EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

chapter 1 - ECONOMIC COHESION

TABLE OF CONTENTS

Key messages    

1.1.     INTRODUCTION    

1.2.     ECONOMIC TRENDS AMONG EU REGIONS AND MEMBER STATES    

1.2.1 Convergence back on track    

1.2.2 Less developed regions maintain a strong manufacturing sector, but their agriculture needs to modernise    

1.3.     PRODUCTIVITY IN LESS DEVELOPED MEMBER STATES IS CATCHING-UP    

1.3.1 Productivity and employment contribute to the economic recovery in the EU    

1.3.2 Capital metropolitan regions more prone to boom and bust than in other regions    

1.3.3 GDP growth in rural and intermediate regions proved to be more resilient during the crisis years    

1.4.     THE ECONOMIC DEVELOPMENT CLUBS OF EUROPEAN REGIONS AND THE MIDDLE-INCOME TRAP    

1.5.     COMPETITIVENESS OF EU REGIONS    36

1.5.1 Firms in EU Capital metro regions tend to be larger and to grow at a faster pace.    

1.5.2 Innovation remains spatially concentrated    42

1.5.3 The number of people with tertiary education keeps increasing, but large disparities persist    53

1.5.4 Improving market access does not always generate growth    58

1.5.5 Digital networks are spreading, but closing the gap between urban and rural areas represents a major challenge    62

1.6.     CAPITAL AND METRO REGIONS ARE THE MAIN DRIVERS OF REGIONAL COMPETITIVENESS IN EUROPE    67

1.7.     REFERENCES    74

Figure 11: Coefficient of variation (index, 2000 = 100), GDP per head in PPS, employment rate, unemployment rate, EU-28 NUTS 2 regions, 2000-2016 Figure 1 1: Coefficient of variation (index, 2000 = 100), GDP per head in PPS, employment rate, unemployment rate, EU-28 NUTS 2 regions, 2000-2016

Figure 12: GDP per head growth rates of regions in less developed or moderately developed Member States, 2001-2008 Figure 1 2: GDP per head growth rates of regions in less developed or moderately developed Member States, 2001-2008

Figure 13: GDP per head growth rates of regions in less developed or moderately developed Member States, 2009-2015 Figure 1 3: GDP per head growth rates of regions in less developed or moderately developed Member States, 2009-2015

Figure 14: Growth of GDP per head in real terms, 2001-2018 Figure 1 4: Growth of GDP per head in real terms, 2001-2018

Figure 15 Employment shares in Industry (excluding construction), 1995-2014 Figure 1 5 Employment shares in Industry (excluding construction), 1995-2014

Figure 16 GVA shares in Industry (excluding construction, 1995-2014 Figure 1 6 GVA shares in Industry (excluding construction, 1995-2014

Figure 17 Employment shares in Agriculture, 1995-2014 Figure 1 7 Employment shares in Agriculture, 1995-2014

Figure 18 GVA shares in Agriculture, 1995-2014 Figure 1 8 GVA shares in Agriculture, 1995-2014

Figure 19: Evolution of total employment (number employed) in metro regions, 2000-2014 Figure 1 9: Evolution of total employment (number employed) in metro regions, 2000-2014

Figure 110 Employment shares in Industry (excluding construction), by club, 1995-2014 Figure 1 10 Employment shares in Industry (excluding construction), by club, 1995-2014

Figure 111 GVA shares in Industry (excluding construction), by club, 1995-2014 Figure 1 11 GVA shares in Industry (excluding construction), by club, 1995-2014

Figure 112: Firms density by metro region, 2014 Figure 1 12: Firms density by metro region, 2014

Figure 113: Employees per firm by metro region, 2014 Figure 1 13: Employees per firm by metro region, 2014

Figure 114: Enterprise birth rate per metro region, 2013 Figure 1 14: Enterprise birth rate per metro region, 2013

Figure 115: Death rate of enterprises, 2012 Figure 1 15: Death rate of enterprises, 2012

Figure 116: Number of high growth firms per 1000 inhabitants, in 2014 Figure 1 16: Number of high growth firms per 1000 inhabitants, in 2014

Figure 117: Patents by metro regions, average 2009-11 Figure 1 17: Patents by metro regions, average 2009-11 43

Figure 118: Total expenditure on R&D, 2014 Figure 1 18: Total expenditure on R&D, 2014 46

Figure 1 19: population (aged 25-64) with tertiary education, 2016 54

Figure 120: People’s levels of digital skills, by level of economic development, 2015 Figure 1 20: People’s levels of digital skills, by level of economic development, 2015 57

Figure 121: Households with access to Next generation Access (NGA) broadband, by type of area, 2012 and 2016. Figure 1 21: Households with access to Next generation Access (NGA) broadband, by type of area, 2012 and 2016. 64

Figure 122 - Regional competitiveness index, 2016 Figure 1 22 - Regional competitiveness index, 2016 68

Figure 123: Relationship between RCI and GDP per head (in PPS), by level of economic development Figure 1 23: Relationship between RCI and GDP per head (in PPS), by level of economic development 72

Figure 124: Relationship between RCI and the birth rate of firms (relative to population), by level of development Figure 1 24: Relationship between RCI and the birth rate of firms (relative to population), by level of development 73

Table 11 Change in employment and GVA by NACE sector per group of Member States, shares in 2016 and changes 2000-2016 Table 1 1 Change in employment and GVA by NACE sector per group of Member States, shares in 2016 and changes 2000-2016

Table 12: Decomposing average annual change in GVA per head per MS, 2001-2008 and 2009-2016 Table 1 2: Decomposing average annual change in GVA per head per MS, 2001-2008 and 2009-2016

Table 13 Change in GDP per head, productivity and employment per head by type of metropolitan region, 2001-2008 and 2009-2014 (average % per year) Table 1 3 Change in GDP per head, productivity and employment per head by type of metropolitan region, 2001-2008 and 2009-2014 (average % per year)

Table 14: Real GDP per head, productivity and employment per head growth by urban-rural typology, 2001-2008, and 2009-2014 Table 1 4: Real GDP per head, productivity and employment per head growth by urban-rural typology, 2001-2008, and 2009-2014

Table 15: European regions, by income club: some stylised facts Table 1 5: European regions, by income club: some stylised facts

Table 16: Total R&D expenditure and the distance to the EU2020 target, EU-28 regions, 2014 Table 1 6: Total R&D expenditure and the distance to the EU2020 target, EU-28 regions, 2014 45

Table 17: Population aged 30-34 with a tertiary education, by cohesion policy groups of regions, 2016 * Table 1 7: Population aged 30-34 with a tertiary education, by cohesion policy groups of regions, 2016 * 56

Map 11 GDP per head in real terms, 2015 Map 1 1 GDP per head in real terms, 2015

Map 12 Change in GDP per head index, 2000 vs 2015 Map 1 2 Change in GDP per head index, 2000 vs 2015

Map 13 Change in GDP per head index, 2000 vs 2008 Map 1 3

Map 14 Change in GDP per head index, 2008 vs 2015 Map 1 4

Map 15: Risks factors linked to globalisation and technological change Map 1 5: Risks factors linked to globalisation and technological change

Map 16: The economic development clubs of EU regions Map 1 6: The economic development clubs of EU regions

Map 17: Patent applications to the European Patents Office, 2010-2011 Map 1 7: Patent applications to the European Patents Office, 2010-2011 44

Map 18: Total expenditure on R&D, 2014 Map 1 8: Total expenditure on R&D, 2014 47

Map 19: Regional Innovation Scoreboard, 2017 Map 1 9: Regional Innovation Scoreboard, 2017 49

Map 110: Population aged 25-64 with tertiary education, 2016 Map 1 10: Population aged 25-64 with tertiary education, 2016 55

Map 111: Expected change in road accessibility due to TEN-T network composition, by NUTS3 region Map 1 11: Expected change in road accessibility due to TEN-T network composition, by NUTS3 region 60

Map 112: Average speed of direct rail connections, 2014 Map 1 12: Average speed of direct rail connections, 2014 60

Map 113: Rail accessibility during morning peak hours, Map 1 13: Rail accessibility during morning peak hours, 61

Map 114: Accessibility to passenger flights by NUTS3 region, 2013 Map 1 14: Accessibility to passenger flights by NUTS3 region, 2013 61

Map 115: Next generation access coverage by NUTS3 regions, 2016 Map 1 15: Next generation access coverage by NUTS3 regions, 2016 63

Map 116: Households with a broadband connection, by NUTS2 regions 2016 Map 1 16: Households with a broadband connection, by NUTS2 regions 2016 66

Map 117: Regional Competitiveness index, 2016 Map 1 17: Regional Competitiveness index, 2016 69

Map 118: Changes in RCI, 2016-2013; 2013-2010 and over the whole period, 2016-2010. Map 1 18: Changes in RCI, 2016-2013; 2013-2010 and over the whole period, 2016-2010. 71

Key messages

·After the double dip recession in 2008 and 2011, the EU economy is now growing again, with growth being particularly high in low-income countries.

·The crisis reversed the long-term trend towards a narrowing of regional disparities in GDP per head and employment. However, the first signs of convergence resuming are evident, though in many regions GDP per head and employment remain below their pre-crisis levels.

·GDP per head in the less developed regions is converging towards the EU average through both faster productivity growth and increased employment.

·The regions with high GDP per head have grown faster than the EU average, in part because they have benefited from the agglomeration economies from the national capital or a large city being located there. These benefits can be further extended by improving links between a large city and its rural hinterland or between smaller cities to enable specialised services to be shared and economies of scale to be realised.

·The regions with a GDP per head between 75% and 120% of the EU average seem stuck in a ‘middle-income trap’. Between 2000 and 2015, their GDP per head growth was far below the EU average. Their manufacturing sectors are smaller and weaker than those in regions with a lower or higher GDP per head. Their costs are too high and their innovation systems not strong enough to be competitive at the global level.

·Innovation in the EU remains highly concentrated. In north-western EU countries States, however, good regional connections, a skilled labour force and an attractive business environment have enabled surrounding regions to benefit from proximity to highly innovative ones. In southern and eastern EU countries, the most innovative regions are less strong and, accordingly, other regions close to them enjoy little benefit.

1.1.INTRODUCTION

Regional economic divergence has become a threat to economic progress in the EU (Iammarino et. al., 2017) at a time when globalisation poses new challenges to economic cohesion. While the evidence suggests that the EU economy as a whole has benefited and continues to benefit from globalisation, these benefits are not automatically and evenly transmitted to all European regions (European Union, 2017a).

Cohesion Policy has invested heavily in reducing economic disparities across EU regions. It has co-financed investment in innovation, education and digital and transport networks, so helping to create a single market that boosts growth, productivity and specialisation in areas of comparative advantage in all regions. As such, it strengthens the position of EU enterprises in global markets where they have to compete with both firms from low-cost locations and highly innovative ones.

The crisis has been highly disruptive in many parts of the EU. It has reversed the long-term trend towards a narrowing of regional disparities. It has led to reductions in economic activity and employment in many Member States. However, the first signs of the convergence process resuming can be detected. Nevertheless, many regions still have a GDP per head and employment rate below their pre-crisis level.

Cohesion Policy has made a substantial contribution to economic cohesion. In the years between 2007 and 2014, around 400,000 SMEs received support under cohesion policy and more than 1 million new jobs were directly created. Nevertheless economic disparities still remain, requiring substantial amounts of investment beyond the present programming period if they are to be reduced.

This chapter describes recent trends in economic cohesion in regions and cities in the EU. It covers the differential trends in GDP per head across the EU and in the impact of globalisation as well as the factors underlying regional competitiveness, such as the extent of tertiary education, entrepreneurship, innovation and digital and transport networks. It also presents an aggregate indicator, the Regional Competitiveness Index, intended to summarise the different dimensions of competitiveness of EU regions.

The main concern throughout is to highlight the performance of the less developed regions and of different types of area, cities and rural areas, in particular.

1.2.ECONOMIC TRENDS AMONG EU REGIONS AND MEMBER STATES

1.2.1 Convergence back on track

In 2015, more than one in four EU residents (27%) lived in a NUTS 2 region with a GDP per head, in PPS terms,  1 below 75% of the EU average ( Map 1 1 ).

Most of them are located in central and eastern EU Member States, Greece, Portugal, Spain, and southern Italy, Portugal. They also include most of the outermost regions 2 . In Bulgaria and Romania, GDP per head is below 50% of the EU average in all regions, except for the capital city regions of Yugozapaden and Bucureşti-llfov.

Between 2000 and 2015, GDP per head increased relative to the EU average in all the regions in the central and eastern Member States ( Map 1 2 ). Growth was particularly high over the period in the capital city regions in Romania (from 56% of the EU average to 136%) and Bulgaria (from 38% to 76% of EU average in 2015).

In Greece, the situation deteriorated. In 2008, three of the 13 regions had a GDP per head above 75 % of the EU average, in 2015, just two - the capital city region Attiki (95 %) and Notio Aigaio, the southern Aegean islands (75 %).

In Portugal, only two regions in 2015 had a GDP per head above the 75% threshold, Lisbon (103 %) and Algarve (79 %), in both substantially lower than in 2008 before the crisis.

There are signs that the long-run process of regional convergence, which was interrupted by the economic crisis, has resumed. Prior to the crisis, disparities in GDP per head in the EU were shrinking (the coefficient of variation falling by 12 % between 2000 and 2008), mainly due to regions with the lowest levels of GDP per head growing faster than average ( Figure 1 1 ). In the crisis years, between 2008 and 2014, however, regional disparities widened slightly (the coefficient of variation increased by 4 % between 2008 and 2014, but remained well below the level in 2000). In 2015, disparities started to narrow again, though it is too early to say if this will be sustained.

Regional disparities in employment rates narrowed from 2013, though this was preceded by a significant increase as the result of the crisis and disparities in 2016 were much wider than in 2008. By contrast, reflecting the increased participation in the labour market, disparities in regional unemployment rates continued to widen, though at a slower pace than before 2012.

Figure 11: Coefficient of variation (index, 2000 = 100), GDP per head in PPS, employment rate, unemployment rate, EU-28 NUTS 2 regions, 2000-2016

Source: Eurostat and DG REGIO calculations.

Note: The coefficient of variation is weighted by the total regional population

Between 2000 and 2008, all the regions in the EU13 except Malta converged to the EU average ( Map 1 3 ), with big strides (more than 20 index points) in the capital regions of the Bulgaria, Czech Republic, Hungary, Romania and Slovakia as well as in the Baltic States. Most of the Greek regions converged, while the Italian regions and mainland Portugal diverged.

Between 2008 and 2015, all the regions in the EU13 converged except Cyprus and Praha. ( Map 1 4 ). The Baltic States who were hit hard by the crisis still converged. Greek regions experienced big reductions in their GDP per head relative to the EU average, more than reversing the convergence achieved between 2000 and 2008. Almost all Portuguese and Italian regions continued to diverge. Spain was also affected by the crisis and diverged, but not to the same extent as Greece.

Overall, the biggest relative increases in GDP per head between 2000 and 2015 occurred in the EU13, while the biggest reductions were in Greece and Italy, in the latter both before and after the crisis ( Map 1 2 ). But also a few regions in Belgium, the Netherlands, France and the UK also experienced big drops.

Note: Annual average change on previous year, in %

Figure 12: GDP per head growth rates of regions in less developed and moderately developed Member States, 2001-2008

Figure 13: GDP per head growth rates of regions in less developed and moderately developed Member States, 2009-2015

Note: capital regions are indicated in red. Regions are ranked by growth rates for the period 2001-2008 in both figures

Mainstream economic growth theories predict that the lower the initial GDP per head the higher growth will be. Indeed, growth was higher than average in both the less developed and transition regions (located mostly in less developed and moderately developed Member States 3   Figure 1 2 ), with GDP per head in regions in less developed and moderately developed Member States growing at a faster pace than the EU average.

The economic and financial crisis led to a reduction in GDP per head between 2009 and 2015 in around 40% of regions, located mainly in Ireland, Italy, Spain, Portugal and Greece; in most Greek regions, the reduction amounting to over 3% a year. The process of convergence was halted with several of the less developed and transition regions growing more slowly than the EU average ( Figure 1 3 ).

From 2000 onwards convergence was mainly driven by the catching up of the less developed economies. GDP per head, therefore, grew faster in real terms in the less developed Member States than in others over the period 2001-2016, except in 2010 and 2011, and it is forecast to continue to do so in 2017 and 2018 ( Figure 1 4 ).

From 2011 to 2013 the average growth rate in the moderately developed Member States was below that in the highly developed Member States, i.e. diverging. Only in 2014 did it overtake the rate in the highly developed Member States and growth in their GDP per head is forecast to be around 2.5% in both 2017 and 2018 (as against 3.5% in less developed Member States).

Figure 14: Growth of GDP per head in real terms, 2001-2018

Source: Eurostat, DG ECFIN, DG REGIO calculations

According to a European Commission reflection paper (European Union, 2017a), globalisation has a highly differentiated impact on EU regions. While some are well positioned to take advantage of the new opportunities it offers, others are hit by job losses, stagnating wages and shrinking market shares due to low-cost competitors moving into more technologically advanced sectors.

The best response to globalisation is a continuous effort to move up the value chain. This requires innovation, entrepreneurship, knowledge transfer and continuous upgrading of the skills of the labour force. Regions that are innovative and have a large share of high-skilled jobs and a highly educated work force are less likely to be hit hard by heavy job losses than others.

There are four important risk factors linked to globalisation and technological change: (1) a large share of employment in low-tech manufacturing, (2) rapidly increasing unit labour costs in manufacturing over the past decade which may compromise competitiveness and reduce market share, (3) a large share of working-age population with low educational attainment, and (4) a decline in employment in industry between 2000 and 2013 ( Map 1 5 ). Some 9% of EU regions, located in 7 different Member States, are at risk from globalisation by being exposed to up to four of these factors. Most are located in southern or central and eastern Europe, though there are also high risk regions in Denmark, France, Ireland and the UK. In many Member States, the situation is diverse with some regions being subject to three or four risks and others only one or none at all. These risks may diminish over time, though probably only slowly since changes in innovation or education attainment levels take time to be accomplished.

Map 15: Risks factors linked to globalisation and technological change

1.2.2 Less developed regions maintain a strong manufacturing sector, but their agriculture needs to modernise

In 1995, industry, excluding construction (i.e. mainly manufacturing), accounted for around 21% of both total employment and gross value-added (GVA) in the EU. The rise of services, automation in manufacturing and the relocation of parts of it to emerging economies has led to a steady reduction in both shares since then, to 19% in the case of GVA and 16% in the case of employment ( Figure 1 5 and Figure 1 6 ).

In less developed regions, the share of both GVA and employment in industry is, on average, larger than in the more developed and transition ones 4 . Moreover, the share of GVA increased over the period (from 21% to 24%) while the share of employment declined - though by less than in other regions –implying an increase in productivity in industry relative to other sectors.

The reduction in the share of employment in agriculture in the EU over the 20 year-period has been substantial, especially in less developed regions. In 1995, it accounted for around 9% of total employment and by 2012, the share had fallen to 5% when, because of low productivity – partly reflecting subsistence farming in EU-13 countries – the share of GVA was under 2% ( Figure 1 7 and Figure 1 8 ). In less developed regions, the share fell from 22% to 14% between 1995 and 2014 and as productivity increases, it is likely that it will fall further.

As the number of jobs in less productive segments of agriculture and industry declines, more jobs may be created in services and more advanced areas of industry and agriculture. Regions can indeed choose not to abandon agriculture and industry. Within global value chains, economies can increase their productivity by upgrading to higher value segments within the same sector (Sheperd, 2013). In addition, automation has made labour costs less relevant and may bring back some manufacturing firms to the EU, but the jobs they will offer will be different from those that were moved away in past years (European Union, 2017, OECD, 2016 and Eurofound, 2016). Training may help workers losing their jobs to gain news ones as the structure of economic activity shifts, but there is a limit to what it can achieve.

1.3.PRODUCTIVITY IN LESS DEVELOPED MEMBER STATES IS CATCHING-UP

Less developed Member States tend to have a different economic structure than the others, with more employment in agriculture and industry ( Table 1 1 ). 5 In 2016, the share of employment in agriculture was 11 percentage points higher in less developed Member States than in highly developed Member States (13% vs 2%). In 2016, the share of their employment in industry was around 21% (i.e. the same as in less developed regions), and 7 percentage points larger than in highly developed Member States (14% as in more developed regions).

Both agriculture and industry lost employment between 2001 and 2008 and between 2009 and 2016. The pattern for agriculture was the same: the less developed Member States had the fastest reduction in agricultural employment, followed by the moderately developed, with the slowest reduction in the highly developed Member States. GVA in agriculture on the other hand grew fastest in the less developed Members States between 2001 and 2008, but it did not grow at all between 2009 and 2016.

Industrial employment remained constant in the less developed MS between 2001 and 2008, while it shrank in the other groups of Member States. Joining the EU and the single market has created more potential for specialisation in higher value-added sectors, so less developed Member States may have been able to maintain a larger share of employment in industry because the balance between labour costs, productivity and accessibility represented an attractive location for manufacturers. Industrial GVA in less developed Member States grew three times faster than in highly developed Member States between 2001 and 2008 and four times faster between 2009 and 2016.

Employment and GVA in construction grew quickly, especially in the less developed countries in the run up to the crisis and fell sharply between 2009 and 2016 in all three country groups.

Over the period 2001-2008, GVA in industry in these countries increased by more than in other sectors, by much the same as in the business and financial sector (K-N). It increased even over the crisis years, 2009 to 2013, whereas it declined in both moderately developed and highly developed Member States.

By contrast, shares of employment and GVA in business and financial sector in the less developed Member States, which used to be small, increased towards those in the highly developed countries. The impact of the crisis was limited, both employment and GVA continuing to grow after 2008 but at slower rates than between 2000 and 2008.

The restructuring and modernisation of agriculture is still ongoing in the less developed Member States. In 2016, it accounted for 13% of employment – as against only 2% in the highly developed Member States – but for only 3.5% of GVA. Both shares are tending to decline as restructuring takes place and, along with the shares in moderately developing countries, are converging towards those in highly developed countries. 6  

Table 11 Change in employment and GVA by NACE sector by group of Member States, shares in 2016 and changes 2000-2016

Note: blue bars indicate positive changes, red bars indicate negative changes. Annual average change on previous year, in %

Note: Less developed: BG, EL, EE, HR, LV, LT, HU, PL, RO; Moderately developed: CZ, CY, PT, SK, SI; Highly developed: BE, DK, IE, ES, FR, DE, IT, LU, MT, NL, AT, FI, SE, UK.

1.3.1 Productivity and employment contribute to the economic recovery in the EU

In the years before the crisis, from 2001 to 2008, GVA per head in the EU grew by 1.7% a year in real terms, fuelled primarily by productivity growth of 1.2% a year, with increases in the employment rate adding another 0.6% a year ( Table 1 2 ). Productivity growth was also the main source of growth in GVA per head in less developed Member States, though both were substantially higher than the EU average, especially productivity growth (4% a year).

Between 2009 and 2016, GVA per head in the EU grew slightly (by 0.3% a year) productivity grew faster (by 0.6% a year) and the employment rate by less (0.2% a year, while the share of working-age population declined (by 0.4% a year) as opposed to it remaining unchanged as it did between 2001 and 2008. The number of Member States with a declining share of working-age population increased markedly between the two periods, from 8 to 27, Luxembourg being the only exception.

Over the 2009-2016 period, the less developed Member States had the highest growth in GVA per head (0.9% a year) mainly driven by an increase in productivity (1.2% a year) with only a slight increase in the employment rate (0.1% a year) but offset by a reduction in the share of working-age population (0.4% a year). The moderately and highly developed Member States followed a similar pattern, but with lower growth in GVA per head (0.4% and 0.2% a year, respectively) and productivity (0.7% and 0.4% a year).

Between 2009 and 2016, GVA per head grew in all of the less developed Member States except Greece (where it fell by 3.2% a year) and Croatia (by 0.7%a year). Productivity growth was relatively high (between 1.4% and 2.8%) in five of the nine countries, but employment rates either fell or increased only slightly, except in Lithuania and Hungary.

Among the five moderately developed countries, GVA per head declined in Cyprus (by 1.8% a year), Portugal (by 0.2% a year) and Slovenia (by 0.3% a year) mainly due to a fall in employment rates.

Among the highly developed countries, only Italy and Finland had a decline in GVA per head (1% a year) between 2009 and 2016. Both experienced a reduction in productivity and the employment rate fell as well in Italy.

Table 12: Decomposition of annual average change in GVA per head by Member State, 2001-2008 and 2009-2016

Source:    EUROSTAT, DG REGIO calculations; for Malta, real GDP was used instead of real GVA.

Note:    Blue bars indicate increases, red bars indicate reductions. Annual average change on previous year, in %.

1.3.2 Capital metropolitan regions more prone to boom and bust than in other regions

In 2014, metropolitan (metro) regions accounted for 58% of population in the EU, 61% of employment and 67% of GDP.

Accordingly, they are major centres of employment and business activity with higher productivity than elsewhere.

In both the EU-15 and EU-13, real GDP per head in metro regions grew faster than in other regions in the pre-crisis years between 2001 and 2008 ( Table 1 3 ). Growth rates in capital city regions were especially high, mainly fuelled by higher productivity growth in the EU-15 and higher employment growth in the EU-13.

The crisis had a different effect on the metro regions in the EU-15 than on those in the EU-13. In the EU-15, GDP per head in the capital metro regions declined at the same rate as in other regions between 2009 and 2014. In the EU-13, it was rather stable in the capital metro regions, whereas it grew in the other regions, mainly fuelled by increases in productivity. In both the EU-13 and EU-15, there was a reduction in employment in all types of regions.

In the EU-13, growth of GDP per head in non-capital metro regions over the period 2009-2014 was, on average, twice the EU-13 average as a result of high productivity growth while employment remained unchanged. Whether this launches a period of higher growth outside the capital regions, so a narrowing of the gap in GDP per head with the latter, remains to be seen.

Table 13 Changes in GDP per head, productivity and employment per head by type of metropolitan region, 2001-2008 and 2009-2014 (average % per year)

Source: EUROSTAT, DG REGIO calculations

Note: green bars indicate positive changes, red bars indicate negative changes. Annual average change on previous year, in %

Employment in both metro and non-metro regions generally increased between 2000 and 2008, though at a faster rate in capital city regions than others and by more in other metro regions than non-metro ones ( Figure 1 9 ). In the next two years, it declined markedly in all regions, but it then began to recover in the capital city regions, continuing to grow up to 2014 when the number employed was much the same as before the crisis. In the other metro regions, recovery was more hesitant and by 2014, employment was still below the level in 2008. In the non-metro regions, employment continued to decline up to 2013 and began to increase only in 2014.

Figure 19: Evolution of total employment (number employed) in metro regions, 2000-2014

Source: Lavalle et al. (2017)

1.3.3 GDP growth in rural and intermediate regions proved to be more resilient during the crisis years

Between 2001 and 2008, real GDP per head in rural regions ( Table 1 4 ) in the EU-28 grew by 1.9% a year, slightly higher than in other types of region. At the same time, productivity grew faster, while employment relative to population rose more slowly.

In the EU-15, GDP per head grew in all types of region, fuelled in equal parts by increases in productivity and the employment rate, though in rural regions more by productivity.

In the EU-13, in the years before the crisis, economic growth was mainly driven by increases in productivity, especially in rural regions, where increases were accompanied by a decline in employment. The two may be linked, insofar as higher productivity growth was due to catching up in the use of technology and more efficient methods of working, including in agriculture, which in turn led to a reduction in employment.

The crisis had a different effect on rural regions than others, since construction and industry were most affected and these are less present in rural areas. Accordingly, the reduction in GDP per head between 2009 and 2014 was less pronounced in rural regions than in urban ones, particularly in the EU-15. In the EU-13, GDP per head grew over this period in all types of region and at much the same rate, but in all cases by much less than before the crisis.

Employment declined in all types of region, but more in urban and intermediate ones in the EU-15 and in urban and rural ones in the EU-13.

Productivity continued to grow in both the EU-15 and EU-13 and, as in the pre-crisis period, by more in the latter than the former, though the difference in rates was much smaller.

In 2014, GDP per head in rural regions in the EU-15 was, on average, some 72% less than in urban ones, while in the EU-13, the difference was much wider, the level in urban regions being only 42% of that in rural ones.

Table 14: Real GDP per head, productivity and employment per head growth by urban-rural typology, 2001-2008, and 2009-2014

Source: EUROSTAT, DG REGIO calculations

Note: green bars indicate positive changes, red bars indicate negative changes. Annual average change on previous year, in %

1.4.THE ECONOMIC DEVELOPMENT CLUBS OF EUROPEAN REGIONS AND THE MIDDLE-INCOME TRAP 7

Economy-wide forces together with differences in the characteristics of economies mean that it is possible to divide countries, regions and cities by their level of economic development. They can be said to belong to different ‘development clubs’, each of them characterised not only by different income levels but also by different structural features, such as the education level of population, infrastructure endowment, innovation capacity and institutional quality.

Clubs differ systematically across these dimensions and for each club there are specific needs and challenges related to its starting point. Grouping EU regions into income clubs is a way of generating insights into economic development and provides a distinctive perspective on regional policy. It brings out the uneven path of regional development that occurs and helps to identify means of overcoming the barriers to development in lagging regions. For this purposes, EU NUTS 2 regions can be divided into four groups according to their GDP per head in 2013 (see box below).

Most of the very-high and high income regions are located in a band from London through the Benelux and Germany down to northern Italy, with a few capital city regions outside this area ( Map 1 6 ). There are two other broad areas, a large middle-income part in the west of the EU and a low income part in the south and the east.

The very-high income club is dominated by a few very large urbanized or capital city-regions, and by a number of smaller but highly urbanized inter-connected ones (e.g. Rhine-Ruhr in Germany or Randstad in the Netherlands), specialized in the production of high-quality goods and services.

Map 16: The economic development clubs of EU regions

Source: EUROSTAT, REGIO-GIS elaborations.

The high-income regions share many characteristics with the very-high income ones but tend to be less city-centered. The medium-income club is vast and consists mainly of regions in the north-west of Europe outside the very-high and high income clubs. The low-income club is concentrated in the east and south of the EU.

The structural characteristics of the four clubs differ markedly ( Table 1 5 ).

Table 15: European regions, by income club: some stylised facts

Source: EUROSTAT, Cambridge Econometrics, DG REGIO calculations based on the latest available data.

Total population change varies with the club gradient, with people moving to higher-income regions and away from low income ones. Many high-income regions experienced high rates of population increase over the period 2001-2015, except for those in Germany. In many low-income regions in the east and south of the EU as well as in declining industrial parts of north-eastern France and northern England, population declined. While some low-income regions experienced population growth over the period, these tend to be those with extensive amenities and a low cost of living.

Examining the labour market in the different clubs provides further insights. Employment declined between 2001 and 2014 while it increased in the other regions, especially in the very-high income ones. The share of employment in industry (excluding construction), as observed above, is largest in low-income regions. In all clubs, however, employment in industry declined over the period, the more so in the middle-income ones. Low income regions, as also seen above, have the highest unemployment.

Patenting activity, which is an indicator of innovation, is highly concentrated in very-high and high income regions.

Very-high and low income regions experienced the highest growth in GDP per head over the years 2001-2015. In the former, this is mainly due to their level of competitiveness and specialisation in the production of high-quality goods and services, while low-income regions are catching-up, taking advantage of their ability to mobilise low-cost capital and labour to capture activities for which this gives them a competitive edge. Middle-income regions had the lowest growth and face a particular challenge – the so-called 'middle-income trap' – because they are neither very cheap, nor are they particularly innovative or productive. Their manufacturing sector tends to be smaller and weaker than in regions with either a higher GDP per head or lower one ( Figure 1 10 and Figure 1 11 ) and their costs are too high to compete with the former, their innovation systems not strong enough to compete with the latter.

The main challenges for regions in each club can be summarised as follows:

1. Very-high income club: many of these regions are attracting population, though some of them have high unemployment rates and have under-performed since the beginning of the economic crisis (Dijkstra et al., 2015). The main need is to keep pace with global competitors. They need to maintain their specialisation in high-wage activities and their comparative advantage by continuing to push the boundaries of innovation and technology.

2. High income club: regions in this group share many characteristics with the very-high income ones. Their employment rates are high and the challenge is to remain innovative, but they are more vulnerable to competition from the lower income regions. They are particularly vulnerable to standardisation of what they produce which can allow firms to move to regions with lower costs and less-skilled labour. Their challenge is to innovate in their areas of specialisation and to expand into high value-added activities related to this.

3. Medium income club: this is a large group consisting of two sub-groups, each with specific challenges. One consists of regions that have lost manufacturing jobs and in which the education level of the work force is below that in higher income regions. In general, they are fragile economically because of this. The other consists of regions experiencing population growth, but mainly of older people who move there because of the local amenities and low cost of living. Such inward movement mainly stimulates employment in non-tradeable local services, which gives rise to limited skill development, innovation capacity and export capability. Regions, in both sub-groups, risk falling into a 'middle-income trap'. As productivity and wages rise, they become less attractive for labour-intensive, low-skilled activities. Moving up the value chain requires higher investment per worker than in earlier stages of development, because of the need for a better educated labour force and new business models. To become attractive for higher value-added activities, regions have to improve the quality of their institutions and business ecosystem, become more innovative and improve the skill sets of their labour forces through better education and training.

4. Low income club: these regions suffer from having low levels of technology and business organization and a work force with limited skills, but they have the advantage of offering low-cost land and labour. They tend to lose talented people and well-educated young people to higher-income regions, while at the same time being unable to attract firms and talent from outside, so encouraging an outward movement of population.

1.5.COMPETITIVENESS OF EU REGIONS

1.5.1 Firms in EU Capital metro regions tend to be larger and to grow at a faster pace.

In the 2014-2020 period, Cohesion Policy is focused heavily on supporting smart growth with particular emphasis on innovation and high growth firms and with programmes aimed at increasing the innovative capacity of SMEs. In previous periods too, a substantial share of Cohesion Policy funding was devoted to improving the business environment and supporting entrepreneurship. In the 2007-2013 programming period, for example, some EUR 47.5 billion, 24% of the total ERDF, was allocated to support of SMEs. 8

In 2014, the largest number of firms with at least one employee 9 relative to population was in capital metro regions in most countries (the exceptions are France, Italy, Austria, and Spain ( Figure 1 12 ). There are, however, large variations across regions in the same Member State, particularly in Romania, Slovakia and Hungary. Firms, especially large firms, may locate in more urbanised areas to benefit from agglomeration economies, the three main sources of these being matching, sharing and learning (Puga, 2010). Cities, therefore, tend to have larger labour markets allowing a better matching between labour demand and supply, a better sharing of inputs, such as infrastructure, in the production process and more people working and living in close proximity enabling them to learn more easily from each other.

Figure 112: Firms density by metro region, 2014

At the same time, firms operating in urban areas face more competition, since larger markets attract more firms. This tends to result in less competitive firms being forced out of business (Melitz and Ottaviano, 2008; Combes et al., 2012). The data, indeed, show that firms in metro regions, particularly in capital city ones, are on average, larger in terms of employment than those in non-metro regions, apart from in Latvia and Hungary ( Figure 1 13 ) 10 .

Figure 113: Employees per firm by metro region, 2014

The birth of enterprises is one of the main drivers of job creation and economic development. Young enterprises are often innovative and tend to increase the competitiveness of a region both directly and indirectly by pushing competitors to become more efficient.

In 2013, (the latest year for which data are available) newly-created enterprises were more numerous in capital metro regions, both in more developed and less developed Member States, except in Spain and Italy, the highest birth rates being in Bratislava and Budapest ( Figure 1 14 ).

Figure 114: Enterprise birth rate per metro region, 2013

High birth rates often go together with high death rates (Figure 115), as in Bratislava and Budapest. However, some regions have high rates of start-up but low death rates, as in Copenhagen, hinting at local features which nurture the birth of new enterprises while also keeping them profitable.

Figure 115: Death rate of enterprises, 2012

High growth enterprises (those growing by 10% a year or more) 11 play an important role in the economic growth of cities and regions through their contribution to productivity and innovation (Acs et al., 2008).

In 2014, high growth firms were found mainly in metro regions, except in Portugal and Italy, though there were marked variations in their incidence within countries (Figure 116). In a number of Member States - Slovakia, Hungary and the Czech Republic in particular –, the large variation between regions is due mainly to the large number of high-growth firms operating in the capital metro region. 12

Figure 116: Number of high growth firms per 1000 inhabitants, in 2014

Source: Eurostat, Business Demography, DG REGIO calculations

Note: France, Denmark, 2013

(1)

   Gross Domestic Product (GDP) per head in Purchasing Power Standards is the total value of all goods and services produced per inhabitant. Purchasing Power Standards (PPS) adjusts for differences between countries in purchasing power due to differences in price levels.

(2)

   The European Union includes 9 Outermost Regions, which are a long way from the European continent. They are: Guadeloupe and La Réunion, Mayotte, French Guiana and Martinique, Saint-Martin (France); Madeira and Azores (Portugal) and Canary Islands (Spain).

(3) See the Lexicon section for the list of less developed and moderately developed Member States.

(4) See Lexicon for a definition of ‘less developed’, ‘transition’ and ‘more developed’ regions.

(5) This section analyses data at the country level because of the unavailability of regional data on sectoral employment for 2015 and 2016 and partly for 2014.

(6) However, in some Member States agriculture has a social function as it absorbs labour in times of crises. Of course, this social cushioning muddles the real productivity figures of the sector.

(7) Prof. Simona Iammarino, Prof. Andrés Rodríguez-Pose, and Prof. Michael Storper substantially contributed to the content of this section.

(8) European Union (2016), Support to SMEs – Increasing Research and Innovation in SMEs and SME Development. Ex post evaluation of Cohesion Policy programmes 2007-2013, focusing on the European Regional Development Fund (ERDF) and the Cohesion Fund (CF), Executive Summary, available at: http://ec.europa.eu/regional_policy/sources/docgener/evaluation/pdf/expost2013/wp2_final_en.pdf .

(9) The terminology employer firms will be used throughout the chapter to indicate firms employing at least one employee.

(10) Some care is needed in interpreting this result. Some large enterprises may be composed of multiple local units which may be located in different regions, but with their employment registered in the head office, often located in the capital of a country. This may inflate the number of employees that are counted as working in the capital city.

(11) A high-growth enterprises are those in which employment increased by 10 % a year or more over a three-year period and which had at least 10 employees at the beginning of the period.

(12) As indicated above, perhaps at least partly because of employment in local units being registered in the head office.

EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

1.5.2 Innovation remains spatially concentrated

As widely documented in the economic literature, research and innovation play a crucial role in determining the economic performance of countries and regions. Innovation, understood in the broad sense to include product, process, market and organisational innovation, is identified as one of the major engines of economic growth, employment and environmental sustainability and accordingly is of critical importance for social progress as well as prosperity.

In particular, innovation is an important driver of long-run productivity growth and, as such, is crucial in maintaining the competitiveness of firms. This is particularly true for firms in Europe which have increasingly to compete with firms located in less developed parts of the world in emerging economies. The latter are not only catching up fast in terms of technology but they also still benefit from cheaper labour due in part to lower labour standards, a lack of social protection for workers and lower income expectations, though low labour costs tend to be offset by lower productivity. From this perspective, innovation, as well as the capacity to assimilate innovation produced elsewhere, is an important condition for maintaining the specific features of the European social model. In addition, contrary to growth from restructuring, growth from innovation is in principle without bounds, which is why it is central to sustaining growth over the long-term.

Measuring innovation is difficult, the number of patent applications being one of the few indicators available and the most commonly used one. Although it is imperfect because it covers only innovations which are patentable and, in the case of the EU, only those registered at the European Patent Office, there is a lack of alternatives. Over the two years 2010 and 2011 (the last data available), an average of some 113 patent applications per million people was made to the European Patent Office ( Map 1 7 ). While there are large variations in applications across regions, there is a clear spatial pattern, with those with most applications – i.e. the most innovative - being located mostly in the Netherlands, Germany, Austria, Denmark and Sweden. At the NUTS 3 level, Eindhoven, in the Netherlands, had the highest number of applications (1 731per million inhabitants in the period), followed by Heidenheim in Germany (1 049) and Rheintal-Bodenseegebiet in Austria (832). 1

Metropolitan areas tend to offer an environment which is particularly conducive to the introduction of new ideas, products and processes. A vast body of literature explains why urban areas are likely to be more innovative than others, such as the presence of a creative and skilled work force, specialised clusters of economic activity, universities and research institutes. 2

There are not only clear-cut differences in patenting activity between metro regions (around 140 applications per million inhabitants) and non-metro regions (around 86 per million) (Figure 117), but there is less variation between them (as measured by the coefficient of variation), suggesting that they offer a more favourable environment. 3

Figure 117: Patents by metro regions, average 2009-11

Source: Eurostat, DG REGIO calculations.

Map 17: Patent applications to the European Patents Office, 2010-2011

Source: Eurostat, REGIO-GIS

One of the main indicators for assessing the capacity to innovate is the level of expenditure on R&D which tends to be essential for technical progress to take place. 4  

Expenditure on R&D in the EU-28 amounted to around 2% of GDP in 2014 (the latest data available) and only marginally increased over the previous two decades (1.8% of GDP in 1995), not by nearly enough to close the gap with other highly developed economies, especially Japan (where expenditure in 2014 amounted to 3.5% of GDP) or the US (where it stood at 2.7% of GDP in 2013).

Regions in the EU-15 have on average slightly higher expenditure in R&D (2.1% of GDP in 2014) than those in the EU-13 (1.8% of GDP). There are, however, wide variations across NUTS 2 regions, from over 6% of GDP in Brabant Wallon in Belgium and Braunschweig and Stuttgart in Germany to only 0.1% of GDP in Centru in Romania and Severen Tsentralen in Bulgaria (Map 18 and Figure 118).

Expenditure on R&D in 2014 exceeded the Europe 2020 target of 3% in only 30 regions, all of them in the EU-15 ( Table 1 6 ). In the more developed regions, it was less than 1 percentage point below the target on average – though still a significant amount given the generally slow rate of increase over recent years – while in less developed regions, it was slightly over 2 percentage points below.

Table 16: Total R&D expenditure and the distance to the EU2020 target, EU-28 regions, 2014

In general, therefore, regions with the highest expenditure on R&D are the most highly developed ones, and in most cases those where the capital is located (Germany, France, and the UK are exceptions). Of the 20 regions with the highest expenditure, 19 regions have a GDP per head above 100% of the EU average and regions with low levels of expenditure tend to be either in southern, central and eastern Member States or are the low GDP per head ones in western Member States.

Figure 1-18: Total expenditure on R&D, 2014

Source: Eurostat, DG REGIO calculations.

Source: Eurostat, REGIO-GIS.

A 2017 European Commission report highlights the key role innovation plays in the development of regions, and not only the high-tech ones. 5 The Regional Innovation Scoreboard (RIS), an extension of the European Innovation Scoreboard, assesses the performance of regions in this respect on the basis of a limited number of indicators. For 2017, it covers 220 regions across 22 EU Member States and Norway while Cyprus, Estonia, Latvia, Lithuania, Luxembourg, and Malta are covered at country level.

The most innovative region in the EU by this measure is Stockholm, followed by Hovedstaden in Denmark, and South East England (Map 19).

Despite regional variations within countries, the ranking of regions largely matches that of countries. Most of the regional Innovation Leaders are in countries also identified as Innovation Leaders and almost all of the regional Moderate and Modest Innovators are located in countries categorised in the same way. However, regional ‘pockets of excellence’ are evident in some Moderate Innovator countries (such as, Praha in the Czech Republic, Bratislavsky kraj in Slovakia, and Pais Vasco in Spain), while some regions in strong innovation countries lag behind.

The assessment of regions in terms of innovation has changed over time. Between 2011 and 2017, 128 regions (60% of the total) improved their performance, while for 88, performance worsened. Although 75% of Innovation Leaders improved their performance, only 30% of Modest Innovators did so, implying a widening gap between them.

Performance declined in particular in more peripheral regions, in all regions in Romania and for more than half of those in Denmark, Finland, Germany, the Czech Republic, Hungary, Portugal, and Spain. It increased in all regions in Austria, Belgium, France, the Netherlands, Norway, Slovakia, Switzerland, and the UK and in more than half of those in Greece, Italy, Poland, and Sweden.

Map 19: Regional Innovation Scoreboard, 2017

Source: Regional Innovation Scoreboard 2017, REGIO-GIS.

In general, the RIS confirms the wide diversity of regions in terms of innovation performance, so highlighting the fact that innovation has a strong regional dimension. Given this wide variation, measures for supporting innovation, including Cohesion Policy programmes, need to take explicit account of the regional or local context when devising the kind of support to provide. The smart specialisation approach is helping in this regard.

1.5.3 The number of people with tertiary education keeps increasing, but large disparities persist

A well-educated work force is the key to economic development and prosperity. Higher education boosts upward social mobility and improves employment prospects. In 2016, people aged 25-29 with a tertiary education had an employment rate of 81%, compared to 74% for those with an upper secondary education and 54% for those without an upper secondary education. 6

The link between educational attainment and employment rates is also strong for the population aged 25-64. Only 54% of those without an upper secondary education are employed in 2016 as against 75% of those with upper secondary qualifications and 85% of those with tertiary education. Moreover, the gap in employment rates between those with tertiary education and those with only basic schooling has widened over time (from 28 percentage points in 2006 to 31 percentage points in 2016). 7

The share of people aged 25-64 with tertiary education, however, varies markedly across regions (Map 110 and Figure 119).

Metropolitan areas, especially larger ones, tend to have a more highly educated population than other areas. 8 Demand for highly skilled labour attracts those with such qualifications and makes it easier for them to find a job matching their skills. At the same time, firms are also more likely to find the skills they need in such areas. In 2016, around 41% of those aged 25-64 had tertiary education in capital metro regions and 32% in metro regions generally, as compared with an average of 30% in the EU as a whole. The highest figures were in Inner London, Brabant Wallon in Belgium and Helsinki, the lowest (below 20%) in regions in Italy and Romania (Map 110).

Figure 119: Population aged 25-64 with tertiary education, 2016

The Europe 2020 strategy has a target of increasing the share of the population aged 30-34 with tertiary education to 40% by 2020. The situation in 2016, however, varies markedly between regions, largely according to their level of economic development. Over half the 81 more developed regions had already achieved the target (some before it was set in 2010). Some 22% of transition regions had also achieved the target (as compared with none up to 2013), while 29% of less developed regions had done so (Table 17). 9

Table 17: Population aged 30-34 with a tertiary education, by regional groups, 2016 *

Ensuring that everyone has the right skills for an increasingly digital and globalised world is essential for an inclusive labour market and to spur innovation, productivity and growth (OECD, 2016). In 2015, around 25% of those aged 25-64 reported having a low level of digital skills and 29% a basic level, while 28% reported having a higher level than basic. The situation at EU level, however, hides marked differences between Member States, particularly between those with different levels of economic development, digital skills tending to increase with the latter (Figure 120). Whereas 35% of those in highly developed Member States reported their digital skills to be above basic, in less developed Member States, the figure was only 21%.

Figure 120: People’s levels of digital skills, by level of economic development, 2015

Source: Eurostat, DG REGIO calculations.

1.5.4 Improving market access does not always generate growth

Investment in transport infrastructure is widely used to promote economic development, but its actual impact on the economy is complex and hard to predict. In a number of cases across the EU, projections of transport demand made before the infrastructure had been built to justify the investment concerned have proved to be too optimistic. This is demonstrated by several severely under-used motorways, airports and high-speed railway lines (Flyvbjerg et. al., 2003, European Union, 2014b).

In principle, lowering transport costs should boost trade and economic growth. The new economic geography theory of regional development, however, warns that improving transport connections between two cities may not necessarily help both even if it improves overall productivity. For example, if a city with less efficient firms is connected to one with more efficient firms, the latter might capture the market in the other city, leading to a reduction of economic activity there.

Regional access to markets by road is mainly determined by the spatial distribution of population. A remote region will always have a small market even with large-scale road investment. Accordingly, transport investment, especially in areas with a mature network, cannot radically alter market access. Potential accessibility by road is the highest in regions and cities in the centre of the EU (European Union and UN-HABITAT, 2016). Many regions in central and eastern Member States, however, are not yet connected by an efficient road network and will only have better access to markets after the completion of the Trans-European Transport Network ( Map 1 11). 10

The speed and frequency of trains is also much lower in central and eastern EU countries (Poelman and Ackermans, 2016). While some countries, such as the Czech Republic and Hungary, have a relatively dense rail network, the frequency and speed of service on many of the lines makes it an unattractive alternative to travel by car (Map 112).

Accessibility by rail is very high in the areas in and around the highly urbanised parts of the UK, the Netherlands, Belgium, northern France and the Rhine-Ruhr region in Germany. This is due to the combination of a high concentration of population, a dense rail network, high-speed rail connections and relatively high frequency of service (Map 1-13). Accessibility is still high in and around cities in western and eastern France, many parts of Germany, the north of Italy and some parts of Spain. It is relatively low in Austria and Switzerland due to the mountainous terrain and lower still in more peripheral western parts of the EU, in Ireland, Portugal and Spain, and in the Nordic countries, where there are longer distances between cities and low population density. In most of the eastern parts of the EU, as noted above, accessibility is low because of low frequency of service and slow speeds.

By 2050, the EU intends to complete a European high-speed rail network, the aim being for rail, both high and normal speed, to account for at least 50% of all medium-distance passenger travel. 11 This will require substantial investment, especially in countries where the network is not very dense and the service tends to be slow and infrequent.

Access to passenger flights is highly uneven across the EU, ranging from London and surrounding areas where people have access to over 3000 flights a day to regions in eastern Poland and Romania without any flights within 90 minutes driving time (Map 1-14).

(1)

   Note that the patent applications relate mainly to technological innovations in the manufacturing sector and do not a capture innovation in services, which are often intangible. It is therefore liable to be biased (Morrar, 2014).

(2) European Union and UN-HABITAT (2016).

(3) The coefficient of variation calculated on the average number of patent applications in 2010 and 2011 in metro regions is 1.1, as against 1.4 for non-metro regions.

(4) One should note however that R&D expenditure is likely to underestimate innovation activities, particularly in sectors outside manufacturing where non-technological innovation is frequent.

(5)    Regional Innovation Scoreboard 2017, available at: : http://ec.europa.eu/growth/industry/innovation/facts-figures/regional_en .

(6) COM(2016), Education and Training Monitor 2016.

(7) COM(2016), Education and Training Monitor 2016, page 27.

(8) European Union and UN-HABITAT (2016).

(9) European Union (2014a).

(10) The map depicts expected changes relative to the situation in 2012.

(11)

   European Commission 'White Paper, Roadmap to a Single European Transport Area – Towards a competitive and resource efficient transport system.' COM(2011) 144 of 28.3.2011.

EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

1.5.5 Digital networks are spreading, but closing the gap between urban and rural areas represents a major challenge

Access to high capacity telecommunication networks is vitally important for competitiveness and growth. The use of digital services and the capacity to operate successfully in a global business environment increasingly rely on fast and efficient broadband connections. ICT infrastructure is therefore a major determinant of the development potential of EU regions. The most prosperous regions are in general already well-endowed in this regard, though there are still serious gaps in many of the less prosperous ones and pronounced disparities between urban and rural areas.

Over 214 million EU households (98%) had access to at least one of the main fixed or mobile broadband technologies (excluding satellite) in mid-2016. If satellite coverage is included, basic broadband services are now available to every household in the EU, so that the European Commission’s Digital Agenda for Europe target of basic broadband for all has been achieved. 1

The coverage of Next Generation Access (NGA) 2 is expanding fast. In 2016, around 76% of households across the EU have access to at least one NGA network, up from 68% at the end of 2014, though there are wide variations in coverage between and within Member States ( Map 1-15 ).

Access to fast broadband services in rural areas remains a challenge. Even though 99% of rural households across the EU28 had access to at least one broadband technology at the end of June 2016, only 39% (12 million households) had access to NGA broadband ( Figure 1-21 ), with almost no households with access in rural areas in Greece (0.3%). Substantial progress has been made since 2012 ( Figure 1-21 ). The funding provided under rural development policy to an expected 4 400 projects to install ‘last-mile’ connections to larger broadband projects co-financed by other EU funds is planned to improve access to ICT infrastructure and services for an estimated 18 million people living in rural areas.

Coverage is almost complete in most urban areas and cities, though there are a number of areas where it is well below the EU average (of 82% in urban areas), mostly in Greece (55%) and France (50%).

Map 1-15: Next generation access coverage in NUTS 3 regions, 2016 

Source: DG CONNECT: Europe's Digital Progress Report 2017, REGIO-GIS.

Figure 1-21: Households with access to Next generation Access (NGA) broadband, by type of area, 2012 and 2016.

Source: European Commission, 2016, Broadband Coverage in Europe 2016.

Data are for the end of 2012 and mid-2016.

Household take-up of broadband has increased markedly in recent years along with coverage. While in 2009, only around 56% of households in the EU had a broadband subscription, the figure was over 72% in 2012 and it had increased to 83% in 2016. However, large differences remain between regions ( Map 1-16 ). In 2016, the proportion of households with broadband was below 60% in Kentriki Ellada in Greece and Severozapaden and Yugoiztochen in Bulgaria, while it was over 95% in the large majority of regions in the Netherlands and in Helsinki-Uusimaa in Finland, South-East England and Luxembourg.

Map 1-16: Households with a broadband connection, 2016

Source: Eurostat, REGIO-GIS.

1.6.CAPITAL AND METRO REGIONS ARE THE MAIN DRIVERS OF REGIONAL COMPETITIVENESS IN EUROPE

The Regional Competitiveness Index (RCI) is designed to capture the different dimensions of competitiveness for NUTS 2 regions and is the first measure to provide an EU-wide perspective on this. The 2016 edition follows the two previous ones published in 2010 and 2013 (Annoni and Kozovska, 2010; Dijkstra, Annoni and Kozovska, 2011, Annoni and Dijkstra, 2017a). All three of them are built on the same approach as the Global Competitiveness Index of the World Economic Forum (GCI-WEF). The 2016 index is based on 74 mostly regional indicators covering the 2012-2014 period though with a number of indicators for 2015 and 2016.

The index is based on a definition of regional competitiveness from the perspective of both firms and residents (Dijkstra et al., 2011):

Regional competitiveness is the ability of a region to offer an attractive and sustainable environment for firms and residents to live and work in.

The RCI results for 2016 are in line with those for 2013. Once again, a polycentric pattern is evident with capital and other metro areas being the main centres of competitiveness. Spill-over effects are evident in most of the north-west of the EU, but less so in the in the east and south. As in 2010 and 2013, there is substantial variation both between countries and within them, the latter, in many cases, due to the capital city region significantly out-performing others in the country ( Map 1-17 ).

The so-called ‘Blue Banana’, a highly urbanised, industrialised corridor defined in 1989 by a group of French geographers led by Roger Brunet, with Greater London at one end and Lombardia at the other and encompassing the Benelux countries and Bavaria, is not evident on the RCI map. On the contrary, the RCI shows strong capital and other metro regions in many parts of Europe. In some countries, capital city regions are surrounded by others that are similarly competitive, indicating the presence of spill-over effects, but in many other countries, the regions neighbouring the capital are far less competitive. An important question for the future is whether the strong performance of the capital and other metro regions concerned will help to strengthen the performance of neighbouring ones or whether the gap between them will widen.

London and its commuting area, which includes seven NUTS 2 regions, 3 is ranked top in 2016, ahead of Utrecht in the Netherlands –for the first time is not the most competitive region – which is ranked joint second with Berkshire, Buckinghamshire and Oxfordshire in the UK. 4 As in 2010 and 2013, most of the top-ranked regions include either capital cities or large metropolitan areas which help to boost their competitiveness. The regions at the other end of the scale are mainly in Greece and Romania with one in Bulgaria.

Capital city regions tend to be the most competitive in their countries ( Figure 1-22 ). The only exceptions are in Germany, Italy and the Netherlands. In the last, the capital city region is ranked second and in Italy, Lombardia continues to be the most competitive one as in previous years. In Germany, many regions are more competitive than Berlin, which may be due to the relatively short time it has been the capital of a reunited country.

The gap between the capital city region and others is particularly wide in some countries, especially in Romania, Greece, Slovakia, Bulgaria and France. A big gap of this kind is generally a reason for concern as it puts substantial pressure on the capital city region while possibly leaving resources in other regions underutilised.

The gap between the capital city region and the second highest-ranking one is relatively small in the UK, Austria and Belgium. However, a small gap does not necessarily mean that the whole country is highly ranked. For example, in Belgium and the UK, variations between regions are relatively wide, highlighting the limitations of a national-level analysis. Such variation raises questions over whether gaps in regional competitiveness are harmful or not for national competitiveness and how far t they can, and should, be reduced.

Figure 1-12 - Regional competitiveness index, 2016

Source: Annoni et al. (2017b).

Map 1-17: Regional Competitiveness index, 2016

Source: Annoni et al. (2017b), REGIO GIS.

The changes over time in the RCI scores, as opposed to the rankings, are informative. 5 Even though the index is not entirely consistent between years because of recurrent and often unavoidable revisions of regional indicators and the NUTS classification, the three editions of the RCI provide a unique means of monitoring and assessing the development of regional competitiveness across the EU. Map 1-18 shows the regions where the scores changed by more than 5% of the difference between the highest and lowest scores across the three editions (i.e. the maximum score range). The three maps show the changes between 2013 and 2016, 2010 and 2013 and over the period as a whole. Between 2013 and 2016, competitiveness improved in around 10% of regions and weakened in another 10%, while between 2010 and 2013, it improved in many more regions (26%) than it weakened (11%).

Between 2010 and 2013, competitiveness improved in most Belgian and German regions. While it remained largely unchanged between 2013 and 2016 in most of the latter, it weakened in several Belgian regions, including in the capital city region. Competitiveness also deteriorated significantly in Greek and Irish regions between 2010 and 2013, and failed to improve over the following three years. In regions in many countries (Austria, Bulgaria, Czech Republic, Denmark, Spain, Finland, Hungary, Poland, Portugal, Romania, Sweden and Slovakia), competitiveness as measured remained largely unchanged over the 6 years.

In the other countries, there were quite a few changes. In France, competitiveness improved in 12 regions between 2013 and 2016 and four between 2010 and 2013. Conversely in the UK, it improved in many fewer regions between 2013 and 2016 (4) than between 2010 and 2013 (9).In Italy, it deteriorated in four regions in the first period and remained unchanged in all regions over the following three years. In the Baltic countries, competitiveness improved between 2013 and 2016 in Latvia and Lithuania, while it remained unchanged at a relatively high level in Estonia

Map 1-18: Changes in RCI, 2016-2013; 2013-2010 and over the whole period, 2016-2010. 

Note: Regions with an increase of over 5 % in the RCI range (z-scores) are categorised as improving in terms of competitiveness and with a reduction of over 5 % as deteriorating.

As might be expected, there appears to be a positive relationship between regional competitiveness and GDP per head, which is evident for both those both with high levels of the latter and those with low levels ( Figure 1-23 ).

Figure 1-23: Relationship between RCI and GDP per head (in PPS), by level of economic development

There is some evidence that regions which are more competitive have higher rates of start-ups, at least those which are most highly developed and those which are least developed ( Figure 1-24 ).

Figure 1-24: Relationship between RCI and the birth rate of firms (relative to population), by level of development

1.7.REFERENCES

Acs, Z. Parsons, W., and Tracy, S. (2008), High Impact Firms: Gazelles Revisited, Office of Advocacy, U.S. Small Business Administration.

Annoni, P. and Kozovska, K. (2010), EU Regional Competitiveness Index 2010, EUR 24346, Luxembourg: Publications Office of the European Union.

Annoni P., Dijkstra L. (2017a) Measuring and monitoring regional competitiveness in the European Union. In Huggins, R. and Thompson P. (Eds.): Handbook of Regions and Competitiveness - Contemporary Theories and Perspectives on Economic Development. Edward Elgar Publishing.

Annoni, P., Dijkstra, L., and Gargano, N. (2017b), EU Regional Competitiveness Index: RCI 2016, Working papers 02/2017, Directorate General for Regional And Urban Policy, European Commission.

Combes, P.P, Duranton, G., Gobillon, L., Puga, D., and Roux, S. (2012), The productivity advantages of large cities: distinguishing agglomeration from firm selection, Econometrica, 80 (6), 2543-94.

Dijkstra, L., Annoni P., Kozovska, K. (2011), A new European Regional Competitiveness Index: theory, methods and findings, Working Papers 02/2011, Directorate General for Regional and Urban Policy, European Commission.

Dijkstra, L., Garcilazo, E., and McCann, P. (2015), The effects of the global financial crisis on European regions and cities, Journal of Economic Geography, 15(5): 935-49.

Eurofound (2016), ERM annual report 2016: Globalisation slowdown? Recent evidence of offshoring and reshoring in Europe, Publications Office of the European Union: Luxembourg.

European Union (2014a), Investments for jobs and growth, promoting development and good governance in EU regions and cities. Sixth report on economic, social and territorial cohesion, Publications Office of the European Union: Luxembourg.

European Union (2014b), EU-funded airport infrastructures: poor value for money, Report of the European Court of Auditors, Publications Office of the European Union: Luxembourg.

European Union (2017a), Reflection Paper on Harnessing Globalisation, Publications Office of the European Union: Luxembourg.

European Union (2017b), Reflection Paper on the Future of EU Finances, Publications Office of the European Union: Luxembourg.

European Union and UN-HABITAT (2016), The State of European Cities 2016. Cities leading the way to a better future, Publications Office of the European Union: Luxembourg.

Flyvbjerg, B. Bruzelius, N., and Rothengatter, W. (2003) Megaprojects and risks, Cambridge University Press: Cambridge.

Iammarino, S., Rodriguez-Pose, A., Storper, M. (2017) Why Regional Development Matters for Europe's Economic Future, Working Papers 07/2017, Directorate General for Regional and Urban Policy, European Commission.

Ketels, C. and Protsiv, S. (2016) European Cluster Panorama 2016. European Cluster Observatory.

Lavalle, C., Pontarollo, N., Batista E Silva, F., Baranzelli, C., Jacobs, C., Kavalov, B., Kompil, M., Perpiña Castillo, C., Vizcaino, C., Ribeiro Barranco, R., Vandecasteele,I, Pinto Nunes Nogueira Diogo, V., Aurambout, J., Serpieri, C., Marín Herrera, M., Rosina, K., Ronchi, S., and Auteri, D. (2017) European Territorial Trends for Cities and Regions Ed. 2017, Publication Office of the European Union, Luxembourg, JRC107391.

Melitz, M.J and Ottaviano, G.I.P. (2008), Market Size, Trade, and Productivity, Review of Economic Studies, 75(1): 295-316.

Moretti, E. (2010) Local Multipliers, The American Economic Review, 100 (2), 373-77.

Moretti, E., and Thulin, P. (2013) Local Multipliers and Human Capital in the United States and Sweden, Industrial Corporate Change, 22 (1), 339-62.

Morrar, (2014), Innovation in Services: A Literature Review, Technology Innovation Management Review, 4(4): 6-14.

OECD (2014) Innovation and Modernising the Rural Economy, Paris: OECD Publishing.

OECD (2015) The Metropolitan century: Understanding Urbanisation and its Consequences. Paris: OECD Publishing.

OECD (2016) Automation and independent work in a digital economy, OECD Policy Brief on the Future of Work, May 2016.

OECD (2016b) Regional Outlook: Productive Regions for Inclusive Society. Paris: OECD Publishing.

Poelman, H. and Ackermans, L. (2016) From rail timetables to regional and urban indicators on rail passenger services, Working papers 02/2016, Directorate General for Regional and Urban Policy, European Commission.

Puga, D. (2010)The magnitude and causes of agglomeration economies, Journal of Regional Science, 50: 203–219

Shepherd, B. (2013), Global Value Chains and Developing Country Employment: A Literature Review, OECD Trade Policy Papers, No. 156, OECD Publishing, Paris.

(1)

   Broadband Coverage in Europe 2016, available at:    
https://ec.europa.eu/digital-single-market/en/connectivityy .

(2)

   Next Generation Access Networks are defined as wired access networks which consist wholly or partly of optical elements and which are capable of delivering broadband access services with enhanced features, (such as higher throughput) as compared with those provided over existing copper networks.

(3)

   Table A.1.1 of the Appendix in Annoni et al. (2017b) lists the NUTS 2 regions comprising London and its commuting areas.

(4)

   It is important to note that, due to the margins of error in the set of indicators included in the index, the difference between some of the scores may not be statistically significant.

(5)

   Comparing the RCI over time is complicated because each edition of the index incorporates improvements and slight modifications. These do not affect the overall structure of the index, but they limit the possibilities of measuring change over time. The reasons for the modifications are various: new indicators become available at the regional level, while others are not updated or no longer fit the statistical framework of the index. In addition, methodological improvements, especially between the first and the second editions, and changes in the definition of NUTS regions complicate the exercise. Nevertheless, there remains a fair degree of continuity in the indicator list – changes between 2013 and 2016 are listed in Table A.3.1 in the Appendix in Annoni et al. (2017b).

EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

Chapter 2: Social cohesion

Contents

2.    Social cohesion    

2.1.    Population change is increasingly determined by migration    

2.1.1.    Two thirds of the EU-13 population live in a region of population decline    

2.2.    Employment rates are higher for those born in another EU country than for the native-born    

2.3.    Asylum seekers and refugees    

2.4.    The employment rate surpassed its pre-crisis level, but unemployment rates are still too high    

2.5.    Education and training    

2.6.    Adult proficiency in literacy and numeracy needs to be raised in several EU Member States    

2.7.    Poverty and social exclusion is declining in the EU-13 but growing in cities in the EU-15    

2.8.    Moving at different speeds to the Europe 2020 targets    

2.9.    More women are studying, working and being elected to regional assemblies    

2.10.    Life in the EU is among the longest in the world but regional disparities persist    

2.11.    Measuring social progress at the regional level    

Figure 2.1 Natural change and net-migration in EU-28, 1960-2015    

Figure 2.2: Population born outside the EU, 2011 and 2016    

Figure 2.3: Population born in another EU country, 2011 and 2016    

Figure 2.5: Employment rate gap between non-EU born men and women (aged 20–64,), 2016    

Figure 2.6 First-time asylum applications in the EU-28 by gender and age, 2016    

Figure 2.7: Literacy proficiency of adults (aged 16-64), 2016    

Figure 2.8 Numeracy proficiency of adults (aged 16-64), 2016    

Figure 2.9 The proportion of people at risk of poverty or social exclusion by degree of urbanisation, 2015    

Figure 2.10 Change in the proportion of people at risk of poverty or social exclusion by degree of urbanisation, 2008-2015    

Figure 2.11 The at-risk-of-poverty rate by degree of urbanisation, 2015    

Figure 2.12 Change in the at-risk-of-poverty rate by degree of urbanisation, 2008-2015    

Figure 2.13 Proportion of people living in Low work intensity households by degree of urbanisation, 2015    

Figure 2.14 Change in proportion of people living in low work intensity households by degree of urbanisation, 2008-2015    

Figure 2.15 Proportion of people with severe material deprivation by degree of urbanisation, 2015    

Figure 2.16 Change in the proportion of people with severe material deprivation by degree of urbanisation, 2008-2015    

Figure 2.18: Europe 2020 achievement index by degree of urbanisation, 2015    

Table 2.1 Population share in regions by determinants of population change, 2005-2015    

Table 2.2 Population change, natural change and net migration in capital metro, other metro and non-metro regions, 2005-2015    

Table 2.3 Population change, natural change and net migration in urban, intermediate and rural region, 2005-2015    

Table 2.4: Division of regions by median age in 2016 and direction of net migration, 2005-2015 (% of total)    

Table 2.5: Population age in capital metro, other metro and non-metro regions, 2016    

Table 2.6: Employment and unemployment by category of region    

Table 2.7 Youth unemployment, those not in employment, education or training (15-24) and participation in education and training (25-64), 2008-2016    

Table 2.8: EU 2020 regional achievement index, 2010-2015    

Map 2.1: Natural population change in NUTS 3 regions, 2005-2015    

Map 2.2: Net migration in NUTS 3 regions, 2005-2015    

Map 2.3: Total population change in NUTS3 regions, 2005-2015    

Map 2.4: Median age of the population in NUTS3 regions, 2016    

Map 2.5 Employment rate (20-64), 2016    

Map 2.6 Change in employment rate (20-64), 2008-2016    

Map 2.7 Unemployment rates, 2016    

Map 2.8 Change in unemployment, 2008-2016    

Map 2.9: Population aged 15-24 not in employment, education or training (NEET), 2016    

Map 2.10 Participation of adults aged 25-64 in education and training, 2016    

Map 2.11 Early school leavers aged 18-24, average 2014-2016    

Map 112 Low achievers in mathematics, reading and science    

Map 2.13 The EU-2020 achievement index, 2015    

Map 2.14 Change in the EU-2020 achievement index, 2010-2015    

Map 2.15 Difference between female and male employment rates (20-64), 2016    

Map 2.16 Difference between female and male employment rates, 2016    

Map 2.17 Gender balance of population 30-34 with a tertiary education, average 2014-2016    

Map 2.18 Gender gap for early school leavers, average 2014-2016    

Map 2.19 Women in regional assemblies, 2017    

Map 2.20 Change in the share of women in regional assemblies, 2007-2017    

Map 221 EU Life expectancy at birth, 2015    

Map 222 Infant mortality, 2015    

Map 2.23 Road fatalities, 2015    

Map 2.24: Road traffic fatalities in cities    

Map 2.25: The EU-SPI 2016 (0=lowest level of social development; 100=highest level of social development)    

Map 2.26 EU Social Progress index, three sub-indices    

Summary

·In 2016, the employment rate of those aged 20-64 reached 71% which is above the pre-crisis level but still well below the 75% target set by the Europe 2020 strategy. The situation varies markedly across the EU. In Spain, the rate was still 5 percentage points below the 2008 level, in Cyprus 8 percentage points lower and in Greece, as much as 10 percentage points less.

·Unemployment in the EU has fallen from a high of 10.9% in 2013 to 8.6% in 2016, still above the 7% it was in 2008. In the Czech Republic, Germany, Hungary, Malta, Poland and the UK, the rate is lower than in 2008, in Greece, Spain, Italy and Cyprus, at least five percentage points higher. Youth unemployment followed a similar pattern and remains above 40% in Greece and Spain. Regional disparities in unemployment rates have not narrowed as yet, but they have largely ceased to widen.

·The risk of poverty or social exclusion in the EU has fallen back to its pre-crisis level, but it remains higher in EU-15 cities, while it is significantly lower in EU-15 rural areas, as it is in all types of areas in the EU-13.

·Big differences in unemployment and income between regions encourage people to move to find better job opportunities and/or to escape poverty. In several regions, this has led to large reductions or increases in population, putting pressure on public infrastructure and services. A major task of regional development strategies is to tackle the factors pushing people to move.

·The EU has recently seen a big increase in asylum-seekers, reaching 1.2 million in 2015 and in 2016. Although this represents only 0.5% of working-age population, their distribution across the EU is far from even. The effective integration of the people concerned is important for cohesion and future prosperity.

·

2.    Social cohesion

2.1    Population change is increasingly determined by migration

As natural population growth in the EU slowed down in the early 1990s, migration overtook it as the main source of overall population growth ( Figure 2.1 ). In the 1960s, natural growth added more than 3 million people a year to the EU-28 population, in the 2000s, it added only 350 000. In 2015, for the first time, there was a natural reduction in the EU population. The impact of migration in the 1960s was small, adding only about 100 000 a year, while in the 2000s, it added over a million a year on average. In 2015, migration increased population in the EU by 1.8 million, a figure which does not include all the asylum seekers who arrived during the year as they are typically included in the population figures only after 12 months of residence or after being granted international protection.

Figure 2.1 Natural change and net-migration in EU-28, 1960-2015

In 2016, 10.7% of the EU population were born abroad, either outside the EU or in another EU country, an increase of 0.7 of a percentage point compared to 2011. Two-thirds of the people concerned were born outside the EU, the number of whom rose from 6.3% of total population in 2011 to 6.9% in 2016 ( F igure 2.2 ). The increase was 2 percentage points or more in Luxembourg, Finland and Sweden. In contrast, the share declined by over 2 percentage points in Cyprus and Slovenia and by around 1 percentage point in the Baltic States, because of outward migration among the people concerned and/or because they passed away.

The share of people born in another EU Member State (other-EU-born) barely increased between 2011 and 2016 (from 3.7% to 3.8%), though it rose by over 2 percentage points in Luxembourg and Slovenia ( Figure 2.3 ). The only countries where it declined were the Czech Republic (by 2 percentage points), Germany (1 percentage point) and Ireland (0.5 of a percentage point).

Figure 2.2: Population born outside the EU, 2011 and 2016

Figure 2.3: Population born in another EU country, 2011 and 2016

Two thirds of the EU-13 population live in a region of population decline

In the EU, 43% of the population live in a NUTS 3 region that lost population due to a natural reduction between 2005 and 2015. In the EU-13, the share was much larger (66%). The largest reduction occurred over the period in Eastern Bulgaria (a decline of more than 10%) ( Map 2.1 ). In many countries, rural and intermediate regions experienced fewer births than deaths. This was particularly so in Romania, Hungary, the Baltic States and Germany, where there was a natural reduction in population in almost all regions except metropolitan ones. The same was true in large parts of Portugal, Spain, France, Poland and the UK.

A smaller proportion of population in the EU, 31%, live in a region that lost population due to net outward migration, more people leaving the region than people entering the region, between 2005 and 2015. In the EU-13, however, the figure was much higher, 66%, as compared with only 22% in the EU-15 ( Table 2.1 and Map 2.1 ). Lithuania, Latvia and some Romanian regions have been particularly affected. Metropolitan regions in these countries were the only ones with net inward migration, more people entering than leaving the region over this period, although in some cases with a shift of population from the city centre to the surrounding region.

The highest growth in total population (7.7% on average) occurred in regions where there was both a natural increase in population and net inward migration ( Map 2.3 ). Almost half the EU-15 population live in such regions, but only 19% of the EU-13 population. The biggest reductions (7.2% on average) occurred in regions where there was both a natural population decline and net outward migration. Only 11% of the EU-15 population live in such a region as against 49% of the EU-13 population.

Because net inward migration in many cases fully offset a natural reduction in population, only 31% of the EU population live in a region that experienced a net loss of population over these ten years. The regions with a declining population are mostly located in EU-13 Member States. Reductions were particularly large in the Baltic States, Bulgaria, Hungary, Romania, Slovakia and Croatia, except in capital city or neighbouring regions or in those with major cities. Regions in Germany, especially in the eastern part, and Portugal also experienced substantial reductions in population.

Table 2.1 Population share in NUTS 3 regions by determinants of population change, 2005-2015, in %

Migration from outside the EU and mobility 1 between and within EU Member States is affected by differences in living conditions, unemployment and wage levels as well as the extent of discrimination (ESPON 2017).

Table 2.2 Population change, natural change and net migration in capital metro, other metro and non-metro regions, 2005-2015

Capital metropolitan (metro) regions have experienced the highest population growth, especially in the EU15 Member States, where population increased by 8% between 2005 and 2015, mainly driven by a natural increase in population (5%) ( Table 2.2 ). In the EU-15, population also increased in other regions (by 4%), mostly driven by net inward migration (which added 3% to the total). In the EU-13, population increased in capital metro regions as well (by 5%), entirely as a result of net inward migration, but both the other metro and non-metro regions lost population, mainly due to net outward migration.

Table 2.3 Population change, natural change and net migration in urban, intermediate and rural region, 2005-2015

Rural regions tend to have slower population growth than urban ones, but faster growth than intermediate regions in both the EU-13 and the EU-15. In the EU-13, intermediate regions have the highest net outward migration rate, in the EU-15, the lowest net inward migration rate. As a result, in the EU-28, population in intermediate regions remained unchanged, while it increased by 6%in urban regions and by 1% in rural ones.

Comparison between net inward migration ( Map 2.2 ) and the median age of population ( Map 2.4 ) indicates that younger people are more mobile than older ones. In the regions with net outward migration, the average age of the population living in the region tends to be higher and vice-versa. At the NUTS 3 level, 77% of regions with net inward migration over recent years are also those with the youngest populations, with a median age of less than 40 ( Table 2.4 ), while 68% of regions with a median age above 50 experienced net outward migration. Regions of net outward migration in Portugal, central France, southern Italy, Greece, Bulgaria, Hungary, southern Romania, eastern Germany, Finland and the Baltic States tend, for the most part, to have an older than average population. On the other hand, regions of net inward migration in southern Spain, northern France, London and surrounding areas, north-eastern Scotland and southern Sweden and Finland have a younger than average population, in many cases, migrants being attracted by dynamic urban centres. Accordingly, net outward migration tends to push up the median age of population, since it is disproportionately younger people who move, which also tends to reduce the birth rate so reinforcing the effect on the median age.

Table 2.4: Division of NUTS 3 regions by median age in 2016 and direction of net migration, 2005-2015 (% of total)

The largest shares of young people are in the capital metro regions in the EU-15 – almost 23% of the population was below 20 in 2016 - while those of 65 and older accounted for only 16% ( Table 2.5 ). Many young people come to the capital to study or to find a job. The elderly, who are mostly retired, do not need to be close to employment opportunities and often opt for a more peaceful and a lower cost location outside the capital.

The tendency is the same, even if less pronounced, in other metropolitan regions. In the EU-15, there are about the same number of elderly as young people (21% of both in 2016). Those below 20 are more numerous than those of 65 and older in all three types of region in the EU-13.

Table 2.5: Population age in capital metro, other metro and non-metro regions, 2016

2.2.    Employment rates are higher for those born in another EU country than for the native-born 

People born in the EU have the right to live and work wherever they choose in the Union, enabling them to gain work experience in other Member States for short periods as well as to move there on a long-term basis. In the EU as a whole, the employment rate of people aged 15-64 born in a different EU country averaged 70% in 2016, slightly higher than that of the native-born (67%) and substantially higher than that of people born outside the EU (59%) (Figure 2.4). In Portugal, Hungary, Luxembourg, Latvia, Croatia and the UK, the employment rate of other EU-born was markedly higher than that of the native-born.

Figure 2.4: Employment rate by country of birth (15-64), 2016

People born outside the EU, on the other hand, face multiple challenges to find a job. In most Member States, for which there are reasonably reliable data, the employment rate of non-EU born was lower than that of either the native-born or other EU-born, including in countries with a large share of non-EU born such as Sweden, Belgium, the Netherlands and France. Speaking the local language, having the right qualifications and having them recognised are only some of the difficulties the people concerned face in finding a job.

In most EU countries, the rate of employment of the native-born is higher than that of those born outside the EU, regardless of education level, whether basic, upper secondary or tertiary 2 . In some countries (Cyprus, Czech Republic, Spain, Greece, Italy, Luxembourg, Malta, Slovenia) the rate of employment of non-EU born, according to data for 2016, is higher than that of the native-born, but only for those with basic education.

Gender also plays a role. Employment rates of men are higher than for women in all countries, irrespective of the country of birth, but especially so for the non-EU born ( Figure 2.5 ). In Belgium, Greece, the UK, Bulgaria, Poland Italy and Malta, the difference for the latter was over 20 percentage points in 2016, reflecting in part cultural norms, lack of opportunity and inadequate wages in respect of the women concerned.

Figure 2.5: Employment rate gap between non-EU born men and women (aged 20–64,), 2016

2.3    Asylum seekers and refugees

In 2015, EU Member States received 1.2 million first-time applications for international protection and the same number again in 2016. As a share of the current non-EU born population, the yearly inflow in 2015 and 2016 together amounted to 7% at EU level (18% in both Germany and Finland, 16% in Sweden) and 0.5% in terms of total population (1.8% in Sweden and 1.5% in Austria). If confined to the number of positive first instance asylum decisions, it was only around 0.1% of the population (being highest in Sweden and Germany at 0.7% and 0.5%, respectively). 3 The increase in asylum seekers brought with it an increased flow of the most vulnerable group seeking asylum, namely unaccompanied minors, 4 whose numbers in the EU almost doubled between 2013 and 2014 (from 13 000 to 23 000) and almost quadrupled in the following year (92 205 in 2015, 59% of whom were hosted in Sweden and Germany). Although it declined in 2016, it was still at a relatively high level (63 280). By their nature, those concerned require additional protection and integration assistance in order for the most viable and sustainable solutions to be found which are in their best interest.

The distribution of asylum seekers across the EU is highly uneven. Germany, in particular, received more first-time asylum applications than all other EU countries combined in 2016. Not all these have been, or will be, granted refugee status and not all want to stay. Accordingly, at this stage, it is too early to say how many will remain in the EU.

Figure 2.6 First-time asylum applications in the EU-28 by gender and age, 2016

Recent asylum seekers are predominantly young and male, a disproportionate number being men aged 18-34 ( Figure 2.6 ).

The rapid influx represented a challenge for the local authorities to provide asylum seekers with food and shelter in the areas where they arrive. Integrating them into EU society will require language training, education and help in finding a job or setting up a business. The evidence from an ad hoc LFS survey in 2014 is that refugees face considerable problems in integrating into the labour market, as reflected in their significantly lower employment rates than other non-EU born residents and the EU -born population in most Member States (ESDE 2016). Low participation rates among women, a large proportion of people without upper secondary education and low levels of proficiency in the local language underlie this tendency (ESDE 2016, Dumont et al 2016). While the chances of refugees and others born outside the EU being employed increases significantly with their education level, the increase is smaller than for the native born or other-EU born (ESDE 2015, 2016).

2.4.    The employment rate surpassed its pre-crisis level, but unemployment rates are still too high

In 2016, the EU employment rate for those aged 20- 64 ( Map 2.5 ) exceeded the pre-crisis level for the first time. At 71%, it is higher than the previous high in 2008 of 70%, though only slightly. The rate has not recovered, however, in all parts of the EU. In Greece, it is still 10 percentage points lower than before the crisis, in Cyprus 8 points lower and in Spain 5 ( Map 2.6 ). On the other hand, it was 10 percentage points higher in Hungary and Malta.

Only 6 Member States (Sweden, Germany, Denmark, UK, Estonia and Netherlands) had an employment rate in 2016 above the Europe 2020 target of 75%. In more than half of Member States it was below 70% and in Greece, Spain, Croatia, France and Italy, below 65%. The impact of the crisis on employment rates has made it unlikely that the target will be reached by 2020.

The rate, however, varies markedly between types of region. The average employment rate in more developed regions 5 was 74.2% in 2016, quite close to the 75% target ( Table 2.6 ). In the less developed regions the average rate was well below the target, at only 65%. While it increased slightly in these regions between 2008 and 2016, in the transition regions, it has not increased at all. The increases in employment rates in regions where rates are low at least means that after several years of divergence, regional disparities in employment have started to narrow again.

Table 2.6: Employment and unemployment by category of region

Between 2008 and 2016, unemployment increased at the same time as employment rates went up, which means that the rate of job creation lagged behind the rise in the labour force. Although the unemployment rate fell from a high of 10.9% in 2013 to 8.6% in 2016 ( Map 2.7 ), this was still higher than in 2008 (7%). While in some northern and eastern parts of the EU, rates were lower than before the crisis, in the southern Member States, rates were up to 10 percentage points higher ( Map 2.8 ). In several regions in Greece, Italy and Spain and in the French outermost regions, rates were still over 20%.

The youth (15-24) unemployment rate declined from a high of 23.7% in 2013 to 18.7% in 2016, but it remains well above the level before the crisis of 15.9% in 2008 ( Table 2.7 ). The rate in 2016 was particularly high in the less developed regions (averaging 24%) but it was even more so in the transition ones (27%). The share of young people neither in employment nor in education or training (the NEET rate) has also declined, in this case from a high of 13.2% in 2012 to 11.5% in 2016, only slightly above the 2008 level (10.9%). The NEET rate was also highest in the less developed and transition regions.

Table 2.7 Youth unemployment, those not in employment, education or training (15-24) and participation in education and training (25-64), 2008-2016

Map 2.9: Population aged 15-24 not in employment, education or training (NEET), 2016

2.5.    Education and training

In a fast-changing, technology-driven world, people need to have access to opportunities continuously to update and improve their skills as well as to acquire new ones. This is vital not only to enable them to remain in employment and advance in their careers but also to boost productivity and the competitiveness of the economy.

To this end, EU Member States set a target in 2010 that by 2020, 15% of those aged 25-64 should be taking part in continuing training as compared with only just over 9% at the time. Progress towards this target, however, has been slow. By 2016, the figure had risen to only just under 11%. The target had been reached or exceeded in only 7 Member States and there were pronounced disparities not only between but also within countries, especially in Italy, France, the Czech Republic, Hungary and Germany ( Map 2.10 ).

Stronger efforts are needed to encourage low-qualified adults in particular to participate in training since there is a larger proportion of people with only basic schooling in the EU than in other industrialised economies. Because such people are the least likely to participate in training, engaging them is particularly challenging. The New Skills Agenda for Europe includes recommendations to tackle this issue (see Box).

One of the Europe 2020 targets is to reduce the share of early school leavers to 10% or less. At the EU level, the share of those aged 18-24 with no qualification beyond basic schooling and no longer in education or training in the 2014-2016 period was 11%, close to the target, but with wide differences between and within countries ( Map 2.11 ). In Spain, Portugal, Italy, Bulgaria and Romania, for example, the share in almost all regions is far from the target, whereas in Belgium, Germany, the UK and Greece, there is a large variation between regions, with some close to the target or below and others far away. In the Bruxelles-Capital region, for instance, 15% of 18 to 24 year-olds were early school-leavers against a country average of just below 10%.

High rates of early school leaving may be linked to pockets of socio-economic deprivation, often with high concentrations of migrants, where schools are of low quality and are less capable of retaining students. This is particularly the case in larger cities.

Education and continuing training have recently been confirmed to be among the main drivers of economic growth, a larger proportion of poorly educated people being more detrimental to growth than a smaller proportion of highly educated ones.

The results of the 2016 PISA (the OECD Programme for International Student Assessment) survey of 15 year-olds shows, in line with previous surveys, that competence in maths is particularly problematic in the EU, with over 22% of those tested having a low proficiency ( Map 2.12 ). Around 20% of those tested in the EU also had insufficient understanding of what they read and a low proficiency in science. The largest proportions with low proficiency (over 35% in all three disciplines) were in Bulgaria, Romania and Cyprus, while at the other end of the scale, Finland, Estonia and Ireland had reached the Europe 2020 target of no more than 15% of low achievers in the three disciplines, and Denmark and Slovenia were close to it.

(1) Article 45 of the Treaty on European Union and the Treaty on the Functioning of the European Union

(2) Data come from the Labour Force Survey. ‘Basic' is lower secondary education or less (i.e. ISCED levels 0, 1 and 2); 'upper secondary' includes upper secondary and post-secondary, pre-tertiary (i.e. ISCED levels 3 and 4) and 'tertiary' is university and equivalent (i.e. ISCED levels 5-8).

(3) Hungary has seen a large inflow (2.1% of its total population of 2015 and 2016) but mostly as a transit country, as the ratio of asylum decisions to applications was only 2%, indicating many people absconding and highlighting the need to consider asylum decisions as well when measuring asylum seeker inflow.

(4) Unaccompanied minors are generally defined as those under the age of 18 who arrive without parents, other adult relatives or guardians (UNHCR).

(5) See Lexicon for the definition of less developed, transition and more developed regions.

(6) Council Resolution of 27 November 2009 on a renewed framework for European cooperation in the youth field (2010-2018), OJ C 311, 19.12.2009, p. 1.

(7) European Commission 'The New Skills Agenda for Europe', COM(2016) 381 of 10.6.2016

(8) OJ C 484, 24.12.2016, p.1.

EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

Map 1.12 Low achievers in mathematics, reading and science

Vocational education and training (VET) can improve job-specific and transversal skills, facilitating the transition to employment and maintaining and updating the skills of the work force. Over 13 million people enrol in initial VET programmes every year in the EU. Yet labour market forecasts indicate an upcoming shortage of people with VET qualifications in a number of Member States. Those with recent VET qualifications at upper secondary level generally have a smoother transition from education to the labour market and higher employment rates than those with upper secondary qualifications from general education pathways who do not go on to complete tertiary education 1 .

The evidence suggests that VET programmes lead to better employment outcomes than non-tertiary general oriented ones. In 2015, those who had recently completed initial VET had an average employment rate of 73% in the EU, as against one of 61% for those who had recently completed upper-secondary general education and had not gone on to tertiary education. The biggest difference was in Belgium, Germany, Estonia, and Cyprus. Only in 6 countries (the Czech Republic, Ireland, France, Malta, Finland and UK) 2 was the average employment rate of those with VET qualifications similar or lower than those completing general upper-secondary programmes.

Despite this, for many young people and their parents, initial VET is not seen as an attractive option, suggesting perhaps a need to improve the labour market relevance of VET programmes. Too few programmes at present fully exploit the potential of work-based training or provide opportunities to progress to tertiary education. As a response, Member States agreed in 2015 3 to further strengthen key competences in VET curricula and provide more effective opportunities to acquire or develop these skills.

2.6.Adult proficiency in literacy and numeracy needs to be raised in several EU Member States

The ability to read and understand both literary and numerical information is essential for full participation in society and the economy. Without adequate skills of these kinds, people are likely to remain at the margins of society and to face significant barriers in finding a decent job.

In practice, in most Member States, substantial numbers of people have low levels of proficiency in reading and maths, as indicated by the Survey of Adult Skills (PIAAC) (carried out by the OECD with support from the European Commission), which assesses the ability of people aged 16 and over in these respects ( Figure 2.1 and Figure   2.2 ). According to the survey, the highest levels of literacy and numeracy are in Finland, the Netherlands and Sweden together with Japan. By contrast, levels are relatively low in Spain, Greece and Italy. The survey shows, moreover, that high levels of inequality in literacy and numeracy are related to inequality in the distribution of income.

Figure 2.1: Literacy proficiency of adults (aged 16-64), 2016

Figure 2.2 Numeracy proficiency of adults (aged 16-64), 2016

2.7.Poverty and social exclusion is declining in the EU-13 but growing in cities in the EU-15 

Clear signs of a general improvement in the social situation in the EU are emerging, though divergences among Member States remain. In 2015, almost a quarter (23.7%) of people in the EU were recorded as being at risk of poverty or social exclusion, the poverty indicator targeted by Europe 2020 (see Box). The proportion increased during the crisis between 2008 and 2012 but then fell back to the 2008 level. This reduction, which was common to most Member States, followed increases in incomes as a result of the recovery in economic activity, improvements in labour markets and reductions in those affected by severe material deprivation and those living in low work intensity households (two of the components of the indicator). The proportion at risk of poverty, on the other hand was 1 percentage point higher in 2015 than in 2008 5 .

Despite positive signs, the risk of poverty or social exclusion remains a key challenge especially in the Baltic and southern Member States. The risk remains high despite improvements in Bulgaria, Croatia, Latvia, Lithuania, Romania and Greece, and it has been rising in Cyprus and Italy. Together with an increase in inequality in many Member States, it is one of the main challenges to social cohesion.

In the EU-13, the proportion of people at risk of poverty or social exclusion is considerably larger in rural areas (34%) than in cities (20%) ( Figure 2.3 ). In the EU-15, the pattern is the opposite, the proportion being larger in cities (24%) than in rural areas (21%), though the difference is much smaller. Between 2008 and 2015, the proportion fell in all areas in the EU-13, the difference between cities and rural areas narrowing. In the EU-15, the proportion fell only in rural areas while it increased in cities, towns and suburbs ( Figure 2.4 ).

Figure 2.3 The proportion of people at risk of poverty or social exclusion by degree of urbanisation, 2015

Figure 2.4 Change in the proportion of people at risk of poverty or social exclusion by degree of urbanisation, 2008-2015

There is some difference in the incidence of the three indicators combined in the aggregate measure across the EU, though there are also similarities since each of them is measuring an aspect of poverty or social exclusion. In 2015, 17.3% of the EU population was recorded as being at risk of poverty ( Figure 2.5 ). As in the case of the aggregate indicator, there was a somewhat larger proportion of households at risk in rural areas across the EU (19.8%) than in cities (16.7%) or towns and suburbs (16.0%). At the same time, rural areas have a smaller proportion of households with very low work intensity, which suggests that their higher risk of poverty is not mainly due to their lower employment but to their lower incomes, or perhaps to their incomes needing to support larger households. The difference in the risk of poverty between cities and rural areas at EU level is due to the big difference in the EU-13 (26% as against 11%), while in the EU-15, the proportion at risk is slightly smaller in rural areas than in cities. Moreover, the proportion fell between 2008 and 2015, in rural areas solely in the EU-15 ( Figure 2.6 ).

Figure 2.5 The at-risk-of-poverty rate by degree of urbanisation, 2015

Figure 2.6 Change in the at-risk-of-poverty rate by degree of urbanisation, 2008-2015

Figure 2.7 Proportion of people living in Low work intensity households by degree of urbanisation, 2015

Figure 2.8 Change in proportion of people living in low work intensity households by degree of urbanisation, 2008-2015

In line with the pattern of change in unemployment, the proportion of people living in households with very low work intensity in the EU in 2015 was higher than in 2008 (10.6% as against 9.2%) but lower than the peak in 2014 (which in fact relates to 2013). In contrast to the risk of poverty, the proportion was much higher in the EU-15 than in the EU-13, especially in cities (18%), whereas in the EU-13, it was higher in rural areas (6%) than in cities (4%) ( Figure 2.7 ). The situation in the EU-15 may seem surprising as employment opportunities tend to be greater in cities. But it is also the case that a larger proportion of people live alone than in other areas and if they become unemployed, household work intensity immediately falls to zero, whereas in households with two or more people, the other person(s) in the household may continue to be employed. It is also the case that the proportion of non-EU born in EU-15 cities is four times that in rural areas, which because of their lower employment rates also tends to increase the number of households with low work intensity.

In addition, the crisis hit cities in the EU-15 harder than other areas, the proportion of people living in low work intensity households increasing by 8 percentage points as a result, whereas it remained unchanged in rural areas. In the EU-13, by contrast, the proportion declined by 3 percentage points in both rural areas and cities and by 2 percentage points in towns and suburbs ( Figure 2.8 ).

Figure 2.9 Proportion of people with severe material deprivation by degree of urbanisation, 2015

Figure 2.10 Change in the proportion of people with severe material deprivation by degree of urbanisation, 2008-2015

The severe material deprivation indicator identifies people who cannot afford any four of 9 basic items included in the EU-SILC. The proportion concerned in the EU-13 was more than twice that in the EU-15 in 2015 (14% as against 6%), reflecting the much lower income levels. In the EU-13, in the same way as the risk of poverty, it was larger in rural areas than cities (16% as against 12%, Figure 2.9 ), but the difference is narrowing. Between 2008 and 2015, the proportion fell by 9 percentage points in rural areas and 5 percentage points in cities ( Figure 2.10 ).

In the EU-15, severe material deprivation is more common in cities than rural areas (affecting 7.4% of the population in 2015 as against 4.4%) and has become more so over time (increasing by 1.3 percentage points while remaining unchanged in rural areas). Although many cities in the EU-15 have high levels of GDP per head, they also have, in many cases, high levels of inequality, as reflected in at-risk-of-poverty rates, higher concentrations of deprivation than other areas and more households with low work intensity.

2.8.Moving at different speeds to the Europe 2020 targets

The Europe 2020 strategy sets out five headline targets to be reached by 2020, covering employment, education, poverty, innovation and climate change. The targets for reducing greenhouse gas emissions and increasing renewable energy have been translated into legally-binding national targets. In the other cases, there are optional national targets.

Portugal, Spain, the south of Italy, Croatia, Greece, Bulgaria, Romania and eastern Hungary are furthest away from achieving the targets ( Map 2.1 ). Intra-country variation is, however, pronounced. Apart from the traditional north-south divide in Italy, in France, Germany, Belgium, the UK, the Czech Republic and Denmark there are both regions with high values of the index and those with low values.

Between 2010, when the targets were set, and 2015, almost all regions in central and eastern Member States made progress towards achieving them ( Map 2.2 ). The score on the index for the less developed regions increased on average from 36 to 46. The score for the transition regions, on the other hand, rose only marginally, reflecting the impact of the crisis. The score also increased for the more developed regions, from 76 to 80, but at this rate even these will not reach the targets by 2020 ( Table 2.1 ).

Constructing the Europe 2020 achievement index

Table 2.1: EU 2020 regional achievement index, 2010-2015

Source: Dijkstra and Athanasoglou, 2015.

In general, cities are closer to achieving the targets ( Figure 2.11 ) than towns and suburbs or rural areas. In Sweden, Czech Republic and Luxembourg, cities have reached or surpassed the employment, education and poverty reduction targets - indeed; some had already done so in 2010. The difference between cities and other areas is very wide in some cases, in Bulgaria, Romania, Spain, Hungary and Poland, in particular, in all of which rural areas are lagging well behind.

In some countries, especially in the EU-15, towns and suburbs score better than cities. In France, the UK, Austria, Malta and, in particular, Belgium, cities score poorly, primarily due to low employment and high poverty rates.

While progress was made towards the targets in almost all countries between 2010 and 2015, if by not enough to meet them by 2020, in Greece and Cyprus, the situation deteriorated in all three types of area ( Figure 2.1 ). The achievement index was also lower in 2015 than in 2010 in Danish and Belgian cities, in towns and suburbs in France and in rural areas in Spain.

Figure 2.11: Europe 2020 achievement index by degree of urbanisation, 2015

Figure 2.19: Change in the Europe 2020 achievement index between 2010 and 2015

2.9.More women are studying, working and being elected to regional assemblies

Equality between women and men has been enshrined in the EU Treaties from the very beginning and is part of the 2009 Charter of Fundamental Rights.

In 2016, the employment rate of men aged 20-64 in the EU was 12 percentage points higher than that of women ( Map 2.3 ). In 2001, the gap was 18 percentage points and has narrowed every year since then, including over the crisis years. Employment rates of men are higher than for women in all EU regions except Övre Norrland in Sweden and Corse in France. The difference is over 20 percentage points in Malta and several Greek, Italian and Romanian regions. In Malta, Greece and Italy, the difference narrowed between 2001 and 2016, but in Romania, it increased by 5 percentage points.

At the EU level, unemployment rates of men and women are much the same, the rate for women being only 0.4 of a percentage point higher than for men in 2016 ( Map 2.4 ). This implies that the employment gap is primarily due to more women not participating in the work force. The Commission's Strategic engagement for gender equality has identified a number of way of increasing employment rates of women:

•make it easier to balance caring and professional responsibilities;

•share time spent on care and household responsibilities more equally;

•provide childcare for 33% of children under 3 and 90% of children between 3 and mandatory school age (the targets set under the Barcelona agreement in 2002);

•provide support for care of other dependants;

•encourage more women to become entrepreneurs;

•promote gender equality in research;

•improve the integration of women migrants into the labour market.

More of the women aged 30-34 have tertiary education than men in the EU and this is the case in all regions, except in several German ones and a few others scattered across the EU ( Map 2.5 ). On average, 43% of women in this age group had this level of education in 2014-2016 as opposed to only 34% of men. In Latvia, northern Sweden, Slovenia, some Polish regions and Molise in Italy, the share of women with tertiary education was 20 percentage points or more larger than for men.

Figure 2.20 Difference between female and male graduates by field of tertiary education, 2015

While more women than men have tertiary education, their fields of study differ substantially, which may partly be a factor underlying their lower employment rates. In particular, far more men than women opt for a natural science, mathematics, ICT or engineering degree in all Member States ( Figure 2.20 ).

Women aged 18-24 are also less likely to have left education and training before completing upper secondary schooling than men. ( Map 2.6 ). There are many reasons why young people may decide to leave school early. Personal or family problems, learning difficulties, a fragile socio-economic situation are all potential reasons but the school environment, teacher-pupil relations and the quality of teaching may also play an important role. The highest rates of early school-leaving are in regions in Spain, Portugal and Italy, mostly because of young men leaving early. In Sardegna, for example, around 28% of young men left education before completing upper secondary education as against just under 15% of young women. While more men than women leaving education early is the norm across the EU, there are a few regions (around 10% of the total) scattered across northern, central and eastern parts of the EU (but in Bulgaria especially), where the reverse is true, though only marginally so in most cases.

(1)

   The indicator measures the employment rates of persons aged 20 to 34 having completed education 1-3 years before the survey with a diploma from upper secondary education (ISCED 3) or post-secondary non tertiary education (ISCED 4), and who are currently not enrolled in any further formal or non-formal education or training, out of the people in the same age group.

(2)

   European Commission, 2016, Education and Training Monitor 2016, https://ec.europa.eu/education/sites/education/files/monitor2016_en.pdf .

(3)

   'Riga Conclusions 2015 on a new set of medium-term deliverables in respect of VET for the period 2015-2020'.Conclusions of the Council of Ministers in charge of vocational education and training. Available at: http://www.izm.gov.lv/images/RigaConclusions_2015.pdf .

(4)

   European Commission 'Investing in Europe's youth' COM(2016) 940 of 7.12.2016.

(5)

   2015 and 2008 refer to the years of the survey. The income being measured actually relates to the previous years, i.e. 2014 and 2007.

(6)

   ‘Equivalised’ means that income is adjusted for differences in the size and composition of households.

(7)

   The 9 items are a colour TV, a washing machine, a telephone, a car, a meal of meat or fish or the equivalent every other day, a week’s annual holiday away from home, an ability to avoid being in arrears on mortgage payments, rent, utility bills, hire purchase instalments or loans, an ability to make ends meet and an ability to keep the house adequately warm.

(8)

   The Pillar was published as a Commission Recommendation and as a proposal for an inter-institutional Proclamation with the European Parliament and the Council.

EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

In 2017, women made up half or more of members of regional assemblies across the EU in only 17 out of 297 cases. Five regional assemblies in Hungary, Italy and Romania have no women members at all and in several regional assemblies in these three countries as well as in Slovakia, less than 10% of members were women. Women were most represented in assemblies in Belgium, Spain, France, Sweden and Finland, where they accounted for 40% or more of members ( Map 2.19 ).

The average regional assembly in the EU had only 29% of members that were women in 2017, only slightly more than in 2007 (27%), so that at this rate of progress, it would take 100 years to reach 50%. There is also no indication of a larger increase in countries with a small share of women members than in others ( Map 2.20 ).

In some countries, the share of women has increased without the need for a gender quota. In Sweden, for example, most political parties ensure that every second candidate for election is a woman. In Belgium, France, Spain, Portugal and Ireland, however, quotas have been used to raise the number of women at national and/or regional level of government (Ireland does not have any regional assemblies and Portugal has regional assemblies only in the Acores and Madeira).

2.10.Life in the EU is among the longest in the world but regional disparities persist 

The EU has one of highest life expectancies at birth in the world, 80.6 years in 2015. Spaniards and Italians have the longest expectancy in the EU (83.0 and 82.7 years at birth, respectively), while Lithuanians have the shortest (74.6 years). Most EU Member States have a life expectancy higher than in the United States, which is ranked only 31st in the world in this regard, with an expected life span of 79.3 years in 2015 (WHO 2017).

Differences between regions across the EU are, however, marked ( Map 2.21 ). Life expectancy at birth is below 75 in many parts of Bulgaria and Romania and the eastern regions of Hungary as well as in Latvia and Lithuania. In 20 NUTS 2 regions (mainly located in France, Italy and Spain but also including the wealthiest part of London - Inner London West - which includes Westminster), life expectancy is over 83. Regional disparities in infant mortality ( Map 2.22 ) and, to a lesser extent, road fatalities ( Map 2.23 ) can partly explain the differences.

In 2015, an average of 3.6 children per 1 000 born alive died before reaching one year of age in the EU, a reduction from 3.8 in 2012. Infant mortality, however, was above 6 per 1 000 in 21 NUTS 2 regions in Romania and Bulgaria – all except the capital city ones – all the French overseas regions, the Spanish regions of Ceuta and Melilla (on the North coast of Africa), the most eastern region in Slovakia and the English region of Shropshire and Staffordshire in the West Midlands. By contrast, the rate was 2 per 1 000 or less in 18 regions scattered across the EU – in two or more in Austria, Finland, the Czech Republic, Slovenia and Spain and one each in Belgium, Germany, Greece, Spain, the Netherlands and the UK .

Road traffic fatalities vary equally widely across the EU. Although they declined overall by 45% between 2004 and 2014, the number still averaged 51 per one million inhabitants in 2015, though with large differences between regions ( Map 2.23 ). (For comparison, the US figure was twice as high in 2015, at over 100 per million.) .

The regions with the highest figures, with over 99 deaths per million, are mostly in Bulgaria, Romania, Greece, Croatia and the north-eastern part of Poland though also in Portugal, Corse and, above all, the Belgian province of Luxembourg, where as many as 210 road fatalities per million inhabitants were recorded in 2015, 38% more than in 2010.

Road fatalities are, in many cases, less in capital city regions than in other parts of the country. The safest capital cities in the EU in which to drive are Stockholm and Wien, in both of which the number of road deaths was below 10 per million in 2015, while in London, Copenhagen, Paris, Madrid, Berlin and Prague, fatalities are less than in other regions ( Map 2.24 ). This reflects in part low traffic speeds and good public transport, which gives people the option of not driving.

Cities, however, do not have lower fatality rates than other areas everywhere. In Romania, Italy, Belgium, Lithuania, Latvia and Poland, rates are relatively high in cities. In Bucharest and many other Romanian cities, there were more than 90 deaths per million in 2013-2014, far above the target of 31 deaths per million set by the European Road Safety Action Programme for 2020. In 2015, this target was reached in only 16% of regions. Further efforts and more investment are, therefore, needed in most regions to improve road safety.

2.11.Measuring social progress at the regional level 

Social progress can be defined as a society’s capacity to meet the basic human needs of its citizens, to establish the basis for citizens and communities to improve and sustain their quality of life and to create the conditions for people to reach their full potential. This definition underlies the Global Social Progress Index which measures social progress at the national level in about 130 countries worldwide 1 . In an attempt to measure social progress at the regional level in the EU, the European Commission recently published the EU Regional Social Progress Index (EU-SPI) that builds on and adapts the Global Social Progress Index. The EU-SPI is based on a set of 50 social and environmental indicators, drawn primarily, though not only, from Eurostat data. The EU-SPI is aimed at providing consistent, comparable and policy-relevant measures of the social and environmental situation in all NUTS 2 regions 2 . It covers three dimensions of social progress - basic human needs; the foundations of well-being and opportunity - each of which is broken down into four underlying components ( Figure 2.21 ).

Economic indicators are deliberately excluded which means that the EU-SPI measures social progress rather than economic performance and can be compared with economic indicators.

Figure 2.21: The framework of the EU-SPI index

The index has been built to identify social and environmental strengths and weaknesses, to inform regional development strategies and to support peer learning between regions. It scores the various aspects covered on a scale from 0 to 100, where 0 represents the lowest possible level of social progress and 100 the highest. Results show that social progress in the EU is highest in Nordic and Dutch regions and lowest in Romanian and Bulgarian regions ( Map 2.25 ). Social progress is also moderately high in Austria, Germany, Luxembourg, Ireland and the UK. Belgium and France score well too, though both show large internal differences. The largest regional variation is in Italy where central regions score better than the rest of the country ( Figure 2.22 ).

According to the SPI, except for some regions in Member States which joined the EU in 2004 or after, basic human needs are being met in almost all regions ( Map 2.26 ). The ‘Foundations of well-being’ dimension shows greater variation with only the Nordic Member States, the Netherlands and Ireland scoring well in all regions ( Map 2.26 ). The largest differences relate to ‘Opportunity’, with low scores in many regions in the southern and central eastern countries ( Map 2.26 ).

Map 2.25: The EU-SPI 2016 (0=lowest level of social development; 100=highest level of social development)

Map 2.26 EU Social Progress index, three sub-indices

Figure 2.22: Degree of within-country variability of the EU-SPI

There is a close link between the EU-SPI and regional GDP per head, although the relationship indicates that at every level of economic performance there are opportunities for more social progress but also risks of less ( Figure 2.23 ). In low GDP per head regions, every extra euro of GDP tends to lead to more social progress, while for high GDP per head regions, this is much less true. Among the high GDP per head regions, some regions such as the Nordic regions and most of the Dutch regions score higher than would be expected given their GDP her head.

In a small number of regions, commuting across NUTS 2 boundaries has a distorting effect on GDP per head of some significance since commuters increase GDP without being counted in the population. This is the case in Brussels and London, in particular, where around half the people working there live elsewhere. In these regions, GDP per head is an especially poor proxy for income and this may partly explain why some score poorly relative to GDP per head. Many other issues, however, make GDP per head a poor proxy for median disposable household income, such as the variable share of GDP going to wages (which on average has been shrinking), the differing degree of inequality of earnings and the varying extents of redistribution through taxes and social benefits, both between people and between regions.

Figure 2.23: Relationship between EU-SPI and GDP per capita

References

ESPON (2017) The Geography of new Employment Dynamics in Europe. Delivery 2 – Interim Report

Dijkstra L. and Athanasoglou S. (2015) The Europe 2020 index: The progress of EU countries, regions and cities. Regional Focus Working Paper 01/2015.

Dumont J.-C., Liebig T., Peschner J., Tanay F., Xenogiani T. (2016) How are refugees faring on the labour market in Europe? OECD-EC Working Paper 1/2016.

European Commission (2016). 'Labour Market Integration of Refugees' in Employment and Social Developments in Europe (ESDE) 2016 – available at: http://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=7952&type=2&furtherPubs=yes

European Commission (2016). 'Labour Mobility and Migration in the EU' in Employment and Social Developments in Europe (ESDE) 2015 – available at: http://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=7859&furtherPubs=yes  

Huber P and Bock-Schappelwein J (2014) The Effects of Liberalizing Migration on Permanent Migrants' Education Structure. Journal of Common Market Studies. 52: 268-284.

(1)

   For more information on the Global Social Progress Index: https://www.socialprogressindex.com .

(2)

   For more information on the regional EU-SPI:     http://ec.europa.eu/regional_policy/en/information/maps/social_progress .

EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

CHAPTER 3 - TERRITORIAL COHESION

Table of Contents

3.1. Introduction    

3.2. Energy Union and climate change    

3.2.1. Increasing energy efficiency    

3.2.2. Reducing greenhouse gas emissions    

3.2.3. Increasing the share of renewable energy    

3.2.4. Climate change    

3.3. State of environment    

3.3.1. Water    

3.3.2. Waste    

3.3.3. Sustainable transport    

3.4. Sustainable cities    

3.4.1. Cities can be environment friendly    

3.4.2. Changes in land use per person    

3.4.3. Urban transport    

3.4.4. People living in cities suffer more from pollution    

3.4.5. Access to green spaces    35

3.4.6. River flooding    41

3.5. Cross-border cooperation and territorial dimension of cohesion policy    44

3.5.1. Border regions…………………………………………………………………..    44

3.5.2. Other types of cooperation: interregional, transnational and macro-regional    50

3.5.3. Local, urban and metropolitan development    53

Figure 1 Primary energy consumption, change 2005-2015 and distance to target    

Figure 2 Change in greenhouse gas emissions outside the Emissions Trading Scheme, 2005-2011 and Europe 2020 targets    

Figure 3 Share of renewable energy in gross final energy consumption, 2006, 2015, target 2020    

Figure 4 Share of waste landfilled in selected EU Member States, 2014    

Figure 5 Passenger travel by transport mode, 2014    

Figure 6 Freight transport by mode, 2014    

Figure 7 Distance to services by type of municipality in the EU    

Figure 8 Population living within 1km distance of different services, by city size and degree of urbanisation in the EU    

Figure 9 Access to public transport in large European cities, 2014-2016    

Figure 10 Access to public transport in mid-size European cities, 2014-2016    

Figure 11 People reporting that they live in an area with problems relating to pollution, grime or other environmental problems, by degree of urbanisation, 2015    

Figure 12 Proportion of people who are satisfied with the level of noise in their city, 2015    

Figure 13 Share of the Natura 2000 network which intersects with Functional Urban Areas    

Figure 14 Population of border areas having access to rail passenger services, 2014    

Table 1 Population of border areas having access to rail passenger services, 2014    

Map 1 Electricity generated from hard coal and lignite - Electricity generated from renewable energies    

Map 2 Negative climate change impacts under a 2°C warming scenario    

Map 3 Projected increase in multi-hazard exposure    

Map 4 Main water bodies with less than good ecological status or potential    

Map 5 Urban, wastewater not collected, 2012    

Map 6 Residential, industrial and commercial areas per inhabitant by city, 2012    

Map 7 Change in residential, industrial and commercial areas per inhabitant by city, 2006-2012    

Map 8 Concentrations of airborne particulate matter (PM10) in cities, 2014    

Map 9 Concentration of ground-level ozone (O3) in cities, 2014    

Map 10 Access to green urban areas in cities, 2012    

Map 11 Share of NO2 concentration removed by vegetation in cities, 2010    

Map 12 Population flooded in the case of the biggest 100 year flood in FUAs    

Map 13 Border regions, NUTS 3    

Map 14 Cross-border road network efficiency in border areas    

KEY MESSAGES

·Substantial progress has been made in the EU in limiting energy consumption and greenhouse emissions. Most Member States are close to reaching the targets set under the Europe 2020 strategy.

·Part of the progress, however, is explained by the slowdown in economic activity during the crisis and there is a risk that the current recovery will make it more difficult to maintain momentum towards meeting the targets.

·In transport, there needs to be a major shift towards using less energy, cleaner kind of modes and more efficient use of infrastructure to reach the EU objectives for greenhouse gas emissions.

·The impact of climate change is likely to be considerable for a number of EU regions, particularly the outermost regions, regions around the Mediterranean, along coastlines generally and mountainous ones.

·Adapting to changes caused by global warming is generally costly and further investment is needed to make EU regions and cities more resilient to the consequences of the changes concerned.

·Despite general progress in reducing environmental pressures (notably as regards waste water and waste treatment), more efforts are needed to meet key environmental objectives of the EU.

·Pollution is often more of a problem in cities than in other areas. Air pollution is a particular concern and nature-based solutions, such as the development of green urban spaces, can provide an efficient means of mitigating the problem.

·Cities can also be more efficient in the use of resources than other places and can make it possible to adopt a low-carbon life style.

·Cross-border cooperation, a major policy objective of the EU, has helped to mitigate the adverse effects of internal borders, Support for cooperation has led to improvements in cross-border security and concrete achievements in transport, education, energy, healthcare, training and other areas.

·National borders, however, still constitute obstacles to the movement of goods, services, people, capital and ideas and substantial gains to the regions concerned as well as to the EU as a whole could be obtained if the remaining restrictions were removed.

3.1. Introduction

As argued in the 5th Cohesion Report, territorial cohesion highlights various issues which are central to cohesion policy. Among these are the environmental dimension of sustainable development and the promotion of a flexible functional geographies approach to territorial development. The latter aims to adapt the geographical level of analysis and implementation of policy to the challenges to be addressed. Depending on the issue at stake, this ranges from macro regions, such as the Baltic Sea or the Danube region, to metropolitan and cross-border areas. This chapter therefore covers the major environmental challenges affecting the development of EU regions, on the one hand, and a number of major issues addressed by various territorial cooperation schemes, on the other.

Environmental challenges are increasing in number and importance. Global warming and the associated climate change is likely to have fundamental consequences for the EU economies and societies, notably with the increase in the frequency of extreme natural events that is expected to accompany the general rise in temperature. The extension of human settlements, built-up areas and industrial activities accentuates the pressure on the environment with effects notably in the form of air pollution, a deterioration in the quality of water bodies and the fragmentation of natural habitat, while the production of waste has reached levels which require a radical change in approach.

A large share of cohesion policy resources has always been invested in measures to improve the quality of the environment or to tackle key environmental challenges. The policy is geared towards supporting the shift to a low-carbon economy while at the same helping Member States and regions to improve their capacity to mitigate the negative impact of climate change.

Cohesion policy invests heavily in the installation of facilities to improve the quality of drinking water and to treat wastewater and in waste management and recycling schemes as well as in measures to increase energy efficiency. It also helps to develop ’green’ infrastructure across the EU and to establish a network of protected natural areas as part of Natura 2000, while supporting a shift towards more environmentally-friendly modes of transport, all with the objective of .ensuring a sustainable path of development throughout the EU.

For the 2014-2020 period, around EUR 78 billion of Cohesion Policy funding has been allocated to supporting the shift towards a low-carbon economy (thematic objective 4), adaption to climate change and risk prevention (thematic objective 5) and improving environmental protection and resource efficiency (thematic objective 6). This amounts to almost a third of ERDF and Cohesion Fund resources, the two sources of financing most concerned with environmental issues.

Territorial Cooperation is a key objective of cohesion policy, focusing on joint action and exchange of policy ideas and experience between national, regional and local authorities in different EU Member States. It helps to reduce the obstacles to development which stem from national borders and supports the adoption of common strategies to solve common problems. Around EUR 10 billion have been allocated to such cooperation for the 2014-2020 period.

3.2. Energy Union and climate change

The EU has the objective of making a transition to a low-carbon economy and of ensuring that Europe has access to secure, affordable and climate-friendly energy. The Energy Union is a European priority project in which five dimensions are closely interlinked: energy security, solidarity and trust; a fully integrated European energy market; energy efficiency to moderate demand; action on climate change to decarbonise the economy; and research, innovation and competitiveness.

As part of this, targets have been set for reducing greenhouse gas emissions progressively up to 2050. These are included in both the 2020 climate and energy package and the 2030 climate and energy framework.

The 2020 climate and energy package is aimed at achieving a 20% cut in greenhouse gas emissions, a 20% improvement in energy efficiency (both from 1990 levels) and a 20% share of renewables in final energy consumption. The 2030 climate and energy framework is more ambitious, increasing these targets to 40% for the first and to 27% for the other two 1 .

Cohesion policy plays a central role as regards the Energy Union. By helping Member States achieve EU climate and energy targets, cohesion policy investments tackle energy poverty and enhance energy security, while furthering regional development, competitiveness, growth and jobs. By supporting the Energy Union, the policy also contributes to reducing air pollution which, according to the WHO, is one of the main environmental hazards facing us.

For the 2014-2020 period, around 21% of the ERDF and Cohesion Fund resources are allocated to climate related interventions. While the ESF is by its nature less oriented towards this area, 1.4% of its resources still go towards combating the effects of climate change.

Cohesion policy supports a comprehensive range of climate-related measures, such as improving energy efficiency in public buildings, housing and small and medium-sized enterprises, smart grids; renewable energy sources; clean urban transport, railways, cycle tracks and footpaths; research on climate change and adaptation to it, including resilient infrastructure and risk prevention and management.

3.2.1. Increasing energy efficiency

Increasing energy efficiency is critical for reducing the energy dependence of the EU economies and protecting the environment. Energy efficiency can be improved at all stages of the energy chain, from generation to final consumption. EU measures focus on areas where the potential for savings is greatest, buildings, in particular. Increasing energy efficiency is one of the main objectives of the Energy Union and one of the primary targets of the Europe 2020 strategy. The aim is to lower EU primary energy consumption to less than 1483 million tonnes of oil-equivalent (Mtoe) a year and final energy consumption to less than 1086 Mtoe  2 .

Between 2005 and 2015, EU primary energy consumption fell by 11% from 1713 Mtoe in 2005 to 1530 Mtoe in 2015 3 ( Figure 3-1 ). Primary energy consumption fell in all Member States over this period, except Estonia and Poland where it increased (by 15% and 3%, respectively). Reductions were largest (20% or more) in Lithuania, Greece and Malta.

In 2015, primary energy consumption in the EU as a whole was still around 3% above the 2020 target. In Malta, France, Germany, the Netherland and Bulgaria, substantial reductions in energy consumption are still needed to meet the indicative national targets set in 2013. In 18 Member States, on the other hand, consumption was already below the targets 4 .

Figure 3-1 Primary energy consumption, change 2005-2015 and distance to target

Source: Eurostat

Final energy consumption in the EU fell by more than 9% between 2005 and 2015, from 1191 Mtoe to 1082 Mtoe, i.e. to slightly below the 2020 target. The largest reductions were in Greece (22%), Spain (18%) and Portugal (16%), countries in which GDP either declined over this period (Greece and Portugal) or increased relatively little. Final consumption increased only in Lithuania (by 4%), Poland (6%) and Malta (50%). Final consumption in 2015 was below the national 2020 targets in 16 Member States but still needed to be reduced further in the others, especially in Malta, Lithuania, Slovakia and Hungary.

Recent analysis shows that the reduction in energy consumption is a result not only of improvements in energy efficiency but also of structural changes in electricity generation and of the downturn in the economic activity from 2008 5 . The economic recovery now underway might, therefore, give rise to an upsurge in energy consumption across the EU if GDP growth were to be particularly high, so putting the achievement of targets at risk.

Heating and cooling in buildings and industry account for half of EU energy consumption. For the most part, the energy concerned is from fossil fuels and only 16% comes from renewables. A sharp reduction in both and in the use of fossil fuels would contribute greatly to meeting the EU's climate and energy goals. This would require significant investment which can be supported to a major extent by cohesion policy in the majority of Member States.

3.2.2. Reducing greenhouse gas emissions

The EU emissions trading system (ETS) is a major means of cutting greenhouse gas emissions from power and heat generation, industry and aviation, covering around 45% of such emissions in the EU. The 2020 target requires a reduction in emissions in the areas concerned of 21% on the 2005 level, while the target for 2030 requires a cut of 43%.

In the other, non-ETS sectors, namely housing, agriculture, waste and transport (excluding aviation), Member States have set binding targets for cutting emissions under the Effort Sharing Decision (ESD). These differ between countries according to their national wealth, varying from a 20% cut relative to the 2005 level for the wealthiest to a 20% increase for the least developed. To achieve the 2030 target of a 40% reduction in EU greenhouse gas emissions, the ESD areas would need a cut of 30% (relative to 2005). It is in these areas that Cohesion policy funding can help Member States to achieve their targets.

Some Member States have already reduced emissions markedly in ESD sectors ( Figure 3-2 ). Between 2005 and 2015, they were reduced by 22% in Portugal and 27% in Greece. In other countries emissions increased, notably in Lithuania (by 12%) and Malta (by 34%). Variations in economic growth explain part of these differences, but other factors are important as well. For example, emissions were reduced by almost 21% in Sweden yet GDP grew on average by 1.7% a year over the period.

Figure 3-2 Change in greenhouse gas emissions outside the Emissions Trading Scheme, 2005-2011 and Europe 2020 targets

In 18 Member States, the level of emissions in 2015 was lower than the target set under the ESD, most especially in Croatia, which committed to limiting the increase in emissions to 25% relative to the 2005 level but actually cut them by 16%. Some of the other countries have gone a long way to achieving the target and have only a little more to do. In particular, in the UK and Austria, emissions need to be reduced by less than 1%. In Ireland, on the other hand, they need to be reduced by almost 10%, while in Malta, emissions rose by much more than the increase agreed.

3.2.3. Increasing the share of renewable energy

The EU objective is to increase the share of renewables in energy consumption to 20% by 2020 (10% in the transport sector) and to 27% by 2030. Under the Renewable Energy Directive 6 , EU Member States have set binding targets for increasing their national shares by 2020, which vary from 10% in Malta to 49% in Sweden, reflecting differences in both the prevailing share and the potential for expanding it. In some Member States, therefore, the share is already large - almost 54% in Sweden in 2015 and 34% in Latvia - while it is well below 10% in Malta, Luxembourg and the UK ( Figure 3-3 ).

Figure 3-3 Share of renewable energy in gross final energy consumption, 2006, 2015, target 2020

In 2015, 11 Member States had already exceeded their targets and in another three, the share needed to be increased by less than 3 percentage points to meet them. In 10 countries, however, the required increase was more than this and in four of them – the UK, Ireland, France and the Netherlands - 7 percentage points or more.

The potential of countries or regions for producing renewable energy depends on their geo-physical characteristics. For instance, coastal regions generally have a high potential for producing wind energy, especially those along the shores of the North and Baltic Seas and some Mediterranean islands. The potential for solar energy production is obviously higher where there are large amounts of sunshine, while the production of hydroelectricity also requires suitable geo-physical features. Realising whatever potential exists, however, depends on the policies implemented.

Accordingly, the production of renewable energy varies markedly from one region to another. This is well illustrated by electricity production. In some regions, electricity generation is still largely dependent on coal and lignite. This is particularly the case in most regions in Poland but also in Germany, the UK, Italy, Ireland, Spain, Romania and Croatia ( Map 3-1 ). In contrast, in other regions electricity is principally produced from renewables, notably in Cyprus, Greece, Austrian, Sweden, Finland and France, hydroelectricity, biogas, biomass and wind energy being the main sources 7 .

Map 3-1 Electricity generated from hard coal and lignite - Electricity generated from renewable energies

3.2.4. Climate change

European regions differ widely in relation to the challenges they face from climate change. Mediterranean regions are likely to experience significant increases in days of extreme heat, growing risk of droughts, declining crop yields and more multiple climatic hazards 8 . Coastal areas face the risk of rising sea levels, increasing sea temperatures 9 and growing numbers of ‘marine dead’ zones 10 . The Atlantic region will experience increasing instances of heavy rainfall and more risk of river and coastal flooding and damage from winter storms. Mountain regions are expected to suffer higher increases in temperature than the European average, a shift of plant and animal species to higher ground and a greater risk of some of them becoming extinct as well more chance of rock falls and landslides and reduced potential for hydro-electricity generation.

At the same time, climate change might create opportunities, such as an expected reduction in energy demand for heating in Northern European and Atlantic regions or new possibilities for exploiting natural resources and sea transport in Artic regions. But in general, climate change will have major adverse effects on the environment which it will be necessary, and often costly, to adapt to.

Vulnerability to climate change varies widely from one region to another. According to meta-analysis integrating assessments covering multiple areas (water, agriculture, tourism, ecosystems and so on) 11 , Italy, Spain and southern and central France are likely to have the highest number of areas adversely affected, along with parts of south-eastern Europe (Map 3-2).

Map 3-2 Adverse climate change effects under a 2°C warming scenario

Climate change is also expected to increase the occurrence of natural hazards throughout the EU in the coming decades. Recent studies 12 show that places where the effects are likely to be particularly severe (i.e. affected by increase in the probability of hazard occurrences of at least 20% for three or even four of the 7 hazards considered) will progressively extend northwards to Central and Western Europe in the coming decades, covering, by 2050, many areas of the Netherlands, the UK and Ireland as well Spain, France, Italy, Bulgaria and Romania
( Map 3-3 ).

Map 3-3 Projected increase in multi-hazard exposure

Estimating the economic costs of climate change is particularly challenging, but most studies indicate that these costs could be high even for modest changes in climate 13 . The PESETA II study estimates total damages in the EU of up to EUR 190 billion by the end of the 21st century under a high economic growth scenario 14 , mostly from heat-related deaths and losses in agriculture and coastal areas.

The costs are expected to be far from evenly distributed across Europe, and much higher in southern Europe than elsewhere (the CIRCE project estimates that Mediterranean countries could lose an average of just over 1% of GDP by 2050 notably from damage to tourism and energy) 15 .

3.3. State of environment

3.3.1. Water

One of the objectives of the Seventh Environment Action Programme (7th EAP) is to ensure the good status of transitional 16 , coastal and fresh water by 2020. Surface water 17 is a major component of fresh water and improving its ecological state is critical to achieving this objective.

The Water Framework Directive 18 (WFD) and other water-related ones have contributed to improving water protection in the EU. In general, people throughout the EU can safely drink tap water and swim in many of the coastal areas, rivers and lakes. However, reducing pollution to meet the objectives of the WFD requires as a pre-condition that several other Directives and regulations are fully implemented 19 .

Although progress in wastewater treatment and reductions in agricultural inputs of nitrogen and phosphorus have helped to improve the quality of surface water in the EU, pollution from agriculture (particularly nitrogen losses) as well as from urban and industrial wastewater remains significant. According to the EEA, in 2015, only 53% of water bodies are estimated to have good ecological status, making it unlikely that the objective of achieving good status of all water will be met by 2020 20 .

Member States differ substantially in terms of the ecological status of their river basins ( Map 3-4 ). In Belgium, northern Germany and the Netherlands, over 90% of surface water is reported to be in a ‘less than good’ ecological state. In the Czech Republic, southern England, northern France, southern Germany, Hungary and Poland, 70% to 90% of freshwater bodies (lakes and rivers) are reported to be in a similar state. The ecological status of coastal and transitional water is also poor in the Black Sea and greater North Sea regions. On the other hand, a much larger share of surface water is in good ecological state in Northern regions of Sweden and Finland and some regions of Northern Italy, Northern Spain, Latvia and Greece.

Map 3-4 Main water bodies with less than good ecological status or potential

To achieve good status, Member States will have to do more to reduce the pressure on water bodies. This will require substantial investment in ways of reducing pollution or tackling over-abstraction of ground water and morphological and hydrological changes 21 . Such investment can be supported by cohesion policy (in the 2007-2013 programming period around EUR 17.8 billion of the ERDF and Cohesion Fund was allocated to wastewater infrastructure in 22 Member States 22 ).

Appropriate collection and treatment of wastewater to remove organic matter, nutrients (nitrogen and phosphorus) and other hazardous substances it contains is essential for improving the ecological status of water bodies (marine and freshwaters) as well as to reduce the risk to human health and biodiversity.

The Urban Waste Water Treatment Directive 23 (UWWTD) sets minimum requirements in respect of urban wastewater treatment, making it mandatory for settlements with the equivalent to 2 000 inhabitants or more. Since its adoption in 1991, it has led to a considerable reduction in discharges of major pollutants but its implementation still needs to be improved in a number of Member States and regions.

The level of treatment required in the UWWTD depends on the sensitivity of the areas of discharge and on the size of the settlements. Sensitive areas are those where the environmental risks due to the adverse effects from wastewater discharge are particularly high (e.g. risk of eutrophication by excess of nutrients) or which require specific protection, such as drinking water abstraction areas and waters for bathing and those where shellfish live. Secondary (biological) treatment, which decomposes most of the organic matter responsible for the oxygen depletion, is the minimum requirement in ‘normal' or non-sensitive areas. Tertiary (or more stringent) treatment, which removes nutrients and disinfects the water, is required in large settlements (with the equivalent of 10 000 inhabitants or more) discharging into sensitive areas.

According to the UWWTD 9th Reporting Exercise (2014), high compliance rates are generally observed in most EU-15' Member States, especially in Austria, Germany and the Netherlands, which have largely implemented the Directive. However, there are still a number of EU-15 countries which have compliance gaps and have delayed the implementation of necessary measures. This is notably the case for Italy, Spain, Belgium, Luxembourg and Ireland.

The picture is different for EU-12 Member States (i.e. excluding Croatia, for which the deadline for compliance is 2018). This is partly a result of their later accession and the transitional periods for compliance which have been granted to them. The last available results, however, show a substantial improvement in compliance with collection obligations compared to previous years. The compliance rate is generally high, except for Cyprus (61%), Slovenia (65%) and above all Bulgaria (26%) and Romania (3%). Some Romanian regions as well as several Bulgarian regions and Eastern Slovenia show compliance rates below 40%, and even below 20% as in the case of Romania and of south-western Bulgaria. This regional concentration of not compliant agglomeration has significant implication for the water quality of the affected river basins such as the Black Sea Basin.

The same applies to wastewater treatment. In the majority of EU-12 Member States, Secondary treatment of wastewater shows high compliance rates of above 85% for eight of the countries, the exceptions being Romania, Bulgaria, Malta and Slovenia which have much lower compliance rates. In some regions, like Principado de Asturias (ES), Sicilia (IT), Slovenia and most Bulgarian regions, the share of agglomerations where secondary treatment is not taking place is even below 40%. In these regions, human and ecosystem health is critically threatened due to the low degree of compliance.

Compliance rates are also high in most cases in respect of stringent treatment, where applicable, varying between 50% and 100%, except in Romania, Bulgaria and Malta, where there are substantial delays in implementing the necessary measures.

3.3.2. Waste

Solid waste affects human health as well as the environment since it generates emissions of polluting substances into the air, soil, surface water and groundwater. It also presents major challenges for management as the quantity of waste produced per person has increased steadily over time. A transition to a more circular economy requires action throughout a product’s life-cycle: from production to the creation of markets for waste-derived materials. Waste management is one of the main areas where further improvements are needed and which are within reach. Accordingly, reducing the generation of waste and promoting its reuse and recycling are key objectives of the EU action plan for the circular economy 24 .

In 2014, an average of 4.9 tonnes of waste per person were generated in the EU. Much of this was produced by construction and demolition, mining, quarrying and manufacturing. Households also produced a substantial amount of waste, an average of 411 kg per person. Marine litter, escaping from waste management systems, is a growing concern. The total amount of waste generated (including mineral waste) in the EU increased by around 2% between 2010 and 2014 though there are wide variations between Member States.

Increasingly, waste is recycled or energy is recovered from it. Between 2010 and 2014, the proportion of treated waste (excluding mineral waste) recycled increased only slightly from 53% to 55%, while the proportion incinerated with energy recovery rose from 11% to 14%. The increase in recycling occurred against a background of measures designed to stimulate it, including EU and national legislation, support from the structural funds, landfill taxes and pay-as-you-throw schemes.

In 2014, the proportion of waste (excluding mineral waste) disposed of in landfill fell from 28% to 25% in the EU ( Figure 3-4 ). There are, however, marked variations between Member States. Over 80% of waste is still landfilled in Bulgaria and Greece and over 50% in Estonia, Cyprus, Malta, Romania and Slovakia. By contrast, less than 5% goes to landfill in Belgium, Denmark and the Netherlands.

Figure 3-4 Share of waste landfilled in selected EU Member States, 2014

3.3.3. Sustainable transport

Besides making transport more competitive and increasing the quality of the network, EU transport policy has also sought to reduce dependence on oil, greenhouse gas and other emissions (such as SOx, NOX and fine dust), to limit congestion and to improve safety.

Over the past 20 years, the volume of goods and number of passengers transported within the EU has grown steadily, apart from during the global recession in 2008-2009. Between 1995 and 2014, both passenger and freight transport increased on average by just over 1% a year 25 . Transport increasingly faces serious social and environmental challenges. It is second only to energy in greenhouse gas emissions, accounting for 23% of the total and, unlike energy, its emissions have raised since 1990 (by around 20%). Transport may also have significantly damaging effects on the quality of the environment, such as by increasing fragmentation of natural habitats.

The aim, therefore, is to establish a 'sustainable mobility' model of transport, to develop an efficient and competitive transport sector as a key element of the EU internal market while at the same time reducing costs from road accidents, respiratory diseases, climate change, noise, environmental damage and traffic congestion. The model entails fostering environmentally-friendly modes of transport as well as combined and inter-modal transport.

In its 2011 White Paper on the future of transport up to 2050, the Commission set the objective of reducing greenhouse gas emissions from transport by at least 60% in relation to 1990 levels by 2050. The interim aim is to reduce emissions by 20% in relation to 2008 levels by 2020-2030, requiring a fundamental shift towards the use of less and cleaner energy and more efficient utilisation of transport infrastructure. To achieve these objectives, the White Paper called for a shift of 30% of freight being transported over 300 km by road to rail or water by 2030 and one of over 50% by 2050, a tripling of the length of the existing high-speed rail network by 2030 and a move of the majority of medium-distance passenger travel to rail by 2050. It also targets the establishment of a fully functional multimodal TEN-T in the EU by 2030 and a high-quality and high-capacity network by 2050. In many places, achieving these objectives implies improving markedly the quality of transport infrastructure and new construction. Transport is the main beneficiary of the Connecting Europe Facility which has a budget of EUR 24 billion for the period up to 2020.

Cars remain by far the predominant mode of passenger transport in the EU. In 2014, they accounted for over 83% of all inland passenger km travelled in the Union 26 , varying from 68% in Hungary to almost 90% in Portugal and Lithuania ( Figure 3-5 ).

Buses accounted for 9% of passenger km travelled, the share varying from 3% in the Netherlands to 23% in Hungary. Trains accounted for 8%, though the figure varies according to the size and state of the rail network. In France, Austria and Sweden, which have fast and frequent trains, around 10% of travel was by rail, while in Greece, Estonia, Lithuania, where the rail network is limited and of low quality, the figure was less than 2%.

Figure 3-5 Passenger travel by transport mode, 2014

In the case of freight, around 75% of goods were transported by road in 2014 ( Figure 3-6 ). In Cyprus, Malta, Ireland and Greece, all or almost all were. Only 18% on average went by rail, though in Austria, the proportion was 44% and in Latvia, 59%. In Romania, Belgium and the Netherlands, there is an extensive network of inland waterways and these carried around 20% of freight in the first two and almost 40% in the last.

Figure 3-6 Freight transport by mode, 2014

These figures have been remarkably stable over time both for passenger and freight transport, except in a few Member States, particularly Romania and Estonia, where the share of freight going by road increased by 10 and 18 percentage points, respectively, between 2011 and 2014. Significant effort is, therefore, needed to achieve a shift to more environmentally-friendly modes of transport.

3.4. Sustainable cities

3.4.1. Cities can be environment friendly

Cities are often considered to be inherently harmful for the environment. In practice however, cities are not just a source of pollution but also a potential solution to current environmental challenges. While urban areas in the EU generally face more environmental challenges than other places, they can often prove to be more resource and energy efficient than other areas where low-density settlements, energy-intensive buildings (e.g. detached houses) and the level of dependency on the car for transport are generally more common. Housing in cities tends not only to occupy less land but also more frequently takes the form of apartments and townhouses which generally require less energy to heat and cool.

Cities also offer more possibility adopting a low carbon lifestyle. Living in cities tends to make it possible to access a large number of services using less energy-consuming modes of transport. People generally prefer to be close to the services and facilities they regularly have need of, such as schools, healthcare services, childcare, cultural and sports facilities and shops. The average distance to such services is usually much less for people living in cities than in towns and suburbs or rural areas.

On average in the EU, the distance to access services by road is 4.5 times greater in rural areas (almost 9 km) than in cities (less than 2 km). In countries which are more urbanised, the difference is smaller, as in Malta (1.4 times greater in rural areas), the Netherland (2.3 times), Belgium (2.9 times) and the UK (2.8 times). In countries where urban areas are more dispersed, the difference can be much larger (at the extreme, in Finland, it is 13 times greater).

The difference between cities and other areas in terms of accessing services varies according to the service concerned. Local services (such as schools, general health services, childcare, sports facilities and shops) are generally available in all types of municipality, even though they take longer to reach in rural and suburban areas ( Figure 3-7 ). The difference is greater for ‘sub-regional’ services, such as high schools, hospitals, theatres, cultural facilities and supermarkets, and greatest of all for regional services, such as specialised education and healthcare centres, large sports and cultural facilities or government offices. The average distance to reach such services in the EU is 48 km in rural areas, 38 km in towns and suburbs and less than 10 km in cities with a population of more than 250,000.

Figure 3-7 Distance to services by type of municipality in the EU

Accordingly, large cities offer the possibility of accessing services by walking or by bicycle while in rural areas or in smaller towns, it is much more difficult, or impossible, to do so. For instance, the average share of population in the EU living within 1 km of local services increases rapidly with the degree of urbanisation and the size of city, rising from 12% in rural areas to over 80% in cities of more than 5 million inhabitants ( Figure 3-8 )

Figure 3-8 Population living within 1km distance of different services, by city size and degree of urbanisation in the EU

Cities also tend to be more efficient in their use of land. Built-up areas per person in cities are only a quarter of those in rural areas. This reflects the fact that the availability of land and its cost make cities more attractive for less land-intensive activities, such as services, company headquarters or leisure facilities, than suburbs or rural areas. Land scarcity also increases the incentive to economise on land use for housing, which is generally smaller in cities than in other areas where the average area occupied per household tends to be much larger.

Although land use per inhabitant is usually greater in large cities than in smaller ones, there are wide variations across the EU. In particular, cities in Northern and Western Europe are often more densely populated than in southern and central-eastern EU countries and the built-up area per inhabitant, therefore, tends to be smaller ( Map 3-6 ). This difference tends to increase over time. Between 2006 and 2012, the built-up area per inhabitant increased most in cities in the southern and central-eastern EU while it declined in a number of large cities in northern and western Europe ( Map 3-7 ).

Map 3-6 Residential, industrial and commercial areas per inhabitant by city, 2012

Map 3-7 Change in residential, industrial and commercial areas per inhabitant by city, 2006-2012

3.4.2. Changes in land use per person

The process of urbanisation is driven by a range of factors that can be influenced by various types of policy, including cohesion policy. According to a recent study 27 , land use per person in the EU increased steadily from 0.94 of a hectare per 100 people in 1975 to 1.3 hectares in 2010. The overall increase in land use per person is consistent with an ‘urban sprawl’ phenomenon, or the rapid, and sometimes uncontrolled, expansion of built-up areas around towns and cities, creating widespread and relatively low density urban suburbs, often inefficient in terms of energy and land consumption 28 .

The observed increase in land use per person, however, seems to be running out of steam as urban areas in many EU regions have become more densely populated over more recent years. The main increase in land use per person occurred over the period 1975-1990. In the period 2000-2010, despite a continuing slight increase at EU level, many regions experienced decreases.

The main developments in land use per person in different types of EU region are as follows:

·Metro and capital city regions: a NUTS 3 region which is a metropolitan area or part of one is more likely to experience increases in population density, and even more so if it contains the national capital city.

·Rural regions: a rural NUTS 3 region is likely to experience a decline in population density, which means that built-up areas are expanding at a faster pace than population.

·Increases in population, GDP per head, employment and accessibility are all positively associated with growth of population density. In general, socio-economic factors are major determinants of a region's attractiveness.

·Regions with a high Percentage of Available Land (PAL) have few or no physical constraints on development which discourages growth of population density. Pressure on land prices is likely to be low and so extensive land development is relatively inexpensive. Conversely, regions with limited space for development tend to experience upward pressure on land prices, leading to denser urban development.

·Places with high initial levels of population density generally experience lower growth of density, suggesting that further densification may be discouraged in such regions. This could be because of two possible complementary reasons: a concern to avoid or reduce the diseconomies resulting from densification and technical or legal constraints on population growth.

3.4.3. Urban transport

Public transport is equally more accessible in large cities. In the vast majority of large cities, the share of the population with high or very high access to public transport is above 60%, and up to 98% in Madrid ( Figure 3-9 which shows the situation in a sample of large cities) 29 . The only exception is Dublin where the figure is only 38%. The figures tend to be slightly lower for mid-size cities. The proportion of inhabitants with high or very high access to public transport is less than 50% in Toulouse and Vilnius but close to 90% or more in Bologna, Sevilla and Edinburgh ( Figure 3-10 ).

Figure 3-9 Access to public transport in large European cities, 2014-2016

Figure 3-10 Access to public transport in mid-size European cities, 2014-2016

3.4.4. People living in cities suffer more from pollution

In 2015, the proportion of people in cities in the EU reporting to live in an area with environmental problems (19%) was larger than for those in towns and suburbs (13%) and rural areas (8%) ( Figure 3-11 ). The proportion for those in cities was particularly large in Malta (34%), Germany (33%) and Greece (30%), while it was only around 10% or less in Ireland, Cyprus, Denmark, Croatia and Finland, where environmental problems seem less common 30 . 

Figure 3-11 People reporting that they live in an area with problems relating to pollution, grime or other environmental problems, by degree of urbanisation, 2015

Air pollution remains a major environmental concern in the EU. Nine out of 10 people in urban areas in the EU are exposed to pollution concentrations above the levels recommended by the World Health Organisation (WHO). Air pollution has a major impact on human health, with an estimated 400 000 premature deaths each year due to high levels of fine particles and ozone. It also has a significant effect on ecosystems. Excessive nitrogen deposits (eutrophication) and ozone concentrations adversely affect biodiversity and crop yields and cause other material damage in over half of the EU.

At the same time, emission of air pollutants, notably of carbon monoxide, sulphur oxides and lead, has declined markedly in the EU over the years, partly as a result of EU legislation 31 . The application of European standards has also been successful in reducing vehicle emissions (such as after the introduction of the diesel particle filter), and the progressive renewal of the vehicle fleet means that air quality in the EU is likely to improve over the long-term. However, more needs to be done to address the issue, such as introducing regional or local incentives to favour very low pollutant emitting vehicles or even zero emission ones.

Some areas are still far from complying with agreed EU air quality standards 32 . This is notably the case in cities, where the majority of the EU population lives and where levels of sulphur dioxide, nitrogen oxides (NO2), volatile organic compounds, ammonia, fine particulate matter (PM2.5 and PM10 33 ) and ground level ozone (O3) remain high.

Air pollution is severe in a number of cities in southern and central Poland, the Czech Republic, Romania and Bulgaria ( Map 3-8 ) but also in Southern Europe (Po Valley, Naples, Cyprus and Greece). According to the EEA, in 2014 around 17% of the urban population in the EU was exposed to PM10 levels above the daily limit and 9% to PM2.5 levels above the EU target 34 .

To a large extent, concentration of airborne particulate matter is caused by emissions from diesel engines or from coal mining and other heavy industry. It is also affected by atmospheric conditions, pollution levels rising with sunshine and hot temperatures. These factors explain the geographical distribution of high PM concentrations. In 2013, for example, the average concentration rose above 40 μg per cubic metre in 9 cities in Bulgaria (including Sofia), peaking at 62.2 per cubic metre in Plovdiv, the second city 35 . The Czech cities of Havírov, Karviná and Ostrava in the coal mining region of Moravia-Silesia also recorded very high concentrations of PM. At the other end of the spectrum, most cities with relatively low levels of air pollution are located in the Nordic and the Baltic Member States.

Concentration of ground-level ozone can cause breathing and cardiovascular problems, asthma and lung disease. High concentrations occur mostly in cities in Northern Italy, Spain (e.g. Jaén and Toledo), the East and South of France (e.g. Sophia-Antipolis, Martigues, Mulhouse, Colmar and Aix-en-Provence) and Southern Germany (e.g. Freiburg im Breisgau, Karlsruhe, Hanau, Friedrichshafen and Heidelberg) ( Map 3-9 ). Around 15% of the urban population in the EU lives in areas in which the EU O3 target threshold for protecting human health was exceeded in 2013 36 .

(1) On 30 November 2016, the Commission proposed an update to the Energy Efficiency Directive including a new 30% energy efficiency target for 2030.

(2) Note that these energy targets are not straight-forward to interpret in energy efficiency terms. The main determinants of energy use are GDP growth and the share of (heavy) manufacturing in the economy and in general, changes in energy consumption per se say very little about energy efficiency as such.

(3) Primary energy consumption is the energy supplied to industry, transport, households, services and agriculture, including generation/ transformation losses, consumption of the energy transformation sector and network losses.

(4) In most cases, targets reflect the objective to reduce energy consumption by 2020. However, for some countries the target allows an increase in primary energy consumption.

(5) See European Commission (2015), Securing Energy Efficiency to Secure the Energy Union - How Energy Efficiency meets the EU Climate and Energy Goals, JRC Science and Policy Report EU 27450.

(6) Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources, OJ L 140, 5.6.2009, p. 16.

(7) Note that renewable energy is not necessarily environment friendly: solar, wind, biomass or hydropower projects may have significantly adverse effects on e.g. biodiversity or water bodies (including through intensive land use and reduced connectivity of rivers). In consequence, strategic and integrated planning with early stakeholder involvement, in line with relevant EU legislation (SEA, EIA, WFD, Birds and Habitats Directives) is essential to maximize renewable energy production while reducing environmental impacts.

(8) The number of hot days (those exceeding the 90th percentile threshold of a baseline period) has almost doubled since 1960 across Europe. Since the beginning of the 21st century, Europe has experienced several extreme heat waves (in 2003, 2006, 2007, 2010, 2014 and 2015). Under a high emissions scenario, very extreme heat waves are projected to occur as often as every other year in the second half of the 21st century (EEA (2017), Climate change, impacts and vulnerability in Europe 2016, An indicator-based report, EEA Report N° 1/2017).

(9) An increase in sea temperature is likely to have important consequences in term of biodiversity. Wild fish stocks are responding to changing temperatures and food supply by changing their distribution which can affect local communities dependent on them.

(10) Dead zones are hypoxic (low-oxygen) areas caused by excessive nutrient pollution from human activity coupled with other factors that deplete the oxygen required to support most marine life in bottom and near-bottom water.

(11) See EEA (2017), Climate change, impacts and vulnerability in Europe 2016, An indicator-based report, EEA Report N° 1/2017 for a meta-analysis.

(12) Forzieri, G., Feyen, L., Russo, S., Vousdoukas, M., Alfieri, L., Outten, S., Migliavacca, M., Bianchi, A., Rojas, R. and Cid, A. (2016), "Multi-hazard assessment in Europe under climate change", Climatic Change 137, 105–119 (doi: 10.1007/s10584-016-1661-x).

(13) Ciscar, J. C., Feyen, L., Soria, A., Lavalle, C., Raes, F., Perry, M., Nemry, F., Demirel, H., Rozsai, M., Dosio, A., Donatelli, M., Srivastava, A., Fumagalli, D., Niemeyer, S., Shrestha, S., Ciaian, P., Himics, M., Van Doorslaer, B., Barrios, S. (2014), Climate impacts in Europe: The JRC PESETA II Project, JRC Scientific and Policy Reports EUR 26586 EN, JRC87011, European Commission, Joint Research Centre, Seville.

(14) In 2000, the Inter-governmental Panel on Climate Change (IPCC) published the Special Report on Emissions Scenarios (SRES) which describes greenhouse gas emission scenarios used to make projections of possible future climate change. The SRA1B scenario assumes rapid economic growth, a global population that reaches 9 billion in 2050 and then gradually declines, the quick spread of new and efficient technologies, a convergence of world income and way of life and extensive social and cultural interactions worldwide.

(15) Navarra, A. and Tubiana, L. (2013), Regional assessment of climate change in the Mediterranean, Advances in Global Change Research, Springer Netherlands, Dordrecht.

(16) Transitional waters are bodies of surface water in the vicinity of river mouths which are partly saline as a result of their proximity to coastal waters but which are substantially affected by freshwater flows.

(17) Surface water is water on the surface of the planet in rivers, lakes, wetlands and oceans, in contrast to groundwater and atmospheric water.

(18) Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy, OJ L 327, 22.12.2000.

(19) This includes the Urban Waste Water Treatment Directive, the Nitrates Directive, the Directive on Sustainable Use of Pesticides and the Industrial Emissions Directive.

(20) Note that the picture is similarly bleak for marine ecosystems. In their initial assessments for the Marine Strategy Framework Directive, Member States indicated that only 4% of marine species and habitats have a 'good environmental status', while 80% are categorised as 'unknown'. The indication is that marine resources are being used unsustainably and as a number of human off-shore and on-shore activities depend on the health, cleanliness and productivity of the seas, there is a need for them to be used responsibly.

(21) River morphology corresponds to the shapes of river channels. It is determined by a number of processes and environmental conditions, including the composition and erodibility of the river bed and banks, vegetation and the rate of plant growth, the availability, size and composition of sediments and human interaction. River hydrology refers to the movement, distribution and quality of water.

(22) COM(2016) 105 final of 4.3.2016.

(23) Council Directive 91/271/EEC 21 May 1991 concerning urban waste water treatment, OJ L 135, 30.5.1991, p. 40.

(24) European Commission 'Closing the loop - An EU action plan for the circular economy', COM(2015) 614 of 2.12.2015. A circular economy is one in which the value of products, materials and resources is maintained for as long as possible, minimising waste and resource use.

(25) European Commission (2016), Statistical Pocket Book – EU Transport in Figures, Luxembourg: Publications Office of the European Union, 2016 (https://ec.europa.eu/transport/facts-fundings/statistics/pocketbook-2016_en).

(26) Passenger-kilometre represents one passenger travelling a distance of one kilometre. The share is the percentage of transport by passenger cars in total inland passenger transport, measured in passenger-kilometres.

(27) Batista e Silva F, Alvarez M, Vizcaino P, Jacobs Crisioni C, Ghisetti C, Pontarollo N, Lavalle C, D’Hombres B (2017) Determinants of urban land densification in Europe (forthcoming).

(28) See for instance Jaeger J., Bertiller R, Schwick C. and Kienast F., (2010) "Suitability criteria for measures of urban sprawl", Ecological Indicators 10(2): 427–441.

(29) No access: it takes more than 5 minutes to walk to a bus or tram stop and over 10 minutes to reach a metro or train station; Low access: it takes less than this to walk to a public transport stop – i.e. people can easily walk there – with less than four departures an hour; Medium access: it people can easily walk to a public transport stop with between 4 and ten departures an hour; High access: people can easily walk to a bus or tram stop with more than 10 departures an hour OR people can easily walk to a metro or train station with more than 10 departures an hour (but not both); Very high access: people can easily walk to a bus or tram stop with more than 10 departures an hour AND a metro or train station with more than 10 departures an hour. See Dijkstra, L. and Poelman, H. (2015), "Measuring access to public transport in European cities", Regional Working Papers N° 01/2015.

(30) Note that these figures relate to perceived problems which might differ from actual problems as a result of differences in expectations about the state of the environment.

(31) Directive 2010/75/EU on industrial emissions, Directive (EU) 2015/2193 on medium combustion plants, Directive (EU) 2016/2284 on national emission ceilings and Directive 2008/50/EC on ambient air quality.

(32) Directive 2008/50/EC on ambient air quality and cleaner air for Europe fixes air quality standards, with a limit of 40 μg/m³ for the annual mean concentration of nitrogen dioxide. For fine particles, the limit is not more than 35 days per year with a daily average concentration exceeding 50 μg/m³ and a mean annual concentration not exceeding 40 μg/m³. For ozone, the limit is a daily 8-hour mean concentration not exceeding 120μg/m³ on more than 25 days per year.

(33) Particulate matter (PM) are microscopic solid or liquid matter suspended in the atmosphere. Subtypes of atmospheric particulate matter include respirable particles with a diameter between 2.5 and 10 micrometres (μm).

(34) European Environment Agency (2015), Air quality in Europe — 2015 report, Report No 5/2015. http://www.eea.europa.eu/publications/air-quality-in-europe-2015 .

(35) EUROSTAT (2016), Urban Europe, Statistics on cities, towns and suburbs, Edition 2016, Publications office of the European Union, Luxemburg.

(36) European Environment Agency (2015), Air quality in Europe — 2015 report, Report No 5/2015. http://www.eea.europa.eu/publications/air-quality-in-europe-2015 .

EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

Map 3-8 Concentrations of airborne particulate matter (PM10) in cities, 2014

Map 3-9 Concentration of ground-level ozone (O3) in cities, 2014

Other types of pollution are also important in an urban environment, including noise pollution. A perception survey on the quality of life in 79 European cities conducted in 2015 1 suggests that in most cities, the level of noise is not a major problem ( Figure 3-12 ). In 62 cities, less than half of respondents reported a problem with noise levels, though the proportion was generally larger in capital cities than in others. The proportion was particularly small in the Nordic Member States (Oulu, Finland, 12%; Aalborg, Denmark, 13%) and the UK (Tyneside and Belfast, 14%). However, in a number of other cities noise pollution seems to cause discomfort and stress, particularly in Bucharest, Palermo and Athens, in each of which around two-thirds reported problems.

Figure 3-12 Proportion of people who are satisfied with the level of noise in their city, 2015

3.4.5. Access to green spaces

Green Infrastructure (GI) is a strategically planned network of natural and semi-natural areas with other environmental features designed and managed to deliver a wide range of ecosystem services. Ecosystem services are the contributions of nature to human well-being, such as the provision of clean air and water, pollination of fruit and vegetables by bees and the recreation provided by natural areas.

The EU Habitats and Birds Directives have given rise to Natura 2000 areas, the EU network of protected areas, which is the backbone of EU Green Infrastructure deployment, and is designated to protect the most threatened habitats and species. Natura 2000 also provides opportunities, for the development of tourism, recreation, agriculture, forestry, sustainable fisheries and aquaculture as well as nature-based means of controlling floods, mitigating and adapting to climate change and producing other ecosystem services. Recent studies have shown that the economic benefits generated by the Natura 2000 network can be substantial 2 .

The establishment of Natura 2000 is to a large extent complete on land (with more than 18% of the EU’s landmass protected as a result). Progress in designating of marine areas for protection has been slower, though 6% of EU seas and oceans are now covered.

Improving the environment in less favoured regions increases their attractiveness for external investors and tourists and helps to strengthen their regional identity, but there remain shortcomings in the implementation of the Directives concerned, partly as a result of a lack of adequate funding 3 .

Deploying Green Infrastructure in rural areas in the EU can give rise to a wide range of ecosystem services, but more investment is needed in it in and around urban areas in order to increase the beneficial effects of the services it produces, even though the costs are likely to be higher for a given level of nature protection 4 .

Green urban spaces are a good example of this general principle. Green urban spaces can mitigate pollution problems and help to absorb carbon from the atmosphere as well as rainwater. They also offer shade and so help to limit temperature increases, as well as being important places for social interaction and for the quality of life in general. Access to green urban areas varies widely across EU cities ( Map 3-10 ). In many cities in Western, Central and Northern Europe, people have access to vast areas of green space. In Chomutov-Jirkov in the Czech Republic, for example, over 13 000 hectares of green space can be accessed in less than 10 minutes walking. On the other hand, such spaces are less present in many Eastern and Southern EU cities, partly because of the climate which often makes it costly to maintain them, given the need for extensive watering systems.

Map 3-10 Access to green urban areas in cities, 2012

Urban green spaces also play an important role in regulating air quality, as evidenced by many studies (Escobedo and Nowak. 2009, Litschke and Kuttler, 2008, Nowak et al. 2006, Nowak et al., 2013). The latest (Vizcaino et al., 2017), which focuses on European functional urban areas (FUAs) 5 , finds that the contribution of green urban spaces to reducing NO2 concentration varies widely across the EU. In a number of Swedish cities (Gothenburg, Uppsala, Umeå, Örebro and Jönköping), Târgovişte in Romania, Vilnius in Lithuania and Ioannina in Greece, more than 50% of NO2 concentration is removed by green spaces ( Map 3-11 ). By contrast, in many cities in the Southern UK, Belgium, the Netherlands and Northern Italy, because of low levels of vegetation, only a small fraction is removed.

Map 3-11 Share of NO2 concentration removed by vegetation in cities, 2010

3.4.6. River flooding

There is a significant risk that large parts of Europe will be confronted with an increase in the occurrence and frequency of floods as a result of climate change. Effective water management, as required by the WFD, will help Member States to prepare for extreme weather events which can cause substantial damage 6 .

Since the WFD, the Floods Directive 7 , adopted in 2007, is intended to create a pan-European framework that can support Member States to identify, assess and tackle flood risk. Since its introduction, the management of flood risk has been strengthened and new models and methods for assessing and/or managing the risk have been developed. A more systematic, coordinated and holistic implementation of management plans has been achieved with a better understanding of priorities, along with a more focused discussion and improved awareness of the risk and the development of partnerships, involving spatial and land use planning and civil protection, to reduce it.

River flooding is a frequently occurring natural hazard in Europe. It is of particular concern in urban areas, where physical and human losses can be high. The flood impact indicator developed by Lund et al. (2013) 8 enables the impact of flooding at both regional and urban level to be assessed. The methodology takes account of both the estimated natural risk and the capacity of the region or city to mitigate the event and recover from it. When applied to Europe's major FUAs, it shows that, though the degree to which areas are affected varies greatly depending on its location and the hydrological characteristics of its surrounding (upstream) area, the risk of flooding exists in many cities right across the EU ( Map 3-12 ). In a large number of FUAs in the Netherlands, Italy and Hungary, over 50% of the population is at risk in the event of the biggest flood in the last 100 years reoccurring. There is also a high risk in FUAs in Southern Germany, Poland, Romania, Spain and France.

Map 3-12 Population flooded in the case of the biggest 100 year flood in FUAs reoccurring

3.5. Cross-border cooperation and territorial dimension of cohesion policy

The EU is facing an increasing number of new global challenges which have a significant impact on the economic, social and territorial cohesion in Europe. To respond to many of these challenges, European territorial cooperation enables countries and regions to identify solutions to common problems in border regions and other functional areas of co-operation.

3.5.1. Border regions

For analytical purposes, border regions are defined as NUTS 3 regions located along or very close to land and maritime borders between EU Member States and other countries. There are two types of border region: internal ones, i.e. regions located on borders between EU Member States and/or European Free Trade Area (EFTA) countries, and external ones, i.e. those located on borders between an EU country and a non-EU or EFTA one ( Map 3-13 ).

As the severity of border effects is likely to diminish with the distance from the border, the definition of border regions is complemented by that of border areas, which are those covering a 25 km zone on both sides of the border. Indicators can be defined for border regions or border areas or for a combination of both. NUTS 3 regions not being formally along land borders but which lie at least partly inside the 25 km wide area along borders are also considered to be border regions.

Map 3-13 Border regions, NUTS 3

In the last few decades, integration among EU Member States as well as with neighbouring countries has been progressively extended. However, despite the elimination of many institutional and regulatory barriers, borders still continue to obstruct the movement of goods, services, people, capital and ideas, which prevents the benefits of integration from being fully realised.

In this context, European Territorial Cooperation has played an important role in mitigating the adverse effects of internal borders and has realised many concrete achievements with regard to cross-border security, transport, education, energy, health care, training and job creation. For the 2014-2020 period, EUR 6.6 billion was allocated to 60 cross-border cooperation programmes 11 .

In 2014, around a third of the EU population lived in land border regions, the GDP of which was some 28% of the EU total, implying a GDP per head of 88% of the EU average. This average hides wide variations, reflecting the differences between different parts of the EU, with border regions with a high GDP per head being located in the North and West and those with a low level being located in Central and Eastern Europe.

Recent research has identified some of the main obstacles to the development of border regions. There are often socioeconomic disparities between regions on the two sides of the border which reduce the opportunities to cooperate and hinder integration. For some regions, physical obstacles and poor transport infrastructure limit access to markets and services on the other side of the border, while cultural and language differences can restrict interaction between people or businesses. Legal and/or administrative difficulties can also limit the scope for regional integration and labour mobility even in places which are potentially functional regions.

A recent study 12 suggested that if only 20% of the existing legal and administrative obstacles were removed, border regions could gain up to 2% in GDP. Regions located along borders in central EU and EFTA countries may have a lower GDP due to these obstacles ( Map 3-13 ).

(1)

   European Commission, Directorate-General for Regional and Urban Policy (2016), Quality of Life in European Cities 2015, Publications office of the European Union, Luxemburg.

(2)

   European Union (2013), The Economic benefits of the Natura 2000 Network, Publications Office, Luxembourg.

(3)

   Special Report No 1/2017: More efforts needed to implement the Natura 2000 network to its full potential, the European Court of Auditors.

(4)

   Vallecillo, S., Polce, C., Barbosa, A., Perpiña Castillo, C., Zulian, G., Vandecasteele, I., Rusch, G. and Maes, J. (2016), Synergies and conflicts between the delivery of different ES and biodiversity conservation: Spatial planning for investment in green infrastructure and ecosystem restoration across the EU, OpenNESS Deliverable D3.3/WP3.

(5)

   The functional urban area consists of a city plus its commuting area; see the EU-OECD FUA definition at http://ec.europa.eu/eurostat/statistics-explained/index.php/European_cities_%E2%80%93_the_EU-OECD_functional_urban_area_definition .

(6)

   Under a no-adaptation scenario (i.e. assuming continuation of the current protection against river floods up to a current 100-year event), EU damages from the combined effect of climate and socioeconomic changes are projected to rise from EUR 6.9 billion a year to EUR 20.4 billion a year by the 2020s, EUR 45.9 billion a year by the 2050s and EUR 97.9 billion a year by the 2080s. See Rojas et al. (2013) Climate change and river floods in the European Union: Socio-economic consequences and the costs and benefits of adaptation, Global Environmental Change 23, 1737–1751:    
    http://www.sciencedirect.com/science/article/pii/S0959378013001416# .

(7)

   Directive 2007/60/EC of the European Parliament and of the Council of 23 October 2007 on the assessment and management of flood risks, OJ L 288, 6.11.2007, p. 27.

(8)

   Lung T., Lavalle C., Hiederer R., Dosio A. and Bouwer L. M. (2013), A multi-hazard regional level impact assessment for Europe combining indicators of climatic and non-climatic change, Global Environmental Change, 23, p. 522-536.

(9)

    http://ec.europa.eu/eurostat/web/gisco/geodata/reference-data/administrative-units-statistical-units/urban-audit#ua11-14 .

(10)

   Natura 2000 sites (2016)  ( https://www.eea.europa.eu/data-and-maps/data/natura-8#tab-metadata ).

(11)

   In the case of external border regions, the Instrument for Pre-Accession Assistance (IPA) supports cross-border co-operation between candidate countries, potential candidate countries and EU Member States while the European Neighbourhood Instrument (ENI) provides support to EU regions bordering Neighbourhood countries to the East and the South.

(12)

   Camagni et al. (2017), Quantification of the effects of legal and administrative border obstacles in land border regions, Final Report to the European Commission, Politecnico de Milano.

EUROPEAN COMMISSION

Brussels, 9.10.2017

SWD(2017) 330 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

My region, My Europe, Our future:
The seventh report on economic, social and territorial cohesion

{COM(2017) 583 final}

Map 3-13 Percentage GDP loss in EU NUTS 3 land border regions due to cross border obstacles

The state of the cross-border road network varies considerably across the EU. In some places, it is good, even better than elsewhere in the region, such as along the Belgian-French or Belgian-Dutch borders ( Map 3-14 ). In other places, the cross-border road network is poor and limits the capacity of the regions to develop. This can be due to geophysical barriers, such as the mountain chain which forms the border between France and Spain, but it can also reflect the orientation of transport policy.

Map 3-14 Cross-border road network efficiency in border areas

Access to cross-border transport also varies across the EU. While in some cases access to cross-border rail services is as good as to services elsewhere in the region (observations on the diagonal of Figure 3-14 ), in many others, it is more limited (observations above the diagonal).

Figure 3-14 Population of border areas having access to rail passenger services, 2014

A number of border regions face more serious demographic challenges than other areas. Many located in the EU-13 have experienced substantial loss of population over recent years as a result of both a natural reduction (reflecting their older population) and outward migration ( Table 3-1 ). Between 2005 and 2015, population in the EU-13 land border regions fell by 3.5% as against 1.2% in non-border regions, outward migration reducing population by 1.9% combined with a natural reduction of 1.5%.

The situation is different in the EU-15, where population increased in border regions as in non-border ones, though at a slightly slower pace partly because of less inward migration.

Table 3-1 Change in population of land borders regions and other regions in the EU-15, EU-13 and EU-28, 2005-2015 (% change)

3.5.2. Other types of cooperation: interregional, transnational and macro-regional

Interregional cooperation is needed to tackle in an effective way common problems which affect most regions to differing extents, to enable examples of good practice and know-how to be shared, to build networks and to support analysis of major territorial development issues. Four interregional cooperation programmes are currently in operation (Interreg EUROPE, INTERACT, URBACT and ESPON) which cover all EU Member States and a number of third countries and which are allocated around EUR 1 billion for the 2014-2020 period.

There are, in addition, 15 transnational cooperation programmes which group together regions in different EU countries to tackle issues that are of common concern to them and which together have been allocated EUR 2.1 billion for the present period ( Map 3-15 ). They support a range of projects relating to innovation, the environment, transport, communication and sustainable urban development. Transnational Cooperation can help to establish functional links in a given territory, such as sea basin strategies, the arctic framework or macro-regional strategies. Under the ESF, transnational cooperation helps improve the delivery of employment and social policies and contributes in the implementation of reforms, by enabling the stakeholders to learn from experiences and good practices in other countries.

Macro-regional strategies are a form of territorial cooperation between countries which help to improve the implementation of EU policies. They are equally designed to tackle common problems, such as relating to the environment or climate change. Macro-regional strategies can also provide an appropriate framework for cross border institutional co-operation. They are not, however, directly financed under cohesion policy but they focus on an optimal use of existing financial sources (e.g. the ESIF, Horizon 2020, COSME, LIFE), better implementation of existing legislation and better use of existing institutions.

Since the European Council endorsed the EU Strategy for the Baltic Sea Region (EUSBSR) in 2009, three further macro-regional strategies have been developed: the EU Strategy for the Danube Region (EUSDR) in 2011, the EU Strategy for the Adriatic and Ionian Region (EUSAIR) in 2014 and the EU Strategy for the Alpine Region (EUSALP) in 2016 ( Map 3-15 ).

At present 19 EU and 8 non-EU countries are involved in macro-regional strategies which have become an integral part of the EU policy framework. They have increased interest in territorial cooperation and cohesion and awareness of its added value. They have led to increased coordination and strengthened cooperation in a number of areas (such as navigability, energy and climate change) and intensified cooperation with non-EU countries, helping to mitigate possible adverse effects on the EU’s external borders.

Each macro-regional strategy has achieved specific results:

·EUSBSR: the quality of the Baltic Sea water is being improved and nutrient inflows reduced through projects such as PRESTO or Interactive water management (IWAMA), while the SUBMARINER Network is further encouraging the innovative and sustainable use of marine resources;

·EUSDR: the coordinated management of water in the Danube river basin, though projects such as SEERISK is reducing the risk of damage by floods, while projects such as like FAIRWAY and DARIF are reducing bottlenecks to navigability and improving the safety of navigation;

·EUSAIR: cooperation with EU countries on issues of common interest is helping Western Balkan participating countries prepare for EU accession; green/blue corridors linking land and sea in the Adriatic and Ionian Sea have been identified as areas where strategic projects should be undertaken to achieve sustainable economic growth respectful of the environment,

·EUSALP: projects such as ‘mountErasmus’ are helping to establish a cross-border educational space for dual vocational training in the Alpine region, while ‘AlpinfoNet’ is being developed into a cross-border information system to improve passenger transport in the region.

Map 3-15 Transnational cooperation programmes 2014-20

3.5.3. Local, urban and metropolitan development

Cohesion policy promotes integrated and place-based approaches to foster economic, social and territorial cohesion, while at the same time recognising the role of sustainable urban development in realising overall EU objectives. To allow more flexibility in tailoring the provision of ESI funds to territorial needs, new and improved delivery mechanisms were put in place for the 2014-2020 programming period, in particular, Integrated Territorial Investment (ITI) and Community-Led Local Development (CLLD).

Almost 9% of the cohesion policy budget (around EUR 31 billion) is allocated to integrated territorial and urban development in the current period, the ERDF contributing the largest part (EUR 25.5 billion) and the rest coming from the other ESI funds over half the total is being provided using the new territorial instruments. Overall, the new territorial provisions are used in around 150 programmes, creating better links between the local strategies and the thematic objectives set out in the programmes.

The rationale for applying integrated, place-based approaches relates either to territorial integration, to thematic integration, to the blending of different financial resources or to institutional knowledge.

-Territorial integration: around half of the integrated strategies are using a functional approach, under which horizontal coordination arrangements help to improve the governance of a functional area and promote urban-rural or even cross-border links, though often it also requires new coordination arrangements between the administrative units involved.

-Thematic integration: ITI was specifically designed to combine investment under different priority axes or from different programmes, since a strategy supported through an integrated multi-thematic priority axis can only by financed through one programme. As a result, strategies implemented through ITI include on average more thematic objectives than those implemented through a priority axis.

-Blending different financial sources: the ERDF provides in most cases the bulk of financing together with the ESF, but the other ESI funds, other EU instruments and national or regional public and private funding can also make a significant contribution in some Member States, especially for ITI strategies. In most cases, the strategies will be funded by non-repayable grants, but financial instruments are also important in several strategies or for particular types of investment, such as for improving energy efficiency.

-Institutional knowledge: the strategic planning process and, more especially, the delegation requirements for sustainable urban development and CLLD have led in a number of Member States to the creation of new collaborative arrangements and bodies responsible for project selection and other tasks. In other Member States, this delegation has also resulted in capacity building and advisory measures, such as the establishment of new bodies or internal departments to support urban authorities' decision making.

Empowering cities: Sustainable Urban Development

The urban dimension is at the heart of Cohesion Policy. For the 2014-2020 period, at least 50% of the ERDF is invested in urban areas. Around EUR 14,5 billion (8 %) of the total ERDF budget has been allocated directly to support over 900 integrated sustainable urban development strategies, with considerable additional financing from the ESF and from other EU or domestic sources in a number of Member States.

Three options were provided for Member States to implement sustainable urban development strategies in the current period – through a dedicated multi-thematic priority axis, a dedicated programme or the use of the new ITI instrument. The ITIs have been relatively slow to be taken up but have been adopted in in 13 Member States, where well-functioning domestic programming and spatial planning arrangements were already in place or technical assistance was provided to help develop the strategies concerned.

Going beyond administrative boundaries: Integrated Territorial Investments

Cohesion policy pays particular attention to the specific socio-economic characteristics of functional area, making a wide range of investments available and promoting the adoption of integrated strategies targeted at specific needs.

Despite its novelty, ITI is being used flexibly for multidimensional place-based interventions for tackling complex territorial problems in 13 Member States. It has been adopted by around 150 different territorial strategies, which were developed not only for administrative regions to replace regional programmes but also for functional area such as remote and sparsely populated rural areas, islands and coastal areas, environmental protection sites and functional urban areas.

Strengthening local communities: Community-led Local Development

Community-led Local Development (CLLD) has been introduced under cohesion policy as a voluntary instrument, extending the existing LEADER approach for rural development and fisheries policies and its territorial focus depends very much on the coordination with the EAFRD and EMFF. Complementary arrangements usually target rural areas with small or medium sized towns or cities nearby, while in some Member States, the ERDF and ESF are used to support urban participatory measures targeted at social inclusion and urban regeneration.