territory of Częstochowa. |

National Road nr 79 in the city of Jaworzno (2,92 km) | 16880016 | 761766 (4,5 %) | 12900 | New alignment of the road DK in the city of Jaworzno, bypassing the transit traffic from the city centre. Was procured as a lump sum contract though no unit cost information is available. |

SPAIN Andalucia | Mediterranean Motorway A7 section Tramo Castell de Ferro–Polopos (3,6 km) | 28174391 | 17891652 (64 %) | 8472 | Section of the new Motorway A7 from Algeciras to Barcelona. Used the alignment and part of the roadway of an existing national road. |

Mediterranean Motorway A7 section La Herradura– Almunecar (9,12 km) | 280150747 | 108939032 (39 %) | 7454 | Section of the new Motorway A7 from Algeciras to Barcelona. The road is on one end not connected to a motorway, although the construction was ongoing at the time of the audit. |

New bridge over the river Genil and a railroad overpass in the town of Loja (1,68 km) | 6850627 | 3925293 (57 %) | 8889 | A new (second) bridge was built into the town of Loja over the river Genil and a same level railway pass was upgraded into an overpass. |

SPAIN Extremadura | Motorway A66 section Caceres North– Aldea del Cano (29,343 km) | 96689964 | 58896574 (67 %) | 10690 | Section of the new motorway from northern Spain to Seville, in the south. |

Motorway A66 section Enlace de Hinojal–Caceres North (21,422 km) | 88065756 | 50841178 (58 %) | 12594 | Section of the new motorway from northern Spain to Seville, in the south. |

Ronovation of the Local Road EX – 385 in Jaraceijo (14,617 km) | 6104166 | 4135445 (68 %) | 154 | Upgrade and partly new built local road that enhanced the accessibility to the Manfragüe Wildlife Protection National Park. |

|

TOTAL | | 3058523384 | 1355781378 (44 %) | | |

[1] As provided by the road authorities at the time of the audit.

[2] Average of the results of three measuring points.

[3] Average of the results of two measuring points.

--------------------------------------------------

ANNEX II

COMPARISON OF TOTAL COST, CONSTRUCTION COST AND ROADWAY CONSTRUCTION COST OF THE PROJECTS AUDITED PER 1000 m2 (IN EURO)

+++++ TIFF +++++

--------------------------------------------------