Regions Benefiting from Globalisation and Increased Trade [PDF]

Regions Benefiting from Globalisation and Increased Trade Study for DG Regio Final Report Vol. I Politecnico di Milano Bocconi University, Milan IGEAT – ULB

Milan, 15 October 2009

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Research staff Politecnico di Milano Prof. Roberta Capello (project director) Prof. Roberto Camagni Dr. Ugo Fratesi (project coordinator) Dr. Antonio Affuso Dr. Andrea Caragliu

Bocconi University Prof. Laura Resmini (unit research director) Dr. Laura Casi Dr. Rachele Salmona

IGEAT, Free University of Brussels Prof. Christian Vandermotten (unit research director) Dr. Gilles Van Hamme Moritz Lennert Pablo Medina-Lockhart Dr. Mohamed El Aydam

National experts Dr. Rafael Boix Prof. Daniela Constantin Prof. Grzegorz Gorzelak Dr. David Grover Prof. Charlie Karlsson Dr. Gilles Van Hamme

2

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Table of Contents of Final Report – Vol. I EXECUTIVE SUMMARY .....................................................................................................6  1. Recent globalization patterns and benefiting regions ........................................................................................... 6  2. Globalization processes ........................................................................................................................................... 8  3. What are the EU regions most influenced by globalisation processes? Identification of global regions........ 10  4. Are the recent globalisation trends present in EU regional economies? Globalization trends in EU regions11  5. What are the factors distinguishing between successful and unsuccessful regions? ........................................ 15  6. What is the future of the European regional economies in light of the different globalisation strategies that may emerge over the next twenty years? A prospective analysis .......................................................................... 16  7. Regional policy implications ................................................................................................................................. 18 

1. GLOBALIZATION AND THE RESHAPING OF REGIONAL ECONOMIES..................23  1.1 The new globalization paradigm ............................................................................................................................. 23  1.2 Globalization and the new opportunities for regional economies......................................................................... 25  1.3 Benefiting regions: a definition................................................................................................................................ 27  1.4 Research questions and structure of the research work........................................................................................ 34 

2. GLOBALIZATION PROCESSES: TRADE AND FDI TRENDS ....................................36  2.1. A major geographical shift of trade at the world level ......................................................................................... 36  2.1.1. World trade trends.............................................................................................................................................. 36  2.1.2. A new division of labour? .................................................................................................................................. 37  2.1.3. The growing importance of service trade for EU............................................................................................... 38  2.1.4. EU countries in the global trade........................................................................................................................ 39  2.2. Geographical Patterns of FDI inflows.................................................................................................................... 43  2.2.1 General trends ..................................................................................................................................................... 43  2.2.2. Foreign firms in the EU27: geographical and sectoral patterns ......................................................................... 45  Annex 2.1. The geographical zones used in the analysis of trade ............................................................................... 49  Annex 2.2. FDIRegio dataset: its structure and its consistency with official data .................................................... 49 

3. IDENTIFICATION OF EUROPEAN GLOBAL REGIONS .............................................52  3.1 Aim of the chapter .................................................................................................................................................... 52  3.2. Globalization Index ................................................................................................................................................. 52  3.2.1. Regional participation in the global networks.................................................................................................... 52  3.2.2. Attraction of activities........................................................................................................................................ 53  3.2.3. Attraction of the workforce................................................................................................................................ 54  3.2.4. Air connections with extra-European space....................................................................................................... 55  3.3. Toward a synthetic globalisation index.................................................................................................................. 55  3.3.1. Methodology ...................................................................................................................................................... 55  3.3.2. Globalization index according to relative indicators.......................................................................................... 56  3.4. Open sectors ............................................................................................................................................................. 58  3.5. Regional taxonomy in front of globalization ......................................................................................................... 60  3.5.1. Theoretical taxonomy ........................................................................................................................................ 60  3.5.2. The taxonomy applied to European regions....................................................................................................... 61  3

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

3.6. Structural profile of global players ........................................................................................................................ 65  3.7. Towards an interpretation of the performance of global regions: next steps..................................................... 74 

4. SPATIAL PATTERNS OF GLOBALIZATION TRENDS ...............................................75  4.1. Introduction ............................................................................................................................................................. 75  4.2. Globalization trends in different groups of regions .............................................................................................. 76  4.3. Regional adaptation to global trends ..................................................................................................................... 82  4.3.1. A new globalization paradigm: local functions as conditions for benefiting from globalization....................... 82  4.3.2 The next global shift: presence of local skills and attraction of new functions................................................... 89  4.3.3. Deindustrialization and the rising of services economy: industry and service location patterns ....................... 91  4.3.4. Decentralization of inter-twined functions: the role of innovation in manufacturing location .......................... 94  4.4. Conclusions............................................................................................................................................................... 96 

5. SPATIAL FDI PATTERNS ............................................................................................99  5.1. Introduction ............................................................................................................................................................. 99  5.2. Spatial patterns of FDI across regions: does globalization matter? .................................................................. 102  5.3 Spatial patterns of globalization trends: does FDI matter? ................................................................................ 109  5.4. Determinants of foreign firm patterns of location .............................................................................................. 113  5.4.1 FDI’s location-specific determinants: a first selection...................................................................................... 115  5.4.2. The location specific determinants of FDI: the relative importance ................................................................ 117  5.5. Concluding remarks .............................................................................................................................................. 131  Annex 5.1 ....................................................................................................................................................................... 133 

6. FUNCTIONAL TRENDS AT REGIONAL LEVEL........................................................147  6.1. Regional position in the division of labour: trends in professions ..................................................................... 147  6.1.1. Data and indicators .......................................................................................................................................... 147  6.2. Regional functional specialisation ....................................................................................................................... 148  6.4. Conclusions............................................................................................................................................................. 154  6.5. Regional functional dynamics of open sectors..................................................................................................... 154  6.5.1. Data.................................................................................................................................................................. 154  6.5.2. A descriptive analysis of the open sector ......................................................................................................... 154  6.6. Sectoral/functional trends and the participation in the global networks.......................................................... 158  6.7. Conclusions............................................................................................................................................................. 159 

7. SUCCESS FACTORS OF BENEFITING REGIONS ...................................................160  7.1. Introduction ........................................................................................................................................................... 160  7.2. Regional growth patterns and structural characteristics................................................................................... 162  7.2.1. Definition and description of the patterns followed by European regions ....................................................... 162  4

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

7.2.2. Growth patterns and structural features of global players................................................................................ 168  7.2.3. Growth patterns and structural features of regional players............................................................................. 170  7.3. Common behaviours among global regions: winning strategies ....................................................................... 173  7.4. An in-depth analysis of winning strategies: a case-study approach .................................................................. 179  7.4.1. Aim of the qualitative analysis......................................................................................................................... 179  7.4.2. Selection of case studies .................................................................................................................................. 180  7.5. Major lessons from the case studies ..................................................................................................................... 184  7.5.1 General remarks ................................................................................................................................................ 184  7.5.2 In order to benefit from globalization, regions have to ... ................................................................................. 184  ... take advantage from their settlement structure, ... ................................................................................................. 184  ... shift towards services, ... ........................................................................................................................................ 185  ... shift from low-level to command and control functions, ... ................................................................................... 186  ... shift towards the knowledge economy, ... .............................................................................................................. 187  ... hold a high level of physical connections with the rest of the world, ... ................................................................ 188  ... have an economic structure which supports them in the markets, ... ..................................................................... 189  ... be able to attract and retain FDI, ........................................................................................................................... 190  ... be able to implement the good supporting policies, .............................................................................................. 191  ... and be able to avoid the major threats.................................................................................................................... 191  7.6. Conclusions............................................................................................................................................................. 192 

8. A PROSPECTIVE ANALYSIS.....................................................................................194  8.1. Setting the scene: main driving forces for a prospective analysis...................................................................... 194  8.2. An aggressive Europe in a high-quality competitive world and in a fast recovery from the economic crisis 198  8.3. A defensive Europe in a price-competitive world and in a slow recovery from economic crisis .................... 203  8.4. National results of the simulation exercise .......................................................................................................... 205  8.5. Regional results of the simulation exercise .......................................................................................................... 206  8.5.1 Results by regional typology............................................................................................................................. 206  8.5.2. Baseline scenario ............................................................................................................................................. 207  8.5.3. Regional results of scenario A ......................................................................................................................... 210  8.5.4. Regional results of scenario B.......................................................................................................................... 213  8.6. Main findings for the global regions .................................................................................................................... 216 

9. CONCLUSIONS AND POLICY IMPLICATIONS.........................................................217  9.1. Retrospect: the main lessons from our empirical analysis ................................................................................. 217  9.2. Regional strategies: winning behaviours ............................................................................................................. 218  9.3. The impact of national trends and performance on regions .............................................................................. 219  9.4. The impact of macroeconomic and structural policies on regions .................................................................... 222  9.5. The rationale for regional policies........................................................................................................................ 223  9.6. Regional policy implications ................................................................................................................................. 226 

REFERENCES ................................................................................................................229  5

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Executive summary 1. Recent globalization patterns and benefiting regions A new globalisation paradigm is underway… In recent years, the globalisation process has undergone reinvigorated expansion which has involved three closely interwoven processes: the increasing planetary integration of markets for goods and services (“internationalization”), markets for the location of economic activities (“multinationalization”), markets for production factors such as technologies and information. Whilst the phenomenon of globalisation is not new in itself, what is indeed new in recent times has been the form that it has assumed, recently labelled as the new globalisation paradigm.1. Before the advent of this new paradigm, firms and sectors were the finest level at which globalisation’s impact was felt: more open trade explained the fortune of some firms and sectors exploiting economies of scale in the production of final products in low-wage areas. Today, competition increasingly generates effects within the firm, at the functional and task level, on a task-by-task basis, imposing new competitive strategies on firms based on the strong reallocation of production phases at the spatial level. The off-shoring and out-sourcing of production phases become the usual corporate strategies by which firms seek to obtain productivity gains (see chapter 2 of the report). All this inevitably gives rise to increasing complexity in spatial terms: “there is undoubtedly something profoundly new about new globalization”.2 The North/South, Centre/Periphery model of the past – which implied the moving of activities and employment towards low-cost countries and exporting low-value-added manufacturing goods from those countries – was a model able to respond to competition on a firm-to-firm basis through competitive advantages based on low labour costs. Today, this is only a simplifying picture of real processes. In fact, the new globalisation paradigm requires firms to adopt complex strategies intended to optimize multiple trade-offs: between labour costs and temporal proximity to customers; between transportation and inventory costs; among accesses to knowledge sources, to skilled labour and unskilled labour; between commodification of existing products and the need to accelerate product design and time-to-market. New cross-border relationships emerge, reflecting the fact that firms do not seek labour cost advantages alone, but also additional and more value-added local assets, like knowledge, creativity and an entrepreneurial spirit, the flexibility (rather than cost) of labour markets, and the presence of relational and social capital (see chapter 1 of the report).

As a result regional competition increases… It is clear that globalisation trends affect the division of labour of sub-national economies within advanced countries, and that regional economies increasingly compete to grasp the opportunities offered by globalisation. Globalisation is an unavoidable process for regional economies, and all the more so because of its pervasive nature. Globalisation influences emerging and developed, old and new economies, areas specialised in advanced and traditional sectors, and in high and low value-added activities and tasks. Globalisation’s pervasiveness suggests that the best strategy for regional economies to cope with it is to adapt to change. Postponing this strategy entails high risks for a local economy obliged to engage in world competition as a latecomer with respect to its 1

Baldwin R., Globalisation : the Great Unbundling, Prime Minister’s Office, Economic Council of Finland, 2006. Kenney, M. and Florida R. (2004), Locating global advantage: industry dynamics in the international economy, Stanford University Press, Stanford Ca., Preface. 6 2

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

competitors, increasingly locked in old forms of production organisation, with limited complementary skills developed for an adaptation strategy, and in the meantime missing important opportunities offered by the globalisation processes. Globalisation is not only a source of threats. It is primarily a source of opportunities for national and regional economies. Firstly, there are evident positive macroeconomic effects. Secondly, an important impact of a spatial reorganisation of production is natural selection among firms in favour of the most efficient ones. Globalisation is heightening the competitive climate within which firms strive against each other. In order to cope with this climate, and with the consequent increase in dynamic uncertainty (concerning markets, technologies, and successful organisational models), firms rely more and more on high-quality human factors, access to information, devices or “operators” enabling rapid information assessment and transcoding, and forms of coordination and cooperation. As a consequence, directly or indirectly, through explicit location decisions or through the selective effects of competition, firms favour and support those territories that can supply these new “relational” factors. Territories, understood as collective actors, may help firms to be competitive by enhancing the presence of these new strategic production factors, thus bringing benefits to their “stakeholders” (local populations). We may accordingly say that regional economies compete with each other, no automatic device being sufficient to assure long-term development and well-being. … and some regions benefit... Globalisation heightens the competitive climate among European regions, some of which will be able to benefit from increased competition by enhancing their relative positions in the European economy. The aim of the project presented in this report is to highlight the structural conditions which enable local economies to improve their relative competitive positions. ‘Benefiting regions’ are defined here as those regions that maintain and even improve their competitive positions in the European economy thanks to globalization processes, i.e. those regions able to increase their relative production capacity (measured in terms of GDP) more than the European average amid globalization. This may result from the following different economic growth conditions (see chapter 7 of the report). a. Employment growth takes place in both high and low value-added functions, the former having a greater effect in quantitative terms than the latter. If this occurs, both employment and productivity increase, and therefore GDP as well. b. Employment losses take place in low value-added activities and are more than off-set by high value-added functions. In this case, employment losses are more than off-set by productivity increases, and GDP increases. c. Employment increases in low value-added functions, accompanied by a limited loss, if any, of high-skilled jobs. GDP increases despite the loss of productivity. The relationship between regional growth patterns and globalization trends is measured indirectly by identifying regions where globalization’s impact is felt first and more strongly. These regions – labeled global regions in this project – are empirically identified, and their economic dynamics analyzed and compared to other regions. The overall aim of this study is to identify what regions benefit from globalization, and the reasons for their success. The specific research questions are the following: 7

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

-

what are the European regions most influenced by globalisation processes? are the recent globalisation trends present in European regional economies? what regions have benefited most from globalisation trends over the past ten to fifteen years? what are the factors distinguishing between successful and unsuccessful regions? what is the future of the European regional economies in light of the different globalisation strategies that may emerge over the next twenty years?

These research questions are investigated within the framework of recent globalization processes, as now outlined.

2. Globalization processes Trade increases globally ... World trade has undergone a major geographical shift during the past decade: the poles of the triad (USA, Japan and Europe) have seen their share in the world trade decline to the benefit of emerging countries, especially Pacific Asia. This decline is very apparent in the case of the EU over the past decade: when intra-block trade is excluded, the share of EU-27 in world exports has decreased from 26.6% in 1995 to 22.6% in 2006 (see chapter 2 in the report). World trade can still be understood in terms of the classic division of labour conceptualised by the so-called centre/periphery approach. The means that specialization in terms of sectors and products is still correlated with the level of development: technological and upmarket products are highly specific to the most developed countries (EU-15, USA and Japan), while emerging countries are more specialized in less technological products, labour-intensive production, or primary products. In this regard, the EU maintains a good position in some of the most highly technological sectors and with some upmarket products, mainly chemicals (DG), machinery (DK) and transport equipment (DM). The EU is also very competitive in services, which account for 29% of extra-EU exports and have considerably increased the net balance for services in transportation, financial services, computer and information services, and other business services. However, there are evident signs of a new division of labour whereby emerging countries (mainly belonging to the “old generation”, for instance South Korea and Malaysia) are improving their positions in components and capital goods but not in upmarket finished goods for consumption, while China and India play a growing role as assemblers of consumer goods. Within the EU-27, the main change has been a modest decentralization process to the benefit of new member-states but also of peripheral Western countries, mainly Ireland and Spain, whose initial shares of EU-trade were very small. Among the main European countries, only Germany has been able to maintain its dominant position in the EU benefiting sectors, while Italy, France and especially the UK have globally declined in both intra and extra-EU trade in merchandises. The geographical pattern as regards trade in services is very different from that of manufacturing industries. In 2003, the United Kingdom was the leading service exporter in the EU, with 16.5% of total EU trade in services, followed by Germany (14.5%), France (11.2%), Italy (10.6%) and the Netherlands (9.4%). In nearly all types of services, the new member-states still have very limited shares of services exchanges. 8

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

... as well as FDI... Moreover, the EU has been a major player at world level in terms of FDI. At the same time it has been one of the main sources of, and destinations for, capital flows. As a destination region, the EU has been able to improve its capacity to attract FDI despite the emergence of new destinations of interest around the world, such as China and India. The main factor responsible for this success is the integration process, in both its vertical and horizontal dimensions (see chapter 2 in the report). The integration process has offered new business opportunities not only to extra-European multinationals, whose number has exponentially increased over time, but also, and mainly, to European multinationals, which have massively exploited the new markets opened by progressive liberalizations and enlargements. In 2005-07, intra-European foreign firms represented about 67% of all multinationals operating in the EU. One may therefore conclude that, in Europe, FDI-driven globalization is more a regional phenomenon than a global one.

… which, however, remains mostly an intra-EU phenomenon. The distribution of FDI in Europe replicates most of the characteristics of the globalization process apparent at world level. In particular, foreign firms are highly geographically concentrated in a small number of countries, which have changed over time and also across intra- and extra-Europe FDI. In 2005-07, the latter concentrated mainly in the UK and Ireland in EU15, and in Romania in EU12. Whilst the former countries offer very friendly environments for extra-EU firms, the latter has recently emerged among new EU member-states as favourite locations for foreign firms, regardless of their region of origin, owing to their recent structural reforms (see chapter 2 in the report). The concentration of FDI is marked at sector level as well. Again, the patterns are similar to those recorded at global level. In particular, as the share of FDI in services has dramatically increased over time, so the share of FDI in manufacturing has decreased. Although this phenomenon is widespread throughout Europe, the relative concentration of FDI in manufacturing is still higher, and all the more so in EU12, while the relative concentration of FDI in services increases in EU15.

At regional level, capital regions attract more FDI… The picture of the >location pattern of FDI across EU regions generally corresponds to the evidence on MNE location patterns across Europe. There is a strong concentration of FDI in the more advanced EU regions, from the UK to the north of Italy, including regions on the border between France and Germany, Ireland, Belgium, and the Netherlands. FDI in the new EU memberstates is largely concentrated in Romania and also, to some extent, in Polish and some Hungarian regions, and in Estonia. These patterns, however, have changed over time, and at the beginning of the period considered, the favourite locations for FDI were Czech and Hungarian regions. Generally speaking, capital regions seem better able to attract more new foreign firms than other regions in each country (see Tables 5.8 and 5.9 chapter 5 in the report).

… while most regions, notably in the southern coountries of the EU and at the Eastern external borders, are clearly at the margins of FDI location patterns . By contrast, most regions, notably in the southern countries of EU and on its eastern external 9

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

borders, have remained clearly marginal to the location patterns of MNEs in Europe. One possible explanation for these trends may be that the globalization process has driven the location of foreign firms. And, in fact, we found that the distribution of foreign firms differs substantially not only between globalized and non-globalized regions but also within globalized regions, with regions specialized in open service sectors attracting, on average, more FDI than regions specialized in open manufacturing sectors (See Table 5.3, chapter 5 in the report).

Differences in the distribution of FDI across and within types of regions are mainly driven by foreign firms operating in service sectors. FDI in the manufacturing sector is more evenly distributed, and this tendency has increased over time (See Table 5.3 and Table A5.1, chapter 5 in the report).

3. What are the EU regions most influenced by globalisation processes? Identification of global regions

The regions most strongly affected by globalization trends are those with high specialisation in open sectors and high accessibility … Regions expected to be more influenced by globalisation trends are global regions. These are identified as: -

regions with competitive and dynamic sectors (sectors in search of new markets, more open to competition, and better able to gain advantages from this competition); regions with high accessibility to large markets, and accessible from large markets.

Only those regions well endowed with physical connections and with the appropriate specialization in competitive and dynamic sectors have the potential to be global players. They are consequently those regions with the highest potential to benefit from globalization processes.

…and on the basis of these two features a taxonomy of European regions can be created. A taxonomy of regions can be created on the basis of two indicators: the physical accessibility of regions, and their specialisation in dynamic and open sectors. Four regional clusters are identified on the basis of these two dimensions (Table 1). 1. regional players. These are regions with a good specialization mix but without structural connections with other areas in the world. These regions are therefore expected to take advantage of their specialization, but they are also expected to be somewhat passive, because their good mix does not result from their behaviour, but more from the chance circumstance of being specialized in the right sectors at the right time. 2. global players. These are the regions whose role in globalization is maximum: they are structurally open and have all the connections; moreover, they are specialized in sectors which gain benefit from globalization itself. These regions are therefore those expected to be able to lead Europe and drive patterns of response to globalization also for the other regions of the EU. 3. pure gateways. These are regions whose behaviour is puzzling: they have structural connections other areas in the world, but at the same time they are specialized in closed 10

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

sectors. For this reason, they act as gateways to the world for neighbouring regions which are instead specialized in export sectors. 4. cut-off regions. This is the residual category made up of regions which have neither the structural elements to connect with the world nor the appropriate specialization in open growing sectors. These regions have no chance of playing a role in globalization processes and will hence be omitted from the analysis conducted in the chapters that follow.

Table 1. Regional taxonomy of possible globalization impacts Accessibility Index Location quotient in open growing sectors higher than 1 (right specialization) Location quotient in open growing sectors lower than 1 (wrong specialization)

Below average

Above average

1 Regional players

2 Global players

4. Cut-off regions (out of the analysis)

3 Pure gateways

When applied to European regions, the typology yields interesting results. There are many fewer global players (58 regions) than regional players, but the regions represented include all the major metropolitan and capital areas of Europe. Unlike regional players, these regions are mainly service areas: 29 out of 58 are specialized in service open growing sectors and 10 more are specialized in both service and manufacturing open growing sectors. Empirically, pure gateways appear to form an almost irrelevant category: in fact, only five regions belong to this group, confirming the theoretical intuition that this group is somewhat improbable. Closer inspection of the location quotients of these five regions shows that they are very close to 1 (normally above 0.95) for service open sectors, whereas they are much lower for manufacturing open sectors. As a consequence, these regions are classified among the service open global players, so that their number increases from 29 to 34. Interestingly, a minority but substantial group of global players are specialized in manufacturing open growing sectors, which signals that manufacturing is still important for globalization notwithstanding the economy’s increasing service trend. This group accounts for 14 regions out of 58; but if consideration is also made of those regions specialized in both service and manufacturing open growing sectors, it emerges that more than 41% of global players are manufacturing specialized.

4. Are the recent globalisation trends present in EU regional economies? Globalization trends in EU regions Globalisation trends emerge in Europe and especially in global regions ... The main findings of this analysis confirm that deep and decisive new globalisation patterns are ongoing in Europe. European regions exhibit a clear shift towards services activities. Employment has grown rapidly in services, whereas employment in manufacturing has grown much less, and has often decreased. This trend is more sizeable in global players, whereas for regional players and the other regions, the growth is substantial but to a lesser extent. 11

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

The shift towards services activities of the regional economies is much stronger in global players compared with regional players, and even more so compared with the other regions. We may therefore assume that the shift in the globalization process towards the services sector in the European economy is affecting global regions much more than other ones. The slow growth of productivity in services, which is the sector accounting for almost all employment growth, signals that there is an ongoing process of manufacturing restructuring, and that the services sector also acts as a channel of jobs creation. However, this increase is at the expense of productivity gains because it is taking place in low-skilled jobs. Productivity increases have been larger in global players, both in services (where employment has mostly grown) and manufacturing. Manufacturing productivity has also been growing rapidly in regional players specialized in open growing sectors, whereas productivity increases have been consistently lower in the other regions, signaling that, at least apparently, the globalization forces are drivers of productivity increases either because of technological transfer or because of the competition that forces the most open regions to react by increasing productivity. Total value added has been increasing thanks to productivity increases and despite employment decreases. Globalization forces may have played a role in this regard; despite the same increase in total value added, the pattern exhibited by the non-global other regions is completely different: much lower productivity increases and a maintenance or even an increase in employment levels. Non-global regions, de-specialized in open growing sectors, have therefore been the locations of lower manufacturing production phases, whereas the most open regions have had to shift to phases with higher value added and cut back and delocalize the lower phases. The patterns followed by services are much simpler: global players have been growing more rapidly in both employment and productivity, so that also total value added has been growing more in this group of regions than in any other group. Hence, global players appear to have benefited from their position and socio-economic structure, and they have developed their services activities to a greater extent, especially by capturing larger shares of the economy in this sector and the higher value-added functions. The other regions, by contrast, have experienced lower services employment growth restricted to the lowest value-added functions, those that can be more easily decentralized.

Global regions have local conditions which turn threats into opportunities, thanks to ... Globalization processes exhibit mega trends, each of which constitutes a major threat for regional economies. However, certain spontaneously occurring conditions may turn threats into growth opportunities, and these conditions are especially present in global regions.

12

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

... the presence of high-value functions... A first trend is the globalization of functions and tasks. A prime advantage expected to derive to the home area from this new production fragmentation is the natural selection of the most efficient local firms, functions and activities, with a consequent increase in productivity at aggregate level for the home areas. Some risks exist for areas on both sides of the process. Home areas, especially manufacturing regions, risk losing local core functions, local specific expertise, and skills that qualify local competitiveness and explain the role of the local economy in the international division of labour. This risk entails a particular threat to local economies: that of losing core functions which can be substituted by low value-added activities. Host regions compete strongly to attract both lowlevel job creating activities and high-value, productivity increasing, functions. Our expectation is that territories able to attract not only low-value but also high-value functions are those that gain the most in terms of economic performance. This applies to both manufacturing and service activities. Our empirical analysis shows that global regions have: -

-

a positive relationship between the presence of high value functions and the growth of service productivity. This positive effect on productivity is not expected to happen for medium-to-low value functions. In fact, the presence of SME managers is positively linked with services employment growth and negatively linked with services productivity growth. Medium-level functions seem to act on employment creation but do not achieve productivity gains ; no significant relationship between managers of SMEs and services employment growth, meaning that, in global regions, small and medium-sized enterprises are not the main channels of employment creation.

As mentioned above, the risk is that regional economies may become the containers of low–skilled activities with low value added. In fact: -

-

-

productivity growth in services is negatively correlated with the presence in the region of service functions of medium/low level, implying that the regions whose main functional specialization is in clerks have experienced a significantly lower services productivity growth than the others. This effect is somehow linked with globalization, because it is not present in the other regions, but is strong and significant in those regions directly affected by globalization (see Chapter 4); the relationship between the growth of services productivity and the growth of customer services clerks is negative and significant overall, but this is especially so within the global players group. This witnesses that increase in low tertiary activities leads to a decrease of tertiary productivity increases; the growth of tertiary employment has a strict link with the growth of low tertiary functions in global players, whereas the relationship is not significant overall or in the other regional groups, although it is nearly significant in the regional players. It seems that the more regions are exposed to globalization, the more they increase their low-value tertiary employment. This employment increase is not sufficient to compensate for the loss of productivity which accompanies low tertiary functions, and therefore GDP decreases.

... attraction of high-value service activities ... A second trend is off-shoring and outsourcing of service sectors and functions. The global shift in services offers large potential benefits for countries at both ends of the process: receiving countries 13

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

gain jobs, skills, access to foreign skills, while the sending ones improve their competitiveness by moving to higher-level activities. Since most of the off-shoring and out-sourcing has taken place among developed countries, this underscores that it does not primarily concern a “North-South” divide, and that it mainly affects regional economies in developed countries. In theory, the most advantaged areas in Europe, given this global shift, are urban specialized service areas, where the likelihood of finding capable local firms is higher. However, the advantages that a host local area can obtain depend on its ability to attract not only low value-added activities, like back-office functions or customer interaction services, but a balance between high and low-value activities in services. This can guarantee both an increase in the number of jobs and a positive effect on the local economy’s overall productivity. This likelihood is probably greater in urban service areas with high-skilled professions which attract high-value service activities and functions. The results of our analysis show the following. -

-

-

The concentration of medium-to-low services professions, measured by the ratio between their average regional share in the period 1996-98 and the growth of their share, shows a negative correlation overall, implying that those regions with a lower presence of these services at the beginning of the period have increased this presence more than the others. The increase of low-tertiary service activities in the economy takes place more in regions less specialised in services, most probably as a result of a loss of manufacturing activities. The risk of these regions to substitute manufacturing with low-value added tertiary activities is that these low-value added activities have lower productivity growth. They risk that the mere quantitative substitution of jobs is not enough to obtain a positive GDP growth. There is a negative correlation overall, meaning that regions specialized in lower phases are also increasing the skill content of their economic activities. However, this relationship is not significant in the other regions, witnessing the fact that the risk mentioned before of becoming a pure container of low-value activities exists for this kind of regions. Global and regional players, on the contrary, find a way to develop a catching-up process of their functions. By the same token, the share of blue collars, whose decrease represents the other side of the service increase, decreases more in the regions which initially had more of them, which completes the information provided by the evidence on white collars, which grow more in those regions with fewer service functions. However, the blue-collar pattern only applies to the global players and the other regions, whereas the correlation is not significant for the regional players. It is likely that the de-industrialization process is less strong in those regions best specialized in those sectors in which Europe is competitive worldwide. Finally, it is only among the global players that the concentration of business services is positive and significant, indicating that it is when regions are global that tertiary high-level manufacturing support functions concentrate in order to exploit economies of scale, whereas this is not necessary for those regions unaffected by globalization processes.

… substitution of manufacturing activities with high-value service functions. A third trend is deindustrialization and the advent of a service economy. Dealing with deindustrialization is a challenge for regions in a globalized setting. It is particularly important for regions not to lose manufacturing completely, but rather to keep the highest value-added phases and functions. The importance of manufacturing activities has two reasons: i) innovation trends take place at more rapid pace in manufacturing than in services, and the presence of manufacturing therefore guarantees a higher pace of innovation, which is the channel for real productivity increases; ii) if we deal with areas specialized in manufacturing, these have to avoid to lose their core competences, on which their competitive advantages rest. The maintenance of high-value 14

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

manufacturing activities is easier when regions find a balance between core manufacturing activities and other related activities, which often belong in the services sector. This effect is difficult to capture empirically, and in this section we only highlight the relationships between manufacturing activities and other related activities belonging to the services sector. In fact: -

-

-

There is a significant positive relationship between a region’s endowment of business services and the growth of manufacturing productivity, so that regions with more business service activities have been able to increase manufacturing productivity more, i.e. by keeping the highest phases. The relationship is significant overall, but it is higher and significant in the global players, which evidences that this factor is more important for regions with greater openness to globalization. The growth of business services is linked with the growth of manufacturing productivity. The relationship is particularly significant for global players. By contrast, it is not significant for regional players and the other regions, highlighting that, in dynamics, this process is even more closely linked with globalization trends. It is particularly significant in those regions specialized in business-related services, i.e. sector K (real estate, renting and business activities). The balance between services and industry is not enough if regions are unable to maintain a certain level of innovation. The relationship between the presence in the region of human resources in science and technology and the growth of manufacturing productivity is positive. It is evident that having a certain level of innovative capabilities is important for regional productivity growth overall, but especially so (higher correlation and higher significance) in global players.

5. What are the factors distinguishing between successful and unsuccessful regions? Within global regions, some benefit from globalisation… Notwithstanding their similar degree of exposure to the external world, the capacity of global regions to gain opportunities from globalization varies greatly. This result receives two main explanations in our analysis. The first concerns the fact that open regions are present in all European countries, and they are subject to strong national effects in their growth patterns. For example, the leading role of Eastern countries impacts on all regions in those countries, and not merely on open regions; in Western countries, similarities in growth trajectories are apparent among regions belonging to the same country. The second reason is linked to the clear evidence that global regions differ in their economic dynamics because of their different endowments of strategic assets like human capital, innovation capability, and high-level functions. Their different territorial specificities explain much of the difference among their performances.

… thanks to endogenous factors… Structural elements are at the basis of the different economic performances of regions, in the case of both global and regional players. As regards global players, those with a greater capacity to gain economic opportunities from globalization are characterized – with respect to the losing global players – by high-level functions measured by the presence of a large share of private managers and a large share of highly-educated people in the labour force. These features characterize those global regions able to increase both employment and productivity. Interestingly, the features which characterize winning regional players with respect to losing regional players are different from those of global regions. In particular, the capacity to grasp opportunities from globalization – and 15

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

therefore to record above-average employment growth as well as productivity gains – is, for regional players, closely correlated with a change in the structural mix of manufacturing sectors and specialization in medium-level functions. It therefore emerges quite clearly that regional players win if they avoid competing with the same weapons as global players, and instead find themselves an important role as leaders in medium-tech industries and functions.

... and to winning strategies/behaviours. Among global regions, our analysis has identified a group of regions that perform relatively better in terms of GDP growth. These are labelled “benefiting regions” because their relative competitive position improves. Our analysis shows that the economies of these regions share certain strategic behaviours. They avoid a simple quantitative substitution effect between industrial and service jobs; they accompany an increase in service jobs with an increase in private service productivity; they deal with a negative increase in industrial employment by increasing industrial productivity; they are able to attract FDI to their sector of specialisation; and they are endowed (especially in Western countries) with a relatively high share of command and control functions. Evidence for all these findings is provided by both the quantitative and the qualitative (case studies) analyses. In a purely inductive way we infer from our analysis that the benefiting regions are those that adopt winning strategies, namely: - increasing manufacturing productivity in the same sectors of specialization. This is achieved by means of new technologies, organizational and managerial innovation, or, in some cases, corporate adaptation, especially vertical integration (with suppliers and customers) and horizontal integration (with similar firms in order to achieve economies of scale). This strategy protects and supports manufacturing employment growth. - reconversion of regions to higher phases of the production process, i.e. decentralizing low-level production phases to areas with lower wages and production costs. This strategy preserves the region’s specialization (especially in terms of value added), generally at the expense of job losses. - reconversion of the regional sectoral structure from low value-added sectors to high valueadded ones, moving to high-level service activities, and avoiding the mere quantitative substitution effect between manufacturing and service jobs. Confirmation of these observations was provided by a more in-depth and qualitative analysis, i.e. by using a case-study methodology.3

6. What is the future of the European regional economies in light of the different globalisation strategies that may emerge over the next twenty years? A prospective analysis Different alternative globalization patterns can be envisaged… The prospective analysis now developed is based on alternative competitive strategies pursued in the future by three main blocks of countries: the EU member-countries and the emerging countries, taking account of the European Commission’s strategies as a third discriminating element. The interplay among these strategies will depict different scenarios with which European regional economies will have to deal. 3

See chapter 7 for a summary and Volume 2 of the final report for the detailed case studies. 16

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Two scenarios will be built by taking the two above-mentioned possibilities into consideration and by developing forward-looking analyses for European regions, assuming opposite strategies adopted by the emerging countries: a riskier, and probably more expansionary strategy based on determination to compete on a world scale by undertaking strong and decisive internal restructuring, reconverting and modernising processes; and a more protective strategy whereby BRIC reinforce present tendencies. Comparison is made of the future opportunities and risks of growth for European regions. A further element which cannot be omitted from a prospective scenario is the economic crisis that, since August 2008, has heavily affected the economic system. It is impossible to say when and how this negative cycle will end, but what can be done is to construct scenarios under different assumptions concerning the speed of recovery from the present recession of the world economy. We construct the following two scenarios (see section 8.1 of the report): -

-

a scenario combining a reactive strategy by the EU 27, a modernising strategy by BRIC, and an excellence-based EU policy developed in a general economic setting of a short-term recovery from economic crisis (expressed in Fig 8.4 by a light colour for the setting). In this scenario the recovery from the crisis in the short term obtained through efficient Keynesian macroeconomic policies makes it possible to move to excellence-based policies thanks to a new wave of private investments that takes place after the end of the crisis; a scenario based on opposite strategies – a defensive EU27, a price-competitive strategy by BRICs, and a cohesive policy of the European Commission – developed in a general economic setting of recession and deep long-term crisis (expressed in Fig 8.4 by a dark colour), in which the consumption of durable goods is strongly depressed. Macroeconomic policies remain Keynesian throughout the scenario period; structural policies are oriented to cohesion aims in order to limit the effects of the crisis.

… which give rise to different national growth patterns. The courageous strategy scenario is overall more expansionary, both for the Old 15 and for the New 12 countries of Europe, with the former benefiting slightly more than the latter. The earlier exit from the crisis and the more aggressive strategies of EU countries pay off. Overall, the growth rate of New 12 member countries remains more sustained with respect to that of Old 15 countries, implying that for these countries an aggressive modernizing strategy enables them to be competitive, with exit from the crisis. As far as manufacturing is concerned, scenario A is not substantially different from the baseline, since it is assumed that manufacturing cannot be the main driver of growth for Europe, not even in the most positive scenarios. The sector driving growth is on the contrary the service sector, with considerably higher employment growth rates both in the East and the West. Job creation in services is consequently what can be expected in scenario A, and it helps explain the positive GDP performance (see Table 8.2 in the report). The protectionism scenario is less expansionary for all the EU 27 countries owing to the delayed end of the economic crisis and the defensive strategies implemented by the European countries. In this scenario, too, the New 12 member-countries outperform the Old 15 in GDP growth, but less substantially so with respect to the other scenarios. In fact, it appears that the Eastern countries following a cost-competitive defensive strategy suffer from the crisis even more than the Western countries do. As far as employment is concerned, de-industrialization in this scenario is accelerated with respect to the baseline in all Europe, and more so in the New 12 countries than in the Old 15 countries, because the former need to restructure their economies to withstand competition, and in this case they fail to do so. Also as regards services, this scenario is more restrictive with respect to 17

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

the baseline, with lower employment growth rates in all Europe, and particularly in the New 12 countries, which suffer most from the crisis owing to their failure to modernize their economies (see Table 8.2 in the report).

At a regional level, global regions show their vital role for the future of the EU economy… A prospective analysis shows the vital role of global regions for the future of the EU economy: in a scenario of courageous strategies – able to drive European countries towards a knowledge-economy profile and to modernise economies in emerging countries – and of relatively rapid recovery from the present economic crisis, global regions record the highest GDP growth rates, and among them are those at present benefiting from globalisation. An opposite scenario of protectionism with slow recovery from the economic crisis shows that global regions are able to lose less than the other regions and maintain their relative role of leadership in the EU economy (see Table 8.3 in the report).

… and owing to the economic downturn, regional disparities increase. An important result from the prospective analysis is that, overall, between the initial simulation period and 2020, disparities in the EU27 increase, doing so mainly because of the economic downturn, which strongly hits both core and peripheral regions, the latter having less capacity of reaction. The economic downturn annuls the advantages of rapid processes of inter-national catching-up (mainly by New Member States), with a slight decrease in intra-national disparities. But this general trend is generated by two different evolutions: interregional disparities will increase because of a markedly widening intranational dualism between core and periphery regions insufficiently counterbalanced by slowly decreasing inter-national disparities. In all cases, intranational disparities are likely to widen. Overall, considering that a new phase of continental integration is beginning (and that our starting year for long-term comparison is 2005), Williamson’s prediction may come true – once again (see section 8.5 in the report). These tendencies are reinforced in the scenario with rapid recovery from the economic downturn mostly due to the capacity of strong areas to take advantage of the economy’s revitalization. The slow recovery scenario instead records a lower increase in regional disparities, probably due to the pervasiveness of the negative effects of the crisis in all sectors and activities in all strong areas. A “two-speed growth model” is an appropriate expression to denote the way in which European regional growth will develop in the future, and it indubitably calls for appropriate regional policies (see Fig. 9.2).

7. Regional policy implications The empirical analysis yields the following important policy implications. A) Given the good economic performances of global regions, regional policies should first and foremost support regions in opening their local economies so that they can exploit the advantages deriving from an integrated world economy. Policies should first be addressed to those regions already endowed with activities belonging to open sectors but which lack international attractiveness. Regional areas, as they are labelled in this report, are the most suitable regions for the application of such policies in order to increase their attractiveness to international headquarters, 18

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

high-quality human capital and specialized labour through local quality-of-life interventions, as well as through fiscal and labour costs incentives and efficient territorial policies. B) Regional policies should seek to reinforce the winning strategies of local economies by avoiding generic assistance policies in support of employment, and by developing job-creating policies only in those sectors and activities that show a capacity to increase productivity growth. The latter is more likely to take place either in sectors in which the region is specialized, where core competencies and specific local assets are present, or in high-value sectors like advanced services. Figure 1 depicts these two regional trajectories; the first one (a) envisages a development from an increase in employment growth in indigenous activities to an increase in productivity growth in these sectors, while the second one (b) highlights a trajectory of increased productivity growth in new sectors and activities, followed by increased employment in the newly-established sectors. Both trajectories end with a situation comprising a higher productivity growth rate and a higher employment growth rate. However, trajectory (b) requires the decisive rationalization of local activities and a regional policy supporting the creation of new high-value sectors and activities through restructuring of the local economy. Trajectory (a) is rooted in indigenous activities and preserves local core competences. For this reason, between the two policy trajectories, (a) is the best strategy to put in place when an implicit related risk is overcome. The risk of this strategy is in fact that a higher increase in employment will be achieved in indigenous activities, thanks to assistance policies, with no effort made to achieve higher growth rates in indigenous activities, so that a move to the upper quadrant of Fig. 9.4 is prevented. As our empirical analysis shows, a choice of this kind does not generate the competitive conditions for a local economy to withstand fiercer international competition. C) The move of a local economy from a simple policy of increased employment growth rates to productivity growth rates in indigenous activities is relatively simple if that local economy is specialized in capital-intensive advanced sectors and activities. When, instead, indigenous activities are related to labour-intensive sectors, regional policies must be devoted to facilitating a transition to higher-efficiency functions and activities, enlarging the filière to include advanced complementary activities (e.g. design, product innovation, marketing) located upstream and downstream from the core production. Many success stories can be found in the real world; in our case studies, the capacity to move upstream of the production chain and specialize in knowledgeeconomy activities has been the key to the success of local economies like Stockholm and Barcelona. Fig. 1. Possible regional trajectories

Productivity growth rate

b a

+

-

-

+

Growth rate of employment in indigenous activities

19

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

D) Increased productivity growth rates can also be achieved through local policies supporting innovation; as our empirical analysis shows, it is not merely the degree of R&D expenditure and of technological development that is linked to winning strategies. Innovation has to be understood as all efforts devoted to increasing knowledge so as to foster sectoral transformation, develop local capabilities to cooperate synergically with other regions, and invent new organizational solutions at both the firm and public governance levels. E) Whilst the above are suggestions arising from our empirical analysis on the past, our prospective analysis suggests some policies with which to support local economies amid the present economic crisis. Keynesian, demand-oriented policies are necessary to tackle the present economic downturn, and they should be put in place by national governments in the most efficient way possible. Our analysis shows that macroeconomic policies have different impacts at the regional level, and that in any case economies will emerge from the crisis with a “two-speed growth model” that will reinforce regional disparities. The economic downturn annuls the advantages of rapid processes of inter-national catching-up (mainly by New Member States), with a slight decrease in intra-national disparities (see Table 9.2 in Chapter 9). But this general trend is generated by two different evolutions: interregional disparities will increase because of a strongly widening intra-national dualism between core and periphery regions insufficiently counter-balanced by slowly decreasing inter-national disparities (see Fig. 9.2 in Chapter 9). Intra-national disparities are likely to widen. Overall, considering that a new phase of continental integration is beginning (and that our starting year for long-term comparison is 2005), Williamson’s prediction may come true – once again. These tendencies are reinforced in the scenario with rapid recovery from the economic downturn mostly due to the capacity of strong areas to take advantage of the revitalization of the economy. The slow recovery scenario, instead, records a lower increase in regional disparities, probably due to the pervasiveness of the negative effects of the crisis in all sectors and activities in all strong areas (Fig. 2). A “two-speed growth model” is an appropriate expression to denote the way in which European regional growth will develop in the future. Fig. 2. Differences in the evolution of per capita income disparities among scenarios, 2005-2020

Another important explanation for the need of regional policies is that our empirical results show that the capacities of reaction to the economic downturn differ even among global regions; within this group, the cluster of benefiting regions is the one able to perform best in each scenario (Table 2). These results express two major messages: a) there are strong regions which, like the conservative regions, require particular help in turning their potentialities into real growth 20

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

opportunities; b) some sort of cumulative reinforcing learning process takes place in governance and public policies, as witnessed by the constantly good performances of benefiting regions.

Table 2. Average annual GDP growth rates between 1996 and 2020

Baseline Scenario A Scenario B

Benefiting regions

Catching-up regions

Conservative regions

2.75 3.74 2.01

2.77 3.85 1.19

2.03 2.99 1.28

Global regions will perform better than the others by recording either higher growth rates in a scenario of rapid recovery or a smaller economic downturn in a scenario of a long economic crisis. In this framework, structural, long-term, supply-side policies must complement public investment to limit the territorial asymmetries caused by the recession period. In global regions, structural policies must be devoted to reducing the social costs of the crisis, while in weak areas they must support the few indigenous activities that are present (like entrepreneurship, innovation capabilities). This also applies to the EU’s cohesion policies, which must be devoted to limiting the regional disparities generated by the economic downturn by complementing public investments to foster endogenous capacities, and by implementing “smart investment policies” designed to maximize the efficiency of returns on investments. F) The integration of short- and long-term policies managed at different governmental levels is a key to long-run success. Sectoral regional long-term policies must be devoted to increasing efficiency, innovativeness, and competitiveness in the long run, in those sectors where short-term policies are intended to increase consumption. Public incentives in the automobile sector, for example, should be coupled with structural policies oriented to the development of hydrogen cars. Likewise, structural policies should support social reforms with an impact on internal demand. In Eastern countries, pension reforms could allow precautionary savings to turn into internal consumption. G) All regional structural policies must be based on the uniqueness of places by reinforcing the assets and core competences already present in the area. Turning Territorial Diversity into Strength (the title of the Green Paper) summarizes this notion as stated some fifteen years ago in a similar terms (“turning specificities into assets”) to synthesize the idea that regions should build their competitive advantages from below, investing in their particular competences, resource endowments and even weaknesses (such as inaccessibility, if this means conservation of an unspoilt environment), renovating their productive ‘vocations’ and cultural assets. This is still valid today; and even more so in an era of globalization in which local specificities help regions to emerge and compete in a global economy. Regional policies should help regions not according to their needs, but according to their innovative ideas concerning their uniqueness. H) From this it follows that different policies should be devoted to global and regional players, because their degree of success in globalization depends closely on different strategic assets. Global players must be helped to position themselves at the highest levels of the international division of labour by attracting technologically highly advanced functions and activities in both industry and service sectors. By contrast, regional players should not be forced to play the same role as global players; they do not possess the right structural assets to compete with global players, and a direct 21

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

competition with them would be at a loss. In the case of regional players, by contrast, policies should be designed to help them play an important role in the international division of labour, becoming the most attractive areas for medium-tech activities and functions both in industry and services. A way to keep competition in medium to low functions into consideration, a certain level of innovative capacity is necessary. This helps regions to renovate their manufacturing activities (product innovation), to maintain their core competences and use them to achieve and conquer market niches. I) As a final remark, our empirical results bear witness to the importance of FDI for regional development. This holds not only for Eastern countries but also for Western ones. The case of London’s outperformance is emblematic in this regard. Regional policies should be oriented to the attraction of multinational companies. Especially in regions with clear identities and productive vocations, incentives to attract FDI should be oriented to the filière, reinforcing and enlarging already-present core competences to encompass upstream and downstream value-added activities. Globalization is a process that cannot be avoided; regional policies should help local economies to turn threats into opportunities, and to put winning strategy trajectories in place.

22

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

1. Globalization and the reshaping of regional economies 1.1 The new globalization paradigm Globalization is not a state of the world but an evolutionary process: the increasing planetary integration of markets for goods and services (« internationalization »), markets of location for economic activities (« multinationalization »), markets of such production factors as technologies and information. The three processes are strictly interwoven, and reinforcing each other in a cumulative way, highly facilitated by progress, and cost abatement, in transport and communication activities. In fact, the nature of international trade has evolved from pure exchange of final goods among national production systems to exchange of intermediate goods and components within production networks organized on a world-wide scale.4 Local production systems find themselves increasingly tied together and interdependent, mainly through the global strategies of multinational corporations. Just to give some quantitative indications: since almost thirty years, international trade has been steadily growing at a rate which is the double of world GDP (and three times after 2000); foreign direct investments (FDI), on their turn, have grown - up to the slump of 2000/01 due to the bursting of the dot.com bubble and 9/11 - at rates that are the double with respect to international trade, and four times those of world GDP5. Trade of intermediate goods have steadily increased over the last ten years in the EU both within and outside Europe (Fig. 1.1). While the phenomenon of globalisation is not new in nature, what is new in recent times is the form through which this takes place, recently labelled as the new globalisation paradigm6. Before this new paradigm took place, firms and sectors were the finest level at which globalisation’s impact was felt: more open trade explained the fortune of some firms and sectors, exploiting economies of scale in the production of final products in low-wage areas. Nowadays, competition generates effects increasingly inside the firm, at the functional and task level, on a task-by-task basis, calling for and imposing new competitive strategies to firms based on a strong reallocation of phases of production at the spatial level. Off-shoring and out-sourcing of phases of production become usual corporate strategies, through which firms look for productivity gains. All this inevitably leads to an increasing complexity in spatial terms: “there is undoubtedly something profoundly new about new globalization”7. The North/South, Centre/Periphery model of the past - implying moving activities and employment towards low-cost countries and exporting low-value added manufacturing goods from these countries – was a model which could rightly respond to competition on a firm-to-firm basis, looking for competitive advantage based on low labour cost. 4

See, among others, Godart O., Görg H. and Görlich D. (2009), “Back to Normal? The Future of Global Production Network After the Crisis”, Kiel Policy Brief, n. 9; Abraham K. G. and Taylor S. K. (1996), “Firms’ use of outside contractors: Theory and evidence”, Journal of Labor Economics 14, 3: 394-424; Amiti M. and Wei S. J. (2005), “Fear of service outsourcing: is it justified?”, Economic Policy 20, 42: 308-347; Feenstra R. C. (1998) “Integration of Trade and Disintegration of Production in the Global Economy”, Journal of Economic Perspectives, 12(4): 31-50. 5 Sources : OECD, Economic Outlook, dec., various years ; UNCTAD, World investment report, various years ; World Bank, World development report, various years ; Camagni R. “Policies for spatial development”, in OECD Territorial Outlook, Paris, 2001, 147-169. Clark G. L., Palaskas T., Tracey P. and Tsampra, M. (2004), “Globalisation and Competitive Strategy in Europe’s Vulnerable Regions: Firm, Industry and Country Effects in Labour-Intensive Industries”, Regional Studies, Vol. 38, n. 9, pp. 1085-1100. 6 Baldwin R., Globalisation : the Great Unboundling, Prime Minister’s Office, Economic Council of Finland, 2006. 7 Kenney, M. and Florida R. (2004), Locating global advantage: industry dynamics in the international economy, Stanford University Press, Stanford Ca., Preface. 23

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Nowadays, this is only a simplifying picture of real processes. In fact, the new globalisation paradigm calls for complex firms strategies, trying to optimize along multiple trade-off: between labour costs and temporal proximity to customers; between transportation and inventory costs; between access to knowledge sources, to skilled labour and unskilled labour; between commodification of existing products and necessity to shorten product design and time-to-market. New cross-border relationships emerge, reflecting the fact that firms do not look only for labour cost advantages, but require additional and more value added local assets, like knowledge, creativity and entrepreneurial spirit, flexibility of the labour markets (rather than cost), presence of relational and social capital8. Fig. 1.1. Volume of intermediate goods trade 2000-2009

The new globalisation paradigm is also reflected in the new spatial trends of FDI. Most of these investments are directed towards developed countries (80% in the years 1986-90, around 60% in years 1993-97 and 65% more recently in 2006) and look particularly attracted by accelerations in economic integration processes: in fact, EU15 countries, at the top of the process of creation of the Single market in 1991-92, received up to 50% of world FDI, and similar accelerations were evident in the case of Eastern European countries after the accession. Secondly, services since 1990 accounted for the majority of total FDI; by 2005 they accounted for almost two thirds of the total, while manufacturing for 30% and primary sectors for less than 10%. Services still keep a relevant share in greenfield FDI (42% in 2006, with manufacturing accounting for 54%), greenfield FDI representing 1/3 of total FDI.

8

Camagni R. (2008), « Towards a concept of territorial capital », in Capello R. et al., Modelling regional scenarios for the enlarged Europe, Springer Verlag, Berlin. 24

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

All this means that a huge part of cross-border relationships regard directly advanced countries, and, within them, core and metropolitan regions where the service sector locates preferably. Manufacturing industries too, though threatened by changes in trade patterns and multinational location decisions, still find in advanced and EU countries favourable contexts for subsistence and development, at the condition of renovating industrial and territorial strategies.9 Territorial concentration and clustering, focalization on top quality products and segments, engagement in continuous technological and organizational innovation. A recent OECD Report shows how the supposed de-industrialization process is often only the result of industrial restructuring and productivity increase more that the effect of trade challenges; ten over the eleven top manufacturing countries in the world are developed countries, and five belong to EU-15 (Germany, France, UK, Italy and Spain)10. Between firms, industries and regions a complex interplay is emerging, where firms and industries choose - and create – “places”, but also places and regions attract and develop economic activities thanks to their factor endowment. In the case of advanced countries, these factors may be labelled as “the new territorial capital”: accessibility to large markets, agglomeration economies, presence of knowledge, creativity and entrepreneurial spirit, flexibility of the labour markets (rather than cost), presence of relational and social capital. Regions - considered as sort of collective agents, implicitly or explicitly defining specific development trajectories – compete in the global economy, building on their historical strengths and identifying opportunities for diversification and enlargement of their specialisations by strengthening their know-how and knowledge base11. This report assumes the point of view of regions (in the advanced European Union), and addresses the general question of which regions gain and which lose in a globalised world. The main idea is that a win strategy is to adapt regional economies to changes, by helping major global trends to take place locally in a way that guarantees to turn threats into opportunities of growth. How this can be done is a subject matter of this research.

1.2 Globalization and the new opportunities for regional economies As mentioned before, globalisation is an evolutionary process, which has called profound changes in the spatial organisation of production over the last twenty years, affecting drastically regional economies. The influence of globalisation patterns was no longer simply affecting the division of labour between emerging and advanced countries, developed and developing economies. It is 9

On the relocation of industrial activities, see Kenney, M., and Florida, R. (2004), Locating global advantage: industry dynamics in the international economy, Stanford University Press, Stanford Ca., Preface; Marin, D. (2006), ‘A new international division of labor in Europe: Outsourcing and offshoring to Eastern Europe’, Journal of the European Economic Association 4, 2-3: 612-622; Storper M., Chen Y. And De Paolis F. (2002), “Trade and the Location of Industries in the OECD and European Union”, Journal of Economic Geography, n. 2, pp. 73-107; Tavares, A.T. and Teixeira A. (2006), (eds.) Multinationals, Clusters and Innovation, Palgrave Mc Millan, New York; Wai-chung Yeung H. (2009) “Situating Regional Development in the Competitive Dynamics of Global Production Networks: An EastAsian Perspective”, Regional Studies, Vol. 43, n. 3, pp. 325 – 351. 10 OECD (2007), Globalisation and regional economies, Paris, ch. 1 and 2; Dunford M. (2003), “Theorising regional economic performance and the changing territorial division of labour”, Regional Studies, v. 37, pp. 839-854; Gardiner B., Martiner, R. and Tyler P., (2004), “Competitiveness, Productivity and Economic Growth across the European Regions”, Regional Studies Vol. 38, n. 9, pp. 1045-1067. 11 Camagni R. (2002), “On the concept of territorial competitiveness: sound or misleading?”, Urban Studies, n. 13, 2395-2412. 25

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

nowadays clear that globalisation trends affect the division of labour of sub-national economies within advanced countries, and that competition increases among regional economies to grasp the opportunities offered by globalisation. Regional competition becomes even a more critical issue if one reminds that the ricardian law of comparative advantage does not hold in case of comparison among local economies (inter-regional trade), and consequently the ricardian conclusion that each country will always be granted some specialisation and role in the interregional division of labour is not valid for regions.12 In fact, a region can well be pushed “out of business” if the efficiency and competitiveness of all its sectors is lower than that of other regions. In fact, at the inter-regional level the two equilibrating forces that in principle allow passing from an “absolute advantage” to a “comparative advantage” regime either do not work properly or do not exist. The “classical” equilibrating process relies on downward flexibility of prices and wages, which is widely hampered by the existence of national wage contracts in both private and public structures and by the homogeneity of import prices (we remind that regions are very open economies). The second, “modern” process relies on the devaluation of the currency, and it is automatically excluded in an inter-regional context (Camagni, 2002). Therefore regions compete on the basis of an “absolute” advantage principle, and, whenever noncompetitive, they do not have any automatic mechanism on which to rely in order to maintain some export specialisation (and to pay for their imports). Their fate is, in this case, mass unemployment and, in case of insufficient public income transfers, emigration and possible desertification. Globalisation is an unavoidable process for regional economies due to its pervasive nature. Globalisation influences emerging and developed, old and new economies, areas specialised in advanced and traditional sectors, in high and low value-added activities and tasks. Its pervasive nature suggests that the best strategy for regional economies to face globalisation is their adaptation to change (Fontagné and Lorenzi, 2005). Postponing such a strategy entails high risks for a local economy, obliged to play in the world competition as a late comer with respect to its competitors. Late comers remain more and more locked in old forms of production organisation, with limited complementary skills developed for an adaptation strategy, and losing in the meantime some important opportunities offered by the globalisation processes. Globalisation is not only a source of threats. It is first and foremost a source of opportunities displayed to national and regional economies. First of all, there are evident positive macroeconomic effects.13 From the consumers side, a traditional advantage (in terms of increasing purchasing power in real terms) stems from international trade between advanced and emerging countries, with a consequent positive effect of a limited inflation rate in advanced economies. From the production side, outsourcing and off-shoring in low-cost areas allow final goods to be sold at low prices for the imported country; this means that firms are able to import intermediate goods from developing countries at lower prices, obtaining productivity gains.14 One part of this productivity gain is reflected in wage increase, and one part in lower relative prices of 12

For a in depth explanation of why the Ricardian law does not hold at regional level, see Camagni, op. cit.; Capello R. (2007), Regional Economics, Routledge, London, chap. 6. 13 Mankiw, G. N., and Swagel, P. (2006), ‘The Politics and Economics of Offshore Outsourcing’, Journal of Monetary Economics 53, 5: 1027-1056. 14 See among others, Baldwin R. (2006), “Globalisation: the great unbundling(s)”, Paper for the Finnish Prime Minister’s Office, Economic Council of Finland as part of EU Presidency; Bhagwati J., Panagariya A., and Srinivasan, T. N. (2004), ‘The muddles of outsourcing’, Journal of Economic Perspectives 18, 4: 93-114; Feenstra R. C. and Hanson G. (1999), ‘The impact of outsourcing and high-technology capital on wages: Estimates for the United States, 1979-1990’, Quarterly Journal of Economics 114, 3: 907-940; Grossman, G. M. and Helpman, E. (2005), ‘Outsourcing in a global economy’, Review of Economic Studies 72, 1: 135-159; Grossman, G.M., Rossi-Hansberg, E. (2008), Trading Tasks: A Simple Theory of Offshoring, American Economic Review, 98-5, 1978-1997. 26

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

manufacturing goods, in favour of competitive gains and wider final consumption. Secondly, outsourcing and off-shoring of production in developing areas increase the purchasing power of these countries, which become a growing potential market for the final production of developed economies. Besides macroeconomic effects, one important impact of a spatial reorganisation of production is a natural selection among firms in favour of the most efficient ones. In spatial terms, this selection takes place especially in those areas where the industrial tissue does not renew itself, where the mobility of resources is limited by the rigidity of the economy, where investments are weak, where technological change at the world scale is not captured by local R&D activities or by a local absorption capacity of external R&D, where advanced services are present in a limited way, where skills and know-how remain locked into an outdated production organisation. For all these reasons, the best way to face globalisation is, as mentioned before, a gradual adaptation to change. Globalisation is raising the competitive climate within which firms are confronting each other. In order to cope with this condition, and with the consequent increasing level of dynamic uncertainty (about markets, technologies, successful organisational models), firms more and more rely on high quality human factors, accessibility to information, devices or “operators” allowing fast information assessment and transcoding, and forms of coordination and cooperation. As a consequence, directly or indirectly, through explicit locational decisions or through the selective effects of competition, they favour and support those territories that supply these new “relational” factors. Territories, understood as collective actors, may help firms to be competitive, enhancing the presence of these new, strategic production factors, bringing benefits to their “stakeholders” (local populations). In this sense, we can say that regional economies compete with each other, no other automatic device being in a measure to assure long term development and wellbeing (Camagni, 2002).

1.3 Benefiting regions: a definition Globalisation raises the competitive climate among European regions; some of them will be able to benefit from increased competition by enhancing their relative position in the European economy. This project is devoted to highlight under which structural conditions local economies are able to increase their relative position. In this project benefiting regions are defined as those regions that maintain and even reinforce the competitive position in the European economy thanks to globalization processes, i.e. those regions that are able to increase their relative production capacity (measured in terms of GDP) more than the European average in a period of globalisation. This can be the result of different economic growth conditions: d. employment growth takes place both in high and low value added functions, the former having a higher effect in quantitative terms than the latter. If this is the case, both employment and productivity increase, and therefore GDP; e. employment losses take place in low value added activities and are more than compensated in high value added functions. In this case employment losses are more than compensated by productivity increases, and GDP increases; f. employment increases in low-value added functions, accompanied by a limited loss, if any, in high skilled jobs. GDP increases despite the loss in productivity.

27

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

If we put the indicators of relative productivity and relative employment growth, for statistical reasons the 45° negative slopped line represents the situation in which regional GDP growth equals the European average GDP growth (Figure 1.1). Regions that position above the 45° negative line register a higher than average GDP growth. This competitive situation is achieved through different dynamics of productivity and employment dynamics, as explained by the three cases mentioned above. The identification of regional growth patterns for each European region and the structural regional elements explaining these performances are the subject matter of this project (chapter 7). The relationship between regional growth patterns and globalization trends can be measured only indirectly, by identifying regions where globalization impact is felt first and more strongly (chapter 3). Figure 1.1. Regional growth patterns Case b

Relative regional productivity growth Case a

Relative regional employment growth Case c

Regional GDP growth = European growth

Legend: Case a: increase of both low and high value added jobs. Case b: increase of low to high value added jobs, losing a relatively low number of skilled jobs. Case c: increase of high to low value added jobs, losing a relatively high number of skilled jobs.

1.1.

Globalization trends and the new threats/opportunities for regional economies in the EU

Globalization processes develop following some mega trends, each of which presents a major threat for local economies nowadays. If these trends develop under certain spontaneous conditions, they can turn into growth opportunities. Major trends are here analyzed; they are associated to: a) the major threats they embed; b) the conditions to turn them into opportunities and c) the typologies of regions that are expected to be favored by these trends (Table 1.1). The conditions to turn threats into opportunities are those who guarantee that the competitive productive capacity remains higher than the European average, i.e. that regional economy do not lose productivity gains or that productivity losses are more than counterbalanced by employment growth.

28

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Table 1.1. Globalisation trends: threats and opportunities to regional economies Globalisation trends

Threats

Conditions to turn threats into opportunities

Favoured regions

a)A new globalization paradigm: - Relocation of functions and tasks instead of sectors.

Increase of low value manufacturing functions only.

Attraction of high-value manufacturing and service activities.

Regions with high command and control functions. Regions already highly specialized in high-value service and manufacturing activities.

b)The next global shift: - Increase of service offshoring and out-sourcing instead of manufacturing off-shoring.

Increase of low value added service activity.

Attraction of high-value service functions.

Regions with high command and control functions. Regions already highly specialized in high-value service activities and endowed with good human capital.

c)Deindustrialization and the rising of service off-shoring and outsourcing: - Re-orientation of production towards new growth industries and activities. Deindustrialisation in favour of services in manufacturing areas

d)Decentralization of inter-twined functions (manufacturing and related services)

Loss of core activities.

Loss of innovation from industrial activities.

Attraction of low value added activity with limited R&D capacity.

Attraction of new value added activities related to core activities.

Attraction of new value added activities related to core activities.

Attraction of intertwined production/R&D functions.

Regions already highly specialized in manufacturing with highlevel functions.

Regions with a certain level of innovation activities. Regions with a certain level of high value functions and innovation rate.

a) a new paradigm: globalization of functions and tasks Globalization patterns are nowadays much more than simply moving employment and activities from developed to emerging areas; competitive gains are based on intra-industry reallocation of resources more than on inter-industry reallocation. As said in section 1.1, corporate strategies are much more complex, trying to combine different goals at the same time, and to achieve multiple trade-off: between labour costs and temporal proximity to customers; between transportation and inventory costs; between access to knowledge sources, to skilled labour and unskilled labour; between commodification of existing products and necessity to shorten product design and time-tomarket. 29

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Corporate strategies are therefore based on a horizontal division of labour of their activities, rather than a vertical division of labour, externalising (outsourcing) or re-locating (pure off-shoring) in other international location some of their functions, notwithstanding the position of each function in the value chain (Fontagné and Lorenzi, 2005) (Table 1.2). This brings to a fragmentation of production in both manufacturing and service activities. Location strategies are much more complex than the simple aim to achieve low-labour cost; they respond to different and wider aims, like proximity to the final market, decrease of production costs, subdivision of the value chain in search for higher specialization of functions, within each single foreign affiliate, or through the externalization of some functions to highly specialized local firms.

Table 1.2. Modern forms of organisation of production chain Internalised production Home country

Domestic production within a firm through increasing production efficiency and/or decentralisation to a different location

Foreign country (off-shoring)

Production by foreign affiliates (intra-firm trade)

Externalised production (out-sourcing) Production out-sourcing to a domestic firm

Production by foreign local firms (intra-industry trade)

Source: Our elaboration on World Investment Report, 2004, p. 148

a’) threat/opportunities of the new globalization paradigm A first expected advantage for the home area of this new production fragmentation is that of a natural selection of the most efficient local firms, functions and activities, and a consequent increase of productivity at aggregate level for the home areas.15 Some risks exist for areas at both sides of the process. Home areas, especially manufacturing regions, risk to lose local core functions, local specific know-how and competences that qualify local competitiveness and explain the role of the local economy in the international division of labour. Productivity and job losses are highly probable in a scenario of this kind. For host regions, a strong competition takes place for attracting both low-level job creating activities, and high-value, productivity increasing, functions. Our expectation is that territories that are able to attract not only low-value tasks but also high-value functions are the ones that gain the most in terms of economic performance. This is true for both manufacturing and service activities.

b) the next global shift: off-shoring and outsourcing of service sectors and functions

15

Dunford M. (2006), “Industrial Districts, Magic Circles, and the Restructuring of the Italian Textiles and Clothing Chain”, Economic Geography, January 2006, v. 82, iss. 1, pp. 27-59; Humprey, J. and Schmidz, A. (2002) “How does insertion in global value chains affect upgrading in industrial clusters?”, Regional Studies, Vol.36, pp.1017-1027. 30

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Nowadays, globalization more and more takes place in the service sector, through both off-shoring and out-sourcing strategies. FDI in service sectors has been underway for some time, but it has assumed new dimensions and patterns since the 1990s. The world’s import stock of services quadruplicated between 1990 and 2002, from an estimated 950 billion US dollars to over 4 trillion US dollars (World Investment Report, 2004). This explosion is certainly linked to the liberalization of FDI policies, which began around mid 1980s and gathered momentum during the 1990s. All this entails important consequences if one thinks that services are the largest productive sector in most economies, and their competitive (and efficient) production is critical to the welfare of a society as a whole. The growth of service FDI has gone hand in hand with the industry mix of such FDI. Until 1990s, services FDI was concentrated in trade and finance, accounting for 25% and 40%, respectively, of total inward FDI stock in service (World Investment Report, 2004). Since the 1990s, other services have seen more dynamic FDI growth, among them are telecommunications and electricity, water supply and business services. This increasing tendency in service off-shoring is seen as a major trend over the next years, if one thinks that service off-shoring is, compared to manufacturing offshoring, simpler in terms of resources, space and equipment requirements, and might therefore be more footloose given the lower sunk costs involved; it affects firms in all sectors, and may therefore have higher implications for the host economy than the fragmentation of manufacturing; it affects mainly white-collar workers, while manufacturing off-shoring involves primarily blue-collar workers and generally creates jobs in the host area without destroying them in the home area. This change of mix of service industries reflects also different reasons behind off-shoring activities. Finance and retail trading represented traditional host-country market oriented services; nowadays, more complex strategies are put in place, in search for efficiency gains based on inter-affiliate division of labour, relying on foreign affiliates producing components – not necessary for their parent firms, but for other affiliates that specialize in other components. Therefore, also in service we assist to the breaking up of service activities into components that are produced wherever it is more convenient to do so, with the result that certain foreign affiliates perform back-office functions of various kinds for their parent company, or for other foreign affiliates. If in Europe 45% of the largest firms with off-shoring experience has off-shored activities to their foreign affiliates, 48% of the companies have out-sourced activities to third party service providers (UNCTAD, 2004), witnessing that the phenomenon of service out-sourcing is also common. The choice between off-shoring and outsourcing in service activities, in favour of the first, depends first of all from the need of keeping strict control on that activities. For example the financial services industry appears to rely almost exclusively on internalize models of off-shoring. Moreover, offshoring is preferred when the level of internal interaction with other functions matters. Service, manufacturing and R&D activities call for strong interaction in order for the firm to be efficient; in contrast, back-office functions and customer interaction services can easily be outsourced. Outsourcing, in any case, is strongly subject to the existence of capable local firms; more than one example exists in which off-shoring was chosen because of the lack of efficient and reliable local companies in the host country.

b’) major threats/opportunities from the next global shift The global shift in services offers large potential benefits for countries at both ends of the process: receiving countries gain jobs, skills, access to foreign skills, while the sending ones improve their competitiveness by moving to higher level activities. Since most of the off-shoring and out-souring 31

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

has taken place among developed countries, this underscores that it is not primarily a “North-South” divide, and that it mainly affects regional economies in developed country. In theory, the most advantaged areas in Europe given this global shift are urban specialized service areas, where the possibilities to find capable local firms are greater. The advantages that a host local area can reach, however, depend on its ability to attract not only low value added activities, like back-office functions or customer interaction services, but also high-value activities in services, so to guarantee an increase in GDP, through productivity gains. A right balance between low and high value activities in services is therefore required, i.e. a balance that guarantees either productivity increases or limited losses of productivity compensated by more than proportional increases in employment. This balance is probably easier to be reached in urban service areas endowed of highskilled professions; the latter represent in fact an attraction element for high-value service activities and functions.

c) deindustrialization and the rising of service economy Globalization is generally associated with a deindustrialization process. In advanced countries and in regions specialized in manufacturing the new forms of production organization put in place by firms imply a shift of functions and tasks outside the area, with the expected consequence of job losses in industrial employment. Some confusion exists on this aspect in the literature, being a process associated only to industrial employment losses. On the contrary, pure industrial employment decrease is not enough to identify a deindustrialization process; deindustrialization takes place when industrial employment losses are associated with industrial productivity losses, and with a real industrial GDP decline. While strategies of outsourcing and off-shoring easily impact on employment, cutting some blue collar activities in the traditional industrial regions of advanced countries, their effects on industrial productivity is contradictory. It might be the case that industrial productivity increases thanks to a dropping out of non efficient functions and tasks, or thanks to the specialization of the region into higher value added functions. However, this process has to be efficient enough to guarantee an increase in industrial GDP at the local level. c’) major threats from deindustrialization and the rising of service economy In front of this globalization trend, an important balance has to be found by regional economies through the spatial reorganization of production; the losses in industrial employment have to be counterbalanced locally by a higher increase in industrial productivity, to guarantee at the same time an increase in real industrial GDP. This goal is reached through the ability of regional economies to re-orient their specialization towards new-growth industries and activities in related sectors. Examples of such transitions include the transformation from telephone handset production to mobile internet system design, or from vehicle production to GPS, road sensing and safety equipment (OECD, 2007). More important, in a period of rapid rise of service fragmentation of production, is a shift to the service sector, which is highlighted as a possible counterbalanced effect to industrial employment losses. Between 1998 and 2004, an OECD Report shows that most regions experienced large job losses in manufacturing (an average of 20.000 manufacturing jobs disappeared in each OECD country) and these jobs were usually, although not always, offset by growth in service employment (OECD, 2007). This process of substitution of industries with service employment presents some threats to regional economies too, and imposes the search for the right balance between industries 32

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

and services. In fact, major effects on the real local economy are registered when the new service jobs are high-value added jobs, generally in “producer services” (working for industries, outside them). On the other hand, when regional specialization moves towards low-value added services, mainly in “consumer oriented” activities or “low-profile functions” (e.g. call centers), the net advantage for the regional economy may be limited or even negative. At least a part of the present slowdown of aggregate productivity growth in advanced countries is linked to this kind of trend. The risk associated to the move towards service activities is that innovation trends, through which real productivity gains are achieved, take place in manufacturing more than in service activities. Regional economies are obliged to look for a balance between industrial and service sectors so to keep a certain level of innovation and productivity rate. A trivial quantitative substitution between numbers of jobs lost and re-created is a dangerous strategy; high-quality skills have to be protected, with true possibilities of productivity gains, as a main goal. Regions favored by this globalization trend are those regions that have already developed a certain level of R&D activities; innovative skills can be reoriented towards new production activities, facilitated by cumulative self-reinforcing knowledge.

d) decentralization of intertwined functions If this is a well known behaviour of firms nowadays, what is relatively new is the fact that delocalisation of activities is at present usually occurring for “packages” of intertwined functions: R&D activities follow nowadays production off-shoring, with the aim to absorb the needs of new local markets, and develop products that respond to these high growth markets (OECD, 2007). This is especially important for high-tech industries, like telecommunications, where the technological advances have to be tailored upon local consumers’ needs. There is abundant evidence of innovation off-shoring. The share of manufacturing R&D expenditure accounted for by foreignowned companies can be very high indeed: a recent OECD research reports this share around 50% in old EU 15 countries, and more than 50% for EU Eastern countries (OECD, 2006). This logic has been pushed to an extreme by some firms, keeping only a few internal functions like strategy, research and branding. Nike is a clear example of this “empty” or “hallow” firm, producing no product within its organisation, but having the “Nike” branding as a property asset, together with the design of the new products. This is also true for IBM and Compaq, two competitors sharing the same out-sourcing company, Ingram, as producer and assembler of their products (Fontagné and Lorenzi, 2005). When extrapolated for a number of firms and industries in a region, this corporate strategy may lead to increase in productivity for the region where these highvalue activities are kept, putting in place a process of local specialisation in “command and control” functions. This is also true for some service activities, that are in general off-shored and out-sourced together with production activities, and sometimes also alone. This is the case of pure R&D off-shoring in software production Bangalore is the well known case example).

d’) threats/opportunities from decentralization of intertwined functions The major threat related to this trend is the capacity of regions to become the loci of attraction for these combined functions. Higher skilled labour force and high innovation regions are expected to gain in the competition to attract these functions. 33

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

1.4 Research questions and structure of the research work The general aim of our study is to identify which regions benefit from globalization, and which are the aim reasons behind such a success. The specific research questions are the following: -

which are among the European regions the most influenced by globalisation processes? are the above mentioned globalisation trends present in European regional economies? which are the regions that mostly benefit from globalisation trends over the past ten to fifteen years? which are the new success factors, distinguishing successful and problem regions? which is the future of the European regional economies according to different globalisation strategies that can emerge over the next twenty years?

In order to accomplish our research task, and be able to reply to these questions, we proceed with some methodological steps (Fig. 1.2). First of all, the research aims at identifying the most open European regions to globalization; these are the regions where globalisation impact is felt first and / or more strongly. Global regions are identified on the basis of two major elements: -

regions that have a higher degree of openness, i.e. a higher degree of physical accessibility from outside Europe; at the same time, open regions are those regions that are specialised in service or manufacturing productions that are particularly open to external markets.

The first part of the project aims that identifying these regions, with the help of major trends in sectoral national trade and sectoral national foreign direct investments (chapter 2), and with the measurement of an indicator of regional globalization, capturing the degree of physical accessibility to each region (chapter 3). Once these regions are identified, the major globalization trends are analyzed in spatial terms, in order to see whether these trends influence more directly global regions (chapter 4). Our expectations are that among open regions, the ones benefiting the most from globalizations are those in which spatial trends generate growth opportunities, i.e. regions in which the above mentioned threats of globalizations are kept under control and turned into growth opportunities (sec. 4). Moreover, open regions are the ones able to attract not only extra-EU FDI, but internal EU FDI; they are therefore able to gain from European integration more than the others; benefiting regions are expected to be among the global regions those able to capture the highest advantages from the presence of FDI (chapter 5).

34

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Figure 1.2. Structure of the project

Chapter 2 WP 1.1 International trade patterns at sectoral level

WP 1.2 FDI geographical (re-)orientation at national regional and sectoral

WP 1.3 Identification of sectors benefitting from globalisation

Chapter 3

WP 2.2 Regional globalisation index

WP 2.1 Regional specialisation in open sectors

WP 2.3 Identification of open regions

WP 2

Chapters 4, 5 and 6

Chapter 7

WP 2.4a Spatial patterns and spatial effects of globalisation trends

WP 2.5 Case studies

WP 2.4b Spatial patterns and effects of FDI

WP 2.6 Synthesis of quali-quantitative success factors

WP 3.1 Scenarios

Chapters 8 and 9

WP 3.2 Policy implications

35

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

2. Globalization processes: trade and FDI trends 2.1. A major geographical shift of trade at the world level 2.1.1. World trade trends The call stipulates that “The new emerging international division of labour stimulates higher levels of specialisation. The wider access to markets increases the importance of agglomeration economies. An analysis of trade outside and within the EU could show whether an increase in specialisation and spatial concentration has occurred” (tender specifications, page 2). In this chapter, we will examine whether we observe this process of specialization/concentration at both world and European level according to classical data of trade. We will thus examine the major trends observed in the trade of goods and services from 1995 onwards, and put the accent on the EU position in the world trade.16 In the last years, the growth rate of the trade has been very high, around 5.5% in annual average between 2000 and 2007 to be compared with the 3% of annual growth of the merchandise production17. From the trade point of view, the globalization process is still on good trails. One of the reasons of this globalization by trade is the development of intra-sectoral trade and, more specifically, the intra-firms trade. Indeed, the major firms are locating the different segments of production in the different parts of the world, and this process explains the increase of the vertical intra-sector trade. However, this process should not be overestimated: while the intra-sector trade has considerably increased inside EU as a result of the integration process, the inter-sector trade still accounts for the major part of the trade for extra EU trade, as well as for USA or China (CEPII, 2004). In the same time, the world trade has gone through a major geographical shift: while the poles of the triad (US, Japan and Europe) have seen their share in the world trade decline, emerging countries – especially from Pacific Asia – have increased their share. This decline is very clear for EU in the last decade: when intra-block trade18 is excluded, the share of EU-27 in the world exports has decreased from 26.6% in 1995 to 22.6% in 2006. The decline has been even more dramatic for NAFTA, from 16.2 to 11.2 during the same period of time (Table 2.1). In the same period, the intra-EU trade in the world trade has also significantly declined: from 28.0% in 1995 to 25.6% in 2006. This is mainly due to higher growth rates in the world trade outside Europe (mainly in Southern and Eastern Asia) and does not correspond to the end of the integration process: the share of EU in the EU trade has indeed remained very stable from the nineties onwards (around 65%), after decades of a commercial integration process.

16

A detained analysis is reported in Annex 1 of Vol. III. These data are deflated (WTO, 2008). 18 Geographical areas are defined in Annex 2.1 of this chapter. 17

36

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

2.1.2. A new division of labour? The world trade can still be understood in terms of the old classic division of labour, the so- called centre/periphery. It means that specialization in terms of sectors and products is still correlated to the level of development: technological and upmarket products19 are very specific of the most developed countries (EU-15, USA, Japan), while emerging countries are more specialized in less technological products, labour intensive production or primary products. In this context, EU keeps a good position in some of the most technological sectors and some upmarket products. Indeed, the EU decline does not affect some sectors in which EU is increasingly specialized, mainly chemical (DG), machinery (DK) and transport equipment (DM) (Table 2.1; CEPII, 2004). In these sectors, EU has increasingly positive balances of trade. While EU has also been able to maintain its position in the electric and optical equipment sector (DL), it is not specialized and has a very negative trade balance in this sector. The other sectors in which EU has managed to increase its world market share are of minor importance. Looking into sectors in greater detail, we observe high heterogeneity inside these benefiting sectors. EU has generally declined in the labour intensive or less technological sectors (heavy chemical industry; electric appliance for example), but has been able to maintain its position for some more technological productions (refined chemicals, motors, precision instruments). The literature also suggests that EU remains very competitive in upmarket products for consumption goods, such as luxury goods (CEPII, 2004).

Table 2.1. Share of the world exports (%) for the three main economic blocks according to the economic sector, 1995-2006. EU-27 Sector Food Mining Manufacture of textiles and textile products Manufacture of wood and wood products Manufacture of pulp, paper and paper products; publishing and printing Manufacture of chemicals, chemical products and man-made fibres Manufacture of rubber and plastic products Manufacture of other non-metallic mineral products Manufacture of basic metals and fabricated metal products Manufacture of machinery and equipment n.e.c. Manufacture of electrical and optical equipment Manufacture of transport equipment Manufacturing n.e.c. Electricity, gas and water supply Non classified Total (TT)

NAFTA

A+DA C

20,8 3,8

21,0 2,8

21,2 2,1

19952006 0,4 -1,6

DBDC

21,1

17,1

15,6

DD

19,9

25,4

28,2

DE

37,9

36,5

DFDG

33,6

DH DI

Code

1995

2000

2006

1995

2000

Pacific Asia

19952006 17,1 -8,7 4,0 -3,4

2006

25,8 7,4

22,5 3,9

-5,5

5,5

5,0

2,9

8,3

12,3

9,2

4,3

42,4

4,5

34,3

29,3

32,8

32,6

-1,1

17,9

32,2

28,9

30,2

-2,1

51,8

43,5

38,8

-13,0

DJ

30,8

26,1

26,3

-4,5

DK

42,9

38,6

43,0

DL DM DN E

21,0 34,0 31,7 66,3 36,1 26,6

20,1 33,9 29,4 59,6 23,0 23,1

21,1 33,0 25,3 35,2 18,2 22,6

1995

2000

19952006 5,0 1,9 1,1 0,2

2006

3,1 0,9

3,8 0,8

-2,6

32,2

34,8

43,1

10,9

-8,0

6,3

9,7

21,5

15,2

22,0

-12,3

7,4

9,8

13,4

6,0

15,4

13,5

-4,5

12,2

12,6

14,3

2,1

19,7

18,6

14,9

-4,9

29,8

31,1

32,8

3,0

10,8

10,8

9,3

-1,6

21,2

22,7

30,9

9,7

10,1

8,1

6,9

-3,2

21,6

23,6

26,1

4,4

0,0

20,1

22,1

16,4

-3,7

25,1

26,5

27,8

2,8

0,0 -1,0 -6,5 -31,2 -17,9 -3,9

18,6 19,3 12,7 0,0 9,8 16,2

18,8 19,4 11,6 0,1 14,9 14,8

13,1 19,3 10,3 0,2 7,2 11,2

-5,5 0,0 -2,4 0,2 -2,6 -5,1

42,0 38,8 24,6 0,7 5,5 22,9

38,5 36,8 29,8 0,3 13,1 23,5

44,8 38,7 33,7 0,9 7,8 24,2

2,7 -0,1 9,1 0,3 2,3 1,2

Source: CHELEM

However, signs of change can be observed in this old division of labour. First, at the world scale and the level of sectoral division we used, no sign of a concentration/specialization process can be observed but rather a major shift from EU and NAFTA 19

Upmarket products are characterized by innovativeness and the skilled labour needed (CEPII, 2004). 37

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

to Asia and some other emerging countries. It means that in nearly all sectors and all segments of production (including R&D for example), emerging countries have taken a growing share of the world trade. But, as already stated, this is particularly true for all intensive labour manufacturing activities which have been massively offshored to Eastern Asia. Second, the literature suggests that a new division of labour is taking place. In this new division of labour, emerging countries (mainly of the “old generation” such as Southern Korea, Malaysia…) are improving in components and capital goods but not in upmarket finished goods for consumption, while China and India play a growing role in assembling consumer goods (CEPII, 2004).

2.1.3. The growing importance of service trade for EU In 2007, the trade of services account for 29% of the EU exports if we only take into account the extra-EU trade, while service only represent 20% of the total world exports. It means that EU-27 has a much higher world market share in services than in the merchandises, respectively 20% and 13% (Table 2.2)20. In the last years, the growth has not been much higher for services than for manufacturing trade: the annual growth of trade in merchandises has reached 12% in annual average between 2000 and 2007 and 13% for the services21. However, the EU-15 net balance for services has considerably increased between 1995 and 2005 (Table 2.3)22. From the data, it is clear that EU has been able to increase its competitiveness in the world for certain types of services. In particular, EU-15 has considerably increased its exchange balance with the rest of the world, mainly through four types of services: transportation (NACE 60 to 62), financial services (65), computer and information services (72+), and other business services (74).

Table 2.2. Exports of merchandises and services in 2007* Exports in billion $ EU-27 Agricultural Fuel and mining Manufactures Total Merchandises Transportation Travel Other commercial services Total Services Total

EU-27 (only extra-EU27)

Structure of the exports (%) World

EU-27 total trade (%)

EU-27 (only extra-EU27)

Share of EU in the world

W orld trade (%)

EU-27

EU-27 (only extra-EU27)

487,7 291,5 4249,1 5028,4 337,2 362,0

108,7 84,2 1406,5 1599,4 163,3 105,0

1127,7 2038,4 9499,5 12665,6 750,0 855,0

7,4 4,4 64,6 76,5 5,1 5,5

4,8 3,7 62,1 70,6 7,2 4,6

7,1 12,8 59,5 79,4 4,7 5,4

43,3 14,3 44,7 39,7 45,0 42,3

9,6 4,1 14,8 12,6 21,8 12,3

846,4 1545,6 6574,0

398,9 667,2 2266,6

1685,0 3290,0 15955,6

12,9 23,5 100,0

17,6 29,4 100,0

10,6 20,6 100,0

50,2 47,0 41,2

23,7 20,3 14,2

Source: WTO: http://www.wto.org/english/res_e/statis_e/its2008_e/its2008_e.pdf * The share calculated in the last column is different from Table 2.1 because intra-blocks trade has been excluded form the total.

20

These figures do not exclude intra-block trade in the non European parts of the world. Data from WTO (2008), not deflated. 22 Figures for EU-27 are only available for the last years and also show a growingly positive balance. As for the new member states they have a very little contribution in this international trade of services. 38 21

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Table 2.3. Service trade balance by type of services for EU-15 in 1995 and 2005, in millions $.

SERVICE

NACE1

NACE 2

Construction services

F

Transportation Travel Communication

I I I

45 60-62 63 64

Transportation and communication Insurance

J

Financial services

J

I 66 65 J

Finance and insurance Computer and information

K

raoyalties and license fees

K

Other business services Business services

K

Givernment services Personal, cultural and recreationnal services TOTAL SERVICES

72 73-74 K LQ MNOP

Extra EU-15 extra EU-15 Net balance Credit

Evolution of extra EU-15 Extra EU-15 the extra EUDebit Net balance 15 net balance 2005 2005 1995-2005 6914 4872 535 89313 13421 12585 95944 -22990 -23133 7843 -612 -105

1995 4337 836 144 -507

2005 11786 102734 72954 7231

472 2776 3659 6435 -303 -4165 4183 -285 1815

182919 6757 34857 41613 17534 23480 120391 161405 8099

193100 8128 13526 21654 8580 31093 88832 128506 6691

-10181 -1371 21331 19959 8954 -7614 31559 32899 1408

-10653 -4148 17672 13524 9257 -3448 27376 33184 -406

2664 12017

4448 413356

365723

4448 47633

1784 35616

Source: OECD

2.1.4. EU countries in the global trade The geographical pattern of trade inside the EU-27 has gone through complex evolutions. Here again, there is no clear pattern of concentration/specialization of the trade. The main evolution has rather been a modest decentralization process to the benefit of new member States but also of peripheral Western countries, mainly Ireland and Spain whose initial share in the EU-trade was indeed very weak. Among the major European countries, only Germany has been able to keep its dominant position in the EU benefiting sectors while Italy, France and most of all UK have globally declined in both intra and extra-EU trade of merchandises. However, the pattern has been different according to the sectors: while transportation equipment, metal and wood manufacturing industries have experienced a concentration process, the opposite is observed for chemical industry. The evolution of DG toward decentralization appears very clear from the map (Fig. 2.1 and 2.2): major countries have lost market shares in both intra- and extra-EU27 trade, while small countries of NMS, Ireland, Belgium as well as Spain have increased their weight in extra-EU27 trade. Machinery (DK) has been evolving quite differently: the dominant position of Germany has been relatively stable, while the other important countries have significantly declined, especially France and UK. If evolutions are positive in many NMS, we have to keep in mind that they still represent a very limited share of the extra and intra-EU27 trade. In the electric and optical equipment sector (DL), we observe contradictory evolutions in intra and extra-EU trade. On the world markets, Germany, Ireland, Netherland, Finland, Austria and Hungary have increased their market shares. Yet these evolutions were based on different types of production: for example, Germany is specialized in precision instruments, Finland in telecommunications instruments and Hungary in assembling different types of electronic production. In the transport equipment industry, geographical evolutions have been very clear: Germany has reinforced its position; all other Western countries have lost market shares to the benefit of some new member States (Slovenia, Slovakia, and the Czech Republic among others). 39

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

These major evolutions can be synthesized in the following way: - while Germany has been able to maintain or improve its position in some benefiting sectors, the other big countries have lost market shares in both intra and extra-EU trade (France, Italy and above all United Kingdom); - most NMS have increased their market shares in these benefiting sectors but at a low level; - Ireland, and to a lesser extent Spain, have increased their market shares in some benefiting sectors. As far as trade of services is concerned, the geographical pattern is very different from manufacturing industries. In 2003, United Kingdom is the first service exporter in the EU with 16.5% of the total EU trade of services, before Germany (14.5%), France (11.2%), Italy (10.6%) and The Netherlands (9.4%) (Fig. 2.3). UK performs best in financial and business services. In financial services, Luxemburg and Ireland are, together with Germany, the other main services suppliers. In nearly all types of services, the new member States still have a very limited share of the services exchanges. To a certain extent, this is also true of other more peripheral countries from Northern (Finland, Denmark) or Southern Europe (Spain, Portugal).

40

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Fig. 2.1. Evolution of the market shares of EU countries in the extra-European trade, between 1995-2006 for some sectors. INTRA

EXTRA

Chemicals and chemical products

Share of Eu trade (%) 20 10 5

Share of Eu trade (%) 20 10 5

Relative evolution ((2006-1995)/1995) -0.61 - -0.45 -0.45 - -0.22 -0.22 - 0 0 - 0.04 0.04 - 0.46 0.46 - 1.2

Relative evolution ((2006-1995)/1995) -0.61 - -0.5 -0.5 - -0.2 -0.2 - 0 0 - 0.25 0.05 - 0.9 0.9 - 1.83

Share of Eu trade (%)

Share of Eu trade (%)

Machinery and equipment

20 10 5

20 10 5

Relative evolution ((2006-1995)/1995) -0.22 - -0.2 -0.2 - -0.05 -0.05 - 0 0 - 2.14 2.14 - 4.76 4.76 - 6.66

Relative evolution ((2006-1995)/1995) -0.3 - -0.2 -0.2 - -0.05 -0.05 - 0 0 - 0.43 0.43 - 1.31 1.31 - 3.42

Share of Eu trade (%)

Share of Eu trade (%)

Electrical and optical equipment

20 10 5

20 10 5

Relative evolution ((2006-1995)/1995) -0.41 - -0.35 -0.35 - -0.08 -0.08 - 0 0 - 0.49 0.49 - 4.2 4.2 - 6.99

Relative evolution ((2006-1995)/1995) -0.33 - -0.2 -0.2 - -0.1 -0.1 - 0 0 - 0.88 0.88 - 2.71 2.71 - 9.32

Share of Eu trade (%)

Share of Eu trade (%)

Transport equipment

20

20

10 5

10

Relative evolution ((2006-1995)/1995) -0.33 - -0.21 -0.21 - 0.07 0.07 - 0.25 0.25 - 0.44 0.44 - 3.87 3.87 - 4.88

5

Relative evolution ((2006-1995)/1995) -0.57 - -0.35 -0.35 - -0.2 -0.2 - 0 0 - 1.31 1.31 - 3.43 3.43 - 6.69

Source: UNCTAD

41

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Fig. 2.2. Evolution of the market share of EU countries in the total EU-27 trade between 1995 and 2006. INTRA

EXTRA

Share of Eu trade (%)

Share of Eu trade (%)

20

20

10

10

5

5

Relative evolution ((2006-1995)/1995) -0.21 - -0.18 -0.18 - -0.03 -0.03 - 0 0 - 0.69 0.69 - 1.15 1.15 - 1.38

Relative evolution ((2006-1995)/1995) -0.21 - -0.18 -0.18 - -0.03 -0.03 - 0 0 - 0.69 0.69 - 1.15 1.15 - 1.38

Source: UNCTAD Fig. 2.3. Services trade of EU countries in 2003 Transport, storage and communication

Financial intermediation

Share of Eu trade (%) 20 10 5

Share of Eu trade (%) 20 10 5

Balance trade -26.3 - -10 -10 - 0 0 - 20 20 - 55

Balance trade -65.6 - -43 -43 - 0 0 - 20 20 - 62

Real estate, renting and business activities

Total services

Share of Eu trade (%) 20 10 5

Share of Eu trade (%) 20 10 5

Balance trade -32.7 - -25.3 -25.3 - 0 0 - 13 13 - 34.9

Balance trade -20.5 - -15 -15 - 0 0 - 25 25 - 52

Source: UNCTAD 42

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

2.2. Geographical Patterns of FDI inflows The last wave of globalization that has affected world economy since the beginning of the past decade has seen multinational enterprises (MNEs) play a leading role in shaping and driving crossborder integration of production through the transfer of production facilities, functions and or technology across space. These trends have been reinforced by the liberalization of new markets, especially in the service sectors, the reduction of capital movement restraints, and the creation of a friendly environment for Foreign Direct Investment (FDI) in a growing number of countries.23 The EU has actively participated in this process, being at the same time one of the most important source of and destination for capital flows. MNEs’ location and production decisions have contributed to reshape the geography and industry structure within and across countries. Some of the key facts concerning FDI and its spatial and sectoral trends include:  The main driver for FDI in the EU has been the integration process, in both its vertical and horizontal dimension;  FDI is mainly an intra-EU phenomenon, with more than two thirds of capital flows occurring among European countries;  Although the share of FDI in manufacturing is still significant, FDI flows in the service sectors are becoming more and more important;  Patterns of FDI show a clear spatial and sectoral concentration;  There is a clear divide between Western and Eastern European countries, with the former attracting relatively more FDI in the service sectors than the latter, which are a favourite location for FDI in manufacturing. 2.2.1 General trends The last decades have seen two important waves of increasing FDI inflows at world level – the first occurring in the mid-1980s, and the second in the mid-1990s – driven by a worldwide sustained economic growth, an acceleration of the technical progress, especially in the information and communication technologies, as well as privatization programs and liberalization of new markets all around the world. The EU has been a major player in these waves, since they coincided with three important milestones of the European integration process, e.g. the single market program, the introduction of the Euro and the East enlargement. Despite the cyclical character of FDI flows and their dependence from economic fundamentals, inward FDI stocks in the EU have increased exponentially since the 1980s, reaching their peak in 2007, with more than 7000 billions of USD and a percentage of world stocks of about 45%.24 Better information on the capacity of an area to attract FDI can be drawn by adjusting inward FDI stocks with GDP. As shown in Fig. 2.4, the EU’s capacity to attract FDI has increased over time and overcome world average from 1990 onwards, thus suggesting that the EU has been able not only to maintain but also to further improve their attractiveness for foreign investments, despite the emergence of new interesting destinations all around the world, such as China, India and Brazil.

23

A detailed analysis is contained in Annex 2 of Vol. III. See UNCTAD, World Investment Reports, various issues for an in-depth analysis of FDI flows and stocks at European and world levels.

24

43

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Fig. 2.4. FDI inward stocks over GDP 45.0 World

40.0

European Union

35.0 30.0 25.0 20.0 15.0 10.0 5.0

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

-

Source: UNCTAD database

The dynamic of the distribution of FDI inward stocks within EU27, highlights a persistency in geographical concentration of FDI across EU countries, while confirming an overall improvement in FDI penetration (Fig. 2.5). In particular, three phases, one for each decade, can be clearly distinguished. In the 1980s FDI was quite concentrated in a few number of countries, as indicated by the gap between the minimum and the maximum values recorded, which is always very large, with negative minimums (Cyprus) and maximums (Ireland) exceeding 100. In the 1990s, instead, FDI was more geographically dispersed, thus involving a larger number of countries, while in the 2000s, concentration increased again, though no negative values were recorded. These patterns may be explained by the enlargements of the EU that have opened to FDI more and more markets over time, thus generating inequalities among new and old destinations. It is also worth noticing that, on average, FDI penetration has increased over time, especially from 1990 onwards. This trend coincides with the openness of the new markets of Central and Eastern European countries, which have attracted a consistent share of both intra and extra EU FDI. Fig. 2.5. FDI stocks over GDP in the EU27, 1980-2007 200.00 150.00 100.00 50.00 0.00 max

-50.00

min EU average

-100.00 1980

1983

1986

1989

1992

1995

1998

2001

2004

2007

Source: UNCTAD database

44

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

2.2.2. Foreign firms in the EU27: geographical and sectoral patterns Although useful for a comprehension of the main facts and trends concerning FDI, international statistics on inflows and inward stocks at country level suffer from many shortcomings and distortions, which make them less significant in in-depth and more disaggregated analyses on the structure and evolution of foreign investments at sector and region levels. At this purpose, information drawn from the FDI Regio dataset is much useful, since it allows to analyse the activity of MNEs with reference to the economic sector, the origin within or outside Europe, and the location within each destination countries.25 As expected, FDI penetration in the EU has increased over time, given the progressive liberalization of markets and the reduction of transportation costs which make easier and more efficient for firms to exploit new strategies of production based on off-shoring and outsourcing abroad. During the considered period, the number of new foreign firms established in the EU27 grew from 19,410 of the 1997-99 period to 109,155 of the 2005-2007 period (Table 2.4). About one third of these firms come from non-EU countries. Therefore, globalization in Europe is more regional in nature, being sustained by the regional integration process. This trend has reinforced over time, as indicated by the share of intra-EU FDI on total FDI, which has increased from 58% of the end 1990s to 67% of the mid-2000s. Table 2.4 New foreign firms by periods 1997-99

2001-03

2005-07

EU27

20400

96956

109155

EU15

17903

74964

66349

EU12

2497

21992

42806

EU27

58%

62%

67%

EU15

60%

65%

65%

EU12

51%

53%

69%

total number of new foreign firms

Intra EU FDI (%)

main recipient countries (%) Uniuted Kingdom

23%

42%

32%

Romania

2%

16%

31%

France

14%

9%

8%

Germany

13%

4%

3%

Poland

1%

1%

5%

Source: Own elaboration from FDIRegio database

Despite that, the apparent paradox of globalization, e.g. global industry highly geographically concentrated, is confirmed by the EU experience. The geographical distribution of FDI within Europe is in fact very uneven, with EU15 collecting 61% of total FDI flows and EU12 the remaining 39%. However, it is worth noticing that in 1997-99, only 12% of foreign firms chose to locate in Central and Eastern European countries. The improvement in the attraction capacity of EU12 is due mainly to Romania, where flows of FDI have recently increased substantially, driven by the improved economic conditions of the country and the accomplishment of the reform process. In 2005-07, the most important recipient countries for FDI were the United Kingdom (32% of total FDI) and Romania (31%), followed by France (8%) and Poland (5%). 25

FDI Regio dataset has been built up starting from Amadeus database. Annex 2.2 of this chapter provides more details on FDI Regio dataset, including main variables, its primary source and the criteria we used to construct it, its major advantages and shortcomings with respect to official aggregated data, as well as its level of significance in explaining general trends in FDI. 45

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Since larger countries tend to attract more FDI than smaller ones, international comparison requires adjusting absolute flows with country size. Therefore, total number of firms has been normalized by population. As Fig. 2.6 shows, top five recipient countries include now small countries, such as Ireland, Denmark, Austria and Belgium, while large countries show a very poor performance in term of FDI. Moreover, new member states move up the rank over time, and in 2005-07 two of them, namely Romania and Poland belong to the top 10 performers in terms of FDI. It is finally worth noticing that though globalization has reduced the importance of where to locate, geographical concentration of foreign firms tend to increase over time, at least at country level. This pattern is more marked for extra than for intra-EU FDI. Fig. 2.6. FDI inflows over GDP in EU27 1997-99

2005-07 ro

dk

ie

ie

uk

at

lu

uk

at

lu

be

be

ee

es

nl

nl

se pl

se

fr

fr

bg

lv

hu

cz

dk

hu

fi

de

lv pt

pt

cz

fi

es

ro

de

it

sk

gr

it lt

pl

gr

bg

si

EU12 EU15

extra EU FDI

EU27

total FDI

0.00

50.00

100.00

EU12

extra EU FDI total FDI

EU15 EU27

150.00

0.00

500.00

1000.00

1500.00

Source: Own elaboration from FDIRegio database

One of the most striking characteristics of the globalization process has been the reduction of the importance of the manufacturing sectors and the emergence of the service sectors as pillars of the economic structure of many countries and regions. FDI does not represent an exception to this respect. Recent liberalization and de-regulation of service markets, as well as the rapid expansion of demand in several service markets have opened up new business opportunities in home markets, together with the ability to expand internationally in certain service activities. This general 46

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

phenomenon is common to several countries, and Europe does not make an exception. As the share of FDI in the service sectors has dramatically increased over time, from 58% of late 1990s to 75% of mid 2000s, the share of FDI in manufacturing activities has decreased from about 41% to 24% of total FDI in the 2005-2007. This pattern is particularly marked in EU12, where FDI in the manufacturing sectors accounted for about 60% of total FDI in the late 1990s. Despite that, FDI in the manufacturing sectors is more important in the EU12 rather than in the EU15. FDI in primary sectors – e.g. agriculture and mining and quarrying – remains negligible, accounted for less than 2% of total FDI (Fig. 2.7). Fig. 2.7. FDI by sectors (percentages) 100% 90% 80% 70% 60%

Tertiary sectors

50%

Secondary sectors

40%

Primary sectors

30% 20% 10% 0% EU27 EU15 EU12 EU27 EU15 EU12 EU27 EU15 EU12 1997-99

2001-03

2005-07

Source: Own elaboration from FDIRegio database

At country level, the increase in the importance of the service sectors for the world economy yielded to changes in patterns of FDI relative concentration at sectoral level in favour of the tertiary sector. As it is shown in Fig. 2.8, relative concentration of FDI in services has increased over time, bringing the sectoral distribution of FDI between manufacturing and service sectors across EU countries in the 2000s more closed to the EU average than it were at the end of the 1990s. Despite that, a clear divide emerges between EU15 and EU12 member states with the former showing a relative concentration of FDI in the service sectors higher than that in the manufacturing sectors (Fig. 2.9). The opposite trend characterizes EU12 countries. In particular, Czech R., Hungary, Poland, Romania and Latvia further increase their attractiveness in FDI in the manufacturing sectors, while, Bulgaria has been gaining positions as potential location for FDI in the service sectors, because of the recent liberalization process.

47

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Fig. 2.8. FDI sectoral concentration by country

LQs in services

1.50

1.00

0.50

1997-99 2005-07 2001-03

0.00 0.00

0.50

1.00

1.50

2.00

2.50

3.00

LQs in manufacturing

Source: Own elaboration from FDIRegio database

Fig. 2.9. FDI sectoral specialization in EU27 countries FDI in service sectors 1.20 lu 1.00 2005-07

hu

lv

uk

ie

EU15

bg

ro

fr

EU12

pt cz

gr

at se

fi

be

de es

0.80 pl

it

0.60

0.40 0.40

0.50

0.60

0.70

0.80

0.90

1.00

1.10

1.20

1997-99

FDI in manufacturing sectors 2.00

pl

1.80

2005-07

1.60 cz

de

1.40

es

1.20 1.00

pt

fi

be

fr

dk

0.60

nl

at

hu

bg

se gr

0.80

EU12 ro lv

EU15 uk

ie lu

0.40 0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

1997-99

Source: Own elaboration from FDIRegio database

48

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Annex 2.1. The geographical zones used in the analysis of trade

SUB-SAHARIAN AFRICA NAFTA LATIN AMERICA ASIA and OCEANIA NORTH EUROPA SOUTH EUROPA CHINA - JAPAN - ASIAN DRAGONS ex USSR MIDDLE EAST EU 27

Annex 2.2. FDIRegio dataset: its structure and its consistency with official data In this report, we use firm-level information on inward FDI. The data are taken from the Amadeus database collected by the Bureau Van Dijk (www.bvdep.com). The database consists of company accounts reported to national statistical offices concerning 11 million public and private companies in 41 European countries. For each company the database provides the year of creation, the country/region and the ownership structure by nationality. The data also includes the region where the firm where founded, as well as the sector of activity. Foreign firms were selected when they were newly created during the periods considered, i.e. 1997-1999 and 2001-2003 and when the percentage of assets owned by non-resident was at least 10%. We also considered all EU27 countries as host countries, while distinguishing between European and non European foreign investors.26 For the first period there were no data available for Cyprus, Malta and Slovenia, while for the second one only Cyprus and Malta were missing. Moreover, in the 1997-1999 period no foreign direct investments were recorded in Estonia, Lithuania and Slovakia. A limitation of this data for studying the geographical patterns of foreign firms is that data contain either plant or firm level information. This can potentially bias the location of FDI in favour of regions and/or countries where headquarters tend to locate. An advantage of this approach is instead represented by the fact that we do not need to estimate the regional distribution of foreign firms starting from national data. This top-down approach, in fact, is based on the simplifying assumption that the sensitivity of foreign firms to employment data – or whatever it is used to regionalize patterns of FDI – is constant across foreign firms, regardless the internationalization strategy they 26

We considered as European investors also Norway, Island, Lichtenstein because they have signed agreements with the EU which allow them to participate to the single market without being member of the EU. 49

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

pursue (efficiency, market and resource seeking FDI) the country of origin and the role the foreign affiliates can play within the group (productive vs. research units). Comparing UNCTAD data on inward FDI flows with the total number of foreign firms extracted from Amadeus following the criteria just described, we can notice that the correlation between the two measures of FDI flows is quite high in all periods, as indicated by Fig. A2.1. Thus, by considering number of foreign firms instead of values of FDI we do not introduce in the sample any significant distortion, though foreign investments in some destination countries have a relative importance that is different in terms of number of firms with respect to the value of FDI inflows.

50

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Fig. A2.1 Number of foreign firms vs. Inflows of FDI a) 1997-99 250000 200000

5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0

Inw ard FDI flow s n. of foreign firms

150000 100000 50000 0 at be bg cy cz de dk ee es fi fr gr hu ie it lt lu lv mt nl pl pt ro se si sk uk

Pearson correlation coeff.: 0.871 – p>0.000

b) 2001-2003 160000

inw ard FDI flow s

140000

n. of foreign firms

120000 100000 80000 60000 40000 20000 0

45000 40000 35000 30000 25000 20000 15000 10000 5000 0

at be bg cy cz de dk ee es fi fr gr hu ie it lt lu lv mt nl pl pt ro se si sk uk

Pearson correlation coeff.: 0.415 – p>0.032

c) 2005-2007 590000 490000

Inflow s of FDI n. of foreign firms

390000

40000 35000 30000 25000

290000

20000 15000

190000

10000 90000 -10000

5000 at be bg cy cz de dk ee es fi fr gr hu ie it lt lu lv mt nl pl pt ro se si sk uk

0

Pearson correlation coeff.: 0.626 – p>0.000

51

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

3. Identification of European global regions 3.1 Aim of the chapter The aim of this chapter is to identify typologies of regions that can potentially benefit from globalization in different ways. As mentioned in Chapter 1, regions expected to be more influenced by globalisation trends are global regions. These are identified as: -

regions with competitive and dynamic sectors (sectors in search of new markets, more open to competition, and better able to gain advantages from this competition); regions with high accessibility to large markets, and accessible from large markets.

The degree of openness of a regional economy influences its possibility to potentially benefit from globalization. Only those regions which are well endowed with physical connections and which have the right specialization in competitive and dynamic sectors have the potential to be global players, defined as those regions where globalization’s impact is felt first and most strongly. As we will see in the next chapters, global players are able to benefit from globalisation if the risks linked to globalisation are kept under control and threats turned into opportunities. This chapter will concentrate on the methodology to cluster European regions into homogeneous groups according to their degree of openness and therefore to their potential source of advantages that they can obtain from globalisation, leaving the identification of their performance trends to chapter 6. In the next sections we will present the two main indicators on which we base the identification of open regions. The indicator of physical accessibility of regions will be studied in section 3.2, whereas the specialisation in dynamic and open sectors will be investigated in section 3.3. Section 3.4 will finally cross the two dimensions to identify the global players and the regional players, the two types of regions which will be at the basis of the analysis of the following chapters.

3.2. Globalization Index 3.2.1. Regional participation in the global networks Regions have different levels and types of participation in the current globalization process and in the global networks. The objective of this section is to evaluate this regional participation in the extra-European global networks. From an economic point of view, the insertion in the global world occurs through three main channels: trade (exchanges), attraction of activities (FDI) and attraction of workforce. Trade indicators do not exist at the regional level and is treated in a more indirect way in Section 3.4. As a consequence, the globalization indicators are indeed related to the two other dimensions. The indicators and sources to build a globalisation index we will use are the following ones: -

the number of offices of advanced services firms, which allow to capture the capacity of a region to insert in the networks of advanced services which have been important for economic growth in the recent years (GAWC); 52

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

-

the headquarters of transnationals are an indicator of the commanding functions of a region (Fortune; IGEAT); the extra-EU FDI in the region is an indicator of the capacity to attract extra-EU capital (FDIRegio database); the extra-EU airflows connections is an indicator of the insertion of a region (city) in the global networks and not an accessibility indicator (OAG); the extra-European born population indicates the capacity to attract foreign labour force (Census data for Eurostat completed by LFS for Greece; National statistics for Belgium and Germany).

However, these indicators give a partial and biased picture of the position of the region in the world networks. This is why we will opt for a synthetic indicator extracted from a principal component analysis on the different variables of global insertion. Map 3.1. presents all indicators used in the PCA, which are discussed hereafter.

3.2.2. Attraction of activities The attraction of activities is measured through two different indicators: a global indicator of FDI, focusing on extra-EU FDI only, and the number of offices of leading firms in advanced services. FDI indicator is the penetration index of extra-EU FDI, i.e. it represents the number of new inward FDI firms per million inhabitants over the period 2001-2003. Because of strong national effects that characterise this phenomenon, we choose to use this indicator with respect to the national mean. While less general than the FDI indicator, the offices in advanced services27 is perfectly in line with the global cities theory. In this theory, cities are the nodes of the firms’ network. In this framework, the number of offices is an indicator of the capacity of the nodes (cities) to attract activities in highlevel services which are one of the key of the growth in the post-fordist capitalism. As illustrated by map 3.1, the highest numbers of advanced services offices are to be found in the capital-cities. This is especially true in the New Member States where these cities seem to concentrate all the national potential of attractiveness for advanced services. In Western Europe, beside the major poles (London, Paris, Brussels, Amsterdam, Madrid…), secondary national and international poles also appear in the firms’ networks. Transnational headquarters have been used as an additional indicator to approximate the commanding capacities of the regions which, to a certain extent, could give an idea of the stability of the regional insertion in the world economy. As compared to the capacity to attract activities in high-level services, the commanding functions show an even more concentrated geography. Big world cities of London and Paris are at the top of these commanding functions in Europe. Amsterdam show high level of commanding functions compared to the size of the economy. Germany, and to a lesser extent Italy, has a much less concentrated geographical pattern of commanding functions, sign of their urban polycentricism. The New Member States are nearly absent from these Fig.s: it means that these regions are attractive for investors but they do not host headquarters of transnational companies. 27

For this indicator and transnational headquarters, the location is the city; we convert the location into a regional location even if the activities are very much concentrated inside the central areas of the cities. 53

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Map. 3.1. Single indicators of the globalisation index Transnational headquarters

Offices of advanced services firms

Number of employed of the transnational whose headquarters is in the region

Number of offices 300 150 75

5000000 2500000 1250000

Number of employed for 1000 jobs 0.93 - 82 82 - 207 207 - 478 478 - 1163

Number of offices for 1000000 jobs 4 - 28 28 - 60 60 - 105 105 - 353

Extra-liberalized European air space connections

Extra-European Population

Number of seats

Extra-Eur Population

2000000 1000000 500000

1000000 500000 250000

Share of extra-eur. born population % 0-2 2-5 5-8 8 - 12 12 - 26

Number of seats for 1000 inhab. 0 - 200 200 - 400 400 - 600 600 - 4100

0

500

1000 Km

3.2.3. Attraction of the workforce The attraction of workforce is evaluated to the share of Extra-European born population28. Since the aim of the indicator is to evaluate the capacity to attract foreign qualified workforce, we compare our indicator with the share of the high-level graduation extra-EU born population according to LFS, which is much more incomplete than the census data we used. For all common observations between both indicators (N=185), the correlation is high. The comparison suggests that metropolitan areas are underestimated by our general indicator because of their specific capacity to attract highly qualified workforce. While still reflecting the capacity of the rich North-Western Europe to attract foreign population, the map also illustrates the high number of extra-European foreign-born population in former outmigration Mediterranean countries such as Spain, Italy, Portugal and Greece. These figures are particularly impressive because they represent a new phenomenon in these regions. The map also highlights the presence of nearby foreign-born population at the margins of the EU-space, especially the Russian population in the Baltic States.

28

Extra-European space is defined as all space commonly considered as Europe, which include all the Balkan states but not Turkey and not the ex-USSR, except the Baltic states. 54

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

The concentration of the extra-European born population in the metropolitan areas – in both relative and absolute terms – is very clear from the map.

3.2.4. Air connections with extra-European space Air connections could be considered as synthetic indicator since all the dimensions (workforce, economic exchanges…) of the globalization process produce connections with the rest of the world.29 However, despite this synthetic aspect, this indicator presents a major bias since extra-European airflows are concentrated into several major hubs, which mean that majors hub cities are overestimated and the others are underestimated. The geography of airflow connections is an indicator of the cities position in the global networks. It reflects relatively well the position of European cities in these networks, while overestimating of the international hubs cities located in London, Paris, Amsterdam and Frankfurt. In the same time, it is exactly the same cities which have high number of transnational headquarters and also host the most important stock exchanges in Europe.

3.3. Toward a synthetic globalisation index 3.3.1. Methodology To synthesize the different indicators, we will produce a Principal Component Analysis (PCA). It will enable us to capture the most relevant common pattern of the different indicators. It means that the different types of bias and specificities of our indicators will be to a certain extent limited by the synthetic indicator. To run this PCA analysis, we can use relative or absolute indicators. Relative indicators are related to jobs or population and take into account the relative capacity of a region to attract international activities and workforce and to connect with the rest of the world. However, the denominator is very much dependent on the size of the NUTS area according to the main cities. As a result, large NUTS2 areas compared to the cities (Lombardia compared to Milano for example) are underestimated in regard with small urban NUTS2 areas (Brussels or Hamburg for example). Since three of the 5 indicators are indeed strictly urban-located, the ratio with population or employment of the NUTS2 area introduces some biases. Absolute figures on the contrary are well reflecting the capacity to attract from the rest of the world or to connect with the rest of the world. However, for two indicators which are less metropolitanoriented (extra-European born population and FDI), large NUTS2 areas have – all things being equal – higher values.30

29

Extra-European space is defined as the non European liberalized airspace which include Iceland, Norway and Switzerland. The location is the airport and we convert into a regional location. In the few cases where the NUTS2 location of the airport does not correspond to the city to which this airport is related, we locate the airport in the NUTS2 of the main city rather than in the NUTS 2 of the airport location (Brussels, London). 30 Annex 4 of Vol. III contains all logical PCA run on absolute or relative indicators, and demonstrates that the degree of correlation between the different outcomes is very high. 55

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

In order to define global regions, we opted for relative indicators with a threshold derived from absolute figures. By combining relative and absolute indicators, we avoid the bias related to the dimension of the Nuts 2 area (because relative indicators are overestimated in small NUTS2 and absolute indicators are overestimated in big NUTS2). This solution is also in line with the theory (notably global city theory): to benefit from globalization, especially in services, it is also necessary to have a critical mass. Clustering in specific service activities cannot survive without this critical mass and density which provides the necessary range of other services in the environment. 3.3.2. Globalization index according to relative indicators In this analysis all indicators are relative to the number of jobs or the population of the area. FDI indicator is relative to the national mean since this indicator presents a strong national effect. In the Principal Component Analysis, the first component takes into accounts 57% of the initial information and is the only one whose Eigen value is higher than 1 (which means that it takes more information than the initial variables) (Table 3.1). Moreover, this first component takes into account in a relatively homogeneous way the different variables (Table 3.2). It means that the globalization indicators we entered in the analysis are relatively well correlated with each other. As a consequence, the scores of the regions on this first component are a good synthesis of the different globalization indicators. The geography of the globalization index is presented in Map 3.2. As already stated, to select the global regions, we also take into consideration the globalization index according to absolute figures in order to capture only regions with the necessary critical mass to benefit from globalization (see also Map 3.2).

Table 3.1. Share of information taken into account by each components Component 1 2 3 4 5

Eigen value 2,85 0,73 0,60 0,44 0,38

% of Variance 56,92 14,67 12,04 8,86 7,51

Table 3.2. Relationship between the 5 globalization indicators and the 2 first components

Advanced services offices transnational headquarters Air connections/pop Share of Extra-eur born population FDI/jobs in regard with national mean

Component 1

Component 2

0,744

0,417

0,793 0,836

-0,165 -0,134

0,694

-0,554

0,696

0,455 56

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Map 3.2. Globalisation index (2000-2005) Globalization index (2000-2005)

score on the first component of a PCA on 5 relative globalization indicators

-0.198 - 10.536

s r e t e m o l i K 0 0 8

0 0 4

0

-0.8 - -0.436 -0.436 - -0.175 -0.175 - 0 0 - 0.748 0.748 - 1.531 1.531 - 3.569 3.569 - 7.13

Global regions according to absolute indicators

Author: IGEAT

57

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

3.4. Open sectors The second determining aspect of the external openness of regions is their specialization. In fact, being specialized in growing sectors, relatively more open to trade than the EU average, and performing relatively better than the EU average in periods of high globalization processes, is an important channel by which regions can take advantage of globalization trends. On the contrary, a specialization in closed and/or declining sectors, makes regions less able to take advantage of external opportunities. Given the lack of sectoral trade data at regional level for all EU countries, the identification of regions endowed with dynamic open sectors is identified in a two step procedure. The first step is useful to determine which sectors are the open and growing ones. To achieve this goal, we used data from the analysis on the external (extra-EU) performance of European sectors at NACE1 level, both on trade and on FDI. The sectoral analysis is reported in Chapter 2 and in details in Annex 2 of Vol. III. Open growing sectors are defined as those which have a good performance in either FDI or ExtraEU trade31. As far as FDI are concerned, we consider as positively affected sectors those which have a growing share of extra-EU FDI. This applies (Table 3.3) to only two small manufacturing sectors (DD and DN) and five service sectors (G, H, I, K, MNOP), this because there has been a strong orientation towards service FDI so that, overall, the share of goods sectors has decreased of 24.8% while the share of service sectors has increased of 24.4% between the periods 1997-99 and 2001-03. As far as trade is concerned, one single indicator is not able to capture all needed features. In fact, we need three complementary aspects: 1. the openness of the sector has increased over time. The classic indicator for this is the sum of export and imports, relative to GVA to avoid biases due to the changing importance of sectors in the international economy; 2. exports in the sectors have been growing and, since this applies to almost all sectors due to increased international trade flows, they have to grow more than the EU average. This can be captured by the fact that a sector has increased its share in EU exports; 3. the European trade balance has been improving over time, signalling that European goods and services in that sector have been increasingly competitive worldwide. The trade patterns are a good indicator for that, but also these ones need to be made relative to sectoral GVA in order to avoid biases due to the changing importance of sectors on international trade. A sector which complies with all three requirements at the same time is an open sector in trade activities. Any sector which complies either with the FDI requirements or with the trade requirements, is according to our definition an open sector. However, since a sector whose GVA decreases over time is hardly a sector which benefits from globalization, we also require that sectors, to qualify, have been growing in real terms (Table 3.3). As it can be observed in Table 3.3, however, only a few European sectors have decreased their real GVA over the period 1995-2005, and none of them complies with trade or FDI requirements. 31

Details of this analysis are in the Annex 3 of Vol. III. 58

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Table 3.3 Identification of open growing sectors Sector

Mining Food Manufacture of textiles and textile products Manufacture of wood and wood products Manufacture of pulp, paper and paper products; publishing and printing

Nace code

T1: FDI1: Share of the FDI open increasingly growing different sectors [criterium open sectors sectors in the quota of FDI1] [absolute total EU extra EU FDI change in GVA in 2005 [absolute (X+M)/GVA change in (%) ] FDIi/FDIeu]

GVA Sectors T3: T2: export Open Trade open criterium which improving growing growing sectors sectors [yearly European [criteria T1 comply with sectors growth 1995- (TRADE or either [absolute trade balance + T2 + T3] FDI) + GVA 2005] TRADE or [absolute change in FDI Xi/Xeu] change in (XM)/GVA]

C A+DA

0.8 4.0

-0.5

9.4

-1.9

0.8

DB-DC

0.8

-0.4

117.7

-1.7

-64.3

DD

0.4

12.0

0.1

0.3 DD

5.0 DD

1.9 -0.6 -1.8 DD

DE

1.5

-2.3

4.0

-0.6

DF-DG

2.1

-4.9

100.0

3.9

14.6 DF-DG

DF-DG

DH

0.8

-1.1

39.1

0.0

11.0 DH

DH

DI

0.8

-0.6

9.3

-0.5

-1.7

-0.3

DJ

2.3

-1.9

39.1

0.1

-6.2

0.5

DK

1.9

-4.0

35.7

-1.3

13.4

0.5

DL

2.0

-7.6

96.3

0.3

-9.2

0.4

DM DN

1.8 0.7 19.1

52.0 59.5 48.3

0.3 -0.9 -2.2

2.2 6.0 11.4 2.9 7.1 5.7 22.0 6.3

-0.2 0.0 3.3 1.1 0.6 -0.8 16.9 0.0

3.1 -2.0

0.1 -0.7

Government services

E F G H I J K L-Q

Personal, cultural and recreationnal services

M-N-O-P

16.5 71.9

3.1 M-N-O-P 24.2

Manufacture of chemicals, chemical products and man-made fibres Manufacture of rubber and plastic products Manufacture of other non-metallic mineral products Manufacture of basic metals and fabricated metal products Manufacture of machinery and equipment n.e.c. Manufacture of electrical and optical equipment Manufacture of transport equipment Manufacturing n.e.c. Goods Electricity, gas and water supply Construction Trade (retail and wholesale) Hotels, restaurants Transportation and communication Finance and Insurance Business services

Services

-1.3 0.2 DN -24.8

G H I K

11.3 5.4 5.0 -0.6

-0.9 0.8 3.4 -0.4

0.0 2.8

-0.1 2.8

6.1

0.2 DD

8.8 DM -16.9 1.7

-0.1

DM DN

-1.6 -0.4

-1.5 1.4 J 1.4 K -0.2 0.0 0.3

G H I J K M-N-O-P

0.8 DF-DG 0.4 DH

1.1 DM 0.3 DN 0.1 1.5 2.1 1.8 3.2 2.4 3.3 3.3 1.4

G H I J K

2.6 M-N-O-P 2.6

Legend: coloured sectors meet the requirements for each indicators. In the last column coloured are the sectors that comply with all requirements to be defined “open growing sectors”.

Our analysis highlights 11 open growing sectors. For what concerns manufacturing sectors, we highlight as open growing sectors: -

DD, manufacture of wood and wood products; DF-DG, manufacture of chemicals. Chemical products and man-made fibres; DH, manufacture of rubber and plastic product; DM, manufacture of transport equipment; DN, manufacturing not else classified.

The open growing service sectors, which are less disaggregated, represent overall an higher portion of European GVA. This was expected since, as Chapter 1 illustrates, the most recent globalization trends affect services more than manufacturing. Especially due to FDI, service sectors are more open than manufacturing sectors; -

G, trade (retail and wholesale); H, hotels and restaurants; I, transportation and communications; J, finance and insurance; K, business services; MNOP personal, cultural and recreational services. 59

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

The sectors identified are instrumental for the second step of the methodology, i.e. to determine which regions have the good specialization in front of globalization. This is easily calculated through a location quotient in open sectors; if it is higher than 1, regions are specialized in these sectors and, ceteris paribus, are expected to get an advantage from their specialization due to the composition of sectors present in the analysis, called MIX effect in a traditional shift-share analysis. On the contrary, the location quotient in open growing sectors being lower than 1, the region is disadvantaged by its specialization.

3.5. Regional taxonomy in front of globalization 3.5.1. Theoretical taxonomy We are now able to build a typology of regions in front of globalization, based on the crossing of the two indicators: the sectoral specialization indicator and the globalization indicator. Table 3.4. reports the logical taxonomy that we obtain if we cross the two indicators. On the horizontal axis we have the regional specialization in open sectors (either service or manufacturing specialisation), expressed by the location quotient. On the vertical axis the globalization index is applied (as built in Section 3.2), distinguishing between regions with an higher or lower than regional average globalisation index. Table 3.4. Regional taxonomy in front of globalization Globalization Index Location quotient in open growing sectors higher than 1 (right specialization) Location quotient in open growing sectors lower than 1 (wrong specialization)

Below average

Above average

1 Regional players

2 Global players

4. Cut-off regions (out of the analysis)

3 Pure gateways

On the basis of these two dimensions, four regional clusters are identified: 1. regional players. These are regions with a good specialization mix but without structural connections with other areas in the world. These regions are therefore expected to take advantage of their specialization, but they are also expected to be somewhat passive, because their good mix does not result from their behaviour, but more from the chance circumstance of being specialized in the right sectors at the right time. 2. global players. These are the regions whose role in globalization is maximum: they are structurally open and have all the connections; moreover, they are specialized in sectors which gain benefit from globalization itself. These regions are therefore those expected to be able to lead Europe and drive patterns of response to globalization also for the other regions of the EU. 3. pure gateways. These are regions whose behaviour is puzzling: they have structural connections other areas in the world, but at the same time they are specialized in closed 60

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

sectors. For this reason, they act as gateways to the world for neighbouring regions which are instead specialized in export sectors. 4. cut-off regions. This is the residual category made up of regions which have neither the structural elements to connect with the world nor the appropriate specialization in open growing sectors. These regions have no chance of playing a role in globalization processes and will hence be omitted from the analysis conducted in the chapters that follow.

3.5.2. The taxonomy applied to European regions The theoretical definition of regions has been applied on actual European Regions at NUTS2 level, using for the globalization index the results of section 3.2 and for the specialization the location quotients obtained from the IGEAT matrix, which contains the employment and value added disaggregated by NACE1 sector for all regions of the EU. We expect empirically to have different behaviours by regions specialized in manufacturing open growing sectors with respect to regions specialized in service open growing sectors. For this reason, instead of calculating one location quotient for the sum of all open growing sectors, we calculated one for manufacturing and one for service. Any region having higher than 1 one (or both) location quotients qualifies to be a regional player or a global player depending on its score in the globalization index. In this way, we are also able to identify regional players specialized in manufacturing open growing sectors, regional players specialized in service open growing sectors and regional players specialized in both. In the same way, we are also able to identify global players specialized in manufacturing open growing sectors, global players specialized in service open growing sectors and global players specialized in both. The typology of regional players is depicted in map 3.3. It can be observed that regional players account for about a half of EU regions. Most of them (91 out of 132 regions) are specialized in manufacturing open growing sectors, and only a smaller number (35 out of 132) in service open growing sectors. Almost negligible (6 regions out of 132) is the presence of regions specialized at the same time in manufacturing and service open growing sectors. Regional players are diffused throughout Europe, and are never the very core of their countries, even if sometimes they are the neighbouring regions of their capitals. The typology of global players is depicted in Map 3.4. Their number is much smaller (58 regions) than the one of regional players, but the regions represented include all major metropolitan and capital areas of Europe. Differently from regional players, these regions are mainly service areas: 29 out of 58 are specialized in service open growing sectors and 10 more are specialized in both service and manufacturing open growing sectors. Empirically, pure gateways appear to be an almost irrelevant category: in fact, only 5 regions belong to this group, confirming the theoretical intuition that this group is somehow improbable. By looking more carefully at the location quotient of these 5 regions, it can be observed that they are very close to 1 (normally higher than 0.95) for service open sectors, whereas they are much lower for manufacturing open sectors. This makes us chose to classify these regions within the service open global players, rising this number from 29 to 34 (Map 3.4). 61

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Interestingly enough, a minority but consistent group of global players is specialized in manufacturing open growing sectors, signalling that manufacturing remains relevant for globalization despite the increasing service trend of the economy. This group accounts for 14 regions our of 58, but if one also considers those regions specialized in both service and manufacturing open growing sectors, one observes that more than 41% of global players are manufacturing specialized.

Map 3.3. Regional players in Europe

Politecnico di Milano - June 2009 Typology of Regional Players Regional players specialized in open service sectors Regional players specialized in open manufacturing sectors Regional players specialized in open service and open manufacturing sectors

62

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Map 3.4. Global players in Europe

Politecnico di Milano - June 2009 Typology of Global Players Global players specialized in open service sectors Global players specialized in open manufacturing sectors Global players specialized in open service and open manufacturing sectors

The empirical inconsistency of the pure gateways allows us to conclude the chapter with the following classification to put forward for the empirical analyses in the next chapters (see Map 3.5): 1. Global players; 2. Regional players; 3. Other regions.

63

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

What is interesting to observe, is that all regions which have higher than average structural openness, are specialized in open growing sectors, signalling that, for global players, productive structure and connection structure go hand in hand.

Map 3.5 Typology of regions in front of globalization

Politecnico di Milano - June 2009 Typology of Regions Facing Globalization Global Players Regional Players Other regions

64

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

3.6. Structural profile of global players In the previous section, we have identified three types of regions in front of globalization. In this section we describe their structure, in order to evidence that their characteristics are considerably different. To do this, we applied an Anova analysis on the means of structural variables over the three groups and found a large number of statistically significant differences. We decided to leave considerations on structural differences among groups of regions to Chapter 6 when the performance is analyzed. The results of the Anova exercise are reported in Table 3.5. First of all, we may observe that the global players have a GDP per person which is considerably higher than the others. Being a global region is hence associated with higher wealth, even if this analysis is not able to test the direction of the causal relationship. Given their higher degree of richness, global players are less keen to be supported by development policies. Structural Funds expenditure per capita is much lower in fact much lower. Global players are also characterized by lower unemployment rates, but also regional players have lower unemployment rates with respect to the other regions. Being specialized in open growing sectors hence appears to be correlated with occupational levels. Concerning the territorial structure, global players include most Megas (i.e. Metropolitan European Growing Areas as defined by Espon project 1.1.1), they are hence those with the stronger urban structure. This outcome was expected since they hold the world connections typical of urban structure. Being global regions mainly urban ones, population density is much higher than in the other regions and the share of rural population is considerably lower. Because of density, also total population is higher in global players, and this is not because of the regional area, which is similar in local and global players whereas the other regions are much larger and rural. Regarding the productive structure, global players are mainly (but not only) tertiary regions, so it was expected to have there higher value added per employee in the tertiary sector, but also in manufacturing the productivity of labour is higher than in the other regions. Also the technological level is higher in global regions, since total factor productivity (estimated as the residual of a regression of GDP over capital and labour) is much higher in global players with respect to the other regions (Map. 3.6). In particular, if we look at the indicators of the presence in the region of the pre-conditions for the knowledge economy, global regions have more human resources in science and technology, whatever measure is used to measure it (core or extended definition). Also, significantly higher is the percentage of people with higher than degree qualification (ISCED 5 and 6).

65

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

Table 3.5. Quantitative characteristics of global and regional players

Structural variables Growth of real GDP (2002 -2005) Regional DIF (2002-2005) Regional GDP per inhabitant (2000, 1995 prices) Regional GDP per inhabitant in PPS (1995, 2005 prices) Regional unemployment rate (2000) Dummy megas (2000) Human Resources in Science and Technology (% of total pop. in 2000) Human Resources in Science and Technology (% of active pop. in 2000) Total Factor Productivity (residual of ln(y)=αln(k)+βln(l) in 2000) Percentage of people with ISCED 5 or 6 instruction Population density (2000) Share of rural population (2000) Rurality index Growth spillovers (2000) Kms. of roads over regional area (2000) Structural fund expenditure per 1000 inhabitant 1994-1999 Population growth (1999-2002) Population growth (1995-2002) Labour productivity (Value added/total employment in 2000) Labour productivity in the primary sector (2000) Labour productivity in the manufacturing sector (2000) Labour productivity in the service sector (2000) Employment growth in the manufacturing sector (1999-2002) Railways (2000) Roads (2000) % of citizens trusting others "a lot" or "quite" (2000) % of citizens who believe others evade taxes "a lot" or "quite" (2000) % of citizens who believe others bribe public officers "a lot" or "quite" (2000) Infrastructure endowment (tr. infr. indicator from ESPON Kten in 2000) Regional population (2000) Overall R&D expenditure (2003) Location quotient (VA) for the primary sector (sectors A/B in 2004) Location quotient (VA) for the mining sector (sector C in 2004) Location quotient (VA) for the manufacturing sector (sector D in 2004) Location quotient (VA) for the electricity sector (sector E in 2004) Location quotient (VA) for the construction sector (sector F in 2004) Location quotient (VA) for the retail trade sector (sector G in 2004) Location quotient (VA) for the hotel and restaurants sector (sector H in 2004) Location quotient (VA) for the transport sector (sector I in 2004) Location quotient (VA) for the financial intermediation sector

Regional globalization typologies Global Regional Other players players regions 2.763 2.465 2.093 0.156 -0.151 -0.410 22296.056 14712.425 13579.491 54881.979 18783.930 20509.137 0.072 0.086 0.105 0.707 0.153 0.093 28.518

20.333

18.129

38.815 29.154 26.700 0.329 0.031 -0.041 1.231 0.892 0.912 740.698 172.923 99.338 0.131 0.338 0.351 0.531 1.421 1.397 15.936 2.057 1.057 0.200 0.109 0.073 159566.190 291983.530 666291.410 0.900 0.566 0.164 0.268 0.192 0.049 39.136 30.799 29.946 22.551 25.724 21.015 43.846 34.140 32.838 41.041 31.914 32.063 -1.192 -0.240 -0.031 0.096 0.060 0.043 0.271 0.155 0.105 0.358 0.285 0.323

F-test 2.05 4.38 18.93 40.670 5.8 56.08

sig. 0.131 0.014 0.000 0.000 0.004 0.000

45.75 0.000 44.25 11.960 17.40 21.71 29.06 8.660 3.730 25.99 13.000 1.830 2.560 6.270 0.180 4.68 5.380 2.640 19.590 15.650 4.660

0.000 0.000 0.000 0.000 0.000 0.000 0.025 0.000 0.000 0.163 0.080 0.002 0.838 0.010 0.005 0.073 0.000 0.000 0.010

2.274

2.235

2.318

1.550 0.214

2.656

2.629

2.573

1.480 0.232

0.189 2909.658 1857.760

0.071 1484.740 406.823

0.044 1597.975 304.419

13.830 0.000 24.050 0.000 45.690 0.000

0.506 0.841

1.573 1.680

2.464 2.848

30.790 0.000 2.710 0.068

0.897 1.025

1.151 1.221

0.947 1.400

11.700 0.000 3.430 0.034

0.896

1.117

1.250

17.230 0.000

1.075

1.013

0.943

4.170 0.017

0.859 1.229 1.261

1.191 0.962 0.645

1.190 0.919 0.627

1.850 0.160 9.940 0.000 51.430 0.000 66

Regions Benefiting from Globalisation and Increased Trade: Final Report 15 October 2009

(sector J in 2004) Location quotient (VA) for the real estate sector (sector K in 2004) Location quotient (VA) for the public administration sector (sector L in 2004) Location quotient (VA) for the education sector (sector M in 2004) Location quotient (VA) for the healthcare sector (sector N in 2004) Location quotient (VA) for the OP sector (sector OP in 2004) Location quotient (VA) for the food and beverages sector (sector DA in 2004) Location quotient (VA) for the textiles and clothing sector (sectors DB/DC in 2004) Location quotient (VA) for the wood sector (sector DD in 2004) Location quotient (VA) for the paper sector (sector DE in 2004) Location quotient (VA) for the oil and chemicals sector (sectors DF/DG in 2004) Location quotient (VA) for the rubber sector (sector DH in 2004) Location quotient (VA) for the non metallic products sector (sector DI in 2004) Location quotient (VA) for the basic metals sector (sector DJ in 2004) Location quotient (VA) for the machinery sector (sector DK in 2004) Location quotient (VA) for the electrical components sector (sector DL in 2004) Location quotient (VA) for the transportation equipment sector (sector DM in 2004) Location quotient (VA) for the manufacturing of miscellaneous manufacturing (sector DN in 2004) Location quotient (VA) for the manufacturing of miscellaneous manufacturing (sectors M/N/O/P in 2004) Location quotient (VA) for manufacturing open growing sectors (in 2004) Location quotient (VA) for services open growing sectors (in 2004) Location quotient (VA) for all open growing sectors (in 2004) Location quotient (VA) for manufacturing high technology sectors (in 2004) Location quotient (VA) for manufacturing medium-high technology sectors (in 2004) Location quotient (VA) for manufacturing medium-low technology sectors (in 2004) Location quotient (VA) for manufacturing low technology sectors (in 2004) Growth of labour productivity in the manufacturing sector (20012005) Growth of labour productivity in the service sector (2000-2005) Herfindahl index for the manufacturing sector (1995) Herfindahl index for the the whole economy (2004) Variation of the Lawrence index in the manufacturing sector (1995-2004) Variation of the Lawrence index in the whole economy (19952004) Number of regions

1.036

0.790

0.761

32.810 0.000

0.971 0.912

1.050 1.099

1.227 1.219

11.150 0.000 20.220 0.000

0.863 1.066

0.992 0.843

1.039 0.753

4.310 0.014 26.320 0.000

0.910

1.327

1.362

10.330 0.000

0.753 0.882 1.102

0.993 1.730 0.827

1.730 1.709 0.775

10.430 0.000 7.380 0.001 8.400 0.000

1.122 0.832

1.111 1.325

0.544 0.814

9.690 0.000 16.580 0.000

0.696

1.387

1.561

16.380 0.000

0.708

1.176

1.111

8.400 0.000

0.773

1.001

0.827

2.710 0.069

0.966

1.005

0.757

2.530 0.082

0.925

1.286

0.466

18.010 0.000

0.797

1.247

1.057

8.100 0.000

0.932

0.986

1.018

2.640 0.073

0.966

1.254

0.695

30.870 0.000

1.049 1.042

0.903 0.932

0.883 0.867

39.150 0.000 58.590 0.000

0.966

1.005

0.757

2.530 0.082

0.947

1.131

0.611

21.970 0.000

0.732

1.248

1.136

15.000 0.000

0.927

1.154

1.235

7.690 0.001

2.242 1.495 0.150 0.109

2.609 1.295 0.144 0.088

2.452 0.868 0.156 0.087

0.153

0.167

0.142

2.910 0.056

0.104

0.132

0.127

5.250 0.006

58

132

75

0.100 2.950 1.060 32.290

67

0.904 0.055 0.347 0.000

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 3.6. Total Factor Productivity in 2000

Politecnico di Milano

Standardized Total Factor Productivity Blank- No data 0.01 - 0.30 0.31 - 0.48 0.49 - 0.65 0.66 - 0.83 0.84 - 1 Global_regions

Source: our elaboration

Also for what concerns infrastructure endowment the global players are different from all other regions. They are more endowed with roads, highways and railways per squared kilometre (Map. 3.7).

68

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 3.7. Infrastructure endowment (roads per sqm in 2001)

Politecnico di Milano

Infrastructure endowment 0.00 - 0.05 0.06 - 0.13 0.14 - 0.28 0.29 - 0.51 0.52 - 0.91 Global_regions

Source: ESPON database

Concerning their demographic structure, population is growing more in global players, probably also because of the fact that (see the globalization index) they attract migration more than other regions. We can finally have a look at the specialization of regions: global players are overall despecialized in manufacturing, which is concentrated in regional players, and also in agriculture, mining and electricity (which all characterize the other regions). Global regions are instead specialized in many service sectors: wholesale and retail trade (G, probably due to the higher purchasing power); transport, storage and communication (I); real estate, renting and business activities (K) (Map 3.8); other community, social and personal service activities and activities of households (O+P). More than on any other sector, the relative specialization of global regions is in financial intermediation (J), being the natural location of headquarters of these activities (Map 3.9). 69

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 3.8 Degree of specialisation in real estate, renting and business activities in 2004

Politecnico di Milano Location quotient of the real estate, renting and business activities industry 0.00 - 0.48 0.49 - 0.69 0.70 - 0.87 0.88 - 1.12 1.13 - 1.59 Global_regions

Source: IGEAT matrix

In the other service sectors, the global players show no clear specialisation patterns, namely hotels and restaurants (H); public administration and defence and compulsory social security (L); education (M); health and social work (N). Global players are hence regions whose economy depends much less than the others on the public sector. Within the manufacturing sectors, global players are de-specialized in any sector, with the exception of manufacture of coke, refined petroleum products and nuclear fuel and manufacture of chemicals, chemical products and man-made fibres (DF+DG) (probably due to the presence of headquarters and research rather than production), whose location quotient is however similar between local and global regions and much lower in the other regions. The other exception is sector DE, manufacture of pulp, paper and paper products, publishing and printing, which is a sector strictly linked to information and culture, which is a feature of large urban areas. 70

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 3.9. Degree of specialisation in financial intermediation in 2004

Politecnico di Milano

Location quotient of the financial intermediation industry 0.00 - 0.46 0.47 - 0.74 0.75 - 1.28 1.29 - 2.28 2.29 - 4.13 Global_regions

Source: IGEAT matrix

Finally, looking at the stability of specialization in global regions over time, we can observe that the specialization structure of global regions is significantly more stable than in all other regions, since the change in specialization measured by the Lawrence index is lower. This is a signal of strength if it is due to endogenous ability to keep value added levels in front of threats, but a signal of weakness if this is due to the un-ability to change specialization to react to technological and demand changes. The first reply is more likely, as it will become evident in Chapter 6 devoted to the analysis of the economic performance. As evidenced in Chapter 1, functions are more important than sectoral specialization in the regional ability to compete in front of globalization. In this section, we show that global regions are also different from the other in the functions that they carry on. The data which more closely allow to

71

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

see the regional functions are those of professions coming from the European Union’s European Labour Force Survey32. Again with Anova testing, we investigated the professions of regional workers on a large number of typologies at 1, 2, 3 digit ISCO-88 (International Standard classification of occupations) classification33. Statistically significant differences arise in professions which can be put in some relation to the functions which regions carry out in the international division of labour. First of all, global regions have a significantly larger share of corporate managers (Isco 12), which is a signal that they carry out command and control functions. The transnational headquarters are in these regions (see Section 3.2) and, as evidenced in Fig. 3.1, this pattern is reflected in the command functions of all corporations (both national and transnational). The change in corporate managers, does not show a significant pattern, even if it is overall increasing, because of the increasing importance of coordination with respect to production. Another important aspect for regional competitiveness is the ability to carry out the highest phases of the production processes, those which need more qualifications and produce higher value added. If we look at the physical, mathematical and engineering science professionals (Isco 21) we see that also in this case these high qualification professions are significantly more diffused in global regions. Also, they are constantly increasing in global regions, while the increase in regional players and the other regions is a more recent trend. The highest production phases need the presence in the region of sufficient technology and innovation. In the impossibility to disentangle the private researchers from the labour force survey, we analyzed the university researchers (Isco 231), whose presence is often considered necessary to complement in-firm R&D. Also these professions are significantly more present in global regions taken as a whole (the tiny sample prevents from obtaining reliable results for the individual regions). Their increase, on the contrary, is a pattern common to all European regions. In Chapter 1 it has also been evidenced the importance of related service activities for the growth of core regional productions. In the labour force survey it is possible to analyze the business professionals (Isco 241, which includes: accountants, personnel and careers professionals, and business professionals not elsewhere classified). These professions are not only considerably more diffused in global regions, but also increasing at a statistically significant faster pace in these regions. A different role is played by the lower classified white collars. These can be measured through office clerks (Isco 41) and customer service clerks (Isco 42). Both are significantly more diffused in global regions, because of the higher incidence of services in these regions. We can however observe that office clerks are decreasing in all Europe, replaced by other higher value added qualifications, and this patterns is stronger in global regions. As far as customer service clerks are concerned, these are overall quite stable, but recently increasing in those regions which are not global players nor regional players. These regions are hence probably attracting these low-scale functions due to lower labour cost and the increasing simplicity of telecommunications. 32

This survey is administered annually, and involves a large number of questions to individuals. Since data are not stratified enough to allow precise annual estimates of regional professions, we use three year means to increase the significance of results. 33 Since the Labour Force survey is not stratified for professions, it is somehow hazardous to use its information at 3digit ISCO level. However, we use here 3-years averages and averages over groups of regions, where the smaller group has 58 members. For this reason, it is much likely that the result are reliable. One first check of this reliability is the stability in time of the patterns evidenced in Figure 3.1. 72

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 3.1. Regional shares in selected professions 1996-2007 12 Corporate managers

21 Physical, mathematical and engineering science professionals 5.0%

7.0%

4.5% 6.0% 4.0% 5.0%

3.5% 3.0%

4.0%

isco_12_9698 isco_12_9901 isco_12_0204 isco_12_0507

3.0%

isco_21_9698 isco_21_9901 isco_21_0204 isco_21_0507

2.5% 2.0% 1.5%

2.0%

1.0% 1.0% 0.5% 0.0%

0.0% global

local

global

other

231 College, university and higher education teaching professionals

local

other

241 Business professionals

0.7%

2.5%

0.6% 2.0% 0.5%

1.5%

0.4%

isco_231_9698 isco_231_9901 isco_231_0204 isco_231_0507

0.3%

isco_241_9698 isco_241_9901 isco_241_0204 isco_241_0507 1.0%

0.2% 0.5% 0.1%

0.0%

0.0% global

local

other

global

41 Office clerks

local

other

42 Customer services clerks 2.5%

14.0%

12.0% 2.0% 10.0%

1.5%

8.0%

isco_41_9698 isco_41_9901 isco_41_0204 isco_41_0507

6.0%

isco_42_9698 isco_42_9901 isco_42_0204 isco_42_0507 1.0%

4.0% 0.5% 2.0%

0.0%

0.0% global

local

global

other

local

other

8 Plant and machine operators and assemblers 12.0%

10.0%

8.0%

isco_8all_9698 isco_8all_9901 isco_8all_0204 isco_8all_0507

6.0%

4.0%

2.0%

0.0% global

local

other

Finally, we found significant differences in the share of blue collars (Isco 8). These are much less diffused in global regions (which are more tertiary and which, also in manufacturing, appear to tend to keep the command functions and delegate the production phases). The dynamics is also significantly different: in front to a nearly stable pattern in regional players and other regions, global players register a significant decrease in blue collars, signalling that they are increasingly delegating the production phases of manufacturing.

73

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

3.7. Towards an interpretation of the performance of global regions: next steps This chapter has allowed us to highlight which regions in Europe are open regions, and therefore have potentially higher probability to benefit from globalization. The so called “global players” are those regions that have full potentialities to benefit from globalization processes, both in terms of specialization in dynamic and open sectors, and in terms of physical attractiveness and accessibility. In the next chapters we are interested in analyzing the way in which global regions are able to face globalization processes with respect to other kinds of regions. In particular, we are interested in analyzing how and to which extent the major globalization trends highlighted in Chapter 1 - like the relocation of functions and tasks instead of sectors; the increase of service off-shoring and outsourcing instead of manufacturing off-shoring; deindustrialization in favour of services in manufacturing areas; decentralization of inter-twined functions (manufacturing and related services) - take place more in global than in regional players or other regions. Moreover, our attention is focused on whether global regions are more able than other kinds of regions to find the necessary balance between high and low level functions, to attract high-level service functions, to innovate despite the growing service activities; in other words, whether global regions are more able than other regions to turn threats stemming from globalization processes into growth opportunities. To achieve this goal, we proceed according to some logical steps. We first analyze the spatial patterns of globalization mentioned in Chapter 1, and see whether these patterns are present under certain features in global vs. regional players. In particular, we analyze in Chapter 4: -

if major globalization patterns develop differently in global regions with respect to other kinds of regions; if global regions are able to adapt their economies to globalization trends differently with respect to other kinds of regions. In this case, we are able to highlight which kinds of regions risk in front of some particular trends.

A particular analysis is devoted to the capacity of the different typologies of regions to attract FDI. Regional spatial patterns of FDI are analyzed and specificities of single typology of regions (global players vs. regional players) analyzed (Chapter 5). Chapter 6 analyzes the performance of global players in terms of employment and labour force composition. Chapter 7 is devoted to the analysis of the determinants of the economic performance at regional level, with the aim to determine whether particular success factors explain the economic behaviour of global regions. To this end, also case study analyses will be run in order to identify in details the replicable or non-replicable factors of economic success. Chapter 7 of this report contains the logical structure of the case studies. Chapter 8 of this report presents the scenario methodology, the scenario building and the scenario results. Chapter 9 of the report is devoted to the policy implications of the whole project.

74

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

4. Spatial patterns of globalization trends 4.1. Introduction The main aim of this chapter is to describe the main globalization trends at a spatial level. In particular, in this empirical analysis we are interested to highlight the geographical patterns of the mega trends that occur nowadays, and that the literature highlights as the main threats that economic systems have to face. The main globalization trends have been highlighted in Chapter 1, and can be briefly recalled here: -

-

-

a new globalization paradigm, from sectoral to functional and task international reallocation. In particular, we will investigate if it is true that local functions are preconditions for benefiting from globalization; the next global shift, intended as the increase of service activities. Here our interest mainly lies in whether the presence of some particular local skills attracts functions based on these skills; if this is true, this would not assure a balance between low and high value functions, but represents a cumulative and self-reinforcing process in the international divisions of tasks at the regional level which does not favour balanced opportunities of growth among European regions; deindustrialization and the rising of services economy. In this case we are interested to see whether the location of manufacturing is influenced by the location of the service activities; decentralization of inter-twined functions, an in particular whether manufacturing activities are attracted by innovative environments where they have higher probability to increase their own innovation activities.

Our interest is to describe: -

-

whether these trends take place with different intensity in different types of regions. We expect that global players, more exposed to globalization, are more affected by these trends than regional players and other regions; whether the structural preconditions for turning these trends into growth opportunities are present in the different regions. Especially the presence of high value functions is expected to attract other high value functions, in a cumulative self-reinforcing process. At the same time, the attraction of high value functions is expected to benefit regions more than the presence of low value functions; moreover, the attraction of low value function is expected to generate employment growth, but not productivity gains in the long term.

These elements are difficult to be captured empirically. Difficulties lie in the availability of good proxies for functions, and in the distinction between local endogenous effects and effects due to globalization trends. For what concerns the first difficulty, the analysis is based on the labour force survey data on professions, available at NUTS 2 for 2 or even 3 digit code disaggregation. These data represent the only available data; they require however some attentions when applied since the sample is not regionally stratified. We applied an average data of a three year period, and at 2 digit code. Changes between two periods is used, but with cautiousness. For what concerns the distinctions between endogenous changes in the local economy and local changes due to globalization effects, nothing specifically can be done to separate out these effects. The logic behind our analysis is however that the globalization effects should be concentrated in global, and to a lesser extent in regional players, than elsewhere, given their natural exposure to 75

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

globalization. Therefore, all results are analyzed particularly by comparing global vs. local and other regions. Finally, in this chapter the analysis of the trends and of the structural preconditions necessary to afford globalization processes are described through statistical methodology like Anova or correlation indices, and the main results mapped. No cause effect chain is included in this kind of methodologies, but variables are correlated taking into consideration a time lag which expresses a cause-effect logic.

4.2. Globalization trends in different groups of regions In this section we analyze some important globalization trends mentioned in Chapter 1, namely the shift towards services activities and the different productivity increases in manufacturing and services. In addition to global forces, these trends are also pulled by local endogenous forces. We are not therefore able to distinguish how much they take place because of the one or because of the other. However, we will show that they take place differently in regions with different exposure to globalization, in particular in the three groups identified in Chapter 3 of the report, and will therefore be allowed to make some conjectures on the link between globalization and these trends. In Fig. 4.1a, the patterns followed by the three main indicators of regional growth, namely employment, productivity and value added are represented for the period 1995-2005. Only two of them are really independent, and data for the third one can always be found by combining the other two. The three indicators are represented in the lines of the figure, whereas vertically, Fig. 4.1 represents the same indicators for the three types of regions: the global players, the regional players and the other regions. Each indicator is represented separately for manufacturing and services, to capture the different patterns followed by the two major sectoral activities of the economy, being the third one, agriculture, a marginal industry in advanced economies. All patterns are represented as indices with year 1995 used as numeraire and put to 100 in order to allow comparisons between different regional groups and different variables. By looking at employment growth, we can observe a clear shift towards services activities: employment has been growing in services at a fast pace, whereas employment in manufacturing has been growing much less, and has often decreased. The same negative dynamics is registered in the agricultural sector, though the trend seems to be affected by some data inconsistency, given the sudden strong downturn in 2001. By looking at the globalization characteristics of regions, we observe that services employment growth has been growing sooner and more sizeably in global players, whereas for regional players and the other regions the growth is consistent but less sizeable. Manufacturing employment, on the contrary, has been decreasing in both global and regional players at a comparable pace. Being specialized in open growing sectors is hence not enough to maintain manufacturing employment levels. In other regions than global and regional players, manufacturing employment has remained substantially stable.

76

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The gap between the two indicators can represent the shift towards services activities of the regional economies; it is easy to observe that it is much stronger in global players with respect to regional players and, even more, with respect to the other regions. We can hence assume that the new shift in the globalization process towards the services sector in the European economy is taking place, hitting much more global than other regions. The second line of Fig. 4.1a represents productivity growth. We can first observe that, in all types of European regions, manufacturing productivity has been growing more than services productivity. The slow growth of services productivity, sector accounting for almost all employment growth, is a signal that there is an ongoing process of manufacturing restructuring and that the services sector also acts as a channel of jobs creation; however, this increase takes place at the expenses of productivity gains, taking place in low qualified jobs. This is supported by the observation of the agricultural sector, where restructuring is strongest and the decrease of jobs induces high productivity increases but total value added decreases. Productivity increases have been larger in global players both in services (where they most grew in employment) and manufacturing. Manufacturing productivity has been growing fast also in regional players, which are specialized in open growing sectors, whereas productivity increases have been consistently lower in the other regions, signaling that, at least apparently, the globalization forces are a driver of productivity increase either because of technological transfer or because of the competition that forces the most open regions to react by increasing productivity. In the third line of Fig. 4.1a the patterns of value added are represented. These patterns can be simply obtained by multiplying employment and productivity patterns; however they are interesting because they allow to understand if some effects are purely statistical or they hide important economic trends. In fact, productivity increases can be obtained by cutting the less productive jobs, and in this case total value added would also be decreasing; however, it is also possible that, due to technological or organizational innovations, the restructuring process can achieve higher total value added with lower employment levels. It is the second possibility the one which appears to take place in manufacturing, in both global players and regional players: total value added has been increasing thanks to productivity increases and despite employment decreases. Globalization forces might have played a role in this respect; despite the same increase of total value added, the pattern followed by the non global other regions is completely different: much lower productivity increases and a maintenance or even and increase in employment levels. Non-global regions, de-specialized in open growing sectors, have hence been the collector of manufacturing lower production phases, whereas the most open regions have had to shift to phases of higher value added and cut and delocalize the lower phases. The patterns followed by services are much simpler: global players have been growing faster in both employment and productivity, so that also total value added has been growing there more than in any other group of regions. Second come the regional players, which had slower services employment growth and services productivity growth, implying lower total value added growth. Latest the other regions, with even lower value added growth coming from much lower productivity growth and also lower employment growth. Hence, global players appear to have benefited from their position and socio-economic structure in order to develop their services activities more, especially by capturing larger shares of the economy in this sector and the higher value added functions. The other regions, on the contrary, have experienced a lower services employment growth, limited to the lowest value-added functions, those that can be more easily decentralized. 77

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

To evidence possible differences between the two main areas of the European Union, the same analysis has been done separately for regions belonging to the Old 15 member countries and to the 12 New Member countries. These results are presented respectively in Figures 4.1b and 4.1c. If the trends followed by regions belonging to the Old 15 countries (Fig. 4.1b) are almost identical to the ones of the whole EU, due to the much larger weight of these countries in terms of population and economy, for the trends in New 12 member countries (Fig. 4.1c), some differences emerge. First of all, a lower employment growth in services, for all groups of regions, but in particular for regional players and the other regions, where tertiarization appear to be still a recent and non much relevant process (Fig. 4.1c, on top). The second difference concerns the loss of manufacturing employment, which is in the New 12 countries slighly larger than in the Old 15 countries for global and regional players, but for the other regions, which in the West were maintaining their industry employment level, there is a decisive decrease of industry employment. The third difference concerns the loss of agricultural employment, which is stronger in the East than in the West but remains stronger for regional players. The patterns followed by productivity are even more markedly different in the East with respect to those in the West (Fig. 4.1c, middle). The patterns followed by industry productivity are much steeper in the New 12 member countries with repect to their Old 15 counterparts. In particular, this is true for Eastern global players, which not only outperform Western global players, but also outperform considerably Eastern regional players, whereas in Fig. 4.1b it could be observed a much more similar pattern between Western global and regional players. Also in the East, in any case, the other regions are the worst performing in terms of industry productivity, which leads to the conclusion that the process of manufacturing restructuring in Eastern regions has been intensively linked to globalization. Also service productivity has been growing much more in the East than in the West in the 10 yearlong period of analysis (Fig. 4.1c), and also in this case, this process has been much more evident in global players, whereas Eastern regional players and other regions are almost at the same level, though still larger than in the West. Services in the East, therefore, have not only been creators of employment, but have also considerably contributed to the increase of productrivity. Finally, though with kinks which put some doubts on the full reliability of data, agricultural productivity which has increased in the east more than in the West but less than the indutry one. In this case, it appears that the process of agricultural restructuring has been fast in global and regional players, whereas in the other regions, where less employment has been lost, lower productivity increases have been achieved. The last indicator to confront is the one of value added (Fig.4.1c on bottom): the growth of value added, in services and in manufacturing - but not in agriculture - has been much faster in the East than in the West of Europe. In services and, differently than in the West, also in manufacturing, Eastern global players have considerably outperformed regional players, not to mention the lagging other regions. Global players in New 12 countries, therefore, not only have led their countries through a tertiarization process, but have also led them with manufacturing growth, something which has not at all happened in Old 15 global players, which on the contrary have outperformed the rest of regions only in services. 78

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 4.1a Globalization trends in different groups of regions Regional Players

Global Players

Other regions

130.00

130.00

130.00

120.00

120.00

120.00

110.00

110.00

Industrial employment Service Employment Agricultural Employment

100.00

110.00

Industrial employment Service Employment Agricultural Employment

100.00

90.00

90.00

90.00

80.00

80.00

80.00

70.00

70.00 1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

1995

2005

1996

1997

1998

1999

2000

2001

2002

2003

2004

70.00

2005

1995

150.00

140.00

140.00

140.00

130.00

130.00

Industrial productivity Service Productivity Agricultural Productivity

120.00

120.00

2000

2001

2002

2003

2004

1995

2005

1996

1997

1998

1999

2000

2001

2002

2003

2004

135.00

130.00

130.00

125.00

125.00

125.00

120.00

120.00

Industrial value added Service Value added Agricultural Value Added

110.00

1998

1999

2000

2001

2002

1998

1999

2000

2003

2004

2005

2001

2002

2003

2004

2005

115.00

Industrial value added Service Value added Agricultural Value Added

110.00

95.00

90.00

90.00 1997

1997

100.00

95.00

95.00

1996

1996

105.00

100.00

100.00

2005

120.00

115.00

105.00

105.00

2004

Other regions

130.00

110.00

2003

Industrial productivity Service Productivity Agricultural Productivity

1995

135.00

Industrial value added Service Value added Agricultural Value Added

2002

90.00

2005

135.00

115.00

2001

120.00

Regional players

Global players

1995

2000

100.00

90.00

90.00 1999

1999

110.00

100.00

100.00

1998

1998

130.00

Industrial productivity Service Productivity Agricultural Productivity

110.00

110.00

1997

1997

Other regions

150.00

1996

1996

Regional players

Global players 150.00

1995

Industrial employment Service Employment Agricultural Employment

100.00

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

90.00 1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

79

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 4.1b Globalization trends in different groups of regions – Regions belonging to Old 15 member countries. Regional Players

Global Players

Other regions

130.00

130.00

130.00

120.00

120.00

120.00

110.00

110.00

Industrial employment Service Employment Agricultural Employment

100.00

90.00

80.00

110.00

Industrial employment Service Employment Agricultural Employment

100.00

90.00

90.00

80.00

80.00

70.00

70.00

70.00 1995

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

1995

150.00

140.00

140.00

140.00

130.00

130.00

130.00

Industrial productivity Service Productivity Agricultural Productivity

120.00

110.00

100.00

Industrial productivity Service Productivity Agricultural Productivity

120.00

110.00

100.00

100.00

90.00

2000

2001

2002

2003

2004

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

1995

135.00

130.00

130.00

130.00

125.00

125.00

125.00

120.00

120.00

120.00

115.00

115.00

Industrial value added Service Value added Agricultural Value Added

110.00

105.00

100.00

95.00

Industrial value added Service Value added Agricultural Value Added

110.00

90.00 2000

2001

2002

2003

2004

2005

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

Industrial value added Service Value added Agricultural Value Added

110.00

105.00

100.00

100.00

95.00

95.00

90.00 1995

1999

2004

115.00

105.00

90.00

1998

2003

Other regions

135.00

1997

2002

Industrial productivity Service Productivity Agricultural Productivity

Regional players

Global players

1996

2001

2005

135.00

1995

2000

90.00 1995

1999

1999

120.00

110.00

90.00 1998

1998

Other regions

150.00

1997

1997

Regional players

Global players

1996

1996

2005

150.00

1995

Industrial employment Service Employment Agricultural Employment

100.00

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2005

80

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 4.1c Globalization trends in different groups of regions – Regions belonging to New12 member countries. Regional Players

Global Players

Other regions

130.00

130.00

130.00

120.00

120.00

120.00

110.00

110.00

Industrial employment Service Employment Agricultural Employment

100.00

90.00

80.00

110.00

Industrial employment Service Employment Agricultural Employment

100.00

90.00

90.00

80.00

80.00

70.00

70.00

70.00 1995

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

1995

190.00

180.00

180.00

180.00

170.00

170.00

170.00

160.00

160.00

160.00

150.00

150.00

Industrial productivity Service Productivity Agricultural Productivity

140.00 130.00 120.00 110.00 100.00

1999

2000

2001

2002

130.00

130.00

120.00

120.00

110.00

110.00

100.00

100.00

2003

2004

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

170.00

170.00

165.00

165.00

165.00

160.00

160.00

160.00

155.00

155.00

155.00

150.00

150.00

150.00

145.00

145.00

140.00

140.00

Industrial value added Service Value added Agricultural Value Added

135.00 130.00 125.00 120.00 115.00 110.00 105.00 100.00 95.00 90.00 2002

2003

2004

2005

1997

1998

1999

2000

2005

2001

2002

2003

2004

2005

140.00

135.00 130.00

Industrial value added Service Value added Agricultural Value Added

135.00 130.00

125.00

125.00

120.00

120.00

115.00

115.00

110.00

110.00

105.00

105.00

100.00

100.00 95.00

90.00

2001

2004

145.00

Industrial value added Service Value added Agricultural Value Added

95.00

2000

2003

Other regions

170.00

1999

1996

Regional players 175.00

1998

2002

Industrial productivity Service Productivity Agricultural Productivity

1995

175.00

1997

2001

90.00 1995

2005

Global players

1996

2000

140.00

175.00

1995

1999

150.00 Industrial productivity Service Productivity Agricultural Productivity

140.00

90.00

90.00 1998

1998

Other regions

190.00

1997

1997

Regional players

Global players

1996

1996

2005

190.00

1995

Industrial employment Service Employment Agricultural Employment

100.00

90.00 1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

81

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The last difference arises for regional players, which in the West have been shifting towards services whereas in the East have been growing more in manufacturing than in services.

4.3. Regional adaptation to global trends 4.3.1. A new globalization paradigm: local functions as conditions for benefiting from globalization As seen in Chapter 1, globalization no longer concerns sectors and activities but functions and tasks. In the impossibility to get data on tasks, we are however able to investigate the professions, which are a good proxy for the tasks performed in each region due to the international division of labour. Among structural elements, especially stressed in the literature is the role played by command and control functions, which should allow regions to attract the highest level tasks and phases of production, attracted by local high skilled labour force availability. In other words, the presence of high value functions represents one of the conditions to turn globalization threats into growth opportunities, allowing to capture the highest and most productive functions and tasks of the new international division of labour. In this part of the analysis we describe the relationship between regional preconditions that are in theory claimed to be necessary for a region to get advantages from the new globalization trends and the economic performance of regions. We expect these linkages to be stronger in global players, since they are more exposed to globalization. We can approximate the command and control functions with the share of corporate managers (Isco 12)34. It can be observed in Map 4.1 that their distribution is not uniform across Europe, being higher in regions whose colours are green. Interestingly enough, these regions are not only the capital ones but also other second level regions. Once the presence of command and control functions is correlated with tertiary productivity (with a time span), we can speculate on the fact that the presence of command and control functions positively act on service productivity growth and therefore on growth opportunities. The relationship between the presence of corporate managers and the growth of private tertiary productivity35 is depicted in Map 4.1. We can observe (and check with the statistical results in Table 4.1) that there is a significant relationship between the share of corporate managers in a region and the growth of services productivity. Interestingly enough, this relationship is even more positive for global players. Once a sectoral breakdown in service activities is taken into consideration, the relationship of command and control functions with tertiary productivity, which is on aggregate significant, becomes especially significant within the sub-groups of regions which are specialized in sectors G (Wholesale and retail trade; repair of motor vehicles, motorcycles and personal and household goods), H (Hotels and restaurants) and, more than in any other specialization, K (Real estate, 34

Due to the limits of the labour force survey, which is not stratified for professions, we only used professions with a minimun threshold of employment, i.e. we restricted ourselves to the two digit level apart from a few cases in which the three digit profession contains a consistent percentage of employment. Data, however, were not that bad and, since we also used three years averages, data show a significant stability through time, which is an indication of reliability. See the Annexes for a confront of the percentages of people employed in the relevant professions in our three year periods. 35 In all cases, we will restrict statistics to private services when investigating service productivity, employment or value added. Public services value added, in fact, is not determined by the market, but by the political decision on how much to pay public officers, and is hence independent from the forces of globalization, though these forces might contribute to shape the preferences of the political class. 82

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

renting and business activities). This means that the regions specialized in real estate, renting and business activities which have a high share of corporate managers have significantly outperformed their counterparts which are not endowed with high value functions. The positive effect on productivity provided by high value functions is not expected to happen for medium to low value functions, at least in theory. In order to prove whether this general statement holds true, an analysis of the relationship between productivity and other medium to low value functions has been developed. The effect of small and medium enterprise managers, which are persons in charge of smaller scale decisions and hence are not normally linked to the regional command functions, is very different from the one of corporate managers. In particular, at it can be seen in Maps 4.2a and 4.2b, their presence is positively linked with services employment growth and negatively linked to services productivity growth. Medium level functions seem to act on employment creation and not to achieve productivity gains. The correlation between managers of SMEs and private services employment growth (Map 4.2a) is positive and significant overall, but in particular in regional players, which do not have the connections with the external world. On the contrary, in global players, there is no significant relationship, meaning that in these global regions the small and medium enterprises are not the main channel of employment creation. The relationship is strong and significant especially within the regions specialized in sector H (Hotels and restaurants) which are, especially when non global, mainly touristic ones. The correlation between managers of SMEs and private services productivity (Map 4.2b) is on the contrary overall negative and significant and in particular this is true within the regional players. Also in this case, global players do not appear to be affected (neither positively, nor negatively) if SMEs managers are present as the relationship is not significant within the group. The relationship, on the contrary, is very important for those regions which specialize in H (Hotels and restaurants), I (Transport, storage and communication), L (Public administration and defence; compulsory social security) and, more than in any other sector, M (Education).

83

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 4.1. Results of the correlation analysis Map

Variables

General correlation

Correlation by group of regions Other regions

Regional Players

Global Players 0.34 * *

4.1

Quota of corporate managers 1999-01 vs. growth of service productivity 2002-05

0.26 * * *

0.26 * *

0.18 * *

4.2a

Quota of SMEs managers 1999-01 vs. growth of service employment 2002-05

0.22 * * *

0.23 * *

0.31 * * *

-0.07

4.2b

Quota of SMEs managers 1999-01 vs. growth of service productivity 2002-05

-0.23 * * *

-0.35 * * *

0.14

4.3

Quota of clerks 1999-01 vs. growth of service productivity 2002-05

4.4a

Quota of clerks 1996-98 vs. growth quota of clerks 1996-2007

-0.56 * * *

4.4b

Quota of Physical, mathematical and engineering science professionals 1996-98 vs. growth quota of Physical, mathematical and engineering science professionals 19962007

-0.25 * * *

4.4c

Quota of business services professionals 1996-98 vs. growth quota of business services professionals 1996-2007

4.4d

Quota of plant and machine operators and assemblers 1996-98 vs. growth quota of plant and machine operators and assemblers 1996-2007

4.5

Quota of business services professionals 1999-01 vs. growth of manufacturing productivity 2002-05

0.31 * * *

0.16

4.6

Share of human resources in science and technology on population in 2000 vs. growth of manufacturing productivity 200205

0.06

0.11

4.7

Share of human resources in science and technology on population in 2000 vs. growth of manufacturing value added 200205

-0.12 *

-0.06

0.04

0.21 * * *

-0.15 * *

-0.13

0.31 * * *

-0.03

-0.28 * *

-0.62 * * *

-0.46 * * *

-0.58 * * *

-0.18

-0.41 * * *

-0.41 * * *

-0.21 *

-0.47 * * *

0.04

0.27 * *

-0.09

-0.27 * *

0.32 * * *

0.49 * * *

-0.16

0.53 ***

-0.31 ***

0.30 **

*= significant at 10%; **= significant at 5%; ***= significant at 1%.

84

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 4.1. Relationship between command and control functions and service productivity growth

500 km

Politecnico di Milano

4.1 © EuroGeographics Association for the administrative boundaries

Growth of service productivity 2002 - 2005

Quota of corporate managers 1999 - 2001

-7.05 0.47

0.48 2.05

2.05 8.90

0.000 0.029 0.029 0.042 0.042 0.171

No data Global regions

85

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 4.2a. Relationship between medium functions and service employment growth

Map 4.2b. Relationship between medium functions and service productivity growth

86

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

On the other side of the coin there is the related threat for regions which are not able to maintain the highest phases and tasks; the risk is that they become the container of low qualified activities with low value added. This can be observed in Map 4.3, where the productivity growth in private services is correlated with the presence in the region of tertiary functions of medium/low level (Isco 4, clerks). Overall, no significant correlation between the two variables can be observed, but the effect is higly dependent on globalization. In fact, within global players, regions with higher functional specialization in clerks have experienced a significantly lower services productivity growth than the others, meaning that for regions highly exposed to globalization, having a specialization in medium-low level tertiary activities is a weakness. The effect is not present within regional players and is present with opposite sign within the other regions. This means that, for those regions which are excluded by globalization processes, specializing in services of low-medium level can still be a viable development strategy. Interestingly, the relationship is significant and highly positive within the sub-group of regions specialized in sector H (Hotels and restaurants), since maybe for tourist regions the presence of services activities is helpful to complement the touristic offer. It is instead negative and significant for those regions specialized in sector I (Transport, storage and communication), i.e. the sector more related to ICTs.

87

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 4.3. Relationship between productivity growth in services and tertiary functions of medium/low level

500 km

Politecnico di Milano

4.3 © EuroGeographics Association for the administrative boundaries

Growth of service productivity 2002 - 2005

Quota of clerks 1999 - 2001

-7.05 0.47

0.48 2.05

2.05 8.90

0.000 0.091 0.091 0.132 0.132 0.195

No data Global regions

88

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

4.3.2 The next global shift: presence of local skills and attraction of new functions Section 4.2 has already shown a clear trend towards an increase in the service economy, and that globalization is a process strictly linked with it. To face this process, it is important for regions to attract and retain high value functions, and avoid to specialize in the lower phases in order to avoid the loss in productivity gains. The importance of high value functions for the achievement of service productivity gains is already shown in Map 4.1. Regions which are not able to attract these functions are condemned to remain marginal in the global economy. We can here analyze if these processes are cumulative or not, by looking at the regional concentration patterns of a selected number of professions. This is the transposition at regional level of the trends already observed at more aggregate level in Fig. 4.1. In Map 4.4a we represent the process of concentration of medium-low services professions, measured by Isco 41 (office clerks), by analysing the relationship between their average regional share in the period 1996-98 and the growth of their share (Table 4.1). It can be observed that there is a negative correlation overall, implying that those regions with lower presence of these services at the beginning of the period have increased this presence more than the others. The increase in service activities in the economy takes place more in regions less specialised in service, in which the increase of service activities increases faster with respect to regions which are already specialized in services. We can investigate if this process concerns the most skilled services functions rather than only the basic, low value added service functions. To do that, in Map 4.4b we represent the share in 1996-98 versus the change in share of Isco 21 (physical, mathematical and engineering science professionals), who are highly skilled white collars. A negative correlation appears overall, meaning that regions specialized in lower phases are also increasing the skill content of their economic activity. However, this relationship is not significant within the other regions, which are cut-off from globalization processes. For this reason, it appears that, without the impact of globalization forces, regions are less able to modernize their economic structure. Even more interesting is the relationship for business professionals (Isco 241), which represent those advanced services which are more linked with the activities of other businesses. As shown by Map 4.4c, they exhibit a positive and significant concentration pattern overall; a cumulative selfreinforcing process takes place pushing regions which are more endowed becoming even more endowed. The concentration pattern, however, hides very differentiated patterns within the three typologies in front of globalization. In particular, no pattern exists within the regional players, whereas the pattern within the other regions is negative and hence more similar to the one followed by skilled workforce (see Isco 21 above). It is only within the global players that the concentration of business services is positive and significant, inducing that it is when regions are global that tertiary high level support functions concentrate to exploit economies of scale, whereas this is not necessary for those regions which are not affected by the globalization processes.

89

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 4.4a. Concentration of medium-low services professions over time

Map 4.4b. Concentration of skilled services functions (scientists) over time

500 km

Politecnico di Milano

4.5b

Growth quota of Physical, mathematical and engineering science professionals 1996-2007

growth quota of clerks 1996-2007 Quota of Physical, mathematical and engineering science professionals 1996-1998

-0.059 - -0.017 - 0.005 -0.017 0.005 0.054 Quota of clerks 1996-1998

500 km

Politecnico di Milano

4.5a

0.016 0.079 0.079 0.114 0.114 0.196

No data Global regions

-0.017 - 0.003 0.003 0.007

0.007 0.058

0.003 0.014 0.014 0.029 0.029 0.071

No data Global regions © EuroGeographics Association for the administrative boundaries

Map 4.4c. Concentration of business professionals over time

Map 4.4d. Concentration of plant and machine operators over time

500 km

Politecnico di Milano

Politecnico di Milano

4.5c

-0.010 - 0.002 0.002 0.005

0.005 0.033

0.000 0.003 0.003 0.011 0.011 0.037

No data Global regions

500 km

4.5d

Growth quota of plant and machine operators and assemblers 1996-2007 Quota of plant and machine operators and assemblers 1996-1998

Quota of business services professionals 1996-1998

Growth quota of business services professionals 1996-2007

-0.073 - -0.015 - 0.000 -0.015 0.000 0.072 0.037 0.079 0.079 0.101 0.101 0.200

No data Global regions © EuroGeographics Association for the administrative boundaries

90

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Finally, it is also interesting to have a look at the blue collars (Isco 8, plant and machine operators and assemblers), whose decrease represent the other side of the service increase. In Map 4.4d, it can be observed that the overall correlation between the share of plant and machine operators and assemblers decreases more in the regions which initially have more of them, completing the information provided by the evidence on white collars, which grow more in those regions where less service functions are present. However, the blue collar pattern only applies to the global players and the other regions, whereas the correlation is not significant within the regional players. Since regional players are specialized in open growing sectors, and most of these regions also are specialized in manufacturing open growing sectors (see Chapter 3), it is plausible to assume that the deindustrialization process is less strong in those regions which have the best specialization in those sectors in which Europe is competitive worldwide.

4.3.3. Deindustrialization and the rising of services economy: industry and service location patterns Facing deindustrialization is a challenge for regions in a globalized setting. In particular, it is important for regions not to lose manufacturing completely but keep the highest value added phases and functions. This is easier when there is a balance between core manufacturing activities and other related activities, which often belong to the services sector. This effect is hard to capture empirically; our attempt in this section is to highlight the relationships that exist between manufacturing activities and other related activities, belonging to the services sector. In Map 4.5, represented is the relationship between the endowment of business services (Isco 241), in the regions and the growth of manufacturing productivity. It is possible to observe that there is a significant positive relationship between the two, so that the regions with more business service activities have been able to increase more manufacturing productivity, i.e. by keeping the highest phases. The relationship is significant overall, but higher and significant in the global players, signalling that this factor is hence more important for regions which have an higher openness to globalization.The effect is positive and significant within regions specialized in almost all services sectors, with the only exception of those specialized in sector H (Hotels and restaurants). The balance between services and industry is not enough if regions are not able to maintain a certain level of innovation. In Map 4.6 we represented the relationship between the presence of human resources in science and technology in the region and the growth of manufacturing productivity. It is evidenced that having a certain level of innovative capabilities is important for regional productivity growth, overall but especially (higher correlation and higher significance) in global players. Among these regions, those which have the higher human resources in science and technology are also those whose manufacturing productivity growth is higher. Going to regional specialization, this effect is positive and significant in regions specialized in sectors J (Financial intermediation) and K (Real estate, renting and business activities). It hence appears that the effects of innovation on productivity growth are higher in global players and higher in those regions which have a specialization in business related services.

91

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 4.5. Relationship between the endowment of business services and the growth of manufacturing productivity

500 km

Politecnico di Milano

4.6 © EuroGeographics Association for the administrative boundaries

Quota of business services professionals 1999 - 2001

Growth of manufacturing productivity 2002 - 2005 -4.17 0.62

0.62 3.29

3.29 14.59

0.000 0.006 0.006 0.012 0.012 0.052

No data Global regions

92

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 4.6. Relationship between the presence of human resources in science and technology and the growth of manufacturing productivity

Politecnico di Milano

500 km

4.7 © EuroGeographics Association for the administrative boundaries

Share of human resources in science and technology on population in 2000

Growth of manufacturing productivity 2002 - 2005 -4.171 - 1.328 3.616 1.328 3.616 14.592 0.039 0.131 0.131 0.183 0.183 0.318

No data Global regions

93

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

4.3.4. Decentralization of inter-twined functions: the role of innovation in manufacturing location Facing deindustrialization is a challenge for regions in a globalized setting. In particular, it is important for regions not to lose manufacturing completely but keep the higher value added phases, guaranteeing a continuity also in both the core activities, in knowledge and in local competences. The processes of off-shoring and out-sourcing increasingly take place through the decentralization of inter-twined functions, so that not only firms decentralize lower phases of production but often they move various production phases, often including the conception of the products i.e. R&D phases. R&D phases are more easily decentralized in regions where there are research institutions; the latter allow in fact direct interaction with firm R&D offices and also make available locally a cohort of highly educated people which can be hired by the firms to perform research tasks. Map 4.7 represents the share of people in research and development related activities correlated with the increase of total manufacturing value added over the following period. There is a slightly negative and weakly significant relationship overall, partly due to the fact that any statistics of human resources in science and technology fail to capture the innovation activities actually carried out by the employees. Most interesting, however, is the fact that this correlation is strongly different among groups of regions. In fact, there is no correlation within the other regions, there is a negative and significant relationship between regional players but there is a strong and highly significant relationship within the global players. This allows us to speculate that the relationship between the presence of human resources in science and technology in the area and the growth of manufacturing value added is due to the capability of technologically endowed regions to attract manufacture, and not only to the local effects of R&D on firms productivity. Should the effect be only linked to local factors and not also to the decentralization of functions, in fact, the correlation should be present within all three groups of regions and not only within the global players. Finally, it is interesting to observe that also the correlations of human resources in science and technology with manufacturing GVA growth are linked to the regional specialization. In particular, it is interesting to observe that this correlation is negative and significant for those regions specialized in sector O+P (Other community, social and personal service activities + Activities of households), and this is not surprising since this is obviously not a technologically advanced service sector.

94

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 4.7. Relationship between the presence of human resources in science and technology and the growth of manufacturing value added

Politecnico di Milano

Share of human resources in science and technology on population in 2000

Growth of manufacturing value added 2002-2005

500 km

4.8 © EuroGeographics Association for the administrative boundaries

-4.072 - 0.408 2.832 0.408 2.832 12.930 0.039 0.131 0.131 0.183 0.183 0.318

No data Global regions

95

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

4.4. Conclusions The main messages that stem out of this analysis confirm most of our expectations since: -

-

-

-

-

-

a clear shift towards services activities is taking place: employment has been growing in services at a fast pace, whereas employment in manufacturing has been growing much less, and has often decreased, and this trend is more sizeably in global players, whereas for regional players and the other regions the growth is consistent but less sizeable; the shift towards services activities of the regional economies is much stronger in global players with respect to regional players and, even more, with respect to the other regions. We can hence assume that the new shift in the globalization process towards the services sector in the European economy is taking place, hitting much more global than other regions; the slow growth of services productivity, sector accounting for almost all employment growth, is a signal that there is an ongoing process of manufacturing restructuring and that the services sector also acts as a channel of jobs creation; however, this increase takes place at the expenses of productivity gains, taking place in low qualified jobs; productivity increases have been larger in global players both in services (where they most grew in employment) and manufacturing. Manufacturing productivity has been growing fast also in regional players, which are specialized in open growing sectors, whereas productivity increases have been consistently lower in the other regions, signaling that, at least apparently, the globalization forces are a driver of productivity increase either because of technological transfer or because of the competition that forces the most open regions to react by increasing productivity; manufacturing total value added has been increasing thanks to productivity increases and despite employment decreases. Globalization forces might have played a role in this respect; despite the same increase of total value added, the pattern followed by the non global other regions is completely different: much lower productivity increases and a maintenance or even and increase in employment levels. Non-global regions, de-specialized in open growing sectors, have hence been the collector of manufacturing lower production phases, whereas the most open regions have had to shift to phases of higher value added and cut and delocalize the lower phases; the patterns followed by services are much simpler: global players have been growing faster in both employment and productivity, so that also total value added has been growing there more than in any other group of regions.

Hence, global players appear to have benefited from their position and socio-economic structure in order to develop their services activities more, especially by capturing larger shares of the economy in this sector and the higher value added functions. The other regions, on the contrary, have experienced a lower services employment growth, with also lower productivity growth (Fig. 4.1). A more in-depth analysis on the capacity of some regions to grasp globalization opportunities led us to the following messages. For what concerns local functions as structural preconditions for benefiting from globalization, we mainly found that: -

the relationship between the presence of corporate managers and the growth of tertiary productivity turns out to be positive and significant for global regions; the positive effect on productivity provided by high value functions is not expected to happen for medium to low value functions. The presence of small and medium enterprise 96

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

-

-

managers is positively linked with services employment growth and negatively linked to services productivity growth. Medium level functions seem to act on employment creation and not to achieve productivity gains; the correlation between managers of SMEs and services employment growth is positive and significant overall, but in particular in the regional players, which do not have the connections with the external world. On the contrary, in global players, there is no significant relationship, meaning that in these global regions the small and medium enterprises are not the main channel of employment creation; the correlation between managers of SMEs and services productivity is on the contrary overall negative and significant and in particular this is true within the regional players. Also in this case, global players do not appear to be affected (neither positively, nor negatively) if SMEs managers are present as the relationship is not significant within the group.

On the other side of the coin there is the related threat for regions which are not able to maintain the highest phases and tasks; the risk is that they become the container of low skilled activities with low value added. In fact: -

-

-

productivity growth in services is negatively correlated with the presence in the region of service functions of medium/low level, implying that the regions whose main functional specialization is in clerks have experienced a significantly lower services productivity growth than the others. This effect appears to be somehow linked with globalization, because it is not present in the other regions, but is strong and significant in those regions directly affected by globalization ; the growth of tertiary employment has a strict link with the growth of low tertiary functions in global players, whereas the relationship is not significant overall or in the other regional groups, although it is nearly significant in the regional players. It seems that the more regions are exposed to globalization, the more they risk attracting low-skilled tertiary functions when they want to increase their employment. the relationship between the growth of services productivity and the growth of customer services clerks is negative and significant overall, but this is especially so within the global players group.

For what concerns the presence of local skills and attraction of new functions, the following results emerge: -

-

-

a process of concentration of medium-low services professions, measured through the relationship between their average regional share in the period 1996-98 and the growth of their share, which shows a negative correlation overall, implying that those regions with lower presence of these services at the beginning of the period have increased this presence more than the others. The increase in service activities in the economy takes place more in regions less specialised in service, in which the increase of service activities increases faster with respect to regions which are already specialized in services; a negative correlation between the presence of physical, mathematical and engineering sicne professionals and their growth rate, meaning that regions specialized in lower phases are also increasing the skill content of their economic activity. However, this relationship is not significant within the other regions, which are cut-off from globalization processes. For this reason, it appears that, without the impact of globalization forces, regions are less able to modernize their economic structure; it is only within the global players that the concentration of business services is positive and significant, inducing that it is when regions are global that tertiary high level support 97

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

-

functions concentrate to exploit economies of scale, whereas this is not necessary for those regions which are not affected by the globalization processes. finally, a decrease in the share of blue collars (whose decrease represent the other side of the service increase), more pronounced in those regions which initially had more of them, completing the information provided by the evidence on white collars, which grow more in those regions where less service functions are present. However, the blue collar pattern only applies to the global players and the other regions, whereas the correlation is not significant within the regional players; it is plausible to assume that the deindustrialization process is less strong in those regions which have the best specialization in those sectors in which Europe is competitive worldwide.

Facing deindustrialization is a challenge for regions in a globalized setting. In particular, it is important for regions not to lose manufacturing completely but keep the highest value added phases and functions. This is easier when there is a balance between core manufacturing activities and other related activities, which often belong to the services sector. This effect is hard to capture empirically; our attempt in this section is to highlight the relationships that exist between manufacturing activities and other related activities, belonging to services sector. In fact: -

-

-

there is a significant positive relationship between the endowment of business services, in the regions and the growth of manufacturing productivity, so that the regions with more business service activities have been able to increase more manufacturing productivity, i.e. by keeping the highest phases. The relationship is significant overall, but higher and significant in the global players, highlighting that this factor is hence more important for regions which have an higher openness to globalization; the growth of business services is linked to the growth of manufacturing productivity. The relationship is particularly significant within global players; on the contrary, is not significant in regional players and in the other regions, highlighting that in dynamics, this process is even more linked to globalization trends. This process is particularly significant in those regions which are specialized in business related services, i.e. sector K (Real estate, renting and business activities); the balance between services and industry is not enough if regions are not able to maintain a certain level of innovation. The relationship between the presence of human resources in science and technology in the region and the growth of manufacturing productivity is positive. It is evidenced that having a certain level of innovative capabilities is important for regional productivity growth, overall but especially (higher correlation and higher significance) in global players.

98

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

5. Spatial FDI patterns 5.1. Introduction This chapter focuses on spatial patterns of Foreign Direct Investment (FDI). In particular, we are interested in understanding whether and to what extent the main trends of globalization and their spatial patterns have been driven by FDI. This implies to explore mutual interrelations between globalization and foreign direct investments. As discussed in previous chapters, the main globalization trends directly affecting FDI can be summarized as follows: -

-

-

-

Recent liberalization has made possible for Multinational Enterprises (MNEs) to enter markets that were previously protected. At sectoral level, this phenomenon has interested mainly the service sectors, contributing to a generalized de-industrialization of many economies in the world. This opening process has given MNEs more opportunities for business, but also made them more selective in the choice of the potential investment sites; Globalisation has affected MNEs themselves. In particular, their ownership advantages have changed with technical progress and shrinking economic space. The deployment of new technologies, to take advantages of changing costs, as well as logistic needs and specific market demands have become more important. Moreover, because of the complexity of information flows, MNEs had to organize their activity and management differently, changing the relationships with clients, suppliers and competitors to manage better the process of technical change and innovation. Given the new context, MNEs have also changed their portfolio of mobile assets in order to better match with the immobile assets of different locations. These portfolios now include some functions that create ownership advantages, such as R&D, training and strategic management. While a large domestic market remains an important factor to attract FDI, MNEs servicing global markets look for other attributes, world class infrastructures, skilled and productive labour and agglomeration of efficient suppliers, clients, and competitors, as well as support institutions and services.

Given these issues, we focus on the distribution of FDI across European regions and the factors explaining it. In particular, we are interested in understanding: -

-

-

Whether spatial patterns of FDI differ across different types of regions, sectors and sources of foreign investments. We expect that global regions, because of their intrinsic characteristics, attract a consistent share of FDI either from outside or inside Europe. Moreover, we expect that most of these new foreign firms operate in service rather than manufacturing sectors. Whether FDI has affected main trends of globalization. Since FDI is one of the main drivers for globalization, we expect that most of the trends detected in chapter 4 occur, or magnify in presence of FDI. Whether and to what extent traditional FDI determinants are still at work. We expect that their capacity to attract FDI is stronger in regional players and other regions than in global regions, where we expect that other factors of attraction, directly linked to globalization trends, become relevant.

We start our analysis by examining possible patterns of FDI concentration across and within groups of regions, i.e. global regions, regional players and other regions. This simple exercise allows us to 99

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

address the first research objective mentioned before and have a synthetic picture of the role played by each group of regions in FDI tournament. Given the multiple dimensions of the dataset at the base of our analysis, we also analyse whether and to what extent spatial patterns of FDI differ across sectors and according to source country of foreign firms. Table 5.1 summarizes the results. Table 5.1 Breakdown of regions by FDI (number of regions) FDI FDI below the mean above the mean

Total

total FDI other regions

68

7

75

regional players

116

15

131

global regions

18

35

53

Total

202

57

259

other regions

69

6

75

regional players

122

9

131

global regions

18

35

53

Total

209

50

259

other regions

68

7

75

regional players

114

17

131

global regions

17

36

53

Total

199

60

259

other regions

72

3

75

regional players

119

12

131

global regions

19

34

53

Total FDI in manufacturing

210

49

259

other regions

61

14

75

regional players

93

38

131

global regions

14

39

53

Total

168

91

259

extra-Europe FDI

intra-Europe FDI

FDI in services

Note: Global regions are 58. Five of them have been excluded because they coincide with national boundaries. See footnote 1 for further explanations.

On average, regions with a concentration of FDI above the mean are only 57 out of 259, that is, about 22% of the all sample. 36 This share increases up to 35% when only FDI in the manufacturing 36

Since we can not exclude that country-specific effects may affect the distribution of FDI across European regions, the number of foreign firms per millions of inhabitants has been weighted by the corresponding national means. This implies that smaller countries for which it was not possible to use a NUTSII regional classification have been dropped 100

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

sectors is considered, while it remains almost unchanged in the case of FDI in the service sectors, which, therefore, appears to be more spatially concentrated than FDI in manufacturing. Also, intraEurope FDI is more dispersed than extra-Europe FDI, given that a larger number of regions shows a concentration of FDI above the mean. As expected, most of the regions with a concentration of FDI above the mean are global regions (60% on average). Not surprisingly, this percentage is higher for extra-Europe FDI and for FDI in services (about 70% each), and lower (43%) for FDI in manufacturing. As far as regional players are concerned, they seem to be attractive for FDI in manufacturing only. They in fact represent about 42% of regions with a concentration of this kind of FDI above the mean, a share very similar to that of global regions. Although 44 out of 53 global regions locate in Western Europe, their dominant role in FDI tournament remains almost unchanged in New Member States (NMS), as it is shown in Table. 5.2. Global regions, in fact, still represent about 60% of regions with a concentration of FDI above the mean, with the only exception of FDI in manufacturing.

Table 5.2 Breakdown of regions by FDI: former EU15 vs. NMS (number of regions) EU15 regions FDI below the mean

FDI above the mean

total FDI others local global Total

54 85 16 155

extra-Europe FDI others local global Total

NMS regions

Total

FDI below the mean

FDI above the mean

Total

7 10 28 45

61 95 44 200

14 31 2 47

0 5 7 12

14 36 9 59

56 87 16 159

5 8 28 41

61 95 44 200

13 35 2 50

1 1 7 9

14 36 9 59

intra-Europe FDI others local global Total

55 83 15 153

6 12 29 47

61 95 44 200

13 31 2 46

1 5 7 13

14 36 9 63

FDI in services others local global Total

58 87 17 162

3 8 27 38

61 95 44 200

14 32 2 48

0 4 7 11

14 36 9 63

FDI in manufacturing others

48

13

61

13

1

14

by the sample. We refer here to the three Baltic Republics, Luxembourg and Cyprus. Malta was not considered because we have no data on FDI. 101

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

local global

71 13

24 31

95 44

22 1

14 8

36 9

Total

132

68

200

36

23

63

Overall, we can conclude that, as expected, global regions, wherever they are located in, attract on average more FDI than other kinds of regions. Their attractiveness further reinforces in case of extra-Europe FDI and in the service sectors. Therefore, globalization processes matter for FDI location patterns. This issue is further explored in the next sections.

5.2. Spatial patterns of FDI across regions: does globalization matter? At geographical level, spatial patterns of FDI highlight a strong concentration37, though they tend to change over time, as it clearly emerges by comparing the distribution of FDI across regions at the beginning and the end of the considered period (see maps 5.1 and 5.2, respectively).38 In particular, in the late 1990s, most of new foreign affiliates were located in the EU’s core, i.e. the area ranging from the UK to the North of Italy, including regions along the French-German border, Belgium, the Netherland, Denmark and Austria. The data also show that other areas, such as the Spanish regions of Madrid, Cataluña, Basque country as well as Central Greece were able to attract a consistent flows of FDI, especially when compared to the general attractiveness of their respective countries. As far as Central and Eastern European regions are concerned, most of FDI was directed to Czech and Hungarian regions, that is very close to the former EU15 border (see Map 5.1). Map 5.1 The location pattern of newly established foreign affiliates in EU in 1997-99.

37

Previous studies on the location of MNEs in Europe highlight geographical concentration. See EC, (2005),“The Location of Multinational across EU Regions”, in The EU economy 2005 review, pp. 237-249. 38 Since large regions tend to attract more FDI than small ones, the number of new foreign firms has been normalized by population. Note also that the identification of global regions relies on 2000 regional characteristics; therefore, this classification can not be applied to previous periods. 102

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The location pattern of newly established foreign affiliates in EU in 1997-1999

500 km

4.8

Total FDI penetration index 1997-1999

© EuroGeographics Association for the administrative boundaries

0 - 14 14 - 44 44 - 103 103 - 189 189 - 385 No data Global regions

Source: Own elaboration from FDIRegio database

These patterns have changed over time. In the mid-2000s new recipient regions emerged, especially in the East periphery, while others, belonging mainly to the South periphery, have become marginal in the location patterns of FDI in Europe.39 Overall, the concentration of FDI seems to be stronger in capital regions than in other regions (Map 5.2).

Map 5.2 The location pattern of newly established foreign affiliates in EU in 2005-07.

39

We refer here to Italy, Greece and Portugal, whose regions have attracted a very low number of foreign firms, as compared to other western European countries. 103

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The location pattern of newly established foreign affiliates in EU in 2005-2007

500 km

4.8

Total FDI penetration index 2005-2007 0 0 - 39 39 - 89 89 - 177 177 - 301 301 - 526 526 - 877 877 - 1909 1909 - 6810 No data Global regions

© EuroGeographics Association for the administrative boundaries

Source: Own elaboration from FDIRegio database

These patterns do not differ substantially between manufacturing and service sectors, though FDI in manufacturing concentrates mainly in Central and Eastern European regions and FDI in service in Western regions. These patterns emerge clearly from Maps 5.3 and 5.4 reported below. One possible explanation for these dynamics could be that the globalization process has driven the location patterns of MNEs. If it was the case, globalized regions should show a distribution of FDI quite different from that of other regions in terms of concentration of foreign firms operating there, sectors interested by the presence of FDI and origin of foreign firms from outside Europe. Table 5.3 shows the results of this simple exercise.

Map. 5.3 The distribution of FDI across EU regions: manufacturing, 2005-07

104

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The distribution of FDI accross EU regions : manufacturing 2005-2007

500 km

4.8

Manufacturing penetration index 2005-2007 0 - 14 14 - 33 33 - 58 58 - 89 89 - 136 136 - 231 231 - 654 654 - 1146 No data Global regions

© EuroGeographics Association for the administrative boundaries

As expected, the distribution of FDI in globalized regions is statistically different from that of FDI in non-globalized regions, with MNEs strongly concentrated in the former, as indicated by the mean of the statistic distribution, which is always higher in globalized than in non-globalized regions (see Table 5.3).40 This result is driven by both extra- and intra-Europe FDI, though the distribution of the latter differs across regions in the service sectors only. These facts are not surprising since a growing share of extra-Europe FDI as well as the progressive tertiarization of local economies indicate that globalization is in place. Therefore, we can conclude that not only globalization drives the location of FDI, but also that the European integration process and the globalization process are complements rather than substitutes.

40

In particular, we found that the distribution of FDI differs between globalized regions and regional players and other regions, while no difference has been observed in the distribution of FDI between the two types of non-globalized regions identified in the previous chapter. 105

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 5.4 The distribution of FDI across EU regions: services, 2005-07 The distribution of FDI accross EU regions : services 2005-2007

500 km

4.8

Services penetration index 2005-2007 0 - 26 26 - 69 69 - 140 140 - 269 269 - 532 532 - 845 845 - 1256 1256 - 5664 No data Global regions

© EuroGeographics Association for the administrative boundaries

Source: Own elaboration from FDIRegio database.

Quite surprisingly, the distribution of intra-Europe FDI in the manufacturing sectors does not differ across types of regions, meaning that globalised and non globalized regions show a similar capacity to attract this kind of FDI. This was not true in the previous analysed period, i.e. 2001-03. (see Table A5.1). This result indicates that over time European foreign investors in manufacturing perceive differences across regions as less marked than non-European foreign investors and investors in the service sectors. It is worth noticing that FDI is unevenly distributed also within globalized regions. In particular, foreign firms are more concentrated in regions specialized in open service sectors. This pattern is constant over time, for all groups of foreign investors, with the exception of European foreign investors in manufacturing. Overall, these results indicate that globalization exerts a stronger effect on location patterns of FDI in services than in manufacturing.

106

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.3. The distribution of FDI across types of regions: descriptive statistics (2005-07) TYPE OF REGION Total FDI Total Extra EU FDI Total Intra EU FDI Total Manufacturing FDI Extra EU Manufacturing FDI Intra EU Manufacturing FDI Total Services FDI Extra EU Services FDI Intra EU Services FDI

Other 76,22 15,36 60,85 28,6 5,14 23,47 47,61 10,23 37,39

Local 133,11 29,73 103,37 44,21 8,48 35,72 88,9 21,25 67,65

Global 462,03 186,21 275,82 75,39 26,54 48,84 386,65 159,66 226,99

Significance *** *** *** ** ***

GLOBAL REGIONS' SPECIALIZATION Total FDI Total Extra EU FDI Total Intra EU FDI Total Manufacturing FDI Extra EU Manufacturing FDI Intra EU Manufacturing FDI Total Services FDI Extra EU Services FDI Intra EU Services FDI

Other 112,4 24,51 87,89 38,53 7,27 31,26 73,87 17,24 56,63

Tertitary 655,23 286,48 368,76 97,58 37,81 59,78 557,64 248,67 308,98

Manufacturing 185,55 37,15 148,4 43,4 8,52 34,88 142,146 28,62 113,52

Both 192,22 53,95 138,27 44,68 13,48 31,2 147,53 40,47 107,07

Significance *** *** *** ** ***

REGIONAL PLAYERS' SPECIALIZATION Total FDI Total Extra EU FDI Total Intra EU FDI Total Manufacturing FDI Extra EU Manufacturing FDI Intra EU Manufacturing FDI Total Services FDI Extra EU Services FDI Intra EU Services FDI

Other 244,47 89,86 154,6 49,01 14,47 34,53 195,46 75,39 120,07

Tertitary 138,82 38,36 100,46 34,23 9,73 24,5 104,58 28,62 75,96

Manufacturing 119,55 22,07 97,47 46 7,14 38,86 73,55 14,94 58,61

Both 306,43 97,16 209,27 73,62 21,84 51,78 232,81 75,32 157,49

Significance

*** *** ***

*** *** ***

Reported coefficients are means. The last column indicates whether and to what extent the FDI distribution differs across types of regions. In particular, ***, **, * indicate a significance level of 1%, 5% and 10%, respectively. Results have been obtained through an analysis of variance (Anova). Source: Own calculation from FDIRegio database

In non-globalized regions FDI seems to be more evenly distributed, though in 2001-03 there was a weak preference for regions with a mixed specialization (both manufacturing and service open sectors). This difference, however, disappeared over time.41 41

A more disaggregated analysis shows that while FDI in all open service sectors is unevenly distributed both within and across types of regions, differences in the distribution of FDI in the manufacturing sector are mainly driven by the manufacture of refined petroleum products, chemicals and chemical products (DF and DG branches), and to a lesser extent, transport equipments (DM). See Tables A5.2 in the Annex. Over time, sectoral differences become more 107

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

relevant, especially in Western EU regions, while they weaken in Central and Eastern European regions (See Tables A5.3 and A5.4 and tables A5.6 and A5.7 in the Annex). 108

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

5.3 Spatial patterns of globalization trends: does FDI matter? As it has been discussed in the previous chapters, globalization is a complex phenomenon, being the result of several economic, social and technological factors. For that reason, its measurement and the identification of its main impacts are not easy tasks. In this section we try to understand whether and to what extent FDI, one of the most direct and measurable expressions of globalization, has driven spatial patterns of globalization in Europe. Following what has been discussed in chapter 4, we thus expect that the linkages between preconditions necessary for regions to take advantage from globalization and the economic performance of regions are stronger the larger the share of FDI attracted by each region. These relationships and their statistical significances are summarized in Table 5.4.42 As expected, most of the globalization trends seem to be driven by FDI, as indicated by the coefficients of correlation, most of which are significant only in regions with a concentration of FDI above the corresponding national means. Moreover, when the coefficients of correlation are significant with the expected sign in both types of regions, they are always larger in regions with a concentration of FDI above the national means, indicating that the larger the share of FDI, the more intense is the corresponding relationship. The phenomena mainly driven or reinforced by the presence of FDI are the following ones: - the (positive) relationship between the presence of command and control functions and the performance of the region, measured by the increase in the private sector productivity; - the increase in service activities in regions less specialized in those activities. This phenomenon regards both medium-low functions and high skilled service functions. - the capacity of business professionals to enhance manufacturing productivity; - the increasing de-industrialization, as indicated by the negative dynamic of the share of plant and machine operators and assemblers, whose growth is faster the smaller the share of this kind of workers on total employment. Our results thus suggest that regions well endowed with command and control functions have also attracted a large share of FDI, thus further increasing the performance of the private service sector (see Map 5.5). Also, private sector productivity grows faster in regions where the share of employment in medium-low functions is lower, and a strong presence of FDI further reinforce this relationship. This implies that foreign firms contribute to increase the performance of the service sectors by choosing to locate in regions relatively more specialized in advanced and control functions rather than in medium-low functions. Moreover, FDI further reinforce the deindustrialization process of European regions.

42

Since global regions represent about 60% of regions with FDI above the mean, we can conclude that the results shown in the last column of Tables 5.4 and 5.5 are mainly driven by global regions, while results shown in the column before the last of the same tables refer, on average, to what happened in regional players and other regions, which represent the majority of regions with a concentration of FDI below the mean. 109

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 5.5 Command and control functions, private service productivity and FDI

500 km

Politecnico di Milano

4.1 © EuroGeographics Association for the administrative boundaries

Growth of service productivity 2002 - 2005

Quota of corporate managers 1999 - 2001

-7.05 0.47

0.48 2.05

2.05 8.90

0.000 0.029 0.029 0.042 0.042 0.171

No data or countries with only one region FDI above the national mean

It is interesting to observe that most of the correlation depicted above are driven by FDI in services. We refer here to the positive relationship between corporate managers and productivity growth in private services, the dynamics of the employment in business professionals and plant and machine operators and assemblers, as well as the positive relationship between business professional and the productivity growth in the manufacturing sector. On the opposite, FDI in manufacturing, seems to be the driver for the negative dynamic of employment conditions for Physical, mathematical and engineering science professionals and clerks, as well as the decline of private service productivity in regions with a strong specialization in medium level functions (see Table A5.8 in the Annex).

110

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.4 Main trends of globalization: the role of FDI Map

Variables

FDI below the mean

FDI above the mean

4.1

Quota of corporate managers 1999-01 vs. growth of service productivity 2002-05

0,22 ***

0,38 ***

4.2a

Quota of SMEs managers 1999-01 vs. growth of service employment 2002-05

0,22 ***

0,26 *

4.2b

Quota of SMEs managers 1999-01 vs. growth of service productivity 2002-05

-0,21 ***

4.3

Quota of clerks 1999-01 vs. growth of service productivity 200205

4.4a

Quota of clerks 1996-98 vs. growth quota of clerks 1996-2007

-0,54 ***

-0,62 ***

4.4b

Quota of Physical, mathematical and engineering science professionals 1996-98 vs. growth quota of Physical, mathematical and engineering science professionals 1996-2007

-0,35 ***

-0,27 *

4.4c

Quota of business services professionals 1996-98 vs. growth quota of business services professionals 1996-2007

4.4d

Quota of plant and machine operators and assemblers 1996-98 vs. growth quota of plant and machine operators and assemblers 1996-2007

4.5

0,08

0,03

-0,35 *** 0,16

0,22

-0,11

-0,44 ***

Quota of business services professionals 1999-01 vs. growth of manufacturing productivity 2002-05

0,03

0,35 **

4.6

Share of human resources in science and technology on population in 2000 vs. growth of manufacturing productivity 2002-05

0,01

0,23

4.7

Share of human resources in science and technology on population in 2000 vs. growth of manufacturing value added 2002-05

-0,19 **

-0,011

Source: Own calculation from FDIRegio database

Needless to say, we would also expect to find that the previously mentioned globalization trends are driven mainly if not exclusively by extra-Europe FDI. This is, on average, confirmed by our data, with the exception of the increased specialization in medium-low level functions, and the consequent negative impact on private service productivity. On the opposite, extra-Europe FDI plays a dominant role in enhancing the positive impact exerted on productivity growth, both in manufacturing and service sectors, through an increase in regions’ specialization in medium-high functions (see Table 5.5).

111

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.5 Main trends of globalization: extra- vs. intra-Europe FDI Extra-Europe FDI Map

Variables

Intra-Europe FDI

FDI below the mean

FDI above the mean

FDI below the mean

FDI above the mean 0.29 **

4.1

Quota of corporate managers 1999-01 vs. growth of service productivity 2002-05

0.21 ***

0.39 ***

0.24 ***

4.2a

Quota of SMEs managers 1999-01 vs. growth of service employment 2002-05

0.24 ***

0.17

0.24 ***

4.2b

Quota of SMEs managers 1999-01 vs. growth of service productivity 2002-05

-0.27 ***

-0.07

-0.20 ***

4.3

Quota of clerks 1999-01 vs. growth of service productivity 2002-05

0.05

4.4a

Quota of clerks 1996-98 vs. growth quota of clerks 1996-2007

-0.6 ***

Quota of Physical, mathematical and engineering science professionals 1996-98 vs. 4.4b growth quota of Physical, mathematical and engineering science professionals 1996-2007

-0.35 ***

4.4c

Quota of business services professionals 199698 vs. growth quota of business services professionals 1996-2007

Quota of plant and machine operators and assemblers 1996-98 vs. growth quota of plant 4.4d and machine operators and assemblers 19962007 4.5

Quota of business services professionals 199901 vs. growth of manufacturing productivity 2002-05

4.6

Share of human resources in science and technology on population in 2000 vs. growth of manufacturing productivity 2002-05

4.7

Share of human resources in science and technology on population in 2000 vs. growth of manufacturing value added 2002-05

0

-0.16 **

0.31 ***

0.21

0.13 *

-0.49 ***

-0.49 ***

-0.29 *

-0.33 ***

0.21

-0.42 ***

0.04

-0.13 *

0.2

-0.39 ***

-0.04

-0.7 ***

-0.31 **

0.23

-0.25 *

0.43 ***

0.32 ***

0.41 ***

-0.01

0.27 *

0.02

0.13

-0.22 ***

0.09

-0.20 **

0.01

Source: Own calculation from FDIRegio database Before concluding this section it is also worth spending a few words on globalization trends not driven by FDI. We refer here to the following trends previously identified: - the relationships between low functions – proxied by clerks – and private service productivity. - the cumulative self-reinforcing process concerning business professionals, whose share on total employment tend to increase more in regions already well endowed with this function. 112

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

-

The positive impact exerted by human resources in science and technology on productivity and value added in the manufacturing sector.

While the first fact is not surprising, since FDI is supposed to bring into regions high level functions, the last two do. However, it is interesting to notice that they are driven by FDI in service and extra-Europe FDI, respectively. Therefore, we can conclude that different types of FDI are conducive for different forms of globalization. In particular, technology and knowledge transfer seems to rely more on extra-Europe rather than intra-Europe FDI.

5.4. Determinants of foreign firm patterns of location Since FDI is not neutral with respect to globalization and its different dimensions, as well as their potential impact on regions’ performance, it becomes relevant, at least from a policy perspective, to understand how foreign firms take their location choices.43 This would allow policy makers to implement more effective FDI promotion policies and, in so doing, attract those foreign investments which assure the optimal response to globalization challenges. In order to achieve this research objective, we explore empirically which regional characteristics are more likely to exert a positive impact on FDI flows. In so doing, we distinguish between traditional determinants, as identified by the economic theory,44 and non-traditional determinants of FDI, whose importance has increased over time because of the deepening and the widening of the globalization process.45 Of course, we expect that new determinants of FDI matter more in globalized rather than in non-globalized regions. Both kinds of determinants are summarized in Table 5.6. Table 5.6 The determinants of FDI: traditional vs. new location advantages Traditional determinants New determinants Demand side: Specialization in services; Market size, potential and accessibility; Quality of human capital; Supply side: Technological level Input (labour) costs Skilled labour force; Specialization (manufacturing); Agglomeration economies The size of the local demand, both actual and potential, is a crucial factor for foreign investments, especially those made in order to serve the local host market. Moreover, its geographical and economic proximity to the main economic centres will allow foreign firms to export to and import from the core cheaply. Hence, we expect that regions with a large and growing local market as well as a good geographical and economic accessibility to the European economic core are attractive to 43

On the importance of location for multinational enterprises, see Dunning, J. (1998), “Location and the Multinational Enterprise: a Neglected Factor?”, Journal of International Business Studies, 29 (1), pp. 45-66. 44 Critical surveys on the main theoretical and empirical literature on traditional FDI determinants can be found, among many others, in Barba Navaretti G. and Venables A., Multinational Firms in the World Economy, 2004, Princeton University Press; Resmini L., “The determinants of FDI in Central, Eastern and South eastern Europe: lessons from the past and prospects for the future”, in Liebscher, K., Christl, J., Mooslechner, P., Ritzberger-Grünwald, D. (eds.), Foreign direct investment in Europe: a changing landscape, 2007, Edward Elgar, Cheltenham, UK and Northampton, MA, USA.; Bloningen B. (2005), “A review of the empirical literature on FDI determinants”, NBER working paper series n. 11299. 45 OECD, Globalisation and Regional Economies, 2007, provides a thorough discussion on how globalization has changed MNEs’ needs in terms of location advantages. 113

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

foreign firms and therefore, are likely to receive more FDI than other regions. We use GDP and GDP growth rates as controllers for local demand size and growth potential, while market accessibility of each region has been computed as the sum of GDP of all other European regions weighted by the distance between the latter and the host region.46 Efficiency seeking FDI, instead, is more sensitive to labour costs, which are generally considered to be a negative host-specific location determinant, particularly for FDI in labour intensive industries. However, in case of intensity in highly skilled and very productive labour activities – such as financial services – where wages are relatively higher, labour costs may not matter. Therefore, no prior expectations can be made about the sign of this variable since it strongly depends on the technology adopted by foreign firms and by the level of productivity of local labour force. Skilled labour force is traditionally proxied by the share of human resources devoted to science and technology activities, and R&D expenses in total GDP. However, as it will be clearer later in this chapter, there are other new and more significant proxies for human capital competencies, which can better capture foreign firms’ needs. The role of agglomeration economies in explaining firms’ location choice is well acknowledged by regional and international economics. Various theoretical concepts suggest that clustering economic activities results in cost-savings and productivity gains through several channels, ranging from client-supplier relationships, labour pooling, and competition effects.47 Several variables can be considered as a possible measure of agglomeration economies. Urbanization economies, which arise when firms benefit from the overall scale and diversity of cities, can be proxy by population density. We expect that urbanization economies matter more for local-market oriented FDI, such as those specialized in non tradable services (retail trade and other personal services), than for other types of FDI. The importance of localization externalities, arising when firms benefit from the presence of clients, suppliers and competitors in the same industry, can be captured through industry-specific agglomeration measures, while the Herfindahl index is a proxy for diversity externalities and inter-industry knowledge spillovers. Since an existing concentration of foreign firms facilitates the gathering of information via business relationships or because it demonstrates the economic potential of a region, foreign agglomeration may also be an important factor of attractiveness at regional level. With the ongoing globalization process, other forms of agglomeration economies have become relevant. We refer here to service agglomeration economies and endowments of highly skilled labour force, which may help multinational firms to reduce production costs and further strengthen their firm-specific advantages. Easy access to, as well as competition among various local services – such as professional services and command and control functions – and a wide range of cultural diversities – such as the presence of university and scientist professionals – may help foreign firms in overcoming several problems related to opaque bureaucracies, unreliable communication infrastructures, immature financial institutions, and cultural issues. Therefore, while low-cost labour force and market accessibility remain important location advantages, the increasingly sought-after advantages are competitive combinations of wages, skills and productivity. We expect that globalization-specific advantages are more important in global

46

GDPk 2 , where j denotes the regions for which the index is calculated and k refers to all k ≠j DIST jk

In formula: MA j = ∑

other European regions included in the sample; GDP is region k’s Gross Domestic Product and Djk is the squared timedistance between j and k. 47 See Fujita M. and J. Thisse ( 2002), Economics of Agglomeration, Cambridge University Press. 114

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

regions than in non-globalized regions, which keep on relying on traditional factors to attract consistent shares of FDI. Needless to say, the importance of location-specific determinants of FDI, both traditional and globalization-specific ones, depends on at least two factors: the motive for FDI – e.g. market, resource or efficiency-seeking investment – and the type of investment, e.g. manufacturing or services.48 Therefore, in order to understand whether and to what extent they are at work across European regions, we compare pairwise correlation coefficients between each of the above mentioned factors and different kinds of FDI (intra- and extra-Europe FDI, manufacturing and services) in different types of regions (global and regional players and other regions). This strategy allows us to identify location advantages able to exert a positive effect on inflows of FDI in Europe. 5.4.1 FDI’s location-specific determinants: a first selection Table 5.7 reports the correlation coefficients and their statistical significance of traditional and globalisation-specific determinants of FDI as suggested by the literature. Their analysis highlights several interesting facts that deserve further analysis. Generally speaking, both traditional and new factors of attraction seem to be at work in Europe. Taken in isolation, traditional location-specific factors seem to be more important than globalization-specific factors, which only in a few cases are able to exert a positive effect on inflows of FDI in Europe. This fact indicates that globalization has modified location advantages, from single socio-economic characteristics to a competitive combination of traditional and new location advantages. Secondly, it is interesting to notice that the importance of each factor – or groups of factors – changes across types of regions. In particular, global regions attract FDI because of their market size, growth potential and accessibility, as well as their specialization in service activities and, mainly in financial intermediation. Finally, the positive signs of the Herfindahl index and the population density indicate that also agglomeration economies make them attractive for foreign investors. Overall, these results suggest that foreign investments in global regions are more market than efficiency oriented and seem to be more specialized in service activity rather than manufacturing. Foreign firms de-localizing in regional players and/or other regions, instead, seem to be motivated by efficiency reasons. Market characteristics are not correlated with FDI, with the exception of growth prospects, while supply characteristics do. In particular, foreign firms seem to be interested in a well developed and highly productive economic sectors, both in manufacturing and services, such as low tech manufacturing sectors and telecommunication and transportation services. Low labour costs do not seem to play any role in attracting FDI, with the only exception of extraEuropean foreign investors delocalising in the least globalized regions (i.e. “other regions” group). More interestingly, extra and intra-European FDI are attracted by different factors in regional players only, with the former sensitive to market accessibility, urban economics, financial intermediation services, command and control functions and college, university and higher education teaching professionals.

48

See Barba Navaretti and Venables, op. Cit. and Markusen, J. (1995), The Boundaries of Multinational Enterprises and the Theory of International Trade”, Journal of Economic Perspectives, 9 (2), pp. 169-89. 115

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Finally, it is worth noticing that, on average there does not seem to be any systematic correlation between technology support characteristics and FDI, with the exception of teaching professionals and extra-Europe FDI in regional players. This fact is quite surprising, though it can be explained by the less relevance and strong geographical concentration of high tech foreign firms in Europe. As it will be demonstrated in the next section, once the geographical location of regions in old and new EU member states is accounted for, determinants of FDI emerge more clearly. It is important to observe that the correlation between location specific determinants and FDI changes not only across regions but also across sectors. More interestingly, while regional differences weaken with a finer disaggregation of manufacturing FDI, they strengthen with a finer disaggregation of FDI in the service sectors. This implies that factors of attraction of FDI in services are region-specific, while those of FDI in manufacturing are sector-specific (see Tables 5.8 and 5.9). Table 5.7. FDI and regional characteristics: correlation analysis GLOBAL PLAYERS total FDI

extra EU

intra EU

REGIONAL PLAYERS total FDI

OTHERS

extra EU

intra EU

total FDI

extra EU

intra EU

Traditional determinants market accessibility

0.27

0.27

0.26

0.14

0.24

0.10

0.00

-0.05

0.02

market potential (GDP growth)

0.29

0.29

0.29

0.38

0.54

0.32

0.50

0.61

0.42

labour cost

-0.06

-0.06

-0.07

-0.10

0.05

-0.13

-0.18

-0.15

population density

0.48

0.46

0.49

0.09

0.20

0.05

0.05

-0.22 0.01

specialization in manufacturing (lt)

0.19

0.20

0.18

0.35

0.21

0.36

0.43

0.32

0.43

Herfindal, all

0.25 0.12

0.26 0.09

0.24 0.15

0.02

0.09

0.00

-0.13

0.00

-0.17

Productivity in services (growth)

0.60

0.70

0.57

0.69

0.71

0.63

Productivity in private services (growth)

0.17

0.15

0.18

0.33

0.36

0.31

0.32

0.34

0.30

Industry productivity (growth)

0.18

0.16

0.19

0.05

0.06

0.05

0.25

0.14

0.27

specialization in services (j)

0.32

0.30

0.34

0.01

0.15

-0.03

-0.13

-0.21

-0.09

specialization in services (i)

0.16

0.14

0.18

0.17

0.12

0.18

0.27

0.36

0.22

Corporate managers

0.19

0.17

0.22

0.20

0.51

0.11

-0.01

-0.03

0.00

SMEs' managers

-0.28

-0.28

-0.26

-0.29

-0.26

-0.28

-0.32

-0.31

-0.30

-0.03

-0.09

-0.13

-0.07 -0.19 -0.25

0.05

Globalization-specific determinants

Human capital

business services professionals

-0.06

-0.07

-0.05

-0.02

0.03

Scientific professionals

-0.19

-0.19

-0.19

-0.10

0.11

-0.15

-0.20

-0.17

clerks (spec. Index)

-0.32

-0.33

-0.30

-0.21

-0.03

-0.25

-0.29

-0.36

office clerks (spec. Index)

-0.13

-0.15

-0.11

-0.22

-0.08

-0.25

-0.29

-0.35

-0.25

0.14

-0.08

-0.14

-0.20

-0.11

customer services clerks (spec. Index)

-0.31

-0.32

-0.30

-0.03

workers (spec. Index)

0.20

0.22

0.17

0.08

-0.16

0.14

0.27

0.26

0.25

R&D expenditures

-0.15

-0.15

-0.16

-0.02

0.13

-0.05

-0.09

-0.11

-0.07

Human resources in S&T

-0.06

-0.05

-0.07

-0.16

-0.15

-0.16

-0.12

-0.17

-0.10

College and university teachers

0.14

0.15

0.13

0.07

0.36

0.00

-0.11

-0.10

-0.11

Technology support

Coefficients in bold are statistically significant at least at 10%. Source: Own elaboration from FDIRegio database. 116

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

In particular, high and medium high tech FDI is associated with market accessibility and growth prospect, urban externalities, foreign agglomeration and the presence of command and control functions and technology support measures in almost every kind of regions, while FDI in low and medium-low tech sectors seem to be attracted by market growth prospects and low labour costs, linkages with industry-specific and foreign location externalities, transportation and communication services, and skilled workers in almost all regions. As far as FDI in services is concerned, it is worth noticing that they respond positively to demand side characteristics in all types of regions, though their capacity of attraction weakens in other regions. Foreign agglomeration is also important in all regions and for all types of FDI with the exception of FDI in telecommunication and transportation services. Local specialization both in manufacturing and services matters as a determinant of FDI only in regional players and other regions. These facts indicate that regional players and other regions are more attractive for foreign firms providing business services rather than consumer services. The latter, in fact, tend to concentrate more in global regions than in other types of regions. Finally, FDI in private services are, on average, positively linked with advanced and scientific functions in global and regional players and with low level functions in the other kinds of regions.

5.4.2. The location specific determinants of FDI: the relative importance The previous section has been devoted to better understand whether the determinants of foreign investments, as suggested by the theory of FDI and the more recent literature on globalization, were at work in the EU regions. We tested the association with several traditional and non-traditional potential factors of attraction with different measures of FDI penetration, taking into account the sector of activity of multinational enterprises and their origin inside or outside Europe. Quite interestingly, we uncover that these relationships may differ in terms of intensity, sign, and significance across different types of regions and sectors. In carrying out that analysis, however, we considered each factor of attraction in isolation, in order to understand whether it was important for inward FDI in EU regions or not. What we still do not know is the relative importance of those factors, i.e. which characteristics each region has to work with in order to improve its capacity of attraction of FDI, according to either the sector of activity of foreign firms or their geographical origin within or outside Europe. This sub-section addresses these issues.

117

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.8. Factors of attraction of FDI: manufacturing sectors GLOBAL PLAYERS Traditional location determinants market accessibility market potential (GDP growth) labour cost population density local, manufacturing (lt) FDI (ht) FDI (mht) FDI (mlt) FDI (lt) Herfindal, all Productivity in services (growth) Productivity in private services (growth) Industry productivity (growth) Globalization specific determinants local, services (j) local, services (i) FDI, services Human capital Corporate managers SMEs' managers business services professionals professionals (all) clerks office clerks customer services clerks workers (spec. Index) Technology support R&D expenditures Human resources in S&T Teaching professionals

REGIONAL PLAYERS

HT FDI

MHT FDI

MLT FDI

LT FDI

0.31 0.42 0.07 0.35 0.16

0.33 0.31 0.09 0.42 0.20 0.90

0.09 0.30 -0.20 0.23 0.32 0.76 0.85

0.05 0.23 -0.25 0.19 0.32 0.70 0.74 0.96

0.90 0.76 0.70 0.33 0.18

0.85 0.74 0.23 0.09

0.96 0.01 0.15

0.24 0.20

0.16 0.09

0.22 0.02 0.47

OTHER REGIONS

HT FDI

MHT FDI

MLT FDI

LT FDI

HT FDI

MHT FDI

MLT FDI

LT FDI

0.21 0.35 0.02 0.09 0.10

0.23 0.47 0.00 0.15 0.21 0.66

0.00 0.40 -0.28 0.04 0.40 0.58 0.76

-0.09 0.30 -0.30 -0.05 0.46 0.58 0.62 0.90

0.20 0.31 0.08 0.16 0.37

0.35 0.35 0.13 0.29 0.32 0.66

0.19 0.37 -0.10 0.14 0.38 0.75 0.81

-0.08 0.44 -0.29 -0.01 0.44 0.66 0.64 0.86

0.00 0.12

0.66 0.58 0.58 0.32 0.38

0.76 0.62 -0.02 0.54

-0.06 0.63

0.66 0.75 0.66 -0.24 0.52

0.90 -0.13 0.60

0.81 0.64 -0.04 0.65

0.86 -0.10 0.64

-0.06 0.68

0.07 0.16

0.06 0.16

0.30 0.02

0.32 0.14

0.36 0.20

0.29 0.09

0.40 0.17

0.50 0.17

0.35 0.21

0.25 0.26

0.25 -0.01 0.49

0.09 0.18 0.49

0.10 0.26 0.46

0.17 0.46 0.31

0.09 0.03 0.21

-0.14 0.08 0.08

-0.20 0.20 0.06

0.08 0.13 0.51

-0.06 0.16 0.38

-0.16 0.23 0.29

-0.21 0.28 0.05

0.54 -0.25 -0.03 -0.03 -0.13 -0.17 0.12 -0.01

0.43 -0.26 -0.16 -0.12 -0.06 -0.07 0.03 0.09

0.09 -0.32 -0.23 -0.31 -0.35 -0.34 -0.17 0.43

-0.04 -0.30 -0.20 -0.33 -0.44 -0.40 -0.26 0.46

0.41 -0.24 0.14 0.16 -0.02 -0.06 0.11 -0.13

0.45 0.28 0.21 0.16 0.00 -0.03 0.11 -0.06

0.01 -0.33 -0.02 -0.20 -0.35 -0.34 -0.17 0.28

-0.13 0.32 -0.19 -0.30 -0.48 -0.44 -0.28 0.31

0.30 -0.32 -0.03 -0.06 -0.13 -0.17 0.12 0.11

0.28 -0.38 -0.16 0.03 -0.06 -0.07 0.03 0.10

0.09 -0.36 -0.23 -0.12 -0.35 -0.34 -0.17 0.22

-0.16 -0.30 -0.20 -0.25 -0.44 -0.40 -0.26 0.34

-0.08 -0.15 0.35

-0.16 -0.22 0.28

-0.17 -0.17 0.03

-0.16 -0.08 -0.10

0.10 0.03 0.23

0.10 -0.04 0.28

-0.11 -0.26 -0.04

-0.16 -0.28 -0.16

0.08 0.06 0.15

0.15 0.24 0.03

-0.01 -0.02 -0.16

-0.16 -0.19 -0.21

Coefficients in bold are statistically significant at least at 10%. Source: Own elaboration from FDIRegio database. 118

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.9. Factors of attraction of FDI: service sectors g+h

GLOBAL PLAYERS i+k2 j+k1 l m+n+op

g+h

REGIONAL PLAYERS i+k2 j+k1 l m+n+op

g+h

OTHER REGIONS i+k2 j+k1 l m+n+op

Traditional location determinants market accessibility market potential (GDP growth)

0.03

0.32

0.45

0.35

0.45

0.11

0.19

0.29

0.06

0.16 -0.04

0.02

0.29

0.53

0.19

0.23

0.30

0.12

0.32

0.30

0.41

0.22

0.46

-0.25

-0.01

0.16

0.22

0.11 -0.18 -0.01

0.14

0.22

0.06 -0.23 -0.09

population density

0.18

0.57

0.67

0.56

0.67 -0.03

0.15

0.09

0.17 -0.01

local, manufacturing (lt)

0.29

0.17

0.01 -0.17

0.06

0.39

0.25

0.16

0.02

0.21

FDI, manufacturing

0.96

0.80

0.65

0.71

0.95

0.85

0.77

0.08

0.79

labour cost

0.26

0.07

0.49

0.03

0.31

0.49 -0.08 -0.07

0.01 0.49

0.04

-0.01

0.02

0.02 -0.05

0.03

0.44

0.31

0.41 -0.15

0.30

0.97

0.91

0.94 -0.05

0.83

Productivity in services (growth)

0.12

0.12

0.12

0.04

0.09

0.61

0.53

0.48 -0.01

0.54

0.70

0.67

0.67 -0.07

0.62

Productivity in private services (growth)

0.02

0.25

0.25

0.18

0.24

0.30

0.33

0.30

0.08

0.29

0.28

0.34

0.36 -0.04

0.38

0.05

0.25

Industry productivity (growth) Herfindal, all

0.17

0.20

0.20 -0.08

0.17

0.04

0.02

-0.03 -0.06

0.25

0.22

0.09

0.27

-0.02

0.29

0.49

0.30

0.45

0.01

0.02

0.11 -0.03

0.04 -0.12 -0.09

-0.19

0.12

-0.17

0.11 -0.03

0.06

0.03

0.16

0.11

0.16

0.01

0.10

0.27

0.25

0.10

0.23

Globalization specific determinants local, services (j)

0.29

0.24

local, services (i)

0.08

0.48

0.48

0.63

0.43 -0.11

0.08

0.25

0.17

0.24 -0.16 -0.05

-0.04 -0.07

0.00

FDI, services

0.85

0.95

0.88

0.33

0.91

0.96

0.91

0.92

0.23

0.88

0.99

0.92

0.98 -0.04

0.86

managerial skills (spec. Index 12)

-0.10

0.21

0.44

0.24

0.37

0.03

0.25

0.52

0.37

0.50 -0.10

0.14

0.11 -0.04

0.26

managerial skills (spec. Index 13)

-0.30

-0.30 -0.02

-0.27

Human capital

business services professionals (spec. Index) -0.19

-0.17 -0.18

0.11

-0.21 -0.29 -0.17

-0.22 -0.04

-0.20 -0.30 -0.29

0.04

0.09

0.18

0.06 -0.08

0.06

0.10

0.10

0.03 -0.14

0.00

-0.03

0.03

0.07

0.00

0.12

0.03

0.09 -0.25 -0.11

-0.15

0.19

-0.05

Science professionals

-0.32

-0.10

0.00

0.12

-0.05 -0.20

clerks (spec. Index)

-0.45

-0.26 -0.09

0.28

-0.16 -0.32 -0.14

0.11

0.22

0.00 -0.37 -0.17

-0.18

0.00

-0.08

office clerks (spec. Index)

-0.41

-0.26 -0.11

0.35

-0.17 -0.30 -0.16

0.04

0.13

-0.06 -0.36 -0.19

-0.21

0.02

-0.13

customer services clerks (spec. Index)

-0.27

-0.11

0.26

0.36

workers (spec. Index)

0.06 -0.15

0.48

0.04 -0.16 -0.30

-0.12

-0.20 -0.12 -0.16

0.00 -0.17 -0.06

0.03

0.18

0.00

-0.15 -0.08

0.05

-0.18 -0.22

0.20 -0.21 -0.11

0.32

0.00

0.01 -0.09

0.14

0.12

0.19

0.02

0.08

-0.02 -0.05

-0.01

-0.07

0.14

-0.05

-0.01 -0.07

0.20

Technology support R&D expenditures Human resources in S&T

-0.04

0.00 -0.06 -0.09

Teaching professionals

-0.16

0.17

0.43 -0.06

-0.06 -0.17 -0.06 0.38 -0.08

0.04

0.14

0.14

0.11 -0.12 -0.04

-0.04 -0.03

-0.14 -0.16 -0.08

0.34

0.23

0.34 -0.19

0.04

g+h = retail and restaurant industry; i+k2= transports and business services; j+k1=financial intermediation and real estate; l=public sector; m+n+op=services to citizens 119

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009 Coefficients in bold are statistically significant at least at 10%. Source: Own elaboration from FDIRegio database.

120

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

As a first attempt to identify the main determinants of inward FDI in Europe, a simple OLS estimation has been considered.49 The dependent variables used in the various regressions are different specifications of a foreign direct investment penetration index. The latter has been defined as the ratio of the number of new FDI undertaken in a given region during the period from 20052007 normalized by population.50 All factors whose correlation with FDI turned out to be significantly different from zero in the correlation analysis described in the previous section have been considered as potential explanatory variables.51 These variables have been organized in similar groups and tested separately one from the other. In order to obtain a general model explaining patterns of inward FDI in Europe, we start by testing different blocks of explanatory variables, organized according to homogenous regional characteristics – market characteristics, specialization in various manufacturing and/or service sectors, endowments of human capital, technological characteristics. Then, we check the effectiveness of a more general specification including at least one variable for each of the above mentioned groups of regions’ characteristics. We do it first for all regions included in the sample, and then by distinguishing either different kinds of regions (global vs. regional players) or different sources of FDI (extra- vs. intra-Europe FDI). Finally, we consider a sector-based perspective, analyzing separately services and manufacturing FDI. Table 5.10 shows the results of the first econometric exercise. Ccolumns 1 and 2 focus on the traditional determinants of FDI. We also add dummies in order to control for and compare not only the capacity to attract FDI of different groups of regions, i.e. global vs. regional players and other regions, but also of regions belonging to the outstanding receivers of FDI in the 2005-07 period, i.e. United Kingdom and Romania. Note that the two country-specific dummies are positive and significant in all specifications. This indicates that, despite globalization, foreign firms are still sensitive to some country-specific factors and that the latter should be better identified across Europe. We obtain only a weak evidence that global regions attract, all other regional characteristics being equal, more FDI than other kinds of regions. Quite interestingly, once controlling for service specialization, global region dummy is no longer significant, while it is controlling for functions. This implies that what really makes the difference between globalized and not-globalized regions, at least as far as foreign firms’ attractiveness is concerned, is service specialization rather than function specialization.52

49

The hypothesis that location patterns of FDI may be affected by some form of spatial autocorrelation has been rejected by spatial autocorrelation diagnostic (See Table A5.9 in the Annex). 50 In this way, we accounted for the size effect, consisting in the capacity of a region to attract more FDI just because it is bigger, in economic term, than other regions. Country specific effects, instead, have been captured through specific dummy variables. 51 In constructing each block of explanatory variables we control for multicollinearity and heteroskedasticity effects. Since we also control for spatial spillovers, only variables without missing values have been considered. The selected explanatory variables have been lagged one period in order to avoid potential endogeneity problems. 52 The reader should remember that functions and tasks are common to all sectors of activities, thus also regions highly specialized in manufacturing should also possess a good endowment of advanced and low functions. 121

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.10. FDI location patterns: traditional vs. new determinants all regions - Total FDI Variables GDP (millions PPP) GDP growth Labour costs Market accessibility

(1) 0.050 0.338 *** -0.134 ** 0.385 ***

(2) 0.329 -0.134 0.432

Manufacturing LT Manufacturing MLT Manufacturing MHT Manufacturing HT

-

Services K Services J Services I Services H Services G Services ONP

-

-

Plant and machine operators and assemblers Clerks Professionals Corporate managers SME managers

-

-

Dummy RO Dummy UK Global Regional

2.453 0.820 0.697 0.005

0.586 -0.230 -0.346 0.004

*** *** ***

2.198 0.724 0.758 0.102

*** ** *** *

Specifications (3) 0.291 *** -0.072 0.311 *** -0.418 0.784 1.112 -0.248 2.489 0.331

*** *** ***

*** *** ***

(4) 0.245 -0.051 0.183

(5) *** ***

0.165 -0.014 0.108

-

0.040 0.339 0.010 -0.072

-

1.033 0.451 0.343 2.479 -

3.290 0.517 1.374 26.565 -5.641

*** *

7.393 0.798 3.197 24.025 -8.658

***

***

***

**

*** **

2.463 0.707 0.155 -0.049

*** ***

3.337 -0.590 0.438 -0.013

*** ** ***

3.521 -0.496 0.009 -0.061

*** **

FDI (t-1) 0.350 *** 0.358 *** 0.414 Adjusted R2 0.70 0.71 0.77 OBS. 252 252 252 OLS estimation also includes a constant term. ***, **, * indicate significant levels at 1, 5 and 10 percent respectively.

***

0.303 0.78 252

***

0.378 0.83 252

***

Generally speaking, specifications (1) and (2) indicate that European regions attract more market rather than efficiency seeking FDI. This is indicated by the relative intensity of the estimated coefficients, which is higher for demand variables than for supply variables. As stated at the beginning of this study, the globalization process we are experiencing is characterized by some important innovative trends: a clear shift from manufacturing to services offshoring and outsourcing and a globalization of functions and tasks. Specifications (3) and (4) focus on the role played by these facts as potential “new” determinants of foreign firms’ location patterns. We found that inflows of FDI are larger the more a region is specialized in services, and, particularly, in financial intermediation, transports and communication services and retail sales. Also functions, especially the advanced and control ones, are important determinants in the location 122

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

choice of foreign firms, as the positive and significant coefficient of the corporate manager variable indicates. It is worth noticing that when traditional and new determinants of FDI are considered together, as we did in specification (5), traditional determinants lose their significance and new globalizationspecific determinants seem to be the only factors able to explain location patterns of FDI.53 In particular MNEs seem to look for different kinds of human capital, ranging from plant and machine operators and assemblers to corporate managers.54 A notable exception is given by SMEs managers, whose regional concentration seems to exert a negative effect on foreign firms. This result indicates that foreign firms, on average, do not consider a location advantage the presence of a well developed local industrial fabric, made of small and medium enterprises with which to establish input-output linkages. This result is not a surprise since what matters in a globalized world is the participation to international rather than local networks of production. The positive and significant sign of the coefficient of the already established foreign firm variable support this consideration. These results hold on average and do not consider potential differences across regions and or foreign firms. In order to check whether and to what extent these possible sources of heterogeneity matter, we re-estimate specification (5) by considering potential differences between regions belonging to Western and Eastern Europe on the one hand, and between intra- and extra-Europe FDI, on the other hand. The results of this exercise are reported in Table 5.11. Few differences emerge and are worth to discuss. Extra- and Intra-Europe FDI seem to be attracted by the same factors. However, intra-Europe FDI is also sensitive to potential market demand, while extra-Europe FDI positively reacts to the presence of professionals, and, in particular, scientific professionals, too. Finally, it is worth noticing that, ceteris paribus, UK regions attract more extraEurope FDI than other regions, while their capacity of attraction of intra-Europe FDI is lower than that of other European regions. Romanian regions, instead, represent a good location for both kinds of foreign firms.

53

Potential demand – proxied by GDP growth rate – seems to be the only notable exception. These aggregate results hide differences within functions. In particular, foreign firms are not attracted by regions specialized in workers performing elementary occupations. Moreover, they prefer regions with high endowments of customer services clerks rather than offices clerks. Among professionals, MNEs look for business and teaching professionals. These disaggregated results are available upon request. 123

54

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.11. FDI location patterns: the role of geography Variables GDP growth labour costs market accessibility

Extra-Europe FDI (5) 0.086 0.028 0.075

Intra- Europe FDI (5) 0.172 *** -0.020 0.086

Manufacturing LT Manufacturing MLT Manufacturing MHT Manufacturing HT

-0.578 -0.076 0.192 0.242

**

0.127 0.386 -0.029 -0.152

Services J Services I Services H

0.876 0.006 0.141

***

Services G

1.139

plant and machine operators and assemblers clerks professionals corporate managers SME managers

6.967 0.267 3.885 22.852 -7.599

Dummy RO Dummy UK global regional

4.547 0.585 0.176 -0.156

*** ***

FDI Adjusted R2 OBS.

0.278 0.88 252

***

Western regions (5) 0.104 * -0.056 0.175 **

0.340 0.045 -0.041 -0.091

1.011 0.398 0.387

*

***

2.750

***

-0.451

***

7.747 1.855 3.277 23.260 -8.366

**

11.487 5.798 1.702 28.102 -5.830

** *** *

***

*** **

3.363 -0.755 -0.010 -0.037

*** ***

0.376 0.80 252

***

0.652 1.037 0.661

** *** ***

*** ** ***

-0.622 0.010 -0.120

**

0.386 0.87 252

***

NMS regions (5) 0.104 * -0.056 -0.408 ***

0.340 0.045 -0.041 0.698

*

0.652 -0.380 -1.098

** * ***

5.068

***

12.320 5.240 3.400 5.007 15.938

*** *

5.323

***

** **

0.010 -0.120 0.386 0.87 252

***

Note: coefficients reported in the last two columns have been estimated simultaneously, through interaction effects with dummy variables. The reported coefficients are not slope differentials but the final effects. *** = significant at 1% level ** = significant at 5% level * = significant at 10% level

By distinguishing between regions belonging to Western and Central and Eastern Europe we uncover other interesting facts. Demand factors and specialization in services are important location advantages for Western regions, while they exert a weaker effect on the attractiveness of regions in NMS. The latter, instead, is strongly affected by the share of SMEs manager on total employment. Overall, these results indicate that foreign firms de-localizing in NMS’ regions are interested in exploiting emerging local markets. Establishing client/supplier relationships with local firms may in fact help in providing products and services particularly suitable to the needs of local consumers. This consideration is further reinforced by the negative and significant sign of the market accessibility variable. Foreign firms operating in Western regions, instead, seem to have a larger 124

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

market horizon, as indicated by the fact that market accessibility has a positive effect on inflows of FDI and SME manager coefficient is negative, while the coefficient of corporate managers is much higher for former EU15 regions than for NMS regions. A sector-based perspective, though potentially useful, does not allow us to further enrich previous results (see tables 5.12 and 5.13). The most striking differences with respect to the general trends described above are the following. First of all, we found that while FDI in manufacturing is attracted only by other already existing manufacturing foreign firms, FDI in service is sensitive to foreign agglomeration, regardless of the sector of activities of the incumbent foreign firms. Secondly, the average effect found in previous regressions that regions well endowed with corporate managers attract more FDI is driven by FDI in services and by FDI in manufacturing but only in EU15 regions. In NMSs’ regions, instead, foreign investments in manufacturing are more numerous the larger the share of SMEs manager on regional total employment. This implies that local connections matter more than international linkages in manufacturing and in NMS. This result reinforce the idea that Central and Eastern European regions are more attractive for FDI in manufacturing than in services and that most of these investments have been undertaken to exploit local emerging markets. Table 5.12 The determinants of FDI: manufacturing All regions Variables GDP growth labour costs market accessibility

(5) 0.075 -0.042 0.062

Manufacturing LT Manufacturing MLT Manufacturing MHT Manufacturing HT

0.067 0.553 -0.057 -0.126

Services J Services I Services H Services G

0.506 0.220 0.027 2.086

All regions (6) 0.080 0.017 0.025

**

*

***

-0.136 0.399 0.066 0.014 0.823 0.549 0.023 2.199

plant and machine operators and assemblers clerks professionals corporate managers SME managers

8.046 2.172 1.653 20.404 -8.341

Dummy RO Dummy UK global regional

3.237 -0.208 -0.084 -0.001

***

3.143 -0.471 -0.001 0.055

0.227

***

-

FDI

***

*** **

4.634 2.715 0.164 19.854 -10.116

Western regions (5) 0.014 -0.226 0.135

*

*** ** ***

**

*** *** *** **

**

0.362 0.220 -0.089 -0.002 0.069 0.958 0.473 -1.563

13.448 9.299 1.939 24.365 -5.160

NMS regions (5) 0.014 -0.226 -0.446

***

0.362 0.220 -0.089 -0.002

*** *** ***

*** *** ***

0.120 -0.920 -1.799 4.712

13.448 9.299 1.939 -2.151 19.636

*** *** ***

*** *** ** **

4.396 -0.295 -0.059 -0.052

***

3.800 -0.295 -0.059 -0.052

***

0.219

***

0.387

** 125

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

FDI manufacturing FDI services Adjusted R2 OBS.

0.73 252

0.601 -0.237 0.77 252

*** ***

0.82 252

0.82 252

Note: coefficients reported in the last two columns have been estimated simultaneously, through interaction effects with dummy variables. The reported coefficients are not slope differentials but the final effects. *** = significant at 1% level; ** = significant at 5% level; * = significant at 10% level

Table 5.13 The determinants of FDI: services All regions Variables GDP growth labour costs market accessibility

(5) 0.189 -0.012 0.064

Manufacturing LT Manufacturing MLT Manufacturing MHT Manufacturing HT

-0.038 0.211 0.098 -0.113

Services J Services I Services H Services G

***

All regions (6) 0.193 0.006 0.055

***

-0.122 0.125 0.142 -0.055

1.188 0.564 0.412 2.522

*** ** ** ***

1.349 0.653 0.429 2.660

plant and machine operators and assemblers clerks professionals corporate managers SME managers

5.969 2.005 3.122 25.094 -7.685

**

4.717 1.808 2.351 24.735 -8.531

Dummy RO Dummy UK global regional

3.720 -0.552 0.046 -0.077

*** **

3.582 -0.693 0.079 -0.045

FDI FDI manufacturing FDI services Adjusted R2 OBS.

0.460 0.84 252

***

0.402 0.133 0.84 252

*** **

Western regions (5) 0.139 0.109 0.100

***

0.243 0.043 0.063 -0.325 *** ** ** ***

*

** *** *** ***

*** *

NMS regions (5) 0.139 0.109 -0.409

*** **

0.243 0.043 0.063 -0.325

0.877 0.867 0.583 0.508

*** ** ***

0.877 0.867 0.583 4.949

*** ** *** ***

7.825 4.581 0.378 29.253 -6.642

**

7.825 4.581 0.378 29.253 -6.642

**

*** ***

*** ***

6.047 -0.723 0.034 -0.135

*** ***

6.047 -0.723 0.034 -0.135

*** ***

0.474 0.86 252

***

0.474 0.86 252

***

Note: coefficients reported in the last two columns have been estimated simultaneously, through interaction effects with dummy variables. The reported coefficients are not slope differentials but the final effects. *** = significant at 1% level ** = significant at 5% level * = significant at 10% level

126

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

As a further inspection of FDI location mechanisms, we now consider a more detailed sector-based perspective, applying the same model as before to a more disaggregated sector classification, in order to control for possible specificities in the location choice of multinational enterprises investing in Europe. Table 5.14 below shows the result of a more disaggregated analysis for FDI in manufacturing, while Table 5.15 focuses on several service sectors.

127

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.14. The determinants of manufacturing FDI: breakdown by sectors Variables

HT

MHT

GDP growth labour costs market accessibility plant and machine operators and assemblers clerks professionals corporate managers SME managers Dummy RO Dummy UK eu12 global

0.029 0.078 0.044 -3.083 1.729 -4.079 21.263 -1.433 1.265 *** -0.710 *** 0.618 *** 0.200

0.043 -0.008 0.153 -1.442 4.433 -1.805 14.523 -0.719 2.242 -0.407 0.431 0.440

regional FDI manufacturing HT FDI manufacturing MHT FDI manufacturing MLT FDI manufacturing LT FDI water and energy FDI trade, hotels and restaurants FDI financial intermediation and real estate FDI transports and business services FDI services to citizens Adjusted R2 OBS.

0.059 0.308 *** 0.043 0.069 0.104 0.042 -0.116 * -0.147 * 0.109 -0.037 0.63 252

0.133 -0.073 0.532 *** 0.238 *** -0.156 ** 0.089 -0.091 -0.072 -0.018 -0.077 0.67 252

*** * *** *** *** ** ***

MLT 0.071 0.006 0.138 -0.362 2.724 -1.139 15.564 -1.835 2.256 -0.370 0.626 0.021

LT

**

*** *** ***

-0.028 0.023 0.064 0.569 *** -0.028 -0.090 -0.121 * -0.130 0.030 -0.075 0.62 252

0.146 0.028 0.097 -1.933 3.630 -3.068 23.446 -5.706 3.270 -0.787 0.595 0.174

** *

*** * *** *** ***

-0.010 -0.043 0.004 0.125 0.340 *** -0.031 -0.121 -0.041 0.022 -0.078 0.70 252

*** = significant at 1% level ** = significant at 5% level * = significant at 10% level

First of all it is important to notice that globalization matters mainly for services FDI, as indicated by the coefficients of the global region dummy, which are positive and significant in almost all tradable services but not in manufacturing, with the exception of FDI in medium-low tech sectors. Secondly, it is interesting to observe that agglomeration forces work mainly within and not across manufacturing sectors. In the case of services, instead, foreign firms react not only to within sector externalities, but also to benefits arising by co-locating with manufacturing firms. Thirdly, SME manager concentration becomes a factor of attraction for FDI in services, while it remains unattractive for FDI in manufacturing. This implies that most of FDI in services are motivated by the needs to exploit the local market. Finally, note that Central and Eastern European regions collect, on average, more FDI than Western European regions in all economic sectors.

128

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 5.15 The determinants of service FDI: breakdown by sectors Variables

water and energy

trade, hotels and restaurants

financial intermediation and real estate

0.285 *** 0.004 0.058

0.050 -0.116 0.145 **

transport and business services

services to citizens

0.101 0.011 0.030

0.138 -0.007 0.061

GDP growth labour costs market accessibility

-0.076 -0.107 * 0.053

plant and machine operators and assemblers clerks professionals corporate managers SME managers

-0.036 -2.297 -0.162 -1.953 8.266 ***

-3.957 4.445 -2.497 30.409 *** 3.691

-0.393 5.418 * 1.082 18.883 *** 13.373 ***

-2.746 5.181 -0.122 *** 21.900 *** 1.465

Dummy RO Dummy UK eu12 global regional

-0.097 0.298 0.289 * 0.126 -0.087

3.661 -1.324 0.871 0.344 -0.100

*** *** *** **

4.712 *** 0.222 0.698 *** 0.482 *** -0.089

3.460 *** 0.142 0.586 *** 0.455 *** 0.032

2.939 0.677 0.411 0.338 -0.006

FDI manufacturing HT FDI manufacturing MHT FDI manufacturing MLT FDI manufacturingLT FDI water and energy FDI trade, hotels and restaurants FDI financial intermediation and real estate FDI transports and business services FDI services to citizens Adjusted R2 OBS.

0.122 0.086 -0.022 -0.052 0.173 *** -0.013 -0.093 0.086 0.044 0.2 252

-0.011 0.398 *** 0.059 0.068 -0.137 * 0.023 -0.034 0.148 *** -0.129 0.76 252

-0.017 0.053 0.076 -0.185 ** -0.157 ** 0.073 0.326 *** 0.087 -0.099 0.75 252

-0.047 0.215 ** 0.082 -0.073 -0.018 0.008 0.032 0.347 *** -0.018 0.82 252

0.044 -0.002 0.099 -0.052 0.016 -0.057 0.100 0.090 0.107 0.79 252

-3.664 * 1.691 -3.006 13.693 *** 7.010 ** *** *** ** **

*** = significant at 1% level

129

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009 ** = significant at 5% level * = significant at 10% level

130

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

5.5. Concluding remarks This chapter focused on FDI patterns of location across EU regions. In particular, it explored the potential interrelationships between FDI and globalization, by analysing, first, potential differences in the distribution of FDI between globalized and not-globalized regions, and, secondly, the role played by FDI in shaping and driving globalization trends. Finally, this chapter discussed the main determinants of inflows of FDI in Europe during the period from 2005 to 2007. Several interesting lessons have been learnt. In particular: 

Globalization plays a role in shaping spatial patterns of FDI. Foreign firms concentrate the most in globalized regions. This trend is driven mainly by extra-Europe FDI and by foreign investors operating in service sectors, which tend to concentrate in a very small number of regions and countries. These facts suggest that the European integration process and the globalization process are complements rather than substitutes.



Globalization has affected mainly foreign firms operating in service sectors, which prefer to locate in global regions already specialized in the tertiary sector. This fact holds for both intra and extra-Europe FDI. Therefore, there is a complementary relationship between European integration process and globalization trends.



Intra-Europe FDI in the manufacturing sectors is more evenly distributed and this tendency reinforces over time. In 2005-07, the distribution of FDI in manufacturing did not differ either across or within types of regions. This result indicates that European foreign investors perceive differences across regions as less marked than European foreign investors operating in services and extra-European foreign investors.



Globalization trends have been driven or reinforced by FDI. In particular, FDI has enhanced the ongoing de-industrialization process and strengthened the positive impact effect exerted by command and control functions or high level functions on regional productivity growth rates. Both intra- and extra-Europe FDI are involved in these trends. However, we demonstrated that extra-Europe FDI are more conducive for those processes involving technology and knowledge transfer.



Globalization has changed the relative importance of traditional and not-traditional determinants of FDI, with the former loosing importance in favour of the latter. Regression analysis demonstrate that regions specialized in manufacturing sectors are less attractive than regions specialized in service sectors. FDI determinants differ across regions and sectors. However, all kinds of regions rest on a set of factors of attraction which include either traditional or globalization-specific location characteristics. Among the former, the most significant is the market demand potential, measured by the GDP growth rate. The latter include specialization in services – mainly financial intermediation and transportation and communication services – and in medium-high and command and control functions.





Despite these general trends, global regions are more attractive for market seeking FDI, while regional players are the preferred location for efficiency seeking FDI. Moreover, foreign firms investing in Central and Eastern European regions seem more interested in exploiting emerging local markets, and, consequently, to joint local networks of production, while foreign firms investing in Western regions have a larger market horizon, giving their strong preference for international networks of production. 131

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009



On a sectoral base, it is worth noticing that manufacturing FDI exploits within sectors foreign agglomeration externalities, while service FDI also reacts positively to across sector agglomeration forces.



On a regional base, it is interesting to observe that global regions attract, all else equal, more FDI in tradable services than other European regions. Central and Eastern European regions attract, on average, more foreign firms than all other European regions both in manufacturing and in service sectors, regardless their position in the globalization process. Finally, UK regions, all else equal, attract more extra-Europe foreign firms than other European regions. Finally,

132

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Annex 5.1

Table A5.1 The distribution of FDI across types of regions: descriptive statistics (2001-03) TYPE OF REGION

Other

Local

Global

Significance

Total FDI Total Extra EU FDI Total Intra EU FDI Total Manufacturing FDI Extra EU Manufacturing FDI Intra EU Manufacturing FDI Total Service FDI Extra EU Service FDI Intra EU Service FDI GLOBAL REGIONS' SPECIALIZATION

103.87 18.65 85.22 43.29 7.28 36.01 55.88 10.45 45.43 Other

137.66 33.49 104.17 45.46 10.85 34.62 79.47 18.02 61.45 Tertitary

504.67 208.12 296.55 91.81 32.47 59.34 360.90 148.94 211.96 Manufacturing

Both

*** *** *** *** *** *** *** *** *** Significance

Total FDI Total Extra EU FDI Total Intra EU FDI Total Manufacturing FDI Extra EU Manufacturing FDI Intra EU Manufacturing FDI Total Service FDI Extra EU Service FDI Intra EU Service FDI LOCAL REGIONS' SPECIALIZATION

125.42 28.11 97.31 44.68 9.55 35.12 70.92 15.28 55.65 Other

671.38 309.24 362.14 115.29 46.09 69.20 491.76 225.78 265.98 Tertitary

161.00 18.30 142.70 44.68 4.55 40.13 107.33 12.07 95.27 Manufacturing

418.97 130.05 288.92 77.92 25.24 52.68 270.98 79.30 191.68 Both

*** *** *** *** *** *** *** *** *** Significance

Total FDI 278.65 176.27 108.22 358.92 ** Total Extra EU FDI 101.28 50.24 21.01 125.02 Total Intra EU FDI 177.38 126.03 87.21 233.90 *** Total Manufacturing FDI 64.45 46.58 41.75 95.32 ** Extra EU Manufacturing FDI 18.26 14.26 7.99 34.19 * Intra EU Manufacturing FDI 46.18 32.32 33.75 61.12 * Total Service FDI 188.90 112.52 59.27 193.00 ** Extra EU Service FDI 70.84 26.98 11.57 63.51 Intra EU Service FDI 118.06 85.54 47.70 129.49 ** Reported coefficients are means. The last column indicates whether and to what extent the FDI distribution differs across types of regions. In particular, ***, **, * indicate a significance level of 1%, 5% and 10%, respectively. Source: Own calculation from FDIRegio database.

133

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table A5.2 The distribution of FDI across types of regions by open sectors: descriptive statistics (2001-03) TYPE OF REGION TOTAL FDI dd dfdg dh dm dn g h i j k1 k2 m n op EXTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op INTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op

Other

Regional Global

2.07 2.91 2.47 1.43 1.66 30.62 2.24 3.94 1.94 2.41 10.29 0.18 0.49 1.85

1.36 3.79 2.23 2.17 2.89 37.37 3.49 6.28 2.92 3.14 21.05 0.22 1.07 3.48

0.84 8.52 3.27 2.41 4.91 152.89 6.85 22.28 26.15 20.58 107.21 1.02 2.01 19.47

0.38 0.59 0.34 0.25 0.39 6.36 0.35 0.62 0.29 0.41 1.71 0 0.1 0.41

0.33 1.16 0.48 0.71 0.82 7.47 0.29 1.18 0.81 0.59 5.68 0.07 0.67 1.06

0.24 1.99 1.21 0.98 2.26 71.55 3.01 6.73 8.72 6.5 41.61 0.36 0.87 8.8

1.68 2.31 2.12 1.18 1.27 24.25 1.88 3.32 1.65 2 8.58 0.18 0.39 1.44

1.02 2.63 1.74 1.45 2.06 29.89 3.19 5.09 2.1 2.56 15.36 0.15 0.41 2.42

0.59 5.53 2.05 1.43 2.64 81.33 3.84 15.54 17.42 14.08 65.6 0.67 1.14 10.67

Significance

*** * *** *** ** *** *** *** *** *** ** ***

*** *** *** *** ** *** *** *** *** *** *** ***

***

* *** *** *** *** *** *** *** 134

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Reported coefficients are means. The last column indicates whether and to what extent the FDI distribution differs across types of regions. In particular, ***, **, * indicate a significance level of 1%, 5% and 10%, respectively. Source: Own calculation from FDIRegio database

Table A5.3 The distribution of FDI across types of EU15 regions by open sectors: descriptive statistics (2001-03) TYPE OF REGION TOTAL FDI dd dfdg dh dm dn g h i j k1 k2 m n op EXTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op INTRA EU FDI dd dfdg dh dm dn g h i j

Other

Regional Global

1.04 2.76 2.38 1.52 1.58 26.39 2.03 3.3 2.31 2.32 11.33 0.19 0.54 1.76

0.63 4.63 2.33 2.48 3.2 42.65 4.03 7.24 3.87 3.89 27.49 0.27 1.39 4.42

0.48 9.36 3.13 2.78 5.23 109.03 6.32 23.69 31.89 23.2 127.33 1.11 2.23 23.38

0.02 0.5 0.31 0.26 0.32 2.4 0.13 0.23 0.35 0.42 1.93 0 0.09 0.35

0.1 1.54 0.59 0.84 1.06 7.41 0.19 1.36 1.1 0.78 7.67 0.09 0.91 1.42

0.04 3.36 1.14 1.18 2.49 29.18 2.56 7.32 10.97 7.89 51.32 0.39 0.94 10.86

1.03 2.31 2.07 1.26 1.26 23.99 1.9 3.07 1.96

0.52 3.08 1.73 1.63 2.14 35.24 3.84 5.88 2.76

0.44 5.99 1.98 1.59 2.73 79.86 3.76 16.37 20.91

Significance * *** * *** *** *** *** *** *** *** * *** * *** *** *** *** *** *** *** *** ** *** *** *** ** ***

* *** *** *** 135

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

k1 k2 m n op

1.9 9.4 0.19 0.44 1.41

3.11 19.81 0.18 0.48 2.99

15.31 76.01 0.72 1.28 12.52

*** *** *** *** ***

Reported coefficients are means. The last column indicates whether and to what extent the FDI distribution differs across types of regions. In particular, ***, **, * indicate a significance level of 1%, 5% and 10%, respectively. Source: Own calculation from FDIRegio database.

Table A5.4 The distribution of FDI across types of EU12 regions by open sectors: descriptive statistics (2001-03) TYPE OF REGION TOTAL FDI dd dfdg dh dm dn g h i j k1 k2 m n op EXTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op INTRA EU FDI dd dfdg dh dm

Other

Regional Global

6.52 3.57 2.83 1.05 1.99 49 3.13 6.73 0.32 2.85 5.79 0.12 0.25 2.24

3.26 1.59 1.97 1.35 2.05 23.41 2.05 3.73 0.4 1.15 4.05 0.06 0.23 1.02

2.06 5.61 3.74 1.14 3.79 304.7 8.7 17.38 6.27 11.47 37.57 0.7 1.27 5.94

1.99 1.22 0.48 0.21 0.68 23.62 1.3 2.35 0.04 0.39 0.75 0.02 0.14 0.68

0.93 0.16 0.19 0.37 0.21 7.62 0.56 0.71 0.06 0.07 0.42 0.01 0.02 0.1

0.95 1.7 1.43 0.28 1.45 218.24 4.58 4.69 0.94 1.68 8.01 0.22 0.62 1.66

4.52 2.34 2.35 0.83

2.32 1.43 1.76 0.97

Significance

***

* ** *** *** *** *** ** **

**

** * ** *

1.11 3.91 *** 2.3 0.86 136

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

dn g h i j k1 k2 m n op

1.31 25.38 1.82 4.38 0.28 2.46 5.04 0.09 0.11 1.56

1.84 15.79 1.49 3.01 0.35 1.08 3.62 0.04 0.21 0.92

2.33 86.45 4.12 12.69 5.32 9.79 29.56 0.47 0.64 4.28

*** ** *** *** *** *** *

Reported coefficients are means. The last column indicates whether and to what extent the FDI distribution differs across types of regions. In particular, ***, **, * indicate a significance level of 1%, 5% and 10%, respectively. Source: Own calculation from FDIRegio database.

137

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table A5.5 The distribution of FDI across types of regions by open sectors: descriptive statistics (2005-07) TYPE OF REGION TOTAL FDI dd dfdg dh dm dn g h i j k1 k2 m n op EXTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op INTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op

Other

Regional Global

1.49 1.8 1.53 0.93 1.67 21.74 2.42 3.65 0.73 3.35 12.45 0.13 0.34 1.95

2.11 2.9 2.25 1.9 2.63 34.78 3.25 6.61 2.8 5.98 29.76 0.32 0.91 3.64

1.03 6.57 2.93 2.23 3.87 126.58 6.79 21.41 20.57 27.75 160.43 1.13 2.69 17.34

0.24 0.49 0.33 0.19 0.4 5.08 0.58 0.84 0.15 0.76 2.13 0.03 0.1 0.41

0.28 0.9 0.46 0.56 0.52 7.64 0.69 1.03 0.61 1.32 8.44 0.06 0.3 0.98

0.38 2.41 1.13 0.77 1.55 60.09 3.02 7.55 8.09 8.56 62.6 0.42 1.32 7.65

1.25 1.32 1.2 0.74 1.27 16.66 1.84 2.81 0.58 2.59 10.32 0.1 0.23 1.53

1.83 2 1.79 1.34 2.11 27.14 2.55 5.58 2.19 4.66 21.32 0.26 0.62 2.66

0.66 4.17 1.79 1.46 2.32 66.48 3.77 13.86 12.48 19.19 97.83 0.71 1.37 9.69

Significance

*** *** ** *** *** *** *** *** *** ***

*** * ** *** * * *** ** *** *** *** *** ***

*** ** *** *** *** *** *** *** *** *** 138

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Reported coefficients are means. The last column indicates whether and to what extent the FDI distribution differs across types of regions. In particular, ***, **, * indicate a significance level of 1%, 5% and 10%, respectively. Source: Own calculation from FDIRegio database

139

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table A5.6 The distribution of FDI across types of EU15 regions by open sectors: descriptive statistics (2005-07) TYPE OF REGION TOTAL FDI dd dfdg dh dm dn g h i j k1 k2 m n op EXTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op INTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op

Other

Regional Global

Significance

0.39 1.79 1.19 0.77 0.89 12.62 1.15 1.82 0.88 2.32 10.83 0.12 0.24 1.65

0.35 3.03 1.54 1.6 1.65 26.28 1.27 5.66 3.7 4.47 33.75 0.31 0.92 3.75

0.4 6.8 2.03 2.25 3.45 68.36 4.08 18.46 24.85 18.85 174.33 1.05 1.94 18.31

*** *** *** *** *** *** *** *** *** *** ***

0.06 0.48 0.22 0.18 0.17 1.11 0.23 0.2 0.18 0.62 1.91 0.03 0.06 0.39

0.05 1.09 0.42 0.59 0.53 5.18 0.21 1 0.81 1.48 10.8 0.06 0.32 1.2

0.11 2.6 0.73 0.8 1.44 21.02 1.52 7.14 10.13 6.5 72.63 0.45 0.95 8.62

*** *** *** *** *** *** *** *** ** *** *** *** **

0.33 1.31 0.98 0.6 0.72 11.5 0.93 1.62 0.7 1.71 8.91 0.08 0.18 1.26

0.3 1.94 1.12 1 1.12 21.09 1.07 4.66 2.89 2.99 22.95 0.25 0.59 2.56

0.3 4.2 1.29 1.45 2 47.33 2.57 11.32 14.72 12.35 101.7 0.59 0.99 9.69

***

*** *** *** *** *** *** *** *** *** *** *** *** 140

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Reported coefficients are means. The last column indicates whether and to what extent the FDI distribution differs across types of regions. In particular, ***, **, * indicate a significance level of 1%, 5% and 10%, respectively. Source: Own calculation from FDIRegio database

141

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table A5.7 The distribution of FDI across types of EU12 regions by open sectors: descriptive statistics (2005-07) TYPE OF REGION TOTAL FDI dd dfdg dh dm dn g h i j k1 k2 m n op EXTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op INTRA EU FDI dd dfdg dh dm dn g h i j k1 k2 m n op

Other

Regional Global

Significance

6.3 1.86 3.01 1.63 5.08 61.49 7.94 11.63 0.07 7.81 19.53 0.2 0.76 3.24

6.76 2.57 4.14 2.7 5.22 57.22 8.45 9.14 0.43 9.97 19.23 0.34 0.91 3.34

3.23 5.79 6.03 2.19 5.32 328.12 16.18 31.62 5.77 *** 58.57 * 112.33 * 1.39 5.28 13.96

1.04 0.49 0.84 0.28 1.41 22.37 2.11 3.63 0 1.37 3.09 0.05 0.31 0.53

0.9 0.38 0.58 0.49 0.49 14.11 1.97 1.12 0.9 0.89 2.21 0.05 0.23 0.42

1.14 1.73 2.51 0.67 1.91 195 8.22 8.96 1.03 *** 15.69 27.89 * 0.29 2.59 4.27

5.26 1.36 2.18 1.36 3.67 39.12 5.83 8 0.07 6.45 16.44 0.16 0.46 2.71

5.86 2.19 3.56 2.21 4.72 43.11 6.49 8.02 0.34 9.08 17.03 0.29 0.68 2.92

1.92 4.06 3.53 1.52 3.42 132.77 7.95 22.67 4.74 *** 42.88 * 84.45 * 1.09 2.69 9.69 142

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Reported coefficients are means. The last column indicates whether and to what extent the FDI distribution differs across types of regions. In particular, ***, **, * indicate a significance level of 1%, 5% and 10%, respectively. Source: Own calculation from FDIRegio database

143

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table A5.8 Main Globalization trends: the role of manufacturing vs. service FDI Map

Variables

manufacturing FDI

service FDI

FDI below the mean

FDI above the mean

FDI below the mean

FDI above the mean

4.1

Quota of corporate managers 1999-01 vs. growth of service productivity 2002-05

0.28 ***

-0.30 **

0.20 **

0.37 ***

4.2a

Quota of SMEs managers 1999-01 vs. growth of service employment 2002-05

0.50 ***

0.27 **

0.22 ***

0.23 **

4.2b

Quota of SMEs managers 1999-01 vs. growth of service productivity 2002-05

-0.29 ***

-0.55 ***

4.3

Quota of clerks 1999-01 vs. growth of service productivity 2002-05

-0.30 ***

-0.39 ***

4.4a

Growth of service employment 1998-02 vs. absolute change in quota of customer service clerks 2002-07

4.4b

Absolute change in quota of customer service clerks 1999-04 vs. growth of service productivity 2002-05

4.6a

Quota of business services professionals 1999-01 vs. growth of manufacturing productivity 2002-05

0.46 ***

-0.22

0.38 ***

0.19 *

4.6b

Absolute change in quota of business services professionals 1996-04 vs. growth of manufacturing productivity 2002-05

0.20 **

-0.10

0.08

0.20 *

4.7

Share of human resources in science and technology on population in 2000 vs. growth of manufacturing productivity 200205

0.20 **

-0.16

0.05

0.22 **

4.8a

Quota of univeristy teachers 1999-01 vs. growth of manufacturing value added 200205

0.19

-0.08

0.19 **

4.8b

Quota of univeristy teachers 1999-01 vs. growth of manufacturing productivity 200205

0.19 **

-0.28 **

0.11

4.5a

Quota of clerks 1996-98 vs. growth quota of clerks 1996-2007

-0.40 ***

-0.73 ***

-0.60 ***

-0.60 ***

4.5b

Quota of Physical, mathematical and engineering science professionals 1996-98 vs. growth quota of Physical, mathematical and engineering science professionals 19962007

-0.03

-0.44 ***

-0.54 ***

-0.06

4.5c

Quota of business services professionals 1996-98 vs. growth quota of business services professionals 1996-2007

4.5d

Quota of plant and machine operators and assemblers 1996-98 vs. growth quota of plant and machine operators and assemblers 1996-2007

0.09

-0.18 **

0.27 ***

-0.19 **

-0.17 **

0.00

-0.29 **

0.29 **

0.16

0.16 *

-0.04

-0.02

-0.22 **

-0.14

0.09

-0.18

0.21 **

-0.09

-0.04

0.05

0.21 **

-0.52 ***

144

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

145

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table A5.9 Spatial autocorrelation diagnostic Moran's I Spatial error: Lagrange multiplier Robust Lagrange multiplier

Statistic -0.65

df 1

p-value 1.484

0.95 0.85

1 1

0.330 0.356

Spatial lag: Lagrange multiplier 0.14 1 0.713 Robust Lagrange multiplier 0.04 1 0.843 These tests have been computed on specification (5) in table 5.8.

146

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

6. Functional trends at regional level By analyzing the geographical distribution of professions, we will assess the regional position in the division of labour and its evolution. This approach is in line with the whole study, which puts into the fore the idea that regional structures cannot only be understood in terms of sectoral specialization. Indeed, the tasks and segments of production inside each sector have to be considered as well. Analyzing professions will enable us to assess the regional position in the division of labour in terms of tasks rather than sectors (section 6.1). However, this raises many difficulties as far as methodology is concerned, mainly in that professional data are based on limited samples of the Labour Force Survey. This analysis will be completed by a more classic sectoral analysis mainly focused on open sectors (section 6.2), which should also be considered as indicative of the regional position in the division of labour – as the correlations analysis between sectoral specialization and professions will show – and offers the great advantage to rely on more robust statistics at regional level.

6.1. Regional position in the division of labour: trends in professions 6.1.1. Data and indicators The source of the data on professions is the Labour Force Survey from 1995 to 2007. Data before 1999 are difficult to use because of geographical changes in NUTS 2 and many missing data. Moreover, before 1999, most of the countries did not provide data at digit-3 level. To identify high level professions, we work from the most refined available level, that is to say ISCO-digit3, and we aggregate professions in order to obtain a synthetic and more robust indicator. To get more stable statistics, we also used three years average data for the following periods: 19992001, 2002-2004 and 2005-2007. Despite these three years average periods, data show strong national differentiations. It seems that an important part of these national discrepancies are strongly related to the inadequacy between national classifications in use and the standard international classifications. We used three different approaches to produce a synthetic indicator of high level professions from the most refined classification of professions: the first is based on a principal component analysis on the share of high level professions at regional level; the second is based on the EGP classification of professions; the third is a list of what could a priori be considered as high level professions. These different indicators are well correlated with each other and give the same general picture of the geography of high level professions in Europe (see Annex 5 for methodological precisions). However, the evolution of our different indicators between 2000 (1999-2001) and 2006 (20052007) shows low correlations between each other. It means that the different indicators of evolutions give a rather different geographical picture of the evolution of the regional position in the international division of labour. Several reasons can explain these low correlations: -

evolutions on a relatively short period are not strong enough to be captured through surveys; 147

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

-

evolutions in the classification of professions – and other statistical issues - could have greater impact than real evolutions and explain the strong national discrepancies observed; the complexity of the geographical pattern of professional evolution itself: at least some of the regional trends are probably not strong enough to be highlighted through surveys.

Our main analysis will be based on high level professions according to the EGP classification. This classification of professions is originally used in social classes’ analyses. However, since it is based on the type of contract, on the autonomy and qualification of the work, it provides a good indicator of the hierarchical level of professions. Usually, the implementation is made at ISCO-digit4 level but we had to work from digit-3 level55. Our indicator of high level professions is finally based on the share of higher managerial, lower managerial and professional workers (EGP I and II). As far as evolutions are concerned, we will focus on evolutions regarding national averages to limit the statistical problems which produce strong national effects.

6.2. Regional functional specialisation The level of professions is strongly related to the participation of the region in global networks. Table 1 illustrates the huge differences in the share of high level professions between global, regional and other regions. The same difference can be observed when data are calculated according to national averages: global players have 3% more high level professions than the national average56 (Table 6.1). However, the gap between the global regions and the others is much deeper in Eastern Europe, where capital cities are concentrating most of the high level functions. Table 6.1. Share of high level professions according to the type of regions in 2006

Type of regions

Eastern Europe*

EU-15

EU*

High level Share of professions EGP high level professions according EGPb to national averagec

other regional players global players other regional players global players other regional players global players

22,9 25,7 32,1 25,8 27,7 33,3 25,6 27,2 33,2

-2,9 -1,2 7,7 -2,4 -1,8 3,2 -2,5 -1,7 3,5

Source: LFS b. High level professions are defined according to the EGP classification (level 1 and 2) c. Data show deviations from national averages * Romania and Bulgaria not included as well as Ireland in Western Europe

55 56

See Ganzeboom, Treiman for the implementation. The choice of the indicator does not influence the results. 148

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The share of high level professions in the regional employment gives an interesting picture of the labour division inside Europe (Fig. 6.1). First, we observe strong national differences but it is difficult to say to which extent these differences are related to problems of professional classification. For example, the fact that all UK and Swedish regions have higher shares than Dutch regions could illustrate the differences in the classification problems rather than in the division of labour. Second, strong geographical patterns still appear from the map: -

North-western and Scandinavian countries have higher shares of high level professions compared to Mediterranean and Eastern countries; Metropolitan and global regions have higher shares of high level professions, which indicate their specific position in the national and international division of labour; Some peripheral areas have relatively high shares of high level professions, especially Northern Scandinavia. Fig. 6.1. High level professions at the regional level in EU, around 2006

High level professions, in 2006

High level professions according to national average, 2006

Share of high level professions in the employment, %

Difference with national share of high level professions 2006, % -27 - -6 -6 - -1.8 -1.8 - 1.4 1.4 - 7 7 - 22.1

5 - 20 20 - 24.9 24.9 - 28.5 28.5 - 32.4 32.4 - 37.4 37.4 - 50.5 Global regions

Global regions

0

1000

2000 Kilometers

Source: LFS

The level of professions is correlated with our indicator of globalization and also shows correlations with the share of high level services (Table 6.2). This is of course in coherence with the theoretical framework of the project: global regions are specialized in some high level services which require high level professions such as corporate managers, higher and lower controllers or scientific professionals. But the relationship is not unilateral: the high level of the workforce qualification also attracts high level functions very well embedded in the global city networks. This relationship between sectoral and functional specialization does not only reflect the difference between global and other regions: similar figures are obtained among the global regions only. However, among 149

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

global players, regions still specialized in manufacturing industry have lower shares of high level professions.

Table 6.2. Correlation between high level professions and regional economic structure around 2005 All regions

Global regions

Regional players

Agriculture Chemicals

-0,44** 0,18**

-0,29** -0,17

-0,19* 0,27**

Transport, electronic and machinery equipment

-0,06

-0,36**

0,00

Trade and hotelsrestaurants

-0,01

-0,03

0,14

High level services (J and K)

0,70**

-0,65**

0,31**

Non market services (L, M and N) -0,02 0,08 0,21* ** Significant at 0.01; * Significant at 0.05 Correlations are weighted by the total employment of the region Source: LFS 2005 to 2007; IGEAT matrix 2004

6.3. Evolution of the regional position in the division of labour The question raised here is whether a high participation in the global networks and high level service specialization is related to a structural shift to higher functions. Since our indicators give an unstable picture of the evolution of professions at regional level, we focus on the evolution regarding the national average, which gives a more stable picture. According to Table 6.3, global regions did not significantly differ from the others in the structural shift to high level functions and professions. In terms of evolution, we can observe complex patterns. Firstly, the evolution of the share of high level professions is not significantly higher in global players. However – and secondly –, global players have created around 45% of the jobs in high level professions while they only represent 40% of the total employment at regional level. This is of course due to the initial concentration of these types of jobs in the global regions. Thirdly, when evolutions are measured with regard to the national mean, global regions appear to have, on average, higher growth rates in high level professions than the other regions of the country. This could mean that when the statistical biases related to national average are eliminated, global regions do perform better than other regions of their country. Finally, in Eastern Europe, the picture is rather different and the gap between global regions and the others have considerably increased during the 2000-2006 period: it means that the capital regions concentrate more and more high level functions in Eastern Europe.

150

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 6.3. Evolution in high level professions according to the type of regions, 2000-2006

Type of regions

Eastern Europe*

EU-15

EU*

other regions regional players global players other regions regional players global players other regions regional players global players

Total share Evolution of of each high level Evolution of the share in type in the professions high level growth of EGP according professions high level EGPb, % professions, to national % averagec, % 1,3 2,5 4,2 3,3 2,5 3,1 3,1 2,5 3,3

0,9 7,9 6,1 22,9 24,3 38,0 23,8 32,2 44,1

-1,4 -0,2 2,3 -0,3 -0,1 0,3 -0,4 -0,1 0,4

Source: LFS d. High level professions are defined according to the EGP classification (level 1 and 2) e. Data show deviations from national averages * Romania and Bulgaria not included as well as Ireland in Western Europe

The correlations between the initial economic structure and a shift to higher level of professions (according to national average), are not significant (Table 6.4). However, they hide differences between Eastern and Western Europe: in the former, the growth of high level professions has been higher in capital regions with higher globalization index, contrary to Western Europe where the geographical pattern is far more complex. In other words, Eastern capitals are increasing the gap against non global Eastern regions and narrowing the gap vis-à-vis Western metropolitan regions. However, when evolutions are evaluated in absolute rather than relative figures, correlations with the specialization in high level services are significant (Table 6.4). Indeed, although global regions do not specialize more than the others in higher functions, they create more jobs in higher functions because most of these jobs are initially concentrated there.

151

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 6.4. Correlation between economic structure and functional evolution compared with the national average, 2000-2006 Evolution with respect to the national average, 2000-2006 Agriculture Economic structure in 1995 (share of added value in %)

High level services (J and K) Non market services (L, M and N)

Absolute evolution 2000-2006

0,00

-0,25**

0,03

0,46**

-0,01

-0,27**

Economic growth 1995High level services 2004 (J and K) -0,05 ** Significant at 0.01 Correlations are weighted by the total employment of the region Source: LFS 2005 to 2007; IGEAT matrix 2004

-0,12

At the regional level, despite the statistical uncertainties related to the data, two major evolutions should be underlined (Fig. 6.2): - the positive evolution in most Eastern regions towards a level which is now similar to (or even higher than) Mediterranean regions; - the fast evolution in Eastern capitals, which have increased the gap against the rest of the nation. We do not observe such a process in the Western metropolises, which show more complex and heterogeneous evolutions as regards their national environment. Finally, Fig. 6.3 illustrates the geographical evolution pattern of high level professions in both absolute and relative terms.

152

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 6.2. Evolution of high level professions with regard to the national average, between 2000 and 2006 Evolution of the profession level

s r e t e m o l i K 0 0 0 1

0

Difference in the share of high level profession between 2000 and 2006, relative to national average -5.2 - -2 -2.- -0.5 -0.5 - 0.5 0.5 - 2 2-7 No Data Global regions

Fig.6.3. Evolution of high level professions in absolute figures, between 2000 and 2006 Evolution of high profession level 2000 - 2006

© 2009 IGEAT

500 km

© EuroGeographics Association for the administrative boundaries

Employement

Origin of data: Labour Force Survey

400 200 100

Relative evolution (%) -30.4 - 0 0 - 11.9 11.9 - 24.2 24.2 - 42.3 42.3 - 84.5 No data

153

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

6.4. Conclusions 1. The Labour Force Survey provides a good picture of regional functional specialization across Europe through professions. However, it suffers from statistical biases which make the evolution of professional specialization difficult to evaluate. It confirms the analyses on sectoral data which had highlighted that the evolutions measured through LFS or through regional accounts could show significant differences. 2. Regions’ functional specialization is strongly related to the sectoral specialization in high level services and to higher participation in the global networks. These three processes are highly correlated between each other. In other words, global regions are specialized in high level services and concentrate the higher functions whatever the sector. 3. Because of the concentration of higher functions in global regions, these regions have created most of the jobs in the upper professions: around 44% of the jobs in higher professions have been created in the global regions whose share in total employment is about 40%. 4. However, we found no empirical evidence of a relationship between an initial specialization in upper functions, high level services (finance and business services) and strong participation in the global networks, on the one hand, and a shift to upper functions, on the other hand. Put differently, the global commanding regions with high functions have not increased their specialization in the higher functions more than the others. 5. Yet there is clear sign that in Eastern Europe the global capital regions have reinforced their position compared to the rest of the country and concentrate these higher functions even more than before.

6.5. Regional functional dynamics of open sectors The objective of this section is to understand the sectoral dynamics related to globalization within Europe, especially in the open sectors. Moreover, we will try to understand these dynamics in relation with the regional participation in global networks and the role in the international division of labour. 6.5.1. Data Dynamics have been evaluated in the 1995-2004 period using the IGEAT matrix. These data based on Eurostat statistics have been completed with national statistics to provide a database of a relatively detailed sectoral structure at different points in time. Data have been deflated on a national basis. 6.5.2. A descriptive analysis of the open sector We present here several maps of the main open sectors. Each map shows the initial added value in the sector (1995) and the annual average growth between 1995 and 2004. Each sector is characterized by specific dynamics but we have to keep in mind that the most dynamic regions are able to reach high growth levels in most of the sectors. The chemical sector is very concentrated. While negative growths have been observed in some of the major poles such as Lombardia, Ile-de-France, Northern England, the German chemical sector remains relatively dynamic. The growth has been particularly high in Ireland whereas new member States have not benefited from considerable offshoring in this often very qualified sector. The transport equipment industry has a different geographical pattern of growth, with major growth 154

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

poles in Germany, England – though with a more dispersed industry – and in some regions in the new member States which have benefited from massive offshoring from Western European automotive companies, notably to Slovakia, Slovenia, Poland or Romania. On the other hand, Belgian, French, Spanish, Portuguese or Italian activity has severely dropped to the benefit of new locations in the Central and Eastern Europe. Fig.6.4. Evolution of chemical industry and transport equipment, 1995-2004

© 2009 ULB-IGEAT

500 km

Added value 1995 (millions €) - Chemistry 5000 2500 1250

Annual average Added value growth rate 1995 - 2004 - Chemistry -24 - -11 -11 - -3 -3 - 0 0-7 7 - 14 14 - 34

500 km

© 2009 ULB-IGEAT

Added value 1995 (millions €) - Transport equipment 5000 2500 1250

Annual average Added value growth rate 1995 - 2004 - Transport equipment -15 - -4 -4 - -1 -1 - 0 0-5 5 - 10 10 - 33

Origin of data: Eurostat, National Statistical Offices. © EuroGeographics Association for the administrative boundaries

The trade and hotels/restaurants sectors are mainly basic market services for population. Dynamics are globally related to the general economic growth, even if some specific national conditions explain better performances in some countries. This is why the highest growth rates are found in the new member States, United Kingdom, Ireland or Spain. In general, the major poles – such as capital regions – enjoyed moderate growths compared to the rest of the nation.

155

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 6.5. Evolution of the trade and hotels/restaurants sector, 1995-2004

© 2009 ULB-IGEAT

50000 25000 12500

Annual average Added value growth rate 1995 - 2004 - Trade -10 - -2 -2 - -0.5 -0.5 - 0 0-3 3-6 6 - 16

500 km

© 2009 ULB-IGEAT

500 km

Added value 1995 (millions €) - Trade

Added value 1995 (millions €) - Hotels and restaurants 50000 25000 12500

Annual average Added value growth rate 1995 - 2004 - Hotels and restaurants -9 - -1 -1 - -0.5 -0.5 - 0 0-3 3-7 7 - 23

Origin of data: Eurostat, National Statistical Offices. © EuroGeographics Association for the administrative boundaries

Transport and communication as well as the financial sectors have shown very specific dynamics. These very central sectors at both European and national scales had the highest growth rates in the European major metropolitan areas and in the main national urban areas. In the financial sector, we can underline the growth of Paris, London, Amsterdam, Milan and Frankfurt in regard with the rest of the country. National economic capitals of Mediterranean or Eastern Europe also enjoyed systematically better growth rates than the rest of the national territory. These evolutions are to be related with the evolution of this sector which has privileged international development within the global networks to the detriment of more traditional activities in relation to the financing of the local/national activities and the relation to clients. Finally, business services and health sectors have been the most dynamic in Europe during the last decade. In spatial terms, the metropolitan areas do not benefit from the fastest growth rates and – in many cases – more peripheral areas have higher growth rates. In both sectors, this could be interpreted as catching up mechanisms.

156

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 6.6. Evolution of transportation/communication and financial sectors, 1995-2004

© 2009 ULB-IGEAT

500 km

© 2009 ULB-IGEAT

500 km

Added value 1995 (millions €) - Transport and communication

Added value 1995 (millions €) - Financial intermediation

50000

50000

25000 12500

25000 12500

Annual average Added value growth rate 1995 - 2004 - Transport and communication -14 - -2 -2 - 0 0-2 2-4 4-6 6 - 15

Annual average Added value growth rate 1995 - 2004 - Financial intermediation -15 - -6 -6 - -2 -2 - 0 0-3 3-6 6 - 28

Origin of data: Eurostat, National Statistical Offices. © EuroGeographics Association for the administrative boundaries

Fig. 6.7. Evolution of business services and health sector, 1995-2004

© 2009 ULB-IGEAT

© 2009 ULB-IGEAT

500 km

Added value 1995 (millions €) - Real estate, renting and business activities 50000

50000

25000 12500

25000 12500

Annual average Added value growth rate 1995 - 2004 Real estate, renting and business activities -7 - 0 0-2 2-4 4-6 6 - 10 10 - 22

500 km

Added value 1995 (millions €) - Health

Annual average Added value growth rate 1995 - 2004 - Health -2 - 0 0-2 2-4 4-7 7 - 10 10 - 21

Origin of data: Eurostat, National Statistical Offices. © EuroGeographics Association for the administrative boundaries

157

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

6.6. Sectoral/functional trends and the participation in the global networks The relationship between sectoral dynamics and globalization is complex. For example, open sectors show very different patterns of growth according to the level of globalization of the region. In addition, these patterns are sometimes very different in Western and Eastern Europe (Table 6.5). The manufacturing industry has had lower growth rates in global regions in the Western parts of Europe, whereas this was not necessarily the case in the new member States. While in Western Europe, deindustrialization in the metropolitan areas has continued, the Eastern metropolitan/capital regions have been able to attract investments in both manufacturing industries and high level services. Eastern capital regions have been particularly dynamic in the open manufacturing sectors, especially the chemical and automotive industries. Apart from the usual advantages of metropolitan regions, this is probably due to the presence of cheap medium or highly qualified workforce in these regions. Table 6.5. Economic structure and economic growth according to the type of region, 1995-2004

Agriculture Extraction Manufacturing industry Food industry textile Wood Paper, printing and edition Chemical industry Plastic and rubber Mineral Metal Machinery Electric and electronic equipment Transport equipment Miscellaneous industry Energy Construction Trade Hotels/restaurants Transport/communication Financial intermediation Business services Administration Education Health Personal and cultural services Total

Sectoral structure EU-15, % in 2005

Sectoral structure EU-12, % in 2005

Average annual growth in EU-15, 1995-2004

Global Regional regions players 0,9 3,3 0,2 0,3 16,2 19,8 1,6 2,4 0,6 0,6 0,2 0,5 1,9 1,5 2,6 2,1 0,7 1,0 0,4 0,8 1,9 2,8 1,8 2,4 2,3 2,2 1,9 2,6 0,5 0,8 1,8 1,9 4,8 6,1 11,6 11,2 2,6 2,9 7,7 6,2 7,2 4,1 25,4 20,0 5,9 6,5 4,4 5,6 6,2 8,1 5,1 4,0 100,0 100,0

Global Regional regions players 2,9 5,3 0,4 1,6 16,1 24,5 2,5 3,5 1,2 1,4 1,0 1,0 1,6 1,4 2,0 2,7 0,7 1,4 0,7 1,4 1,3 3,5 0,8 1,9 2,2 2,7 1,4 2,6 0,7 1,1 2,9 3,4 5,3 6,0 15,3 14,7 1,9 1,6 12,1 7,7 5,8 3,0 18,3 13,1 7,1 6,2 4,3 5,1 3,5 4,1 4,0 3,6 100,0 100,0

Global Regional regions players -0,84 -0,03 -3,45 -4,66 0,31 0,60 0,97 1,28 -3,08 -3,49 1,04 0,98 0,41 -1,74 0,41 1,14 1,66 1,32 -2,15 -1,43 -1,24 -0,69 0,70 2,49 0,46 0,91 2,12 2,84 -0,70 -1,10 -0,28 -0,43 1,64 1,88 2,41 2,03 3,46 3,04 3,06 2,24 3,11 2,54 3,97 3,91 1,36 1,48 2,90 2,84 3,81 4,27 3,62 2,93 2,47 2,34

Other 4,0 0,9 17,6 2,7 1,5 0,6 1,3 1,2 0,8 1,2 2,7 2,0 1,7 1,1 0,9 2,2 7,3 11,1 3,4 6,2 3,9 18,8 7,1 6,0 7,8 3,6 100,0

Other 10,5 1,2 19,1 4,2 2,7 1,1 1,1 1,7 0,8 1,4 1,8 1,4 1,4 1,0 0,7 3,6 5,5 15,8 1,5 7,3 2,8 12,5 7,2 5,7 4,2 3,2 100,0

Other -0,93 1,83 0,77 1,53 -3,07 0,91 -0,53 1,46 1,01 0,51 1,20 1,97 1,40 1,50 0,76 0,48 2,76 1,30 2,97 1,71 1,39 4,39 2,17 2,49 4,16 2,24 2,16

Average annual growth in EU-12, 1995-2004 Global Regional regions players -2,47 -1,87 2,40 0,00 3,64 2,66 1,18 0,01 1,06 -2,72 9,87 2,84 3,32 2,41 -1,92 4,50 11,92 6,47 1,48 2,22 5,87 1,69 0,78 1,66 10,39 5,90 11,18 9,54 4,65 0,01 3,56 -3,36 3,42 -0,67 6,11 6,63 0,39 2,94 6,37 1,21 2,98 -1,06 6,58 6,26 6,44 9,25 6,20 7,35 5,32 6,63 4,08 2,99 4,76 3,08

Other -1,44 -0,32 1,70 -0,40 2,52 1,70 2,50 2,25 8,06 3,20 1,75 1,81 2,26 4,71 -2,18 -2,56 0,80 7,63 3,31 1,64 -2,73 4,71 12,83 9,22 8,29 3,08 3,08

Source : IGEAT matrix In yellow: the open sectors

Low qualified market services (trade and hotels/restaurants) show higher growth rates in the global regions in EU-15, while the contrary is true in EU-12. This clearly highlights the fact that global regions do not only grow in highly qualified sectors but also in lower qualified market services such as retail trade, hotels/restaurants and personal services. In Eastern Europe, we observe a catching up process in rural/peripheral areas where some basic services were underdeveloped. Transport/communication and financial intermediation are specific to global regions, which also enjoyed higher growth rates between 1995 and 2004. This means that these sectors tend to concentrate more and more in global regions. We have already underlined the specific reason behind this process in the financial sector. Business services show a very different geographical pattern: while these are very characteristic of the most global regions, the non global regions had higher growth rates in this sector in Western Europe, contrary to Eastern Europe where we observe an increasing gap between capital regions and the rest of the country. This is in line with the results of the analysis through professions. 158

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Finally, non market services – administration, education and health – enjoyed higher growth rates in non global regions. At least to some extent, this can be interpreted as a catching up process of regions which are more deprived in terms of public services, especially in the new member States. For example, new universities or hospitals often locate in more peripheral areas. 6.7. Conclusions 1. Not surprisingly, open sectors do not show homogenous patterns of growth: we can indeed observe great differences between manufacturing, basic market services, and high level services. 2. High level services (transport and communication, finance and business services) are very specific to global regions where we find high correlation rates between the share of these sectors and the globalization index, as well as the share of high level professions. 3. In line with what has been found with the evolution of professions, there is no empirical evidence that global regions have performed better than others in high level services, except in most of the Eastern capital regions. 4. Indeed, inside the high level services, we have found different spatial patterns between financial and business services. The first have concentrated more and more in the global regions, while the second – which represent a much higher share of employment – have not. In the financial sector, international functions have been privileged while in the business services, less global regions have sometimes been the most dynamic ones. 5. The business services sector is of course very large, and it would have been very interesting to distinguish between higher functions (such as advice or research) and less qualified services (such as security or call centres).

159

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

7. Success factors of benefiting regions 7.1. Introduction In the previous two chapters we have described the globalization trends occurring at the regional level. Chapter 4 contains a general relationship between structural conditions and globalization trends in the different types of regions; Chapter 5 has focused more on the capacities of FDI attraction of European regions. Chapter 6 has been devoted to the analysis of the position of regions in the international division of labour according to tasks and not to sectors. As a general conclusion of the empirical part so far developed, we can say that most of the theoretical conditions that favour growth are mostly present in global regions, with no difference between Eastern and Western countries (Table 7.1). Attractiveness of FDI is higher in global regions (Table 5.1); once a sectoral breakdown is analyzed, attractiveness of FDI in global regions is especially high value-added sectors (Table A.5.2 and A.5.3). Global regions are also mostly endowed of high-value functions (Table 6.1) and attract high value functions more than their national average (Table 6.3). Again, this is true for both Eastern and Western countries; Eastern global regions even register a more marked endowment and evolution of high value functions with respect to the national average compared to Western global regions, as easily expected given the concentration of FDI and high-value activities in capital regions (Table 3.5).

Table 7.1. Theoretical reasons for competitiveness and empirical evidence Theoretical reasons for competitiveness: winning features in globalization trends

Regions with a higher presence of winning features

Empirical evidence so far achieved

Attractiveness of FDI

Global regions

Table 5.1

Attractiveness of FDI in high-value added sectors

Global regions

Tables A.5.2 and A.5.3

Endowment of high-value functions

Global regions

Table 6.1

Attractiveness of high-value functions

Global regions

Table 6.3

Endowment of R&D activities

Global regions

Table 3.5

Human capital (High education labour force)

Global regions

Table 3.5

We now turn to some considerations on the general economic performance of regions in Europe, with the aim to highlight whether regions, and in particular global regions, that demonstrate a higher economic performance, are also the ones that have been able to adjust to globalization trends in the best way, being endowed of the structural preconditions, in terms of functions, human and social capital, that are highlighted in the literature to be the main success factors in front of globalization. In particular, this chapter will analyze which regions benefit from globalisation, and which are the conditions allowing them to be as such.

160

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

‘Benefiting regions’ are defined here as those regions that maintain and even improve their competitive positions in the European economy thanks to globalization processes, i.e. those regions able to increase their relative production capacity (measured in terms of GDP) more than the European average amid globalization. This may result from the following different economic growth conditions: g. Employment growth takes place in both high and low value-added functions, the former having a greater effect in quantitative terms than the latter. If this occurs, both employment and productivity increase, and therefore GDP as well. h. Employment losses take place in low value-added activities and are more than off-set by high value-added functions. In this case, employment losses are more than off-set by productivity increases, and GDP increases. i. Employment increases in low value-added functions, accompanied by a limited loss, if any, of high-skilled jobs. GDP increases despite the loss of productivity. Since all three types of economic growth conditions lead to an higher than average regional GDP growth rate, it is first useful to investigate if global regions are benefiting to a larger extent than regional players and other regions. In Table 7.2 we represent the average annual GDP growth rates in two periods of time of the three typology of regions in front of globalization, as well as a test to analyze if these growth rates are significantly different. It turns out that in the first period of time, i.e. 1999-2002, global players significantly outperformed the other typess of regions in terms of GDP performance. This held true for both regions in the Old 15 member countries and in the New 12 member countries. Interestingly enough, in western regions regional players were the second performers, close to global players, whereas in eastern regions global players by far outperformed the other regions (second best performed) and the regional players. In the second period of time, again global players were the best performers among European regions, but not significantly overall and in Western countries. In Eastern countries, on the contrary, the growth rate of global players was significantly higher. Because of the existence of strong national effects, the performance of the three regional typologies has been analyzes also with respect to their national averages, in term of average annual differential growth rate of these regions with respect to their countries. In this way, it is possible to see if global players have been leading their respective countries in terms of growth rates. In Eastern countries, the differential of global players with respect to their countries is high and significantly different from the one of the other regions in both periods. In western countries the differential growth rate is larger in both periods but significantly larger only in the first period. If all European regions are taken together, being a global players appears to increase significantly the possibility to be a benefiting region and to lead the country in terms of growth.

161

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 7.2 Growth performance of the three types of regions, 1999-2005. All regions Global Regional Other players players regions

F

Old 15 regions Global Regional Other players players regions

F

New 12 regions Global Regional Other players players regions

F

Growth rate 1999-2002

3.17

2.22

1.87

8.21 ***

2.54

2.25

1.62

4.04 **

6.22

2.14

3.20 24.28 ***

Growth rate 2002-2005 Differential growth rate with respect to the country 1999-2002 Differential growth rate with respect to the country 2002-2005

2.76

2.47

2.09

2.06

1.91

1.79

1.74

0.22

6.78

4.43

3.93

9.28 ***

0.37

-0.39

-0.83

7.33 ***

0.12

-0.16

-0.89

5.08 ***

1.54

-1.08

-0.48

9.89 ***

0.16

-0.15

-0.41

4.39 **

0.03

-0.13

-0.24

0.97

0.77

-0.21

-1.30

6.64 ***

The analysis of the factors allowing regions to benefit, will be carried out with two methodological approaches. In the first case, we describe statistically the different economic growth patterns of European regions, and we underline which structural features are statistically significant for the groups of regions with an economic profile of their growth (Sec. 7.2). A second methodological way is to analyze the performance of global players to see if there are indeed different types of global players which allow them to be benefiting (sec. 7.3).

7.2. Regional growth patterns and structural characteristics 7.2.1. Definition and description of the patterns followed by European regions Regions of the three groups that are identified in Chapter 3 (global players, regional players and other regions) will be in this part of the project analysed in terms of their economic performance. Three measures are of interest to analyze performance, related the one to the other: -

growth rate of GDP; growth rate of GDP per employee (i.e. of productivity); growth rate of employment.

The three measures are not independent, since GDP is by definition the product of the multiplication of productivity and employment. At the same time, productivity growth is sometimes obtained through employment cuts.

To illustrate the economic performance of the three groups of regions we analyse the regional patterns of growth through the contemporary inspection of the relative growth of productivity and the relative growth of employment, which are plotted on the vertical and horizontal axis of Fig. 7.1, respectively. We chose to evaluate these variables not in absolute terms but relative to the trend they follow in the whole EU in order to detect the regional specificities beyond the processes happening at general level.57 The two indicators can hence be represented by the following formulas:

57

Camagni, R. (1991), “Regional Deindustrialization and Revitalization Processes in Italy”, in Rodwin L. and Sazanami, H. (eds.) Industrial Change and Regional Economic Transformation, Harper Collins, London, pp. 137167. 162

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

1/ 5

 Pr od r2005    2000   Pr od r 

1/ 5

 Empr2005    2000   Empr 

1/ 5

2005   Pr od EU    2000   Pr od EU 

(7.1)

1/ 5

2005  Emp EU     2000   Emp EU 

(7.2)

This representation has an interesting feature: a 45° negatively sloped line passing through the origin is the locus where the value added of a region grows at the same level of the EU. In fact, a region can have the same growth of value added of Europe if it has the European level in both employment and productivity growth, if an higher than average productivity growth is coupled with a lower than average employment growth and, finally, if it has a lower than average productivity growth coupled with a higher than average employment growth. On the same graph, it is hence possible to control for three indicators at the same time and make six different patterns of growth emerge (Fig. 7.1): 1. virtuous cycle, when higher than average productivity growth generates good performance in both employment and output; 2. restructuring, when a higher than average productivity growth is reached through severe employment cuts, leading nevertheless to good output performance; 3. dropping-out, when productivity growth is reached by closing down inefficient production units, generating lower than average production growth; 4. de-industrialization, defined as a vicious cycle in which employment cuts are unable to restore competitiveness, a condition that perpetuates job losses and low output growth; 5. industrial conservatism, when poor productivity growth is accompanied (and sometimes explained) by a better than average employment growth, generally due to public assistance and productive rescues; 6. sheltered development, when explicit or implicit assistance policies spur the initial development of the areas, notwithstanding low productivity performance. Benefiting regions, defined as in Section 7.1, fall in three quadrants, namely 1, 2 and 6. The regions which position in the virtuous cycle increase both in terms of productivity and employment growth more than the average; the ones in the restructuring area achieve higher than average productivity growth through severe employment cuts, leading nevertheless to good output performance: an apparently good productivity performance is the result of the simple suppression of productive units, with no or scarce positive counterparts. In the case of sheltered development, the good employment condition generally does not come from endogenous regional development but rather from exogenous (somehow artificial) intervention policies which are unlikely to induce a virtuous cycle in the medium and long run. All other situations register a negative growth rate and are hence non-fit for benefiting. In Fig. 7.1, we represent the patterns for all European Nuts 2 regions, with different symbols indicating the different role they play in the globalization process.

163

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

It is immediately clear that there are regions of all three typologies in each of the 6 quadrants.58 This is a signal that the openness to globalization is not by itself enough for a region to have a good economic performance, since other factors are also important. Among the factors which are important in regional performances, probably the most important is the national performance. Two ways of interpreting the relationship between regions and countries are present in the literature: a generative approach in which the country is the sum of the regions which compose it, and a distributive approach where it is national growth which is determined and the regions of the country simply compete between them to get the largest share of country growth (Richardson, 196759). As it will be clear in the chapter which will use the MASST model, we believe in fact that a mixed approach is more valuable, since some factors are playing at national level and other factors are effective at regional level. In any case, country effects are strong at the level of the European union, and there is also a clear country convergence effect which made regions in the New 12 member countries grow more rapidly that their counterparts in the Old 15 member countries. For this reason, we splitted the patterns picture into two, one for the Old 15 countries (Fig. 7.2) and one for the New 12 countries (Fig. 7.3). By looking at the patterns followed by western regions (Fig. 7.2), one can observe that they place on average slightly below the European mean in terms of productivity growth. On the contrary, employment growth has been strong in the period of analysis, which make a relatively larger number of regions falling into the two quadrants of sheltered development and industrial conservatism. Among the best performing (virtuous cycle) regions, there are mainly regional players and global players, signalling that for non global regions it is hard to be growing in periods of increasing globalization. However, there are global players also in the lower quadrants, showing again that being open is not enough to perform well. If we look at the patters followed by Eastern regions (Fig. 7.3)60, we observe that almost all New 12 regions are above the EU mean in terms of GDP growth. Many of them, however, achieve this positive GDP performance with a restructuring process, in which the lower value added jobs are cut and a smaller number of higher value added jobs are created. It is also interesting to observe the large number of global players in the upper part of the virtuous cycle quadrant. Global regions, in the East, appear to have been especially effective in applying their structural and sectoral features to achieve positive economic performance.

58

It is also clear that Polish and Romanian data are rather suspect because it is possible that GVA and employment data are estimated the one using the other withot any correction. 59 Richardson, H.W. (1967) Regional Economics, University of Illinois Press. 60 Again, the analysis does not consider Polish regions because of their strange data. 164

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 7.1 Regional Patterns of Growth 20

Relative productivity growth 2000-2005

ro42 ro31

Other regions Regional players Global players

ro11 ro22

15

ro21 ro12 ro32

sk01

10

hu10

hu31 cz08 cz01 hu33

pl41 pl52

-5

pl34

pl61

hu21 sk04 lt00 hu32 bg42cz03 sk03 hu23 cz04 cz07 cz05 hu22

ee0

bg41

cz06

cz02

sk02

pl12 pl21 pl11 pl33

gr13

5bg32

lv00

gr30 ie02 gr25 gr43 nl11 gr12 gr14 pl42 bg33 be31 es11 si01 gr23 gr41 lu00 nl23 bg34es21 es43 fr10 bg31 gr24 at13 nl42 pt17 nl21 es22 gr22 fr82 es24 es51 pt16 itc2 fr41 fi20 at11 ded2 nl41 uki1 es12es41 be24 fr81 pt15 itf2 fi13 dk01 de50 ukj4 ukk2 nl34 be23 nl33 pt11 fi19 fi1a nl22 be25 deg0 be32 nl32 ukj3 fr61 uke1 be21 itf4 pl43 ite4 itc3ukc2 fi18 nl31 fr23 ded3 ded1 dec0 cy00 ukh2 dee0 be34 nl13 itd4 ukh1 be35 pt18 ite1 de60 ukk1 at31 ukd4 ite3 fr22 ukf3 fr52 itc1 itd3 nl12 de25 ukj1 at21 itd1 fr71 be22 ukf2 ukk4 ukm2 at22 ukf1 ite2 itc4gr42 de80 0 de27 se11 uke3 ukg2 itd2 itd5 be33 itf5 ukh3 at32ukd2 de22 se33 itf3 de91 de24 de26 at34 itg1 at33 ukj2 mt00 de21 uke4 ukd3 de14 de71 de73 ukm3 at12 ukk3 se12 fr42 deb3 ukg1ukl1 fr43 itf6 dea5 dea1 dea3de11 uke2 de92 se31 0de72 deb2 ukn0 fr30 fr83 ukl2 fr24 ukc1 de23 de42 be10 de12 se23 itg2 dea4 dea2 de94 ukg3 fr62 def0 deb1 se32 uki2 fr26 fr25 de41 de93 itf1 fr51 de30 se21 ukd1 de13 ukd5 se22 pl63 pl62 gr11 pl51 pl32gr21 pl31

Relative employment growth 2000-2005

ie01

es62 es42 es70es13 es52 es53 pl22

fr72

fr63 fr21

-5

165

es61 es30 es23

fr53

5

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 7.2 Regional Patterns of Growth in Old 15 member countries 5 gr13

Relative productivity growth 2000-2005 Other regions Regional players Global players

gr30 ie02 gr25

gr11

gr21 nl11

gr43

gr12 be31

ie01

gr41 lu00

gr23

fr10 at13

nl42

pt17

nl23 nl21

gr22 pt15

es11 bg34 es43 es21 es22

fi20

se21

Relative employment growth 2000-2005

gr24

nl41 fr82 uki1 at11 pt16itc2 be24 fr81 dk01 itf2 de50 ukj4 nl34 ukk2 be23 deg0 nl33 pt11 nl22 fi19 fi13 be25 be32 nl32fr61 ukj3 ite4 fi1a ukc2 uke1 be21 itf4 fi18 itc3 nl31 fr23 ded3 ded1 ukh2 dee0 be34 ukh1itd4 nl13 be35 0 dec0 ukj1 pt18 ite1 de60 at31 ukd4 ite3 fr22 fr52 ukf3 nl12 itc1 de25 ukk1 ukf1itd3 at21 itd1 fr71 be22 gr42 ukf2 ukk4 ukm2 at22 ite2 de80 de27 ukg2 itd2 uke3 itc4 itd5 be33 se11 itf5 de22 ukh3 at32 ukd2 se33 de91 de24 de26 at34 itg1 at33 itf3 ukj2 de21 uke4 ukd3 de71 de14 de73 at12ukm3 dea1 se12 fr42 deb3 ukg1 ukl1 itf6 dea3 uke2 fr43 dea5 se31de92 de11 deb2 ukn0 de72 ukl2 fr24fr25 ukc1 de93 de23de12 se23 de42 be10 fr30 dea4 dea2 de94 ukg3 def0 deb1 se32 uki2 itf1 fr26 de41 de30 ded2 fr41

-3

gr14

ukd5 se22 de13

es12 es41

es24

es51 es62 es42 es13 es70 es53

ukk3

fr83 fr62

es23

itg2 fr51

ukd1

fr63

fr53

fr21

166

5

es30

fr72

-5

es61 es52

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 7.3 Regional Patterns of Growth in New 12 member countries 20 Relative productivity growth 2000-2005

ro42 ro31

Other regions Regional players Global players

ro11 ro22 ro21 ro12 ro32

sk01

hu10 hu31 hu33

pl41 pl52

pl12

pl34

cz08 cz01

hu21 sk04 lt00 hu32 bg42 cz03 hu23 cz07 cz04 sk03

hu22

pl33 pl42

cz02

sk02

cz05 lv00

bg32

pl11 pl62

bg41

cz06

pl21 pl61

ee0

pl63

pl51 pl31

pl32

si01

bg31

bg33

bg34

pl43

-5

mt00

0

-5

167

Relative employment growth 2000-2005

cy00 pl22

5

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

7.2.2. Growth patterns and structural features of global players We have seen in Section 7.2.1 that global players, though generally performing well, are not necessarily performing more than the EU average. In this sub-section, we will therefore analyze the structural characteristics associated with the different growth patterns that regions follow. As we have evidenced in Section 7.2.1 that the patterns followed by Eastern and Western regions are different and strictly linked to national effects, the analysis cannot be performed with all regions together, but will be run separately on the Old15 and the New12 member countries. If we re-run the patterns performance by comparing the values of the various regions to the means of their respective area of belonging and count the number of regions belonging to each quadrant, some interesting results emerge (Figure 7.4). First of all, it is evident that being a global player is not enough to be a benefiting region in western countries, since despite being compared to the EU15 mean rather than to the fast growing Eastern regions, Western global players are more often in the dropping out or de-industrialization quadrants than in the virtuous cycle or restructuring ones (Table 7.3). Western regional players are more equally spread, and it even turns out that the highest percentage of benefiting regions is found, in the West, among the other regions. All this analysis highlights some degree of convergence within the Old 15 regions. The results in the New12 countries are much different. 8 out of 10 global players are indeed benefiting regions even if compared with the New12 averages, and 7 of them are in the virtuous cycle quadrant. Also for regional players there is an over-representation in the virtuous cycle quadrant, but less strong and accompained by a large number of regions in the industrial conservatism one (Table 7.3). It hence appears that being a global player, in the East, is strictly linked with being a benefiting region. Table 7.3 Number of regions in each quadrant by typology, New12 and Old15 regions computed separately on respective averages.

Quadrant q1 virtuous cycle q2 restructuring q3 dropping-out q4 de-industrialization q5 industrial conservatism q6 sheltered development TOTAL Benefiting regions (q1+q2+q6) Benefiting regions %

Global players Old15 6 8 10 19 2 2 47

Regional players Old15 17 18 14 24 15 9 97

Other regions Old15 13 15 11 9 8 5 61

16 34.0

44 45.4

33 54.1

Old15 36 41 35 52 25 16 205

Global players New12 7 0 0 1 1 1 10

Regional players New12 9 5 0 6 10 4 34

Other regions New12 2 2 0 3 4 1 12

New12 18 7 0 10 15 6 56

93 45.4

8 80.0

18 52.9

5 41.7

31 55.4

Maps 7.1a and 7.1b represent the benefiting regions among global players (left) and regional players (right), calculated as in Table 7.2 using Old15 and New12 averages.

168

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 7.1b Benefiting regions among regional players

Map 7.1a Benefiting regions among global players

Politecnico di Milano - September 2009

Politecnico di Milano - September 2009

Patterns 2000-2005 est-west 1 Virtuous Cycle 2 Restructuring 6 Sheltered Development

Patterns 2000-2005 est-west 1 Virtuous Cycle 2 Restructuring 6 Sheltered Development

169

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

In order to evidence the characteristics of benefiting regions, we run an Anova analysis on the values that structural variables assume within the benefiting and non benefiting groups of global and regional players in the East and in the West. Unfortunately, for global players in the East the small number of observations (and the fact that almost all these regions are benefiting) precludes the use of this type of analysis. When analyzed are the characteristics which allow global players in Old 15 member countries to benefit (Table 7.4), it is evident that the winning regions are more than average characterized by the presence of high growth metropolitan areas (MEGAs). These areas are also dynamic in terms of attraction of population, so that the efficient urban system appears to be one key of development, as evidenced also by the presence of advanced tertiary sectors (I, Transport, storage and communication and J Financial intermediation) and of high level government functions (public managers). The benefiting global players in the West do not appear to be characterized by an important manufacturing production or productivity, but their manufacturing production is more than average specialized, and especially concentrated in highly growing dynamic manufacturing sectors, where these regions appear to exploit localization economies, and are characterised by the presence of command and control functions (Table 7.4). Finally, also the tertiary specialization is focused on high growth sectors, which certainly contribute with the right mix effect to the positive performance of these regions. Lastly, these regions have been financed more than the others by regional policies, and have probably been able to use the funds effectively. A separate discussion is worth for FDI. Despite being larger in global regions with respect to the rest of EU regions, they appear not to be a clear discriminant between benefiting global regions and non-benefiting global regions. In fact, though larger for the former, they are not statistically different. Table 7.4 Selected characteristics of benefiting global players in Western countries Variable Megas Structural funds per capita Average annual population growth Location quotient in sector I Transport, storage and communication Location quotient in sector J Financial intermediation Herfindahl index in manufacturing sectors Quota of Legislators and senior government officials Quota of Managers of small enterprises Location quotient of high growing manufacturing sectors Location quotient of high growing service sectors FDI on population

Benefiting 0.875 398479 0.711 1.247 1.114 0.168 0.008 0.039 1.177 1.407 4.949

Non-benefiting 0.563 97440 0.253 1.045 1.612 0.138 0.005 0.027 0.874 1.075 3.767

f 4.98 8.36 15.85 4.26 4.28 3.43 4.47 8.07 3.14 7.49 0.33

sig ** *** *** ** ** * ** *** * ***

Unfortunately, global players in the East are not sufficiently numerous to run an analysis such as this one, especially because almost all of them are benefiting regions. Their characterization, however, will be clearer after reading the section (7.3) on the winning strategies. 7.2.3. Growth patterns and structural features of regional players Also among regional players there appear to be a large number of benefiting regions, whose economic performance is higher than the average, either because of the dynamics of productivity 170

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

(restructuring quadrant) or because of the dynamics of employment (sheltered development quadrant) or, finally, because of both effects at the same time (virtuous cycle quadrant). The characteristics which allow regional players to win are not necessarily the same which allow global players to win. The two types of regions, in fact, are different since only global players hold the structural connections with the world, whereas regional players only have the right sectoral mix. Moreover, the winning factors are also different between Eastern and Western regions, as it will be evident below. The result of the analysis for regional players in Old 15 member countries is reported in Table 7.5, where evidenced are the key characteristics which differentiate the winning regions with respect to the others. It is evident that the dynamics of benefiting regional players in the West is led by the tertiary sector, since manufacturing specialization is low, overall and also in the various technological levels. On the contrary, these regions are specialized in some service sectors, though not traditional ones, such as those linked to tourism (H Hotels and restaurants) and the public sector (L Public administration and defence; compulsory social security). These regions probably win with the ability to innovate in products which are otherwise mature (e.g. agri-tourism, balanced coastal tourism). Overall, these regions are specialized, adding do localization economies, and more assisted than their non-benefiting counterparts by public policies and structural funds in particular.

Table 7.5 Selected characteristics of benefiting regional players in Western countries Variable Benefiting Structural funds per capita 554996 Average annual population growth 0.559 Average annual service employment growth 1.950 Location quotient in sector D Manufacturing 0.880 Location quotient in sector H Hotels and restaurants 2.106 Location quotient in sector L Public administration and defence; compulsory 1.116 Location quotient in High Tech manufacturing sectors 0.607 Location quotient in Medium-High Tech manufacturing sectors 0.849 Location quotient in Medium-Low Tech manufacturing sectors 0.929 Herfindahl index in all sectors 0.089 Quota of managers of small enterprises 0.053 Quota of legislators, senior officials and managers 0.106 Quota of physical, mathematical and engineering science professionals 0.023 Quota (%) of people in EGP-2 professions 21.317 Percentage of people with higher than graduate education 0.830 FDI on population 0.466

Non-benefiting 271260 0.119 0.767 1.196 0.834 1.006 1.076 1.282 1.315 0.080 0.028 0.082 0.029 19.209 0.976 0.837

f 6.15 13.48 14.76 15.4 16 2.84 13.67 12.08 8.72 6.72 31.19 8.85 6.59 8.6 4.61 7.41

sig ** *** *** *** *** * *** *** *** ** *** *** ** *** ** ***

The economy of benefiting regional players is more than average characterized by control functions (legislators, senior officials and managers), and in particular those of SMEs. On the contrary, they have a scarce presence of physical, mathematical and engineering science professionals, since these are not needed for their tertiary productions. This datum is corroborated by the scarce presence of people with post-graduate degrees (Isced 5 and 6) and the higher presence of people with secondlevel qualifications (high share of people in EGP-2 professions). Overall, it appears that, among Western regional players, benefiting are those which renounce to be in direct competition with global players, but rather are finding their development pattern in a balance of intermediate level tertiary functions. This conclusion is highly corroborated by the fact that incoming FDI in benefiting Western regional players are significantly lower than non-benefiting, which are therefore more active than the former in the international competition for the attraction of multinationals.

171

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The last analysis performed is for regional players in Eastern regions (Table 7.6) and here a large number of characteristics emerge which differentiate benefiting from non-benefiting regions. First of all, it appears that a convergence process has been taking place and benefiting regional players in Eastern countries were originally poorer than the rest of Eastern regional players, also because of the lower initial productivity in both services and manufacturing, only partly compensated by a larger agricultural productivity. The benefiting regional players in the east were specialized in Agriculture, hunting and forestry (A) Fishing (B), Manufacturing (D) and Construction (F) and have been able to maintain their specialization in time, as evidenced by the high manufacturing Herfindhal index and the low Lawrence index. Among the service sectors, they are only specialized in Transport, storage and communication (I) and particularly de-specialized in some advanced services ones, namely Financial intermediation (J) and Real estate, renting and business activities (K). Table 7.6 Selected characteristics of benefiting regional players in Eastern countries Variable GDP in PPS Agricultural productivity Industry productivity Service productivity Growth of service employment Location quotient in sector A Agriculture, hunting and forestry B Fishing Location quotient in sector D Manufacturing Location quotient in sector F Construction Location quotient in sector I Transport, storage and communication Location quotient in sector J Financial intermediation Location quotient in sector K Real estate, renting and business activities Location quotient in Medium-Low Tech manufacturing sectors Location quotient in Low Tech manufacturing sectors Herfindahl index in manufacturing sectors Lawrence index in all sectors Quota of legislators, senior officials and managers Quota of physical, mathematical and engineering science professionals Quota of clerks Quota of craft and related trades workers Location quotient of high growing manufacturing sectors Location quotient of high growing service sectors Percentage of people with higher than graduate education FDI on population

Benefiting 13021 6.906 7.776 7.511 0.327 3.458 1.569 1.081 1.238 0.316 0.634 1.768 1.657 0.139 0.151 0.046 0.012 0.054 0.214 1.475 0.965 0.733 0.950

Non-benefiting 17616 2.849 9.688 11.344 1.484 2.007 1.259 0.906 0.952 0.617 0.634 1.296 1.383 0.117 0.216 0.068 0.017 0.087 0.183 1.044 1.080 0.950 0.158

f 2.93 36.92 3.79 8.46 4.02 4.66 10.41 17.29 8.72 52.21 18.31 4.99 3.17 6.07 10.75 7.13 3.69 11.04 6.72 5.53 2.93 5.95 3.8

sig * *** * *** * ** *** *** *** *** *** ** * ** *** ** * *** ** ** * ** *

Interestingly enough, the benefiting regional players in the East are not specialized in the most advanced manufacturing sectors, but rather in the low technology ones, with low command and control functions, low physical, mathematical and engineering science professionals and a low quota of people with post-graduate degrees (Isced 5 and 6). Low are also the clerks, which are basically service workers, compensated by a high percentage of craft and related trades workers. The winning development pattern for regional players in Eastern countries appears hence to be one of sub-contracting manufacturing, with benefiting regions characterized by the presence of highgrowth manufacturing sectors and, in a significantly different way, by FDI from multinationals which have arrived in order to install manufacturing production facilities. All these results suggest that a more in-depth analysis is required in order to highlight the different strategies in front of globalization trends put in place by global regions and see whether some we can highlight some strategic behaviours that can be associated to relatively better aggregate economic performance within global players. This is the subject matter of the next section. 172

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

7.3. Common behaviours among global regions: winning strategies Different pro-active or re-active strategies can be put in place by regions as collective actors in front of globalisation. Our interest here mainly lies in the identification of common behaviours in economic performance of global regions and in the analysis of whether common economic performance can reflect some economic strategies. A cluster analysis - a methodology able to group observations (regions) according to the variables inserted - has been run for all 57 global regions. In our case, economic performance variables have been used to identify clusters of regions according to economic performance variables, namely: - annual average productivity growth in services and in industry, over the period 1999-02; - annual average productivity growth in private services, over the period 2002-05; - annual average employment growth rates in industry and service, over the period 1998-02; - levels of productivity in industry and service in 1998; - annual average regional differential growth (with respect to the nation) over the period 2002-05. The choice of the periods are not casual. A certain time span between regional average annual differential GDP growth and productivity growth or employment growth has been chosen in order to have an imposed causality relationship between the economic phenomena and to avoid endogeneity in the results; for this reason regional GDP growth rates are related to a previous period than the ones of productivity and employment growth, either in industry or in services. The exception to the rule is for private service productivity growth which, for data un-availability, has been measured in the same period as GDP growth. Once obtained, clusters have been characterised also in terms of structural factors. Table 7.6 reports the results. Three main clusters are identified, with the addition of a clear outlier (cluster 3, Inner London). Cluster 1 is made of 14 regions and registers a relatively higher differential regional GDP growth with respect to the average of all global regions. This relatively good performance is accompanied by the highest level of service productivity, of manufacturing productivity, of service and manufacturing employment in the previous period. Also private service productivity growth performs well. This group of regions are characterised by the presence of high value functions and a relatively low level of low value functions, as well as R&D capacity and FDI attraction. Cluster 2 is made of 12 regions, prevalently Eastern regions with the exception of Attica and Lisbon. Their low starting point, which is evident from the observation of data on the level of manufacturing and service productivity, implies the highest productivity growth rates and the highest GDP differential growth rate. These regions present the highest share of low skilled workers and this suggests that they are the destination of western manufacturing firms. This is also strengthened by the highest FDI penetration index and by their specialization in the sector of transportation and communication (Table 7.7). Cluster 3 is a typical outlier, inner London, which registers values decisively over the global regions mean. Instead, cluster 4 is rather interesting, especially when compared with cluster 1. Cluster 4 is the most numerous cluster, made of 30 global regions, all from western countries. In this cluster the less performing regions in terms of regional differential GDP growth show a limited capacity of increasing both manufacturing and private service productivity. Interestingly enough, service productivity growth increases, as well as the number of service employment; this suggests that service employment growth takes place especially in public sectors. Industrial employment growth

173

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

is very high in this cluster, but it is again accompanied by a very low productivity growth in manufacturing; assistance job creation policies and attitudes are a way of explaining this outcome.

Table 7.7. Results of the cluster analysis Variables defining clusters GDP regional differential growth (200205) Services productivity growth (1999-02)

Cluster 1

Cluster 2

Cluster 3

Cluster 4

Mean

0.691

2.739

3.607

-0.519

0.537

1.719

5.489

3.186

1.332

2.335

Manufacturing productivity growth (1999-02)

2.224

7.689

2.073

0.513

2.471

Service productivity (1998)

73.139

18.353

199.438

52.899

53.168

Manufacturing productivity (1998)

75.598

15.649

128.659

58.070

54.683

Service employment growth (1998-02) Manufacturing employment growth (1998-02) Private services productivity growth (2002-05)

2.483

1.719

2.250

2.422

2.286

-0.373

-1.959

-1.829

-1.141

-1.165

2.619

3.091

5.838

1.053

1.950

14

12

1

30

57

Cluster 1 4.539

Cluster 2 4.977

Cluster 3 44.353

Cluster 4 2.855

Mean 4.424

Human resources in S&T (2000)

0.163

0.127

0.151

0.131

0.138

Low skilled workers (2000)

0.239

0.357

0.148

0.305

0.298

High skilled workers (2000)

0.015

0.006

0.007

0.008

0.009

14690.940

9025.880

0.000

2434.571

6648.748

n. obs. Variables describing clusters FDI penetration (1999-2001)

Structural funds expenditures in human capital over the period 1994-99 Total structural funds expenditures over the period 1994-99

148835.100 237861.400 107215.700 135931.500 160256.300

These results lead us to a deeper interpretation of these cluster results (Map 7.1). Cluster 2 can easily be called a cluster of “catching-up regions”, able to achieve higher regional differential growth rates thanks to their high FDI attractiveness, and able to put in place through a rationalization of industrial activities (decrease of manufacturing employment growth and increase in manufacturing productivity growth) in front of a high endowment of low functions. Service employment growth is limited, but linked to increases in productivity.

174

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 7.1. A typology of global regions

Politecnico di Milano - June 2009 Clusters 1 2 3 4 Note: cluster 3 is only composed by Inner London

Cluster 1 can easily be labelled that of “benefiting regions”. These regions carry out two contemporaneous strategies. On the one hand, they increase productivity in the same sectors of specialization through new technologies, organizational and managerial innovation, at the expense of employment (Fig. 7.5a and b). This first strategy protects employment growth because it allows to increase high skilled workers and to maintain the level of low qualified workers through increasing competitiveness with respect to cluster 4. Indeed, these regions register at the end of the period a positive GDP differential growth rate compared with the negative one of other western European regions and the lowest decreasing rate of manufacturing employment. On the other hand, they go for an increase in service employment of high quality, registered in an increase of private service productivity growth (Fig. 7.5c and d). The strategy mix is allowed by structural elements of these regions endowed of the highest share of high skilled workers and human resources in S&T which also helps to attract FDI in advanced services, in particular finance and business services. 175

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 7.5a. Relationship between industrial employment growth and industrial productivity growth Industrial employment  growth (1998‐02)

0

Benef itting -0.5

-1

Industrial productivity growth  (1999‐02)

Conservative 0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

6

6.5

7

7.5

-1.5

Catching-up -2

Fig. 7.5b. Relationship between low level functions and industrial productivity growth 7.48

Industrial productivity  growth  (1999‐02)

Catching-up

6.48

5.48

4.48

3.48 0.23

Low level functions  (low qualified workers) 1999‐01 2.48 Benef itting

1.48

0.48

176

Conservative

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

An opposite picture is provided by cluster 4, made of western regions that can easily be labelled “conservative regions” within global regions. In fact, this cluster mainly contains western regions with a negative GDP differential growth rate. Suffering from deindustrialization, they are not able to reconvert their sectoral specialization and increase manufacturing productivity (Fig. 7.5a and 7.5b). Indeed, they are specialized in mature sectors as chemical products and transport equipment. This situation is not offset by a consistent growth in services, on the contrary they have the lowest productivity growth rate of private services (Fig. 7.5c and 7.5d).

177

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 7.5c. Relationship between service employment growth and private service productivity growth 3.5

Services employment growth  (1998‐02) Catching-up 3

Benef itting 2.5

2

1.5

1.7

Private services productivity   growth (2002‐05)

2.2 1

Conservative

Fig. 7.5d. Relationship between high level functions and private service productivity growth Catching-up 3

High level functions  (Directors and chief  executives) 1999‐01

Benef itting 2.5

0.006

Private services productivity   growth (2002‐05) 2

1.5

Conservative

1

178

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

In a purely inductive way we infer from our analysis that the benefiting regions are those that adopt winning strategies, namely: - increasing manufacturing productivity in the same sectors of specialization. This is achieved by means of new technologies, organizational and managerial innovation, or, in some cases, corporate adaptation, especially vertical integration (with suppliers and customers) and horizontal integration (with similar firms in order to achieve economies of scale). This strategy protects and supports manufacturing employment growth. - reconversion of regions to higher phases of the production process, i.e. decentralizing low-level production phases to areas with lower wages and production costs. This strategy preserves the region’s specialization (especially in terms of value added), generally at the expense of job losses. - reconversion of the regional sectoral structure from low value-added sectors to high valueadded ones, moving to high-level service activities, and avoiding the mere quantitative substitution effect between manufacturing and service jobs. Confirmation of these observations was provided by a more in-depth and qualitative analysis, i.e. by using a case-study methodology.

7.4. An in-depth analysis of winning strategies: a case-study approach 7.4.1. Aim of the qualitative analysis Structural changes are not fully reflected in statistics, as they involve qualitative aspects and changes at sub-sectoral level. For this reason, it is helpful to analyze the recent dynamics and the patterns followed by a number of global regions, in order to understand if these regions confirm the intuitions that are provided on aggregate by the statistical analyses. Within the project, there was room for six case studies, which were devoted to investigate the aspects which are most relevant for the study of the impact of globalization on European regions: Each case study hence investigates the following issues: 1. the economy of the region, with a basic economic description of the region. Global trends affecting the region, in particular the description of which major globalization trends have influenced the region over the past 5 to 10 years and the description of the changes which characterize the increase in FDI and trade; 2. regional response to global trends, i.e. how the regional economy reacted to the globalization trends, in particular the description of the regional impacts and of regional reactions that the globalization changes have generated; 3. the role of the region in the new division of labour, highlighting the structural changes that have occurred at the regional level once the impacts and the reactions have been put in place; 4. the policy aspects, in particular Which European / national / local policies have played, if any, a role in: resisting to global threats; turning threats into growth opportunities; make the region able to capture new opportunities; make the region able to upgrade or maintain its role in the international division of labour; attract or maintain FDI;

179

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

5. the prospects for the region in the next few years, in particular focusing on: which threats stemming from globalization processes are the most serious for the region; which opportunities stemming from globalization processes are up to grab for the region; what effects the economic crisis can have on the region (both short-term and long-term); 6. finally, each national expert has filled a table with the replicable and non replicable aspects, analyzing which aspects are replicable in other regions and which ones are instead specific of the case study. All case studies are included fully in the Case-Study Volume of the Report.

7.4.2. Selection of case studies Selecting the case studies is not an easy task, since it is not easy to know in advance how representative is a case of the general phenomena. Moreover, the cluster analysis has evidenced that also global regions are different the one from the other. Global cities in the meaning of which they are referred in the literature, are too trivial and already amply studied in the literature, for this reason they have been excluded. The selection, to be effective, had to cover the ex-ante regional characteristics in the observable items, hence including Old15 and New12 regions, as well as tertiary and manufacturing global regions. Moreover, we had to cover larger and smaller regions, densely populated and more sparse regions. Our choice started from the available indicators and come to chose the following six global regions as case studies: -

be21 es51 pl12 ro32 se11 ukj1

Prov. Antwerpen Cataluña Mazowieckie Bucuresti - Ilfov Stockholm Berkshire, Bucks and Oxfordshire

As evidenced by map 7.2, these regions have a good geographical balance within Europe, covering Eastern and Western, Northern and Southern countries. As evidenced in Table 7.8, the six regions all experienced an high growth rate in recent years, also through a positive differential shift with respect to their own country in all cases but Cataluña. Interestingly enough, five out of six regions hold a Mega, as most global regions do, but after the case study it turned out that also Berkshire, Bucks and Oxfordshire, which does not host a Mega, has the benefits of being close to London. Total population, the area and the consequent population density vary considerably in our sample, this in order to ensure that we did not pick up only one regional typology (Table 7.8).

180

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 7.2 Selected case-study regions

Stockholm

Berkshire, Bucks and Oxfordsh Mazowieckie

Prov. Antwerpen Bucuresti - Ilfov Cataluna

Table 7.8 main characteristics of selected case studies be21

es51

pl12

ro32

se11

ukj1

Prov. Antwerpen 2.50

Cataluña

Mazowieckie

Stockholm

3.13

5.41

Bucuresti Ilfov 6.90

3.93

Berkshire, Bucks and Oxfordshire 3.19

Growth differential with respect to the country 2002-05 Presence of Megas

0.62

-0.19

1.14

0.96

0.82

0.57

Yes, Antwerp

Yes, Barcelona

Yes, Warsaw

Yes, Bucharest

Yes, Stockholm

Population in 2000

1,644,800

6,251,200

5,111,400

2,279,300

1,813,300

No, but bordering with the Greater London 2,088,500

Area

2,867

31,930

35,598

1,821

6,490

5,741

Population density in 2000

574

191

143

1,220

266

364

Growth rate 2002-05

181

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The sectoral specialization is another characteristics which can be investigated in advance, using the IGEAT matrix and calculating the location quotients with respect to the EU mean (Table 7.9). Since global regions are not all tertiary but some of them are still highly manufacturing regions, we picked up regions whith these characteristics. For instance, Mazowieckie, Bucuresti - Ilfov, Stockholm and Berkshire, Bucks and Oxfordshire are all regions specialized in service open growing sectors, whereas for Prov. Antwerpen and Cataluña the specialization is stronger in manufacturing open growing sectors, even if the service location quotient is close to 1. Ex-ante, we chose global regions with different specialization also within the macro-sectors. For instance (Table 7.9), among manufacturing open growing sectors, there is a very high specialization by Prov. Antwerpen in manufacture of chemicals, chemical products and man-made fibres; the manufacture of rubber and plastic products is vert important in Cataluña but almost absent from Stockholm.

Table 7.9 Location quotients in selected sectors (1 = EU mean)

QL in open growing sectors in 2004 QL in manufacturing open growing sectors in 2004 QL in service open growing sectors in 2004

be21 Prov. Antwerpen

es51 Cataluña

pl12 Mazowieckie

ro32 Bucuresti Ilfov

se11 Stockholm

1.09

0.96

1.08

0.92

1.13

ukj1 Berkshire, Bucks and Oxfordshire 1.16

2.44

1.35

0.69

0.63

0.75

0.68

0.96

0.92

1.11

0.95

1.17

1.20

Manufacturing Open Growing Sectors Location Quotients in 2004: DD, manufacture of wood and wood products; DF-DG, manufacture of chemicals. Chemical products and man-made fibres; DH, manufacture of rubber and plastic products; DM, manufacture of transport equipment; DN, manufacturing not elsewhere classified.

0.27

0.48

0.57

1.26

0.40

0.39

5.06

1.74

1.03

0.59

1.40

0.93

1.16

1.44

0.68

1.19

0.13

0.55

1.03

1.17

0.34

0.46

0.57

0.49

0.99

1.03

0.63

0.17

0.19

0.80

Services Open Growing Sectors Location Quotients in 2004: G, trade (retail and wholesale); H, hotels and restaurants; I, transportation and communications; J, finance and insurance; K, business services; MNOP personal, cultural and recreational services.

1.22

1.00

1.65

1.23

1.00

1.15

0.44 1.26

2.53 1.04

0.37 1.71

0.92 2.11

0.59 1.28

0.95 1.09

0.81 0.98 0.78

0.86 0.75 0.79

1.49 0.86 0.81

1.12 0.59 0.66

1.74 1.18 1.12

0.78 1.59 0.94

The transport equipment sector is present in the two manufacturing case-studies, Prov. Antwerpen and Cataluña, whereas it is marginal in the tertiary ones.

182

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Among service sectors, trade, retail and wholesale is a sector of specialization for all six case studies, although with different strengths. On the contrary, only Cataluña, is a region specialized in hotels and restaurants, due to its position and touristic strength. It is very interesting to observe that transport and communications (a sector which include most of ICTs), is a sector of specialization for all our selected case studies. Finance and insurance is a sector of specialization for the four tertiary case studies, whereas business services only characterize the two case-studies which (ex-post) turned out to the most innovative-prone. Finally, only in Stockholm there exist a specialization in personal, cultural and recreational services, as is evidenced by the relative case-study. It is interesting to observe that all specialization of Table 7.9 are coincident with what the National Experts report in their quali-quantitative case studies. Ex-post, it has also been possible to assess the regional characteristics on a further dimension, which was impossible to measure in a statistical qualitative analysis, i.e. the specialization of the region in high-tech or low technology activities. Moreover, the manufacturing or tertiary specialization of regions was qualitatively confirmed by the case-studies. Ex-post, it is hence possible to locate the six case-study regions in a two-way quadrant, depending on their specialization in service or manufacturing activities and on their specialization in high-tech or low-tech (Fig. 7.6). It turns out that two regions are specialized in high-tech tertiary sectors, namely Stockholm (which is ranked number one in Europe in innovation performance by the European Trend Chart on Innovation) and Berkshire, Bucks and Oxfordshire, which is specialized in the creation, transmission and commercialization of knowledge. Fig. 7.6 Ex-post characteristics of case studies Specialization in high-tech

Berkshire, Bucks and Oxfordshire

Stockholm

Cataluña

Specialization in manufacturing

Specialization in services

Prov. Antwerpen

Bucuresti Ilfov

Specialization in low-tech

Mazowieckie

The two Eastern regional case studies are also specialized in similar sectors, with a consistent specialization in ICTs. However, the level of activities performed in these advanced services activities is of a lower level, and often relying on off-shoring from Western firms which find 183

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

abundance of educated people at a lower price; for this reason, we chose to classify them among the tertiary lower technology regions. Finally, the two Western regions which are specialized in manufacturing. None of them is a leader in innovation as their Swedish and British counterparts. Apparently, Cataluña is the one which is more specialized in high-tech, whereas Prov. Antwerpen is still more reliant on its port and on activities with a lower technological content. This might also explain why, in the case studies, it is the latter of the two which is reported to have had more troubles in recent years. Since all six case studies are included in volume 2 of the final report, the analysis that follows focuses on the main messages which can be drawn from the analysis and which can be considered as relevant in general.

7.5. Major lessons from the case studies 7.5.1 General remarks The six case studies confirmed that these regions are those who compete in the global arena. However, the national context remains a major aspect of regional competitiveness, which influences what takes place in our regions, even if all of them are leaders within their countries. For example, the two Eastern regions of Mazowieckie and Bucuresti - Ilfov have taken advantage in respect to other Eastern European competitors, from the size of their countries, since Poland and Romania are considerably larger and more populated with respect to others in the area. Another aspect which has been evidenced is that it is difficult to distinguish what takes place because of globalization and what takes place because of other related processes which are taking place at the same time, processes such as the European integration or (for CEECs) transition and enlargement. In particular, it appears that transition is as much important as globalization for our Eastern regional case studies. A third general aspect which has to be emphasized from the beginning is that global regions take advantage from global forces especially thanks to local forces, since many factors which allow them to be winners in the global competition are the same which allow them to be winners in competition with respect to their neighbouring regions. What follows, in particular, will evidence what strategies have followed our regions to be successful, and which strategies, if pursued, would have allowed them to be more successful. In particular, the results tend to confirm those obtained in the previous sections of the report.

7.5.2 In order to benefit from globalization, regions have to ... ... take advantage from their settlement structure, ... Our case studies differ considerably as far as their territorial and settlement structure, though they are all global regions. For example, Cataluña and Mazowieckie are very wide regions, with strong territorial differences inside, and different economic specializations between the different sub-areas.

184

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Bucuresti - Ilfov is a region which is an enclave within another region, with no direct access to the sea, whereas sea accessibility is a major aspect of the spatial and economic situation of Prov. Antwerpen. The six case studies share the presence (and the positive growth effects) of a Mega city either in the region, or, in the case of Berkshire, Bucks and Oxfordshire, which borders with the Greater London, nearby. Having a second Mega too close does not appear to be a major positive characteristics, since in the only case where it happens, in Prov. Antwerpen, there is strong complementarity of which the Mega of higher level appear to benefit more. The second settlement related positive feature for our case studies is the role of capital city: for all three cases in which the region holds the capital of the country, it attracts the highest level and command functions. This is especially evidenced by the cases of Mazowieckie and Bucuresti - Ilfov but also, in negative sense, by Prov. Antwerpen, which suffers from being the second pole of the country and very close to the capital, since activities such as finance are attracted by the first level city, i.e. Brussels; also Cataluña does not have the command and control functions of the country, but it is in any case successful in attracting research and development, maybe because of strong autonomy and distance from Madrid. The capital city also acts as gateway for the incoming FDI in the country. After the transition to a market economy, there has been a strong concentration of growth and FDI in Eastern capital cities (such as Warsaw in Mazowieckie, and Bucharest, in Bucuresti - Ilfov) a phase which is now followed, at least in Poland, by some signs of de-concentration. This has led to strong national disparities, due to the very high dynamism in capital and Megas cities, with the rest of CEECs experiencing much slower development. The ties between the main city and a dynamic and/or complementary interland are also important for our case study regions, since they often allow to decrease the pressure on land and resources in the core city. For instance, the development of Cataluña has also involved the extension of growth to other provinces beyond Barcelona. Also in the other cases there is evidence of links with the interland, but only within a certain range; for instance, in the case of Mazowieckie, the links between Warsaw and the most peripheral provinces of Mazowieckie are weak, also due to poor infrastructure, but they are not necessary for growth, since longer range ties are more important. In some cases, (such as the city of Bucharest which is within a much less extended region) there has been a re-location to the interland (in this case Iflow) of low value functions, with high level functions such as R&D remaining in the core city.

... shift towards services, ... As expected from what seen in chapters 4 and 6, the shift towards services has been a characteristic of the economic growth of our case studies; it has taken place very strongly in Berkshire, Bucks and Oxfordshire, in Mazowieckie, in Prov. Antwerpen, in Stockholm, in Bucuresti - Ilfov. The shift towards services is not only a characteristic of local firms for our case studies, but also concerns FDI; this is very evident in the case of Stockholm, which has shifted away not only from manufacturing but also from public services. The shift towards service FDI inflows is evident both in Western regions (such as Berkshire, Bucks and Oxfordshire) where these processes started earlier, than in successful Eastern regions which entered the global FDI market only after their transition (Bucuresti - Ilfov and Mazowieckie). Not 185

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

surprisingly, these service FDI investments are different in the various regions, for example, in Mazowieckie FDI in services are often service centres for clients in other countries. The shift towards services in our global case-study regions is also due to externalization of industry functions towards other regions at a lower hierarchical level. De-industrialization has led in many cases to an increase of industry productivity which is higher than the one of services, through however much less manufacturing jobs. In other cases, especially Cataluña, de-industrialization appears not to have taken place, but this has happened also thanks to a shift towards higher-value manufacturing products. It is in fact evident from our regions that to retain manufacturing, global regions need to go upwards in the value chain. One clear example in this case is the chemical-related sector in Prov. Antwerpen; this sector is thriving with respect to others because it has been able to shift to softer activities and productions, such as the pharmaceuticals. The process is shared in the CEECs, for instance also Bucuresti - Ilfov has shifted its industry towards more knowledge intensive activities. Also in Cataluña there has been a transition towards a more knowledge-intensive and creative economy, but in this case with more jobs and not more productivity, due to Spanish development model of the last years. Also in services global regions need to move up in the value chain, since more and more service activities can be externalized. One example particularly successful is the one of Stockholm, which has specialized in knowledge intensive business services, leaving other lower value added services.

... shift from low-level to command and control functions, ... Confirming what observed in chapters 4 and 6, in our global case-study regions, there has been an increase in the number of corporate managers, teaching professionals, research professionals and business professionals; this is especially evident in the case of Berkshire, Bucks and Oxfordshire. Command functions are particularly strong also in the case of the Stockholm region, one of the most successful in upgrading its economy. Paradoxically, the importance of command and control functions is also evidenced by the case of Prov. Antwerpen. The fact that they are weak in this region (because they tend to concentrate in Brussels, which is very close) is reported to be one possible bottleneck which have impeded higher growth. At the same time, the most successful Western European global regions have been off-shoring low functions in order to save costs, as in the case of Berkshire, Bucks and Oxfordshire, which has offshored lower skilled administrative functions, and of Stockholm which has been re-locating lowest level services and keeping the highest ones. Interestingly, Stockholm is now also off-shoring R&D, after the first phase in which only production was off-shored For Eastern global regions there is a quite different situation, since they have been successful also thanks to in-shoring of services from Western regions (this is evidenced in the cases of Bucuresti Ilfov and also of Mazowieckie, where apparently there is not still any risk of off-shoring to other countries. Attracting and keeping the headquarters of multinational is important for global regions, both in the West of Europe (case of Berkshire, Bucks and Oxfordshire) and in the East (case of Bucuresti Ilfov). It has to be observed that having the multinational headquarters is important but it is similarly important to house the headquarters of national firms, as shown by the case of

186

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Mazowieckie. This is a signal of the fact that global and local forces act together and are generally difficult to distinguish in practice. As far as multinationals are concerned, there is not the necessity for European global regions to have the world headquarters of these firms, but having the headquarters for the area and the surrounding countries provides almost the same benefits, as shown by the case of Bucuresti - Ilfov. This leads to the consideration that, in order to have the command and control functions in the regions, it is not very important the absolute position, but the relative one: for instance, its casestudy has evidenced that Stockholm is not a global city, but it is the centre for Northern European countries (including Sweden, Norway, Denmark and Finland) and this gives it a more important role than it could have by itself. Also Bucuresti - Ilfov is shown to be not an absolute winner but a relative winner with respect to surroundings, often playing command and control functions for the Balkan area and other neighbouring countries such as Moldova and Bielarus.

... shift towards the knowledge economy, ... The shift towards the knowledge economy has probably been the most successful strategy for European global regions. For example, the case study evidences that the main specialization of Berkshire, Bucks and Oxfordshire is now the creation, transmission and commercialization of knowledge. For this reason, extremely important is the presence in the region of skilled labour. For example, there is abundance of skilled medical labour and software engineers in Berkshire, Bucks and Oxfordshire, but also in Bucuresti - Ilfov there is a very large number of software graduate for 1000 inhabitant, allowing it to specialize in activities related to the knowledge economy. Also in Stockholm there is strong evidence of a comparatively better educated labour force, with more graduates than the rest of the country (even if Sweden has overall good instruction levels). A counterfactual example is Prov. Antwerpen, which suffers from relatively lower qualifications than Brussels. It has to be noticed that the presence of skilled labour is not a sufficient condition for the region to benefit from it but it is important to have in the region the economic activities able to exploit this potential. For example, Cataluña has plenty of skilled labour but its labour demand is not strong enough for the higher qualifications and the region is hence unable to fully take advantage of its skilled labour force. Among the skills required by the knowledge economy in a global world, our case studies have evidenced the importance of English knowledge. English speaking in Berkshire, Bucks and Oxfordshire, allows the presence of the European headquarters of multinationals and advanced services; in Mazowieckie and Bucuresti - Ilfov, the high level of English as a foreign language, allows the presence of activities which provide services to other countries. Cataluña, on the contrary, suffers from its relatively low-level of English proficiency. However, this region has strong links with Latin America because sharing the same language and the same culture is strongly beneficial to international economic relations. In the knowledge economy, our case studies have shown the importance of regional universities for the two complementary aspects of (i) providing qualified personnel and human capital and (ii) providing high-level research to local firms. The most evident example is the presence of a number of Universities and in particular of the Oxford ones in Berkshire, Bucks and Oxfordshire, where they are at the base of the knowledge economy specialization of the region. Also in cases in which 187

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

there is not the presence of top universities at world scale, as in the cases of Mazowieckie and Bucuresti - Ilfov, it is very important the presence in the region of the most important of the country. Prov. Antwerpen still suffers from not being an important university pole, despite recent progresses in its education system. The Stockholm case signals the fact that research universities are important (there are some of high level) but also helpful and important for the knowledge economy is the presence of other public and private research institutes. This may be one major cause for the measured higher returns from R&D investments in Stockholm with respect to elsewhere. Very important are for all our global regions the in-flows of non-financial forms of capital (people, skills, knowledge and technologies), through in-migration and international links in research. As far as the inward flows of human capital are concerned, an essential feature of global regions is hence the attraction of the most skilled individuals. For instance, Berkshire, Bucks and Oxfordshire, as London, acts a gateway to the whole United Kingdom for foreigners. In other cases, as for Mazowieckie, the attraction is mostly from rest of country rather than international, but those who move are the most skilled. Beyond purely economic factors, also factors related to the quality of life are important in the attraction of foreign skilled individuals; for example, Cataluña has been able to attract these persons thanks to its environment, climate and way of living, with favourable impact in its economy. Finally, our case studies have also evidenced the helpfulness to give to the region (and the main city in particular) a new image and market it as an innovative and cool one, also by entering into new sectors. One clear example is the fashion industry in Antwerp, which has been able to give a new and attractive image to the whole city; its development has been possible through the coordinated action of three poles: the creators, the education system and the public authorities.

... hold a high level of physical connections with the rest of the world, ... All our case studies have evidenced the important of long range connections for global regions. We remember that global regions are also defined in terms of long-range flight accessibility but it is reassuring to know that global regions do benefit indeed from this characteristics. Especially important is airport connectivity: Berkshire, Bucks and Oxfordshire take a strong advantage from the presence very close to its border of Heathrow, the largest European airport61; other case studies (Mazowieckie, Stockholm, Bucuresti - Ilfov) detect the presence of the most important national airport, but support the fact that, would it be bigger and with more connections, the region would take an important benefit. On a lower level with respect to airports, the case studies also evidence the benefits of other connectivity infrastructure, as a port, which was at the basis of industrial development because of assembly industry in Prov. Antwerpen, and other connectivity infrastructure more related to the knowledge economy (case of Berkshire, Bucks and Oxfordshire)

61

Heathrow airport is located in the Greater London. 188

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

... have an economic structure which supports them in the markets, ... The case studies have evidenced a number of market-related factors as important for the competitiveness of global regions. Among them, their diversified economic structure, evidenced as a positive factor in the cases of Stockholm and Mazowieckie. At the same time, the case studies also evidenced that advantages also come from the presence in the region of clusters, which can allow to reach a critical mass in some sectors. Among them, it is possible to cite the clusters of R&D, ICTs, financial services and media and amenities in Stockholm; the traditional chemicals clusters and the new fashion cluster in Prov. Antwerpen, the information technology cluster in Bucuresti - Ilfov. These clusters are allowed by the presence in the regions of some specific qualifications and competences, but at the same time they allow to exploit and re-produce these skill excellences. For example, in Berkshire, Bucks and Oxfordshire, the automotive engineering and motor sport62 cluster is allowed by competences in precision engineering and the ability to access international flows of skilled engineers and technicians, and the biotechnology and pharmaceuticals clusters is allowed by the skills and competences of universities such as Oxford. Among the factors related to the economic structure, the sectoral profile, which can allow benefits either from a traditional shift-share analysis (case of Prov. Antwerpen), or from the possibility to attract and retain activities of higher level with respect to the area where the region is located, as evidenced in the case of Bucuresti - Ilfov, whose sectoral profile is more similar to EU than to the one of the rest of country. Also important is the presence of small and medium enterprises (SMEs) for global regions, though they appear not to be able to make the region thrive by themselves. The supporting layer of selfemployment and SMEs is a positive aspect in the case of Mazowieckie, where however their presence is made positive by the fact that the region also holds a sufficient range of large firms. SMEs are also play a role from as global gateways, and don’t leave this important function only to large firms. For example, they generate a significant part of export from Stockholm. Also, SMEs are very diffused and important in the case of Bucuresti - Ilfov, both on the production side but also in terms of incoming FDI which often come from international small firms. Finally, local SMEs help regions take advantage from globalization forces inasmuch they are able to capture outsourcing from multinationals (evident especially in the case of Bucuresti - Ilfov). The case of Stockholm is the one which most emphasizes the importance of soft aspects such as the “business milieu” and business climate, which allow regional firms to thrive with respect to other counterparts. Another aspects related to entrepreneurship is the presence of start-ups and firm and university spinoffs, which are made easier by the presence of early-stage venture capital. These are very evident in the case of Berkshire, Bucks and Oxfordshire, where financial sources are made available by the nearby presence of London more than by indigenous forces. Science parks and incubators are another important feature of successful global regions, especially those in the specialized in the highest technology activities (case of Stockholm).

62

This cluster accounts for as much as 75% of world output of single seat racing cars. 189

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

... be able to attract and retain FDI, ... Our case-studies reinforce the idea, evidenced in chapter 5, that global regions act as gateways to areas much broader than the regions themselves, and this function plays an important role for them in economic terms, since it gives them the possibility to attract activities also directed to a broader area; for example, Berkshire, Bucks and Oxfordshire, as the South East of Britain in general, acts as a gateway to Europe for extra-European multinationals, whereas Bucuresti - Ilfov acts as gateway not only to Romania but to the whole Balkans and Stockholm acts as gateway to the four Northern European countries. FDI in global regions are not necessarily green-field, on the contrary, especially in the case of advanced Western regions, they are often acquisitions (case of Stockholm). Among the effects of FDI reported in our case-studies, the fact that they increase local expenditure, and also that they allow the development of a local induct. At a smaller, urban, scale, FDI appear to be a very relevant source of demand for office space, with consequences on the urban real-estate market (case of Mazowieckie). The effects of FDI, however, are not only positive, since they can induce dependence from other countries, and this is especially evident in the automotive sector with the manufacturing plants of Opel in Prov. Antwerpen, and Volkswagen and Renault-Nissan in Cataluña, though these drawbacks are not calling for a change in European FDI policies. It is interesting to observe that foreign enterprises are major exporters but also often net importers (i.e. they generate more imports than exports). This is confirmed by the cases of Mazowieckie and Bucuresti - Ilfov, but also in Stockholm the level of exports is low, and this because this region is specialized in command and control functions and headquarters do not generate exports. It hence appears that export seeking is not the best strategy for advanced global regions such as the ones of our case studies. For example, import high-tech firms contributes to innovation in the case of Stockholm and, in the case of Cataluña where the region has successfully followed an exportoriented strategy, this has also drawbacks since: i) an important share of trade still based on cost differentiation and this type of competition faces the emergence of other cheap producers (in the UE or abroad) as well as the higher differential inflation of the Catalan and Spanish economy; ii) in global terms, productivity has not increased from 1995 and an important share of the exports are concentrated on weak demand products. What are the factors which allow global regions to attract FDI in addition to the gateway functions and to their role in the international division of labour? The first reply lies in the presence of a rich and educated labour market, cited as an attraction factor both in the East than in the West (cases of Mazowieckie, Berkshire, Bucks and Oxfordshire, Bucuresti - Ilfov). Also, especially for what concerns CEECs, important are: the presence of capital functions; the size of the national market; the diversified economic structure; the comparatively low labour cost; the European law system and culture and the stability due to the participation to the EU. Our case studies also confirm that infrastructure endowment still plays a role, for instance the case of Bucuresti - Ilfov reports infrastructure better than rest of country and the case of Prov. Antwerpen reports the importance of the port as an historical attraction factors of many FDI.

190

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

... be able to implement the good supporting policies, ... Good supporting policies are reported to be very helpful for the global competitiveness of our casestudy regions. For instance, almost all case studies report the necessity for the regions to receive investments in connectivity infrastructure, but also reported are the benefits of policies, either public or private in their origin, of support of firm entrepreneurship and innovation, with business parks, science parks and incubators. For regions generally active in the knowledge economy, very important are policies towards higher education, allowing them to maintain their human capital at a level higher than their competitors. Then, it is not really important if the origin of the policy is local or if it is a national policy which has stronger than average impact on the region, either because it is the capital region or because it holds important pieces of the national university system (as is the case of Berkshire, Bucks and Oxfordshire, where the U.K. university policies have strong impact because of Oxford and many other university poles). Also the global regions which were lagging in this respects, are trying to catch up, such as Prov. Antwerpen which now has a full university. Among the policies at national level which can positively impact on global regions, it is interesting to highlight the U.K. openness to in-migration, which allowed Berkshire, Bucks and Oxfordshire to be an attractor for international high-level human capital. Other helpful policies have been implemented at metropolitan level, such as the strategic projects of urban renewal (22@bcn, forum2004, Sagrera) which worked well in Barcelona and benefited Cataluña. Sectoral, R&D FDI attraction policies have also been helpful, as has been and the policies towards high-value-added and specific sectors implemented in Cataluña. FDI attraction, though, does not appear to be necessarily linked to policies. For cases (such as Cataluña), which did active policies to attract them, there are others (e.g. Mazowieckie) which attracted them without any policy devoted to it.

... and be able to avoid the major threats. Among the threats which are reported to be present for global regions, an important one is the intrinsic risk associated with a low-price competitiveness. The case of Cataluña is emblematic in this respect; in this region large amounts of jobs in low productivity activities have been created with consequent very limited growth of productivity. This risk is similar to the cases of Eastern regions, where the wage increases due to the success reduce their low-cost advantage, with the consequence that they will soon need to shift paradigm in order to maintain the very high development rates of the past. The openness which is an intrinsic characteristics of global regions makes them more vulnerable to global shocks, a feature which is important especially in a period of global downturn. Another threat for global regions, especially the western ones, concerns off-shoring. In fact, at present, the support activities have often been off-shored, but the headquarters have not; the risk is that, in the future, also headquarters functions, or at least some of them, will be off-shored. And will also research and development activities be off-shored? This trend has already started in Stockholm, which is among the most advanced of the EU for R&D.

191

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

At a metropolitan scale, the major reported threats come from congestion, the saturation of spaces and the consequent increase in land prices, the present lack of enough infrastructure and transport accessibility and the delays in the construction new ones. These are reported in all six case studies, and are made more difficult by the fact that the decision body is often at a level above the regional one, and, in the case of Prov. Antwerpen, the improvement of accessibility infrastructure would also need the involvement of foreign countries nearby. Another threat at urban level is the urban sprawl coming with sustained economic growth, with consequent backwash effects and high house prices, especially evident in the case of Mazowieckie, due also to lack of planning. Policies are reported to be at the wrong scale in many cases. For example, most U.K. policies relevant for Berkshire, Bucks and Oxfordshire are at the level of South-East, but would need to me more tailored on the smaller region. Stockholm, has regional land-use and transportation planning, but few measures to implement the strategy. In particular, it appears that Stockholm is fragmented in many independent municipalities. One case in which the region has been successful in implementing regional strategies with involvement of social, political and economic agents is Cataluña, and this is not surprising knowing the large autonomy of the region so that the target and the decisional level coincide. The need for good policies is reported in many cases, especially for what concerns the strategic focus of these policies. The case-study of Mazowieckie laments the fact that policies have been reactive and not proactive. Also the Stockholm case-study reports the lack of a strategic forum. Again Mazowieckie, reports the lack of policies to diffuse FDI, so that they still concentrate in Warsaw, congesting it and not helping the take-off of other more peripheral areas The coordination of policy levels is also important in this respect, since for example the Bucuresti Ilfov case study assesses that Structural Funds Expenditure is successful only if supported by an effective and rational local policy. Finally, the nature of policies is also a concern for our global regions if they have to be successful in the knowledge economy. In fact, the most knowledge oriented of them, Stockholm, laments the fact that Swedish national policies are still mostly relying on the old manufacturing paradigm, whereas the region only has weak metropolitan policies which would need to focus on the necessity of more agglomeration economies (the region is not dense enough) and the support to the creation of milieu effects.

7.6. Conclusions Global regions demonstrate a higher economic performance on an average, with respect to other regions. This is the result of their higher ability to adjust to globalization trends, being endowed of the structural preconditions, in terms of functions, human and social capital, that are highlighted in the literature to be the main success factors in front of globalization. This means that global players, which are only defined in terms of structure, are also those regions which dynamically lead and drive economic growth processes within their countries, especially but not only in Eastern Europe. Global forces hence appear to pull growth within countries. In this chapter we analysed which structural features characterise the most outperforming regions, both in global and in regional players. In the descriptive analysis, the economy of global regions appears to be more dynamic and structurally stronger. As far as the dynamics is concerned, these 192

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

regions have at the same time a high service employment growth rate and a high increase in service productivity; these features demonstrate that these regions are the ones where service activity increases are virtuous and create high value added jobs rather than lowly qualified jobs. This is probably made easier by the fact that these regions have the highest level of command and control functions, as well as the highest increase of them. On an average, global regions are the ones gaining the most from globalization; Eastern countries especially face increasing competition through the strengthening of their global players, while regional players, and other regions, still find difficult to cope with fearer competition. This is different in the Western countries where a high number of regional players are able to show a relative higher performance than the EU average. Regional players have specific structural features that determine their economic performance which are different from the ones of global players. For western regional players the winning strategy is not to compete with global players by specializing in second-order services and functions, while for the Eastern regional players, the success factors reside in a strong and dynamic manufacturing activity linked to the presence of FDI. In a case-study analysis, we show that the structural features which are more associated with the positive regional behaviour of globalized regions are the ones we expected: the presence of command and control functions, of dynamic sectors at EU level, the ability to insert in the knowledge economy and the attractiveness of FDI are important success factors for global players, whose relative performance is strictly linked to the existence of these elements. It is interesting to see whether in a prospective analysis the role of global players as the leading areas in a period of globalization will last or, on the contrary, they will no longer play this important role. This is the subject matter of the next chapter.

193

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

8. A Prospective Analysis 8.1. Setting the scene: main driving forces for a prospective analysis This part is devoted to the development of what the future will look like under different assumptions on the way globalisation patterns will develop. The main tool to build the prospective analysis is through the use of the MASST model, updated and extended to the sectoral breakdown during the project on “vulnerable regions”63. The MASST (Macroeconomic, Sectoral, Social, Territorial) model is a regional econometric forecasting model with the aim to provide insights on what behavioural trends in GDP, industrial and service employment will look like under different assumptions of the globalisation patterns that can emerge. The most recent version of the MASST model provides the opportunity to apply this forecasting model for a more in-depth inspection of the multiple trade-offs that globalisation patterns bring along at the sectoral level. As mentioned in Chapter 1, the new globalisation trends impact differently on regions according to their sectoral specialisation, to their functional specialisation within the sector, to their labour force composition, all aspects unavailable in the first version of the model. The prospective analysis presented here is based on alternative competitive strategies that three main blocks of countries will pursue over the next ten to fifteen years, up to 2020 (the last data are from 2006): the EU Member Countries and the emerging countries, taking into consideration the European Commission strategies as a third discriminating element. The interplay of these strategies will depict different scenarios with which European regional economies will have to deal with. Two scenarios will be built taking the two above mentioned possibilities into consideration, by developing forward looking analyses for European regions under the assumptions of opposite strategies that the emerging countries can put in place: a more risky, and probably more expansionary strategy based on the willingness to compete on a world scale by undertaking strong and decisive internal restructuring, reconverting and modernising processes; a more protective strategy whereby BRIC reinforce present tendencies (Fig. 8.1). A comparison on future opportunities and risks of growth for European regions will be investigated.

63

On the MASST model see Capello (2007a), Capello et al., (2008), Capello and Fratesi (2008, 2009). 194

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 8.1. Future alternatives for global competition trajectories    Type of globalization process   

PRICE COMPETITIVE BRICs Competitiveness strategy of BRICs strongly oriented to the control of production costs. The present trend is reinforced. Focus on low price low quality products. Low wages and consequent low purchasing power of BRICs consumers. MODERNIZING BRICs Significant modernization of the economies of the BRIC countries. Global customized production and competition based on quality. Significant increase in wages resulting in an increase of purchasing power of BRICs consumers.

Box 8.1. The MASST Model MASST is a macro-econometric regional growth model, whose outcome is mainly GDP growth rates, industrial and service employment growth rates and their spatial distribution. The MASST model is a combination of two different and interactive parts: a pure macroeconomic regional growth model estimated on past-growth; and a simulation algorithm for inspection of the future. The linkage between national factors and regional ones concerning growth is assured by the structure of the model, which interprets regional growth as resulting from a national growth component and a differential regional growth component:

 Yr   Y N  s ; r  N where  Y r and  Y N denote the GDP growth rate respectively of the region and the nation, and s represents the regional differential growth with respect to the nation. National growth depends on the dynamics of the macroeconomic national elements: private consumption growth, private investment growth, public expenditure growth and export and import growth. In its turn, the regional differential component (the shift component, i.e. the relative regional growth) depends on the competitiveness of the local system, this being based on the efficiency of local resources: the increase in the quality and quantity of production factors (like human capital and population) in infrastructure endowment, in energy resources, as well as the sectoral and territorial structure of the regions and the interregional spatial linkages (For details on the MASST model, see Annex 8.1 of this report).

195

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 8.2. Future alternative trajectories for European member States strategies A DEFENSIVE STRATEGY -

protectionism of European economies; attraction of FDI for New Member States countries; international competition on production costs; protectionism especially in vulnerable sectors.

Competitive strategies of Member States countries

A PROACTIVE STRATEGY -

open trade; increased productivity in traditional sectors; customised production and quality competition; increased competition in new sectors, at present influenced in a limited way by globalisation.

Lastly, we assume two possible trends for what concerns the European Commission policies, that can either follow an excellence based policy or a cohesive policy (Fig. 8.3).

Fig. 8.3. Future alternative trajectories for European Commission strategies AN EXCELLENCE BASED COMPETITIVE POLICY European Commission strategies

-

Rigidity in the accompliance of the Lisbon agenda objectives; Infrastructure projects selected on the basis of profitability aims; Structural funds to all regions; 20% budget more than the 2007-2013.

A COHESIVE POLICY -

Flexibility in pursuing the Lisbon agenda objectives; Infrastructure projects selected on the basis of a rebalancing of territorial infrastructure endowment; 30% budget less than 2007-2013 Structural funds only to convergence regions.

A fourth element which in this period cannot be left aside in a prospective scenario is the economic crisis that since August 2008 is heavily affecting the economic scene. It is impossible to be able to say when and how this negative cycle will end. What can be done is to make scenarios under different assumptions on the speed of recovery from the present cyclical negative trends which affects the world economy. In particular, our prospective analysis is built under two main assumptions (Fig. 8.4): -

a situation in which the crisis finds the solution in the short term (up to five years). The demand of consumption goods, like electronic instruments, textile and clothing goods, drops 196

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

-

in the short run, and the negative effects propagate especially in the sectors that produce these goods. Consumption of durable goods like cars recovers in a long time, and therefore sectors affected by durable goods is less affected than in a long term crisis; a situation in which the crisis finds a solution in the long run, affecting the demand also of durable goods and all related.

Fig. 8.4. Assumptions on the general economic setting

CLOSE RECOVERY FROM ECONOMIC CRISIS

LENGTH OF THE ACTUAL ECONOMIC CRISIS DELAYED RECOVERY FROM ECONOMIC CRISIS

The analysis assumes a game-theoretic structure when the different and opposite strategies of the two blocks of countries and of the European Commission are taken into account. Fig. 8.5 presents all possible combinations of the two strategies for each block of Countries which give rise to eight possible scenarios (Fig. 8.5). To the eight possible scenarios we add the fourth dimension, which is expressed in terms of coloured setting of our cube picture, and represents the economic setting we assume. The following two opposite are of especial interest (Fig. 8.5): -

-

a scenario combining a reactive strategy by the EU 27, a modernising strategy by BRIC, and an excellence-based EU policy developed in a general economic setting of a short-term recovery from economic crisis (expressed in Fig 8.4 by a light colour for the setting). In this scenario the recovery from the crisis in the short term obtained through efficient Keynesian macroeconomic policies makes it possible to move to excellence-based policies thanks to a new wave of private investments that takes place after the end of the crisis (Fig. 8.5a); a scenario based on opposite strategies – a defensive EU27, a price-competitive strategy by BRICs, and a cohesive policy of the European Commission – developed in a general economic setting of recession and deep long-term crisis (expressed in Fig 8.4 by a dark colour), in which the consumption of durable goods is strongly depressed. Macroeconomic policies remain Keynesian throughout the scenario period; structural policies are oriented to cohesion aims in order to limit the effects of the crisis (Fig. 8.5b).

197

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 8.5. Proposed scenarios a) An aggressive Europe in a high-quality competitive world and in a fast recovery from economic crisis (scenario A)

An excellence based EU competitive policy

Modernizing BRIC

A cohesive EU policy

An excellence based EU competitive policy

Price-competitive BRIC

Pro-active EU 27 countries

Defensive EU 27 countries

b) A defensive Europe in a price-competitive world and in a slow recovery from economic crisis (scenario B)

Pro-active EU 27 countries

Defensive EU 27 countries

Modernizing BRIC

Price-competitive BRIC

A cohesive EU policy

Through simulations run with the MASST model, how Europe will look like in 2020 in terms of regional GDP growth rates, regional GDP per capita level, industrial and service employment growth rates will be obtained under the assumptions of these two scenarios. The aim of a forward looking analysis through MASST is not to provide precise estimates of future GDP levels and growth rates, but rather to highlight the main tendencies, major adjustments to change, relative behavioural paths that will be at work, given some conditional assumptions about globalisation trends. The two main scenarios are compared to a “baseline scenario”, based on the assumption that the past and present trends affecting growth and the associated policies put in place will continue in the future. This scenario will act as a benchmark for the alternative scenarios presented hereafter. The reason for constructing this benchmark scenario is that it allows comparison of alternative, extreme and imaginary scenarios with a picture of the difference with a structural condition like the present one will look like in 2020.

8.2. An aggressive Europe in a high-quality competitive world and in a fast recovery from the economic crisis The first scenario, an aggressive Europe in a high-quality competitive world, is a scenario built on courageous strategies by all national economies, in Europe and outside Europe, and by the assumptions that counter-cyclical Keynesian policies are so efficient to allow a fast recovery from the present crisis, and that these demand policies can in the short run be substituted by excellence based policies. For the emerging countries, the main assumption is that the strategies put in place are all an endeavour to compete on the basis of quality of products and of labour, and no longer on the basis on prices, as they do at present. Global competition is increasingly based on product innovation, customised production and international specialisation; under these conditions, there is huge

198

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

potential for worldwide development and increasing welfare in all three blocks of countries.64 This market potential generated is limited in the first years by the economic crisis, which also limits the purchasing power of inhabitants. However, after the first years of crisis, efficient Keynesian policies allow for a recovery of internal demand, of BRICs’ purchasing power and of the enlargement of a market for European intermediate and final goods (Table 8.1). Trade flows increase between BRICs and European countries. The increase of competitiveness of BRICs and the high quality standard of living in these countries is reflected in an increase of energy consumption in these countries, only partially counterbalanced by the development of renewable technologies. An increase in energy price is the result of an increase in energy demand. All sectors recover from the losses in production they had during the first years, and the increase in value added comes back to the level of the period before the crisis. In this scenario, the European Member States take into account a pro-active, and aggressive strategy, and compete on external markets on the basis of product innovation. Open trade with external countries in agricultural, industrial and service products is seen as an opportunity for growth more than a risk. This is a common strategy between Eastern and Western countries. To pursue such a strategy, Eastern countries go through a re-structuring of their economies, and move from agricultural to industrial and even service production. The restructuring process is made possible also thanks to the increase in the presence of FDI, that reinforce the increase in internal private investment which is foreseen as a strategy for both Eastern and Western countries. Virtuous public spending and strict compliance with the Maastricht parameters are part of the proactive strategy of the Member states; economic growth is mostly based on private investments, and in general on efficiency principles. Public investments are mainly devoted to R&D and value added functions.

64

A more detailed description of the scenarios is contained in Annexes 4 and 5, respectively for the first and the second scenario. 199

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Table 8.1 Present trends (baseline) and alternative scenarios Driving forces

Baseline scenario (past and present trends)

An aggressive Europe in a high-quality competitive world

Impact of a crisis with a short recovery

A defensive Europe in a pricecompetitive world

Impact of a crisis with a delayed recovery

1 Competitive strategies of BRICs Quality of global competition

Cost-competitive strategy of BRIC countries; Global cost competition on products; Low purchasing power in BRIC countries Constant financial capital demand. Low interest rates, partially balanced by internal demand increase

Globalisation of markets

Energy sources

Significant modernisation of BRIC countries; Global customised production and competition on quality; Significant increase in wages and in purchasing power in BRIC countries. Higher interest rates .

World crisis, partially balanced by internal demand

Lower growth rates everywhere Lower increase of wages in BRICs.

Significant cost-competitive strategy of BRIC countries; Global cost competition on product. Low wages and low purchasing power in BRIC countries. Lower interes rates.

Recovery from world demand crisis.

Limited trade increase

Trade increase due to higher wages and purchasing power in BRIC countries.

Decreases of growth rates with respect to the past in all sectors, especially open ones

No decrease of industrial growth rates with respect to the past.

Persistence of traditional energy source dependence.

Introduction of energy-efficient technologies; diffusion of renewable technologies and lower dependence on traditional energy sources.

Much lower growth rates everywhere, especially in US&Jap; Much lower increase of wages in BRICs as crisis affects also them.

Strong world demand crisis. Constraints to trade by national governments.

Decrease of growth rates in open sectors with respect to the past due to competition from BRIC and the persistence of the economic crisis.

Limited trade increase due to lower wages and purchasing power in BRIC countries

Strong decreases of growth rates in all sectors with respect to the past, especially open ones.

Persistence on traditional energy source dependence. Contained increase in energy prices.

Significant increase in energy prices because of higher oil demand by BRIC countries and recovery from crisis.

Low increase in energy prices, accentuated by the persistence of the economic crisis.

200

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009 2 Competitive strategies of Europe

European competitiveness

World demand crisis strategy

Free trade.

Recovery from world demand crisis.

Limited opening to extra-EU trade;

Free trade.

Maintenance of free trade to recover from crisis.

cost-competitive strategy of New 12 countries; an innovation strategy of the EU 15, only partially successful.

Significant modernisation of New 12 countries; Innovation strategy in entire Europe.

A partial decrease in concentrated intra-national development Macroeconomic conditions

Management of public finance

Globalisation of production (supplier/producer re-organisation effects)

More demand crisis than in the baseline.

Constraints to free trade by European national governments.

Limitation of free trade to protect national productions from crisis.

Cost-competitive strategy of New 12; Low-cost production competition.

Less concentrated intranational development.

More concentrated intranational development.

Revaluation of Euro with respect to the past

Stronger revaluation of Euro.

Lower exchange rates of Euro with respect to the Baseline.

Decrease of inflation

Higher inflation.

Lower inflation.

Increase of public expenditure growth rates;

Virtuous public expenditure: significant decrease in its growth rates.

Weak control on Maastricht parameters.

Strict respect of Maastricht parameters.

Same composition of the labour force

More high-value functions especially in regions already specialized in high-value functions.

Keynesian policies for a limited period, then excellence-based policies to maintain low deficits.

High increase of public expenditure growth rates.

Heavy public expenditure with Keynesian policies.

Low respect of Maastricht parameters. Higher increase in service employment than in the baseline.

Less high-value functions especially in regions now specialized in high-value functions.

Lower increase in industry employment than in the baseline.

201

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Globalisation of ownership

Socio-demographic trends

Lower attraction of FDI with respect to past situation

Increase of FDI thanks to the European strong currency.

Weakening of social tissue due to transformation processes

Higher weakening of social tissue due to strong transformation processes in global areas.

Increasing external inmigration growth rates counterbalancing population ageing.

Openness to external inmigration; greater natural population growth due to higher fertility rates of immigrants; lower unemployment rate due to recovery from crisis.

High unemployment rate

Lower FDI attraction in the short run, then recovery and higher FDI penetration.

Increase of FDI due to low cost area attraction, partially limited by the economic crisis; Lower weakening of social tissue

Much lower FDI attraction due to protectionism.

Maintenance of workers circulation.

Relative closure to external inmigration; low natural population growth.

Limits to workers circulation (explicit or de facto)

Higher unemployment rate.

3 Strategies of the European Commission EU regional policy

Budget as in 2007-2013

20% increase with respect to the 2007-2013 budget distributed to all regions.

30% decrease of the 2007-2013 budget distributed only to convergence regions.

TEN and TINA infrastructure projects

New TINA and TEN networks chosen keeping the present priorities

New TINA and TEN networks chosen with efficiency aims.

New TINA and TEN networks chosen with the aim to rebalance the territorial infrastructure endowment.

202

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Internationalisation processes reinforce the present ones: the best competition strategies for large multinationals is to move packages of intertwined functions, like production together with R&D, or marketing together with design and final parts of production. Large multinationals will therefore be in search for and be attracted by, high-value functions locations. The quick recovery from the economic crisis generates positive employment growth rates in open service sectors at the EU level. Economic growth is based on endogenous, material and non-material, resources and internal production capacities, and not just on exogenous investment and production. Excellence-based policies invest in research and development, in human capital, in innovation and advanced infrastructure. These national policies are reinforced by European Union’s policies that will be mostly devoted to: i) the fulfilment of the goals set by the Lisbon agenda pursued as a ‘must’ for all European countries; ii) an increase in the 2007-2013 structural funds budget of 20%; iii) EU budget devoted to all regions. A socio-demographic consequence of the decisive and vital development strategy of EU countries is a positive attitude towards economic integration, trade openness and market penetration in the external world. Open trade and lower barriers to in-migration stem from this attitude, the consequence being an increase in the natural population growth rate. The fast recovery from the crisis allows to assume that unemployment rates will decrease, despite the restructuring processes taking place in the economies. Overall, we expect this scenario to be more expansionary than the baseline scenario. The way in which the increasing GDP growth rates will be distributed among regions is something less easy to predict. The way in which global regions will behave in this more expansionary scenario is again an interesting empirical question that will be addressed in the simulation exercise.

8.3. A defensive Europe in a price-competitive world and in a slow recovery from economic crisis The second scenario, a defensive Europe in a price-competitive world, implies trends in the driving forces of change which are almost the reverse of those in the previous scenario. The first important assumption is that the crisis will last for a long period, inevitably characterised by demand driven macroeconomic policies. National public expenditure growth rates increase, with the inevitable consequence of high flexibility in the respect of the Maastricht parameters (Table 8.1). BRIC countries opt for a price-competitive strategy, producing low-cost products in low-tech manufacturing industries. International competition in this scenario is based on local low-cost resources (land and labour) which allow for low-price products. The member countries develop a defensive strategy; constraints to trade are imposed by national governments as a strategy to protect their internal markets. In particular, the New 12 countries try to focus on a cost-competitive strategy, with the aim to attract FDI and to choose the role of Europe’s manufacturing belt. National policies are increasingly oriented towards solving internal problems, and towards an economy where public investment is mainly intended to achieve balanced regional development and territorial cohesion, even at the expense of economic growth. Economic crisis does not stimulate innovation and R&D expenditures are not the main choice of national and European policies. In the energy sector, energy production is based on traditional energy sources, causing pressures on the increase in energy price. This latter effect is however counterbalanced by 203

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

the lower growth rates of oil demand by BRICs. The delayed recovery from the economic crisis imposes negative employment growth rates in open industrial sectors at the EU level. Interest rates are low due to the persistence of the economic crisis, the latter keeping also inflation and exchange rates at lower levels than in the baseline scenario. The economic crisis is more acute in open sectors, which register a negative increase in their production growth rates in Europe. Euro exchange rate is lower than in the baseline. Structural funds budget will decrease by 30% and devoted to convergence regions, while the achievement of the Lisbon agenda goals is flexible. The importance of cohesion also determines the choice of the new TINA and TEN networks, which are selected with the aim of rebalancing the territorial infrastructure endowment. Closure characterises in-migration strategies, with the result of lower natural (and total) population growth. Unemployment increases due to the persistence of the crisis, only partially counterbalanced by public jobs creation. This scenario is expected to be less expansionary than the baseline scenario. Who wins and who loses at the regional level is again something that is highlighted through our simulation exercise. To simulate the scenarios in MASST, any qualitative assumption has to be translated into quantitative values, with the procedure described in Box 8.2.

Box 8.2. Moving from qualitative assumptions to quantitative levers in the MASST model The methodology used to construct the quali-quantitative scenarios requires the qualitative assumptions to be ‘translated’ into quantitative levers to be introduced in the MASST model in order to simulate future growth. We describe here the link between the qualitative and quantitative assumptions. In particular, it states the quantitative assumptions behind each scenario that represent the levers of the model. Technically speaking, these represent the target variables to which the model tends in 2020. Although the quantitative assumptions on the target values of the exogenous variables of the model are defined subjectively, they respond to a very strict logic and to solid constraints. General consistency is required – and pursued – in the entire logical chain linking the general characteristics of each scenario to the potential trend of the main macroeconomic, technological and social variables – our so-called ‘driving forces’. The competitive strategies adopted by European countries influence their internal macroeconomic conditions through intertwined changes in unit labour costs, in exchange rates, in inflation rates and in public expenditure growth rates. A devaluation implies an increase in inflation rates; a devaluation assumption in Europe has to be adjusted for the assumption of the behaviour of BRIC countries. Pro-active, restructuring and modernising strategies are in general expected to couple with virtuous public expenditure, revaluation of the currency, a consequent containment of inflation, and an increase in the interest rate due to increased demand for financial capital and a slight increase in unit labour cost variations. The opposite trends are expected to arise from defensive, cost and price-competitive strategies. A fast recovery from the crisis allows a reduction of public expenditure growth rates, a move towards excellence based policies and in influence on all macroeconomic variables, like interest, inflation and exchange and unemployment rates. The presence of a long or a short crisis affects the trends of these variables highlighted by the other assumptions made: sometimes the long or short crisis emphasises the trends of these variables, in other it counterbalances them. More aggressive strategies conceptually imply a larger share of employment in high-value added activities, this being implemented in the model through the share of science and technology employment and the share of tertiary activities. By contrast, under more defensive strategy assumptions, these activities are expected to grow to a lesser extent.

204

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

8.4. National results of the simulation exercise Once simulated in the MASST model, the three scenarios produce the following results at national level (Table 8.2). First of all, we observe that in the baseline scenario, the GDP growth rate is higher for the New 12 member countries with respect to the Old 15 members. The average of EU 27 is obviously much closer to the Western value since these countries represent the largest part of EU population and GDP. In the baseline scenario, there is a continuing trend of deindustrialization, stronger for Western countries, whereas for the service activities there is an increase of employment, which is only substantial for the New 12 member countries. Table 8.2. Aggregate results of the three scenarios in the MASST model Average annual 2005-2020 growth rates of: Manufacturing Service employment employment GDP Baseline scenario EU 27 2.04 Old 15 2.03 New 12 2.26

-1.00 -1.13 -0.58

0.11 0.01 0.68

Scenario A EU 27 Old 15 New 12

-1.00 -1.13 -0.57

1.48 1.38 2.07

2.88 2.87 3.07

Difference between scenario A and baseline EU 27 0.84 0.01 Old 15 0.84 0.00 New 12 0.81 0.01

1.37 1.37 1.38

Scenario B EU 27 Old 15 New 12

-0.06 -0.13 0.37

1.29 1.28 1.40

-1.15 -1.26 -0.77

Difference between scenario B and baseline EU 27 -0.75 -0.14 Old 15 -0.75 -0.13 New 12 -0.86 -0.19

-0.17 -0.14 -0.31

Scenario A is overall more expansionary, both for the Old 15 and for the New 12 countries of Europe, with the former benefiting slightly more than the latter. The earlier exit from the crisis and the more aggressive strategies of EU countries pay off. Overall, the growth rate of New 12 member countries remains more sustained with respect to the one of Old 15 countries, implying that for these countries an aggressive modernizing strategies allows to be competitive with the exit from the crisis. As far as manufacturing is concerned, scenario A is not substantially different form the baseline, since we are assuming that manufacturing can not be the main driver of growth for Europe, not even in the most positive scenarios. The sector which is driving growth is on the 205

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

contrary the service one, with considerably higher employment growth rates both in the East and in the West. Job creation in services is hence what can be expected in scenario A and what helps explaining the positive GDP performance. Scenario B is less expansionary for all the EU 27, due to the delayed end of the economic crisis and the defensive strategies implemented by the European countries. Also in this scenario the New 12 member countries are outperforming the Old 15 in GDP growth, but less sizeably with respect to the other scenarios. In fact, it appears that the Eastern countries, following a cost-competitive defensive strategy suffer the crisis even more than the Western countries. As far as employment is concerned, deindustrialization in this scenario is accelerated with respect to the baseline in all Europe, more in the New 12 countries than in the Old 15 countries, since the former need to restructure their economies to confront with competition and in this case they fail to do so. Also for what concerns services this scenario is more restrictive with respect to the baseline, with lower employment growth rates in all Europe, and in particular in the New 12 countries which suffer most from the crisis due to the failure to modernize their economies.

8.5. Regional results of the simulation exercise 8.5.1 Results by regional typology The MASST model is able to produce results for each Nuts 2 region in Europe. Given this is a project about globalization, before going to the maps, it is interestingly to observe the aggregate results for the three regional typologies defined in Chapter 3, namely the global players, the regional players and the other regions.

Table 8.3. Average annual regional GDP growth rate: aggregate results by regional typology

Global players Regional players Other regions

Baseline scenario 2.42 1.69 1.68

Scenario A 3.39 2.43 2.32

Difference between scenario A and baseline 0.98 0.74 0.64

Scenario B 1.62 0.96 0.99

Difference between scenario B and baseline -0.80 -0.73 -0.69

Table 8.3 reports the average annual regional GDP growth rate of each typology. It is interesting to observe that global players are those who drive European growth in all three scenarios. In particular, by looking at the baseline, global players are expected to significantly outperform regional players and the other regions, with little differentiations between these two groups. In scenario A, global players take full advantage of the aggressive strategies and of the earlier end of the crisis and increase their growth rate by almost one percentage point. Also regional players and the other regions have a benefit from scenario A, but their advantage is smaller, especially for the other regions, whereas for the regional players the advantage is intermediate. This means that the more a region is open to globalization, the more it ought be able to take opportunities from recoveries and openness to trade, also in the future.

206

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

This is confirmed by the results of scenario B. In this case, still global players outperform the rest of Europe in terms of GDP growth rate. However, they are the most affected by the crisis and the defensive strategies, suffering the larger decrease of GDP growth rate with respect to the baseline. In this scenario, also regional players suffer more than the other regions, and this brings their total GDP growth rate to a level which is the lowest. The specialization of regional players in open growing sectors (in the past), and the fact that they do not hold the structural connections with the world, makes them vulnerable to an enduring economic crisis. Finally, although the least affected in relative terms, the other regions are also significantly suffering this crisis and defensive scenario.

8.5.2. Baseline scenario Map 8.1 reports the results of the GDP growth rate in the baseline scenario. It can be observed that the annual average GDP growth rate is positive for most European regions, with only a few exceptions spread in all countries. Interestingly enough, none of these regions is an agglomerated one, nor one which hosts a Mega. The map confirms that global players are regions which on average outperform the others. Most global players are in fact among the most growing regions, including important cities like Frankfurt, Paris, Munich, Stockholm and the Eastern European capitals, including Warsaw, Budapest, Prague, Bratislava. Interestingly enough, global players are not always the best performing within each country, since in some cases also some peripheral regions, such as Highlands and Islands in the UK, Övre Norrland in Sweden have a very good performance. This scenario is a centripetal one, since on average the richest and central regions tend to outperform the poorer and peripheral regions within their countries. This effect appears stronger in Eastern countries, i.e. extending in the future a pattern which has already seen in the past the Eastern capital regions to outperform the rest of their respective countries. However, some countries appear to be more polarized, notably Spain, Romania, Bulgaria, Hungary, Italy (in which the richer North outperforms the South), the United Kingdom, whereas other countries are less polarized, such as Portugal and Germany, where most Eastern German regions are reducing their income gap with the West. Map 8.2a represents the growth rate of manufacturing employment in the baseline scenario. It is useful to remember that this growth rate, in the model, depends strongly on the actual regional specializations in sectors more or less affected by the crisis. The growth rate is more often negative than positive, signalling that the deindustrialization processes will continue in this scenario, but with strong differences at European level and within countries. In particular, negative growth rates are diffused in regions belonging to central Europe, notably in Germany, the Czech Republic, Austria, Hungary, and also in northern countries such as Ireland, Denmark, Sweden, Finland. Positive manufacturing employment growth rates are present only in a number of rural or in any case non-core regions of Southern Europe, in the Baltic countries and in Bulgaria and Romania, the last two member countries of the EU, where however positive growth rates of manufacturing employment do not take place in the tertiary Bucharest - Iflov. Global players appear to have on average lower manufacturing employment growth rates with respect to the rest of Europe, and, among them, only Sofia, the Luxembourg, Cyprus and the Baltic countries have positive growth rates. 207

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 8.1. Average annual GDP growth rate in the baseline scenario

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual GDP growth rate 2005-2020 - Baseline scenario < 0.48 0.48 - 0.844 0.844 - 1.157 1.157 - 1.459 1.459 - 1.753 1.753 - 2.127 2.127 - 2.519 2.519 - 2.951 2.951 - 3.507 > 3.507

208

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map8.2a. Average annual manufacturing employment growth rate in the baseline scenario

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual industrial employment growth rate 2005-2020 - Baseline scenario < -3.238 -3.238 - -2.157 -2.157 - -1.658 -1.658 - -1.293 -1.293 - -1.015 -1.015 - -0.732 -0.732 - 0 0 - 0.26 0.26 - 1.012 > 1.012

Map8.2b. Average annual service employment growth rate in the baseline scenario

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual tertiary employment growth rate 2005-2020 - Baseline scenario < -1.88 -1.88 - -1.248 -1.248 - -0.846 -0.846 - -0.513 -0.513 - 0 0 - 0.344 0.344 - 0.779 0.779 - 1.517 1.517 - 2.464 > 2.464

209

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 8.2b represents the growth rate of service employment in the baseline scenario. Also here, the results depend on the actual regional specialization, as well as other structural and policy variables. It can be observed that country effects appear to take place, since the growth of services is higher in Southern Europe, in Ireland and in most Eastern countries, with the exception of Poland. Notably, global players are the regions which increase most of the increase in services employment concentrates. This is true in almost all countries, and has notable exceptions only in Berlin and Rome, where a lower increase of public services may be expected. For small regions such as Brussels and the Inner London, the positive effect of the increase of services appear to spread around to neighbouring regions.

8.5.3. Regional results of scenario A Map 8.3 presents the difference in average annual GDP growth rate between the scenario A and the baseline. This scenario is more expansionary, due to the exit from the crisis and the aggressive competitive strategies, and this is confirmed by the map at regional level: most European regions present growth rates higher with respect to the baseline. The benefits of this scenario are fairly spread within the countries in the cases of Germany and Britain, whereas in other cases, such as France, Spain, Hungary, Portugal, Bulgaria and Romania, they are concentrated in the core areas with the lagging areas left behind. This is particularly true for some rural French, Polish, Portuguese, British, Eastern German and Italian regions, whose growth rates are even slightly lower than in the baseline scenario. Also the rural regions of Spain, Greece, Austria, Sweden, Romania, Bulgaria and Hungary only have a small advantage from this scenario, but they have in any case a GDP growth rate which is higher than the baseline one. Global regions are the regions which take most advantage of an aggressive scenario and a fast recovery from the economic crisis. They are among the best performers in Europe, and generally also within their countries. In the cases of Athens, Stockholm, Budapest, Bucharest and Sofia they are leading the growth of their own countries. In other cases, among the best performers are global regions of second hierarchical level, such as Leipzig and Hamburg in Germany, or of Rhône-Alpes and Provence-Alpes-Côte d'Azur in France. In Map 8.4a, represented is the difference of manufacturing employment growth rate between scenario A and the baseline. The scenario A assumptions are that at European level, even with the end of the crisis, one cannot expect large employment benefits in this sector. Despite that, some regional differences with the baseline arise due to the different productive and structural specialization of regions. In particular, global players appear to slightly lose employment in manufacturing, since their higher GDP growth rates are led, as will be evident below, by the tertiary activities. Other regions, on the contrary, appear to increase their manufacturing employment; it is the case of some urban non-agglomerated regions in the Eastern countries of Poland, Romania, the Czech Republic and in Western countries such as Germany, Belgium, The Netherlands and the U.K.

210

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 8.3 Annual average GDP growth rate: difference between scenario A and baseline

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual GDP growth rate 2005-2020 - Difference between scenario A and baseline 1.371

211

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 8.4a Annual average manufacturing employment growth rate: difference between scenario A and baseline

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual industrial employment growth rate 2005-2020 - Difference between scenario A and baseline -0.027 - -0.012 -0.012 - -0.004 -0.004 - 0 0 - 0.003 0.003 - 0.006 0.006 - 0.009 0.009 - 0.014 0.014 - 0.021 0.021 - 0.038 0.038 - 0.065

Map 8.4b Annual average service employment growth rate: difference between scenario A and baseline

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual tertiary employment growth rate 2005-2020 - Difference between scenario A and baseline 0.912 - 1.073 1.073 - 1.182 1.182 - 1.236 1.236 - 1.287 1.287 - 1.342 1.342 - 1.402 1.402 - 1.464 1.464 - 1.547 1.547 - 1.647 1.647 - 1.968

212

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 8.4b reports the difference in service employment growth rate between scenario A and the baseline. It is immediately clear that in this scenario tertiary employment grows significantly more for all European regions, especially at the Eastern, Southern and North-Western fringes of the EU. Different patterns appear to take place in the New 12 countries with respect to the Old 15 countries: in the East, tertiary employment growth is more sustained in the capital regions, such as Prague, Warsaw, Budapest, which are able to attract activities and exploit their superior productive system. In the West, on the contrary, the already tertiary capital, Mega or agglomerated regions appear to be outperformed by other more peripheral regions, whose higher growth rates appear to be due to later coming tertiarization. Among the best performers, however, one can also find very strong regions such as Vienna, Utrecht, the Noord-Holland, Brussels, the Outer London. Being a global player does not hence automatically imply a higher or lower service employment growth rate. In fact, global regions in Eastern and southern areas appear to gain more tertiary employment than the average, whereas those in central Europe, notably Germany, but also Lombardy and Piedmont, appear to have high but lower than average gains.

8.5.4. Regional results of scenario B Scenario B is a scenario in which a defensive strategy by the European Union, crosses a costcompetitive strategy by the BRICs and a delayed end of the crisis, for this reason it is no surprise to find that the difference in annual average GDP growth rate with respect to the baseline is negative for almost all European regions (Map 8.5). However, some regions are less than the others affected by the depression in this scenario, in particular in the Eastern and Northern countries. However, also within these countries, the Megas and capital regions are struggling, whereas the less affected regions are the most peripheral and rural ones. Among the regions which survive better from this depression scenario, the rural peripheral regions of Greece, Spain and Portugal, where European structural funds expenditure is larger. Global players are among the worst affected in this scenario, especially Madrid, Cataluña, Lisbon, Leipzig, Athens, and the Eastern capitals of Prague, Bratislava, Budapest, Warsaw. However, other global players survive with middle range results, regions such as Munich, Frankfurt and most western German global regions, Antwerp, Lombardy, Berkshire, Bucks and Oxfordshire and most of the British South-East. The fact that global players are on average the most affected but not by far, may be due to the fact that these regions are more open than the others to the external world, but their strong economic structure allows them to deal somehow with the crisis. In map 8.6a represented is the difference of manufacturing employment growth rate between scenario B and the baseline. This difference is negative for all regions, implying a strong deindustrialization process, with just a few exceptions in some regions of Southern Europe. Overall, in fact, it appears that the Mediterranean regions of Southern Europe, are those which are better able to survive the crisis, in Spain, Portugal, Greece, Southern France and Southern Europe. Rural regions in Eastern European countries, on the contrary, appear to be the most suffering ones, which may be due to the fact that, by applying a cost-competitive defensive strategy, these regions are unable to avoid restructuring through job cuts. In this sense, the most advanced Eastern regions, i.e. the capitals, are less deindustrializing than the rest of their countries.

213

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 8.5 Annual average GDP growth rate: difference between scenario B and baseline

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual GDP growth rate 2005-2020 - Difference between scenario B and baseline < -1.708 -1.708 - -0.922 -0.922 - -0.786 -0.786 - -0.712 -0.712 - -0.663 -0.663 - -0.612 -0.612 - -0.519 -0.519 - -0.341 -0.341 - 0 >0

214

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Map 8.6a Annual average manufacturing employment growth rate: difference between scenario B and baseline

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual industrial employment growth rate 2005-2020 - Difference between scenario B and baseline < -0.353 -0.353 - -0.255 -0.255 - -0.187 -0.187 - -0.158 -0.158 - -0.136 -0.136 - -0.114 -0.114 - -0.091 -0.091 - -0.07 -0.07 - -0.046 > -0.046

Map 8.6b Annual average service employment growth rate: difference between scenario B and baseline

Politecnico di Milano - MASST MODEL - June 2009 Typology of regions Global player Average annual tertiary employment growth rate 2005-2020 - Difference between scenario B and baseline -0.065

215

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Also because of the relative good performance of Eastern capitals, global players are surviving slightly better than the rest of Europe to this contractionary pattern. Regional players, on the contrary, are the worst performing, since they are highly specialized in the most open sectors and suffer from this specialization more than the other regions - which are specialized in other sectors and more than the global players, whose economic structure is stronger. In fact, also in Western countries, global players appear to hold on average better than their own countries, especially in the cases of Paris, Frankfurt, London, Brussels, Rome, the Luxembourg, Utrecht, Noord-Holland and Zuid-Holland. Map 8.6b represents the difference in service employment growth rate between scenario B and the baseline. In this case, the difference is negative for all European regions, with no exception. It is interesting to observe that the pattern appears to be strongly influenced by the national belonging. New 12 countries appear to be among the most suffering ones, with the exception of Bulgaria and Romania, where the increase in service activities is a process which is still in progress. In the Old 15 countries, Spain, Portugal and Greece are those with the lowest progress of service employment with respect to the baseline, remembering however they were among the best performers in the baseline. Service employment is negative but somehow holding on in central Western Europe, in particular in Britain, Denmark, France, Northern Italy and Western Germany. In all these countries, global players are performing better than average, especially in the cases of Utrecht, Bedfordshire and Hertfordshire, Berkshire, Bucks and Oxfordshire, Lombardy, Karlsruhe, Rheinhessen-Pfalz and Ile de France. Also other urban non-agglomerated regions are performing relatively very well, especially in Britain and Western Germany.

8.6. Main findings for the global regions If we look at the most courageous scenario, in which competition increases and the economic downturn vanishes in a short period, the prospective analysis reports the following results for the global regions:  global regions are the regions which take most advantage of an aggressive scenario and a fast recovery from the economic crisis. They are among the best performers in Europe;  global players appear to slightly lose employment in manufacturing, since their higher GDP growth rates are led by the tertiary activities;  being a global player does not hence automatically imply a higher or lower service employment growth rate. In fact, global regions in Eastern and southern areas appear to gain more tertiary employment than the average, whereas those in central Europe, notably Germany, but also Lombardy and Piedmont, appear to have high but lower than average gains. If we look at a less courageous and more protective scenario, in which the economic crisis persists over time, the most interesting results for the vulnerable regions concern:  global players are among the worst affected in this scenario;  also because of the relative good performance of Eastern capitals, global players are surviving slightly better than the rest of Europe in terms of relative employment growth with respect to the baseline;  no difference emerges in the service employment dynamics with respect to the baseline for all regions. 216

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

9. Conclusions and Policy Implications 9.1. Retrospect: the main lessons from our empirical analysis The aim of this last chapter is to draw some conclusions and policy lessons from the main results achieved by the study. We intend to derive regional policy suggestions from interpretations of the results of the empirical analysis, doing so in a deductive way. A first important message from our analysis, and which proves to be of strategic importance for the policy implications, is that despite their similar degree of exposure to the external world, global regions vary greatly in their capacities to gain opportunities from globalisation. This result receives two main explanations from our analysis. The first is the fact that open regions are present in all European countries and are subject to strong national effects in their growth patterns. For example, the leading role of Eastern countries redounds on all regions of countries, and not merely global regions; in Western countries similarities in growth trajectories can be highlighted among regions belonging to the same country. The second reason is linked to clear evidence that global regions have different economic dynamics because of their different endowments of strategic assets, like human capital, innovation capability, and high-level functions. Their different territorial specificities explain much of their different performances. Among global regions, our analysis has identified a group that performs relatively better in terms of GDP growth and which can be termed “benefiting regions”, defined as those regions that maintain and even reinforce their competitive positions in the European economy thanks to globalization processes, i.e. those regions able to increase their relative production capacities (measured in terms of GDP) more than the European average in a period of globalisation. Our analysis demonstrates that the local economies of these regions share some strategic behaviours: they avoid a simple quantitative substitution effect between industrial and service jobs; they accompany an increase in service jobs with an increase in private service productivity; they deal with a negative increase of industrial employment by increasing industrial productivity: they are able to attract FDI in their sector of specialization; and they are endowed (especially in Western countries) with a relatively high share of command and control functions. Evidence for all these findings is provided by both the quantitative and the qualitative (case studies) analyses. Our empirical evidence shows that also in the case of regional players, some are more competitive than others. These regions are defined on the basis of their more open industrial structures relative to the EU average but a lack of physical accessibility with respect to the EU average. This means that the regional players successful in this highly competitive period are those that, notwithstanding the high degree of openness of the industrial structure in which they are specialised, are able to take advantage of a globalised economy. Interestingly, the structural features that characterise this higher performance differ between Eastern and Western regional players, and between regional and global players. The profile of an Eastern winning regional player is that of a regional economy able to reconvert its industrial activities towards the most dynamic industrial sectors (not particularly hightech ones), where GDP growth is achieved, not through productivity increases, but through strong advantages in industrial restructuring and entrepreneurial capacity. The dynamics of industrial sectors are accompanied by a dynamic of a service sector which is not particularly advanced but closely integrated with the industrial system. In Western countries, winning regional players are highly specialised in SMEs, which keep high-level functions, and are thus able to develop product innovation and be competitive. Interestingly, they are not specialised in the high-value functions of large firms, which on the contrary are typical of global players. They specialise in medium-tech industries and are therefore successful because they do not compete with the role of global players. 217

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

The winning global players, for their part, are characterised by dynamic efficient urban systems with high command and control functions, and by the presence in the area of large enterprises playing a leading role in the national economies. Prospective analysis shows the strategic role of global regions in the future of the EU economy. In a scenario of courageous strategies – able to drive European countries towards a knowledge-economy profile and to modernise economies in emerging countries – and of relatively rapid recovery from the present economic crisis, global regions record the highest GDP growth rates, and among them are those that are at present benefiting from globalisation. A reverse scenario of protectionism amid slow recovery from the economic crisis shows global regions able to lose less than the other regions and maintain their relative role of leadership in the EU economy. These differences among the economic behaviours of open regions, both in the past and in the future, can be explained by: -

different strategies put in place by regions, when these are conceptualised as collective actors; different impacts of macroeconomic and structural policies managed at national level on regions; different impacts of structural funds on regions.

9.2. Regional strategies: winning behaviours One of the first explanations for the differing performances of global regions is the different strategies that they put in place as collective actors. Our empirical analysis clearly evidences that the global regions able to benefit from globalization are those that avoid a quantitative substitution of employment growth from industrial to service activities, without keeping productivity growth as an important goal. In the introductory chapter to this report, we envisaged in conceptual terms some major important strategies that local economies should put in place in order to benefit from globalization, namely: -

a balance between high- and low-value functions; non-attraction of only low-value functions; attraction of high-value functions; balance between core activities (manufacturing) and related new activities (services); balance between industry and service sectors and functions maintaining a certain level of innovation.

Our empirical analysis has shown that these strategies are present in global regions (chapter 4). Figure 9.1 shows that most of them are especially present in benefiting regions, because these regions exhibit: -

-

a share of high value functions in one period accompanied by an increase in service productivity growth in the following period, with a much stronger relationship than in conservative regions, these being defined as those regions among the open ones with lower regional differential growth rates (Fig. 9.1a); a share of human capital that, in the following period, is accompanied by an increase in industrial productivity. This is again not the case of the cluster of conservative regions (Fig. 9.1b); 218

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

-

a larger amount of high-level functions, which increase over time (Fig. 9.1c).

This indicates that defensive strategies aimed at a conservative behavior with respect to the past do not pay in the long run; a result which is also confirmed by the prospective analysis of our study: the defensive scenario based on conservative strategies adopted by the main blocks of countries leads to lower economic performance throughout Europe, and especially in the weakest areas of the continent.

9.3. The impact of national trends and performance on regions One of the main features of our analysis is that it pays direct attention to national performance and trends and their impact on regional performance, together with close emphasis on the importance of territorial specificities in determining regional destinies. This applies both to the methodologies applied to data on the past and to MASST, our forecasting model. This emphasis does not derive from purely statistical and econometric reasons: in fact, among the causes of regional success and failure are factors directly linked to certain pervasive and generalized characteristics of the national economy. We refer in particular to institutional factors such as the performance of the high functions of the nation-state – legislative, judicial and governmental; to organizational factors such as the efficiency of services of general interest like education, transportation, communication, health and security services; to economic factors such as general fiscal pressure, the effectiveness of public expenditure, the pervasiveness of environmental regulations, and the efficiency of contract enforcement procedures. Fig. 9.1. Relationships among groups of variables: a comparison among regional clusters a) Relationship between high value functions and private service productivity growth

Catching-up

4.15

High level functions (Corporate managers) 1999-01

3.15

Private services productivity growth (2002-05) 0.036

0.08 2.15

Benefitting

1.15

219

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

b) Relationship between the presence of human resources in S&T and the growth of manufacturing productivity Catching-up 4.1

Share of human capital in S&T in 2000

3.6

Industrial productivity growth (2002-2005)

Benefitting 0.12

3.1

2.6

Conservative 2.1

c) Relationship between high level functions and high level function growth 0.016 High level functions (Business professionals) 1996-98

Catching-up

0.014

0.012

0 0.01

Benefiting

High level functions growth (Business professionals) 1996-98

0.008 Conservative

0.006

A second element linking regional economies to the national economy’s general performance is close inter-regional, within-country integration, relative to inter-national integration, in terms of the exchange of goods, services and production factors due to proximity effects and the absence of any kind of institutional or linguistic barrier. 220

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

All this is reflected in the empirical evidence. By and large, the variance of regional growth rates within countries is lower than the variance of inter-national growth rates, with all regions benefiting from good short-term and long-term performance by their national economy. This fact was evidenced by our empirical analysis on regional growth in the EU between 2002 and 2005, where almost all countries exhibited standard deviations in inter-regional growth rates which were lower than the inter-national standard deviation of growth rates of EU 27 countries. This was particularly the case of the Old 15 countries (Table 9.1). On the other hand, all New 12 countries recorded a greater variability of internal regional growth rates, thus lending support to expectations à la Williamson65 about increasing inter-regional disparities in the first phase of a development or integration process.

Table 9.1. Differentials in inter-regional GDP growth rates between 2002 and 2005 Geographical unit of reference EU 27 Austria Belgium Bulgaria Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden UK

Standard deviation 1.74 0.34 0.80 2.47 1.34 1.10 1.20 0.77 0.54 1.73 1.50 1.37 0.63 1.06 0.98 0.57 1.10 2.40 0.46 0.64 0.85 0.96

Minimum

Maximum

Range

0.54 1.34 0.54 1.15 3.35 3.59 0.15 8.54 0.04 0.59 -0.87 -0.31 1.95 4.45 -0.88 8.52 8.81 3.99 1.08 0.49 2.36 -0.27 3.86 2.44 3.30 2.28 1.58 0.66

8.81 2.48 3.64 7.53 3.35 8.08 3.29 8.54 3.39 3.18 1.81 6.35 6.43 7.12 1.80 8.52 8.81 3.99 1.08 3.82 5.57 1.62 6.94 8.43 4.21 4.80 4.23 4.67

8.28 1.13 3.10 6.37 4.49 3.14 3.35 2.60 2.68 6.66 4.48 2.67 2.68 3.32 3.21 1.89 3.08 5.99 0.91 2.53 2.65 4.01

If aggregate, national development is an important part of regional development, it follows that a first class of territorial development policies consists of sound and consistent policies internal to each country and designed to achieve the pervasive effectiveness of the public administration and to provide public goods and externalities which enhance the development capacities of all local economies.

65

Williamson J.G. (1965), “Regional Inequality and the Process of National Development: a Description of the Patterns”, Economic Development and Cultural Change, vol. 13, pp. 3-45. 221

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

9.4. The impact of macroeconomic and structural policies on regions A clear lesson arising from our simulation results is the importance of macroeconomic and structural policies managed at the national level in exerting strong and differentiated effects on regions. Interest-rate policies, monetary and fiscal policies driving exchange-rate movements. But also such policies as transportation and TENs policies, excellence policies in R&D and competitiveness, and agricultural policies, exert selective effects on different types of regions through a wide array of transmission channels or direct regional targets. The case of agricultural policies is easy to understand, for they provide direct support to rural regions which, at present, are often among the richest ones, at least in many Old 15 countries. But other policies, too, have selective effects: TENs and excellence policies, for example, are naturally biased in favour of stronger regions, where most of the favourable preconditions for successful competitiveness policies are in place, and where the demand for new transportation infrastructure is greater, and guarantees the highest economic return to investment. The disparity-widening effects of such policies were measured by the MASST model and widely illustrated in the previous chapters. Another category of national policies which concern the macroeconomic sphere are usually considered to be neutral in terms of inter-regional disparities. Yet, for inter-sectoral reasons, selective effects are visible in this case as well. In fact, a rise in interest rates is likely to have the greatest impact on regions specialized in capital-intensive sectors such as manufacturing and building and construction. A similar effect may be expected to ensue from a real revaluation of the currency, which generates wider tensions in industries and regions characterized by a greater degree of openness to international trade. Furthermore, revaluation is likely to hit more, coeteris paribus, those regions specialized in labour-intensive industries, because it raises labour costs expressed in international currency (while the cost of capital will remain unchanged at the level determined internationally, especially if the revaluation is indicative of a strong and potentially fast-growing economy). By focusing on the spatial effects of specific national policies and changes in macroeconomic contexts, our MASST model made it possible to disentangle the effects of single policy trends. In particular, presented here are the results of a public expenditure policy put in place for recovery from the economic downturn. The regional efficiency in public expenditure growth rate emerged in the comparison within countries, where the total amount of public expenditure is the same throughout all regions.66 The overall effect of a restrictive fiscal policy on EU growth is negative, and this is true for all countries and regions, as expected. The effects are more marked in Eastern countries, which are more sensitive because of their high dependence on public expenditure in the past. Within some of these countries, like Poland and Hungary, the spatial effect differs between peripheral and capital city regions; the former suffer the most, while the latter are less affected. In the EU15, the restrictive effects are lower than in the New12 countries, with a variation at regional level within countries. In particular, Germany and Italy are generally less affected; in general, in all EU 15 countries the regions suffering less from a restrictive fiscal policy are the

66

Especial attention is required in this regard. The regional effects obtained by our analysis are the induced effects that a national public expenditure generates. In the MASST model, in fact, the direct impact is estimated to be the same in all countries; the regional differentiation is obtained through recursive and cumulative effects through spillovers. In MASST, the regional efficiency of public expenditure therefore depends on the structural features of regions. 222

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

stronger areas. Interestingly, in general, agglomerated and capital city regions are less sensitive to this restriction (Map 9.1). These various considerations warrant careful inspection and monitoring of the regional effects of certain aggregate and structural policies managed at the national and EU level, especially in a period of economic downturn, and close attention by most national governments to the ‘champion’ regions which they regard as the ‘leading edges’ of the country’s competitiveness. Map 9.1. Effects of a lower public expenditure growth rate on regional GDP growth

Politecnico di Milano - June 2009 Average annual GDP growth rate 2005-2020 Effects of public expenditure -0.105 - -0.098 -0.098 - -0.096 -0.096 - -0.094 -0.0943 - -0.0886 -0.0886 - -0.0804 -0.0804 - -0.075 -0.075 - -0.073 -0.0728 - -0.0689 -0.0689 - -0.0663 -0.066 - -0.065 -0.0651 - -0.0638 -0.0638 - -0.0611

9.5. The rationale for regional policies The strongest argument in favour of regional policies lies in the long-term persistence, and even widening, of inter-regional disparities. In fact, the entire history of the European integration process has been apparently characterized by slowly decreasing overall inter-regional disparities; but, in reality, this process generally results from strong catching-up processes among nations, accompanied by increasing disparities at the intra-national level. These last processes of increasing disequilibria within individual countries may exhibit different intensities over time. The Williamson law (Williamson, 1965)67 states that they are naturally 67

Williamson J.G. (1965), “Regional Inequality and the Process of National Development: a Description of the Patterns”, Economic Development and Cultural Change, vol. 13, pp. 3-45. 223

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

stronger in the early stages of the development or integration process, while in a second stage they will slow down or even go into reverse. This prediction has been shown as empirically correct if interpreted in light of the macro-phases of European integration: -

-

increasing disparities in the first development and integration phase of the 1950s until the mid-1960s, followed by a reduction until 1980, mainly due to catching-up by a new category of intermediate-income, ‘third’ regions like the ‘Third Italy’, Flanders, southern Germany, southern France, and East Anglia; an upswing in disparities during the industrial restructuring processes of the 1980s, the new integration phase of the Single Market, and the path towards the single currency, until 19952000, and still no clear evidence afterwards.

The results of the MASST simulations and foresights show a tendency towards concentration and increased disparities also for the future. Overall, between the initial simulation period and 2020, disparities in the EU27 increase, mainly due to the economic downturn, which strongly hits both core and peripheral regions, the latter having less capacity of reaction. The economic downturn annuls the advantages of rapid processes of inter-national catching-up (mainly by New Member States), with a slight decrease in intra-national disparities (Table 9.2). But this general trend is generated by two different evolutions: interregional disparities will increase because of a strongly widening intra-national dualism between core and periphery regions insufficiently counter-balanced by slowly decreasing inter-national disparities (Fig. 9.2).

Table 9.2. Evolution of per-capita income disparities, 2005-2020 Theil index

2005

2020

Difference

Baseline Total Between countries Within countries

0.17766 0.14543 0.03233

0.18568 0.13232 0.05335

0.00791 -0.01310 0.02102

Scenario A Total Between countries Within countries

0.17766 0.14543 0.03233

0.18903 0.13157 0.05746

0.01127 -0.01385 0.02512

Scenario B Total Between countries Within countries

0.17766 0.14543 0.03233

0.18656 0.13462 0.05193

0.00879 -0.01080 0.01960

224

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Fig. 9.2. Evolution of per capita income disparities, 2005-2020 Theil index - Baseline scenario 0.2 0.18 0.16 0.14 0.12

Total baseline

0.1 Between countries baseline

0.08 0.06 0.04 0.02 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Year

In all cases, intra-national disparities are likely to widen. Overall, considering that a new phase of continental integration is beginning (and that our starting year for long-term comparison is 2005), Williamson’s prediction may come true – once again. These tendencies are reinforced in the scenario with rapid recovery from the economic downturn mostly due to the capacity of strong areas to take advantage of the revitalization of the economy. The slow recovery scenario, instead, records a lower increase in regional disparities, probably due to the pervasiveness of the negative effects of the crisis in all sectors and activities in all strong areas (Fig. 9.3). A “two-speed growth model” is an appropriate expression to denote the way in which European regional growth will develop in the future. Fig. 9.3. Differences in the evolution of per capita income disparities among scenarios, 2005-2020 Total Theil index in the three scenarios 0.19 0.188 0.186 0.184 0.182 Baseline

0.18

Scenario A

0.178

Scenario B

0.176 0.174 0.172 0.17 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Year

Another important explanation for the need of regional policies is that our empirical results show that the capacities of reaction to the economic downturn differ even among global regions; within 225

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

this group, the cluster of benefiting regions is the one able to perform best in each scenario (Table 9.2). These results express two major messages: a) there are strong regions which, like the conservative regions, require particular help in turning their potentialities into real growth opportunities; b) some sort of cumulative reinforcing learning process takes place in governance and public policies, as witnessed by the constantly good performances of benefiting regions.

Table 9.2. Average annual GDP growth rates between 1996 and 2020 Benefiting regions

Catching-up regions

Conservative regions

2.75 3.74 2.01

2.77 3.85 1.19

2.03 2.99 1.28

Baseline Scenario A Scenario B

9.6. Regional policy implications Some important policy implications ensue from the lessons and messages obtained by our empirical analysis, as follows:

A) Given the good economic performances of benefiting regions, regional policies should first and foremost support regions in opening their local economies so that they can exploit the advantages deriving from an integrated world economy. Policies should first be addressed to those regions already endowed with activities belonging to open sectors but which lack international attractiveness. Regional areas, as they are labelled in this report, are the most suitable regions for the application of such policies in order to increase their attractiveness to international headquarters, high-quality human capital and specialized labour through local quality-of-life interventions, as well as through fiscal and labour costs incentives and efficient territorial policies. B) Regional policies should seek to reinforce the winning strategies of local economies by avoiding generic assistance policies in support of employment, and by developing job-creating policies only in those sectors and activities that show a capacity to increase productivity growth. The latter is more likely to take place either in sectors in which the region is specialized, where core competencies and specific local assets are present, or in high-value sectors like advanced services. Figure 1 depicts these two regional trajectories; the first one (a) envisages a development from an increase in employment growth in indigenous activities to an increase in productivity growth in these sectors, while the second one (b) highlights a trajectory of increased productivity growth in new sectors and activities, followed by increased employment in the newly-established sectors. Both trajectories end with a situation comprising a higher productivity growth rate and a higher employment growth rate. However, trajectory (b) requires the decisive rationalization of local activities and a regional policy supporting the creation of new high-value sectors and activities through restructuring of the local economy. Trajectory (a) is rooted in indigenous activities and preserves local core competences. For this reason, between the two policy trajectories, (a) is the best strategy to put in place when an implicit related risk is overcome. The risk of this strategy is in fact that a higher increase in employment will be achieved in indigenous activities, thanks to assistance policies, with no effort made to achieve higher growth rates in indigenous activities, so that a move to the upper quadrant of Fig. 9.4 is prevented. As our empirical analysis shows, a choice of this kind 226

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

does not generate the competitive conditions for a local economy to withstand fiercer international competition. C) The move of a local economy from a simple policy of increased employment growth rates to productivity growth rates in indigenous activities is relatively simple if that local economy is specialized in capital-intensive advanced sectors and activities. When, instead, indigenous activities are related to labour-intensive sectors, regional policies must be devoted to facilitating a transition to higher-efficiency functions and activities, enlarging the filière to include advanced complementary activities (e.g. design, product innovation, marketing) located upstream and downstream from the core production. Many success stories can be found in the real world; in our case studies, the capacity to move upstream of the production chain and specialize in knowledgeeconomy activities has been the key to the success of local economies like Stockholm and Barcelona. Fig. 1. Possible regional trajectories

Productivity growth rate

b a

+

-

-

+

Growth rate of employment in indigenous activities

D) Increased productivity growth rates can also be achieved through local policies supporting innovation; as our empirical analysis shows, it is not merely the degree of R&D expenditure and of technological development that is linked to winning strategies. Innovation has to be understood as all efforts devoted to increasing knowledge so as to foster sectoral transformation, develop local capabilities to cooperate synergically with other regions, and invent new organizational solutions at both the firm and public governance levels. E) Whilst the above are suggestions arising from our empirical analysis on the past, our prospective analysis suggests some policies with which to support local economies amid the present economic crisis. Keynesian, demand-oriented policies are necessary to tackle the present economic downturn, and they should be put in place by national governments in the most efficient way possible. Our analysis shows that macroeconomic policies have different impacts at the regional level, and that in any case economies will emerge from the crisis with a “two-speed growth model” that will reinforce regional disparities. Global regions will perform better than the others by recording either higher growth rates in a scenario of rapid recovery or a smaller economic downturn in a scenario of a long economic crisis. In this framework, structural, long-term, supply-side policies must complement public investment to limit the territorial asymmetries caused by the recession period. In global regions, structural policies must be devoted to reducing the social costs of the crisis, while in weak areas they must support the few indigenous activities that are present (like entrepreneurship, innovation capabilities). This also applies to the EU’s cohesion policies, which must be devoted to limiting the regional disparities generated by the economic downturn by complementing public 227

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

investments to foster endogenous capacities, and by implementing “smart investment policies” designed to maximize the efficiency of returns on investments. F) The integration of short- and long-term policies managed at different governmental levels is a key to long-run success. Sectoral regional long-term policies must be devoted to increasing efficiency, innovativeness, and competitiveness in the long run, in those sectors where short-term policies are intended to increase consumption. Public incentives in the automobile sector, for example, should be coupled with structural policies oriented to the development of hydrogen cars. Likewise, structural policies should support social reforms with an impact on internal demand. In Eastern countries, pension reforms could allow precautionary savings to turn into internal consumption. G) All regional structural policies must be based on the uniqueness of places by reinforcing the assets and core competences already present in the area. Turning Territorial Diversity into Strength (the title of the Green Paper) summarizes this notion as stated some fifteen years ago in a similar terms (“turning specificities into assets”) to synthesize the idea that regions should build their competitive advantages from below, investing in their particular competences, resource endowments and even weaknesses (such as inaccessibility, if this means conservation of an unspoilt environment), renovating their productive ‘vocations’ and cultural assets. This is still valid today; and even more so in an era of globalization in which local specificities help regions to emerge and compete in a global economy. Regional policies should help regions not according to their needs, but according to their innovative ideas concerning their uniqueness. H) From this it follows that different policies should be devoted to global and regional players, because their degree of success in globalization depends closely on different strategic assets. Global players must be helped to position themselves at the highest levels of the international division of labour by attracting technologically highly advanced functions and activities in both industry and service sectors. By contrast, regional players should not be forced to play the same role as global players; they do not possess the right structural assets to compete with global players, and a direct competition with them would be at a loss. In the case of regional players, by contrast, policies should be designed to help them play an important role in the international division of labour, becoming the most attractive areas for medium-tech activities and functions both in industry and services. I) As a final remark, our empirical results bear witness to the importance of FDI for regional development. This holds not only for Eastern countries but also for Western ones. The case of London’s outperformance is emblematic in this regard. Regional policies should be oriented to the attraction of multinational companies. Especially in regions with clear identities and productive vocations, incentives to attract FDI should be oriented to the filière, reinforcing and enlarging already-present core competences to encompass upstream and downstream value-added activities. Globalization is a process that cannot be avoided; regional policies should help local economies to turn threats into opportunities, and to put winning strategy trajectories in place.

228

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

References Abraham K. G. and Taylor S. K. (1996), “Firms’ use of outside contractors: Theory and evidence”, Journal of Labor Economics 14, 3: 394-424. Amiti M. and Wei S. J. (2005), “Fear of service outsourcing: is it justified?”, Economic Policy 20, 42: 308-347. Baldwin R. (2006), “Globalisation: the great unbundling(s)”, Paper for the Finnish Prime Minister’s Office, Economic Council of Finland as part of EU Presidency. Baldwin R. E., Francois J. F., Portes R., Rodrik D. and Szekely I. P. (1997), “The Costs and Benefits of Eastern Enlargement: The Impact on the EU and Central Europe”, Economic Policy, Vol. 12, n. 24, pp. 125-176. Barba Navaretti G. and Venables A. (2004), Multinational Firms in the World Economy, Princeton University Press. Bhagwati J., Panagariya A., and Srinivasan, T. N. (2004), ‘The muddles of outsourcing’, Journal of Economic Perspectives 18, 4: 93-114. Bloningen B. (2005), “A review of the empirical literature on FDI determinants”, Atlantic Economic Journal, 33(4), 383-403. Camagni R. (2002), “On the Concept of territorial competitiveness: sound or misleading?”, Urban Studies, n. 13, pp. 2395-2412. Camagni R. (2008), « Towards a concept of territorial capital », in Capello R. et al., Modelling regional scenarios for the enlarged Europe, Springer Verlag, Berlin Camagni R., (2001), “Policies for spatial development”, in OECD Territorial Outlook, Paris, 147169. Camagni, R. (1991), “Regional Deindustrialization and Revitalization Processes in Italy”, in Rodwin L. and Sazanami, H. (eds.) Industrial Change and Regional Economic Transformation, Harper Collins, London, pp. 137-167. Capello R. (2007a), “A Forecasting Territorial Model of Regional Growth: The MASST Model”, Annals of Regional Science, Vol. 41, n. 4, pp. 753-787. Capello R. (2007b), Regional Economics, Routledge, London. Capello, R. and Fratesi, U. (2008) “An Advanced MASST Model: the Social and Sectoral Dimensions Reconsidered”, presented at the North American Regional Science Meeting, New York. Capello, R. and Fratesi, U. (2009) “Modelling European regional scenarios: aggressive versus defensive competitive strategies”, Environment and Planning A, vol. 41(2) pages 481 – 504. Capello, R. and Fratesi, U. (forthcoming) “Globalization and a Dual Europe: Future Alternative Growth Trajectories”. The Annals of Regional Science. Published on-line in July 2009, DOI 10.1007/s00168-009-0295-6. Capello, R., Camagni, R., Chizzolini, B. and Fratesi, U. (2008) Modelling Regional Scenarios for the New Europe. Macroeconomic, Institutional Social and Territorial Driving Forces of Change, Springer-Verlag, Berlin. Clark G. L., Palaskas T., Tracey P. and Tsampra, M. (2004), “Globalisation and Competitive Strategy in Europe’s Vulnerable Regions: Firm, Industry and Country Effects in LabourIntensive Industries”, Regional Studies, Vol. 38, n. 9, pp. 1085-1100. Dunford M. (2003), “Theorising regional economic performance and the changing territorial division of labour”, Regional Studies, v. 37, pp. 839-854. 229

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Dunford M. (2006), “Industrial Districts, Magic Circles, and the Restructuring of the Italian Textiles and Clothing Chain”, Economic Geography, January 2006, v. 82, iss. 1, pp. 27-59. Dunning, J. (1998), “Location and the Multinational Enterprise: a Neglected Factor?”, Journal of International Business Studies, 29 (1), pp. 45-66. EC, (2005),“The Location of Multinational across EU Regions”, in The EU economy 2005 review, pp. 237-249. Feenstra R. C. (1998) “Integration of Trade and Disintegration of Production in the Global Economy”, Journal of Economic Perspectives, 12(4): 31-50. Feenstra R. C. and Hanson G. (1999), ‘The impact of outsourcing and high-technology capital on wages: Estimates for the United States, 1979-1990’, Quarterly Journal of Economics 114, 3: 907-940. Fontagné L. and Lorenzi J-H. (2005), Désindustrialisation, Délocalisations, La Documentation Française, Paris Fujita M. and J. Thisse ( 2002), Economics of Agglomeration, Cambridge University Press. Ganzeboom H.B., Treiman D.J. (1996), “Internationally comparable measures of occupational status for the 1988 International standard classification of occupations”, Social science research, vol. 25, p. 201-239. Gardiner B., Martiner, R. and Tyler P., (2004), “Competitiveness, Productivity and Economic Growth across the European Regions”, Regional Studies Vol. 38, n. 9, pp. 1045-1067. Grossman, G. M. and Helpman, E. (2005), ‘Outsourcing in a global economy’, Review of Economic Studies 72, 1: 135-159. Grossman, G.M., Rossi-Hansberg, E. (2008), Trading Tasks: A Simple Theory of Offshoring, American Economic Review, 98-5, 1978-1997. Humprey, J. and Schmidz, A. (2002) “How does insertion in global value chains affect upgrading in industrial clusters?”, Regional Studies, Vol.36, pp.1017-1027. Kenney, M. and Florida R. (2004), Locating global advantage: industry dynamics in the international economy, Stanford University Press, Stanford Ca., Preface. Mankiw, G. N., and Swagel, P. (2006), ‘The politics and economics of offshore outsourcing’, Journal of Monetary Economics 53, 5: 1027-1056. Marin, D. (2006), ‘A new international division of labor in Europe: Outsourcing and offshoring to Eastern Europe’, Journal of the European Economic Association 4, 2-3: 612-622. Markusen, J. (1995), The Boundaries of Multinational Enterprises and the Theory of International Trade”, Journal of Economic Perspectives, 9 (2), pp. 169-89. OECD (2007), Globalisation and regional economies, Paris. OECD (various years ) Economic Outlook. Resmini L. (2007), “The determinants of FDI in Central, Eastern and South eastern Europe: lessons from the past and prospects for the future”, in Liebscher, K., Christl, J., Mooslechner, P., Ritzberger-Grünwald, D. (eds.), Foreign direct investment in Europe: a changing landscape, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. Richardson, H.W. (1967) Regional Economics, University of Illinois Press. Storper M., Chen Y. And De Paolis F. (2002), “Trade and the Location of Industries in the OECD and European Union”, Journal of Economic Geography, n. 2, pp. 73-107. Tavares, A.T. and Teixeira A. (2006), (eds.) Multinationals, Clusters and Innovation, Palgrave Mc Millan, New York. UNCTAD (various years) World investment report. 230

Regions Benefiting from Globalisation and Increased Trade: Final Report – 15 October 2009

Wai-chung Yeung H. (2009) “Situating Regional Development in the Competitive Dynamics of Global Production Networks: An East-Asian Perspective”, Regional Studies, Vol. 43, n. 3, pp. 325 – 351. Williamson J.G. (1965), “Regional Inequality and the Process of National Development: a Description of the Patterns”, Economic Development and Cultural Change, vol. 13, pp. 3-45. World Bank (various years ) World development report.

231