The impact of globalisation and increased trade liberalisation

“The impact of globalisation and increased trade liberalisation on European regions” IGEAT-ULB (Bruxelles) Politecnico Milano UMS-RIATE (Paris)

Political and scientific objectives of the project - To assess the regional impact of globalization inside EU - To identify the most vulnerable regions - To imagine policy options

I. The storyline Globalization and liberalisation of trade Potential threat on regions specialized in the most « competitive » Sectors. Increased international competition in some manufacturing sectors Potential economic impact : - Positive reaction; - Regional crisis Potential social impact : increased inequalities and social exclusion

The hypotheses 1 – There is a globalization process and its impact on regions is important (winning and losing regions) 2 – Regional specialization in the sectors which are the most vulnerable to globalization is a potential threat for the regions 3 – This can lead to bad economic performances and social degradation

General methodological approach Main difficulty: it is impossible to identify directly regional losers in globalization because the flows between the regions and the world are unknown Consequence: we will use a sectoral approach supposing that sectoral specialization in vulnerable sectors is a potential threat.

The different steps i. ii. To identify vulnerable sectors at the EU level To identify regions potentially vulnerable on the base of the economic structure iii. To cross this potential vulnerability with economic performances iv. To explain the regional diversity of economic performances by qualitative and quantitative analyses v. To explore the relationship with social indicators vi. To develop a prospective analysis vii. Political conclusions

i. Sectors vulnerable to globalization Two criteria in a static and dynamic approach: - Trade balance; - Openess rate to imports. Result : four vulnerable sectors 1°) Textile, clothing (DB) and footwear and leather (DC) 2°) Manufacturing of basic metals and fabricated metal products (DJ) 3°) Electrical and optical equipment (DL) 4°) miscellaneous manufacturing industries (DN).

Why are these sectors vulnerable to globalization? We interpret this by the concepts of life cycle of the products and spatial diffusion over time. Specialization in textile of exports in 1976 and 2006

The specialization in textile exports in the Euro -mediternanean space since 1967

Specialization in mechanic industry in 1976 and 2006

ii. Potentially vulnerable regions Regions specialized in the vulnerable sectors

Potentially vulnerable sectors are beyond a treshold in the vulnerable sectors Remark : NUTS 2/NUTS 3

iii. Regional vulnerability and economic performances We explored this link - At sectoral level - At global level - According to GVA or employment.

No clear correlation within the different vulnerable sectors

No correlation with global economic performances neither!

Differentiating by different European macro-regions, we found no simple relationship between the share of vulnerable sectors and global regional performances. Generally, the impact is negative but very unstable according to sectors and regions Impact of vulnerable sectors on GDP growth, with control for national trends (1995 -2004)

IV. How to explain this diversity of performances? First conclusion : inside the vulnerable regions, some are able to deal with the threat of globalization and others not Two approaches to explain this diversity : - Quantitative (econometric) - Qualitative (case-study)

Qualitative analysis 16 case studies all over Europe and in the three types of regions with winning and losing regions

Textile regions: Kortrijk area and Northern Portugal Kortrijk area* Population 2005 1100000 Northern Portugal 3732550 1995 132 (102) 63 (84) 2000 131 (104) 64 (81) 2005 123 (101) 60 (79) 2005 6, 2 (73) 8, 8 (115) Total 1995 17059 227964 Total 2005 14053 180043 share in 2005 7, 2 16, 6 Share of manufacturing industries 2005 27, 5 (128) 25, 1 (118) Share of high-level services 2005 24, 2 (93) 17, 6 (103) GDP/inhab. according to European average (and national average) Unemployment rate 2005, (and according to national average) Textile employment

Development paths - Historical inheritage: long textile tradition - Development in the last decades Kortrijk area becomes a typical marshallian district after WWII Northern Portugal has developed exogenously from the 80’s onwards - The limits of development Regions are hit by the textile crisis at the end of the 90’s (liberalization) Kortrijk area has been able to adapt, notably through a specialization in carpet production and some technological segments Economic performances are declining because of insufficient R&D, too specialized workforce… In Northern Portugal, the competition of cheaper regions is decisive and we observe a structural crisis (weakening of Porto).

Strengths and weaknesses In Portugal Exogenous control, limited technologies, limited qualification of the workforce, weakness of the metropolitan areas. But signs of diversification In Kortrijk (Wesetrn Flanders) Limits in technological upgrading Limitation in the qualification of the workforce Strong entrepreneurship within a socially coherent tissue High diversfication of the industrial tissue with endogenous control.

« Electronic » regions: Northern Finland West Hungary Oulu* Western Hungary 1995 96 (89) 48 (92) 2000 100 (85) 53 (95) 2005 99 (86) 54 (84) 2005 11, 1 (132) 7, 2 (100) Total 1995 6235 29378 Total 2005 9894 53578 share (%) in 2005 10, 4 6, 2 Share of other manufacturing industries 2005 29, 2 (104) 30, 3 (99) Share of high-level services 2005 17, 1 (96) 16, 6 (107) GDP/inhab. according to European average (and national average) Unemployment rate 2005, (and according to national average) Employment in the electric/electronic sector

Two different development trajectories - Firm size (the role of the big firms) and the question of endogeneity; - Embededness: why is Nokia embedded in Finland not in Hungary? The dependance on a big firm has to be qualified by the embededness of the firm and the local know-how which has been favored/created by its presence.

The strengths and weaknesses: 1. Workforce: “A core explanation to the resilience of the vulnerable sector is its deep knowledge specialization, generated through co-evolution of institutions providing poly-technical education, and corporate actors supporting on-the-job-training and life-long education” In Hungary, the cheap, but qualified workforce has been a major asset but certainly more vulnerable than in the Finnish case 2. Public policy : the example of Triple helix in Oulu: “the Triple Helix cooperation between the university, business and public sector in electronics, ICT and electric-related industries has played a fundamental role” Based on a long tradition (since the 50’s) and the decisive support of the central state. In Hungary, policies focused on tax exemptions.

Conclusions of the case studies • Factors related to firms: – Size of firms – Embeddedness (local networks; endogeneity…) – Innovation • Contextual factors: - Sectoral specialization – Quality of the workforce – entrepreneurship

Synthesis : a typology of risk according to globalization : geographical synthesis of comparative advantages in vulnerable regions Methodology: - From qualitative and quantitative analysis, we identify factors of success (economic structure, workforce, entrepreneurship…) - on this basis, we distinguish amongst potentially vulnerable regions those which are indeed vulnerable and those who are not

Synthesis : a typology of risk according to globalization : geographical synthesis of comparative advantages in vulnerable regions

V. Social performances among vulnerable regions Are vulnerable regions most affected in social terms? • The first hypothesis is that vulnerable regions have seen the labour market situation worsen because they have to face global competition more than other regions. • The second hypothesis is that the vulnerability to globalization will first hit the least qualified persons, and increase the gap between low and high skilled on the labour market.

Conclusions - Vulnerable regions do not significantly differ from the others in terms precariousness in the labour market. - Vulnerable sectors are ejecting more low qualified workers than the rest of the sector. - We observe significant differences between the types of vulnerable regions according to their capacity to integrate low qualified workers: the unemployment gap between low qualified and highly qualified is strongly reduced in textile regions, but higher than average in electr(on)ic regions.

Prospective analysis for policy • Statistical analysis on past evolutions • Policy targets the future • Need foresight on possible evolutions and their consequences • No prediction ! • Raising awareness – Driving forces – Possible levers for policy

Methodological choices • Future cannot be quantitatively predicted • Quantitative foresight offers – Explicit formalisation of often implicit assumptions – Focus on some selected cause-effect relationships – General directions and orders of magnitude • Quali-quantitatif scenarios • MASST – MAcroeconomic, Sectoral, Social and Territorial (MASST) model

MASST Model Politecnico di Milano

MASST Model Politecnico di Milano

Imagining strategies: BRIC

Imagining strategies: Member states

Imagining strategies: European Commission

Choice of scenario hypotheses

Baseline scenario for comparison

Comparing scenarios Scenario A • Almost all regions grow more than in the baseline • Rural vulnerable regions gain less • Vulnerable regions in Western Europe profit of their endowment in urban and tertiary structure • Vulnerable regions in the East generally outperformed by the – non-vulnerable – capital regions • In spite of shift in sectors, vulnerable regions lose more industrial employment => nonvulnerable regions seem to benefit from more tertiary and less specialized structure • vulnerable regions seem unable to replace industry with tertiary activities. Scenario B • Eastern vulnerable regions higher variance in terms of GDP growth than Western • Decisive loss in industrial employment growth in vulnerable regions; • Relative lower loss of service employment growth in vulnerable regions with respect to the others.

VII. Conclusions

Scientific conclusions • Sectoral structure not sufficient to detect vulnerability • No clear information about impact of globalisation and of globalised sectors on regions • Generic approach of regional development more efficient • Segments of production probably more decisive, but very difficult to measure • Situation quite different between GDP and employment

Political conclusions • Need to ensure regional “embeddedness” of firms to increase and prolong impact of their presence • Need to enhance region's capacities of profiting of the presence of large exogenous firms: – Policies to increase intensity and speed of spill -overs – Decisive investments in knowledge transfer, education, etc, often in a very short period of time

Political conclusions • Difficulty to politically create cluster structures • Need to support existing (SME) clusters in the development of more technological innovation • Importance of education, notably basic secondary eduction, for the capacity of a region to profit of opportunities • Need to maintain territorial capital in regions in decline

Thank you ! More information ? Gilles Van Hamme: gvhamme@ulb. ac. be Moritz Lennert: moritz. lennert@ulb. ac. be