Paweł Capik · Magdalena Dej Editors
Relocation of Economic Activity Contemporary Theory and Practice in Local, Regional and Global Perspectives
Relocation of Economic Activity
Paweł Capik Magdalena Dej •
Editors
Relocation of Economic Activity Contemporary Theory and Practice in Local, Regional and Global Perspectives
123
Editors Paweł Capik Keele Management School Keele University Keele UK
Magdalena Dej National Institute for Spatial Policy and Housing Kraków Poland
ISBN 978-3-319-92281-2 ISBN 978-3-319-92282-9 https://doi.org/10.1007/978-3-319-92282-9
(eBook)
Library of Congress Control Number: 2018943255 © Springer International Publishing AG, part of Springer Nature 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Contents
1
Spatial Organisation and Movements of Firms and Industries: The Role of Firm and Place in (Re)Location Research . . . . . . . . . . Paweł Capik and Magdalena Dej
Part I 2
3
4
5
7
Spatial Reconfigurations of Firms and Industries
Between Location, Persistence and Relocation—Early Historical and Spatial Trajectories of the Polish Aviation Industry . . . . . . . . Lech Suwala
15
The Relocation of the Automotive Industry in Brazil and Mexico: Between Corporate Strategies and Industrial Policies . . . . . . . . . . . Mirosław Wójtowicz
33
Industrial Upgrading and Downgrading of Export-Oriented Furniture Firms in the Market Reorientation Towards China’s Domestic Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tianlan Fu and Chun Yang The Changing Firm Landscape and Firm Location Behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sierdjan Koster and Piet Pellenbarg
Part II 6
1
53
71
Firm Relocation Factors and Consequences
Relocation of Economic Activity Within Major Metropolitan Areas in Poland—Scale and Key Characteristics . . . . . . . . . . . . . . Magdalena Dej, Wojciech Jarczewski and Michał Chlebicki
87
Identifying Factors Relevant for Firms’ Location and Relocation. The Case of Ticino . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Federica Rossi
v
vi
Contents
8
The Effects of the 2007 Global Economic Crisis on Firm Relocation Factors: SME Movements from Greece to Bulgaria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Nikos Kapitsinis
9
Location of R&D Abroad—An Analysis on Global Cities . . . . . . . . 145 Davide Castellani and Katiuscia Lavoratori
10 Delocalisation of Enterprises: Qualitative and Quantitative Effects on the Labour Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Ewa Małuszyńska 11 Epilogue: Relocation Factors and Consequences—Towards a Multidisciplinary Research Agenda . . . . . . . . . . . . . . . . . . . . . . . 183 Magdalena Dej and Paweł Capik
Editors and Contributors
About the Editors Paweł Capik lectures in international business at Keele University (UK). His research explores interdependencies between international business strategies and operations and regional development theory and practice, particularly in the context of emerging and transition economies. Paweł’s research areas include inward investment attraction and place marketing, sourcing and shoring strategies, large and small firms’ internationalization and location motivations and their impacts on regional and local economies. Magdalena Dej is a geography researcher at the National Institute for Spatial Policy and Housing (Poland). She received her Ph.D. from Jagiellonian Univerisity and specializes in economic local and regional development, particularly the development of rural areas. She authored and coauthored several books, numerous reports and research papers on the impacts of large companies on local communities as well as on rural areas and their relationship with major metropolitan areas and labor markets. Magdalena led and participated in several international research projects financed by the European Union Interreg IVC, International Visegrad Fund, Polish-German Foundation for Science and National Science Centre, Polish Ministry of Science and Higher Education.
Contributors Paweł Capik Keele Management School, Keele Univeristy, Keele, UK Davide Castellani Henley Business School, University of Reading, Reading, UK Michał Chlebicki Faculty of Philosophy and Sociology, University of Warsaw, Warsaw, Poland
vii
viii
Editors and Contributors
Magdalena Dej National Institute for Spatial Policy and Housing, Kraków, Poland Tianlan Fu School of Geography and Planning, Sun Yat-sen University, Guangzhou, China Wojciech Jarczewski National Institute for Spatial Policy and Housing, Kraków, Poland Nikos Kapitsinis Cardiff University, Cardiff, UK Sierdjan Koster Economic Geography, Faculty of Spatial Sciences, University of Groningen, Groningen, The Netherlands Katiuscia Lavoratori University of Perugia, Perugia, Italy Ewa Małuszyńska Poznań University of Economics and Business, Poznań, Poland Piet Pellenbarg Economic Geography, Faculty of Spatial Sciences, University of Groningen, Groningen, The Netherlands Federica Rossi Institute of Economic Research, Università della Svizzera italiana (USI), Lugano, Switzerland Lech Suwala Institute of Geography, Humboldt-University of Berlin, Berlin, Germany; Institute of Urban and Regional Planning, Technical University Berlin, Berlin, Germany Mirosław Wójtowicz Institute of Geography, Pedagogical University of Kraków, Kraków, Poland Chun Yang Department of Geography, Hong Kong Baptist University, Kowloon Tong, Hong Kong
Abbreviations
BMW BW COP CREDO CWL DB DMI DUNS DWL EMCC EOI EU FDI GEC GL GPN GVA GVC HGV HWS IPI ISI KMSPW
LCV LOPP
Bayerische Motoren Werke Brandenburgische Werke, Warszawa-Paluch Centralny Okręg Przemysłowy (Central Industrial District) Company Relocation and Economic Development Observatory Centralne Warsztaty Lotnicze (Central Aviation Workshops) Daimler-Benz Data Marketing Information Data Universal Numbering System Doświadczalne Warsztaty Lotnicze (Experimental Aviation Workshops) European Monitoring Centre on Change Export-Oriented Industrialisation European Union Foreign Direct Investment Global Economic Crisis GL Francke Werke, Warszawa Global Production Networks Gross Value Added Global Value Chains Heavy Goods Vehicle Harcerskie Warsztaty Szybowcowe, Warszawa Industrialised Product Tax Import-Substituting Industrialisation Koła Mechaników Studentów Politechniki Warszawskiej (Engineering Students’ Circle of the Warsaw Technical University) Light Commercial Vehicle First Liga Obrony Powietrznej Panstwa (League of National Air Defence, later Liga Obrony Powietrznej i Przeciwgazowej (The Airborne and Antigas Defence League)
ix
x
LWL LWS MERCOSUR MNE MNV NAFTA NETS NSTK OECD P&L PWS PZ inz. PZL PZL-WP1 PZL-WP2 PZL-WS1 PZL-WS2 PZS SMEs SWS VAI VAT WLK WLL WWS WWSS ZASPL
Abbreviations
Lwowskie Warsztaty Lotnicze (Lwów Aviation Workshops) Lubelska Wytwórnia Samolotów (Lublinian Aeroplane Plant) Mercado Común del Sur—Southern Common Market Multinational Enterprise Mechanische Neubrandenburgische Werkstätten, Warszwa North American Free Trade Agreement National Establishment Time-Series Nationalsozialistisches Fliegerkorps Organisation for Economic Co-operation and Development Zakłady Mechaniczne Plage i Laśkiewicz (Plage i Laśkiewicz Engineering Establishments) Podlaska Wytwórnia Samolotów (Podlasian Aeroplane Plant) Państwowe Zakłady Inżynierii, Warszawa (National Engineering Works) Państwowe Zakłady Lotnicze (National Aviation Establishments) PZL Wytwórnia Płatowców nr 1, Warszawa (Aircraft Plant No. 1) PZL Wytwórnia Płatowców nr 2, Mielec (Aircraft Plant No. 2) PZL Wytwórnia Silników nr 1, Warszawa (Aircraft Engine Plant, No. 1) PZL Wytwórnia Silników nr 2, Rzeszów (Aircraft Engine Plant, No. 2) Polskie Zakłady Skody (Polish Skoda Works) Small and Medium Enterprises Śląskie Warsztaty Szybowcowe Value Added in Industry Value Added Tax Warsztaty Lotnicze Krakowskie (Kraków Aviation Workshops) Warsztaty Lotnicze Lwówskie (Aviation Workshops in Lwow, not to be confused with LWL) Wojskowe Warsztaty Szybowcowe, Kraków Wielkopolska Wytwórnia Samolotów “Samolot”, Poznań (Wielkopolan Aeroplane Plant ‘The Aeroplane’) Związek Awiatyczny Studentów Politechniki Lwowskiej (Lwow Polytechnic Student Aviation Association)
List of Figures
Fig. 2.1 Fig. 3.1 Fig. 3.2 Fig. 4.1 Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig.
4.2 5.1 5.2 5.3 6.1 6.2 6.3 6.4
Fig. 6.5 Fig. 7.1 Fig. 7.2 Fig. 7.3 Fig. 9.1
Polish aviation works genealogy 1912–1944 . . . . . . . . . . . . . . Development of the automotive industry in the selected Brazilian states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Development of the automotive industry in the selected Mexican states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Upgrading trajectories of export-oriented firms in their market reorientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Location of the Pearl River Delta in China . . . . . . . . . . . . . . . Firm establishments in the Netherlands by type, 1996–2014 . . Types of premise of employing firms and solo firms . . . . . . . . Employing establishments by type of premise by urbanity . . . . Metropolitan areas of Poland’s largest urban centres . . . . . . . . Relocations of companies in the study in 2001–2013 . . . . . . . . Number of relocations of companies in the study. . . . . . . . . . . Percentage of relocations and year of operation of companies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Year of operation and first relocation . . . . . . . . . . . . . . . . . . . . Graphical representation identifying weaknesses and strengths of the territory . . . . . . . . . . . . . . . . . . . . . . . . . . Graphical representation identifying weaknesses and strengths of Canton Ticino . . . . . . . . . . . . . . . . . . . . . . . . Percentage of firms relocating to Ticino from Italy, for which the factors are possible drivers . . . . . . . . . . . . . . . . . Geographical distribution of R&D and DDT FDIs, by destination area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
..
24
..
41
..
45
. . . . . . . .
. . . . . . . .
56 58 73 80 80 95 97 98
.. ..
101 102
..
114
..
117
..
119
..
153
xi
List of Tables
Table 2.1 Table 3.1
Table 3.2 Table 3.3 Table 3.4
Table 3.5 Table 3.6 Table 4.1 Table 4.2 Table 6.1 Table 6.2
Table 6.3 Table Table Table Table
6.4 6.5 6.6 6.7
Process- and path-based approaches towards the evolution of economic landscapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evolution of employment and coefficient of redistribution in particular sections of the automotive industry in Brazil: 1999–2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Participation of selected states in total employment of the Brazilian automotive industry: 1999–2014 . . . . . . . . . . . . . . Participation of selected states in value added in industry (VAI): 1999–2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evolution of employment and coefficient of redistribution in particular sections of the automotive industry in Mexico: 1999–2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Participation of selected states in total employment of the Mexican automotive industry: 1999–2014 . . . . . . . . . . . . . . Participation of selected states in the gross value added (GVA): 1999–2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Domestic sales of furniture in Guangdong Province during 2007–2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Domestic sales strategies and upgrading or downgrading trajectories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Exemplary types of company migrations . . . . . . . . . . . . . . . Structure of the analysed sample compared to structure of companies as a whole according to type of business, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Percentage of relocations among all analysed companies based on year of foundation . . . . . . . . . . . . . . . . . . . . . . . . . Relocation and age of company . . . . . . . . . . . . . . . . . . . . . . Distribution of pairs of periods of individual companies . . . Company size and relocations . . . . . . . . . . . . . . . . . . . . . . . Polish Classification of Activities (PCA) sector and relocations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
..
18
..
38
..
40
..
42
..
43
..
44
..
46
..
60
.. ..
60 93
..
99
. . . .
. . . .
100 101 103 103
..
104 xiii
xiv
List of Tables
Table 7.1 Table 8.1 Table 8.2 Table 9.1 Table 9.2 Table Table Table Table
9.3 9.4 10.1 10.2
Statistical mode values of the firms’ evaluations on relevance and presence/absence of territory characteristics in Ticino . . Significance of elements that affected firm exit from Greece . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Significance of elements that attracted firms in Bulgaria . . . Geographical distribution of R&D/DDT projects and global cities, by destination area . . . . . . . . . . . . . . . . . . . . . . . . . . . Geographical distribution of R&D/DDT projects, by origin area of MNEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Variable list and description . . . . . . . . . . . . . . . . . . . . . . . . Results of conditional and mixed logit models . . . . . . . . . . Classification of sourcing . . . . . . . . . . . . . . . . . . . . . . . . . . . Announced jobs destroyed and created as a result of restructuring in EU28 and Norway between 2002 and 30.11.2017 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
..
116
.. ..
130 136
..
154
. . . .
. . . .
154 156 158 167
..
175
Chapter 1
Spatial Organisation and Movements of Firms and Industries: The Role of Firm and Place in (Re)Location Research Paweł Capik and Magdalena Dej
With the onset of globalisation and ever more progressive liberalisation of international trade and investment regimes, the spatial distribution of firms and economic activities within countries and worldwide is continuously changing. In recent decades, we have often witnessed that what starts as a brave new strategy to ascertain a firm’s competitive advantage soon becomes a global relocation trend, only to evolve again and begin movement in the reverse or yet another direction. Outsourcing and offshoring-driven relocations provide the best global examples of such processes. At the regional and local scale, the fall and successful re-emergence of some of old industrial regions, particularly but not exclusively in Europe, offer relevant cases. While the local, regional and national circumstances are important, what happens in any location is increasingly influenced by its role in and the relationship to the global systems of production, trade and consumption. Rarely has it been more pronounced than at present, when the global economy is yet to fully recover from the late 2000s financial crisis, Brexit is pushing both manufacturing and financial sectors to reconsider its UK bases and national and regional policies offer generous incentives for firms willing to settle within multiple special economic zones world over. Such a dynamic environment presents a unique opportunity for research into new and emerging relocation processes, with the outlook to review and advance the theory underpinning the practice. Location of economic activity traditionally has been at the forefront of economic geography and regional science research. Increasing mobility of globalising firms and industries requires fresh evaluations of motivations, impacts and policy responses guiding their location and relocation choices. This book offers a P. Capik (&) Keele Management School, Keele Univeristy, Keele, UK e-mail:
[email protected] M. Dej National Institute for Spatial Policy and Housing, Kraków, Poland e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_1
1
2
P. Capik and M. Dej
comprehensive, multidisciplinary and multi-method investigation of the phenomenon. It explores current trends in the location and relocation of economic activities, investigates factors behind these processes and examines their developmental consequences at the local, regional and global scale in developed and emerging economies. This opening chapter introduces the debate on the location and relocation of economic activities, its different facets including factors, consequences and theoretical foundations discussed in the initial sections, and leads to the identification of the rationale underpinning the development of this book. The latter parts of the chapter are dedicated to the overview of the volume and the summary of arguments presented in individual chapters.
1.1
Theory in (Re)Location Research
Over time, the studies of the location and relocation of economic activities have led to the development, testing and refinement of numerous theoretical concepts and approaches. While location remains the dominant focus of the majority of research, relocation studies increasingly appear to offer insightful perspectives complementing and adding to the wider context. Consequently, while it is not possible to talk about relocation of firms and economic activities outside the theoretical realms of “location”, it is increasingly difficult to talk about location without paying attention to emerging “relocation” theory and practice. The following paragraphs provide an attempt to trace the conceptual evolution of these two related processes. In the early nineteenth century, Johann Heinrich von Thünen, the precursor of location theory, examining agricultural production considered the principal question of “why does particular activity take place in a particular location?” Von Thünen’s calculations and modelling allowed him to observe the link between land use and changes of rent related to the distance from the central marketplace. With the growing complexity of economies, his insights have been adopted and applied beyond agriculture and shaped early understandings of the spatial differentiation of land use in cities. Building on von Thünen’s ideas, in 1960 William Alonso suggested that in the city different economic actors (residents, firms) experience different rent curves for different land uses. Rent curves for some activities, for example a prestigious office or retail space, are considerably steeper and fall rapidly away from the central location, than for others (e.g. housing). Next, expanding industrialisation and growing spatial differences laid the foundations for an early work on industrial location. In the late 1920s, Alfred Weber endeavoured to explain the location patterns of industries in order to “disclose the causal relationship between them and those large displacing processes which we observe” (Weber 1929: 3 in Wood and Roberts 2011: 20). Relating industries to ubiquitous and localised resources they relied on, Weber explained the location of firms within the manufacturing sector with the least cost model. Weber observed that a manufacturing location is driven by transport cost (the function of
1 Spatial Organisation and Movements of Firms and Industries …
3
weight to distance ratio) of (1) raw materials from their source to the manufacturing plant, and (2) of the finished products to the market. Further, Weber examined the importance of labour costs and argued their influence on location is determined by labour intensity of the manufacturing process and labour cost relative to the materials used (i.e. labour coefficient). Weber’s work laid foundations for the future theorising of the role of agglomeration and subsequently clustering of economic activities. Changing the focus from production to consumption of goods in the early 1930s, Walter Christaller proposed the idea of “central places”. Christaller attempted to explain the spatial patterns of distribution of towns and cities, and factors conditioning it. He argued for a hierarchy of urban places, where larger cities of a “higher order” are the central places with functional relationships to their hinterlands and “lower order”, smaller towns. Central places are fundamentally retail centres providing “central goods”—material or intangible services. Each of them is characterised by a specific threshold or minimum demand needed to maintain its provision and its own range, i.e. the distance up to which the consumer is willing to purchase a good offered in a central place. Next, building on Weber and Christaller’s ideas, in the late 1930s August Lösch set to construct a location model starting with individual decision makers. Lösch argued that the optimal location is a matter of individual firm selecting the “right” location, i.e. the one, where the net profit is the highest. Lösch recognised that the net profit is influenced by a great many factors and expanded his analysis beyond location to include economic regions and trade. The growth and expansion of multinational enterprises (MNEs) in the second half of the twentieth century required new explanations, provided new research opportunities and in that sense laid the foundations for the emergence of new academic disciplines of corporate geography and international business. Building on broad-based traditions in economics, management, strategy and geography, new research began to explore firms’ cross-border expansion, and soon attention shifted to locational adjustments, firm closures and relocations. The process of location adjustment was commonly examined in two perspectives. The first sought to explain adjustment in terms of broad structures such as global competition in product markets, technological change and evolving regulatory systems. The second was concerned with industrial restructuring, structural changes and development (Wood and Roberts 2011: 15–65). Traditionally, location theory was rooted in studies inspired by industrial revolution. However, since mid-twentieth century, a set of new theoretical concepts has emerged, explaining different aspects of the mobility of firms. For example, Vernon’s product life-cycle combined innovation, cost-efficiency of production and demand levels to explain spatial movements of firms. Subsequently, behavioural approaches put emphasis on individual decision makers, their bounded rationality underpinning the perceptions of new locations, and wider non-economic factors (e.g. family situation) guiding the choice of new locations (see further discussion in Koster and Pellenbarg in this volume). Next, exploring the nature of a multinational firm, Dunning proposed an eclectic paradigm combining ownership, localisation
4
P. Capik and M. Dej
and internalisation (OLI) advantages to explain its expansion. More recently, fragmentation and decoupling of MNEs’ value adding activities led to the development of approaches within global production networks (GPN) and global value chains (GVC), both concerned with the spatial organisation of the firm and the governance of value creation (see for example Wójtowicz and Fu and Yang in this volume). Consequently, the theoretical approaches exploring and explaining the location and relocation of firms and industries have been, and indeed remain to be, located within broadly defined spatial sciences and management. More precisely, they are rooted within economic geography and regional development studies, and organisational behaviour, strategy and international business. Naturally for the first set of disciplines, at the centre of theoretical advancements is the focus on place—its characteristics and consequences created by new firms arriving and old ones leaving. Theories emerging from the other group of disciplines concentrate on the motivations driving firms to and from locations, and outcomes of their movements. Such theoretical richness and disciplinary variety are also reflected in this book, where the different perspectives of relocation are advanced through both empirical and conceptual discussion.
1.2
Location and Relocation—Firm and Place Factors
Firm relocation involves either the movement of a company from one location to another (i.e. complete) or an expansion into locations other than the current one (i.e. partial). Relocations comprise locational adjustments to changes in markets, regulations and technology, and consequently, a range of factors has been identified to explain the direction, scope and scale of such movements of firms. While the relative importance of individual factors varies, is firm-specific and reflective of its characteristics and strategy, overall they can be grouped into two broad categories —firm-related and location-related. The former comprise firm’s main characteristics including size, age, ownership mode, the sector of operations, as well as the growth and market expansion strategy, the organisational model and the experience of previous relocations (Pellenbarg and Knoben 2012; see also chapters in the Part II of this volume). Discussing the latter, Kim and Aguilera (2016) and Nielsen et al. (2017) identify the basic characteristics of a destination location as key drivers attracting MNEs. These include transport and communication infrastructures, the demand market size and potential, productivity and wage levels, corporate and other local tax rates, the level of development of formal institutions, and the availability of human capital in terms of labour force, education and skills. Conversely, poor availability of such factors in the current location may provide a push encouraging the firm to look for alternative locations (Fløysand et al. 2017). However, against a long tradition of place-focused research discussed above, recent views appear to suggest that location (as a place) does not matter any longer, and the importance of space and place is said to have become irrelevant.
1 Spatial Organisation and Movements of Firms and Industries …
5
Technology and communication advancements allow mobile companies to evolve into increasingly footloose sectors disregarding the bonds that tie them to places. Equally, long-term locational advantages of places, arguably, are only maintained for short time, as firms move from one destination to another and benefit from proliferation of increasingly similar alternative locations. As this volume demonstrates, such assertions are considerably premature. The analysis presented clearly points out that location remains to be one of the most influential factors determining firms’ profitability, shaping the complexity and reach of their supply chains and markets, and ultimately deciding about their success or failure. In that sense, the firm is always localised, as are the consequences of its spatial behaviours. From a place perspective, “global” processes are actually stretched “local to local” processes, as they unfold in locations with particular and unique characteristics. Firms and industries locate in places due to a complex mix of individual factors identified above (Murray and Overton 2015). Certain firms and activities emerge and remain in particular locales, while others move to agglomerate (cluster) in specific locations. Still others continue to migrate from one destination to the next (e.g. offshoring, nearshoring), while after such explorations some return to their original locations (i.e. backshoring) (Capik 2017). All such patterns, strategies and behaviours behind the search for the best location stress the importance of places, their characteristics and suitability for particular firm and the type of economic activity. Equally, all such movements of individual firms and entire sectors create diverse consequences for home and host locations alike (Dicken 2015)—the key focal points of this volume.
1.3
Focus and the Scope of the Book
This volume provides a thorough analysis of current trends in the location and relocation of economic activity globally, regionally and locally. It explores dominant trends in the mobility and relocation of firms and industries and examines the factors guiding such trends and evaluates their consequences in both developed and emerging economies in Asia, Europe and Latin America. Using robust empirical material, the book offers a multidisciplinary, comprehensive overview, critique and extension of long-established theories underpinning patterns of firm (re)location. The latest global developments—including the recession of the late 2000s, changes in the intensity of economic cooperation within regional trading blocks and resulting from it shifting strategies of multinational enterprises leading to the reconfiguration of global value chains, as well as changing patterns of the use of urban and rural space—call for the re-examination of established theories and research approaches. The book integrates the existing state of knowledge about the relocation of businesses to the fullest extent. It presents the results of the latest research conducted at different spatial scales and includes different national and local contexts. The volume brings together contributions presented during special sessions at Regional Studies Association 2015 Annual Conference in Piacenza (Italy) and
6
P. Capik and M. Dej
developed in the course of the research project “The processes of relocation and spatial expansion of business in metropolitan areas” financed by Poland’s National Science Centre (SONATA grant competition no. 192273, grant period 2012–2017). The book is concerned with a set of wide-ranging issues about relocation of firms and industries, the factors enabling and fostering such movements and the consequences for home and host economies at the local, regional, national and international level. In broad terms, the book deals with the following set of questions: How do historical developments and current affairs shape the economic geography of manufacturing production, service provision and consumption? Why do firms begin to move? How does an isolated relocation event turn into a global process engulfing the entire sectors of economic activity? What are the factors driving relocations of companies? To what extent can they be the predictors of the mobility of industries? What attracts firms to particular locations and detracts them from setting up in other places? What are the short and long-term consequences for firms and places alike? To what extent can they be shaped and corrected by policy? What is the role of policy in the relocation of an individual firm and economic sectors? What scale of policy intervention appears most relevant? How does the “scale” of analysis impact relocation research?
1.4
Book Structure and Overview
The individual chapters are assembled into two distinctive, complementary sections. Contributions in the first section explore major, global trends in the location and relocation of industries and firms in Asian, European and Latin American countries. The second part of the book is dedicated to the discussion of location and relocation factors and the consequences of firms’ movements. Using both quantitative and qualitative methodologies, individual chapters examine the importance of such factors as geography, labour market characteristics and the access to finance in national and cross-border relocations, as well as their spatial and socio-economic consequences for home and host-locales.
1.4.1
Part 1
The opening section of the book, Spatial Reconfigurations of Firms and Industries, comprises four chapters focusing on relocation trends within diverse global industries. In “Between Location, Persistence and Relocation—Early Historical and Spatial Trajectories of the Polish Aviation Industry”, Lech Suwala (HumboldtUniversity of Berlin, Germany) explores the trends dominating the Polish aviation industry through reconstructing its historical and spatial trajectories by means of
1 Spatial Organisation and Movements of Firms and Industries …
7
enterprise genealogy. “Location, persistence and relocation”, contextualised through process* and path-based approaches, provide main framework underpinning the analysis of crucial events in the shifts of diverse parts of the sector. Based on extensive review of archive documents and series of interviews, Suwala offers a fascinating account of historical spatial shifts occurring within the Polish aviation industry in its early years. The analysis demonstrates the necessity of merging process* and path-based approaches in the development of a comprehensive inquiry into relocations of industrial sectors. In the second chapter, “The Relocation of the Automotive Industry in Brazil and Mexico: Between Corporate Strategies and Industrial Polices” Mirosław Wójtowicz (Pedagogical University of Kraków, Poland) applies the Global Value Chains framework to chart the developments and spatial reconfigurations of the automotive industry in two largest markets of Latin America. Through the analysis of regional employment and gross value added data, the author illustrates the transformation of the sector from a nationally isolated to globally integrated one, and explores the role of sub-national scale in the process. Wójtowicz examines the interface between corporate strategies and regional industrial policy over the period of three decades and demonstrates how the two can reinforce each other to benefit places, particularly when confronted with global competitive pressures and liberalisation of trade and investment regimes. The following chapter, “Industrial Upgrading and Downgrading of Export-Oriented Furniture Firms in the Market Reorientation Towards China’s Domestic Market”, offers a somewhat reversed perspective. Tianlan Fu (Sun Yat-sen University, China) and Chun Yang (Hong Kong Baptist University) analyse the influence the demand-driven shift from global to national markets has on upgrading and downgrading behaviours of Chinese furniture producers. In that sense, the authors examine a peculiar type of relocation, namely a situation where the firms remain static but the spatial range of their markets changes as it moves from global to national. Using national, province* and city-level industrial statistics and interview data, the authors identify a number of trajectories taken up by the furniture companies trying to find themselves in a new, post-crisis environment dominated by the domestic demand. Product, process and functional upgrading take places at different pace determined by firm’s individual circumstances and resources, as well as the access to knowledge and capital from state business-support agencies. The analysis demonstrates the importance of emerging economies in (re)shaping the upgrading trajectories of export-oriented suppliers within global supply chains. Fu and Yang conclude with business support and regional policy recommendations, particularly fitting the emerging countries looking to abandon export-oriented growth in favour of developing internal market capacity. In the final chapter of the section, “The Changing Firm Landscape and Firm Location Behaviour”, Sierdjan Koster and Piet Pellenbarg (University of Groningen, The Netherlands) suggest that a small firm relocation reflects the owner’s personal situation rather than the factors in its external environment. The authors question the established logic in current literature that the location is chosen with the best interest of the firm in mind, and instead suggest it is the owner/
8
P. Capik and M. Dej
decision maker’s personal choice, family situation and a broader relational perspective which provides a more accurate context for understanding relocation decisions. Using the setting of the ongoing “Dutch entrepreneurial revolution”, and inspired by the household approach in migration studies, Koster and Pellenbarg suggest that alternative income sources in the owner/decision maker’s household may be important determinants of the firm location. Consequently, the relocation of the firm can then be used to address the “locational puzzle” to optimise the access to all relevant sources of household income. The authors observe that firm location is a part of a wider social bargaining process and goes beyond the attempts to match the firm and location characteristics. The argument is strengthened further with the analysis of business premises used by small firms, which reveals a considerable overlap between home/business locations. Koster and Pellenbarg conclude by identifying policy directions which should consider this new entrepreneurial landscape. With a decreasing size of firms and the growing numbers of jobs in home-based businesses, residential quality but also accessibility and web-infrastructure can be important triggers for local entrepreneurship.
1.4.2
Part 2
The second part of the book is dedicated to the examination of current practice, and in particular, it explores economic and strategy-based factors behind firm and industry relocation at a local, national and global scale. It also examines the diverse impacts and consequences the firm and sector relocation produces in the “new” and “old” locales. In the chapter on “Relocation of Economic Activity Within Major Metropolitan Areas in Poland—Scale and Key Characteristics”, Magdalena Dej, Wojciech Jarczewski (both at the National Institute for Spatial Policy and Housing, Kraków, Poland) and Michał Chlebicki (University of Warsaw, Poland) take a micro-perspective to identify factors driving the mobility of firms within Poland’s largest cities. The authors review a number of theoretical concepts developed in the course of early research undertaken in Western Europe and the USA, including behavioural theories, spatial margins on profitability, global production networks and incubation hypothesis, to argue the suitability of a multi-framework approach to study the relocation of firms in the Central-Eastern European context. Using a robust, purpose-created dataset, Dej et al. explore the influence firms’ size, age and sector of operation have on the distance and direction of their relocation within a metropolitan area. The age of the firm positively shapes the firm’s propensity to change location, as a higher number of older firms experience relocations than the new ones. Also, older businesses relocate over greater distances than the newer ones. Equally, larger companies are more likely to change their location than the smaller ones, but it is the latter type whose relocations cover greater distances. Firms operating within business service sectors relocate more often than those in manufacturing, but manufacturing firms tend to move further away. The authors
1 Spatial Organisation and Movements of Firms and Industries …
9
conclude by stressing the significance of their findings for the spatial revival and economic restructuring of Polish metropolitan areas. In “Identifying Factors Relevant for Firms’ Location and Relocation. The Case of Ticino”, Federica Rossi (Università della Svizzera italiana, Lugano, Switzerland) explores the motivations behind foreign and domestic firms’ migration within one country and across borders. Relying on regional economic theory and using original survey data, Rossi first detects a set of characteristics determining the attractiveness of places (pull factors) and then identifies a set of drivers of firms’ relocation decisions (push factors). Institutional and competitive environments, accessibility and access to resources and inputs, all stand amongst the most important general factors shaping the attractiveness of places to businesses. It is, however, a set of specific drivers which shape the relocation behaviours of firms in the Swiss canton of Ticino—favourable fiscal conditions, political stability and limited bureaucracy, access to funding and R&D infrastructure, as well as the quality of life, the most important amongst them. Rossi concludes that clear identification of both sets of factors provides the policy makers with obvious directions for future development in areas relevant for business, as wide ranging as educational system, urban planning and infrastructure development. Next, in “The Effects of the 2007 Global Economic Crisis on Firm Relocation Factors: SME Movements from Greece to Bulgaria”, Nikos Kapitsinis (University of Cardiff, UK) argues that in the context of the crisis-driven transformations firm relocation factors should not be presupposed, as global economic crisis changes the levels of their significance. The author considers mobility factors similar to those discussed by Rossi, but adopts a longitudinal perspective. Relying on a survey complemented with a series of interviews, Kapitsinis explores how the recent global economic crisis has reshaped the perceptions and fostered the revaluation of relocation factors by small- and medium-sized firms. Factors related to the general economic and institutional environments (e.g. access to finance, local demand, high levels of bureaucracy and legislative perplexity) appear to have increased in significance in the aftermath of the crisis, while the importance of others (e.g. labour costs) has dropped. The resulting conclusion is that any relevant policy intervention should concentrate on wage increase, stability of business environment, technological upgrade and, most importantly, restrict corruption and bureaucracy. In the subsequent chapter, “Location of R&D Abroad. An Analysis on Global Cities”, Davide Castellani (Univeristy of Reading, UK) and Katiuscia Lavoratori (University of Reading, UK) analyse the emerging trends in the co-location of multinational enterprises’ research and development and production activities in global cities. The analysis is set within the context of an apparent paradox. On the one hand, MNEs separate and disperse their value adding activities across locations offering most favourable conditions (external factors). On the other hand, internal factors such as tacit knowledge and the requirement of proximity to final users motivate firms to locate parts of their value chain (e.g. R&D and production) in spatial proximity. The chapter contributes to current literature by furthering the understanding of the international location of R&D function through the analysis of external and internal factors within a multi-country and multi-sector context. The
10
P. Capik and M. Dej
authors base their analysis on fDi Markets database of R&D greenfield investments in nearly 1.5 thousand cities worldwide. Castellani and Levoratori observe that the size and the connectivity of a city are important factors shaping R&D location choices. In particular, R&D appears to be attracted by a combination of good connectivity and moderate congestion, combined with vibrant environment and high quality of life. Further, confirming the role of intra-firm linkages between R&D and production, the authors observe that previous production activities of the MNE increase the probability of it co-locating R&D in the same city. Finally, in “Delocalisation of Enterprises. Qualitative and Quantitative Effects on the Labour Market”, Ewa Małuszyńska (Poznań University of Economics and Business, Poland) focuses on job loss and job creation as the main direct impact of firms’ delocalisation. In this conceptual chapter, the author argues that it is important to unambiguously define the extent and scope of delocalisation of firms’ activities before attributing any consequences to the process, particularly those employment-related. Next, Małuszyńska reviews different methods of assessing labour market impacts relaying either on firm-level data or estimations based on trade-related proxies. Two main consequences of firm delocalisation are observed. First, the changes in the level of employment and the related impact on wages within the locality/industry and proximate localities/industries. Secondly, the changes in the structure of employee skills resulting from fluctuations in demand for skilled and unskilled labour. The review of recent firm delocalisations and labour market changes in the wake of May 2004 EU expansion illustrates earlier observations. Małuszyńska concludes that the influence of delocalisation on the number of jobs available may be either positive or negative depending on the adopted time horizon, the level of development of home and host economies, the choice of delocalisation destinations, the level of sophistication of the relocated production, and the ownership and organisation structure of the firm concerned.
References Capik, P. (2017). Backshoring: Towards international business and economic geography research agenda. Advances in International Management, 30, 141–155. Dicken, P. (2015). Global shift—Mapping the changing contours of the world economy (7th ed.). London: Sage. Fløysand, A., Njøs, R., Nilsen, T., & Nygaard, V. (2017). Foreign direct investment and renewal of industries: Framing the reciprocity between materiality and discourse. European Planning Studies, 25(3), 462–480. Kim, J. U., & Aguilera, R. V. (2016) Foreign location choice: Review and extensions. International Journal of Management Reviews, 18(2), 133–159. Murray, W. E., & Overton, J. (2015). Geographies of globalisation. Abingdon: Routledge. Nielsen, B. B., Asmussen, C. G., & Weatherall, C. D. (2017). The location choice of foreign direct investment: Empirical evidence and methodological challenges. Journal of World Business, 52 (1), 62–82.
1 Spatial Organisation and Movements of Firms and Industries …
11
Pellenbarg, P. H., & Knoben, J. (2012). Spatial mobility of firms. In A. Frenkel, P. Nijkamp, & P. McCann (Eds.), Societies in motion: Innovation, migration and regional transformation (pp. 71–113). Cheltenham: Edward Elgar. Weber, A. (1929). Alfred Weber’s theory of the location of industries. Chicago: University of Chicago Press. Wood, A., & Roberts, S. (2011). Economic geography: Places, networks and flows. Abingdon: Routledge.
Paweł Capik lectures in International Business at Keele University (UK). His research explores interdependencies between international business strategies and operations and regional development theory and practice, particularly in the context of emerging and transition economies. Paweł’s research areas include inward investment attraction and place marketing, sourcing and shoring strategies, large and small firms’ internationalisation and location motivations and their impacts on regional and local economies. Magdalena Dej is a Geography Researcher at the National Institute for Spatial Policy and Housing (Poland). She received her Ph.D. from Jagiellonian University and specialises in economic local and regional development, particularly the development of rural areas. She authored and co-authored several books, numerous reports and research papers on the impacts of large companies on local communities as well as on rural areas and their relationship with major metropolitan areas and labour markets. Magdalena led and participated in several international research projects financed by the European Union Interreg IVC, International Visegrad Fund, Polish-German Foundation for Science and National Science Centre, Polish Ministry of Science and Higher Education.
Part I
Spatial Reconfigurations of Firms and Industries
Chapter 2
Between Location, Persistence and Relocation—Early Historical and Spatial Trajectories of the Polish Aviation Industry Lech Suwala
2.1
Introduction
The recent surprising retroactive withdrawal of a helicopter deal initially won by the AIRBUS Group SE on the part of the new conservative Polish government caused a massive political and media response. Whereas media portrayed different views according to their alignment drawing a mixed picture, AIRBUS addressed an open letter of bewilderment to Polish Prime Minister Beata Szydło, and French President François Hollande even postponed a visit to Poland following the decision in October 2016. Regardless of the political disagreement, the main concern of dispute centred on the existing and prospective locations of the Aviation Industry in Poland. But why is the location so important and what role might Poland play in this realm? In order to understand the high political, economic and media relevance of locations of certain strategic industries and their functions not only for Poland, but also for Europe, the rationale of this chapter is to shed some light with regard to background mechanisms behind locational decisions and trajectories from a historical perspective. It, therefore, portrays and analyses the processes of location, persistence and relocation in the Polish aviation industry since its inception until 1945 and aims to reconstruct historical and spatial trajectories by means of an enterprise genealogy. For this undertaking, general theoretical approaches towards ‘location, persistence and relocation’ that are capable to illustrate such trajectories are explored in the first part. The very essence of the chapter in the second part, however, is the reconstruction of historical and spatial paths within this industry. The latter is based on crucial events that led to pertinent site selections and their impact on the persistence
L. Suwala (&) Institute of Geography, Humboldt-University of Berlin, Berlin, Germany e-mail:
[email protected];
[email protected] L. Suwala Institute of Urban and Regional Planning, Technical University Berlin, Berlin, Germany © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_2
15
16
L. Suwala
and/or relocation of parts of the industry. The whole structure is methodologically undertaken through an extensive analysis of location, persistence and relocation decisions by means of archives (reconstruction of historic decisions) or in-depth interviews of experts (who worked with or have interviewed pre-WW2 key personnel). The results show that while Poland was on its way towards a self-reinforced industrial path in favour of Warszawa as its prime aviation centre, the German invasion decisively turned back time and resettled the industrial landscape.
2.2
Location, Persistence and Relocation
The location, persistence and relocation of economic activities constitute an essential building block in economic geography. The Polish aviation industry portrays an interesting case therein due to its vigorous and capricious nature partly induced by fundamental spatial ruptures in the past. Tracing this dynamic nature of a whole industry, however, is not an easy task and has been entirely neglected in traditional approaches towards industrial location. First, these approaches were static, or in other words, they examine the locational decision during a specific point in time as opposed to a period of time (e.g. Weber 1909). Second, they—with the exception of Weber—only account for independent (e.g. Hotelling 1929) or individual locational decisions (e.g. Smith 1971), and therefore, fail to explain the structure of entire industrial landscapes. Third, they were silent to persistence and relocation issues. The second wave of approaches allowed capturing this dynamic nature of location and relocation by means of a process-oriented view. Hereby, longitudinal analyses of cycles of products (e.g. product life cycle, Vernon 1966) or even whole industries (e.g. the long wave theory, Kondratiev 1935) and their spatial patterns were carried out (e.g. Norton and Rees 1979; Malecki 1991). These cycles were spurred by different types of innovation; however, they could only illustrate mostly recurrent processes and the latter only ex post. Two issues within these process-oriented approaches remained unsolved: locational persistence of industries and the ability to explain spatial ruptures in contrast to recurrent and continuous processes. A first remedy provided the canonical path dependence theory where locational persistence was elucidated by manifold so-called self-reinforcing processes (e.g. localisation economies, increasing returns, etc.) resulting in potential locational ‘lock-ins’ (David 1985; Arthur 1989). Recent extensions of this original framework have also the capacity to tackle spatial ruptures via path creation as a deliberate or even planned process of mindful deviation (Garud and Karnøe 2001; Sydow et al. 2010) from a quasi-predetermined (path dependent) outcome (the lock-in) also with regard to entire industrial landscapes (Martin and Sunley 2006). All in all, tracing historical and spatial trajectories of a whole industry, which take also location, persistence and relocation into account, require a fine-grained and eclectic theoretical approach towards such a dynamic phenomenon.
2 Between Location, Persistence and Relocation …
2.2.1
17
Process- and Path-Based Approaches
The discussion above allows to distinguish between two types of approaches when tackling ‘sequences of happenings’, namely process-based and path-based; both provide different explanations for location, persistence and relocation purposes from a dynamic, mostly novelty-driven (e.g. creativity, innovation, entrepreneurship) perspective. A first fundamental distinction—besides the very nature of processes and paths (see Table 2.1)—is that path-based approaches—in their simplest form— portray basically the idea that future outcomes depend on prior historical events and their sequence (Sewell 1996, 262); or, in short, ‘path (dependence) means that history matters’ (North 1990, 100); and such a system is unable to shake free of its own history (David 1997, 13). Process-based approaches, on the contrary, describe predetermined and continuous trajectories de facto independent of their past. A second fundamental distinction can be drawn with regard to the fact where these events happen, and if future locations depend on prior localised events or, in other words, if ‘geography matters’ (e.g. Cooke 2012, 817). Process-based approaches advocate a rather passive or even absent view on how mechanisms work in order to explain persistence or relocation. It will be assumed that model stages somehow follow a predetermined pattern that is justified by an ex post real-world observation of locational phenomena. Prior locational stages have rare, if any, interrelationships with future locations. Good examples are provided by new industrial spaces as a consequence of ‘windows of locational opportunity’, entirely different locations in novel long waves or in the later stages of the product life cycle. And, if they do, the market or an ‘almost’ invisible hand reorganises and relocates organisational activities towards ‘virgin’ landscapes by chance. Path-based approaches exemplify self-reinforcing mechanisms (e.g. agglomeration economies or spin-off dynamics) or tie ‘related’ knowledge bases to certain locations either able to reconstruct a locational dependency (persistence) or intend reasons for an entire novel location (by creating new paths). Table 2.1 illustrates the differences between both categories of approaches with regard to location, persistence and relocation activities. The process- or path-dependence/creation approaches mentioned in Table 2.1 provide only ‘a label for a particular class of dynamic phenomena, not a theory to account for the way in which such systems behave’ (David 2007, 92). Thus, invoking those concepts alone does not provide a satisfactory explanation for historical and locational trajectories. Constructing an explanatory account requires working out the precise mechanisms through which history or geography influences present and future decisions (Thelen 1999, 391; Pierson 2004, 20–21).
2.2.2
Origins and Mechanisms Behind Locational Dynamics
Let us take a detailed look at those approaches and their explanatory power in the realm of origins and mechanisms with regard to locational dynamics. The common starting point is that novelty (creativity, innovation, entrepreneurship) alters the
18
L. Suwala
Table 2.1 Process- and path-based approaches towards the evolution of economic landscapesa Process-based approaches Process-dependence product life cycle (Vernon 1966; Norton and Rees 1979), long wave theory (Malecki 1994) Process-creation geographical industrialisation (Scott 1987; Storper and Walker 1989) Path-based approaches Path-dependence (Arthur 1989, 1990; David 1985; Sydow et al. 2010)
Location
Persistence
Relocation
Manifold locations with generic assets required, random and chance
No explanation of necessary mechanisms
Locational division of work, entirely new locations
Manifold locations with generic assets required, random and chance
No explanation of necessary mechanisms, however, restructuring possible
Entirely new location (window of locational opportunity)
Random and chance
Endless self-reinforcement, positive or negative lock-in possible Possible lock-in due to branching and local knowledge transfers
Almost not possible, until a firm locates by chance in a novel region Mindful deviation in niches or through branching out of related industries
Path-creation (Martin and Depend on the Sunley 2006), related same or related variety (Boschma and economic Frenken 2003; Frenken activities/ et al. 2007) knowledge bases Source Own compilation a Indicated are only authors who also dealt explicitly with spatial consequences
economic landscape (Suwala 2014). The product life cycle assumes a limited durability of products in single industrial enterprises. This micro-economic approach puts weight on alternating production methods during the lifespan of products and therefore changed locational requirements industrial enterprises have to face. The model highlights four stylised periods: ‘a period of experimentation, a period of rapid growth, a period of diminished growth, and a period of stability or decline’ (Alderfer and Michl 1942, 14), each with definite locational factors. It is silent, however, towards location or persistence measures and provides only a very general real-world observation, as Vernon puts it ‘as there arise new industries of the sort which require external economies in their early stages, the chances seem high that the (New York Metropolitan) region will be a favoured location at first; then as maturity sets in, these industries are likely to spread out to lower-cost locations’ (Vernon 1966, 109). At the same time, it postulates relocation from certain more or less random locations in centre regions—however, with a higher probability where some kind of ‘external economies’ are already in effect in the beginning—towards the ‘periphery’, first in industrial and later in emerging economies with lower-locational cost during the maturity of the product (Norton and Rees 1979, 146). By and large, the approach implies a spatial division of labour between locations in the centre and the periphery where the locations conditions
2 Between Location, Persistence and Relocation …
19
differ (in the maturity stage, e.g. long production runs, stable technologies and scarce demand for novelty, only unskilled cheap labour). As a result, we obtain a stylised cycle—where the new emergence of industries occurs first in locations with certain generic assets (mostly in existing agglomerations with a critical mass concerning infrastructure, human capital and R&D facilities). From this point on, locational dynamics and spatial division of labour accelerate, first to suburban industrial areas, later to industrial peripheries, and finally towards locations in emerging markets or even the Global South—each with particular locational factors (e.g. proximity to markets, cheap labour cost)—as the product matures. Interestingly, the centre or the initial location where the industry originated usually remains in control—often executing headquarter functions—over peripheral sites with outsourced processes along the value chain (e.g. Kulke 1992). Therefore, the approach offers at least some explanatory power with regard to relocation, division of work and hierarchies between locations. Just like the product life cycle, the long wave theory and its spatial application tell us little about the explanatory mechanisms by which locations get chosen, but draw attention to the level of entire industries. It is mostly random and by chance where new industrial regions evolve in the realm of basic innovations (e.g. steam power, railway, electricity etc.) fundamentally challenging the existing industrial structure. Preconditions, however, are significant technical and economic synergies, a swarm of prompt and associated inventions, and sound investment in human and physical capital for the realisation of these inventions, outstanding entrepreneurs ready to take risks, as well as ubiquitous social acceptance (Hall and Preston 1988). The combination of those factors somehow leads to a significant growth of industrial powerhouses. Such industrial regions develop from cyclical, recurrent fluctuations of basic innovations lasting for 50–60 years, are predominantly mono-structured, and always (re-)locate in new industrial spaces. The resulting new industrial spaces are characterised by the following locational factors: no dominant industry, a certain critical mass of human capital, firms and infrastructure, alternatives for entrepreneurs, relatively cheaper locational costs. While the wave is in full swing, the location emanates gravitational forces leading towards a further influx of labour and capital according to market logics. The phase-out leads to rigidity (e.g. limited willingness to change, solely maintenance investments, etc.) and regional sclerosis (Malecki 1991).This approach is empirically well-tested, able to describe locational waves of industries and to shed some light on favourable locational conditions, albeit all only ex post; it, however, stays silent how these locational factors interrelate and offers no necessary mechanism for the relocation of economic activities. Probably, the most interesting insight for our purpose here is that new technological paradigms and radical innovation inevitable lead to relocation and the rise of novel industrial centres. A further, rather process-based approach called ‘geographical industrialisation’ stems from the ‘Californian School’ (Scott and Storper 1987; Storper and Walker 1989). Hereby, it is argued that during the emergence of a new technology or industry a large number of regions allow the industry to locate and grow with roughly equal chances for success. The assumption behind this idea is that all
20
L. Suwala
regions inherit virgin landscapes. A process called ‘selective clustering’ then randomly chooses one or more regions, whereas other fall behind, decline and disappear. It is further assumed that at this stage the industry creates a favourable environment in the selected locations for its purposes endogenously built on agglomeration economies without properly explaining the latter. A third stage emphasises that the novel centre can be threatened by the dispersion of certain industrial activities after a phase of rapid growth due to imminent locational disadvantages arising from higher costs of land (e.g. concentration of commercial activities), labour (already unionised workforce), capital (e.g. rivalling investors) and congestion or pollution effects. As a result, the whole industry either migrates to a different location with a high probability that a former satellite centre where parts of industrial activities dispersed first will be the next centre or stays in the original location if the site is capable to adjust and react upon alternating technological, political, social and economic circumstances. In the course of those four stylised stages, the approach is able to describe relocation and partly also persistence, but fails to explain those issues beyond ex post anecdotal stores. What all process-based approaches have in common is that they inductively describe the trajectory of industries by a stylised generalisation of real-world cases into model stages that indicate relocation ex post. The relocation, however, can either depend on the lifespan of the industrial products or result from intra- or inter-industrial dynamics. Neither are the provided explanations satisfactory for locational mechanisms and measures in general nor for location and persistence in particular. Here, so-called path-based approaches provide further valuable insights by the very fact that paths are unparalleled, unintentional, open and not goal-oriented sequences of happenings—recurrent processes are therefore basically ruled out. At the same time, path concepts lucidly accentuate that future outcomes (i.e. locations) depend on prior historical/localised events and their sequence. Since the sequence in space and time was almost prescribed in process-based approaches, the impact of prior events or locations basically ‘melted in the air’. As a consequence, localisation was driven by randomness and chance, persistence could not be explained and the stylised stages were de facto independent of each other, embodying an atomistic existence. On the contrary, path-based processes permit for a relational conceptualisation of events in time and space and therefore to identify mechanisms (the link between cause and effect) and design measures where location, persistence and relocation happen at least partly for a historical/locational reason. Let us first of all take a look at the traditional (sometimes called ‘canonical’) path-dependence theory. The seminal works of Paul David (e.g. David 1985, 2007) and Brian Arthur (e.g. Arthur 1989, 1994) which are predominantly contemplated as landmark studies on path dependence are interesting for this study as they deal with ‘carriers of history’ (David 1994), ‘locational clusters’ (Arthur 1990), ‘urban systems’ (Arthur 1988), ‘local network structures’ (David and Foray 1994) or both ‘industry location patterns and the importance of history’ (Arthur 1994, Ch. 4). To recall, path dependence—in its simplest form—is basically the idea that future outcomes depend on prior historical/locational events and their sequence, or, in
2 Between Location, Persistence and Relocation …
21
short, that such a system is unable to shake free of its own history or geography. It rests upon four general principles, namely (1) non-predictability, where the outcome is indeterminate; (2) non-ergodicity, where several outcomes (multiple equilibria) are possible and often coincidence decides among impending alternatives; (3) inflexibility, where rigidity occurs and the shift from one selected alternative to another alternative is awkward as the path evolves; (4) inefficiency, where the selective alternative might results in an inferior outcome (see Arthur 1989, 116– 117; Pierson 2000, 253). Hereby, in the realm of increasing returns,1 competition between economics objects2—technologies in the original take, but also applicable to alternative paths or to locations with regard to our interest here—takes a dynamic and evolutionary appeal with a ‘founder effect’ where ‘history’ and/or ‘geography matters’ (Arthur 1989, 128). These increasing returns—indicating increased profits as production expands—occur especially for early adopters of a novel technology due to large fixed costs, learning effects, positive network externalities or self-fulfilling expectations. Now, what do we squeeze out of all those properties and mechanism for purpose? Non- predictability appears to be appropriate only in the beginning, when randomness or ‘small historical accidents’ decide upon the path; hereafter, the path becomes implicitly formed, leading to incremental rigidity until the outcome is fully predictable in the lock-in stage. On the contrary, inflexibility occurs, ceteris paribus, at the end of the process, when the path has ‘grooved in’. In other words, the path-dependence approach stays calm with regard to the issue of location; at the same time, it captures and explains persistence in a very good way. Since current/future events for the first time depend on prior (locational) events, a certain kind of mechanism grasping the link between cause and effect can be conceptualised. Spatially bound self-reinforcing mechanisms which narrow the locational path down here are different types of agglomeration economies (of scale and scope inside the firm, localisation and urbanisation inside the region). The upshot is a locational lock-in (for the good or the bad) that results from an infinite self-reinforcement. The critical or locational junctures, however, responsible for narrowing the path can only be assessed ex post. Potential inefficiency can, therefore, also be only indicated at the later stages of the process, since decisions taken in the beginning may be very efficient and later ones contingently may be selected for reasons not necessarily connected with the initial state. The latter is inevitably intertwined with non-ergodicity or the irreversibility of choices made, bringing historicity and locality of paths to the fore. A relocation alternative does de facto not exist in the system and can, again, only be driven by the fact that a firm localises in a novel region by chance.
1
Cumulative causation, deviation-amplifying mutual casual processes, bandwagon dynamics, virtuous and vicious circles, threshold effects or positive/negative feedback processes are other examples for self-reinforcing mechanism at the very heart of path dependence (Arthur 1994, 112; Pierson 2000, 252; David 2007, 104). 2 This emphasises the passive conception of agency.
22
L. Suwala
The ‘path creation’ approach stems from the critique of the traditional approach (Martin and Sunley 2006) and breaks with its rigidity by introducing new concepts (e.g. ‘path plasticity’ (Strambach 2010), ‘planning for path-dependence’ (Sydow et al. 2010), ‘new path creation’ (Simmie 2013, etc.), but does not depict a coherent framework so far. Despite the predominantly industrial perspective, this approach also advocates an active agency by economic stakeholders, in particular, entrepreneurs who are able to mindfully deviate from existing or well-trotted paths or locations by experimenting in niches (Garud and Karnøe 2001). Political interventions are further explanation mechanisms offered when (locational) paths are broken or reinforced (Sydow et al. 2010). These extensions allow also for the conceptualisation of relocation. Moreover, the approach has the capacity—even if not systematically explored so far—to raise explanatory power for the issue of localisation. An adjacent path-creating approach which tries to resolve the dynamic location-persistence-relocation issue by relating events across time and space, but also advocating deviation from prevailing paths is the concept of related variety (Boschma and Frenken 2003; Frenken et al. 2007). Building on insights of historically informed and localised spin-off processes (Klepper 1997), the approach extends this rather entrepreneurial or firm-based view—that the success of spin-offs (spatially) proximate to their parent companies is higher—towards a sectoral and industrial perspective. The idea is simple, but fascinating at the same time. Again, spatial proximity eases the transfer of routines and especially of tacit knowledge bases. It is assumed that spillovers of those tacit, and therefore, distinctive knowledge bases are a crucial driver for new economic pathways in a particular region. Simultaneously, it is argued that the degree and intensity of those spillovers are more likely if the variety of technologies in a region is related. Hereby, cognitive proximity complements and reinforces spatial proximity (Brinkhoff et al. 2012, 2015). The newly created paths are a result of pure Darwinian evolutionary process of branching where a new sector originates either from an old sector or a recombination of competences from different, but related sectors (Boschma and Frenken 2006). These branching processes ‘are subject to local irreversibilities and share the property of non-ergodicity’ (David 2007, 99). What we obtain from these approaches for our purpose here is that all three phenomena (location, persistence, relocation) can be, even if not fully explained, at least portrayed: location as a historically and spatially bounded trajectory—with the exception when entirely novel landscapes are conquered—depending on the prior variety of technologies and their relations towards each other; persistence where ‘gains from variety at the firm level (economies of scope) and the urban level (Jacobs externalities) provide the central feedback mechanism in economic development generating strong path dependencies in the spatial concentration of industries and the specialization of cities’ (Frenken and Boschma 2007, 635), and relocation where the above-mentioned process starts anew.
2 Between Location, Persistence and Relocation …
2.3
23
Methodology
This part shortly presents the methods used to reconstruct the historical and spatial trajectories of the Polish aviation industry (PAI) between its very beginnings in 1912 until the end of WW2 against the background of the aforementioned location, persistence and/or relocation issues. This undertaking was accomplished through an extensive analysis of location, persistence and relocation decisions by means of a review of archives (where historic location decisions were reconstructed by data triangulation) or by interviewing experts (who interviewed historical key personnel themselves and reviving memories). The archives used include collections of the Polish Aviation Museum in Cracow, the Rzeszow University of Technology, the Warszawa University of Technology, the Wroclaw University Library and in particular the Section of Aviation Engineering, the Wroclaw University of Science and Technology, as well as the Foundation ‘Open Technical Museums’, the Municipality of Swidnik and the personal archive of Andrzej Glass, where all types of materials (books, periodicals, newspapers and company announcements, etc.) were taken into consideration. Experts interviewed included Andrzej Glass (the leading Polish expert on the history of the Polish Aviation Industry), Stanislaw Januszewski (the leading Polish expert on Aviation Archeology in Poland), Wieslaw Lucjanek (former Dean of the Faculty of Power and Aeronautical Engineering, Warszawa University of Technology) who all either worked with or spoke to many of the leading figures of the Pre-Second World War Polish Aviation Industry, among others. It must be freely admitted that first of all no Polish aviation press existed prior to 1911—besides anecdotal tales in daily and weekly newspapers—and therefore, no reliable facts are available; second, that for heavier than air constructions most pre-1939 Polish archives were lost, with few exceptions and oral tradition through key personnel interviewed by Polish historians (acting as experts for this study themselves); third, that Polish communist aviation press is mostly inaccurate and washy due to government censorship and government restrictions (see for a similar appraisal Cynk 1971, XVII; personal communication with Glass 2014). The unit of analysis is solely locations of back then original equipment manufacturers (OEM) and their original engine-producing counterparts (OEPM).
2.4
Historical and Spatial Trajectories of the Polish Aviation Industry
The historical evolution of the PAI can be differentiated into four stages which roughly align with the most significant global ruptures (e.g. Second World War, Fall of the COMECON) of the twentieth century, namely (1) interwar years (prior to 1914/1919–39), (2) Second World War (1939–1945), (3) communist years (1945–1989/90), (4) restructuring period (1990–2017), whereas only the first two stages will be the subject of discussion here (Suwala and Micek 2018).
24
L. Suwala
Fig. 2.1 Polish aviation works genealogy 1912–1944. Source Own compilation
Simultaneously, despite the changing borders of the Republic of Poland and its predecessors during the period investigated, six spatially bounded regions where significant activities of the aviation industry took place can be identified: (1) Central Poland, mainly with its capital region of Warszawa, (2) East Poland in the vicinity of the City of Lublin, (3) Former East Poland with its Capital Lviv, (4) Southeast Poland around the Carpathian Mountains, (5) South Poland with the City of Kraków and Bielsko-Biała, and (6) West Poland with some unevenly distributed locations (Wrocław, Kalisz, Poznań, Lódź). The reason for this choice is the subsequent westward shift of Poland’s borders (see also Fig. 2.1).3
2.4.1
Poland’s Postponed Appearance on World’s Aviation Landscape
Since Poland was not independent during the heyday of industrialisation, it also started to develop its own aviation industry late. Prussia, Russia and Austria-Hungary as occupying forces were only strategically interested in evolving a general infrastructure (communication, rail, streets) during the tripartition of the country; the same holds true for economic infrastructure like company buildings, machinery or any significant industrial activities, as the occupants feared competitors, as well as multiplying effects for these regions; at the same time, the majority of the Polish population and regions were economically suppressed and did not significantly participate in industrial life, only very few Polish projects had an opportunity to evolve beyond the paper stage (Ehlert 1954). For activities dealing with aviation, the worst conditions prevailed in the Prussian Sector, followed by Russian occupied territories, where from 1912/13 onwards a complete
3
All toponyms used here refer to the actual ones in order not to bedevil this chapter. Of course, it must be admitted that during the period of investigation, most place names changed at least once. Rymut (1980) provides an in-depth analysis of all historical place names that either once belonged to Poland or are currently within Polish territories mentioned here.
2 Between Location, Persistence and Relocation …
25
ban was enacted. The relatively best starting points existed on territories administered by the Austro-Hungarian, where only moderate disorders were expected for aviation enthusiasts (Glass 2004, 27). Despite myths twined around Czesław Tański (1861–1942) as ‘the father of Polish Aviation’ (e.g. Tygodnik Illustrowany 1909, 1081), it can be roughly stated that prior to 1918, early Polish aviation pioneers— mostly amateurs—were scattered all over the occupied territories and in foreign cities with a remarkable Polish resettlement (e.g. Vienna, Paris) and started their activities in garages, disused theatre halls or automotive workshops (Cynk 1971, 22–105). Soon, however, infant aviation centres in Warszawa, Poznań, Kraków and Lviv formed around scientific cadres within technological universities (e.g. Cynk 1971, 101), enthusiasts and the foundation of aviation association or flying clubs (Jane 1913, 40), and rather rarely around ‘left-overs’ (e.g. industrial centres, early enterprises, etc.) by former occupants providing ‘the right climate’ (Cynk 1971, 34) (personal communication with Majewski, 2014). The ‘left-overs’ were poor and since almost no photographs or anything that could be planned or started on existed, ideas, human capital (e.g. a significant number of the first generation of constructors were educated at schools outside the country (e.g. the Ecole Superior de Aeronatique in France before 1925), Majewski 2008, 18), and aviation equipment (Deszczynski 2012, 529–539) had to be developed from scratch or imported. Early initiatives were often condemned to bankruptcy because of severe shortcomings within sustainable financing structures (e.g. Dekler 1912, 8–9). From a locational perspective, three crucial events were decisive for the general formation of the spatial trajectory of the PAI until 1939. First, there was an operational consolidation, partly strategically driven locational concentration and a later nationalisation of the emerging Polish aviation industry—all measures that led to an agglomeration in Warszawa. The reasons for that agglomeration were different blows of fate in the remaining locations. For example, despite the initial favourable conditions in Lviv (technical university, aviation association, former Austro-Hungarian maintenance workshop), its production facilities were moved to Kraków due to the Russo-Polish war in 1919–20 and its proximity to the front (Majewski 2008, 14); later the production in Kraków, for its part, was abandoned as the built planes became antiquated (personal communication with Januszewski, 2014). Second, a few private ventures (DWL, Warszawa; Samolot, Poznań; Plage & Laśkiewicz, Lublin; Podlaska Wytwórnia Samolotów, Biała Podlaska) mushroomed in the early 1920s adding some novel locations, especially in East Poland to the aviation landscape (Glass 1992, 58–61). Interestingly, there was a broad variety of conditions that acted as ignition sparks in the respective locations. Some grew out of student associations around universities (Warszawa), other out of mutual initiatives of enthusiasts and patronage (Poznań and Biała Podlaska) and some out of related branches as the automotive industry (Lublin). Due to financial difficulties and aftershocks of the World Economic Crisis in 1929, strong dependence on the Polish Government (represented by the Department of Aeronautics) as the de facto only customer, and deterioration of the relationship between the both parties (the government and the private aviation sector) in the early 1930s, most companies went bankrupt (e.g. Samolot), whereas the rest (with the exception of RWD/DWL) was
26
L. Suwala
nationalised (Cynk 1971, 277). Third, the implementation of a massive industrialisation plan for Central Poland (the so-called Centralny Okręg Przemysłowy [COP, Central Industrial District]) encompassing approx. 1/6 of the former Polish territory, an area slightly bigger than Switzerland, (Radocki 1939, 26) between 1936 and the outbreak of the Second World War gave rise to an aircraft factory in Mielec and an aircraft engine factory in Rzeszów. The general reasons for choosing this triangle among the rivers of Wisła, San and the Carpathian Mountains were manifold. One idea was to revitalise that centrally, economically lagging, but strategically favourable region from a military point of view with low urbanisation rates by providing employment, exploiting natural resources, upgrading agricultural markets and locating some defence-related branches (arms, aviation) (Samecki 1998, 123–126, personal communication with Lucjanek, 2014). There were no particular reasons for choosing Rzeszów and/or Mielec. Although Rzeszów was a small town where a certain capacity of workforce could have been assumed, neither craftsmen nor engineers were available and instead blue-collar workers were brought from surrounding villages or even distant locations. White-collar staff, however, was only available from Universities in Warszawa and Lviv; therefore, the whole undertaking, despite very unfavourable locational conditions, was only possible due to the immense time pressure and military threat from Germany and the Soviet Union and the fact that this political measure was among the governmental initiatives with the highest national priority (Fiszer and Bluj 2007, personal communication with Glass, 2014). In this realm, those greenfield developments could have happened anywhere in the COP. In a nutshell, Warszawa prepared the best fertile ground for the aviation industry until 1939 due to its capital and control function and the respective political and military decision-making units (later even the nationalisation of the industry), the perpetual flow of university graduates, the best environment for financing projects through lobbying institutions (e.g. Liga Obrony Powietrznej i Przeciwgazowej [LOPP], a paramilitary organisation promoting aviation) and governmental orders for aircraft or aircraft engines. Despite the introduced scattered locations in the infant stages of development, where either aircraft or aircraft engine was assembled, Warszawa evolved as the central location for aviation in the newly formed country. Later, the Polish Capital even reinforced its dominant position further, due to economic bottlenecks of the private sector (most aviation enterprises in locations outside Warszawa were organised in such way), especially after the stock market crash in 1929 and its consequences, and later due to the nationalisation of the industry. All in all, all other locations were subsequently degraded to remote workbenches.
2.4.2
Germany’s (Un-)Expected Takeover
The situation suddenly changed in September 1939 as Germany invaded Poland. The new potentates divided the country into a relatively well-developed western
2 Between Location, Persistence and Relocation …
27
part (once belonging to Prussia), which was incorporated into the German Reich, and a middle or eastern part called the German administered district of Poland during World War II (Generalgouvernement, GG). The German Armed Forces convinced Hitler to at least use the industrial potential of the GG for armaments by taking over the entire defence industry (Herrmann 2012, 1). The main objective was to supply the police and the army with weapons, munitions and groceries, degrading the Polish industry into a raw material deposit at the same time (Wolf 2008, 20). Although surprisingly few aviation factories had been entirely destroyed (e.g. Biała Podlaska) during the attack in 1939, the real problem posed the complete divestiture of the internal organisational structures within Polish aviation enterprises, where leading personnel disappeared to neighbouring and amicable countries leaving the whole industry mindless. For this and other reasons like opposition groups, sabotage and general labour shortages, even Polish plants and workshops that have been classified as valuable by the Nazis could not be initially reactivated (Budraß 1998, 619). From an organisational perspective, the Nazis converted former Polish aviation works mostly into low hierarchy workbenches, maintenance and overhauling workshops in a spatially separated Fordist-type production mode with German OEM in the core Reich on top of the pyramid. Therefore, most German aviation giants (e.g. Daimler-Benz, Heinkel, Junkers, etc.) had their ‘labour camps or farms’ in former Polish aviation factories engaging slave labour. In general, less investment was made in aviation locations situated in the east of Poland (e.g. Rzeszów) (Herrmann 2012, 137). Not surprisingly, with the advance of the Red Army, aviation works in Lviv and Lublin were abandoned first and even partly destroyed by the Germans as the Wehrmacht did not believe in having permanent control over those territories. In addition, most buildings at Warszawa-Okęcie—the heart of the Polish Aviation industry—were blown up in 1944, including the majority of aircraft production facilities. All in all, most locations—with the exception of the ones in very far Eastern borderlands—persisted, however, they were degraded, mined or hollowed out, most leading personnel emigrated, machines either taken during retreat of the German Armed Forces or the advance of the Red Army (Glass and Kubalanca 2013, 93–98). Those events left a state of devastation with the industry lying in ashes and rubble. The end of the Second World War implied a restart for the Polish aviation industry and its landscape. Within the limits of ‘westward shift’ of Poland, Lviv got lost while some permanent (Schlesien) or incorporated German territories (Danzig-Westpreußen, Wartheland) with remnants of the aviation industry (e.g. Wrocław, Kalisz, Lódź) belonged to Poland then. Additional former aviation works situated in former long-standing German territories of Ostpreußen (e.g. Gdynia, Malbork, Szczecin) were either destroyed or exploited by German and/or Soviet Armed Forces (Glass and Kubalanca 2013, 103, personal interview with Januszewski, 2014). Despite the poor conditions of most aviation works, German occupants had to leave numerous aviation locations behind that unintentionally helped to set up an aviation industry network in Poland after WW2 and that have survived to date (e.g. Wrocław, Kalisz, Krosno) (Kinder and Suwala 2012).
28
2.5
L. Suwala
Discussion and Conclusions
Now, what have we learnt from the historical and spatial trajectories of the Polish aviation industry until 1945? And how can this example enhance a theory formation? First of all, it can be clearly stated that both—insights from process- and path-based approaches—are necessary in order to adequately reconstruct and illustrate location, persistence and relocation decisions within historical and spatial industrial trajectories. Only a careful eclectic valorisation of both approach families and the respective precise background mechanisms through which history or geography influences present and future decisions are able to fulfil the task. Interestingly, numerous locational decisions can merely be hallmarked as random and accidental (e.g. picking Rzeszów as a site for engine production) even ex post, hence also partly justifying these theoretical approaches. The PAI showcase that its infant years from the foundations (sometime between 1909 and 1912) until the mid-1920s can be characterised by what Storper and Walker (1989) call the localisation phase. Although there were definitely locations with more advantageous conditions than other (e.g. Warszawa or Lviv), early sites of the PAI were randomly scattered throughout the country exploiting windows of locational opportunity in a relatively unrestricted market. Whereas most locations flourished more or less by chance, at least Lublin already had an industrial tradition in automobiles; in this stance, the related variety argument can be raised here. External economies of scale in the Polish Capital, strategic decisions and financial difficulties in other locations, led to a selective clustering of the PAI in Warszawa. To name only a few, the strategic abandoning of Lviv as an aviation centre for production or the collapse of the Samolot Aviation factory Poznań definitely favoured that concentration. Later, however, those were not merely the market (as a functioning market was not in existence in this sector) and agglomeration forces that reinforced a path towards Warszawa as the heart of Polish Aviation Industry with a more advanced labour specialisation (engine manufacturing, assembly) and product differentiation (gliders, planes) than other locations. At the same time, intended decisions of the Government (mainly the Department of Aeronautics, represented by Colonel Ludomił Rayski) towards the nationalisation of the industry were subjugating further locations (e.g. Lublin, Biała Podlaska) under the capital’s control. As the path was increasingly moving towards a locational lock-in, two events—one planned (greenfield initiatives) and one (un-)expected (German invasion)—convulsed the Polish aviation landscape anew. While greenfield initiatives in Rzeszów and Mielec in the realm of the COP, could be best explained either by Storper and Walker’s dispersion stage (process-creation) or the spatial application of the product life cycle (process-dependence) where a locational shift towards the national periphery with workbenches and blue-collar activities occurs due to mass production, increased division of labour, and no vital proximity to R&D facilities— with little change to the status quo of Warszawa, the German invasion in 1939 partly eroded the aviation landscape and started processes and paths anew. Despite a gratuitous acquisition of the PAI and prevalent exploitation and destruction of its
2 Between Location, Persistence and Relocation …
29
physical and human resources, German occupants also founded novel locations (e.g. Krosno) or had unexpectedly to leave locations that belonged to the Reich (e.g. Wrocław) or were incorporated in it (e.g. Kalisz) after the retreat of the German Armed Forces in 1944 and the westward shift of Poland. Interestingly, although the Polish aviation landscape was blighted (e.g. destruction of most aviation buildings in Warszawa in 1944) and the general conditions were rather poor in all locations by the end of WW2, the new locational build-up of the industry in the communist years was not so heavily dominated by random and chance, but rather by persistence, where even heavily destroyed places persisted longer than expected or even still play a significant role in the Polish aviation landscape today.
References Alderfer, E. B., & Michl, H. E. (1942). Economics of American industry. New York: McGraw-Hill. Arthur, W. B. (1988). Self-reinforcing mechanisms in economics (Santa Fe Institute Studies in the Science of Complexity, 5). Redwood City: Addison-Wesley. Arthur, W. B. (1989). Competing technologies and lock-in by historical small events. Economic Journal, 99, 116–131. Arthur, W. B. (1990). Industry location patterns and the importance of history. Mathematical Social Sciences, 19, 235–251. Arthur, W. B. (1994). Increasing returns and path dependence in the economy. Ann Arbor: University of Michigan Press. Boschma, R. A., & Frenken, K. (2003). Evolutionary economics and industry location. International Review for Regional Research, 23, 183–200. Boschma, R. A., & Frenken, K. (2006). Why is economic geography not an evolutionary science? Towards an evolutionary economic geography. Journal of Economic Geography, 6(3), 273–302. Brinkhoff, S., Suwala, L., & Kulke, E. (2012). What do you offer?: Interlinkages of universities and high-technology companies in science and technology parks in Berlin and Seville. In R. Capello, A. Olechnicka, & G. Gorzelak (Eds.), Universities—Cities—Regions (pp. p121– p146). London: Routledge. Brinkhoff, S., Suwala, L., & Kulke, E. (2015). Managing innovation in ‘localities of learning’ in Berlin and Seville. In G. Micek (Ed.), Understanding innovation in emerging economic spaces (pp. 11–31). Farmham: Ashgate. Budraß, L. (1998). Flugzeugindustrie und Luftrüstung in Deutschland, 1918–1945. Düsseldorf: Droste. Cooke, P. (2012). Transversality and transition: Green innovation and new regional path creation. European Planning Studies, 20(5), 817–834. Cynk, J. B. (1971). Polish aircraft 1893–1939. London: Putman. David, P. A. (1985). Clio and the economics of QWERTY. American Economic Review, 75(5), 332–337. David, P. A. (1994). Why are institutions the ‘carriers of history’?: Path dependence and the evolution of conventions, organizations and institutions. Structural Change and Economic Dynamics, 5(2), 205–220. David, P. A. (1997). Path dependence and the quest for historical economics: One more chorus of the ballad of QWERTY (Vol. 20). University of Oxford Discussion Papers in Economic and Social History (November).
30
L. Suwala
David, P. A. (2007). Path dependence: A foundational concept for historical social science. Cliometrica, 1(2), 91–114. David, P. A., & Foray, D. (1994). Dynamics of competitive technology diffusion through local network structures: The case of EDI document standards. In L. Leydesdorff & P. van den Besselaar (Eds.), Evolutionary economics and Chaos theory (pp. 63–78). London: Pinter. Dekler, Z. (1912). Przemysl Lotniczy w roku 1911. Lotnik I Automobilista, 1912(4), 8–9. Deszczynski, M. P. (2012). Import sprzętu wojskowego przez Polskę w latach 1921–1939. Kwartalnik Historyczny CXIX, 3, 507–540. Ehlert, A. (1954). Die Industrialisierung Polens zwischen den Weltkriegen unter besonderer Berücksichtigung der neuen Standorte. Kiel. Fiszer, M., & Bluj, M. (2007). Od Pegasusa do F-100-PW229. Wypisy z siedmiu dekad historii WSK “PZL-Rzeszów” S.A (1937–2007). Magnum-X, Warszawa. Frenken, K., & Boschma, R. A. (2007). A theoretical framework for evolutionary economic geography: Industrial dynamics and urban growth as a branching process. Journal of Economic Geography, 7(5), 635–649. Frenken, K., Van Oort, F. G., & Verburg, T. (2007). Related variety, unrelated variety and regional economic growth. Regional Studies, 41(5), 685–697. Garud, R., & Karnøe, P. (2001). Path dependence and creation. London: Lawrence Erlbaum Associates. Glass, A. (1992). Realizacja koncepcji rozwoju przemysłu lotniczego. In A. Glass (Ed.), Polska Technika Lotnicza do Roku 1939 (pp. 44–79). Warszawa: IHNOit. Glass, A. (2004). Polskie Konstrukcje Lotnicze (Vol. I). Sandomierz: Stratus. Glass, A., & Kubalanca, J. (2013). Polskie Konstrukcje Lotnicze (Vol. V). Sandomierz: Stratus. Hall, P. G., & Preston, P. (1988). The carrier wave: New information technology and the geography of innovation, 1846–2003. London: Unwin Hyman. Herrmann, D. (2012). Führungsverhalten und Handeln reichsdeutscher Unternehmer/Manager und deren Verstrickung in den NS-Terror im Generalgouvernement der besetzten polnischen Gebiete (GG) 1939 bis 1945. Hamburg: Universität Hamburg. Hotelling, H. (1929). Stability in Competition Economic Journal, 39, 41–57. Jane, F. (Ed.). (1913). Jane’s All the World’s Aircraft 1913. London: Sampson Low Marston. Kinder, S., & Suwala, L. (2012). Rzeszów und das “Aviation Valley” – Perspektiven für eine dynamische Wirtschaftsentwicklung in Südostpolen. Bundeszentrale für politische Bildung. Dossier Polen. 6.12.2012. Polen-Analysen, 7(118), 1–13. Klepper, S. (1997). Industry life-cycles. Industrial and Corporate Change, 6, 145–182. Kondratiev, N. D. (1935). The long waves in economic life. The Review of Economics and Statistics, 17(6), 105–115. Kulke, E. (1992). Empirische Ergebnisse zur regionalen Produktlebenszyklushypothese – Untersuchung in Niedersachsen. Die Erde, 123(1), 49–61. Majewski, M. W. (2008). Samoloty i zakłady lotnicze II Rzeczypospolitej. Warszawa: Wydawnictwo ZP. Malecki, E. (1991). Technology and economic development: The dynamics of local, regional and national change. Harlow: Longman. Martin, R., & Sunley, P. (2006). Path dependence and regional economic evolution. Journal of Economic Geography, 6(4), 395–437. North, D. N. (1990). Institutions, institutional change and economic performance. Cambridge: Cambridge University Press. Norton, R. D., & Rees, J. (1979). The product cycle and the spatial decentralization of American manufacturing. Regional Studies, 13, 141–151. Page, S. E. (2006). Path dependence. Quarterly Journal of Political Science, 1(1), 87–115. Pierson, P. (2000). Increasing returns, path dependence, and the study of politics. American Political Science Review, 94(2), 251–267. Pierson, P. (2004). Politics in time: History, institutions and social analysis. Princeton: Princeton University Press. Radocki H (1939) Centralny Okręg Przemysłowy w Polsce, Nakładem Myśli Polskiej, Warszawa.
2 Between Location, Persistence and Relocation …
31
Rymut, K. (1980). Nazwy miast Polski, Wrocław: Zakład Narodowy im. Ossolińskich. Samecki, W. (1998). Centralny Okrȩg Przemysłowy 1936–1939. Wrocław: Wydawnictwo Uniwersytetu Wrocławskiego. Scott, A. J., & Storper, M. (1987). High technology industry and regional development: A theoretical critique and reconstruction. International Social Science Journal, 112, 215–232. Sewell, W. H. (1996). Historical events as transformations of structures: Inventing revolution at the bastille. Theory and Society, 25, 841–881. Simmie, J. (2013). Path dependence and new technological path creation in the economic landscape. In P. Cooke (Ed.), Re-framing regional development—Evolution, innovation and transition (pp. 164–185). New York: Routledge. Smith, D. M. (1971). Industrial location. New York: Wiley. Storper, M., & Walker, R. (1989). The capitalist imperative: Territory, technology and industrial growth. New York: Basil Blackwell. Strambach, S. (2010). Path dependency and path plasticity. The co-evolution of institutions and innovation—The German customized business software industry. In: R. A. Boschma & R. Martin (Eds.), Handbook of evolutionary economic geography (pp. 406–431). Edward Elgar, Cheltenham. Suwala, L. (2014). Kreativität, Kultur und Raum. – Ein wirtschaftsgeographischer Beitrag am Beispiel des kulturellen Kreativitätsprozesses. Springer. Wiesbaden. Suwala, L., & Micek, G. (2018). Beyond clusters? Field configuration and regional platforming: The Aviation Valley initiative in the Polish Podkarpackie region. Cambridge Journal of Regions, Economy and Society. http://dx.doi.org/10.1093/cjres/rsy010. Sydow, J., Lerch, F., & Staber, U. (2010). Planning for path dependence? The case of a network in the Berlin-Brandenburg optics cluster. Economic Geography, 86(2), 173–195. Thelen, K. (1999). Historical institutionalism in comparative politics. Annual Review of Political Science, 2, 369–404. Tygodnik Illustrowany. (1909, December 18). Modele Aeroplanów. Tygodnik Illustrowany, 51, 1081. Vernon, R. (1966). International investment and international trade in the product cycle. The Quarterly Journal of Economics, 80(2), 190–207. Weber, A. (1909). Über den Standort der Industrien. Tübingen: Mohr. Wolf, N. (2008). Ökonomische Zugänge zur Geschichte der deutsch-polnischen Beziehungen. In D. Bingen, P. O. Loew, & N. Wolf (Eds.), Interesse und Konflikt: zur politischen Ökonomie der deutsch-polnischen Beziehungen, 1900–2007 (pp. 9–24). Wiesbaden: Harrassowitz.
Lech Suwala is a Guest Professor for Urban and Regional Economics at Technical University Berlin(Germany), and Senior Lecturer in Economic Geography at Humboldt-University in Berlin (Germany). He is a geographer and economist with working experience in both science and industry. He holds or has held positions as Research Fellow at Simon-Fraser-University, Vancouver, Canada; Lecturer at the University of Namibia, Windhoek and University of Applied Sciences on Technology and Economics, Berlin. His expertise includes spatial creativity, innovation and entrepreneurship research, where he has conducted studies and given manifold speeches on (1) on the very nature of different types of creativity in space, (2) on the management of creativity and innovation in space and (3) on temporary events and institutional change in cultural and creative industries as well as high-technology sectors. One of these sectors encompasses the aviation industry, where he intensively investigated location and relocation dynamics taking the example of Poland. His research has been published in both peer-reviewed journals (e.g. Cambridge Journal of Regions, Economy and Society, Regional Studies) and internationally edited volumes (Ashgate, Routledge, Springer).
Chapter 3
The Relocation of the Automotive Industry in Brazil and Mexico: Between Corporate Strategies and Industrial Policies Mirosław Wójtowicz
3.1
Introduction
In recent decades, the geography of the global automotive industry has undergone a thorough transformation. Since the mid-1980s, this sector has evolved from a number of disparate domestic industries sheltered from excessive competition by protective government policies to a more integrated, oligopolistic global industry (Sturgeon and Florida 2004; Sturgeon and Biesebroeck 2009). At the same time, the tightening of global links is accompanied by the development of strong regional production/operation structures adapted to specific markets (Holmes 1993; Humphrey et al. 2000; Carrillo et al. 2004; Sturgeon et al. 2008). The high motorisation level and market saturation in highly developed countries and political pressure from developing countries exerted on producers to “build where you sell” have persuaded international car companies to develop production in countries offering potentially receptive sales markets. This has led to both a markedly greater dispersion of car production and increased activity of international companies in comparison to the early 1980s. The development of production in new markets has also encouraged regional differences in customers’ preferences, meaning that producers adapt appearance and fittings of vehicles to local expectations and demands, such as the steering wheel position, reinforced suspension and adaptation to natural conditions—tropicalisation (Carillo et al. 2004). Since the mid-1980s, a dominant trend in the global automotive industry has been increasing regional integration, driven by both political and technological factors. The political pressure exerted by the governments of large countries or emerging regional economic groupings leads the major car producers to invest in those countries and regions where the market size and existing or anticipated demand guarantee adequate profits. In spite of the liberalisation of trade taking M. Wójtowicz (&) Institute of Geography, Pedagogical University of Kraków, Kraków, Poland e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_3
33
34
M. Wójtowicz
place since the early 1990s, restricting barriers in trading of automotive sector products, in order to gain recognisability in a given country and achieve a significant share of vehicle sales in this market, companies must reckon with the necessity of developing production there. Only when a manufacturer is present in a country is it likely to win the favour of the government as well as potential consumers, and secure increased interest in buying its vehicles. Production by global car companies in specific regions is also the basis of successful marketing campaigns for a given brand, raising the credibility of a firm in the eyes of local consumers. Political pressure and consumer expectations resulted in a wave of foreign direct investments of global automobile corporations that were not present in Mexico or Brazil before the 1990s (Humphrey et al. 2000; Carrillo et al. 2004; Sturgeon and Biesebroeck 2009), and this will be discussed in depth further in the chapter. Regionalisation has also an impact on technical and technological considerations, as a result of increasingly close cooperation between vehicle producers and their main subcontractors. Major car companies often demand from their most important suppliers that they be present on all markets where the producer is active, and often encourage them to build their factories close to car plants. The result is automotive industry clusters, frequently grouped around the car production factory with suppliers of heavy and specialised parts (engines, transmission, seats) close by, permitting just-in-time delivery. The suppliers of lighter and more common parts, meanwhile, can afford to locate their production sites to exploit cheaper workforce or economies of scale and are therefore not always present in such clusters. This trend also played a major role in the inflow of new investments associated with the car parts production sector to Brazil and Mexico (Pires and Neto 2008; Hasegawa et al. 2013; Juárez 2005; Covarrubias 2011; Carbajal 2015). Global suppliers appeared in these countries, following the investments of the automotive companies with which they worked. Continued collaboration forced them to enter these markets either by takeover and adaptation of local firms to their own needs or by building factories on greenfield sites. Nevertheless, that closer cooperation between vehicle producers and their subcontractors contributed to a dynamic inflow of FDI, connected to the production of car parts and increased employment in this sector, in the two countries (Holmes 1993; Shapiro 1993; Carrillo 2004; Sturgeon and Florida 2004; Amann et al. 2007; Sturgeon and Biesebroeck 2009). The similarities evident in the emergence and development of the automotive industry in Brazil and Mexico, resulting from the countries’ strategies of import substitution industrialisation, gradually became less significant from the mid-1980s onwards. This strategy proved to be incompatible with the continuing processes of globalisation of the world economy (Bennett and Sharpe 1985; Shapiro 1993; 1994). The necessity of joining the global economy forced countries to liberalise gradually and adapt to the prescriptions of the Washington Consensus. Active incorporation of Brazil and Mexico into international trade exchange made the differences in their situation and significance on the world map clear, which played an important role in their further development. The economic development of Mexico was very strongly determined by its position in the shadow of its large
3 The Relocation of the Automotive Industry in Brazil and Mexico …
35
northern neighbour, the USA, and this was further reinforced by the formation of NAFTA in 1994. It was also reflected in the evolution of the Mexican automotive industry, which became increasingly dependent on the export of its products to the US market (Holmes 1993; Mortimore 1995; Carrillo 2004). Brazil, on the other hand, comprises some 48% of the area and is home to 49% of the population of South America, as well as being its largest economy; therefore, it is naturally also a regional leader. Since 1991, and the formation of the regional economic grouping Mercosur, Brazil has been seeking tighter economic and political ties with its neighbours, but also increasingly playing the role of the regional leader for the whole South America (Saraiva 2010; Malamud 2012). The size and role of Brazil in the region make it the largest market for international automotive companies, with very promising prospects of further growth in car sales with improvement of its citizens’ material situation (Laplane and Sarti 2004). Taking into account the global factors and regional specifics of the automotive sector in Brazil and Mexico, we should note that different sets of factors were at play in the dynamic development in each country in recent decades, as a result of their disparate position in the automotive global value chain (GVC). Brazil represents a type of region characterised by a potentially large internal market, low motorisation rate and rapidly developing economy—at least until 2014. This type of region also includes such large countries as China, India and Russia. Mexico, meanwhile, represents a second type of region with dynamic development of the automotive industry, resulting from the process of regional integration with the centre. In the case of Mexico, this means increased integration with the US and Canadian automotive sector within NAFTA. In the first type of regions, the development of the automotive industry mainly results from the needs of the internal market, driving development of production and the inflow of new investments. In the second type, meanwhile, the main factor responsible for increased production and new investments is the access obtained to the receptive markets of the wealthy states of the centre (in the case of Mexico, the USA and Canada), along with the possibility of reducing the costs of car production by relocating it to peripheral states with lower production costs. The transfer of auto parts production is even more extensive than that of finished vehicles, which is politically safe for auto producers, as they kept the final car assembly in the main market country and the supply parts from a country with lower labour costs (Sturgeon and Biesebroeck 2009; Pavlínek and Ženka 2011). The objective of this chapter is to explore the influence of the change in the global and regional conditions of the development of the automotive industry on changes in the distribution of this sector factories in Brazil and Mexico, as well as to point to the emerging growth poles connected to the inflow of international investments of car companies in recent years. In order to achieve these aims, a wide-ranging review of the literature is conducted, documenting the history of transformations of the automotive sector in the two countries in recent decades, applying GVCs as an analytical framework. It is argued that although the two countries occupy a different position in the GVC of the automotive industry, both experienced a profound transition in its spatial distribution. The analysis compares
36
M. Wójtowicz
political and economic factors of the two countries and the international determinants which triggered the relocation processes in this industry. The chapter has four parts. The first presents an overview of global value chains as a framework for analysing the contemporary dynamic of the economic geography of the global automotive industry. The second focuses on charting the contemporary spatial changes in the automotive sector in Brazil. The third covers the spatial transformations in the automotive sector in Mexico, with particular consideration to the period after joining NAFTA. The final, concluding part, draws together the similarities and differences in the evolution of the sector in the two countries. This research is based on data from the last three economic censuses conducted in Mexico by the national statistical office (INEGI 2016) and from the Brazilian Statistics of the Central Register of Enterprises (IBGE 2016a), which provide similar sets of statistical information for further comparative analysis.
3.2
Global Value Chains in the Automotive Industry
In the 1980s, the accelerating globalisation gave rise to a new organisation and governance of global industries, characterised by the emergence of producer-driven and buyer-driven commodity chains. Large, vertically integrated transnational corporations (TNC), which developed the automotive industry in Brazil and Mexico under the import-substituting industrialisation (ISI) model of growth, started to change their production strategy, outsourcing an increasing number of parts needed for car assembly from suppliers. Concurrently, the second oil shock of the late 1970s and the debt crisis that followed it forced the analysed countries to shift to a neoliberal development model, favouring export-oriented industrialisation (EOI), which involved the liberalisation of trade and stronger involvement in global production networks (GPNs) (Gereffi 2014; Gereffi et al. 2005). The above processes allowed automotive companies not only to change their internal structure, by spinning their internal parts divisions and creating autonomous part suppliers, but also to adopt new governance strategies, best explained by producer-driven GVC frameworks, in which the power is exerted by the end-product manufacturer. Final car assemblers are strong enough to require their major subcontractors to be located in the vicinity of their factories, thus favouring the development of industrial clusters (Sturgeon et al. 2008; Sturgeon and Biesebroeck 2011). The emergence of GVCs has redefined the current development of the world automotive industry. Firstly, developing countries such as Brazil and Mexico have gained in importance, attracting a growing proportion of the production of the automotive sector, frequently at the cost of factories forced to close in developed countries. Secondly, there have been significant spatial shifts within the two countries, to a great extent as a result of the concentration of subcontractors in the vicinity of factories producing ready-made vehicles, leading to formation of the “industrial condominium” or “modular consortium” types of spatial concentrations (Pires and Neto 2008; Hasegawa et al. 2013).
3 The Relocation of the Automotive Industry in Brazil and Mexico …
3.3
37
Developments and Relocations Within the Brazilian Automotive Industry
The Sectoral Chamber for the Automotive Industry was established in 1992–1994 in Brazil with the objective of coordinating policies towards the automotive sector, bringing together representatives of the Brazilian government, car producers, suppliers, dealers and trade unionists to develop a joint development strategy. Its first agreement sought to reduce taxes and car prices in order to increase the domestic demand and production and sustain the existing employment levels. A zero-level industrialised product tax (IPI) was introduced on cars powered by engines up to 1000cc. This marked the beginning of a “popular car” strategy, i.e. production of cheap, subcompact vehicles adapted to the financial means of a large swathe of consumers and permitting a production scale ensuring profitability (Quadros and Consoni 2009; Marx and Mello 2014). In 1995, the government launched the so-called automotive regime—AR, which consolidated the industrial policy towards the automotive sector, again increasing its protection from imports and encouraging investment in Brazil. This contributed to a dramatic growth in the domestic market. In 1997, it recorded its highest ever number of sales of vehicles produced in the country, exceeding 1.64 million units. As a result, the TNCs operating in Brazil expanded production, as well as encouraging corporations which were not previously present to enter the market, leading to a new wave of investments. Between 1994 and 2001 car companies invested over 14.8 billion USD in Brazil. New producers appearing on the market included Renault, Honda, Toyota, PSA, Mitsubishi, Iveco and Audi, while those already present invested in constructing new factories adapted to the changing technological demands and organisation of work. The majority of new investments were located away from the traditional regions where the automotive industry was concentrated, taking advantage of the rivalries between regional and local governments keen to attract major foreign investors to increase the dispersion of production (Laplane and Sarti 2004; Rodriguez-Pose and Arbix 2001; Rodriguez-Pose et al. 2001; Marx and Mello 2014). In 1990, some 99.3% of vehicles produced came from the states of São Paulo and Minas Gerais. By 2002, this percentage had decreased to 77.4%. New states with dynamic development of car production appeared on the map of the Brazilian automotive industry: Parana (8.3%), Rio Grande do Sul (6.4%), Rio de Janeiro (4.1%) and Bahia (3.2%) (ANFAVEA 2003). The influx of new investments resulted in increased employment and production in the car industry, which was briefly interrupted in 1998–2000 by the economic crisis, itself echoing the earlier slump in Southeast Asia. From 1999 to 2014, employment in the automotive industry rose by over 2.5 times (Table 3.1). It is worth noting, however, that in the same period production of this type of vehicles grew from 1.28 million to 2.99 million units, indicating increased mechanisation, automation and modularisation of production. This resulted in the reduction of employment in the final assembly of vehicles, shifting more labour-intensive work
38
M. Wójtowicz
Table 3.1 Evolution of employment and coefficient of redistribution in particular sections of the automotive industry in Brazil: 1999–2014 Industry
Employment in 1000s 1999
2009
2014
Change 1999 = 100% 2009 2014
Coefficient of redistribution Qar (1999– 2014)
Automotive industry 192.6 479.4 502.6 248.9 260.9 0.128 overall Production of passenger 61.2 87.4 99.8 142.8 163.1 0.214 cars and LCVs Production of heavy 12.2 23.8 30.2 195.1 247.5 0.312 goods vehicles and buses Production of driver’s 21.5 60.1 70.5 279.5 327.9 0.217 cabins, trailers and bodywork for mechanical vehicles Production of parts and 87.2 292.3 287.6 335.2 329.8 0.126 accessories for mechanical vehicles Regeneration and renewal 10.6 15.7 14.5 148.1 136.8 0.111 of mechanical vehicle engines Source Own compilation based on data from: IBGE (2016a) a Coefficient of redistribution Qr calculated according to the model: Pn t0 t1 1 Qr ¼ 200 i¼1 pi pi where: pi the percentage of workers in industry i-of this state in relation to employment in the country as a whole in the t0 and t1 time slice. The values of the coefficient can vary between 0 and 1; the closer to 1, the greater the changes in the spatial distribution of the industry in question
to subcontractors. The continued technological and organisational changes that came with the modernisation and globalisation of the car industry also contributed to the increased importance of producers of car parts (Sturgeon and Florida 2004; Sturgeon et al. 2008). A useful tool for investigating spatial changes in the distribution of various sectors of the automotive industry is the coefficient of redistribution. The advantages of this simple gauge are the fact it determines the scale of spatial changes and the ease of comparing the obtained results between the various sectors, since its values are contained within an interval between 0 and 1, regardless of the number of employees in each (or other analysed variables, such as production value or gross value added). A deficiency of this measurement, however, is that despite showing the scale of spatial movements, it does not inform us of their directions. Therefore, in order to define the directions of relocation of the various sectors of the automotive industry, subsequently in this chapter I refer to their proportion in the analysed states. Focusing on this made it possible to ascertain both which states gained in importance and which became less significant as a result of this process.
3 The Relocation of the Automotive Industry in Brazil and Mexico …
39
Taking this into account, it is notable that, whereas in the case of production of passenger vehicles and LCVs a relatively high coefficient of redistribution was recorded, suggesting significant changes in the distribution of the industry, for car parts this indicator was lower than the average for the whole sector, which signifies relatively minor changes in its spatial distribution (Table 3.1). The largest spatial shifts took place in the production of HGVs and buses. This branch of the industry saw an increase in employment of almost 150% in the research period, as well as a considerable relocation in comparison to 1999. The smallest changes in distribution, meanwhile, were recorded in the branch dealing with regeneration and renovation of car engines. This branch is characterised by the lowest employment as well as the smallest employment growth in the entire automotive sector (Table 3.1). The spatial changes in the Brazilian car industry show a gradual reduction in the concentration of its manufacturing activity. This deconcentration is linked to the relocation of car factories outside of the traditional regional concentrations around the metropolitan areas of São Paulo and Belo Horizonte, starting in the mid-1990s. This was caused by the endeavours of automobile corporations which had been present in Brazil for some time to increase the production capacity and output of their factories. This usually involved closing down the existing plants and building new ones in less industrialised regions characterised by low activity of trade unions and offering various subsidies, tax breaks and other incentives to attract investors (Rodriguez-Pose and Arbix 2001; Rodriguez-Pose et al. 2001). Some of these investments were located in states in southern Brazil, as a result of the developing production links within Mercosur, which could be perceived as an application of a GVC policy. A particularly dynamic growth was recorded in the states of Parana, Rio Grande do Sul and Santa Catarina (Table 3.2; Fig. 3.1). The second significant direction of relocation targeted previously weakly industrialised regions with a large market potential, and this can also be perceived as an example of the influence of TNCs’ use of development strategies referring to the premises of the concept of GVC. Examples of such shifts are the investments of Ford in Camaçari in the state of Bahia, Mitsubishi in Goias and Fiat in Pernambuco (Laplane and Sarti 2004; Amann et al. 2007; Cruz and Rolim 2010). The deconcentration of employment is especially visible in the case of production of passenger cars and LCVs. São Paulo and Minas Gerais, the dominant states in this branch in terms of the number of workers, experienced a drop in their share from 93.7 to 72.3% of the total between 1999 and 2014 (Table 3.2). The greatest success in the growth of employment in car production, meanwhile, was recorded in the state of Parana, and specifically the town of São José dos Pinhais on the outskirts of Curitiba, where large Renault, Audi and VW factories opened. Further examples of successful relocation are the Bahia, Goias, Rio de Janeiro and Rio Grande do Sul states, where new car factories have also been built (Wójtowicz and Rachwał 2014). Employment in the automobile parts production branch showed a lower spatial concentration, with a decreasing domination of São Paulo. Meanwhile, the most spectacular increase in the share of overall employment in this area was recorded
40
M. Wójtowicz
Table 3.2 Participation of selected states in total employment of the Brazilian automotive industry: 1999–2014 State
Automotive industry
1999
2009
2014
Manufacture of passenger cars and LCV 1999 2009 2014
Amazonas 0.1 0.4 0.5 0.0 0.0 Ceará 0.2 0.5 0.4 0.1 0.6 Pernambuco 0.2 0.5 1.0 0.1 0.1 Bahia 0.3 1.9 1.8 0.0 3.8 Minas Gerais 14.7 14.9 15.7 20.4 16.0 Rio de Janeiro 2.0 2.8 2.9 0.5 3.7 São Paulo 62.8 53.8 50.4 73.3 57.5 Paraná 5.0 8.3 9.0 4.2 10.4 Santa Catarina 2.6 4.5 5.0 0.0 0.0 Rio Grande do Sul 10.3 10.0 10.0 1.4 4.8 Goiás 0.5 1.0 1.4 0.0 2.8 98.7 98.7 98.3 99.9 99.8 Totala Source Own compilation based on data from: IBGE (2016a) a Totals do not add up to 100% because of taking into account
0.1 0.5 1.8 3.9 20.0 2.9 52.3 9.7 0.8 4.1 3.9 99.9
Manufacture of parts and accessories for motor vehicles 1999 2009 2014 0.1 0.1 0.1 0.1 14.6 2.0 66.2 4.8 1.6 10.0 0.1 99.7
0.6 0.2 0.5 1.6 16.3 1.9 58.5 7.1 5.0 8.0 0.2 99.8
0.7 0.3 0.6 1.5 16.1 1.8 55.8 7.6 6.7 8.1 0.2 99.3
only selected states
for Santa Catarina and Parana (Table 3.2). The reasons for these states’ success are on the one hand the relocation of certain factories producing vehicles to southern Brazil, and on the other the development of cooperation within Mercosur and their convenient location, which facilitates export of production parts to Argentina (Laplane and Sarti 2004; Azevedo and Massuquetti 2015). The changes in the spatial concentration of the Brazilian automotive sector are progressing much faster if we consider value added in industry (VAI) data. Between 1999 and 2014, São Paulo state’s share of the VAI of the Brazilian automotive industry shrank from 62.4 to 48.5%. At the other end of the scale were Parana and Rio Grande do Sul, whose share in the industry VAI increased most in the period in question (Table 3.3). The declining significance of São Paulo and Minas Gerais in the formation of the automotive sector’s VAI did not mean a radical reduction of the role of the industry in the economy of these states—it continues to be relatively important. The continuing relocation of production is increasing the significance of this sector in the states of Parana, Rio Grande do Sul, Pernambuco, Bahia and Goias (Table 3.3; Fig. 3.1). The processes of relocation of the Brazilian automotive industry that began in the mid-1990s demonstrate a lasting trend towards further shifts, as proven by announcements of motor companies regarding the construction of new plants away from the previously dominant São Paulo and Minas Gerais. It is important to note, however, that far-reaching relocations of the sector production are restricted by the
3 The Relocation of the Automotive Industry in Brazil and Mexico …
41
Fig. 3.1 Development of the automotive industry in the selected Brazilian states. Source Own compilation based on data from: IBGE (2016a)
distribution of the population, mostly concentrated in the states of the Southeast region. As Brazil is unevenly populated, one of the highest-potential directions of the relocation of the automotive industry might be the states of the Northeast region. These are characterised by a relatively high population potential, and growing as a result of continuing economic development, shedding the stigma of the country’s poorest region. The first corporation to invest in this region was Ford, which constructed plants in the states of Bahia and Ceara. In recent years, Fiat has also shown an interest in the region, opening a new factory in Pernambuco. Another region with a good development potential is the South, with a relatively large population, more affluent than average. It attracts new automotive investments not only with its receptive market, but also thanks to its proximity to the markets of the neighbouring Mercosur states. The states of southern Brazil are particularly attractive to companies planning expansion into the receptive Argentinian market or wishing to integrate their production in the two countries. The third high-potential
42
M. Wójtowicz
Table 3.3 Participation of selected states in value added in industry (VAI): 1999–2014 State
Share of each state in the Brazilian automotive VAI (%) 1999 2009 2014
Share of automotive industry in the VAI of each state (%) 1999
2009
Amazonas 0.18 0.25 0.25 0.43 0.67 Ceará 0.14 0.11 0.13 0.66 0.78 Pernambuco 0.22 0.03 0.82 1.07 0.23 Bahia 0.02 4.35 2.66 0.04 10.31 Minas Gerais 14.64 13.57 12.71 9.27 13.38 Rio de Janeiro 4.13 4.51 4.07 2.85 4.33 São Paulo 62.35 53.65 48.52 8.22 14.25 Paraná 7.85 11.84 16.64 8.46 16.33 Santa Catarina 3.45 1.63 2.50 4.22 3.47 Rio Grande do Sul 6.79 8.22 9.67 4.93 11.82 Goiás 0.05 1.65 1.71 0.26 7.23 99.83 99.83 99.69 6.10b 10.06b Totala Source Own compilation based on data from: IBGE (2016b) a Totals do not reach 100% because of taking into account only selected states b Share of automotive industry in the VAI of whole country
2014 0.58 0.69 3.25 4.91 8.40 2.79 10.76 17.12 3.36 9.97 4.94 7.43b
relocation direction is the states of the Central-West region, and especially the highest-populated one, Goias. This region has enjoyed dynamic population growth and economic development since the 1970s, making it an attractive site for new investments in the long term. The least attractive region for the motor industry is the North. It is still characterised by a relatively low population and sparse road network, making it unattractive to automotive investments. An exception in this region is Manaus in Amazonas state, where the Indian company Mahindra produces off-road and commercial vehicles.
3.4
Current Development and Relocation of the Mexican Automotive Industry
The northward shift of the Mexican automotive industry began in the mid-1980s with the aforementioned wave of new investments in the production of car engines as well as car factories based on new production techniques borrowed from Japanese corporations (Shaiken 1994; Mortimore 1995). The process of development and modernisation of the production potential of the Mexican automotive sector accelerated following the country’s accession to NAFTA. The development strategy also changed at this time, applying the concept of GVCs. As it was geared mainly towards export production for the American and Canadian markets, new plants were opened closer to the US border. Along with
3 The Relocation of the Automotive Industry in Brazil and Mexico …
43
moving closer to the new main market, the second factor responsible for relocation was the desire to shift production from the heavily industrialised regions of Central Mexico, with their strongly unionised and therefore more expensive workforce, to less industrialised areas. Here there were large numbers of potential workers with lower wage demands and accepting the working conditions applying in modern plants, which aimed for the production levels achieved by Japanese corporations (Carrillo 1991; Shaiken 1994; Mortimore 1995; Moreno-Brid 1996). The sharp influx of automotive investments to Mexico in the 1980s, triggered by changes in the industrial policy regarding this sector as well as the economic integration within NAFTA, resulted in the following decade not only in a significant shift in activity towards the north but also in a dynamic increase in employment. A significant shift in the spatial distribution of the industry could also be observed. This is shown by a relatively high value of the coefficient of redistribution at 0.226, suggesting much larger changes compared to Brazil (Table 3.4). A particularly dynamic increase in employment was observed in car parts production, with only small changes in its spatial distribution shown by the low value of the coefficient of redistribution. The reverse situation was observed in the case of manufacturing vehicle bodies and trailers, in which the lowest employment growth was recorded, as well as the highest level of changes in spatial distribution. In car production, the coefficient of redistribution demonstrated a considerably larger relocation of employment owing to the construction of new production plants to the north of the capital (Table 3.4). The motor industry in Mexico in the initial period of development demonstrated a strong spatial concentration in the states located near the important market of the capital. From the 1980s onwards, however, a gradual dispersion could be observed, although in 1999 the Federal District and the four states of the central region surrounding it (Mexico, Morelos, Puebla and Hidalgo) still focused 67.6% of the total workforce in the vehicle production. However, increasing specialisation in
Table 3.4 Evolution of employment and coefficient of redistribution in particular sections of the automotive industry in Mexico: 1999–2014 Industry
Employment in 1000s 1999
2009
2014
Change 1999 = 100% 2009 2014
Automotive 419.2 462.0 722.0 110.2 industry Manufacture of 50.5 50.9 75.0 100.8 cars and trucks Manufacture of 18.1 20.8 23.4 114.9 bodies and trailers Manufacture of 350.5 441.2 623.6 125.9 parts for motor vehicles Source Own compilation based on data from: INEGI (2016)
Coefficient of redistribution Qr (1999–2014)
172.2
0.226
148.5
0.247
129.3
0.290
177.9
0.237
44
M. Wójtowicz
Table 3.5 Participation of selected states in total employment of the Mexican automotive industry: 1999–2014 State
Automotive industry
Manufacture of cars and trucks
1999
1999
2009
2014
2009
2014
Manufacture of parts and accessories for motor vehicles 1999 2009 2014
Aguascalientes 3.5 2.9 3.8 5.0 7.5 9.4 2.4 2.5 Baja California 3.8 2.2 3.7 3.7 4.8 4.1 2.2 1.7 Chihuahua 5.7 20.0 18.6 0.0 0.0 0.0 35.4 23.4 Coahuila 7.8 10.9 14.7 10.9 11.8 9.2 11.8 11.6 Distrito Federal 8.1 2.6 2.4 9.4 6.0 5.8 2.6 2.0 Guanajuato 7.1 4.7 5.5 6.4 6.8 10.5 1.9 3.5 Hidalgo 1.3 0.3 0.1 3.0 0.5 1.4 0.3 0.2 Jalisco 6.6 2.8 2.5 0.8 2.0 3.2 2.4 2.2 Mexico 20.4 9.2 5.2 23.3 17.7 12.2 8.9 7.7 Morelos 1.7 1.0 1.2 5.0 4.7 6.0 0.0 0.4 Nuevo Leon 7.3 8.0 7.5 0.7 0.5 7.0 8.0 8.9 Puebla 10.6 6.7 4.6 26.9 27.1 20.3 3.5 4.9 Queretaro 3.7 4.5 3.8 0.0 0.4 0.9 2.5 5.0 San Luis Potosi 2.5 4.3 3.5 0.1 1.8 3.3 0.9 4.5 Sonora 1.6 6.1 5.5 4.6 8.4 6.8 5.0 6.2 Tamaulipas 5.9 8.4 9.2 0.0 0.0 0.0 9.7 9.4 97.5 94.7 91.9 99.9 99.9 99.9 97.5 94.2 Totala Source Own compilation based on data from: INEGI (2016) a Totals do not reach 100% because of taking into account only selected states
2.6 2.8 19.1 18.2 1.3 5.0 0.1 1.4 5.5 0.4 8.2 3.8 5.1 4.4 5.3 8.8 92.1
production for the US market lowered the share of the central region to 45.7% in 2014 (Table 3.5; Fig. 3.2). A different spatial concentration was evident for car parts production, which adapted to exports to the USA sooner and became concentrated in states lying along the northern border. This was also connected to development in the border area of factories specialising in the manufacture of technically uncomplicated but labour-intensive products (the so-called maquiladora industry) intended wholly for export (Carrillo et al. 2005). Even in 1999, 72.1% of the workers in this branch were focused in the six border states (Baja California, Sonora, Chihuahua, Coahuila, Nuevo Leon and Tamaulipas). By 2014, however, this figure had dropped to 62.4%, as a result of some of the factories moving to the states of the Bajio region (Aguascalientes, Guanajuato, Jalisco, Queretaro and San Luis Potosi). This region therefore increased its share in the concentration of employment in the production of car parts from 10.1 to 18.8% (Table 3.5). The aforementioned six border states, which became less important in terms of the production of car parts, increased their share in employment in vehicle production considerably, from 19.9 to 27.1%, while their share in employment in the
3 The Relocation of the Automotive Industry in Brazil and Mexico …
45
Fig. 3.2 Development of the automotive industry in the selected Mexican states. Source Own compilation
motor industry as a whole also increased, from 32.1 to 59.2% (Table 3.5). This therefore confirmed the fact that this region has become the country’s most important in terms of the concentration of activity of the automotive sector. It is also worth mentioning the states lying between the central and the northern regions, which are conventionally included in the Bajio region. As mentioned above, in recent times they increased their share in the concentration of employment in car part production significantly. This was largely related to the opening of new factories manufacturing cars in these states, which attracted subcontractors cooperating with them. In the researched period, this region increased its share in employment in car production from 12.3 to 27.3% of the total (Table 3.5). The relocation of automobile manufacture from the central region towards the northern region led to its greater share in the gross value added (GVA) in the automotive industry. The share of the northern region share grew from 31.6 to 50.7%, while the other two regions marked reductions, from 35.9 to 19.2% in Bajio and from 31.4 to 28.9% in the central region. In the latter case, this was linked to the strong restructuring of the automotive sector in the Mexico state, with a large amount of production being transferred outside this state. In the Bajio region, most of the states, especially Guanajuato, lost some of their high value-added production despite the increased significance of employment in the automotive sector (Covarrubias 2011; Carbajal 2015). In the central region, only Puebla significantly increases its share in the automotive industry GVA. This was linked to VW’s strategy of modernisation and expansion of the existing plants, rather than relocation of production outside the state (Juárez 2005). The success of this strategy was an outcome of harmonious
46
M. Wójtowicz
Table 3.6 Participation of selected states in the gross value added (GVA): 1999–2014 State
Share of each state in the Mexican automotive industry GVA (%) 1999 2009 2014
Share of automotive industry in the GVA of each state (%) 1999
Aguascalientes 5.5 7.5 4.1 26.6 Baja California 1.9 2.5 3.3 2.6 Chihuahua 2.8 10.4 8.2 3.7 Coahuila 11.4 15.7 15.1 14.6 Distrito Federal −4.1 1.0 0.1 −0.7 Guanajuato 19.9 6.5 6.2 26.2 Hidalgo 0.3 0.2 0.1 2.1 Jalisco 3.8 2.2 1.9 2.8 Mexico 20.2 10.8 9.8 9.1 Morelos 3.4 1.9 2.5 17.1 Nuevo Leon 5.8 7.5 10.1 3.7 Puebla 11.6 14.0 16.4 19.6 Queretaro 4.9 4.6 2.3 12.3 San Luis Potosi 1.8 2.8 3.7 5.4 Sonora 7.5 6.4 11.7 13.4 Tamaulipas 2.2 4.2 2.3 3.5 98.9 98.3 97.8 4.6b Totala Source Own compilation based on data from: INEGI (2016) a Totals do not reach 100% because of taking into account only selected b Share of automotive industry in the VAI of the whole country
2009
2014
35.7 4.2 16.0 22.6 0.2 8.9 1.4 1.8 6.8 10.7 4.6 23.4 11.5 8.4 11.2 5.7 4.2b
30.4 9.2 21.5 26.5 0.0 10.9 0.7 2.4 9.0 18.6 7.5 34.8 8.0 13.9 23.3 5.1 6.1b
states
collaboration of VW managers, local authorities and trade unions. In terms of its share in employment, Puebla became considerably less significant on the national scale as a result of the slower growth of jobs compared to other states (Table 3.5). As for share in GVA production, however, it observed a marked increase. This also affected the increasing importance of the automotive industry in the economy in Puebla, reaching almost 35% of its GVA in 2014. This sector was also very important for the economy in the states of Aguascalientes, Coahuila, Chihuahua and Sonora, providing between 20 and 30% of their GVA in 2014 (Fig. 3.2). In the period in question, these states gained a large inflow of new investments within the sector, making it one of the crucial components of their economy (Table 3.6).
3.5
Conclusion
Since the 1990s, there have been significant changes in the spatial distribution of production plants of the Brazilian and Mexican automotive industries, caused by a number of coinciding factors, which can be divided into three main groups. The
3 The Relocation of the Automotive Industry in Brazil and Mexico …
47
first of these encompasses factors associated with changes in the industrial policies of the two countries. In most general terms, this entailed abandoning a strategy of industrialisation via import substitution in favour of the liberalisation of trade exchanges, as well as a more pro-export orientation of industrial production and improved competitiveness in global markets. The democratisation and decentralisation of power has given more competences to regional and local authorities in terms of fiscal and development policy, leading state and city authorities to actively encourage international companies to choose them as sites for their new investments by offering various tax reliefs, incentives and administrative short cuts. The regional economic integration within NAFTA—and in the case of Mexico also Mercosur—has, particularly for Brazil, also had an impact on decisions regarding sites for new automotive sector investments. The second group of factors concerns global changes in the industry, manifested in growing competition between TNCs. One strategy employed by TNCs in order to increase their competitive advantage is delocalisation or relocalisation of the modules of their value chain. Contemporary research shows that the automotive industry is a sector that makes extensive use of the concept of GVC in its development strategies for the coming decades (Sturgeon et al. 2008; Biesebroeck and Sturgeon 2010). This is because the progress of globalisation both increases competition and encourages finding areas of possible cooperation. On the one hand, rival car companies try to conquer and make a mark on all prospective markets. On the other, though, increasing competition forces them to relocate modules of the value chain, guaranteeing lower production costs while retaining access to the market. Owing to the increased importance of Brazil and Mexico as high-potential markets for new cars, in the last two decades TNCs previously absent on these markets have invested there. These were examples of relocation on global scale, because the creation of new production clusters in those countries was accompanied by production cuts or factory closures in developed countries. New production sites were becoming regionally important because car producers and their first-tier suppliers, following the GVC logic, frequently created large industrial condominiums, enhancing the position of those states which absorbed these investments. The third group of factors was related to changes in the development strategies of various TNCs which had been present in the countries for a long time. These corporations were forced to restructure their production plants, adapting them to the emerging technological changes, new global conditions and growing pressure for reduced production costs, or to relocate their activity to new places with good accessibility to qualified, less unionised workforce, as well as proximity to new regional markets. In both cases—restructuring and new plants—car assemblers renew their cooperation with major suppliers adapting GVC-engaged strategies. Most of these suppliers are global suppliers operating large facilities for both export and shipment to domestic assembly plants. With these three groups of factors in mind, we can attempt to reconstruct the conditions which have led to the spatial changes in the distribution of the automotive industry in Mexico and Brazil. Until the 1980s, the strategy of import substitution industrialisation pursued in both countries encouraged the development of a motor
48
M. Wójtowicz
industry tightly concentrated around the main economic centres. From the 1990s onwards, new investments in the sector in Brazil began to relocate to the adjoining states: Rio de Janeiro and Parana. These were usually made by newcomer corporations (Renault, PSA, Audi) searching for locations relatively close to the main market outlets. The American manufacturers which had been present in Brazil for a long time transferred their investments to peripheral regions with a high development potential, such as Bahia—the most populous state in the Northeast region—or Rio Grande do Sul—a dynamically developing state in the South region also offering attractive conditions for expansion to the neighbouring Mercosur countries. In the first decade of the twenty-first century, new investments were also directed inland, mostly to the states of Goias (Mitsubishi, Hyundai) and Amazonas (Mahindra). In the case of Mexico, the relocation of the automotive industry mostly headed north, as a result of the increasing orientation of the sector towards the US market. Border states, as well as the Bajio region halfway between the capital and the border gained particular importance. The central region—the cradle of the Mexican car industry—underwent profound restructuring and modernisation following the loss of a section of its production potential. This was particularly acutely visible in the case of the diminishing importance of the Federal District and the states of Mexico and Hidalgo. The changing production strategies of TNCs meant that the car factories remaining in the central region now mostly specialise in production for the domestic market. In conclusion, the automotive industry, which developed in the two countries mainly on the basis of TNC investments, in the first development phase under the strategy of ISI, demonstrated a very similar pattern of spatial activity concentration. In both countries, the later change in the global and regional circumstances of the sector operation triggered a far-reaching process of relocation production plants of the industry. From a GVC perspective, the directions of these shifts were strongly determined by the different role which the states played in the regional manufacture systems and the TNCs’ strategies of a regional pattern of the GVC organisation in the automotive industry.
References Amann, E., Haddad, E., Perobelli, F., & Guilhoto, J. (2007). Structural change in the Brazilian automotive industry and its regional impacts. Latin American Business Review, 7(3–4), 97– 119. https://doi.org/10.1300/J140v07n03_05. ANFAVEA. (2003). Anuário Estatístico da Indústria Automobilística Brasileira. Associação Nacional dos Fabricantes de Veículos Automotores – ANFAVEA, São Paulo. Azevedo, A. F. Z., & Massuquetti, A. (2015). Exports from the Brazilian automotive sector to the southern common market: Trade diversion or cost reduction? CEPAL Review, 115, 145–163. Bennett, D. C., & Sharpe, K. E. (1985). Transnational corporations versus the state: The political economy of the Mexican auto industry. Princeton: Princeton University Press. Biesebroeck, J. V., & Sturgeon, T. (2010). Effects of the 2008–09 crisis on the automotive industry in developing countries: A global value chain perspective. In O. Cattaneo, G. Gereffi, & C. Staritz (Eds.), Global value chains in a postcrisis world: A development perspective (pp. 206– 244). Washington: The World Bank.
3 The Relocation of the Automotive Industry in Brazil and Mexico …
49
Carbajal, Y. S. (2015). Evolución, condiciones actuales y retos del sector automotriz en México y en el Estado de México. Toluca: UAEM. Carrillo, J. (1991). Restructuración en la industria automotriz en México. Estudios Sociológicos, 9 (27), 483–525. Carrillo, J. (2004). NAFTA: The process of regional integration of motor vehicle production. In J. Carrillo, Y. Lung, & R. V. Tulder (Eds.), Cars, carriers of regionalism? (pp. 104–117). Houndsmills-London: Palgrave Macmillan. Carrillo, J., Hualde, A., & Quintero, C. (2005). Maquiladoras en México. Breve recorrido histórico. Comercio Exterior, 55(1), 30–42. Carrillo, J., Lung, Y., & Tulder, R. V. (Eds.). (2004). Cars, carriers of regionalism?. Houndsmills-London: Palgrave Macmillan. Covarrubias, A. V. (2011). The Mexican auto industry: From crisis to greater region-centric influence. Norteamerica, 6(2), 115–155. Cruz, M. J. V., & Rolim, C. F. C. (2010). The Brazilian automotive industry in the BRICs context: The case of the metropolitan region of Curitiba. Cambridge Journal of Regions, Economy and Society, 3, 319–334. https://doi.org/10.1093/cjres/rsq027. Gereffi, G. (2014). Global value chains in a post-Washington consensus world. Review of International Political Economy, 21(1), 9–37. https://doi.org/10.1080/09692290.2012.756414. Gereffi, G., Humphrey, J., & Sturgeon, T. (2005). The governance of global value chains. Review of International Political Economy, 12(1), 78–104. https://doi.org/10.1080/09692290500049805. Hasegawa, H. L., Venanzi, D., & Silva, O. R. (2013). Modular consortium (MC) and industrial condominium (IC) in Brazil’s automotive industry. Brazilian Journal of Operations & Production Management, 12(2), 360–367. https://doi.org/10.14488/BJOPM.2015.v12.n2.a15. Holmes, J. (1993). From Three industries to one: Towards an integrated North American Automobile Industry. In M. A. Molot (Ed.), Driving continentally: National policies and the North American auto industry (pp. 23–61). Ottawa: Carleton University Press. Humphrey, J., Lecler, Y., & Salerno, M. S. (Eds.). (2000). Global strategies and local realities: The auto industry in emerging markets. Houndsmills-London: Macmillan. IBGE. (2016a). Cadastro Central de Empresas—CEMPRE 1999–2014. Instituto Brasileiro de Geografia e Estatística—IBGE. Rio de Janeiro. https://sidra.ibge.gov.br/pesquisa/cempre/ tabelas/brasil/2014. Accessed October 10, 2016. IBGE. (2016b). Pesquisa Industrial Anual—Empresa. Instituto Brasileiro de Geografia e Estatística—IBGE. Rio de Janeiro. https://sidra.ibge.gov.br/pesquisa/pia-empresa/tabelas/ brasil/2014. Accessed September 20, 2016. INEGI. (2016). Censos Económicos 1999, 2009, 2014. Instituto Nacional de Estadística, Geografía e Informática—INEGI, Mexico. http://www.inegi.org.mx/est/contenidos/Proyectos/ce/Default. aspx. Accessed September 17, 2016. Juárez, N. H. (2005). Puebla, región anfitriona. Volkswagen, empresa lider. Un agrupamiento automotriz. In: H. N. Juárez, A. R. Lara, & C. C. Bueno (Eds.), El auto global: Desarollo, competencia y cooperación en la industria del automóvil (pp. 169–207). Benemeríta Universidad Autonóma de Puebla, Universidad Autonóma Metropolitana – Xochimilco, Universidad Iberoamericana Ciudad de México, Consejo Nacional de Ciencia y Tecnología, Puebla. Laplane, M. F., & Sarti, F. (2004). MERCOSUR: Interaction between governments and producers and the sustainability of the regional automobile industry. In J. Carrillo, Y. Lung, & R. V. Tulder (Eds.), Cars, carriers of regionalism? (pp. 121–138). Houndsmills-London: Palgrave Macmillan. Malamud, A. (2012). Moving regions: Brazil’s global emergence and the redefinition of latin American borders. In P. Riggirozzi & D. Tussie (Eds.), The rise of post-hegemonic regionalism. The case of Latin America (pp. 167–182). Dordrecht-Heidelberg-London-New York: Springer. Marx, R., & Mello, A. M. (2014). New initiatives, trends and dilemmas for the Brazilian automotive industry: The case of Inovar Auto and its impacts on electromobility in Brazil. International Journal of Automotive Technology and Management, 14(2), 138–157. https://doi. org/10.1504/IJATM.2014.060751.
50
M. Wójtowicz
Moreno-Brid, J. C. (1996). Mexico’s auto industry after NAFTA: A successful experience in restructuring? Kellogg Institute, Working Paper 232. https://www3.nd.edu/*kellogg/ publications/workingpapers/WPS/232.pdf. Mortimore, M. (1995). Transforming sitting ducks into flying geese: The Mexican automobile industry. UN Economic Commission For Latin America and the Caribbean, Santiago de Chile. Pavlínek, P., & Ženka, J. (2011). Upgrading in the automotive industry: Firm-level evidence from Central Europe. Journal of Economic Geography, 11(3), 559–586. Pires, S. R. I., & Neto, M. S. (2008). New configurations in supply chains: The case of condominium in Brazil’s automotive industry. Supply Chain Management: An International Journal, 13(4), 328–334. https://doi.org/10.1108/13598540810882215. Quadros, R., & Consoni, F. (2009). Innovation capabilities in the Brazilian automobile industry: A study of vehicle assemblers’ technological strategies and policy recommendations. International Journal of Technological Learning, Innovation and Development, 2(1/2), 53– 75. https://doi.org/10.1504/IJTLID.2009.021956. Rodriguez-Pose, A., & Arbix, G. (2001). Strategies of waste: Bidding wars in the Brazilian automobile sector. International Journal of Urban and Regional Research, 25(1), 134–154. https://doi.org/10.1111/1468-2427.00302. Rodriguez-Pose, A., Tomaney, J., & Klink, J. (2001). Local empowerment through economic restructuring in Brazil: The case of the greater ABC region. Geoforum, 32(4), 459–469. https:// doi.org/10.1016/S0016-7185(01)00011-2. Saraiva, M. G. (2010). Brazilian foreign policy towards South America during the Lula Administration: Caught between South America and Mercosur. Revista Brasileira de Política Internacional, 53(special edition), 151–168. Shaiken, H. (1994). Advanced manufacturing and Mexico: A new international division of labor? Latin American Research Review, 29(2), 39–71. Shapiro, H. (1993). The determinants of trade and investment flows in LDC auto industries: The cases of Brazil and Mexico. In M. A. Molot (Ed.), Driving continentally: National policies and the North American auto industry (pp. 101–134). Ottawa: Carleton University Press. Shapiro, H. (1994). Engines of growth: The state and transnational auto companies in Brazil. Cambridge: Cambridge University Press. Sturgeon, T., & Biesebroeck, J. V. (2009). Crisis and protection in the automotive industry: A global value chain perspective. In S. J. Evenett, B. M. Hoekman, & O. Cattaneo (Eds.), Effective crisis response and openness: Implications for the trading system (pp. 285–305). Washington: The World Bank. Sturgeon, T., & Biesebroeck, J. V. (2011). Global value chains in the automotive industry: An enhanced role for developing countries? International Journal of Technological Learning, Innovation and Development, 4(1/2/3), 181–205. https://doi.org/10.1504/IJTLID.2011.041904. Sturgeon, T., Biesebroeck, J. V., & Gereffi, G. (2008). Value chains, networks and clusters: Reframing the global automotive industry. Journal of Economic Geography, 8(3), 297–321. https://doi.org/10.1093/jeg/lbn007. Sturgeon, T., & Florida, R. (2004). Globalization, deverticalization, and employment in the motor vehicle industry. In M. Kenny & R. Florida (Eds.), Locating global advantage: Industry dynamics in a globalizing economy (pp. 52–81). Palo Alto CA: Stanford University Press. Wójtowicz, M., & Rachwał, T. (2014). Globalization and new centres of automotive manufacturing—The case of Brazil, Mexico, and Central Europe. Prace Komisji Geografii Przemysłu Polskiego Towarzystwa Geograficznego, 25, 81–107.
3 The Relocation of the Automotive Industry in Brazil and Mexico …
51
Mirosław Wójtowicz holds Ph.D. in socio-economic geography, obtained from the Jagiellonian University. He is Assistant Professor in the Department of Socio-Economic Geography at the Pedagogical University of Kraków (Poland). His research interests focus on issues of urbanisation and industrialisation process in Latin America, with particular emphasis on Argentina, Brazil and Mexico. His other interests are suburbanisation process in Kraków Metropolitan Area. In the years 2013–2015, he was a part of a team that conducted research on this issue, funded by the National Science Center.
Chapter 4
Industrial Upgrading and Downgrading of Export-Oriented Furniture Firms in the Market Reorientation Towards China’s Domestic Market Tianlan Fu and Chun Yang
4.1
Introduction
Global value chains (GVCs) and global production networks (GPNs) theories have been widely used to explain the structural and geographical organisation of fragmented production, distribution, and consumption in the interconnected world (Coe et al. 2004; Coe et al. 2008; Gereffi et al. 2005). In the theory of GVCs, the concept of upgrading is used to account for the ways developing country supplier firms enhance their positions in global production networks or value chains (Gereffi et al. 2005; Humphrey and Schmitz 2002). Most studies have explored the upgrading prospects of developing country suppliers in Western market-oriented production networks (Bair and Gereffi 2003; Frederick and Gereffi 2011). In these production networks, global lead firms or large buyers exercise great control over core competency activities, such as marketing, retail, and design, while developing country suppliers have gained limited upgrading opportunities. Recent studies showed that in the rise of emerging markets, such as China and India, developing country suppliers have turned to strategic recoupling with emerging markets and gained more functional upgrading prospects (Horner 2014; MacKinnon 2012; Yang 2014). Therefore, Blažek (2015) emphasised the needs to explore the changing dynamics of GPNs and the effects of market reorientation on upgrading opportunities of these suppliers in developing countries. Kadarusman and Nadvi (2013) also called for more emphasis on other alternative ways of upgrading in domestic markets. They highlighted the role of local agencies in reshaping industrial upgrading trajectories of local supplier firms in their market shifts (Kadarusman and Nadvi 2013). Moreover, recent studies T. Fu School of Geography and Planning, Sun Yat-sen University, Guangzhou, China e-mail:
[email protected] C. Yang (&) Department of Geography, Hong Kong Baptist University, Kowloon Tong, Hong Kong e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_4
53
54
T. Fu and C. Yang
criticise the fact that the existing research on industrial upgrading has overwhelmingly simplified the paths of upgrading, in which firms always pursue the enhancement of their functional positions in the value chains or production networks while neglecting the diversity of value capture trajectories (Coe and Yeung 2015). Ponte and Ewert (2009) add that analytical attention should be moved beyond normal upgrading and account for different types of upgrading as well as downgrading which entail more significant meaning for firms in developing countries in the changing global economy. Therefore, drawing upon the perspective of GPNs, this chapter attempts to explore industrial upgrading or downgrading trajectories of supplier firms in their market reorientation towards the domestic markets of China. Taking the Pearl River Delta (PRD), particularly Dongguan, one of the most well-known furniture manufacturing clusters in China, as the case, this chapter attempts to explore the industrial upgrading and downgrading trajectories of furniture firms in their market reorientation from exports to domestic sales in China. A particular focus is put on the interaction between the ways in which furniture firms have integrated into domestic market-oriented production networks and industrial upgrading or downgrading trajectories of furniture firms. Based on the intensive field investigation, particularly in-depth interviews with CEOs or marketing managers of furniture firms, the chapter argues that market reorientation towards China’s domestic markets has stimulated industrial upgrading and downgrading trajectories of furniture firms. Some competitive furniture firms achieved functional upgrading when they established own brand names for cooperating with domestic chain retailers which are familiar with the higher end domestic consumers. Meanwhile, some experienced downgrading as they passively responded to the higher requirements of domestic consumers. The chapter adds the literature by incorporating market dynamics in the studies of industrial upgrading or downgrading. It aims to understand the role of emerging markets (e.g. China and India) in reshaping the industrial upgrading or downgrading trajectories in export-oriented industries in developing countries. Following this introduction, this chapter critically revisits the industrial upgrading and downgrading dynamics in the changing global economy. Then, an introduction to the export-oriented furniture industry in the PRD and research methodology is presented. The chapter then explores the ways in which furniture firms achieved industrial upgrading or experienced downgrading in their market reorientation. The chapter concludes with the summary of major findings and theoretical and practical implications.
4.2
Revisiting Industrial Upgrading and Downgrading in the Rise of Emerging Markets in the Global South
In the theory of GVCs and GPNs, industrial upgrading is defined as a move adopted by firms to improve technological competency and to increase profits from participating in value chains or production networks (Gereffi 2009; Gereffi et al. 2005;
4 Industrial Upgrading and Downgrading of Export-Oriented …
55
Kaplinsky and Readman 2005). Four types of industrial upgrading have been identified: (1) process upgrading means that manufacturing firms use superior technology to improve efficiency of the manufacturing process of transforming inputs into outputs; (2) product upgrading refers to introducing by firms more sophisticated product lines to improve the quality of products; (3) functional upgrading regards adopting new functions by firms to enhance their overall skill content; (4) inter-sectoral upgrading means that firms move into other new kinds of productive activities (Humphrey and Schmitz 2002). Industrial upgrading opportunities vary in terms of the way in which local producers fit into global value chains and the coordination relations they articulate with global lead firms (Humphrey and Schmitz 2002). Integrating into quasi-hierarchical relations offers favourable conditions for process and product upgrading but not functional upgrading. However, plugging in market-based relations is conducive for functional upgrading, but it impedes process and product upgrading (Humphrey and Schmitz 2000). For example, local shoe producers in Brazil gained industrial upgrading opportunities through cooperation with global buyers and global lead firms in the form of quasi-hierarchical relation, but they experienced difficulties in achieving functional upgrading opportunities (Humphrey and Schmitz 2002). Similarly, local automotive producers in the Czech Republic achieved process upgrading based on attempts to enhance overall efficiency and productivity, while they encountered constrains on functional upgrading which is limited to global lead firms or higher-tier suppliers (Pavlinek and Zenka 2010). In the rise of emerging markets, such as China and India, recent studies have paid increasing attention to the market shifts of export-oriented industries in developing countries (Cattaneo et al. 2010; Gereffi 2014). The export-oriented supplier firms in developing countries such as China have increasingly integrated into production networks driven by emerging markets (Horner 2014; MacKinnon 2012). Horner (2014) indicates that integrating into domestic market-oriented production networks enabled local pharmaceutical producers in India to gain more functional upgrading prospects. Kadarusman and Nadvi (2013) argue that cooperation with local agencies helped local suppliers acquire and exploit innovative capabilities towards high-skilled and complex functions. From the study of apparel, home appliance and automobile industry in the PRD, China, Liu (2016) suggests that integration into the huge domestic market of China provides alternative ways for local firms to accumulate profits, improve innovation capabilities, and practice marketing and branding. It also enables local firms to coordinate their production networks without violating their cooperation with their global counterparts. Overall, these studies have highlighted the importance of emerging markets (e.g. China and India) in reshaping industrial upgrading trajectories of local suppliers in developing countries. Blažek (2015) indicates that changing market dynamics of GPNs or GVCs have broadened the scope of industrial upgrading analysis. Nevertheless, how integration into emerging markets influences the upgrading or downgrading trajectories of export-oriented suppliers has not received enough attention in the existing literature. Moreover, scholars criticise that the existing research on industrial upgrading has overwhelmingly simplified the trajectories of industrial upgrading (Coe and Yeung
56
T. Fu and C. Yang
Fig. 4.1 Upgrading trajectories of export-oriented firms in their market reorientation. Source Own compilation
2015). The previous explanation of upgrading primarily suggests one possible outcome of many in the participation of firms into new production networks (Coe and Yeung 2015). However, recent studies indicate that upgrading trajectories is more complicated than simply turning to new markets or getting financial supports (Tokatli 2013; Selwyn 2012). Notably, integration into emerging markets might lead to functional downgrading (Ponte and Ewert 2009). Downgrading means that local producers intend to lower the quality of products, thus expanding to lower value-added activities (Smith et al. 2014; Plank and Staritz 2015). A recent study indicates that suppliers experienced economic downgrading when they disintegrated with more stringent higher-wage advanced markets and then recoupled with lower-wage domestic markets in developing countries (Barrientos et al. 2016). Kaplinsky and Farooki (2011) also argue that decupling with EU and recoupling with China’s market undermine the upgrading prospects of export-oriented suppliers in Thailand’s cassava value chain and Gabon’s tropical timber value chain. These studies reflect the effects of market demand from emerging markets on the industrial upgrading opportunities, which have received not enough attention. Different from the market demand of Western advanced economies (e.g. EU and the US), consumers in emerging markets have low requirements in relation to product quality and standards (Yeung and Coe 2015). Lower barriers to enter the markets and less stringent standards in low-income markets decreased the level of value added in products and slimmer margins. However, less has been done to consider the distinct market demands in emerging markets and understand the influence of new market demands on the upgrading as well as downgrading trajectories of local suppliers in developing countries. To fill in the research gaps presented above, this chapter develops an analytical framework to explore the industrial upgrading or downgrading trajectories of export-oriented supplier firms in their market reorientation from exports to domestic sales in China. The framework incorporates export-oriented supplier firms and local agencies which include firm actors (e.g. domestic retailers) and non-firm actors (e.g. local governments) (Fig. 4.1).
4 Industrial Upgrading and Downgrading of Export-Oriented …
4.3 4.3.1
57
Research Context and Method Changing the Dynamics of Export-Oriented Furniture Industry in the Pearl River Delta
Over the past three decades, the PRD has experienced remarkable export-oriented industrialisation (Fig. 4.2), with the export/GDP ratio increasing to 70.9% in 2013 (Guangdong Statistical Bureau 2014). Driven by a large flow of export orders and foreign investment from Hong Kong and Taiwan, numerous labour-intensive industries have dramatically developed in the PRD, including the furniture industry. In Guangdong Province, furniture exports reached US$9 billion, accounting for 40.5% of the national furniture exports in 2007 (China Statistical Bureau 2009; Guangdong Statistical Bureau 2008). In the PRD, Dongguan and Shenzhen City are major furniture exporters (Fig. 4.2). In particular, Dongguan alone recorded 20% of furniture exports of Guangdong Province in 2008, with exports reaching US$2.8 billion (Guangdong Statistical Bureau 2009; Dongguan Statistical Bureau 2010). In Dongguan, Dalingshan Town is well known as the famous furniture export town in China. In 2013, furniture exports contributed to more than 90% of the total furniture industrial outputs in Dalingshan (China National Furniture Association 2014). Data show that there were more than 500 furniture firms concentrated in Dalingshan Town in 2013 (China National Furniture Association 2014). In addition, Houjie Town, a well-known furniture exhibition town, also plays the key role in the development of the furniture industry in Dongguan and even the PRD. It is estimated that over 125,000 foreign and domestic buyers visited Houjie in 2015 during furniture exhibitions (Famous Furniture Fair Report 2015). Through the participation in the bi-annual furniture exhibition, export-oriented furniture supplier firms could reach foreign consumers and domestic consumers as well. However, the export-oriented furniture industry in the PRD has encountered a rapid decline of export orders in the aftermath of the 2008 global financial crisis, leading to the closure and downsizing of furniture firms. Data show that the export growth rate of the furniture industry in Guangdong decreased from 35.7% in 2010 to 3.2% in 2012 (Table 4.1). More than 300 enterprises were closed down during 2008–2012 (Guangdong Statistical Bureau 2009–2013). To mitigate the negative effects of the decline in exports, furniture firms turned to leveraging domestic sales in China. Subsequently, furniture domestic sales ratio in Guangdong increased from 56.6% in 2008 to 64.7% in 2011 (Table 4.1). Notably, rather than totally giving up exports, furniture firms preferred to achieve balance between domestic sales and exports through market reorientation. For the export-oriented furniture firms, China’s domestic market exhibits large consumption power. In China, domestic furniture sales increased to RMB 227.3 billion in 2014 from RMB 76.2 billion in
58
T. Fu and C. Yang
Fig. 4.2 Location of the Pearl River Delta in China
2000 (Castaño 2002), with an average annual rate1 of 41% (HKTDC 2014). Moreover, wood furniture is the most popular furniture products in China, which reached RMB 411.8 billion and contributed to 63.7% of national furniture sales in 2013. A rapid increase in domestic consumption benefits from the rise of middle class and their consumption habits. According to a recent investigation on furniture consumption carried out by Hong Kong Trade Development Council (HKTDC), more than 70% of the investigated China’s consumers prefer to spend more on purchasing or replacing their furniture than before (HKTDC 2015). Moreover, domestic consumers in China show diverse demands, which are different from those of buyers from export markets (e.g. EU and the US) who prefer simple style of furniture. Three types of consumers are categorised: (1) luxury consumers in the high-end segment who prefer high-quality products with brand names; (2) the middle class which pays more attention to ‘quality’ and ‘style’ in their purchase of furniture and care less about the price; (3) average wage-earning consumers who demand low-price furniture in the low-end market segment (HKTDC 2015).
1
The average rate refers to the growth rate of furniture sales outputs of wholesalers and retailers that above designed size. According to the National Bureau of Statistics of China, firms above designed size refer to the firms with the sales outputs of over RMB 20 million.
4 Industrial Upgrading and Downgrading of Export-Oriented …
59
Considering the diverse market demands in China’s market, this chapter will examine the ways in which furniture firms upgrade or downgrade in order to meet the requirements of domestic consumers.
4.3.2
Research Design and Data Collection
In this chapter, export-oriented furniture firms (hereafter ‘furniture firms’) refer to furniture manufacturing suppliers which previously targeted at export markets and then turned to embarking on the domestic market of China in recent years. With the expansion to the Chinese market, furniture firms might become domestic market-oriented furniture firms, if they increase domestic sales ratio to more than 50% of the total turnover. In this research, field investigation, particularly the in-depth interviews, was conducted to collect the first-hand data. In total, 36 firm-level interviews with CEOs or marketing managers of furniture firms were conducted during the period between March 2013 and September 2015. The firm-level interviews were primarily conducted during our visits to the bi-annual furniture exhibitions which were held in March and September in Dongguan and Shenzhen, respectively. At the furniture exhibition, furniture firms were placed in different exhibition zones, according to their ownership, market orientations, and major products, etc. To cover different kinds of furniture, the furniture firms from different exhibition zones were picked up to conduct interviews. The interviews with the CEOs or market managers were conducted during the furniture exhibition or by the follow-up phone interviews to understand the ways in which furniture firms facilitate domestic sales in China and their subsequent industrial upgrading or downgrading strategies. Totally, 36 interviews were completed. Each interview lasted 45 min to one hour. In the investigation, domestic retailers and E-commerce platform managers were also interviewed to understand how the furniture firms interact with domestic retailers and how their collaborations influence the upgrading or downgrading trajectories of furniture firms. Domestic retailers refer to firms or individual merchants who purchase finished furniture goods from furniture supplier firms and then distribute them to the final consumers. Moreover, local governments and industrial associations in Houjie and Dalingshan Town in Dongguan were also interviewed to understand the role of local institutions. The interviews focused on how institutions help and support the industrial upgrading and downgrading of furniture industries in recent years. In addition, the second-hand data were collected from statistical yearbooks and websites of the furniture firms to supplement our analysis. All of the interviews were recorded. The information from the interviews was coded in terms of upgrading and downgrading activities of furniture firms. Following the coding, industrial upgrading and downgrading trajectories of furniture firms were categorised and conceptualised.
60
T. Fu and C. Yang
Table 4.1 Domestic sales of furniture in Guangdong Province during 2007–2013 Year
Outputs (RMB billion)
Exports ($ billion)
Growth rate (%)
Domestic sales (RMB billion)
Share of domestic sales in China (%)
Domestic sales/ outputs (%)
2007 160 9 99.5 26.4 62.2 2008 184.7 10.2 14.1 104.5 56.6 2010 245 13.5 35.7 154 62.9 2011 280 14.9 10.4 182 24.4 65 2012 300 15.4 3.2 201 23 67 2013 339 17.5 13.2 222 65.5 2014 363 19.6 12.6 235 25 64.7 Source Own compilation, based on information from Guangdong furniture association and export data from Guangdong Statistical Yearbook 2008–2014
4.4
Industrial Upgrading and Downgrading Trajectories of Furniture Firms in the Market Reorientation Towards China’s Market
In this research, 66.7% of interviewed furniture firms started to tap into China’s market after 2008, while some started domestic sales even before 2008. In the market reorientation, furniture firms have conducted varied industrial upgrading strategies to facilitate the domestic sales. 72.2% of the interviewed firms (26/36) have achieved functional upgrading, while a small number of furniture firms experienced industrial downgrading (Table 4.2). The following discussion will elucidate the diffident upgrading or downgrading paths of furniture firms. More attention will be paid to the functional upgrading trajectories.
Table 4.2 Domestic sales strategies and upgrading or downgrading trajectories Domestic sales strategies
Upgrading or downgrading trajectories
% of total interviewed firms (36 firms)
I. Cooperation with domestic chain retailers II. Establish physical direct-sale stores
Functional upgrading through branding Functional upgrading through branding and retailing Product downgrading for online sales
26 firms (74%)
III. Establish online shops
17 firms (48%)
9 firms (26%) for online sales; some of these firms are branded firms who experienced downgrading 9 firms (26%)
IV. Selling products Passive attempts at directly to domestic upgrading individual retailers Source Own compilation based on the interviews with furniture firms
4 Industrial Upgrading and Downgrading of Export-Oriented …
4.4.1
61
Strategic Functional Upgrading Through Branding and Retail
According to the interviews, majority of the furniture firms have achieved functional upgrading by expanding to higher value-added activities such as branding and retail after they embarked on the Chinese market. To expand domestic sales, 41.7% (15/36) of the interviewed firms established brand names after 2008. 33.3% engaged in branding even before 2008. Most of these firms are medium- and large-scale firms.2 The move of establishing brand names has been driven by furniture firms’ desire to explore the high-end segment of China’s market in which consumers demand high-quality branded products. Producing and selling branded products in these market segments will significantly increase marginal profits of furniture firms. Explained by the CEO of a furniture firm in Houjie, marginal profit they gain from exports are around 5–10%. But the marginal profit could increase to 20% if they produced products with their own brand name and sell them in the domestic market. A senior marketing manager of a furniture firm in Shenzhen noted: ‘previously (before 2008), our company was one of IKEA’s suppliers. The marginal profit is very low in our cooperation with IKEA, around 5% of the ex-factory price. However, when we sell our branded products, we can increase the marginal profits to 20%. Thus, we prefer to establish our own brands and produce our own products’ (interview in Shenzhen, March 2015). The strategic needs of local partners, namely domestic chain retailers, are another imperative for furniture firms’ establishment of brand names. Due to long-termed engagement in exports, furniture firms have less knowledge about retail. The CEO of a furniture firm in Shenzhen explained: ‘as manufacturers we have no idea about retail or marketing’ (interview in Shenzhen, September 2015). As such, furniture firms preferred to distribute or sell their furniture products in domestic markets through cooperation with domestic chain retailers. In this research, domestic chain retailers are the retailers which primarily purchase products from furniture supplier firms and then sell the products in their localities (e.g. Hunan and Hubei Province). These retailers do not engage in production or design. Compared with furniture firms, domestic chain retailers are much more familiar with Chinese consumers and have skills in retail. Hence, cooperating with local partners has been recognised as the most promising way to establish and expand their sales networks in the domestic market. However, domestic chain retailers
2
According to the standard of small- and medium-sized enterprises designated by the Ministry of Industry and Information Technology of China, the employment of large-scale enterprises in the industry is more than 1000 and business incomes of these enterprises are more than RMB 400 million; the employment of medium-scale enterprises is more than 300 and business incomes of these enterprises are less than RMB 20 million. The employment of small-scale enterprises is less than 20, and business incomes of these enterprises are more than RMB 3 million.
62
T. Fu and C. Yang
tended to cooperate with branded suppliers who could provide high-quality branded products. By getting the franchise of branded suppliers, chain retailers could gain access to the high-end furniture shopping malls such as Red Star Macalline and Easyhome, and establish furniture stores therein. As marked by domestic chain retailer from Hunan Province, as the two well-known furniture shopping malls in China, Easyhome and Red Star Macalline, located in the first or second-tier cities (e.g., Beijing, Shanghai, and Shenzhen), have concentrated numerous high-end consumers. But these shopping malls have high entry barriers (interview in Dongguan, September 2015). Therefore, to meet the strategic needs of domestic chain retailers and then target the high-end domestic consumers, furniture firms have made great efforts in branding, innovation, and design. Firm F is a Hong Kong furniture firm specialising the wood furniture production. Take a famous furniture firm (Firm F3) in Dongguan as a case. Established in 1989, Firm F is a Hong Kong-invested furniture firm focusing on wood furniture production in Dalingshan Town, Dongguan City. Firm F engaged in processing production for global buyers in the past decades. In recent years, Firm F also developed four series products with their own brand names to increase their market share in China’s domestic market. To improve production flexibility and efficiency and lower production cost, the firm also purchased a series of robot arms from Italy, worth RMB 0.3 billion, and thus have developed automated furniture production lines since 2010. These robot arms have been employed to automate the work previously done by low-skill workers, including sawmilling, modelling, and painting. As a result, 30% of human workforce has been saved, which has largely reduced the production cost. Moreover, green timber, imported from Africa and European is used to produce eco-friendly wood furniture products which are favoured by domestic consumers. By September 2015, the domestic sales of this company increased to 50% from 30% of the total turnover before 2008. Firms also made partnerships with over 80 chain retailers after their launch in China’s market (interview in Dongguan, September 2015). The study indicates that integration into domestic market, particularly the high-end segment, forces furniture firms to bypass constrains of functional upgrading they encountered in the Western market-oriented production networks (Horner 2014; Liu 2016). In order to meet the requirements of local retail partners or domestic consumers in the high-end market, furniture firms attempted to establish brand names, thus gaining higher marginal products from domestic sales. The evidence implies the positive effects of market reorientation on industrial upgrading and highlights the importance of local agencies, i.e. domestic chain retailers in the facilitation of upgrading of furniture firms (Horner 2014). This scenario is different from the finding indicated by Kaplinsky and Farooki (2011) who pointed out that the exploration of China’s market undermines the upgrading prospects of export-oriented suppliers in Thailand and Gabon. The relatively
3
The name of the company is anonymous as requested by the interviewee.
4 Industrial Upgrading and Downgrading of Export-Oriented …
63
symmetric power relations between furniture firms and their domestic partners, i.e. domestic chain retailers, give the decision-making power on branding to furniture firms. From this point of view, the functional upgrading of furniture firms benefits from their ways to integrate into domestic market-oriented production networks and governance relations with domestic agencies. In addition, some furniture firms also engaged in retail activities after they had established brand names, indicating another type of functional upgrading. Some branded furniture firms who have strong financial capabilities tended to internalise retail activities by establishing direct-sales stores. According to the interviews, 47.2% (17/36) of furniture firms have established physical direct-sales stores to distribute their products directly to end consumers. These stores are primarily located in the first-tier cities, such as Beijing, Shanghai, and Guangzhou in China or the surrounding cities in the PRD, such as Shenzhen and Dongguan. For furniture firms, controlling domestic sales directly in important markets enables them to capture consumer demand and market trend directly and then quickly adjust their products in relation to design revision, quality improvement, material change, and price and cost adjustment. Furthermore, direct sales allow the reduction of management and retail agent cost. By engaging in direct sales, furniture firms established an intra-firm coordination governance relations in the domestic market. However, furniture firms found it difficult to gain profits from direct-sales stores. Due to the limited experience in retailing and marketing, such as furnishing and decorating stores or training salesmen, furniture firms have weak management of the direct-sales stores. Marketing Manager of a furniture firm in Foshan City claimed: We attempted to operate direct-sale stores by ourselves, but we were unable to make a profit due to the unskilful decoration and placement. As a producer, we are not good at retail activities. The established chain retailers can make good businesses. Retail is an art. (Interview in Dongguan, March 2015)
Take a Hong Kong-invested firm (Firm X4) in Shenzhen as a case. Since 2006, Firm X has established ten direct-sales stores in several first-tier cities (e.g. Beijing, Shanghai, and Guangzhou). However, due to unsuccessful retail strategy and unfamiliarity of local consumption culture, most of them failed to make a profit, leading to the termination of business outside Guangdong Province. Moreover, the high cost of establishing physical stores and high rent for stores also added burden to the furniture firms. For example, the cost of establishing a physical store in Shenzhen reached around RMB 500,000. And the monthly rent for stores in the high-end furniture shopping malls such as Red Star Macalline reach around RMB 100,000 in Shenzhen (interviews in Shenzhen, March 2015). The high cost makes furniture firms fail to make profits on direct-sales stores. Therefore, the study indicates it is easier for furniture firms to achieve functional upgrading by developing brand names and establishing inter-firm cooperation with local retail partners.
4
The name of the company is anonymous as requested by interviewee.
64
T. Fu and C. Yang
In addition, institutions are also key to the functional upgrading of furniture firms in the PRD. For example, the 4-billion policy initiated by Chinese central government which aimed to boost domestic consumption has encouraged the market reorientation of previously export-oriented suppliers. Local institutions such as furniture exhibitions also made furniture firms easily establish connections with the domestic retailers which concentrated at the exhibitions and then achieved functional upgrading. However, it is worth noting that furniture firms rarely gained direct financial support or help from local institutions, such as local governments or industrial associations. From this point of view, institutions tended to provide indirect political supports to supplier firms for their functional upgrading.
4.4.2
Adaptive Product Downgrading Through Domestic Online Sales
Some furniture firms experienced downgrading in their market reorientation as they engaged in online sales. In our study, 26% (9/36) of the interviewed firms built online shops for domestic sales (Table 4.2). Similarly as those engaged in direct-sales in physical stores, these furniture firms also established intra-firm coordination governance relations in the domestic market. Among them, a small number of branded furniture firms deliberately produced low-quality products and sold them in the online platforms (e.g. Tmall, Taobao and JD.com). The study thus argues that the integration into the domestic market might lead to industrial downgrading of furniture firms. The product downgrading of these branded firms was conducted as an adaptive strategy to meet the demand of online consumers and increase market shares. The online sales of furniture firms were encouraged by the rise of Chinese online consumers that might have huge potential in the domestic markets. According to HKTDC (2012), the Internet population in China was more than 510 million while online shoppers accounted for 38% by 2011. Moreover, in recent years, Chinese central government has initiated policies (e.g. ‘E-commerce’ Thirteen-five Development Plan in 2016) to encourage industrial firms to engage in online sales and achieve industrial transformation. Therefore, although the interviewed companies explained that they did not gain direct financial support or help from local intuitions (e.g. local government) for their online sales, to some extent the political support from central government facilitated the process of the industrial downgrading of furniture firms. However, it is observed that integration into domestic market-oriented production networks through intra-firm coordination relations and product downgrading for online sales is not a relatively successful strategy. A senior manager of branded furniture firm in Dongguan remarked: The price of products on line is usually cheaper than that in physical stores because of the lower online operation cost. If furniture firms sell branded products online with a lower price, no consumer is willing to buy the same branded products from physical stores with a
4 Industrial Upgrading and Downgrading of Export-Oriented …
65
higher price. So, to some extent, online sales will destroy our partnerships with chain retailers who operate the chain stores. (Interview in Dongguan, September 2014)
Therefore, branded furniture firms experience essential difficulties in selling high-quality branded products with high prices to online consumers who care most about the price of products but pay little attention to the brand name and quality. Secretary-General of Furniture Industrial Association in Dongguan claimed: Our member firms with their own brands always complained that they could not get profits from online sales due to low-price competition. They are suggested to use low-quality timber to produce some low-end products with lower price for online sales. (Interview in Dongguan, July 2015)
Hence, furniture firms adjusted their products to meet the needs of low-end online consumers. This scenario indicates that product downgrading is an alternative strategy in response to the demand of domestic online consumers (Plank and Staritz 2015; Ponte and Ewert 2009; Smith et al. 2014). Take the Firm F in Dongguan as a case again. This firm has started online sales on an online platform called Meilele (in Chinese) in 2014. To protect the fame of their brands, the firm sells furniture products without their brand names online. However, online sales provided limited contribution to their total domestic sales, which only accounted for around 5% (interview in Dongguan, September 2015). The evidence indicates that product downgrading has not yet facilitated the market reorientation of furniture firms.
4.4.3
Passive Attempts at Industrial Upgrading
Unlike the branded furniture firms which have achieved functional upgrading or experienced product downgrading, some furniture firms show less interest in industrial upgrading in their market reorientation. Results show that 26% (9/36) of the interviewed furniture firms tended to conduct domestic sales by selling their previous export products without redesigning to domestic individual retailers (Table 4.2). These furniture firms are small-scale firms which previously targeted the low-end markets in the Middle East and Southeast Asian countries, in which consumers pursue low price. These firms usually have fewer than 50 workers in their factories and are concerned less about design, quality, and brands. With the lack of brand names which are the preconditions of the high-end segment, furniture firms tended to market transactions with individual retailers which target the low-end domestic consumers (e.g. online consumers). The interviewed firms explained that due to their weak capabilities in branding and innovation, they find it difficult to cooperate with domestic chain retailers which aim to target the high-end markets in China (interviews in Shenzhen, March 2015). Moreover, unbranded furniture firms also blamed weak financial capabilities for hindering their attempts to establish brands and improve innovation. Meanwhile, they gained insufficient financial and political support from local governments, such as tax reduction and
66
T. Fu and C. Yang
branding subsidiaries. One domestic private supplier in Houjie Town, Dongguan complained: We did not get any support from Houjie town government or Dongguan municipal government. It is very difficult for our small firms to get loans from any bank in recent years especially after the 2008 global financial crisis. We do not have capabilities to improve and redesign the previous export products. Registering a brand from governmental institute is easy for us, but making it well-known is very difficult. Without a popular brand name, we cannot get succeed in domestic sales. (Interview in Dongguan, September 2015)
As a result, 25% (9/36) interviewed furniture firms have not yet established brands after turning into domestic sales, reflecting the passive attempts at functional, product and process upgrading. From their point of views, market reorientation was a temporary strategy to mitigate the decline of export orders. Many declared that they are still waiting for the recovery of export markets. An interviewed firm explained: For us, export is easier to get profits because of the large volume of exports. We hope the export markets to recover in the coming years. (Interview in Dongguan, September 2015)
The study indicates that not all of the furniture firms have adopted industrial upgrading or downgrading strategies in response to the consumer demands in their market reorientation to China’s market. However, compared with the firms which have adopted industrial upgrading or downgrading strategies, some firms are still struggling to maintain their businesses in the uncertain global market dynamics. Nevertheless, more empirical studies are needed to examine the effects of market reorientation on export-oriented supplier firms and understand why some firms rise in domestic production networks, while some firms fall.
4.5
Conclusions
The industrial upgrading paths have been widely discussed in the literature on GPNs and GVCs. While numerous studies have been conducted on the industrial upgrading prospects of developing country suppliers in their integration into Western market-oriented production networks, relatively few have been conducted to examine the industrial upgrading or downgrading trajectories of export-oriented producers in the market reorientation towards domestic or host markets in developing countries. The final market shifts from the Global North to the Global South prompted the revisiting of the upgrading trajectories of export-oriented supplier firms in developing countries in the contemporary global economy. The evidence of the export-oriented furniture industry in the PRD proved the insufficient explanation power of industrial upgrading in the previous literature. Indeed, the PRD has witnessed the mix value capture trajectories regarding a combination of downgrading and upgrading driven by domestic market-oriented production networks in the market reorientation of previously export-oriented supplier firms (Smith et al. 2014; Plank and Staritz 2015).
4 Industrial Upgrading and Downgrading of Export-Oriented …
67
Through the perspective of GPNs on industrial upgrading, this chapter sheds light on upgrading or downgrading trajectories of furniture firms and the subsequent effects on their market reorientation towards the domestic market of China. In recent years, the PRD, such as Dongguan, has made great efforts to change its role from low-end exporters to the high-end manufacturers by developing indigenous innovation. The evidence of the PRD elucidates that market reorientation has stimulated the industrial upgrading as well as downgrading trajectories of local suppliers. For example, some competitive suppliers strategically have established brand names and inter-firm cooperation with domestic retail partners, achieving functional upgrading (Horner 2014). The study highlights the importance of domestic chain retailers in helping furniture firms to connect with domestic final consumers and facilitate their process of functional upgrading (Kadarusman and Nadvi 2013). The study further demonstrates that in addition to functional upgrading, some furniture firms deliberately conducted product downgrading to expand online sales, indicating the variegated trajectories of supplier firms in the process of market reorientation (Blažek 2015). Nevertheless, some furniture firms have not yet engaged in product, process, and functional upgrading to target the higher end segment of China’s market, due to their weakness in branding and financing and the lack of direct financial support from local institutions such as local governments. Instead, these firms still showed huge reliance on exports (Tokatli 2013; Selwyn 2012). The evidence of the PRD also demonstrates the importance of emerging markets in reshaping the industrial upgrading and downgrading trajectories of previously export-oriented suppliers. Through integration into domestic markets of China, supplier firms gained more functional upgrading prospects which were dominantly controlled by global lead firms in the global production networks (Horner 2014; Yang 2014). From this point of view, the implication of emerging markets such as China for the transformation of industries in developing countries deserves more attention in future research. The chapter shows that there is still a long way to go for the PRD, the export-processing ‘world factory’, to upgrade and enhance their positions in the GVCs/GPNs. This research has fundamental policy implications for China, which is on a quest for transforming economic development from the export-oriented to the domestic market-oriented model. This chapter provides a vivid evidence of industrial upgrading and downgrading trajectories of export-oriented supplier firms in their market reorientation. More empirical studies are in need to explore the upgrading or downgrading paths in the export-oriented industry in developing countries in the rise of domestic or host markets. Acknowledgements Financial supports from the National Natural Science Foundation of China (41571119), Hong Kong Baptist University Faculty Research Grant (FRG1/15-16/053, FRG1/ 16-17/035, FRG 1/17-18/013 and FRG2/14-15/055) are gratefully acknowledged.
68
T. Fu and C. Yang
References Bair, J., & Gereffi, G. (2003). Upgrading, uneven development, and jobs in the North American apparel industry. Global Networks, 3(2), 143–169. Barrientos, S., Knorringa, P., Evers, B., et al. (2016). Shifting regional dynamics of global value chains: Implications for economic and social upgrading in African horticulture. Environment and Planning A, 48(7), 1266–1283. Blažek, J. (2015). Towards a typology of repositioning strategies of GVC/GPN suppliers: The case of functional upgrading and downgrading. Journal of Economic Geography, 16 (December 2015), 849–869. Castaño, J. (2002). The booming furniture industry in China. International Tropical Timber Organization, 277(3), 2003–2005. Cattaneo, O., Gereffi, G., Staritz, C., et al. (2010). Global value chains in a postcrisis world: A development perspective. Washington, DC: The World Bank. China National Furniture Association. (2014). China furniture yearbook, 2014. Beijing: China Forestry Press. China Statistical Bureau. (2009). China statistical yearbook, 2008. Beijing: China Statistical Press. Coe, N. M., Hess, M., Yeung, H. W-C., Dicken, P., & Henderson, J. (2004). “Globalizing” regional development: A global production networks perspective. Transactions of the Institute of British Geographers, 29(4), 468–484. Coe, N. M., Dicken, P., & Hess, M. (2008). Global production networks: Realizing the potential. Journal of Economic Geography, 8(3), 271–295. Coe, N. M., & Yeung, H. W.-C. (2015). Global production networks theorizing economic development in an interconnected world. Oxford University Press. Dongguan Statistical Bureau. (2010). Dongguan statistical yearbook, 2009. Beijing: China Statistical Press. Famous Furniture Fair. (2015). Famous furniture fair report 2015. Accessed June 13, 2017. http:// www.gde3f.com/wp-content/uploads/2015/08/review.pdf. Frederick, S., & Gereffi, G. (2011). Upgrading and restructuring in the global apparel value chain: Why China and Asia are outperforming Mexico and Central America. International Journal of Technological Learning, Innovation and Development, 4(1/2/3), 67. Gereffi, G. (2009). Development models and industrial upgrading in China and Mexico. European Sociological Review, 25(1), 37–51. Gereffi, G. (2014). Global value chains in a post-Washington Consensus world. Review of International Political Economy, 21(1), 9–37. Gereffi, G., Humphrey, J., Sturgeon, T., et al. (2005). The governance of global value chains. Review of International Political Economy, 12(1), 78–104. Guangdong Statistical Bureau. (2009, 2010, 2011, 2012, 2013, 2014, 2015). Guangdong statistical yearbook, 2008, 2009, 2010, 2011, 2012, 2013, 2014. Beijing: China Statistical Press. Hong Kong Trade Development Council (HKTDC). (2012). Online shopping trends. HKTDC Research. Hong Kong Trade Development Council (HKTDC). (2014). Furniture market in China. HKTDC Research. Hong Kong Trade Development Council (HKTDC). (2015). Furniture market in China. HKTDC Research. Horner, R. (2014). Strategic decoupling, recoupling and global production networks: India’s pharmaceutical industry. Journal of Economic Geography, 14(6), 1117–1140. Humphrey, J., & Schmitz, H. (2000). Governance and upgrading: Linking industrial clusters and GVC research. IDS Working Paper, 120, 1–37. Humphrey, J., & Schmitz, H. (2002). How does insertion in global value chains affect upgrading in industrial clusters? Regional Studies, 36(9), 1–16.
4 Industrial Upgrading and Downgrading of Export-Oriented …
69
Kadarusman, Y., & Nadvi, K. (2013). Competitiveness and technological upgrading in global value chains: Evidence from the Indonesian electronics and garment sectors. European Planning Studies, 21(7), 1007–1028. Kaplinsky, R., & Farooki, M. (2011). What are the implications for global value chains when the market shifts from the north to the south? International Journal of Technological Learning, Innovation and Development, 4(1–3), 13–38. Kaplinsky, R., & Readman, J. (2005). Globalisation and upgrading: What can (and cannot) be learnt from international trade statistics in the wood furniture sector? Industrial and Corporate Change, 14(4), 679–703. Liu, Y. (2016). The dynamics of local upgrading in globalizing latecomer regions: A geographical analysis. Regional Studies, 3404(March), 1–15. MacKinnon, D. (2012). Beyond strategic coupling: Reassessing the firm-region nexus in global production networks. Journal of Economic Geography, 12(1), 227–245. Pavlinek, P., & Zenka, J. (2010). Upgrading in the automotive industry: Firm-level evidence from Central Europe. Journal of Economic Geography, 11(3), 559–586. Plank, L., & Staritz, C. (2015). Global competition, institutional context and regional production networks: Up- and downgrading experiences in Romania’s apparel industry. Cambridge Journal of Regions, Economy and Society, 8(3), 421–438. Ponte, S., & Ewert, J. (2009). Which way is ‘up’ in upgrading? Trajectories of change in the value chain for South African wine. World Development, 37(10), 1637–1650. Selwyn, B. (2012). Beyond firm-centrism: Re-integrating labour and capitalism into global commodity chain analysis. Journal of Economic Geography, 12(1), 205–226. Smith, A., Pickles, J., Buček, M., et al. (2014). The political economy of global production networks: regional industrial change and differential upgrading in the East European clothing industry. Journal of Economic Geography, 14(6), 1023–1051. Tokatli, N. (2013). Toward a better understanding of the apparel industry: A critique of the upgrading literature. Journal of Economic Geography, 13(6), 993–1011. Yang, C. (2014). Market rebalancing of global production networks in the Post-Washington consensus globalizing era: Transformation of export-oriented development in China. Review of International Political Economy, 21(1), 130–156. Yeung, H. W., & Coe, N. M. (2015). Towards a dynamic theory of global production networks. Economic Geography, 91(1), 29–58.
Tianlan Fu is a Postdoctoral Researcher in the School of Geography and Planning at the Sun Yat-sen University (China). She obtained her doctoral degree from Hong Kong Baptist University. Her research interests focus on global production networks, industrial clusters, industrial upgrading, and regional development in China. She has published papers in Growth and Change, and Geografiska Annaler: Series B, Human Geography. Chun Yang is an Associate Professor in the Department of Geography at Hong Kong Baptist University (Hong Kong). Her research areas focus on urban and regional development, global production networks, transnational corporations and foreign direct investment, industrial clusters, regional innovation systems and cross-border regions in China. She is currently an Editorial Board Member of ‘Journal of Economic Geography’ (Oxford University Press), Associate Editor of ‘Regional Studies, Regional Science’ (Taylor & Francis) and Co-Editor of ‘Bandung: Journal of the Global South’ (Springer). She has published extensively in the leading peer-reviewed international journals, such as Economic Geography, Regional Studies, Urban Studies, Geoforum, Environment and Planning A, Environment and Planning C, Cities, Habitat International, Political Geography, Eurasian Geography, and Economics, etc.
Chapter 5
The Changing Firm Landscape and Firm Location Behaviour Sierdjan Koster and Piet Pellenbarg
5.1
Introduction
The Dutch landscape of firms has changed drastically in the last decades. One of the most prolific chances is a still ongoing surge in entrepreneurship since the 1980s. At that time, the country occupied low positions in international rankings of start-up rates. The tables have turned and now start-up rates in the Netherlands are among the highest in Europe. Hartog and Wennekers (2009) summarise this transformation as the Dutch ‘entrepreneurship revolution’. The surge of new and small firms reflects the still ongoing tertiarisation of the economy, as well as an increased cultural acceptance of entrepreneurship as a career choice. Also, the regulatory framework has become more accommodating to entrepreneurship. Finally, globalisation forces and technological development, aided by the liberal labour market regulations, have pushed for an increasingly flexible labour market. Temporary jobs and flexible deployment of personnel are quickly becoming the norm. Partially, the demand for flexibility is fulfilled by thriving temporary employment agencies. Quite specific to the Netherlands, however, is the surge of solo firms—or solo self-employment—in response to the more flexible labour market (OECD 2015a, b). People increasingly offer their skills and experiences on their own account. Solo firms are then hired by other firms on a project basis. Figure 5.1 shows the scale of this shift in the firm landscape. It shows the number of establishments with only one person involved (solo firms) and all other establishments (employers) in the Netherlands over time. The growth of solo firms is significant, and in just under 20 years, the number of solo firms has more S. Koster (&) P. Pellenbarg Economic Geography, Faculty of Spatial Sciences, University of Groningen, Groningen, The Netherlands e-mail:
[email protected] P. Pellenbarg e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_5
71
72
S. Koster and P. Pellenbarg
than tripled from just over 200,000 to almost 800,000.1 In the same period, the number of employing firms has increased by about 10%, while stabilising and slightly declining in the aftermath of the economic crisis of 2004–2007. With the changing landscape of firms, the landscape of firm decisions has also changed. Consequently, the location patterns of firms, as well as their location characteristics, can be affected. Understanding these changes is important for policies aimed at attracting and retaining local firms. At the same time, the changing landscape of firms suggests a reappraisal of the existing conceptual framework in which firm location and relocation is seen. This chapter contributes to the literature by sketching the changes in the Dutch firm landscape and exploring empirically and conceptually the effects on the location decision processes and locational patterns of firms. As a vantage point in this discussion, we take the idea that the declining firm size increases the role of the owner in the location decision process. As such, non-firm considerations regarding the location of the firm may play an increasingly important role in the decision process. This is clearest in solo firms. In fact, the entrepreneur is the main decision-maker in the firm: The firm is the entrepreneur, and the entrepreneur is the firm (Stam 2003; Lazear 2003). In spite of a sole owner of the firm, location decisions are not necessarily solely made by the entrepreneur. Particularly in the case of home-based businesses, competing interests of a partner and possibly children may importantly influence the decision to move shop (Koster and Venhorst 2014). This means that understanding the social context of the firm becomes increasingly important in understanding its location dynamics. This chapter aims to illustrate the role of the social context in the location choice behaviour in the rapidly growing small- and micro-firm segment of the economy. Firm location literature has generally shied away from small businesses exactly because of the idea that firm location decisions may actually not be related to the interests of the firm only. Given the sharp increase in solo firms, it is hard to keep disregarding the group of very small firms. The direct economic impact in employment terms remains relatively small: 90% of all jobs are still in employing firms. It is on the increase, however, and the employment share in solo firms has increased from 3% in 1996 to 10% in 2014, making it a significant group. Regardless of the employment effect, the increase of solo firms has implications for physical work arrangements, making it an increasingly relevant group from the spatial planning perspective. Also, they may form an important flexible layer of suppliers for existing larger firms. Finally, the networks of solo firms may have positive economic effects. Since the economic impact of these micro-businesses is increasing, it is appropriate to integrate their locational behaviour in the analytical framework of firm location and relocation.
1
The definition of solo self-employed differs across databases and other data provide different group numbers as a result. However, the general trend is consistent regardless of the definition and data source used.
5 The Changing Firm Landscape …
73
Fig. 5.1 Firm establishments in the Netherlands by type, 1996–2014. Source LISA (2015)
The remainder of the chapter explores how the increasing importance of small and micro-firms can impact on firm location, the location decisions, the consequent spatial patterns, and the characteristics of the locations. For this, we first review the development of the conceptual arguments how firm location is established from the perspective of the people involved. How are firm location decisions formed and by whom? In this section, we focus on relocation decisions. Relocation choices are made more consciously and tend to be better remembered than most initial location choices. Secondly, we discuss two empirical illustrations which fit in the perspective that acknowledges how the owner and non-firm considerations are increasingly part of firm location decisions. In the first illustration, we focus on firm location decisions of small firms and how they are influenced by the family context and competing locations of economic interest. It turns out that the firm location is the last piece of the locational puzzle. The family does not follow the firm, the firm follows the family. The second illustration explores empirically how the changing firm landscape impacts on the types of premises that firms occupy. This serves as an example of how the changing patterns in the firm population influence the physical infrastructure that they need and the locations that they choose. Finally, tentative implications are formulated which emerge from a perspective in which smaller firms take a more prominent position.
5.2
Firm Location Decisions in the Social Context
Overseeing the historical development of firm location and relocation research, it can be concluded that non-firm considerations and the role of the owner and his or her relevant social context have not played a key role in the explanation of firm locations. Even though the decision-makers or the decision-making team were
74
S. Koster and P. Pellenbarg
considered in the analyses, the motives and driving forces of firm location and relocation were found in the structural, regional level of socio-economic factors. This is a reasonable focus because the decision-makers’ location motives can be related to local, regional or even national location conditions. Knowledge thereof can be used in efforts to stimulate, counter or redirect regional economic trends (see Rossi in this volume). And this was the prime objective of firm migration research in the 1960s and 1970s. Prompted by a small number of trendsetting US studies by McLaughlin and Robock (1949) and Garwood (1953), relocation studies in the 1960s and 1970s were dominated by European policy-driven studies, all inspired by the desire to identify the crucial firm relocation factors that should be managed in policies to reduce regional inequality or to foster urban growth. Extensive overviews of the firm relocation literature of the post-World War II period are given by Pellenbarg et al. (2002) and Pellenbarg and Knoben (2012). In the 1960s and 1970s, most studies had a regional focus, corresponding with the then great interest in regional policies. In the 1980s, attention shifted to the urban level, in line with the then growing interest of policy makers in urban renewal, environmental protection and local economic development. However, the focus was still very much on the regional or local socio-economic characteristics in explaining firm location, reflecting the viewpoint that firm relocation is important mainly as a tool for regional economic development. In the 1990s, the orientation towards firm relocation processes changed. Then, rather than relying on regional, local or urban characteristics, attention shifted to firm internal motives for relocation. The level of analysis shifted from the region to the firm itself and micro-level models was introduced to empirically address location decisions (Van Dijk and Pellenbarg 2000; Brouwer et al. 2004; Knoben and Oerlemans 2008). With the advent of the analyses at the micro-level, it was now possible to zoom in further on the spatial scale and address within-city differences (further explored by Dej et al. in this volume). At the neighbourhood level, in- and out-migration rates of firms tend to show large differences, which suggest that spatial characteristics of neighbourhoods may be important. Sleutjes and Völker (2012), however, find that neighbourhood characteristics just play a modest role. Firm internal factors, such as growth ambitions and the size and quality of the business premises, are more important. Greenhalgh (2008) reported similar results. This indicates that the (local) real estate market may be an important factor for firm location. Risselada et al. (2013) focus on the role of real estate and conclude that real estate characteristics are indeed important determinants for the likelihood of firm relocation. Similar to Van Dijk and Pellenbarg (2000), they find that property ownership has a negative effect on relocation. Residing in old buildings with plenty of floor space also limits firm relocation (Van Dijk and Pellenbarg 2017). With such an overview of firm location literature as a backdrop,2 we can see two consistent elements in the different approaches to understanding firm relocation.
2
We do not strive for a comprehensive overview of the literature. For that, we refer to Pellenbarg et al. (2002) and Pellenbarg and Knoben (2012).
5 The Changing Firm Landscape …
75
First, location and relocation processes of firms tend to be framed in a ‘location factor’ approach. Whether on the regional, urban or neighbourhood level, in a macro- or micro-approach, in a firm internal or a firm external perspective, the idea is that locational conditions relevant to the firm guide relocation decisions (see also Arauzo‐Carod et al. 2010). This line of reasoning echoes the classic theory of the spatial margins of profitability: firms will survive if they are within the borders in which revenue (access to relevant markets) outweighs the costs (access to resources) incurred (Smith 1966). Second, implicitly or explicitly, most studies adopt a ‘decision-makers’ approach in which it is assumed that the firm location is the outcome of a managerial decision process that tries to optimise the outcome for the firm. The decision-makers, first and foremost, have the best interest of the firm in mind. It has been long recognized that such decision processes are far from rational (Simon 1957; Pred 1967). The business owners do not have all information. Even if they had, personal preferences and perception may be more important in the decision than the expected benefits for the firm. As a case in point, many business owners keep on running a business, even if the returns are minimal (Hamilton 2000). Outside factors are generally overlooked in the literature, but the social context of the firm—firms’ employees, business relations, personal friends and acquaintances, a spouse, children and other family members—may importantly influence the decision-making process, the considerations which go into the location decision and the eventual outcome. One important reason why such a ‘relational view’ which includes the indirect decision-makers stayed out of the picture is the fact that classical firm relocation studies concentrated on middle-sized and large manufacturing firms—firms which promised a new future to lagging (agricultural) regions thanks to the process of industrialisation, provided that mobile industrial firms could be persuaded to choose such regions as a relocation target. This was the firm relocation inheritance from the 1950s, 1960s and 1970s. Often, firms with fewer than 10 employees were not even included in research populations. Larger firms were the most interesting ones from a regional policy perspective, and researchers’ interviews with the decision-maker took place in the firm boardroom, taking the view of the firm as it was perceived from there. This approach persisted when firm relocation research shifted attention to the economic sectors of wholesale, transportation and business services which gradually showed more mobility than the manufacturing industry, during the 1980s and 90s. In line with the focus on large firms, there has been a scarce interest in the location dynamics of newly started and young firms. It was readily acknowledged that smaller firms tend to be more mobile than big ones, that small mobile firms are often young firms, and that relocation is to be understood as a typical event in the early stages of a firm’s existence (Stam 2003). Yet, firms in the start-up phase have usually not been a purposefully chosen research category, nor were they identified as a category in broader firm relocation studies. In this tradition, there was little room for location considerations beyond the scope of the firm. The previous pages have highlighted the predisposition of the existing firm location and relocation literature for explaining firm locational dynamics from considerations which relate to the firm only. There has been a strong focus on a
76
S. Koster and P. Pellenbarg
‘location-based’ approach from a ‘decision-makers’ perspective. The implicit assumption in the ‘location-based’ approach to firm location is that the evaluation of various locations is done from the viewpoint of the firm. That is, a location is chosen to optimise the performance of the firm. Now, this is not a ‘historical’ viewpoint. Still, the underlying assumption in much of the location and relocation literature is that the location is chosen with the best interest of the firm in mind. However, with the increasing number of the self-employed and the consequent decreasing average size of firms in the Netherlands, this assumption may become less appropriate. Instead, it is likely that a ‘relational’ approach is also appropriate. In such an approach, the location decision and its outcome is placed in the social context of the entrepreneur. The remainder of the chapter presents two empirical illustrations that are indicative of a ‘relational’ approach to firm location dynamics. The first illustration zooms in on the location and relocation behaviour in relation to other meaningful economic activities within the household. It explores how non-firm-related considerations may be important, particularly for smaller firms. The second illustration explores how new business arrangements are reflected in the type of premises used for the business. Both illustrations have implications for location policies which aim at facilitating local firm development, one of the original motives for studying firm location and relocation processes.
5.3
Empirical Illustration 1: Firm Location Decisions in the Context of the Household3
The organisational structure of a larger firm ensures that decisions are made in the interest of the organisation at large. Subjective as it may be, the decision is made as part of a bargaining process with the best interest of the collective organisation in mind (the decision-maker approach). In smaller firms, the owner has much more leeway to decide on the firm location and it is principally a personal decision (Stam 2003). With this, the location decision is steered away from the sole interest of the firm and other considerations may become more important. Koster and Venhorst (2014), inspired by the household approach in migration research (Cooke 2008), suggest that alternative income sources in the household may be important determinants of the firm location. If there are other important sources of income in the household, accommodating these may trump locating at the most lucrative firm location. Moreover, the location of the firm is relatively flexible as it is determined by the entrepreneur, while the location of a job in wage employment can be assumed to be fixed in the short run. Changing the location of the firm can then be used to solve the ‘locational puzzle’ to optimise access to all relevant sources of household income. This would be an indication that the firm location is a part of a 3
This Section reflects the argument as presented in Koster and Venhorst (2014).
5 The Changing Firm Landscape …
77
social bargaining process that goes beyond optimising the match between firm and location characteristics. Even though, particularly in small firms, the owner has considerable leeway in relocating the firm, there are costs involved. Apart from the costs involved in the actual relocation of the firm, losing access to markets and suppliers may be the most prominent deterrents from relocating a firm. Given the importance of local access to markets and inputs for firm performance (Dahl and Sorenson 2009), self-employed have been assumed to be relatively rooted in space. The changing landscape of firms, with an increasing share of small firms in business services, is likely to reduce the mean costs of firm relocation. With the increasingly footloose nature of self-employed firms, changing the location of the firm may have a limited impact on the market it is active in. Also, small firms in business services have incurred relatively few sunk costs and require few capital investments, making relocation a less costly affair. Additionally, if the firm is small enough, it is likely that it only provides part of the household’s income and it may not be the deciding factor in the locational puzzle. The costs associated with relocating the firm may have become smaller over time with an increasing number of relatively footloose firms. Also, the costs may be relatively small in relation to other meaningful sources of income at the household level. To assess the role of firm (re)location in solving the residential puzzle to optimise all locations of economic interest, Koster and Venhorst (2014) follow the residential and firm moves of Dutch highly educated self-employed over time. For this they use, the Dutch register data at the individual level for the period 2001– 2007, provided by Statistics Netherlands. Using these data, they compare the residential location and the firm location at the municipal level. That is, they assess whether the firm is in the same municipality as where the entrepreneur lives and how these arrangements change over time. Importantly, they are also able to see whether firm relocation or relocation of the residence is used to arrive at the most suitable spatial arrangements of competing locations of economic interest. Two results stand out. Firstly, most entrepreneurs (up to 85%) match the firm location to their location of residence. What is more, the locational match of firm location and residential location increases with the time entrepreneurs spent on the labour market (after graduation). This result resonates with the idea that market information and input of resources are mostly local affairs (Dahl and Sorenson 2009). Running a firm at a distance is costly and entrepreneurs generally avoid this situation. Secondly, Koster and Venhorst (2014) find evidence that the firm location is used as the final piece of the location puzzle that a household faces. It is the firm that is relocated relatively often to obtain a locational match between the firm location and the place of residence; the firm thus trails a residential move. In contrast, households tend to move house even if that means that the firm location is no longer matched to the residential location. Both mechanisms are strongest if other relevant economic locations—jobs in wage employment—are present in the household. This further indicates how firm location is used to accommodate access to other economic meaningful locations.
78
S. Koster and P. Pellenbarg
Clearly, the size of the firm acts as an important mediating factor and it acts as a strong locational anchor: larger firms are more rooted in place. Increased relocation costs are likely to act as a deterrent of relocation. At the same time, larger firms are likely to be more important for the household income, making relocation of a relatively big firm a liability at the household level as well. With the average size of firms declining, it is likely that other criteria than solely the welfare of the firm gain in importance. The location of small firms is also determined by the residential preferences of the household, which include the location of other meaningful economic activities, i.e. other jobs. This comes to show that attracting entrepreneurial small firms is not just a matter of providing the best environment for entrepreneurship. Also, residential quality and the provision of wage jobs may be important.
5.4
Empirical Illustration 2: Firm Premises
The second illustration of the changing firm location arrangement resulting from the changing characteristics of the firm population is the dynamics in the workplaces and the premises of the firms. With a changing landscape of firms, the requirements of the work locations may also change, particularly if the location of the firm serves multiple goals, as suggested above. Understanding the dynamics in the physical locations of firms offers useful policy information as the changing physical requirements impact on the requirements to the built environment. As such, it is an input for local spatial policies, which have a strong tradition in the Dutch context. In the context of surging solo self-employment and mini-businesses (see Fig. 5.1), it can be expected that home-based businesses are increasingly common. This may induce specific policy questions in terms of allowing the mixed use of premises, but also accommodating work-living arrangements in the layout of the building itself. At the same time, the market for ‘traditional’ types of building for firms may change. The link between real estate development and self-employment or entrepreneurship remains somewhat of a blind spot in research into firm dynamics. Some studies have looked at the consequences of working from home, in the light of the advent of teleworking (Baines 2002). However, analyses of the dynamics in the market for firm premises remain scarce, perhaps because entrepreneurs do not see finding suitable premises as a top priority (Smallbone and Welter 2001). At the same time, as explained earlier, real estate characteristics are important driving factors in the relocation decision of firms (Risselada et al. 2013; Sleutjes and Völker 2012). In the remainder, we present new explorative information regarding the dynamics in the premises used by firms, both employers and solo self-employed firms. The data exploration is based on the LISA dataset which is a register of all establishments in the Netherlands. The data are available for the period 1996–2014. The data include information on the number of employees working in each establishment, its precise location and information about the building the
5 The Changing Firm Landscape …
79
establishment is located in. Specifically, we use the information on the function of the building as detailed in the spatial regulations that each municipality has. These regulations specify which function a building may have, ranging from industrial to office space and from a shopping function to residences. Using these data, we can describe the dynamics in the use of premises over time, connected to different types of establishment and to different locations. The data run from 2008 to 2014 and the agriculture sector has been excluded, as it is only reliably documented in the dataset after 2011. Figure 5.2 provides the first hint of the dynamics in the buildings used by firms. The left graph shows the number of establishments which employ more than one person and the right panel focuses on solo firms, firms with only one person involved. Both graphs show the breakdown of the groups into the five largest groups of the premises type (Residential, Industrial, Office, Shops and Multiple Functions) and the ‘Other’ category which includes all other uses including health institutions, educational structures, as well as hotels and restaurants. As the graphs show the count of employing firms as well as solo firms, they in fact reproduce the rightmost part of Fig. 5.1, demonstrating that in the post-crisis period the growth in establishments is fully accounted for by solo firms. Also, the picture for the types of premises is clear, solo firms predominantly operate from home. Surprisingly perhaps, also for employing firms—still mostly small firms— residential buildings are the largest single type of building in which they locate. Just under 30% of all employing firms operate from a residential location. The dynamics in the building types at the level of establishments addresses the spatial and physical challenges the changes imply in terms of work-living arrangements, communication and transportation. The number of employees in the different building types proxies the economic relevance of the dynamics. Measured in employment, the role of residences as places of work is less impressive—the largest firms operate from designated buildings—but still the share of total workplaces located in residences is 16% in 2014. Apart from the big increase in employment in residential locations, employment in all other building types remains very stable, indeed. This suggests, at least for the Netherlands, that there is a pressure to accommodate places of work in residential-type settings rather than accommodating firms in business parks or industry estates, though these fields are generally higher on the policy agenda. Figure 5.3 reveals that there is an important spatial dimension in how firms are distributed across the different building types. Figure 5.3 shows, for five levels of urbanity (see Statistics Netherlands 2017), the distribution of firms across the building types. The graph only includes employing firms. For solo self-employed, the residential location is by far the largest group in each level in the urban hierarchy. For employing firms, there are clear differences, although a residential location is the single largest group at each urban level. Yet, residential locations are far more prevalent in rural areas than in cities. Also, locations that have a mixed use are more common in rural areas. Recall that the agricultural sector was excluded from the analysis, so that the high rate of home-based businesses does not reflect the prominence of the agricultural sector in rural areas. The other categories combined
80
S. Koster and P. Pellenbarg
Fig. 5.2 Types of premise of employing firms and solo firms. Source LISA (2015)
Fig. 5.3 Employing establishments by type of premise by urbanity. Source LISA (2015)
—particularly shops and offices—are more common in urban areas, which reflect their role as functional centres. Contrary perhaps to common belief and to the focus of prior studies and policy debates about accommodating new forms of entrepreneurship in residential areas, it is actually the rural areas in the Netherlands that are spearheading this development, both in terms of establishments and jobs.
5.5
Conclusion
The characteristics of Dutch firms are changing. On average, following a surge in self-employment, firms have become smaller than they used to be, which increases the leverage of the owner on the firm location. This chapter explores how the changing landscape of firms is reflected in firm location dynamics, both in terms of decision-making and in the outcomes of the process.
5 The Changing Firm Landscape …
81
The prime insight is that the borders of the firm become permeable. That is, particularly for smaller firms, aspects outside of the direct considerations of the firm become important in the location decision of firms. Entrepreneurs consider family arrangements and the location of the firm is just one of the factors in determining its location; access to a job for the partner, for instance, is equally important, if not leading. The dynamics in the premises of firms also reflect the permeability of the border between the firm and other domains. Solo firms operate primarily from home, but also (small) employing firms are primarily found in dwellings. In terms of employment, 16% of all jobs in employing firms are in residences. These examples suggest a reappraisal of the current mainstream in firm location research which still relies on its inception period in which large (multinational) firms were the motors of the economy. Understanding their relocation behaviour and the decision processes involved was thus crucial. The fundamental assumption was that the relocation process, messy and fuzzy as it is, was meant to find the optimal location for the firm based on considerations of the firm only. Now, for many firms the location process, if still an optimisation exercise, involves factors in more dimensions than just the firm. This includes preferences and needs of the household. The empirical illustrations also make a case for current policies related to entrepreneurship and location. The increase in entrepreneurship in residential areas can put pressure on the physical infrastructure in terms of work-living arrangements, including access to fast Internet and the accessibility of residential areas. With decreasing firm size and an increasing number of jobs and firms in home-based businesses, residential quality and suitability for business can be important triggers for local entrepreneurship. Interestingly, efforts to accommodate entrepreneurship in the neighbourhood generally focus on cities to battle local socio-economic problems (Sleutjes and Völker 2012). Businesses in rural areas, however, tend to be located at home or in mixed-use premises more often. Over half of all businesses in rural areas are organised as such. In the light of current debates on population decline, the pressure on rural labour markets and sustaining services in low-density areas, a policy shift to rural areas may be called for.
References Arauzo-Carod, J. M., Liviano-Solis, D., & Manjón-Antolín, M. (2010). Empirical studies in industrial location: An assessment of their methods and results. Journal of Regional Science, 50(3), 685–711. Baines, S. (2002). New technologies and old ways of working in the home of the self-employed teleworker. New Technology, Work and Employment, 17(2), 89–101. Brouwer, A. E., Mariotti, I., & Van Ommeren, J. N. (2004). The firm relocation decision: An empirical investigation. Annals of Regional Science, 38(2), 335–347. Cooke, T. J. (2008). Migration in a family way. Population, Space and Place, 14(4), 255–265.
82
S. Koster and P. Pellenbarg
Dahl, M. S., & Sorenson, O. (2009). The embedded entrepreneur. European Management Review, 6(3), 172–181. Garwood, J. D. (1953). An analysis of postwar industrial relocation to Utah and Colorado. Economic Geography, 29(1), 79–88. Greenhalgh, P. (2008). An examination of business occupier relocation decision making: Distinguishing small and large firm behaviour. Journal of Property Research, 25(2), 107–126. Hamilton, B. H. (2000). Does entrepreneurship pay? An empirical analysis of the returns to self-employment. Journal of Political Economy, 108(3), 604–631. Hartog, C., & Wennekers, S. (2009). De twee gezichten van de ondernemerschapsrevolutie en hun betekenis voor Nederland. Zoetermeer: Panteia. Knoben, J., & Oerlemans, L. A. G. (2008). Ties that spatially bind? A relational account of the causes of spatial firm mobility. Regional Studies, 42(3), 385–400. Koster, S., & Venhorst, V. A. (2014). Moving shop: Residential and business relocation by the highly educated self-employed. Spatial Economic Analysis, 9(4), 436–464. Lazear, E. P. (2003). Entrepreneurship. Bonn: IZA. LISA. (2015). Landelijk InformatieSysteem Arbeidsplaatsen (National Information System on Employment). McLaughlin, G. E., & Robock, S. (1949). Why industry moves South: A study of factors influencing the recent location of manufacturing plants in the South. Kingsport, Tennessee: Kingsport Press, National Planning Association. OECD. (2015a). Entrepreneurship at a glance 2015. Paris: OECD Publishing. OECD. (2015b). OECD employment outlook 2015. Paris: OECD Publishing. Pellenbarg, P. H., & Knoben, J. (2012). Spatial mobility of firms. In A. Frenkel, P. Nijkamp, & P. McCann (Eds.), Societies in motion: Innovation, migration and regional transformation (pp. 71–113). Cheltenham: Edward Elgar. Pellenbarg, P. H., Van Wissen, L. J. G., & Van Dijk, J. (2002). Firm relocation. In P. McCann (Ed.), Industrial location economics (p. 148). Cheltenham: Edward Elgar. Pred, A. R. (1967). Behaviour and location. Foundations for a geographic and dynamic location theory, Part 1. Lund Studies in Geography B, 27. Risselada, A. H., Schutjens, V., & Van Oort, F. (2013). Real estate determinants of firm relocation in urban residential neighbourhoods. Journal of Economic and Social Geography, 104(2), 136–158. Simon, H. A. (1957). Models of man, social and rational. New York: Wiley. Sleutjes, B., & Völker, B. (2012). The role of the neighbourhood for firm relocation. Journal of Economic and Social Geography, 103(2), 240–249. Smallbone, D., & Welter, F. (2001). The role of government in SME development in transition economies. International Small Business Journal, 19(4), 63–77. Smith, D. M. (1966). A theoretical framework for geographical studies of industrial location. Economic Geography, 42(2), 95–113. Stam, E. (2003). Why butterflies don’t leave. Location evolution of evolving enterprises. Utrecht: University of Utrecht. Statistics Netherlands. (2017). Regionale Kerncijfers (Primary regional data). http://statline.cbs.nl. Van Dijk, J., & Pellenbarg, P. H. (2000). Firm relocation decisions in the Netherlands: An ordered logit approach. Papers in Regional Science, 79(2), 191–219. Van Dijk, J., & Pellenbarg, P. H. (2017). Firm migration. In: D. Richardson (Ed.), The international encyclopedia of geography: People, the earth, environment, and technology. Section industrial geography. AAAG/Wiley.
5 The Changing Firm Landscape …
83
Sierdjan Koster is an Associate Professor in Economic Geography at the Faculty of Spatial Sciences, University of Groningen in The Netherlands. His research revolves around the theme of regional economic impact of entrepreneurship. Within this wider topic, he aims at understanding dynamics in entrepreneurship at the micro-level (firm or entrepreneur), while connecting this to regional-level outcomes. In line with this approach, his Ph.D. thesis deals with start-ups that spin-off from existing larger firms. Currently, he is also an editor for REGION, the online and open access journal of the European Regional Science Association. Piet Pellenbarg is emeritus professor of Economic Geography and past dean of the Faculty of Spatial Sciences of the University of Groningen, The Netherlands. The main topics of his research and publications are firm location and migration, local and regional economic development, spatial images, and place marketing. Together with his faculty colleague Paul van Steen, he published, since 1987, the continuous series of maps and articles ‘The Netherlands in Maps’ in the International Journal of Economic and Social Geography. In 2008, he published, with Egbert Wever, the book International Business Geography (Abingdon/New York, Routledge). In 2012, he was awarded a honoris causa doctorate at the University of Łódź in Poland.
Part II
Firm Relocation Factors and Consequences
Chapter 6
Relocation of Economic Activity Within Major Metropolitan Areas in Poland—Scale and Key Characteristics Magdalena Dej, Wojciech Jarczewski and Michał Chlebicki
6.1
Introduction
An inherent factor in the operation of a business entity is the need to deal with constant change. The local circumstances relevant from the perspective of running a business in a specific location may undergo change, as may the legal regulations at the national level, as well as consumer preferences concerning needs for particular products. In addition, technological progress ensures that the manufacturing process is also variable over time. Companies therefore face a constant need to adapt in running their business to deal with these dynamic conditions, in order for it to continue to fulfil the criterion of profitability. This process often has a spatial dimension (Pellenbarg et al. 2002), frequently manifested in relocations of business entities. Particularly high intensity of these processes has been observed in metropolitan areas. Owing to the concentration of intellectual potential, capital, the proximity of goods and services important for running the business and a host of other benefits of agglomerations, investors display a high level of interest in locating their companies in such areas. They often focus on areas with a special status as investment zones, formed with the participation and involvement of local authorities with the goal of strengthening the local economy. Metropolitan areas are also the site of incubation M. Dej (&) W. Jarczewski National Institute for Spatial Policy and Housing, Kraków, Poland e-mail:
[email protected] W. Jarczewski e-mail:
[email protected] M. Chlebicki Faculty of Philosophy and Sociology, University of Warsaw, Warsaw, Poland e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_6
87
88
M. Dej et al.
of many companies, particularly in certain sectors, a phenomenon that was identified as early as the 1960s and 1970s in the “incubator hypothesis” (Leone and Struyk 1976). While some firms base themselves in these areas, others leave them, moving within a given metropolitan area (e.g. leaving the centre of a city and relocating to its outskirts) or outside of it. These processes have also been observed in Poland in recent decades, yet no attempt has been made to offer a meaningful analysis taking into consideration their scale and profile. This applies not only to relocation of business entities in metropolitan areas, but also to relocation of companies in Poland in general. Research conducted in Western European as well as North American countries over the last few decades has yielded similar conclusions on companies’ migration directions, although these behaviours vary according to the type of business. In fact, we can distinguish two separate stories of company migrations. In the case of large firms, often with foreign direct investment (FDI), long-distance, frequently international migrations take place, which should be considered on a global scale, or at least a continental one. Migrations of smaller companies, meanwhile, occur over shorter distances, and it is to these that the incubator hypothesis can be applied (Pellenbarg 2010). Many studies on relocation have concentrated on the scale of urban regions [e.g. in Italy (Ortona and Santagata 1983), Great Britain (Elias and Keogh 1982), China (Xiaoxia et al. 2010) and the USA (Iseki and Jones 2014)]. This is because the greatest intensiveness of the relocation process is witnessed in metropolitan areas, which therefore also comprise the area of analysis of this study, in relation to Poland. Research on company relocation has been undertaken in a number of countries, including the USA and the Netherlands, as early as in the 1960s and 1970s, regarded as “the golden era of firm relocation studies” (Pellenbarg et al. 2002). Such studies as a particular variant of the issue of business entity location have commonly been divided, following Hayter (1997) and Machlup (1967), on the basis of the factors influencing the process, into three approaches: neoclassical, behavioural and institutional (Pellenbarg et al. 2002). The neoclassical approach derives from classical economic theory, in which two variables are the most significant for relocation processes: minimisation of costs and maximisation of profits. This approach, whose main premises come from Isard (1956), assumes that companies are generally able to function anywhere where the total income from the business exceeds the total costs. It is possible to determine the spatial limits of profitability that indicate viable and non-viable locations (cf. Smith 1966). In a situation of equilibrium, when a company is located in a specific place, relocation is unnecessary. In a situation of changing conditions, however—regarding both the firm itself and its environment—the spatial limits of profitability might shift, and if a company finds itself outside of the zone of profitability, this might result in relocation. The neoclassical approach therefore assumes that the decision-makers opting for relocation are fully rational and knowledgeable on the subject of the costs and profits resulting from location at a specific site, and that the prime objective of the firm’s operations is to maximise profits.
6 Relocation of Economic Activity Within Major …
89
The behavioural approach has a different emphasis, in terms of both the attitude to the business location itself and relocation. This has four main premises: (1) the role of limited information; (2) the ability to use information; (3) perception and mental maps; and (4) uncertainty (Pellenbarg et al. 2002, in: Pred 1967, 1969). In the behavioural approach, therefore, it is the perception of reality that is crucial, rather than the actual situation. This approach complements the neoclassical perspective by focusing on the motives of the decision-makers themselves which are taken into account in the process of deciding upon relocation. In comparison with the neoclassical approach, it ignores or marginalises economic factors, in particular psychological and sociological issues. The common feature of both approaches, however, is that they treat a company as an active entity making decisions in the conditions of a static environment. In the institutional approach, this perspective was questioned, and attention was switched from the company to the environment. Yet, the most importance is attributed to the sociocultural context in which a business is run. Particular emphasis is placed on the role of the authorities (local, regional, national) and the legal circumstances that determine the rules for running a business in a given place, as well as the real estate market. Previous studies have taken a variety of perspectives on the subject of company relocation. These have included the factors deciding upon the change of location— push and pull factors—the characteristics of the firms undertaking the change, the links between these processes with the features of the local economies, etc. To reiterate, there is very little knowledge on relocation in Poland, and this chapter therefore concentrates on presenting the main characteristics of this process. Consequently, its objective is to track the course of the relocation of businesses located in five Polish metropolitan areas in the context of the companies’ characteristics, including age, size and the type of business operations. The question of the distance of the relocation will also be considered. The results will then be compared with those from previous research conducted on this subject in other countries.
6.2
Theoretical Context
No distinct set of theories or paradigms has been developed to explain relocations of business entities, specifying the spatial rules governing them. Among the reasons for this is the fact that they are usually treated as a specific form of choice of a business location. The only real comprehensive theoretical proposition that refers solely to the processes of relocation is the incubator hypothesis, in its static (Hoover and Vernon 1962) as well as dynamic (Leone and Struyk 1976) form. In the most general terms, this assumes that highly centralised locations constitute a site of incubation for new companies, which is manifested in a disproportionately high number—in comparison with other locations—of new entities and/or employment in such entities (static version). Furthermore, the dynamic
90
M. Dej et al.
variant describes the phenomenon whereby new companies founded in the central areas of cities leave these locations in the first years of their existence, relocating externally in search of better conditions for expanding their productive activity. Alongside pull factors, most studies on relocation processes also identify factors pushing a company out of a particular location, considering them from a neoclassical, behavioural or institutional approach (Pellenbarg et al. 2002). None of these angles, however, appears to be dominant in research on relocation (Scott 2000). The golden age of relocation research came in the 1960s and 1970s, when the field began to develop in Western Europe and the USA. It was on the basis of studies in the USA that the incubator hypothesis was formed and developed, referring to the incubation of companies during their foundation, followed by the processes of relocation in its dynamic variant. As noted by Townroe (1979, p. 13) and Klaassen and Molle (1983), among others, a huge obstacle was posed (which continues to exist in many countries) by lack of access to data at the level of public statistics allowing research to be carried out nationally. As a result, studies concentrated on selected metropolitan regions in states of Ohio and Wisconsin (Pellenbarg et al. 2002, p. 14). Subsequent decades saw decreased interest in relocation studies in the classical form, as a complete change in a company’s business location from point A to point B. The profile of the research also changed. There was less interest in studies on short-distance relocations or relocations within one country, replaced by research in an international perspective, as a result of the considerable intensity of such phenomena. Questions of outsourcing and offshoring became more significant (e.g. Contractor et al. 2010; Marin 2005), as well as the related processes such as global value chains (GVCs) and global production networks (GPNs) (Gereffi et al. 2005; Humphrey and Schmitz 2002). The literature on this subject is vast, and this chapter lacks the space to cite it more extensively. For insightful overviews, see Wόjtowicz and Fu and Yang in this volume. In Poland, and in Central and Eastern Europe more broadly, we can observe major delays in research on relocation processes in the classical form on a domestic scale. The first reason for this is the distinct economic situation in this part of Europe until the 1990s, and a completely different profile of economic processes. The second is the barrier presented by the difficulty with accessing data over the past quarter of a century. This problem was hinted at when discussing the methodology of data acquisition within this study. As described above, the scant resources of public statistics regarding business entities and lack of any generally accessible records on changes in business addresses make it necessary to carry out extremely laborious research, which needs to be done in two stages. First, it is necessary to identify all business entities situated in a given area, before, separately for each company, checking its history for any changes of business address. Together with a huge amount of analytical material, this makes the research process very complicated. In Poland to date, the subject of relocation has not been tackled on a broader scale than within individual municipalities. This study, and the wider project within which it was conducted, is the first of its type concerning business relocations in Poland.
6 Relocation of Economic Activity Within Major …
91
Considering the objective of this chapter, i.e. to determine the main features of the relocation of business entities in Polish metropolitan areas, including firms’ characteristics and the distance of the relocations, the next section provides the discussion of empirical results. Inevitably, the discussion does not refer to one specific theoretical approach, because, as noted above, no coherent set of research approaches has been developed for relocation processes. However, we attempted to present the analysis within the wider context of research carried out within various theoretical traditions and geographies, which proved relevant for interpretation of the results.
6.2.1
Size and Age of Company and Distance of Relocation
The neoclassical approach to relocation emphasises the spatial margins of profitability (Rawstron 1958; Taylor 1970; Smith 1966, 1971; McDermott 1973), which determine the distance a firm is likely to move. On the assumption that the company management wishes to minimise costs and maximise profits and has a suitably wide knowledge of the conditions of its operation, businesses move within these limits to a distance within which they remain economically profitable. Research carried out in this paradigm has demonstrated, for instance, that larger companies are more inclined to relocate than smaller ones, as a result of the costs of floor space. As Pellenbarg, van Wissen and van Dijk note (2002, p. 5): “Within the neo-classical framework of the spatial margins to profitability this means that the level of the cost surface is higher for small firms than for large firms in all locations, and these margins span a larger area for larger firms. Thus, the level of the cost surface for small firms will be higher than the revenue curve everywhere. Therefore, small firms cannot escape failure by relocating to another location, but must grow in order to remain profitable”. However, the aforementioned research by Leone and Struyk (1976) shows that the period of company “incubation”, which lasts about three years, is followed by relocation outside of the central business district in the case of companies whose growth has accelerated the most and who need to hire a larger number of people. It was also observed that in general the distance within which companies relocate is correlated with the growth in employment numbers, whereby businesses whose development in this respect is more dynamic tend to move longer distances. Meanwhile, research conducted on the basis of the behavioural approach, which assumes that companies function in the conditions of uncertainty and limited economic rationality, refers to the frequency of relocation over various distances. Such studies indicated that companies are less likely to choose more distant locations, which are more uncertain and less known, than nearer sites. These conclusions result from research carried out on the basis of mental maps and business representatives’ ideas of the economic space in which they operate (e.g. Pellenbarg 1985). The behavioural approach assigns significance not so much to reality as to the perception of it. The related doubts mean that “[…] Relocation to another geographical market may even be comparable to the inherent uncertainty of a startup” (Pellenbarg et al. 2002, p. 8).
92
M. Dej et al.
Interesting conclusions come from studies on relocations and their distance in relation to company growth, as well as the general economic situation, along with the redistribution policy pursued in Western Europe, which is now widely applied in EU countries. These show that a company’s mobility decreases amid conditions of greater economic uncertainty. This concerns both the situation and economic state of the firm itself and the general condition of the economy in which it operates. Based on his research on the situation in the Netherlands from the 1970s and 1980s, Pellenbarg argues that the redistribution policy places significant restrictions on the potential of peripheral areas to attract companies. This is connected with increasing education levels among residents, resulting in a greater difficulty in finding less qualified staff with lower wage expectations compared to central areas (Pellenbarg 1985). In general, Pellenbarg’s research from later years points to a growth in the number of short-distance relocations compared to long-distance ones, associated with the phenomenon of the “suburbanisation of firms” as a result of the insufficient space within the central city, parking problems and increased congestion. Pellenbarg, van Wissen and van Dijk’s research published in 2002, investigating the connection between company size and inclination to relocate, shows that there are no significant correlations between the two. The empirical studies simply indicate that only companies with fewer than 10 staff members exhibit a greater tendency to change their business location; no such links were identified in the category of larger firms. Among other subjects covered by research in Western Europe were the links between relocation and a company’s lifetime, as well as the length of the actual decision process concerning the change in location. This topic was explored by Townroe (1973), Louw (1906) and Pen (1999, 2000). The last of these studies demonstrated that the average length of the decision process was two years. Partly as a result of this fact, the number of relocations among new companies in the first years of their existence is generally very low. Leone and Struyk’s (1976) analyses seeking to test the incubator hypothesis in the New York metropolitan area led them to the following conclusions: “We did observe both higher growth rates and greater probabilities of establishment relocation for younger plants, but there was little evidence that these activities were contributing disproportionately to decentralization. To the contrary, we found that the younger plants were more likely to relocate within the zone of origin” (Leone and Struyk 1976, p. 13).
6.2.2
Business Types
Whereas initially researchers focused in particular on industrial firms, the changes taking place in the national economies in the 1980s and 1990s led them to increasingly take the relocations of service companies into account. A phenomenon
6 Relocation of Economic Activity Within Major …
93
Table 6.1 Exemplary types of company migrations Case
Migration type
1 X moves totally from A to B (integral move) 2 X moves partially from A to B (partial move) 3 X from A contracts out to Y in B 4 X and Y from A and B start a new joint venture in C 5 X and Y from A and B merge, and locate in A or B 6 X and Y from A and B merge, and locate in C 7 X changes to Y and moves from A to B NB X and Y indicate firms, and A and B their locations
Notes Relocation Spatial expansion
that was observed was the departure of firms providing services to business from the central zone of metropolitan areas to external zones. Research conducted on the Netherlands confirmed that the 1990s were characterised by a very intensive phenomenon of relocation of service companies, especially those providing services to business, from urban cores to business parks in the outskirts of cities, in the suburban zone, thus creating office corridors. Studies analysing firms’ mobility by sector (Kemper and Pellenbarg 1997) showed that those in the category of commercial (business) services are the most mobile group. Construction and manufacturing companies, meanwhile, are less mobile, while the stickiest sector according to the cited research is retail.
6.3 6.3.1
Methodology Defining Company Relocation
First of all, it was necessary to adopt a definition of relocation. The dynamics of economic processes and the number of potential “behaviours” of companies are very high. Investors take a range of decisions concerning their business, some of which also have a spatial dimension connected with migrations of firms. Table 6.1 shows some examples of possible migrations that take place spatially. This study focused on the first case from Table 6.1, i.e. migrations in which a company moves wholesale from location A to location B. It is worth noting that the type identified in point 1 in the table, classical relocation, corresponds to a much greater degree with the behaviours of small and medium-sized single-workplace firms, and much less with the reality of the operation of multi-workplace entities and large enterprises (see Van Dijk and Pellenbarg 2000, pp. 98–99). In this chapter, we use the terms “migration” and “relocation” interchangeably, while bearing in mind that relocation is a semantically narrower category than, and a specific type of, migration.
94
6.3.2
M. Dej et al.
Research Area and Period of Analysis
The study concentrates on analysing relocation in Polish metropolitan areas in which companies’ relocation processes take place with a particularly high intensity. This phenomenon was provisionally identified during previous research in one of the communes (municipalities) of the Kraków Metropolitan Area (Jarczewski and Huculak 2011). Poland does not have a statutory list of metropolitan areas, and so in order to define them, it is necessary to refer to studies which have attempted to delimit such areas. Among the numerous papers on this subject (Dej 2011), we decided to make use of the delimitation by Smętkowski et al. (2008), which is the most complex in terms of the criteria it considers when designating a metropolitan area. This incorporates the following indicators: municipality’s income, migration balance, change in the number of companies with foreign capital and private enterprises listed in the REGON database. The following were also taken into account: distance of the municipality from the metropolitan core, its location in relation to the main transport corridors, the number of inhabitants of the municipality, the existence of branches of companies registered in the metropolitan centre, the level of communication with this centre, elements of communal infrastructure and the area’s compactness criterion (three quarters of the administrative boundaries of the municipality are with municipalities forming part of the metropolitan area). On this basis, the following metropolitan areas were identified: Tri-City, Poznań, Warszawa, Łódź, Wrocław, Silesia and Kraków. Their spatial range is illustrated in Fig. 6.1. From this list, we decided to analyse company relocation in five metropolitan areas, leaving out the Tri-City and Silesian ones. The reason for this omission was that these are polycentric areas in which the processes of relocation, especially concerning companies moving on the city centre (core)–external metropolitan area (ring) line, are characterised by different features from monocentric areas. The study therefore examined the following metropolitan areas: Poznań, Warszawa, Łódź, Wrocław and Kraków. The analysis was conducted for business entities which relocated between 2001 and 2013. These criteria were determined by availability of data. Identification of the companies relocating was based on data from the National Court Register, which has only been registering businesses since 2001. The finishing date was dictated by the fact that the data were collected in 2014. It is also worth highlighting the fact that the study was conducted with reference to Polish metropolitan areas. Previous analyses of the processes of relocation have taken place in Western Europe and the USA, where the circumstances of these processes may differ. For example, the conditions for running a business are different (much longer traditions of enterprise, varying levels of attractiveness for foreign investors, less dynamic economic processes, which in Poland and other post-communist countries constitute something of a reaction to years of restrictions in the economy as a result of the lack of a free-market economy). These differences are also related to the approach to the processes of spatial planning, especially in metropolitan areas, and management of development in these areas. Therefore,
6 Relocation of Economic Activity Within Major …
95
Fig. 6.1 Metropolitan areas of Poland’s largest urban centres. Source Smętkowski et al. (2008)
when comparing the results of this research with previous studies, it is important to bear in mind the specific context of Poland.
6.3.3
Data Sources, Identification Procedure for Relocating Companies and Period of Analysis
In Polish public statistics, there is no register which can be used to identify the business entities that have changed their place of operation. Such companies were therefore identified using a procedure comprising several stages and utilising several data sources. Stage 1: Identification of all companies operating in the five metropolitan areas The first step was to take the municipalities of the external zones of the five Polish metropolitan areas—Kraków, Łódź, Poznań, Warszawa and Wrocław—and
96
M. Dej et al.
identify all of the companies operating there in 2014, in specified size ranges.1 It is important to stress that the analysis did not take into account the category of micro-enterprises (i.e. entities employing fewer than 10 people) as well as sole proprietors. The category of micro-enterprises comprised some 95.6% of all business entities registered in Poland in 2013. The number of individuals acting as sole proprietors in the same year was about 2.96 million. To identify the companies included in the study, we used the Hoppenstedt Bonnier Information (HBI—now Bisnode Polska) commercial company database. This database provided contact details that made it possible to place companies in specific locations. In total, 6240 companies were identified in the area of analysis.2 Data were collected for all business entities, regardless of sector—both industrial and service companies. Stage 2: Determining the scale of company relocation The next step was to obtain reports from the National Court Register for all of the identified companies. These contain information about all the addresses at which each business has been registered. Two relocation categories were used— internal, or intra-municipal, and external, or inter-municipal. Internal relocation applies to changes in location within a municipality, and external relocation to a move to conduct business in a different municipality. We thus differentiated between intra-municipal and inter-municipal relocations. The analysis of relocations encompasses the years 2001–2013, with the starting data being the launch of the National Court Register. For technical reasons, it was not possible to use other sources to determine changes in address in the whole population of companies analysed. As a result, we used this database of companies for information on relocations taking place since 2001. Altogether, we collected information on 5442 business entities. In the case of companies formed in 2001 and later, we also took into account any changes of address, while with those predating the register, we only considered changes after 2001, i.e. not always all the relocations undertaken by a given company. The aim of eliminating this problem led us to stage 3.
1
In the three metropolitan areas—Kraków, Wrocław and Poznań—data were gathered from both the core area (city centre) and the external zone (ring). In the latter case, this concerned companies with more than nine employees, and for the core, more than 49 employees. For two metropolitan areas—Warszawa and Łódź—no data were collected for urban cores, but only for companies in the external zone with more than nine employees. This was dictated by limitations on time and organisation in collecting and analysing such a large amount of data. The authors are aware, however, that this approach entailed a considerable simplification, since according to the incubator hypothesis, the centres of metropolitan areas are home to many companies which then move to suburban districts. Adopting a method of data collection that ignores the centres of these two metropolitan areas can have an impact on the interpretation of results. We therefore endeavoured to keep this limitation in mind whenever drawing conclusions from the results. 2 In Poland, there is currently no extensive source of company contact information that could be used for a larger number of administrative units. Tax offices have access to such data, but do not make it available for research purposes. Furthermore, each tax office operates on an individual basis, and it would therefore be necessary to obtain permission for access to data from the relevant office for each location.
6 Relocation of Economic Activity Within Major …
97 intra-municipal 20.5
intra-municipal 15.2 no relocaƟon 64.3
inter-municipal
intra-municipal
no relocaƟon
Fig. 6.2 Relocations of companies in the study in 2001–2013. Source Own research
Stage 3: Restriction of the set of companies to those set up not before 2001 In this way, we restricted the research material to those entities for which analysis of the relocation was possible throughout their existence.
6.3.4
Sample Size and Main Characteristics
The final sample comprised 2600 companies. Of this group, the vast majority (64.3%), 1672 companies, did not relocate during the period in question (Fig. 6.2). Only slightly over one-third of the businesses relocated—928 companies. A total of 534 business entities relocated at least once between municipalities, while 394 did so with the municipalities in which they were formed. As shown in Fig. 6.3, the majority of companies changing address did so only once (70.2% of businesses). One in five firms relocated twice, and those moving on more than two occasions constituted just 2.3% of relocating companies, or 6.7% of all entities in the study. It is also worth comparing the structure of the analysed sample with that of all the companies in the area of the study from the point of view of the fundamental characteristic of their type of business (Table 6.2). This comparison indicates in particular a clear over-representation of companies from category C (27.2% in the analysed sample, and just 8.9% in the whole population), and under-representation in categories M (6.5%, as compared to 12.0%), as well as S and T (5.71% as compared to 0.41%). In the remaining sectors, the comparisons are similar. No credible analogous comparisons are possible for other characteristics, such as the time of foundation or size of companies, owing to the availability of data in the
98
M. Dej et al.
Fig. 6.3 Number of relocations of companies in the study. Source Own research
0.1%
1.5%
0.1%
3.3%
4.7%
20.0%
70.2%
1
2
3
4
5
6
no data
Central Statistical Office of Poland Local Data Bank and the method of data collection within the study. The analyses conducted in the next part of the article used contingency tables and chi-squared statistical tests. In the cases where the premises did not permit asymptotic methods, exact tests were used—a sign test and Wilcoxon’s signed-rank test.
6.4 6.4.1
Results Company Lifetime and Relocations
Our research demonstrated a statistically significant dependence (chi-squared (v2) = 0.00) showing that in the entire analysed period, i.e. 2001–2013, older companies relocated more frequently than younger ones. External relocations took place considerably more often among companies founded in 2004 and 2005 (Table 6.3). Meanwhile, between 16 and 22% of firms founded between 2001 and 2007 moved within the municipality. Among the companies established in 2008– 2009, we can observe a distinct drop in relocation. None of the businesses set up in 2010–2013 changed their location. This situation is in accordance with the results of earlier research, which showed that there are few relocations among new companies, and the decision process associated with the relocation itself takes around two to three years (e.g. Pen 1999, 2000). At the same time, it is worth considering to what extent the lack of relocations in the last four years of the period of analysis (2010–2013) and the relatively lower number in 2008–2009 were influenced by the general economic slowdown in Poland, an outcome of the global economic crisis itself. After all, such a phenomenon of slowed processes of company relocations as a result of economic recession was observed previously, for instance, in the 1970s.
6 Relocation of Economic Activity Within Major … Table 6.2 Structure of the analysed sample compared to structure of companies as a whole according to type of business, 2013
PCA sector
Type of business Total proportion
99
Analysed sample proportion
A 0.90 0.41 B 0.09 0.16 C 8.90 27.20 D 0.17 0.45 E 0.32 1.32 F 10.71 11.19 G 26.03 28.85 H 6.08 4.73 I 2.70 1.56 J 4.17 2.92 K 3.42 1.93 L 5.07 3.79 M 12.00 6.54 N 3.30 4.53 O 0.23 0.04 P 3.33 0.16 Q 5.33 3.13 R 1.57 0.66 S and T 5.71 0.41 Total 100.00 100.00 NB Entities with fewer than 10 employees were omitted Source Own research and Central Statistical Office of Poland Local Data Bank
As Pellenbarg et al. explain (2002, p. 16): “Firm mobility is related to firm growth, and tends to diminish in an economic recession. Firms are less likely to take far going investment decisions like re-location when the future economic prospects are rather uncertain”. Nevertheless, this research does not provide clear-cut evidence to support this thesis. Relocation is a very important step in the development of a company, bringing enormous costs with it (direct cost of moving, search and information costs of finding new markets, labour, suppliers, etc.). Furthermore, it is necessary to take into account sunk costs in the previous location. These mean that “a move to another geographical market is to a certain extent similar to a start-up, with large investments, and uncertain revenues” (Pellenbarg et al. 2002, p. 4). As a result, companies that decide to relocate do not usually do so more than once. Our research involved analysing the number of relocations of individual companies, and then comparing this information with the lifetimes of the analysed firms. This provided information on companies’ most frequent relocation “behaviours”.
100
M. Dej et al.
Table 6.3 Percentage of relocations among all analysed companies based on year of foundation Company year of foundation 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Total Source Own research
Inter-municipal relocation (%)
Intra-municipal relocation (%)
No relocation (%)
Total (%)
14.6 26.5 25.7 31.1 31.0 24.4 22.4 16.0 12.5 0 0 0 0 20.5
16.5 14.6 21.6 16.0 17.9 18.7 14.3 12.2 11.5 0 0 0 0 15.2
64.0 53.8 52.7 52.9 51.2 56.9 58.3 71.8 76.0 100.0 100.0 100.0 100.0 64.3
100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Companies most commonly decide upon their first relocation between the second and the fourth years of their lifetimes. This was the case of 54.8% of firms relocating for the first time. Some 73.4% of first relocations took place during the first five years of companies’ operations (Table 6.4). Only 246 of the 2600 businesses opted for more than one relocation. Companies changing address for a second time tended to do so between their second and seventh years of activity (74.8% of second relocations), most often between the third and fifth years. Only 60 companies in the sample changed address for a third time. This occurred most often between their fourth and eighth years of their operation (75% of third relocations). Very few firms—only 16 firms—changed address a fourth time. Such relocations took place between the fifth and the 11th years of operation. The low number of cases observed makes it difficult to identify the most frequent period. A fifth change of address was observed in only two cases (in the sixth and eighth years of operation), and this was therefore ignored in the further analyses. As noted, some 73.4% of first relocations took place within the first five years of companies’ operation. This percentage rises to 96.5% of first relocations in the case of the first ten years. The probability that a firm relocates therefore decreases markedly if it has not changed address after the first five years of operation. Companies which have not relocated within their first 10 years can be said with high probability to be unlikely to relocate. Some 87.4% of second relocations occur within five years of the first relocation. Similarly, as many as 93.3% of third relocations happen within five years of the second relocation, and 93.8% of fourth relocations within five years of a business’s third change of address (Fig. 6.4). We
6 Relocation of Economic Activity Within Major …
101
Table 6.4 Relocation and age of company Relocation number
Number of firms
First 897 relocation Second 246 relocation Third 60 relocation Fourth 16 relocation Fifth 2 relocation N = 2600 Source Own research
Mean age of firm at relocation
Median age at relocation
Minimum age at relocation
Maximum age at relocation
3.16
2.00
0
12
4.79
4.00
1
12
5.70
5.50
1
12
6.81
6.50
4
10
6.00
6.00
5
7
30.0
percent
25.0 20.0 15.0 10.0 5.0 .0 1
2
3
4
5
6
7
8
9
10
11
12
13
year of operaon first relocaon
second relocaon
third relocaon
fourth relocaon
Fig. 6.4 Percentage of relocations and year of operation of companies. Source Own research
can therefore assume an approximate limit by which time companies relocate— around five to 10 years from the foundation until the first relocation, and up to approximately five years for further moves. In order to observe the phenomenon of shortening periods between relocations more precisely, we conducted an additional analysis. Since the frequency distribution did not satisfy the requirements of asymptotic tests (lack of normality of distribution—Fig. 6.5, low number of cases), the analyses were carried out on the basis of exact tests—Wilcoxon’s rank-sum test, as well as a sign test. Both tests showed that there are statistically significant differences (value of tests < 0.00)—more frequent was the extension of the periods between the first and the second relocation compared to the period from the foundation to the first relocation. Similarly, more frequent was the extension of the periods between the second and the third relocation than between the first and the second. With the next relocation, meanwhile—the third and fourth compared to the second and third—the differences were not statistically significant (Table 6.4). To sum up, the periods
102
M. Dej et al.
number of companies
250 200 150 100 50 0 1
2
3
4
5
6
7
8
9
10
11
12
13
year of operaon Fig. 6.5 Year of operation and first relocation. Source Own research
between individual relocations up to the third one were more often extended than shortened in the case of individual firms.
6.4.2
Company Size and Relocations
In general, larger companies (with 50 or more employees) relocated more often than smaller ones (10–49 employees). The figure in the former case was 41.7%, as opposed to just 24.8% in the latter (Table 6.5). This difference was statistically significant (chi-squared = 0.00). This dependence confirms the observations made by other researchers following the neoclassical approach. It is explained by spatial margins of profitability and costs of premises, which are not identical for companies of various sizes: “the level of the cost surface is higher for small firms than for large firms in all locations, and this margins span a larger area for larger firms. Then, the level of the cost surface for small firms will be higher than the revenue curve everywhere. Therefore, small firms cannot escape failure by relocating to another location, but must grow in order to remain profitable” (Pellenbarg et al. 2002, p. 5). Our study also suggests that, once smaller firms decide to move, these are more frequently external relocations than with larger companies. These differences were also statistically significant (chi-squared = 0.00). Both dependences were quite weak, however (Phi and V Kramer symmetric measurements at a level of 0.125– 0.228 (Table 6.6). We can venture the hypothesis that this is connected with the greater mobility of smaller entities, which find it easier to change location, e.g. as they have far smaller sunk costs. In the case of larger companies, whose relocation is more likely to be internal, except for the physical change of place of business, they remain in the same environment. Networks of subcontractors and cooperating entities, and especially staff, remain unchanged, as the relocations generally do not exceed a few kilometres. Such a relocation is easier for the company—a smaller “revolution” which does not entail the need to start again effectively.
6 Relocation of Economic Activity Within Major …
103
Table 6.5 Distribution of pairs of periods of individual companies Periods
Number of firms
Period between first and second relocation shorter than that between foundation and first relocation Period between first and second relocation longer than that from foundation to first relocation Period between first and second relations same as that from foundation to first relocation Period between second and third relocation shorter than that between first and second relocation Period between second and third relocation longer than that between first and second relocation Period between second and third relocation same as that between first and second relocation Period between third and fourth relocation shorter than that between second and third relocation Period between third and fourth relocation longer than that between second and third relocation Period between third and fourth relocation same as that between second and third relocation Source Own research
79
Total firms
Test results
246
Statistically significant difference
60
Statistically significant difference
16
Statistically insignificant difference
127 40 13
35 12 5 9 2
Table 6.6 Company size and relocations Company size (number of employees) 10–49 50+ Total N = 2600 Source Own research
Inter-municipal relocation
Intra-municipal relocation
No relocation
Total
21.1 20.0 20.5
8.7 21.7 15.2
70.2 58.3 64.3
100.0 100.0 100.0
More in-depth analyses of the distances of relocations indicate that the lower average distances are, the more relocations take place. More exact analyses, however (sign test and Wilcoxon’s test), did not reveal these differences to be statistically significant.
6.4.3
Type of Business and Relocations
The analyses also showed statistically significant differences (chi-squared = 0.00) between the Polish Classification of Activities (PCA) sector and a company having relocated or otherwise (Table 6.7).
104
M. Dej et al.
Table 6.7 Polish Classification of Activities (PCA) sector and relocations PCA sectors
Inter-municipal Intra-municipal relocation relocation Percentage of companies (%)
No relocation
Total
Number of companies
C 27.4 13.6 54.0 100.0 661 D, E 16.3 16.3 67.4 100.0 43 F 18.0 15.4 66.5 100.0 272 G 22.8 13.6 63.6 100.0 701 H 27.0 11.3 61.7 100.0 115 I, J, K, L, M, N, S 17.8 26.6 55.6 100.0 527 (market services) O, P, Q, R 7.2 7.2 85.6 100.0 97 (non-market services) Total 21.9 16.3 61.8 100.0 2416 Key: C manufacturing, D electricity, gas, steam and air conditioning supply, E water supply; sewerage, waste management and remediation activities, F construction, G wholesale and retail trade; repair of motor vehicles and motorcycles, H transportation and storage, I accommodation and food service activities, J information and communication, K financial and insurance activities, L real estate activities, M professional, scientific and technical activities, N administrative and support service activities, O public administration and defence; compulsory social security, P education, Q human health and social work activities, R arts, entertainment and recreation, S other service activities Source Own research
The most mobile firms are entities from the market services group (PCA sector: I, J, K, L, M, N, S). Companies from this group usually relocate within the municipality within which they are established—this was the case of a quarter of the businesses in the study. Companies from sectors C (manufacturing) and H (transportation and storage), and to a lesser extent sector G (wholesale and retail trade), engaged in inter-municipal relocation to the greatest extent. One should note, however, that this is a very large sector. Meanwhile, the least “mobile” firms were those in “non-market” service sectors O, P, Q, R—the overwhelming majority of which (85.6%) did not change address. Generally, our study confirms the dependence observed in the research carried out for the Netherlands by Kemper and Pellenbarg (1997), which showed that companies in commercial services are the most mobile category of business entities.
6.5
Conclusions
The study showed that relocations of companies in Poland take place with considerable intensiveness. Among the 2600 entities employing at least 10 people in the research areas between 2001 and 2013, one-third (928 firms) relocated in that period, much more often on an external, inter-municipal basis. One-fifth of the
6 Relocation of Economic Activity Within Major …
105
businesses operating in the areas in question undertook such a move. On each occasion, the relocation process represented an enormous challenge for the company, frequently comparable with starting from scratch. As a result, any decisions on changing business location were usually taken only after serious consideration. Therefore, companies which relocated generally did so only once (70% of all relocating firms). Nevertheless, there are also cases of entities which migrated more often—twice or even several times. In extreme cases, particular companies undertook five, or even six relocations. This scale of the phenomenon, where one in three businesses experienced a change of location, suggests that this is a mass phenomenon, not confined to just a small group of companies. It is thus all the more striking that this has previously not been the object of wider interest among researchers. Further in-depth analyses of the characteristics of companies which migrated confirm these dependences and regularities which were previously identified in other countries and described in the literature. It has been possible to prove that companies relocating were those that had to some extent “toughened” since their foundation. In practice, therefore, new firms do not relocate. The main reason for this is the time necessary for making the decision to change location. The research to date indicates that the decision process in this case takes around two to three years. The fact that no entity in the analysed sample relocated within the last four years of the analyses, in 2010–2013, might also be linked to the global economic crisis that had struck in these years. This also had a noticeable effect on the Polish economy, causing a slowdown, a consequence of which was a reduced inclination towards relocation. This is, however, just a hypothesis which was not tested in the statistical analyses. The studies also lead to the conclusion that if a company does not undertake its first relocation within the first five to ten years of its existence, and five years for further changes, in the next years the likelihood of a change of location drops markedly. Further confirmation of the observations from previous studies is the claim that larger firms more often relocate than smaller ones. Among those entities employing at least 50 people, almost 42 relocated, whereas the figure for the group with 10–49 employees was 12% points fewer. Meanwhile, relocations of smaller companies are more likely to take place over greater distances, as shown by the fact that they undertook more external relocations outside of the municipality where they had been formed. This is probably related to their greater mobility and lower sunk costs in the previous location. The previous studies are also confirmed by the statement that the most mobile group of companies comprises those from the market services category. Almost half of firms from this group relocated at least once, but considerably more frequent in their case were intra-municipal than inter-municipal relocations. The change in business location in their case seems considerably simpler than in the case of entities from other sectors, since the main requirement they generally have is a suitable office space, and the involvement of fixed assets is minimal in comparison with companies in other sectors. For obvious reasons, entities providing non-market services (e.g. administration, education, health care) are much less mobile, as their presence in a given place results not from the
106
M. Dej et al.
economic factors of the profitability of a given location, but from having to satisfy the needs of a specific group of inhabitants in a specific space. Interestingly, over a quarter of all industrial firms (section C) undertook an external relocation, making them the most mobile category in that period. The presented results cast significant light on the nature of the relocation processes in Polish metropolitan areas. They indicate the main regularities governing the economic processes in this field. At the same time, though, they leave significant gaps in research. An extremely important question is that of the factors of relocation—both push and pull factors—that lead firms to decide to move. Also significant is the question of the economic consequences of these processes, on a local but also broader, regional scale. It will be possible to answer these questions by conducting further in-depth analyses, which seem necessary from both the scientific and the practical point of view. Acknowledgements The chapter was written as part of a research project financed by Poland’s National Science Centre: The processes of relocation and spatial expansion of business in metropolitan areas, SONATA grant competition no. 192273, grant period 2012–2017.
References Contractor, F. J., Kumar, V., Kundu, S. K., & Pedersen, T. (2010). Reconceptualizing the firm in a world of outsourcing and offshoring: The organizational and geographical relocation of high-value company functions. Journal of Management Studies, 47(8), 1417–1433. Dej, M. (2011). Mechanizmy skutki oddziaływania dużych przedsiębiorstw w obszarach metropolitalnych (Ph.D. thesis, non-published). Jagiellonian University, Kraków. Elias, P., & Keogh, G. (1982). Industrial decline and unemployment in the inner city areas of Great Britain: A review of the evidence. Urban Studies, 1, 1–15. Gereffi, G., Humphrey, J., & Sturgeon, T. (2005). The governance of global value chains. Review of International Political Economy, 12(1), 78–104. Hayter, R. (1997). The dynamics of industrial location. The factory, the firm and the production system. New York: Wiley. Hoover, E. M., & Vernon, R. (1962). Anatomy of a metropolis. New York: Anchor Books. Humphrey, J., & Schmitz, H. (2002). How does insertion in global value chains affect upgrading in industrial clusters? Regional Studies, 36(9), 1–16. Isard, W. (1956). Location and space-economy. New York: Wiley. Iseki, H., & Jones, R. (2014). Analysis of firm location and relocation around Maryland and Washington, DC metro rail stations. Paper submitted for a presentation at a Bi-National Symposium Featuring Paris, France, and Washington, DC on Transit, Transit Oriented Development, and Urban Form. Jarczewski, W., & Huculak, M. (Eds.). (2011). Sukces polityki proinwestycyjnej. Niepołomice 1990–2010. Wnioski dla władz lokalnych, Institute of Urban Development, Kraków. Kemper, N. J., & Pellenbarg, P. H. (1997). De Randstad een hogedrukpan. Economisch Statistische Berichten, 82, 508–512. Klaassen, L. H., & Molle, W. T. M. (Eds.). (1983). Industrial mobility and migration in the European community. Aldershot: Gower. Leone, R. A., & Struyk, R. (1976). The incubation hypothesis: Evidence from five SMA’s. Urban Studies, 13(3), 325–331. Machlup, F. (1967). Theories of the firm: Marginalist, behavioural, managerial. American Economic Review, 57, 1–33.
6 Relocation of Economic Activity Within Major …
107
Marin, D. (2005). A new international division of labor in Europe: Outsourcing and offshoring to Eastern Europe. SSRN Electronic Journal. McDermott, P. (1973). Spatial margins and industrial location in New Zealand. New Zealand Geographer, 29, 64–74. Ortona, G., & Santagata, W. (1983). Industrial mobility in the Turin metropolitan area, 1961-77. Urban Studies, 20(1), 59–71. Pellenbarg, P. H. (1985). Bedrijfsrelokatieen ruimtelijke kognitie (Firm relocation and spatial cognition). Sociaal-geografische Reeks, 33, FRW/RUG, Groningen. Pellenbarg, P. H. (2010). Firm migration. Janglin, China: Northwest A&F University. Pellenbarg, P. H., van Wissen, L. J. G., & van Dijk, J. (2002). Firm relocation: state of the art and research prospects. Groningen: University of Groningen, SOM research school. Pen, C. J. (1999). Improving behavioural location theory; preliminary results of a written questionnaire about strategic decision making on firm relocations. Paper presented to the ERSA Congress, Dublin. Pen, C. J. (2000). Actors, causes and phases in the decision-making process of relocated firms in the Netherlands. Faculty of Spatial Sciences, University of Groningen. Pred, A.R. (1967) Behavior and location: foundations for a geographic and dynamic location theory: Part 1. University of Lund, Lund Studies in Geography B, no. 27 Pred, A.R. (1969) Behavior and location: foundations for a geographic and dynamic location theory: Part 2. University of Lund, Lund Studies in Geography B, no. 28 Rawstron, E. M. (1958). Three principles of industrial location. Transactions of the Institute of British Geographers, 25, 132–142. Scott, A. J. (2000). Economic geography: The great half-century. Cambridge Journal of Economics, 24(4), 483–504. Smętkowski, M., Jałowiecki, B., & Gorzelak, G. (2008). Diagnoza problemów rozwoju obszarów metropolitalnych i rekomendacja delimitacji obszarów metropolitalnych w Polsce. Centrum Europejskich Studiów Regionalnych i Lokalnych EUROREG, University of Warsaw, Warsaw. Smith, D. M. (1966). A theoretical framework for geographical studies of industrial location. Economic Geography, 42, 96–113. Smith, D. M. (1971). Industrial location: An economic geographical analysis. New York: Wiley. Taylor, M. J. (1970). Location decisions of small firms. Area, 2, 51–54. Townroe, P. M. (1973). The supply of mobile industry: A cross-sectional analysis. Regional and Urban Economics, 2(4), 371–386. Townroe, P. M. (1979). Industrial movement, experience in the US and the UK. Westmead: Saxon House. Van Dijk, J., & Pellenbarg, P. H. (Eds.). (2000). Demography of firms; spatial dynamics of firm behavior., Netherlands Geographical Studies, 262 Utrecht/Groningen: KNAG/FRW RUG. Xiaoxia, F., Houkai, W., & Lixue, W. (2010). Analysis of urban industrial relocation’s incentive, approaches and effects—Taking Beijing as an example. In W. Houkai, B. Mei, & W. Yeqiang (Eds.), Series on Chinese Economics Research, Vol. 3. The macro-analysis of regional economy in China. A perspective of firm relocation (pp. 327–344). Singapore: World Scientific Publishing Co.
Magdalena Dej is a geography researcher at the National Institute for Spatial Policy and Housing in Poland. She received her Ph.D. from Jagiellonian University and specialises in economic local and regional development, particularly the development of rural areas. She authored and co-authored several books, numerous reports and research papers on the impacts of large companies on local communities as well as on rural areas and their relationship with major metropolitan areas and labour markets. She led and participated in several international research projects financed by the European Union Interreg IVC, International Visegrad Fund, Polish-German Foundation for Science and National Science Centre, Polish Ministry of Science and Higher Education.
108
M. Dej et al.
Wojciech Jarczewski is the Head of the National Institute for Spatial Policy and Housing in Poland. He is a social and economic geographer who in theoretical and applied research projects seeks to address the question of “How can local, regional and national governments revive and support social and economic development?” He has sought to answer this question by pursuing national and international research projects that explore the following issues: (1) ways for townships to attract investors, (2) geographic factors associated with the establishment of new plants, (3) evaluation of project effectiveness in the case of projects and programmes co-financed by the European Union, (4) urban renewal, (5) creation of new townships, (6) development of recreational geothermal bathing facilities. His research has been published in numerous reports, research papers and books. Michał Chlebicki works at the Faculty of Philosophy and Sociology at the University of Warsaw in Poland. He holds M.A. in Sociology of Economy and Marketing Research and B.A. in Comparative Culture Studies. In addition, he is a former Ph.D. student in sociology at Jagiellonian University. He specialises in sociology of culture and methodology of social research including secondary data analysis, information technology and statistics. He worked or collaborated with, among others, the Centre for Culture Statistics at Statistic Office in Kraków, the National Library of Poland, the Copernicus Science Centre, Institute of Urban Development, where he was coordinating research projects, conducting research and analysis, preparing reports and presentations.
Chapter 7
Identifying Factors Relevant for Firms’ Location and Relocation. The Case of Ticino Federica Rossi
7.1
Introduction and Theoretical Framework
In the regional economics literature, the location decision of firms is one of the most studied topics due to the important consequences that this issue has on the economic development of a territory (McCann 2001). Although it has been studied since the middle of the nineteenth century, the scientific debate on location choice and relocation motivations is ongoing. Indeed, the relevance of these issues goes beyond purely academic curiosity: from an economic policy point of view, knowing qualities and weaknesses of a territory facilitates the prevention of relocation of firms and the promotion of location attractiveness. From a theoretical perspective, a substantial part of the economic literature concentrates on firm location decisions (see McCann 2001 for an overview), analysing relocation as a special case of location theories. Firm relocation can be defined as a particular form of locational adjustment to changes in markets, environmental regulations and technological progress (Pellenbarg et al. 2002). There are two forms of relocation: complete and partial migration. The first one implies the movement of an establishment from address A to another one B, while partial relocation consists of setting up a new local unit, linked to the pre-existing one (Schmenner 1980; Brouwer et al. 2004). In the survey presented in the chapter, only complete relocation is taken into account. Although a lot has been done from a theoretical point of view or using aggregated data, much less attention has been paid to empirical studies on location and relocation at the micro-level, due to the scarcity of firm-level data. Among others, some recent works taking advantage of micro data and explaining location choices are Brülhart et al. (2012) and Alamá-Sabater et al. (2011). While for relocation F. Rossi (&) Institute of Economic Research, Università della Svizzera italiana (USI), Lugano, Switzerland e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_7
109
110
F. Rossi
behaviour, some contributions come from Targa et al. (2006), Brouwer et al. (2004), Hu et al. (2008), and Bodenmann and Axhausen (2012). In particular, Brülhart et al. (2012) analysed if agglomeration economies can reduce firm’s sensitivity to tax differentials, using data on start-up firms in Switzerland. They found that agglomeration externalities actually attenuate the impact of tax differentials on location choices. Willingers and Van Wee (2007) focused, instead, on the specification of accessibility indicators for modelling the location choices of offices in the Netherlands. The analysis showed that the proposed potential accessibility indicators (Tanner and Box-Cox impedance functions) are well capable of explaining the location choices of offices in the Dutch Randstad area. Finally, Alamá-Sabater et al. (2011) explored the role of inter-territorial spillovers in location choices of industrial firms in Spain. They found that human capital, agglomeration economies, and industrial land availability are the main factors driving entrepreneurs’ decisions. Concerning relocation studies, Targa et al. (2006) specified a model capturing the relationship between the migration propensity of the firm and a set of relocation factors (i.e. accessibility, agglomeration economies). They found a positive significant association between transportation supply and firm-level relocation decisions. Brouwer et al. (2004) investigated the determinants of firm relocation behaviour in twenty-one European countries. Results suggested that internal growth factors induce relocation and firms that serve larger markets relocate more often. One of the most recent studies on firms’ relocation in Switzerland is Bodenmann and Axhausen (2012). They modelled firm relocation choices using data from commercial registers of three Swiss Cantons. Results showed that distance, local taxes, and cantonal business development strongly influence relocating firms’ destination choice. Finally, Hu et al. (2008) examined the most important factors for firms’ decisions to relocate in the past and their intention to move in the future. They found that internal characteristics of the firm, features of the location sites and the general economic environment affect past and future choices. This work was particularly interesting for the purposes of the current analysis, since the authors used individual business surveys, where firms rated a set of factors in terms of their importance in the decision to relocate. Indeed, this chapter aims to contribute to the empirical investigation of the location and relocation choices of firms located in Ticino, the southern Canton of Switzerland. Ticino represents an interesting case study. It is a border region: it belongs to Switzerland and therefore enjoys all the benefits deriving from the well-known stable institutional background and from the international reputation of the country as a safe location for investments. At the same time, Ticino takes advantage of its proximity to Italy: as a border region, it has a privileged position to access the southern European markets.
7 Identifying Factors Relevant for Firms’ Location …
111
This chapter has two objectives. Firstly, from a firm point of view, it provides insights into the factors influencing location attractiveness, and secondly, it investigates the drivers of relocation behaviour. Although these two objectives are different, they are interrelated. Indeed, studying the characteristics attracting firms to the territory means focusing on pull factors, as defined in previous studies (De Bok and Sanders 2005). On the other hand, in order to investigate the drivers of relocation behaviour, the fact that one site is substituted for another must be explicitly taken into account; therefore, the characteristics of both origin and destination regions should be considered. In this case, also push factors, which are all the unsatisfying features determining the decision to leave the current location, are important in defining relocation motivations (De Bok and Sanders 2005). Therefore, one factor could act, at the same time, as a push or as a pull factor depending on the point of view of origin or destination location. Previous studies have investigated a large set of factors influencing location and relocation behaviour which can be divided into five main groups: • Institutional environment. The literature shows that institutional environment has a significant impact on firm location and relocation behaviour. In particular, low taxation, modest bureaucracy, political stability and legal order efficiency tend to attract and restrain businesses (Pellenbarg et al. 2002; Salvesen and Renski 2003). At the regional level, there is evidence that quality-of-life aspects, such as the presence of cultural and natural amenities (Love and Crompton 1999), make certain locations more attractive to firms. Moreover, firm relocation could be influenced by the local quality of life situation. Deteriorated physical surroundings reduce the attractiveness of the environment for potential customers, which indirectly influences the sales and profits of local firms (Sleutjes and Beckers 2013). • Competitive environment. Agglomeration economies are one of the key topics in regional economics and, in particular, in explaining firm location choices. Agglomeration of firms within the same industry, interdependence between firms of different sectors and partners’ proximity are often taken into account in location and relocation studies (Manjón-Antolín et al. 2010; Targa et al. 2006; Kronenberg 2013). In particular, Bodenmann and Axhausen (2012) measured the influence of localisation and urbanisation economies on the destination location attractiveness. They defined localisation economies as positive externalities resulting from an agglomeration of same industry firms, while urbanisation economies emerge from interdependences between firms of different sectors. They found that both have a positive effect on location attractiveness. • Accessibility. One of the older theories about firm location is based on the considerations of transportation costs and accessibility (Weber 1909). Accessibility is principally evaluated in terms of distance to physical infrastructure (highway entrances or train stations) and secondly in terms of potential accessibility measures (De Bok and Sanders 2005). Previous studies underline
112
F. Rossi
the importance of these factors in explaining location and relocation decisions (Van Dijk and Pellenbarg 2000; Sleutjes and Beckers 2013; Holl 2004; Targa et al. 2006; Holguin-Veras et al. 2005). • Production inputs. Cost minimisation is a priority for successful firms. In this sense, the costs of the inputs, such as wages, energy and real estate prices, but also space and skilled workforce availability, must be taken into account when looking at the decisions of firms (Manjón-Antolín et al. 2010; Sleuwaegen and Pennings 2006; Holl 2004). Kronenberg (2013) found that low-tech firms paying high average salaries have an increased propensity to relocate, suggesting that these firms move in order to save costs. Moreover, there is evidence that lack of space for expansion is often push-factor number one (De Bok and Sanders 2005). • Additional amenities. According to the literature, some additional facilities, such as airport proximity, research and development (R&D) incentives and proximity to a border, could encourage firms in locating to a particular municipality (Kawamura 2004; Strauss-Kahn and Vives 2009). The remainder of this chapter is organised as follows: the next section is dedicated to the description of the data and the methodology of analysis. The third part highlights and discusses the main empirical results. The final section offers conclusions and identifies future research directions.
7.2
Data and Methods
The analysis presented in this chapter is based on data collected with an online survey among Ticino firms in 2014. The questionnaire was explicitly addressed to the CEO/upper management of the firm. The survey was divided into four main sections. The first part was devoted to general information on the firm, such as sector, age, spatial organisation (head office and branches) and size. In the second section, firms were asked to rate the relevance and the presence/absence of some location factors (e.g. accessibility, cost of labour, business taxes) in Ticino and was the focus of the current chapter. The third part aimed to collect information on business relations, in order to understand frequency, destination and means of transport used for business trips. Finally, in the last section there were some hypothetical questions on future strategical actions of the firm. The questionnaire was sent to a sample of 5890 firms, which was randomly selected from the Bureau van Dijk business listing, reflecting the sectorial composition of Ticino economy (70% of the sampled businesses are in the tertiary sector). The overall response rate was 13%, in line with previous studies using questionnaires with firms (Targa et al. 2006; Kawamura 2004). From the 773 returned questionnaires, incomplete questionnaires and all firms with zero employees (which are mainly foreign branches with only legal residence in Ticino) were excluded, since they are not relevant for the purposes of the analysis. Consequently, 687 firms were included in the analysis.
7 Identifying Factors Relevant for Firms’ Location …
113
Looking at some sample descriptive statistics, 72% of firms operate in the tertiary sector; they are mainly more than 4 years old (74%) and sole proprietorships (80%). Moreover, about 71% are micro-firms (with fewer than 10 workers). These sample characteristics, which seem quite unbalanced, actually reflect the economic structure of Ticino, which is composed of mainly micro-firms (89% of the businesses in Ticino in 2014), sole-proprietorships (87%) and operating in services (81%). Among the 687 firms in the sample, a subsample of 40 firms, which decided to relocate to Ticino from abroad in the past, can be identified. Twenty-five of them came from other locations in Switzerland, 12 from Italy and 3 firms from the rest of Europe, in particular from the UK, Luxemburg and Spain. Twenty-eight firms are in the tertiary sector and 25 are sole proprietorships. Looking at the age of these firms, the sample is balanced: 21 are less than 3 years old. Micro firms represent 80% of this subsample, while only one firm has more than 50 employees. The forthcoming analysis focuses on two main issues: first, the factors influencing location attractiveness from a firm point of view, and second, the drivers of migration behaviour. In order to investigate the first issue, the entire sample (687 firms) was analysed, while the second objective exploits the information coming from the subsample of 40 firms, which relocated from abroad. The factors influencing location attractiveness were analysed proposing a graphical representation (Fig. 7.1) which allows identification of the weaknesses and the strengths of the territory. As in Hu et al. (2008), firms were asked to evaluate the relevance of 24 territory characteristics in their location choice on a Likert scale from 1 (definitely irrelevant) to 7 (definitely relevant). Moreover, firms provided a grade to assess the presence/absence of the same factors at their current location. This evaluation has been done on a Likert scale from 1 (definitely absent in the territory) to 10 (definitely present in the territory). Following the above classification of factors, based on the literature, firms evaluated the following characteristics: • Institutional environment, which includes institutional accessibility, openness and modest bureaucracy, favourable fiscal conditions for firms, political stability, legal order efficiency, financial market efficiency and quality of life. • Competitive environment, which includes proximity to current and potential partners, presence of firms belonging to the same sector, presence of competitors and presence of workers belonging to the same sector. • Accessibility, which includes public transport availability, lack of congestion, accessibility. • Production inputs, which include affordable real estate prices, space availability, low wages, low energy costs, low-skilled workers availability, medium-skilled workers availability and high-skilled workers availability. • Additional amenities, which include proximity to airports, proximity to a border, ease of doing research and development and positive economic outlook.
114
F. Rossi
Fig. 7.1 Graphical representation identifying weaknesses and strengths of the territory. Source Own compilation
Four quadrants may be distinguished from the two aspects as in Fig. 7.1, where the vertical axis represents the evaluation of the presence/absence of factors in the territory, while the horizontal axis indicates the evaluation of the factors’ relevance. In order to place each of the 24 factors in one of the quadrants, the statistical mode was used to synthesise information. In this way, each factor has two coordinates (the modes) in terms of its relevance and its presence in the territory. Thanks to this representation, four categories can be identified: • Strengths of the territory (top right quadrant) that include relevant and present factors. • Weaknesses of the territory (bottom right quadrant) that include relevant, but absent factors. • Negligible characteristics (bottom left quadrant) that include irrelevant and absent factors. • General setting (top left quadrant) that includes factors which are present but irrelevant. Considerations of the factors’ position on the graph could encourage policy makers’ interventions in order to consolidate the strengths or improve the weaknesses of the territory. The analysis could be repeated using subsamples, exploring the heterogeneity among firms belonging to different sectors, having different ages, dimensions or organisational structures.
7 Identifying Factors Relevant for Firms’ Location …
115
The second issue that this chapter investigates regards the drivers of migration behaviour of firms. In order to analyse this topic, a subsample of 40 firms which recently relocated to Ticino was asked to give a grade (from 1 to 10) of the presence/absence of the 24 factors previously listed both at their current and at their past locations. This data collection enables a comparison of the two locations and identifies factors for which the firm has gained (when the evaluation of Ticino was greater than the one associated to the previous location) or has lost (when previous location is better than Ticino for that particular characteristic). However, in order to understand if a firm has moved to Ticino driven by factors for which it has gained, it is necessary to combine this information (gain/loss) with the evaluation of the relevance that the firm attaches to the factor itself. If a firm gains in terms of one territory characteristic and considers it relevant for the location choice, then this particular aspect has encouraged the arrival in Ticino. In this case, the analysis explores the heterogeneity among firms, not only with different characteristics (sector, age, etc.) but also with different countries of origin.
7.3
Canton Ticino Case Study
Table 7.1 shows the values of the statistical mode of the 687 firms’ evaluations regarding the relevance and the presence/absence in Ticino of the 24 territory characteristics considered. Thanks to these values, each characteristic is placed in one of the four quadrants previously described, resulting in Fig. 7.2. In the general setting (top left quadrant of Fig. 7.2), there are the factors related to competitive environment (presence of competitors, firms and workers belonging to the same sector), as well as the proximity to the border. Therefore, in the current case study, firms evaluated the features related to agglomeration economies as being less important. Although agglomeration externalities have a role in explaining location and relocation decisions, as emphasised in the literature (Brülhart et al. 2012), the current results are probably driven by the absence of evident clusters in Ticino, with the exception of the fashion industry and financial services. Negligible characteristics (bottom left quadrant) include the availability of low-skilled workers and the proximity to an airport. Factors such as low wages and ease of doing research and development deserve a deeper discussion. These factors received the value 4 as the most frequent evaluation according to their relevance. This value, central in the scale from 1 to 7, indicates the position of indifference (neither definitely relevant nor the opposite). Therefore, for most of the firms in the sample, these characteristics are somewhat important, but not crucial in their location choice. Among the weaknesses of Ticino (bottom right quadrant), there are the availability of high- and medium-skilled workers, road congestion, energy and real estate prices, fiscal conditions for firms and availability of space around the plant.
116
F. Rossi
Table 7.1 Statistical mode values of the firms’ evaluations on relevance and presence/absence of territory characteristics in Ticino Territory characteristics Institutional environment Institutional accessibility and modest bureaucracy Favourable fiscal conditions for firms Political stability Legal order efficiency Financial market efficiency Quality of life Competitive environment Proximity to current and potential partners Presence of firms which belong to the same sector Presence of workers who belong to the same sector Presence of competitors Accessibility Public transport availability Accessibility Lack of congestion Production inputs Low wages Affordable real estate prices Low energy costs Space availability Low-skilled workers availability Medium-skilled workers availability High-skilled workers availability Additional amenities Proximity to an airport Proximity to a border Ease of doing R&D Positive economic outlook Source Own research
Relevance (stat. mode)
Presence/absence in Ticino (stat. mode)
7
8
7 7 6 6 6
5 10 8 8 8
6
8
4
8
4
8
4
10
5 7 7
10 10 5
4 7 7 7 1 6 6
5 5 5 1 5 5 5
1 1 4 7
5 10 5 7
Therefore, all the aspects related to the firms’ costs as well as the land use are not satisfying in terms of location attractiveness. These results should be considered as alarm bells: firms are unsatisfied with these features of Ticino. Low availability of skilled workers and absence of space for expansion could be convincing push factors for relocation. This is particularly
7 Identifying Factors Relevant for Firms’ Location …
117
Fig. 7.2 Graphical representation identifying weaknesses and strengths of Canton Ticino. Source Own research
true, given the arguments on these issues found in the literature. Bodenmann and Axhausen (2012) found positive effects of the availability of university graduates on municipality attractiveness. Therefore, the labour market, and in particular the educational level of the workforce, is important pull factor. Lack of space for expansion is often push factor number one (De Bok and Sanders 2005). Finally, fiscal policies that aim at attracting firms are crucial (Pellenbarg et al. 2002; Brülhart et al. 2012). Public transport availability, accessibility, positive economic outlook, proximity to existing and potential partners and, in general, the institutional environment make Ticino a welcoming location for economic activities (top right quadrant). The role of institutional setting and governmental policies in firms’ location choices is well known (Sleutjes and Beckers 2013; Van Dijk and Pellenbarg 2000). In particular, Du et al. (2008) examined the impact of economic institutions on the location choice of foreign direct investment. They found that US multinationals prefer to invest in those regions which have better protection of intellectual property rights, a lower degree of government intervention in business operations, a lower level of government corruption, and better contract enforcement. The graphic analysis was repeated for some subsamples, in order to examine the heterogeneity among firms having different characteristics. In particular, for
118
F. Rossi
manufacturing firms, some factors’ position on the graph changes: wages, financial market efficiency and economic outlook became weaknesses of the territory, while medium-skilled workers availability and ease of doing R&D moved to the top right quadrant (strengths of Ticino). Focusing on large- and medium-sized firms (with more than 50 workers), medium-skilled workers availability, ease of doing R&D, real estate prices and proximity to a border became strengths of the Canton Ticino. However, also for these types of firms, low wages are absent, but relevant (bottom right quadrant). Looking at start-up firms’ evaluations, favourable fiscal conditions and low energy costs are attractive features of Ticino, while wages, financial market efficiency and ease of doing R&D moved to the bottom right quadrant. This disaggregated analysis gives some suggestions to researchers and policy makers for designing policies which are able to attract new firms to the territory and prevent the existing ones from escaping. For example, improving access and decreasing bureaucracy for R&D activities will support start-up firms, enhancing their evaluation of this aspect. Moreover, all firms will benefit from policies aimed at increasing human capital, in order to widen the availability of high-skilled professional profiles. Concerning the analysis of the drivers of migration behaviour, 40 firms in the sample relocated to Ticino from abroad. The survey collects their evaluations not only for the current location in Ticino, but also for their city of origin. The analysis of the data enables a comparison of the two locations and identifies factors, which are possible drivers of firms’ relocation. The most interesting results were obtained by analysing this subsample by country of origin, in particular, distinguishing among firms coming from the rest of Switzerland, Italy and the rest of Europe (UK, Luxemburg and Spain). Looking at the 24 location factors previously described, firms coming from the other Swiss municipalities have lost in terms of road congestion, medium-skilled workers availability, proximity to an airport and financial market efficiency. Current and past locations are similar in terms of institutional environment, while firms have gained in terms of wages, proximity to a border and competitive environment. The situation is opposite for the Italian firms: they have mainly gained in terms of institutional environment, ease of doing R&D and proximity to a border, while they lost in terms of competitive environment, affordable real estate prices, congestion and space availability. Firms from the rest of the world have evaluations halfway between the above two groups: they have lost in terms of affordable real estate prices, space availability and wages, but they have gained in terms of competitive and institutional environment, accessibility and energy prices. In order to identify if a particular aspect has contributed to the decision to come to Ticino (it is a possible driver), the information on gain/loss was combined with the firm’s evaluation of the factor relevance. In particular, if a firm has gained in terms of factor X and has given a positive evaluation (greater than four on the Likert scale) to the relevance of this aspect for location choice, then factor X is identified as a possible relocation driver.
7 Identifying Factors Relevant for Firms’ Location …
119
Moreover, if a factor is a driver, it could be considered at the same time a push and a pull factor. Indeed, since the previous location is worse than the current one in terms of a given aspect, this could have pushed the firm to relocate. At the same time, the firm could have been attracted by better conditions in the destination municipality (pull factors). As an example, Fig. 7.3 focuses on the subsample of firms relocating to Ticino from Italy (12 firms). In particular, this figure reports the percentage of firms for which the considered aspects are possible drivers. The main result is that the institutional environment, as well as the positive economic outlook and the ease of doing R&D, was important relocation driver for firms coming from Italy. Indeed, 10 out of 12 Italian relocating firms considered the
Fig. 7.3 Percentage of firms relocating to Ticino from Italy, for which the factors are possible drivers. Source Own research
120
F. Rossi
institutional environment relevant for their location choice and Ticino institutions better than their previous municipalities. The importance of institutions in investment decisions is well known in the literature (Pellenbarg et al. 2002; Salvesen and Renski 2003; Love and Crompton 1999). This case study confirms that differences in the institutions accessibility, openness and modest bureaucracy, political stability and legal order efficiency between two neighbouring countries could drive firms’ relocation in favour of the most convenient location. The same considerations are true for firms from the rest of Europe. On the contrary, firms coming from other Swiss locations have heterogeneous driving factors, and for most of them the competitive environment, although they gain on this aspect, was not a driver for the arrival in Ticino. Instead, low wages and affordable real estate prices were drivers for 9 out of 25 Swiss firms. Indeed, the traditional production inputs, like workforce, real estate and energy, are cheaper in Ticino compared to other Swiss Cantons. Therefore, these firms probably moved in order to save costs (Manjón-Antolín et al. 2010; Sleuwaegen and Pennings 2006; Holl 2004), but at the same time they do not want to leave the Swiss stable institutional background. Moreover, they took advantage of the proximity to the Italian market. In conclusion, the differences among the motivations of firms that relocated to Ticino from different geographical areas clearly emerge. The same reasons could also attract firms to Ticino in the future; therefore, this analysis could support policy makers’ considerations of the regional development measures to be adopted, thereby boosting the attractiveness of the location.
7.4
Conclusions
The main goals of this chapter were to provide insights into the factors shaping location attractiveness from a firm point of view and to shed light on the drivers of migration behaviour of firms in Ticino. Using a rich dataset, the study contributes to the empirical investigation of firms’ location and relocation issues. The proposed approach clarifies the weaknesses and strengths of the territory and, therefore, suggests regional strategies in order to promote location attractiveness and prevent firms from leaving (for comparison, see Dej et al. in this volume). From the theoretical point of view, the chapter highlights the importance that some driving factors have in influencing location attractiveness and firms’ relocation choices in a border region. The collection of data directly from firms represents one of the strengths of the current study. Indeed, the analysis of firms’ opinions enables the researcher and the policy maker to take the pulse of the actual economic situation.
7 Identifying Factors Relevant for Firms’ Location …
121
In particular, the chapter confirms previous literature results about the relevance for location choices of stable institutional environment (Pellenbarg et al. 2002; Salvesen and Renski 2003), good accessibility conditions (Holl 2004; Targa et al. 2006; Holguin-Veras et al. 2005) and availability of cheap and suitable production inputs (Manjón-Antolín et al. 2010; Sleuwaegen and Pennings 2006). The analysis highlights that Ticino benefits from the well-known stable Swiss institutional background, with modest bureaucracy and legal order certainty, which attract firms to the territory, in particular from European countries, such as Italy, where the institutions are not so business friendly. The study also shows the weaknesses of the Ticino territory, in particular the inadequacy of the human capital (medium- and high-skilled workers are not sufficiently available) and road congestion. Although Ticino is just one of the twenty-six Cantons of Switzerland, some results of this case study could be generalised for the entire country. In particular, the strengths of Ticino, such as accessibility, quality of life and institutions, are also famous Swiss features. On the other hand, the costs of production (e.g. wages, real estate prices) are even higher in other parts of Switzerland; therefore, they surely also represent a weakness for other Swiss Cantons. The analysis of weaknesses/strengths helps the policy maker to design specific measures in order to address particular business necessities, from improvements in the educational system to urban development planning and mobility. These considerations lay the foundations for sound regional economic development policies which take into account the various territorial features. Indeed, relocation processes lead to changes in the economic structure, transforming equilibrium and potentially producing spatial conflicts (see Wójtowicz in this volume). Border regions like Ticino have particular features and will represent a favourable business location if they are able to merge the advantages of various countries (in the current case study, Italy and Switzerland). There are limitations to the current analysis; first, the number of relocating firms (only 40 in the sample) should be extended in order to increase the robustness of the results. Moreover, the current analysis does not allow for the establishment of a statistically significant relationship between the push/pull factors and the relocation decisions of firms. This issue could be investigated by applying econometric models. Future research should also extend the analysis to partial relocation, both at national and international levels (i.e. Foreign Direct Investments), in order to understand whether different types of relocation are influenced by the same territory characteristics or not. Acknowledgements The research presented in this chapter was supported by Fondazione Ferdinando e Laura Pica Alfieri, Lugano.
122
F. Rossi
References Alamá-Sabater, L., Artal-Tur, A., & Navarro-Azorín, J. M. (2011). Industrial location, spatial discrete choice models and the need to account for neighbourhood effects. The Annals of Regional Science, 47, 393–418. Bodenmann, B. Z., & Axhausen, K. W. (2012). Destination choice for relocating firms: A discrete choice model for the St. Gallen region, Switzerland. Papers in Regional Science, 91(2), 319– 341. Brouwer, A. E., Mariotti, I., & van Ommeren, J. N. (2004). The firm relocation decision: An empirical investigation. The Annals of Regional Science, 38(2), 335–347. Brülhart, M., Jametti, M., Schmidheiny, K. (2012). Do agglomeration economies reduce the sensitivity of firm location to tax differentials?. The Economic Journal, 122(563):1069–1093. De Bok, M., & Sanders, F. (2005). Firm location and the accessibility of location: Empirical results from the Netherlands. Transportation Research Record: Journal of the Transportation Research Board, 1902, 35–43. Du, J., Lu, Y., & Tao, Z. (2008). Economic institutions and FDI location choice: Evidence from US multinationals in China. Journal of Comparative Economics, 36, 412–429. Holguin-Veras, J., Xu, N., Levinson, H. S., et al. (2005). An investigation on the aggregate behavior of firm relocations to New Jersey (1990–1999) and the underlying market elasticities. Networks and Spatial Economics, 5, 293–331. Holl, A. (2004). Start-ups and relocations: Manufacturing plant location in Portugal. Papers in Regional Science, 83(4), 649–668. Hu, W., Cox, L. J., Wright, J., & Harris, T. R. (2008). Understanding firms’ relocation and expansion decisions using self-reported factor importance rating. The Review of Regional Studies, 38(1), 67–88. Kawamura, K. (2004). Transportation needs, location choice and perceived accessibility for businesses. Transportation Research Record: Journal of the Transportation Research Board, 1898, 202–210. Kronenberg, K. (2013). Firm relocation in the Netherlands: Why do firms move and where do they go? Papers in Regional Science, 92(4), 691–713. Love, L. L., & Crompton, J. L. (1999). The role of quality of life in business (re)location decisions. Journal of Business Research, 44(3), 211–222. Manjón-Antolín, M., Arauzo-Carod, J. M., & Liviano-Solis, D. (2010). Empirical studies in industrial location: An assessment of their methods and results. Journal of Regional Science, 50(3), 685–711. McCann, P. (2001). Modern urban and regional economics. Oxford: Oxford Press. Pellenbarg, P. H., van Wissen, L. J. G., & van Dijk, J. (2002). Firm relocation: State of the art and research prospects. Research Report 02D31, University of Groningen, Research Institute SOM (Systems, Organisations and Management). Salvesen, D., & Renski, H. (2003). The importance of quality of life in the location decisions of new economy firms. Chapel Hill, USA: Center for Urban and Regional Studies, University of North Carolina at Chapel Hill. Schmenner, R. W. (1980). Choosing new industrial capacity: One-site expansion, branching, and relocation. The Quarterly Journal of Economics, 95, 103–119. Sleutjes, B., & Beckers, P. (2013). Exploring the role of the neighbourhood in firm relocation: Differences between stayers and movers. Journal of Housing and the Built Environment, 28(3), 417–433. Sleuwaegen, L., & Pennings, E. (2006). International relocation of production: Where do firms go? Scottish Journal of Political Economy, 53(4), 430–446. Strauss-Kahn, V., & Vives, X. (2009). Why and where do headquarters move? Regional Science and Urban Economics, 39, 168–186.
7 Identifying Factors Relevant for Firms’ Location …
123
Targa, F., Clifton, K. J., & Mahmassani, H. S. (2006). Influence of transportation access on individual firm location decisions. Transportation Research Record: Journal of the Transportation Research Board, 1977, 179–189. Van Dijk, J., & Pellenbarg, P. H. (2000). Firm relocation decisions in the Netherlands: An ordered logit approach. Papers in Regional Science, 79(2), 191–219. Weber, A. (1909). Über den standort der industrien. ersterteil: reinetheorie des standorts. Tübingen, Mohr (English translation: Friedrich, C. J. (1929). Alfred Weber’s theory of location of industries. Chicago: University of Chicago Press.
Federica Rossi currently works as Project Manager of the Observatory of Economic Dynamics (O-De) at the Institute of Economic Research (IRE). She obtained a Ph.D. in Economics in May 2018 (Università della Svizzera italiana, Switzerland), a master’s degree in International Economics and Institutions in November 2012 (Milano Bicocca University, Italy) and a bachelor’s degree in Economics in September 2010 (Milano Bicocca University). Her research interests include accessibility, regional economics, relocation of firms and discrete choice methods.
Chapter 8
The Effects of the 2007 Global Economic Crisis on Firm Relocation Factors: SME Movements from Greece to Bulgaria Nikos Kapitsinis
8.1
Introduction
Firm relocation refers to ‘the movement of an establishment from one location to another’ (Brouwer et al. 2004: 336) and has been well investigated by economic geographers. Early reflections on corporate mobility have shown that labour and transportation cost, as well as taxation, are significant relocation factors (Hayter 1997; Domański 2003; Holl 2004; Jürgens and Krzywdzinski 2009). Reviewing the literature, one is especially struck by the fact that many scholars have predicted rather than explained business mobility by presupposing its determinants (Manjón-Antolín and Arauzo-Carod 2011; Kronenberg 2012; Hong 2013). However, wider processes in economy and society, such as globalisation, clearly affect firm relocation factors (Maskell and Malmberg 1999). The findings on business mobility in the literature quoted above refer to the pre-2007 period of significant technological improvement and high rates of economic growth for many national economies (pre-crisis period); however, the 2007 global economic crisis (GEC) has transformed the conditions under which enterprises operate (Hadjimichalis 2011). In the context of economic downturn, Hudson (2001) has argued that businesspeople move their firms to avoid economic decline. Indeed, since 2007 (post-crisis period), falling profit rate, economic decline and uncertainty have unevenly occurred in many national and regional economies, while austerity policies of different magnitude have been applied by national governments to resolve the crisis. Consequently, great shifts in operational cost have been recorded in different locations, altering the business environment. On these grounds, a significant question is raised that is linked to the way in which the GEC has influenced firm relocation factors.
N. Kapitsinis (&) Cardiff University, Cardiff, UK e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_8
125
126
N. Kapitsinis
Considering the dynamically changing business conditions, this chapter examines the impact of the crisis on the factors that affect firm mobility, analysing the relocation decisions of businesspeople. To this end, the discussion focuses on the relocation factors in the pre- and post-crisis periods and analyses their relative significance, by reviewing the crisis-driven changes in the socio-economic environment. While several authors compare business relocation strategies across space (see for example Castellani & Lavoratori and Koster & Pellenbarg in this volume), a comparison of relocation factors between the two periods could deepen our understanding of the crisis’ effects on firm mobility. Based on a thorough search of the relevant literature, it is believed that this is the first investigation that compares firm movements of a single type (specifically, of Greek companies to Bulgaria) in the period of economic growth with those under the 2007 global economic crisis. It is argued that in the context of the crisis-driven transformations, firm relocation factors should not be presupposed, as GEC may differentiate their significance. The case of the relocation of Greek small and medium enterprises (SMEs) to the neighbouring Bulgaria provides an intriguing context for achieving the research aim of this chapter. Greek business entry into Bulgaria started in the early-1990s during Bulgaria’s transition from a planned to a free market economy, as businesspeople were attracted by low labour cost and geographical proximity to Greece (Labrianidis 1997). The number of Greek enterprises in Bulgaria has recently trebled (Kapitsinis 2017). The explanation for the increased relocation of Greek firms to Bulgaria lies in the dynamics brought to the fore by the pre-crisis conditions of the Greek economy and society and the impact of crisis and austerity policies on the Greek socio-economic conditions. Greece is the European Union (EU) member state which has been mostly affected by the GEC, due to its position on the EU periphery and its weak productive structure and institutions (Hadjimichalis 2011). It is the only developed national economy since the Second World War that has recorded economic recession for six consecutive years (2008–2013), with a total loss of 25% of its gross domestic product. Additionally, the unemployment rate rocketed up to 27.5% in 2013 from 8.7% in 2008 (Eurostat 2017). Although the impact of the crisis in Greece was moderate before 2010, since then the socio-economic conditions have greatly deteriorated due to the austerity policies that the Greek government applied under the rescue plan agreed with the International Monetary Fund, the European Commission and the European Central Bank. These policies involved large cuts in wages and public expenditures and increases in taxation and energy cost (Kapitsinis et al. 2013) and affected the business environment with a rise in business operational cost, a drop in demand, a lack of external financing and an increase in firm indebtedness. On the basis of the high bankruptcy risk, thousands of Greek businesspeople have moved to the Balkans and Cyprus (Kapitsinis 2017). This chapter focuses on firm relocation to Bulgaria as it was the destination country for most Greek companies in both the pre- and post-crisis periods (Labrianidis and Vogiatzis 2013). SMEs are at the heart of the discussion for three reasons. First, they are of great importance to the Greek economy, producing, in 2015, 75% of total value added in
8 The Effects of the 2007 Global Economic Crisis …
127
Greece, whereas EU SMEs contributed 57% (Eurostat 2017). Furthermore, in the same year, 87% of private sector employees in Greece were hired by SMEs, whereas the average number in the EU was 67%. Second, SMEs in Greece have recorded the most vital impact in the aftermath of the GEC, as they present certain characteristics, including low levels of creativity, technology and productivity that have damaging effects on their competitiveness (Giannitsis 2013). Finally, they comprise the vast majority of the Greek enterprises in Bulgaria over time (Karagianni and Labrianidis 2001; Labrianidis and Vogiatzis 2013; Kapitsinis 2017). The remainder of the chapter is structured as follows: the next section scans the literature on firm mobility and its determinants. Next, the methodology that was adopted in the study is explained, before the empirical data are analysed. The last section summarises and outlines the main theoretical contributions and policy recommendations.
8.2
Literature Review on Firm Relocation
Taking advantage of the differentiation of socio-economic conditions, businesspeople move their companies to regions where they could increase the profit rate, determined by aspects of the economic and institutional context (Harvey 1982; Hudson 2001). Such factors are called firm relocation determinants and are unevenly distributed across space (Hayter 1997). Between the early-1970s and the middle-2000s, when most national economies recorded economic growth in the context of greater globalisation of the economy, crucially affecting business strategies across the globe, economic geographers have highlighted several relocation factors (see Castellani & Lavoratori and Wójtowicz in this volume). It is widely recognised that labour cost is the most significant relocation factor for companies of all sizes (Harvey 1982; Hayter 1997; Wilkinson et al. 2001; Domański 2003). High labour cost reduces the profit rate and deters economic activity from a region (Hudson 2001). Chiefly, the significance of labour cost is believed to increase as globalisation equalises the cost of other location factors across space (Maskell and Malmberg 1999). Labour skills are also important, as high-skilled labour attracts companies, especially the technology-intensive ones, to a region (Shaiken 1994; Wilkinson et al. 2001). Additionally, the high cost of transporting raw materials from their sources, or products to the markets, crucially adds to the business operational cost and makes a territory less attractive (Hayter 1997). Closely related to transportation cost, geographical proximity to the home region attracts businesspeople seeking cost reduction (Labrianidis 1997; Brouwer et al. 2004). Another lesson from the literature is that a high level of demand, which depends on the level of local wages and household income, enhances the attractiveness of a region (Domański 2003; Holl 2004). What remains lesser-known is that lack of demand drives enterprises out of a territory. Scholars have considered low taxation
128
N. Kapitsinis
to stimulate business activity in a region by virtue of lower expenditures, whereas high taxation suspends economic activity (Domański 2003; Jürgens and Krzywdzinski 2009). Finally, the access to raw materials and energy cost also affect firm relocation (Karagianni and Labrianidis 2001; Brouwer et al. 2004). Overall, a high formal institutional capacity attracts companies to a territory by setting the basis for undisrupted economic activity, which is particularly benefited by low bureaucracy and efficient legislation (Hayter 1997; Domański 2003; Koster and Kapitsinis 2015). Moreover, informal institutions are significant. High levels of trust and transparency reduce uncertainty and therefore crucially attract firms in a region (Guzik and Micek 2008, see also Rossi in this volume).
8.3
Methodology
Qualitative analysis was employed as it is suitable for explanatory inquiry, while offering a robust methodological link between the empirical data and the theoretical perspectives (Miles et al. 2013). Moreover, along with descriptive statistics, it facilitates an emphasis on the contextual issues that the respondents reveal and a detailed analysis of the socio-economic practices, such as firm relocation (Yin 2014). It is therefore particularly useful for this chapter which examines SME movements. Due to the lack of data, the sample was non-probability, although a high rate of representativeness was achieved. Specifically, there is no official database of Greek firms in Bulgaria, while the businesspeople in Bulgaria are not obliged to register their enterprises with chambers of commerce and industry. Consequently, non-probability convenience and snowball sampling methods were employed as they are suitable for qualitative explanatory inquiry on populations which are difficult to approach (Forza 2002). The representativeness of the sample is believed to be high. The sampling method and features are well defined (Fowler 2008), while the survey was replicated to businesspeople who own companies in all the economic sectors, at a rate similar to their prevalence in the Greek economy (Eurostat data). The analysis draws upon data collected in a fieldwork study undertaken in Bulgaria between May and June of 2014. In detail, 176 structured questionnaires were completed by Greek owners and managers of SMEs located in Bulgaria, specifically in the towns of Sandanski, Petrich, Blagoevgrad, Pernik and Sofia, and the villages of Krupnik, Melnik and Marikostinovo. Seventy-three SMEs had moved in the period from 1989 (when the previous regime in Bulgaria collapsed) to 2006, while the other 103 relocated between 2007 (the year of the beginning of the crisis) and 2014. Respondents were asked to rate the significance of factors that affected their decision to leave Greece and relocate to Bulgaria on a Likert scale of 1 (least significant) to 5 (most significant). Of the 176 respondents, 72 also participated in semi-structured interviews, with 68 of them having moved their companies after 2007 and 4 in the pre-crisis period.
8 The Effects of the 2007 Global Economic Crisis …
129
Interviews with businesspeople are particularly useful on the occasion of economic change, such as the GEC, in order to challenge existing theoretical perspectives (Schoenberger 1991). The interviews were conducted to expand the analysis and explain the meaning of the concepts addressed in the questionnaire on a smaller scale. In this regard, key limitations of the questionnaires were addressed by the interviews since these two methods complement each other efficiently (Johnson and Onwuegbuzie 2004). First, by overcoming the distance between the interviewer and the interviewee, second, by allowing the respondents to explain the reasons why they moved, and third, by discovering issues that were indiscernible within the questionnaires (Schoenberger 1991). In assessing the empirical data, descriptive statistics, comparative analysis and qualitative assessment were employed. The mean of the Likert value for each element that affected firm relocation was calculated. The comparison of relocation factors for firms that relocated in the pre-crisis period with those that have moved after 2007 facilitated the study of the crisis-driven effects. Moreover, the interview data were assessed by content analysis, exploring the way that businesspeople responded and the narrative they used—specific words or concepts, extracted with part of the text (Miles et al. 2013). Finally, the research outcomes were compared with theoretical propositions and literature findings. There are two noteworthy limitations in this study. The first to be acknowledged is the limitation of investigating the single case study of relocation of Greek firms to Bulgaria. However, the aim of this chapter is to explain and analyse an emerging socio-economic phenomenon in depth, seeking to investigate causal processes, rather than generalise (Yin 2014). The second limitation is that questionnaires and interviews were conducted based on a non-probability sample, attributed to the restricted availability of data, which is the biggest obstacle in studying Greek companies that operate in foreign countries. For this reason, the sample was purposive, thus affecting its representativeness.
8.4
Push Relocation Determinants from Greece
A significant increase in firm relocation from Greece to Bulgaria has been observed in the aftermath of the GEC and the austerity policies applied by the Greek government. Almost 1000 Greek companies were located in Bulgaria in 2006, while the number rose to 3000 in 2014 (Kapitsinis 2017). The major economic sector represented among the surveyed companies in the pre-crisis period was manufacturing (47% of sample firms), mainly textiles and clothing, which accords with findings from Karagianni and Labrianidis (2001). In the post-crisis period, most enterprises were in trade (29%), food and accommodation services (16.5%) and construction (12%). The number of Greek firms in Bulgaria has risen since 2007, as the Greek socio-economic environment has significantly changed due to the initial effects of the crisis on the Greek economy. Elements of this environment have influenced
N. Kapitsinis
130 Table 8.1 Significance of elements that affected firm exit from Greece Variable
Pre-crisis mean Likert value
Post-crisis mean Likert value
Difference in classification
High taxation and slow refund rate of VAT High level of bureaucracy High labour cost Difficult access to finance/ credit Difficult access/expensive energy and raw materials Market instability and huge fall of sales The Greek state could seize my bank deposits High level of rent prices Inability to use the post-dated checks Intensive competition
4.21 (1)
4.54 (1)
1 ! 1 (0)
4.12 (2) 3.58 (3) 3.24 (4)
4.3 (2) 3.22 (7) 3.86 (4)
2 ! 2 (0) 3 ! 7 (−4) 4 ! 4 (0)
4.36 −10.05 14.13
3.01 (5)
3.38 (6)
5 ! 6 (−1)
12.29
2.96 (6)
4.13 (3)
6 ! 3 (+3)
34.52
2.89 (7)
3.41 (5)
7 ! 5 (+2)
17.99
2.75 (8) 2.74 (9)
2.92 (9) 3.10 (8)
8 ! 9 (−1) 9 ! 8 (+1)
6.18 13.13
2.52 (10)
2.53 (11)
High transportation cost
2.42 (11)
2.43 (12)
10 ! 11 (−1) 11 ! 12 (−1) 12 ! 10 (+2)
The impact of Greece’s 2.26 (12) 2.81 (10) membership in the Eurozone Comparison between pre- and post-crisis period Source Own research
Difference in level of significance (%) 7.83
0.39 0.41 24.33
entrepreneurs’ decisions to move their companies from Greece. Table 8.1 illustrates the level of significance, i.e. the average Likert value, of each element in both the pre- and post-crisis periods.
8.4.1
Taxation
The descriptive analysis demonstrates that high taxation was a key element since it recorded the highest mean Likert value in both periods. A notable difference (7.83%) in its level of importance was captured in the post-crisis period. This research finding contrasts with data from the Greek Ministry of Finance (2017): the Greek corporate tax rate was higher in 2006 (29%) than in 2011, 2012 (20%) and 2013 (26%). Moreover, in 2013 it was lower than the rate in other EU member states, such as in Germany (30%) and in Italy (31.5%), according to data from Eurostat. Nevertheless, the austerity policies entailed a large number of exceptional
8 The Effects of the 2007 Global Economic Crisis …
131
taxes, such as the special solidarity levy—an exceptional tax based on income,— thus considerably increasing the total taxation. The Greek government implemented these onerous taxation measures attempting to increase public revenues and achieve fiscal balance. Therefore, taxation as more important in the post-crisis period, as high taxes make a region less attractive for economic activity (Jürgens and Krzywdzinski 2009). The rise in the significance of taxation could also be linked to the drop in corporate revenues and the mentality of many Greek SME owners, targeting quick and easy profits (Giannitsis 2013).
8.4.2
Labour Market
Looking at Table 8.1, one is especially struck by the fact that labour cost did not significantly affect firm exit from Greece in the post-crisis period. This is a significant finding considering that labour cost is generally identified as the most important relocation factor for companies of all sizes (Hayter 1997; Hudson 2001; Karagianni and Labrianidis 2001; Domański 2003). In the post-crisis period, labour cost dropped four places in the classification of the importance of the elements that affected firm exit from Greece. Furthermore, the analysis demonstrates that labour cost was the only factor that dropped in significance (−10%). The post-crisis significance of labour cost was the same for all the companies surveyed, regardless of their size and sector. Recent reforms in the Greek labour market are vital in explaining this research outcome. In the pre-crisis period, nominal wages had significantly increased between 1995 and 2005, albeit at a rate below that of labour productivity growth (Giannitsis 2013). The great importance of labour cost accords with evidence from Greek firm relocation to Bulgaria in the middle-1990s (Labrianidis 1997; Karagianni and Labrianidis 2001). However, since 2010, in the framework of the rescue plan, the Greek governments have squeezed wages down by 25%—leading the race in labour cost reduction among EU member states in 2010–2013 (Eurostat data). This has been managed mainly through a reduction of the minimum wage from gross €770 per month to €586 in 2012. Despite this, thousands of entrepreneurs relocated because the Greek business conditions were not conducive to firm profitability. While an entrepreneur who moved to Bulgaria in 2005 stated, ‘labour cost was crucial for my decision’, an informant who established a small trading enterprise in Sofia in 2011 mentioned, ‘it was quite low and, thus, insignificant’. Also, an interviewee located in Sandanski remarked that ‘labour cost was not an important factor.’
8.4.3
Market Conditions
The respondents accorded the greatest difference (40%) in the level of significance to the lack of demand in Greece. It was the sixth most important factor in times of
132
N. Kapitsinis
economic growth but gained three places among those respondents who have relocated since 2007. This is a critical research finding, given that most scholars have focused on the level of demand and the market size in the destination rather than the home region (Domański 2003; Holl 2004). Lack of demand was not mentioned in the interviews as a factor that notably influenced firm relocation decision in the pre-crisis period, since the level of sales under conditions of income growth and adequate leverage was adequate. In the wake of the GEC and recent austerity policies, consumers’ purchasing power was greatly reduced. Specifically, in 2010–2013, wages were squeezed and household income greatly dropped (−30%), while the unemployment rate increased by 190%, greatly affecting citizens’ purchasing power (Eurostat data). Consequently, commodities could not be sold and invested capital lost part of its value. In effect, realisation was restricted, with a resulting decrease in revenues and profits. The owner of a trading company, who moved to Blagoevgrad in 2012, stated that ‘sales fell by 50%, leading to a massive drop in cash reserves. This significantly affected my decision’. Besides, some entrepreneurs made sales on credit, which were never paid for, thereby leading to outstanding transactions. The owner of a company which sells fish products in Sofia explained, ‘customers were asking to pay the next day. We did not trust them, but we had to follow this pattern, as we were obliged to sell our perishable products quickly. However, we were left with many outstanding transactions’.
8.4.4
Finance
Access to external financing proved to be significant for decisions to relocate from Greece, recording the fourth highest mean Likert value in both periods. This is a significant finding since economic geographers have not yet fully explored the interrelationship between firm finance and relocation. Some scholars have examined the role of external finance conditions in attracting corporations to a region (Klagge and Martin 2005; Dunford et al. 2014). However, the way that access to external funding affects firm exit from a region has not received sufficient attention in economic geography literature. Given their characteristics, Greek SMEs have been historically credit-dependent (Giannitsis 2013). A specific understanding of external financing as a means of funding the working capital of Greek SMEs, like paying taxes, wages and suppliers, supports this claim (Labrianidis and Vogiatzis 2013). Overall, access to external financing has been difficult for Greek SMEs, similar to all small and medium enterprises (Giannitsis 2013), thus explaining its importance in the pre-crisis period. External finance conditions were found more significant in firm movement during the post-crisis period as they escalated the illiquidity problem of SMEs. To begin with, Greek SMEs have had significantly higher leverage since 2000 than in the previous period, owing to a decline in interest rates, on the grounds of the high level of competition within the Eurozone (Bank of Greece 2017). In this context,
8 The Effects of the 2007 Global Economic Crisis …
133
banks had increased the provision of loans to SMEs, despite the fact that they neither required business plans nor considered the entrepreneurs’ ability to pay off the debt, in a kind of business subprime lending. However, the rise in interest rates from 4.7% in December 2009 to 7.5% in October 2012, the drop in bank liquidity and the deterioration of SME performance have made access to external financing almost impossible for SMEs (Giannitsis 2013; Bank of Greece 2017). Therefore, external funding conditions recorded a significant difference (19%) in importance regarding their impact on SME exodus. In explaining the decision to move his logistics company to Petrich in 2013, a respondent stated, ‘in the context of the deep recession caused by the crisis and austerity policies, difficult access to external finance was the number one factor in the relocation decision. All the foreign partners required pre-payments but I could not receive a loan’.
8.4.5
Energy Prices
The evidence regarding high prices of energy and restricted access to raw materials explains a critical difference in the mean Likert value of this factor in the post-crisis period (12.3%). Energy has become more expensive in Greece in the post-crisis period. Furthermore, Greece has had the seventh highest electricity cost for industrial consumers in the EU in 2012–2014, recording a 59% increase from 2007 to 2014 (Eurostat data). Finally, in 2014, Greece had the third highest price for petroleum products in the EU. While the price of these products increased worldwide in the first few years after 2007 and then stabilised, in Greece the price has been steadily climbing, owing to the taxes on these products (Giannitsis 2013).
8.4.6
The Wider Formal Institutional Environment
Other formal institutional factors played a critical role in the decision to relocate in both periods. A lack of support from the state and the resulting entrepreneurial uncertainty were certain features of the Greek formal institutional arena that hampered SME economic performance. Sampling concerning these conditions yielded evidence on the importance of the formal institutions on SMEs. The owner of a trading firm in Marikostinovo stated, ‘the state did not back my company at the desirable level’. Moreover, value added tax (VAT) refund from the Greek state was quite slow, typical of the ineffective and dysfunctional structures of the Greek state in the wake of political hesitation for formal institutional improvement (Valyrakis 2008). A respondent who expanded her small manufacturing company to Pernik in 2005 said, ‘the biggest problem in Greece was the inability of the state to refund the VAT’. Delayed VAT refund is closely related to a high level of bureaucracy, which proved to play a notable role in both periods, with a higher level of significance in the aftermath of the GEC.
134
N. Kapitsinis
Several institutional transformations have been triggered by the crisis. Greek state flexibility has declined and legislation has become more complex. The owner of a small service company in Petrich pointed out, ‘more than 800 bills have been voted in the last five years regarding taxation’. Moreover, tax avoidance which was present in Greece even in the pre-crisis period (Valyrakis 2008) has recently increased. For instance, tax exemption of wealthy businesspeople, like shipowners, became more prevalent (Wall Street Journal 2015). This is an important example of increased rent-seeking, while the burden of tax evasion, which is quite high, has fallen on the shoulders of taxpayers who have diligently been paying their taxes (Giannitsis 2013).
8.4.7
Informal Institutions
Moving to cultural factors, norms and values supplemented the formal institutions and significantly affected relocation from Greece. Social trust was the most important of these factors. Particularly prior to 2007, many transactions had been conducted by post-dated bank checks, based on a moderate level of trust within close social networks (Lyberaki and Paraskevopoulos 2002). However, in the post-crisis period, social trust collapsed (Giannitsis 2013). Therefore, businesspeople could neither use nor redeem the bank checks. A respondent who has been running a manufacturing company in Blagoevgrad since 2010 confirmed, ‘the inability to use the post-dated bank checks led to outstanding transactions and further fall of cash reserves. This was a significant aspect of my decision’. Besides, in the context of the recently applied economic policies, businesspeople’s trust in the state, which also has been always low (Lyberaki and Paraskevopoulos 2002), collapsed. The owner of a trading firm in Blagoevgrad stated, ‘although the government did not back my firm in the pre-crisis period, since 2010 it has turned against it’. Corruption, another historical element within the Greek society, grew in the post-crisis period, thereby deteriorating business performance (Giannitsis 2013). Greece fell from the 56th place in the global transparency table in 2007 to the 94th place in 2012 (Transparency International 2014). A respondent who runs a food service firm in Sofia revealed the high level of corruption in Greece by explaining that ‘the government officials had asked for bribes to back me up’. Even though Bulgaria had almost just as bad corruption level, Greek businesspeople still decided that they were better off there than in Greece. In a careful assessment of the Greek business environment, it could be argued that firm relocation was necessary for businesspeople to avoid business failure in the post-crisis period. The high taxation and energy cost, the large drop in demand, the lack of external finance, the rise of bureaucracy and corruption and the collapse of trust have led to a considerable deterioration of business performance. Therefore, after applying internal restructurings, such as reduction of labour cost, dismissals and cuts in investments (Labrianidis and Vogiatzis 2013), that proved inadequate,
8 The Effects of the 2007 Global Economic Crisis …
135
and with conditions deteriorating as time passed, entrepreneurs decided to move seeking business survival.
8.5
The Factors that Attracted Greek SMEs in Bulgaria
Businesspeople strove to exploit the economic and institutional differentiation between Greece and Bulgaria in both the pre- and post-crisis periods. The efforts since 1989, of the Bulgarian state to attract foreign investments, by regulating low business tax rate and labour cost, appear to have been successful (Slaveski and Nedanovski 2002). Furthermore, Bulgaria has been less affected by the GEC, while its government adopted less stringent austerity measures (Petkov 2014), entailing less significant changes in the Bulgarian business conditions than in Greece. Low labour cost and taxation, and geographical proximity to Greece were the elements that mostly attracted Greek SMEs in both periods, recording the highest mean Likert values (Table 8.2).
8.5.1
Tax System
Low taxation proved to be crucial in both periods. In 1999, the Bulgarian corporate tax rate was 23.5%, while in Greece the average business tax rate in the 1990s was above 30% (Eurostat data). In 2007, the corporate tax rate in Bulgaria was reduced to 10%, the lowest in the EU, which was very attractive to Greek businesspeople, as it facilitated a substantial reduction in operational cost for their enterprises. The owner of a wholesale trading company in Sofia explained, ‘low taxation allows me to have high cash reserves and profits’.
8.5.2
Geographical Proximity
Bulgaria’s geographical proximity to Greece was found to be significant in both periods since it would allow entrepreneurs to be close to existing customers and partners, which has been a crucial element in relocation decisions by businesspeople, overall (Labrianidis 1997). Indeed, the geographical proximity to Greece minimised transportation cost for 48% of the respondents addressing the Greek market. The owner of a manufacturing firm in Krupnik said, ‘geographical proximity helped me stay close to customers and reduce transportation cost’.
8.5.3
Labour Features
Labour cost proved to be important in both periods, allowing Greek entrepreneurs to reduce their operational cost. Indeed, the owner of a manufacturing firm in
136
N. Kapitsinis
Table 8.2 Significance of elements that attracted firms in Bulgaria Variable
Pre-crisis mean Likert value
Post-crisis mean Likert value
Difference in classification
Low taxation and prompt refund rate of VAT Geographical proximity to Greece Low labour cost Low level of bureaucracy Market stability and stable economic context Bulgarian state did not have many financial requirements Easy access/cheap energy and raw materials Low level of rent prices Low transportation cost
4.62 (1)
4.52 (2)
1 ! 2 (−1)
−2.16
4.34 (2)
4.57 (1)
2 ! 1 (+1)
5.30
4.16 (3) 3.49 (4) 3.08 (5)
3.71 (4) 3.94 (3) 3.42 (7)
3 ! 4 (−1) 4 ! 3 (+1) 5 ! 7 (−2)
−10.82 12.89 11.04
3.05 (6)
3.50 (6)
6 ! 6 (0)
14.75
2.86 (7)
3.53 (5)
7 ! 5 (+2)
23.43
2.78 (8) 2.75 (9)
3.27 (9) 3.15 (10)
17.63 14.55
2.53 (10)
2.22 (16)
2.53 (11)
3.27 (8)
8 ! 9 (−1) 9 ! 10 (−1) 10 ! 16 (−6) 11 ! 8 (+3)
2.34 (12)
2.63 (13)
2.32 (13)
2.79 (11)
2.22 (14)
2.63 (12)
2.17 (15)
2.50 (14)
2.13 (16)
2.43 (15)
Easy access to finance/ credit Low level of competition in Bulgaria Presence of Greek firms in Bulgaria Strong ties between private firms in Bulgaria and Greece Size of the Bulgarian market The impact of Bulgaria’s non-membership in the Eurozone Cultural connections with Greece Comparison between pre- and Source Own research
Difference in level of importance (%)
−12.25 24.25
12 ! 13 (−1) 13 ! 11 (+2)
12.39
14 ! 12 (+2) 15 ! 14 (+1)
18.47
16 ! 15 (+1)
14.08
20.26
15.21
post-crisis period
Krupnik explained, ‘lower labour cost was significant, meaning that I could greatly increase my profits’. However, it is worth noting that it was slightly underrated in the post-crisis period as the gap of minimum wage between the two countries fell from 1:12 in 1995 to 1:4 in 2012 (Eurostat data). Labour cost was less important for micro companies (sixth place in the classification) than for the small and medium ones (third place). This could be explained by the family character of the Greek micro enterprises (Liargovas 1998). A common feature in the sample was that family members were employed unofficially by businesspeople, testifying to the
8 The Effects of the 2007 Global Economic Crisis …
137
informal economy operating among Greek SMEs (Giannitsis 2013). It should also be considered that the importance of labour cost increases proportionally to the size of the firm (ILO 2015). Contrary to suggestions by Shaiken (1994) and Wilkinson et al. (2001), employees’ skills in Bulgaria had a minor and negative impact on respondents’ decisions to relocate. While in Bulgaria the rate of enterprises and employees involved in training activities is higher than in Greece, the level and years of education are lower (Eurostat data). A respondent who moved his manufacturing enterprise in Blagoevgrad in 2003 stated, ‘I have no trust in employees’ attitude and skills’. However, owners of relocated Greek firms were obliged to hire Bulgarian employees in order to stimulate the demand from the local market. For example, the owner of a retail trading company in Petrich argued, ‘if I did not hire a Bulgarian employee, people would not buy from my company’.
8.5.4
Access to External Finance
For other factors, compelling evidence of crucial differences in their role in attracting companies in Bulgaria was found, with the mean Likert value of access to external finance recording the greatest difference (−12%). This can be explained by two key developments. First, the financing needs of Greek SMEs decreased after relocation to Bulgaria, as a result of the considerable reduction in overall operational cost. This cost dropped by 40% on average before 2007, while an even greater decline (60%) was recorded for entrepreneurs who relocated in the post-crisis period as the Greek business environment became more expensive. This is evident in the view of a respondent who moved his small manufacturing firm from Athens to Sandanski in 2009 and stated, ‘external financing needs were reduced because costs have declined markedly’. Second, Greek entrepreneurs’ changing mentality, in the aftermath of the crisis, towards a lower dependence on external funding was crucial. Several businesspeople decided to change their attitude to unsustainable loans, which escalated the impact of the crisis on SME performance and affected relocation from Greece.
8.5.5
Energy and Transportation Cost
The mean Likert values of affordable energy and transportation cost were found higher in the post-crisis period. The divergence of energy prices between the two countries owing to their significant rise in Greece was crucial for this finding. The owner of a manufacturing enterprise which was set up in Krupnik in 2012 concluded, ‘I decided to move to Bulgaria due to the high availability and low price of raw materials’. Closely connected with energy prices, transportation cost was more significant in the aftermath of the crisis. The recent rise in the price of petroleum
138
N. Kapitsinis
products in Greece, and the fact that Bulgaria had among the lowest price levels for the same products in the EU during the last decade (Eurostat data), determined the difference in the importance of the transportation cost.
8.5.6
The Features of the State
In Bulgaria, the functioning of the state and a modest level of bureaucracy make for a flexible business environment, thus paving the way for low-cost transactions between the state and the citizens and facilitating the relocation of SMEs in both periods, although Bulgaria still has much to improve (Racoviţă 2011). The contrast with Greece where bureaucracy is quite high is important (Valyrakis 2008). Therefore, bureaucracy in Bulgaria was found important relatively to the conditions in Greece. This research outcome contrasts with that of Bitzenis (2006), who found the high level of bureaucracy in Bulgaria to be a significant obstacle to Greek firm entry in the late-1990s. The modest level of bureaucracy also leads to prompt VAT refunds, taking only 32 days, a considerable draw for Greek businesspeople in Bulgaria. Prompt VAT refund is not a new development and has been influenced by the efforts of the Bulgarian government to attract foreign investment (Slaveski and Nedanovski 2002). The manager of a manufacturing company in Sofia argued, ‘here the state is organised; it always refunds VAT on time’. Finally, the Bulgarian state was flexible enough to accelerate the equipping of electronic government. The owner of a manufacturing company in Petrich mentioned, ‘I conduct all the transactions electronically with the online services of the state’.
8.5.7
The Accession to the EU
Among the few options in terms of cross-border relocation to low-cost economies, most of the businesspeople moved to Bulgaria, since according to the respondents, it was a member of the EU. Indeed, Bulgaria’s accession to the EU in 2007 has significantly facilitated the decision to relocate for most respondents (71%). This was an important development for Greek investment in Bulgaria, leading to open borders and the abolition of any trade restriction between the two countries. The owner of a company that develops food products in Marikostinovo described, ‘my decision to choose Bulgaria and not Albania was influenced by Bulgaria’s accession to the EU’. Indeed, Central and Eastern European states’ accession to the EU has facilitated foreign companies’ entry into their markets (Hudson 2001).
8.5.8
Informal Institutional Environment
An environment of ‘entrepreneurial safety’, based on a perceived high level of social trust, positively affected entrepreneurs’ decisions to enter the Bulgarian
8 The Effects of the 2007 Global Economic Crisis …
139
market in both the pre- and post-crisis periods. Accordingly, the level of trust among social actors was revealed to be high and significant according to the respondents, as trust fosters economic growth and regional attractiveness (Guzik and Micek 2008). The owner of a hotel in Melnik explained, ‘in Bulgaria, people’s behaviour makes me feel stable and secure as an entrepreneur’. Nevertheless, the evidence about the level of social trust in Bulgaria is contradictory. While Newton (2001) has indicated that Bulgaria records fragile social trust, Danchev (2005) has shown that the level of confidence in Bulgaria is in a good state with increasing trends since 2000, primarily at the microlevel within local communities. It is worth noting that respondents deemed trust among the economic actors to be higher in Bulgaria than in Greece, as a respondent who established a retail trading company in Blagoevgrad in 2012 said, ‘there is certainty that I will receive payment from customers’. Finally, corruption has been an adverse feature of the Bulgarian society (Pashev 2011). A respondent who has been running a trading enterprise in Sofia since 2011 stated, ‘corruption is a significant problem and constraint. However, it is reasonable since people cannot survive on a gross €200 monthly salary. The government policies concerning minimum wage drive them to ask for bribes’. Surprisingly, corruption was reported to be an obstacle by few Greek businesspeople, in contrast to previous evidence (Bitzenis 2006), highlighting corruption as a significant barrier to Greek entrepreneurs’ decision to set up business in Bulgaria. This research finding can be explained by looking at three issues. First, Greek entrepreneurs crossed the border, despite the massive corruption in the neighbouring country, because of their prior experience with widespread corruption in Greece (Valyrakis 2008). Second, the business conditions in Greece became so hazardous that businesspeople were prepared to do whatever it took to avoid them and stay in business. Third, the transparency index of Bulgaria has increased since 2005, while a drop in the same index in Greece has been recorded (Transparency International 2014). However, corruption in Bulgaria was still high. Overall, the findings about corruption are contradictory. The respondents criticised corruption in Greece, whereas most of them ignored the extended corruption in Bulgaria, highlighting their short-termism (Giannitsis 2013). It is confirmed that the mentality of Greek entrepreneurs remains unchanged in Bulgaria as they overlook crucial issues in search of quick and easy profits.
8.6
Conclusions
This chapter has provided useful insights that deepen our understanding of the way in which the 2007 global economic crisis has affected firm relocation factors. Many studies presuppose the determinants of business mobility (Manjón-Antolín and Arauzo-Carod 2011; Kronenberg 2012; Hong 2013). However, the research findings indicate that, in the circumstances of economic crisis, firm relocation factors
140
N. Kapitsinis
should not be presupposed, due to the rapidly changing economic and institutional conditions. This chapter contributes to the rich literature on firm mobility (Hayter 1997; Labrianidis 1997; Hudson 2001; Domański 2003; Brouwer et al. 2004) by suggesting that the 2007 global economic crisis has significantly influenced the determinants of firm relocation from Greece to Bulgaria. It was indicated that the importance of relocation factors recorded significant differences between the preand post-crisis periods, thus proposing the reconsideration of terminology on firm mobility. Therefore, the results enrich the empirical and theoretical debate on firm mobility by highlighting the emerging changes observed concerning the significance of relocation factors. Regarding the push relocation determinants from Greece, during the post-crisis period, the level of importance of several factors was different compared to the context of economic growth. First, labour cost in Greece was found relatively insignificant for the post-crisis decision to relocate from Greece. This research outcome is of great importance for the theory, since labour cost has been highlighted as the most important factor in the pre-crisis period (Shaiken 1994; Hayter 1997; Wilkinson et al. 2001; Domański 2003). The low level of the importance of labour cost for the post-crisis SME movements from Greece is attributed to the severe wage reduction in the context of the austerity policies. Second, lack of demand emerged as a crucial push determinant for post-crisis relocation from Greece owing to unemployment growth and wage reduction in Greece. This finding adds to the existing body of knowledge on firm mobility, as most studies have focused on the level of demand in the destination territory (Domański 2003; Holl 2004). Third, access to external finance emerged as another major element greatly affecting business relocation from Greece, while its role was upgraded between the two periods, on the grounds of SMEs’ subprime loans and the restricted access to bank credit in the post-crisis period. To date, economic geographers have focused on the importance of external funding conditions in attracting companies in a territory (Klagge and Martin 2005; Dunford et al. 2014). Fourth, the high levels of bureaucracy and legislative perplexity were more important for SME exodus from Greece in the post-crisis period. Fourth, the role of social trust was also more significant among respondents who have relocated after 2007, as it collapsed in the post-crisis context of uncertainty. Summarily, formal and informal institutions have deteriorated in crisis-ridden Greece, thus leading to the SME relocation from the country. This study indicated that it was the economic and institutional environment as a whole that affected firm movement from Greece to Bulgaria. In other words, it is not just a couple of elements which affected firm exit, as the Union of Hellenic Chambers has claimed, focusing on the role of taxation (2016). It was demonstrated that the significance of several factors that attracted Greek businesspeople to Bulgaria has also changed in the aftermath of the crisis. The price of raw materials and transportation cost grew in importance in the post-crisis period, whereas the role of others, such as access to external financing, was downgraded. Overall, geographical proximity to Greece, low taxation and labour cost, and the modest level of bureaucracy in Bulgaria attracted the Greek
8 The Effects of the 2007 Global Economic Crisis …
141
entrepreneurs in both periods, lending support to existing evidence on the crucial role of these factors in firm mobility (Karagianni and Labrianidis 2001; Domański 2003; Jürgens and Krzywdzinski 2009). It is important to notice that some factors in Bulgaria, including the levels of bureaucracy and transparency, demonstrate relative importance compared to the conditions in Greece. Finally, there is a very thin line to define the effects of Bulgaria’s accession to the EU, and the effects of the crisis on the increased firm relocation from Greece to Bulgaria, since they emerged in the same year (2007). The impact of the crisis and the austerity policies deteriorated the economic and institutional context. Therefore, the Greek businesspeople moved their companies to stay in business. The open borders with Bulgaria, after 2007, facilitated this relocation as entrepreneurs chose Bulgaria instead of other neighbouring low-cost economies due to its accession to the EU. Useful suggestions for policy directives can be made based on the findings. Globally, most policy measures adopted to tackle the impact of the GEC focus on labour cost reduction, with governments arguing that this would lead to economic growth and an increase in investments. However, in the case of Greece, this approach was completely ineffective and socially unfair, exacerbating the socio-economic inequalities (Giannitsis 2013). Despite a significant reduction in the cost of labour, economic recession has deepened and entrepreneurs have been moving their firms from Greece because their operations have become unprofitable. Any improvement in the Greek economy will be complex and not an easy process. Contrary to the opinions of several associations which focus just on the reduction of taxation (Union of Hellenic Chambers 2016), any policy changes should not be undertaken piecemeal, but rather as part of a coordinated plan targeting multiple factors. Among other elements, policy interventions should be related to wage increase, full-time employment, technological upgrade and a major restriction of corruption and bureaucracy.
References Bank of Greece. (2017). Online database. http://www.bankofgreece.gr/Pages/el/Statistics/default. aspx. Accessed February 11, 2017. Bitzenis, A. (2006). Determinants of Greek FDI outflows in the Balkan Region: The case of Greek entrepreneurs in Bulgaria. Eastern European Economics, 44(3), 79–96. Brouwer, A., Mariotti, I., & van Ommeren, J. (2004). The firm relocation decision: An empirical investigation. The Annals of Regional Science, 38, 335–347. Danchev, A. (2005). Social capital influence on sustainability of development (case study of Bulgaria). Sustainable Development, 13, 25–37. Domański, B. (2003). Industrial change and foreign direct investment in the postsocialist economy: The case of Poland. European Urban and Regional Studies, 10(2), 99–118. Dunford, M., Liu, W., Liu, Z., & Yeung, G. (2014). Geography, trade and regional development: The role of wage costs, exchange rates and currency/capital movements. Journal of Economic Geography, 14(6), 1175–1197. Eurostat. (2017). Online database. Available at: http://epp.eurostat.ec.europa.eu. Accessed January 28, 2017.
142
N. Kapitsinis
Forza, C. (2002). Survey research in operations management: A process-based perspective. International Journal of Operations and Production Management, 22(2), 152–194. Fowler, F. (2008). Survey research methods (4th ed.). London: Sage. Giannitsis, A. (2013). Greece in crisis (in Greek). Polis, Athens. Guzik, R., & Micek, G. (2008). The impact of delocalisation on the european software industry. In L. Labrianidis (Ed.), The moving frontier. The changing geography of production in labour-intensive industries (pp. 229–251). Aldershot: Ashgate. Hadjimichalis, C. (2011). Uneven geographical development and socio-spatial justice and solidarity: European regions after the 2009 financial crisis. European Urban and Regional Studies, 18(3), 254–274. Harvey, D. (1982). The limits to capital. Oxford: Blackwell. Hayter, R. (1997). The dynamics of industrial location. The factory, the firm and the production system. Oxford: Wiley. Holl, A. (2004). Start-ups and relocations: Manufacturing plant location in Portugal. Papers in Regional Science, 83(4), 649–668. Hong, S. (2013). Agglomeration and relocation: Manufacturing plant relocation in Korea. Papers in Regional Science, 93(4), 803–818. Hudson, R. (2001). Producing places. London: Guildford. ILO. (2015). Small and medium-sized enterprises and decent and productive employment creation. Geneva: ILO. Johnson, R., & Onwuegbuzie, A. (2004). Mixed methods research: A research paradigm whose time has come. Editorial Researcher, 33(7), 14–26. Jürgens, U., & Krzywdzinski, M. (2009). Changing east-west division of labour in the European automotive industry. European Urban and Regional Studies, 16(1), 27–42. Kapitsinis, N. (2017). Firm relocation in times of economic crisis: Evidence from Greek small and medium enterprises’ movement to Bulgaria, 2007–2014. European Planning Studies, 25(4), 703–725. Kapitsinis, N., Metaxas, T., & Duquenne, M. N. (2013). Exploring the coherence and the meaning of territorial competition. Do national states behave in the same way as firms in case of default? The case of Greece and Dubai. Applied Econometrics and International Development, 13(2), 57–72. Karagianni, S., & Labrianidis, L. (2001). The pros and cons of SMEs going international. Greek companies in Bulgaria. Eastern European Economics, 39(2), 5–28. Klagge, B., & Martin, R. (2005). Decentralized versus centralized financial systems: Is there a case for local capital markets? Journal of Economic Geography, 5(4), 387–421. Koster, S., & Kapitsinis, N. (2015). Analysing the geography of high-impact entrepreneurship. In C. Karlsson, M. Andersson, & T. Norman (Eds.), Handbook of research methods and applications in economic geography (pp. 597–613). Cheltenham: Edward Elgar. Kronenberg, K. (2012). Firm relocations in the Netherlands: Why do firms move, and where do they go? Papers in Regional Science, 92(4), 691–714. Labrianidis, L. (1997). The opening of the Greek companies to the Balkan markets might have important negative consequences in the development of the country (in Greek). The Hellenic Review of Political Sciences, 9, 65–101. Labrianidis, L., & Vogiatzis, N. (2013). Restructuring in SMEs: Greece. http://www.eurofound. europa.eu/publications/htmlfiles/ef12475.htm. Accessed April 14, 2017. Liargovas, P. (1998). The white paper on growth, competitiveness and employment and greek small and medium sized enterprises. Small Business Economics, 11(3), 201–214. Lyberaki, A., & Paraskevopoulos, C. (2002). Social capital measurement in Greece. Paper presented at OECD-ONS International Conference on Social Capital Measurement, London, 25–27 September 2002. Manjón-Antolín, M., & Arauzo-Carod, J. (2011). Locations and relocations: Determinants, modelling, and interrelations. Annals of Regional Science, 47(1), 131–146. Maskell, P., & Malmberg, A. (1999). The competitiveness of firms and regions ‘ubiquitification’ and the importance of localized learning. European Urban and Regional Studies, 6(1), 9–25.
8 The Effects of the 2007 Global Economic Crisis …
143
Miles, M., Huberman, A., & Saldaña, J. (2013). Qualitative data analysis: A methods sourcebook (3rd ed.). London: Sage. Ministry of Finance. (2017). Online database. http://www.minfin.gr/portal/en. Accessed May 19, 2017. Newton, K. (2001). Trust, social capital, civil society, and democracy. International Political Science Review, 22(2), 201–214. Pashev, K. (2011). Corruption and accession: Evidence from public procurement in Bulgaria. Public Management Review, 13(3), 409–432. Petkov, V. (2014). Advantages and disadvantages of fiscal discipline in Bulgaria in times of crisis. Contemporary Economics, 8(1), 47–56. Racoviţă, M. (2011). Europeanization and effective democracy in Romania and Bulgaria. Romanian Journal of Political Science, 11(1), 28–50. Schoenberger, E. (1991). The corporate interview as a research method in economic geography. The Professional Geographer, 43(2), 180–189. Shaiken, H. (1994). Advanced manufacturing and Mexico: A new international division of labour? Latin American Research Review, 29, 39–71. Slaveski, T., & Nedanovski, P. (2002). Foreign direct investment in the Balkans. The case of Albania, FYROM, and Bulgaria. Eastern European Economics, 40(4), 83–99. Transparency International. (2014). Corruption perceptions index 2014. http://www.transparency. org/cpi2014/press. Accessed February 22, 2017. Union of Hellenic Chambers. (2016). What pushes out firms from Greece? (in Greek). Press release in the 9th May 2016. http://www.uhc.gr/newsite/info/pdf/2016/09.05.2016.pdf. Accessed January 17, 2017. Valyrakis, S. (2008). Summary of main conclusions of the symposium. In A. Giannitsis (Ed.), Looking for a Greek developmental model (in Greek) (pp. 337–351). Athens: Papazisi. Wall Street Journal. (2015). Greek shipowners prepare to weigh anchor on prospect of higher taxes. http://www.wsj.com/articles/greek-shipowners-prepare-to-weigh-anchor-on-prospect-ofhigher-taxes-1443520314. Accessed March 3, 2017. Wilkinson, B., Gamble, J., Humphrey, J., et al. (2001). The new international division of labour in Asian electronics: Work organization and human resources in Japan and Malaysia. Journal of Management Studies, 38(5), 675–695. Yin, R. (2014). Case study research design and methods (5th ed.). London: Sage.
Nikos Kapitsinis is an early career post-doctoral researcher in the Cardiff Business School in the Wales Public Services project (UK). He was awarded his Ph.D. in Economic Geography in the School of Geographical Sciences at the University of Bristol. His research focuses on regional development, capital mobility, local taxation and spatial structures of production. He has published a number of peer-reviewed papers on regional competitiveness and firm relocation. He is an active member of the Regional Studies Association and received the commendation for the Best Young Scientist Paper at the 4th Pan-Hellenic Congress of Urbanism, Spatial Planning and Regional Development in 2017.
Chapter 9
Location of R&D Abroad—An Analysis on Global Cities Davide Castellani and Katiuscia Lavoratori
9.1
Introduction
The last decades have witnessed an increasing internationalisation of research and development (R&D) activities by multinational enterprises (MNEs), initially to developed economies and then also to emerging countries (OECD 2008; WTO 2008). This motivates the interest of researchers on studying where MNEs locate their R&D activities and identifying which factors drive their location choices. On the one hand, external factors lead firms to disperse R&D activities geographically to exploit the attractiveness of locations. In particular, firms may want to locate close to other firms (co-location between firms). On the other hand, internal factors may drive firms to co-locate their value chain activities (e.g. production and R&D) in the same location (intra-firm co-location). In order to preserve and benefit from intra-firm linkages related to the internal proximity between own activities, the firm needs to concentrate its activities in a same location. The R&D internationalisation process may present a trade-off to the firm between dispersing their activities geographically in search for the best external location factors and concentrating their activities along the value chain in a same location to preserve these internal linkages (Blanc and Sierra 1999; Mariani 2002; Alcácer and Delgado 2016). The importance of external factors to a foreign R&D location decision is rather a well-established fact, and many empirical studies reinforce this view. Instead, while
D. Castellani (&) Henley Business School, University of Reading, Reading, UK e-mail:
[email protected] K. Lavoratori University of Perugia, Perugia, Italy e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_9
145
146
D. Castellani and K. Lavoratori
the existence of intra-firm linkages between different activities of the same firm influencing the R&D location decision is not a new concept, it still lacks large-scale supporting evidence. This study investigates the determinants of the location of offshore R&D activities, analysing the role of external and internal factors as drivers of firms’ location strategy, with a multi-country and multi-sector approach. While traditionally location studies focus the analysis at a country or regional level, this study takes a finer grained unit of location analysis, the global city. This approach is in line with some recent contribution in the International Business and Economic Geography (Iammarino and McCann 2013). Indeed, cities are a particularly interesting unit of analysis, since they are a fundamental engine of economic growth and are particularly attractive locations for MNEs investments, due to their specific characteristics and their degree of global connectivity (Jacobs 1970; Sassen 1991; Goerzen et al. 2013; Castellani and Santangelo, 2016; Belderbos et al. 2017; Du et al. 2017). The reminder of the chapter is organised as follows. The next section discusses a conceptual framework about the location of MNEs innovation activities. Section 9.3 describes the data. Section 9.4 presents the econometric strategy and specification. Section 9.5 describes the empirical findings. Section 9.6 concludes.
9.2
Location of International R&D
Several factors may contribute to explaining the location patterns of R&D activities. The literature broadly distinguishes factors external to the firm and those internal to the firm (see, e.g., Rossi and Kapstinitis in this volume).
9.2.1
The Role of Factors External to the Firm
The foreign R&D location decision may be influenced by factors external to the firm and related to the endowment of a specific geographical area. Agglomeration economies are a key driver for the attraction of MNEs investments. These are the positive externalities generated from the geographical presence of different firms co-located in a given location (Alcácer and Chung 2007). Two main different types of external agglomerations economies exist (Glaeser et al. 1992). Economies associated with the co-presence of firms in the same sector, which generate intra-industry spillovers (called “localisation” or “specialisation” externalities), thanks to the knowledge flows among competitors, specialised labour created by industry demand and specialised suppliers (Marshall 1920), whereas the second type is associated with the inter-industry spillovers (“urbanisation” or “diversification” externalities) created by the co-presence of firms active in different industries, based on the idea that knowledge may arise between complementary
9 Location of R&D Abroad …
147
rather than similar activities (Jacobs 1970). A complementary classification, introduced by Duranton and Puga (2004), recognises three different sources of externalities: “sharing” includes advantages from sharing indivisible goods and facilities, suppliers, customers and risks; “matching”, a source of agglomerations may be the improved matching of labour or firms, an increasing number of agents trying to match is able to improve the quality of each match; “learning” advantages are related to the generation, diffusion and accumulation of knowledge. Studies on FDIs location decisions generally find strongly positive effects of agglomeration economies (Nielsen et al. 2017). However, when firms are heterogeneous, they may have different preferences in terms of “co-agglomeration” strategies, presenting an adverse selection. Firms may differ in the net benefits they receive from locating close to others. Shaver and Flyer (2000) show that firms possessing the “best technologies, human capital, training programmes, suppliers, or distributors benefit less from access to competitors’ technologies, human capital, training programmes, suppliers, or distributors”, whereas “less-capable” firms have a higher propensity to collocate (Alcácer 2006). At the same time, a firm leader in its industry may locate close to other leaders, as in the case of Detroit auto cluster, where the “Big Three” (Ford, General Motors and Chrysler) located their headquarters close to each other and the “stickiness” of local knowledge maintains the cluster in a central position as a global centre of innovative excellence in the automotive industry (Hannigan et al. 2015). At the same time, the activities along the value chain may present different collocation patterns. Alcácer (2006) finds that R&D activities are more likely to be collocated close to other firms because they benefit more from agglomeration economies through knowledge spillovers, rather than sales or production activities that benefit less from agglomerations. Manufacturing activities tend to be more geographically dispersed close to the final market respect to R&D activities which tend to be more concentrated. When the knowledge is locally bounded and embedded, geographical proximity reduces the cost of information and knowledge transfer and the tacit knowledge embedded in a specific area, constituting a potential attractiveness of that location (Jaffe et al. 1993; Lawson and Lorenz 1999). Other location characteristics. Most studies of foreign investment location decisions emphasise the “basic” characteristics of destination location as drivers attracting FDIs. These factors may be demand and market size, productivity and wage levels, corporate tax rates, developed formal institution, but also physical infrastructures and advanced human capital, in terms of education and skills (Nielsen et al. 2017). Foreign R&D activities are also influenced by the presence of natural advantages and external source of knowledge in a specific location. Human capital and a pool of quality R&D personnel, university and adequate education systems, research centres and public knowledge stocks, as well as the specific industrial structure may influence the firms’ location decision (Cantwell and Piscitello 2005; Siedschlag et al. 2013). Geographical distance is generally negatively associated with the location of MNE activities. This is due to the fact that some intermediate inputs need to be transferred between headquarters and subsidiaries, thus locating more distant from
148
D. Castellani and K. Lavoratori
the home country may increase transport costs, and to the fact that more geographical distant location tends also to be less familiar, this may increase sunk costs associated with MNEs’ activities. It has been argued that geographical distance is a lesser obstacle on the location of R&D activities abroad with respect to other functions (such as manufacturing plants), due to the ability of MNEs to transfer (tacit and codified) knowledge within their boundaries but across geographical space, and the emergence of highly specialised knowledge clusters in few (geographically dispersed) locations worldwide. This may drive MNEs to set up R&D labs in distant locations in order to access knowledge (Castellani et al. 2013). The role of global cities as preferred locations of MNEs activities has been recently highlighted (Goerzen et al. 2013; Castellani and Santangelo 2016; Belderbos et al. 2017; Du et al. 2017). These cities allow to reduce the liability of foreignness (Zaheer 1995), thanks to their high degree of interconnectedness to local and global markets, cosmopolitan environment and high levels of advanced producer services. Arauzo-Carod and Viladecans-Marsal (2013) find that the preferred location of high-technology activity (as R&D) is in the centre of the metropolitan area, where there is a higher presence of skilled people and other high-technology firms. Castellani and Santagelo (2016) qualify this finding by showing that R&D is relatively more attracted towards moderately global cities, which can combine connectivity, quality of (technical and scientific) infrastructure and good quality of life, which attracts (among others) young talented researchers. Manufacturing plants instead tend to be located in rural areas and small metro areas (Henderson and Ono 2008; Castellani and Santangelo 2016), in order to facilitate the access to raw materials, suppliers and distributors, but also gain access to lower wages and larger physical spaces. Given the increasing role played by global cities in the attraction of foreign FDIs, this study focuses on the location of international R&D activities in a sample of 110 global cities, following previous insights of Belderbos et al. (2016) based on a smaller subset of 57 global cities.
9.2.2
The Role of Factors Internal to the Firm
In addition to factors external to the firm, the foreign R&D location choice may be driven by the existence of internal (intra-firm) linkages which motivate firms to locate different activities along the value chain close to each other. In an intra-functional perspective, a firm decides to establish a new R&D investment in a location where it is already present with its other activities. This choice may be motivated by the aim to generate economies of scale and scope (Parr 2002; Belderbos et al. 2013). Proximity to other same-firm establishments facilitates local knowledge transfer because new establishments learn from pre-existing establishments’ experience and vice versa, with a positive influence on establishment-level productivity (Rawley and Seamans 2015). At the same time,
9 Location of R&D Abroad …
149
the co-location may be motivated by the reduction of information costs and the uncertainty to operate in a foreign market, in order to reduce the liability of foreignness (Zaheer 1995). The co-location of different investments in similar activities may also be discouraged when firms perceive the risk of redundancy. In other words, it may be too costly for a firm to set up further R&D labs in a given location. In an inter-functional perspective, the existence of internal linkages between R&D and production activities increases the need of co-location within the firm. Internal location factors, as geographical proximity to headquarters and main corporate production units, are important in the R&D location (Howell 1984). A survey conducted by Warrant (1991) shows that 35.7% of responding firms indicate “proximity to manufacturing plant” as the most important factor influencing the choice of R&D location. The literature of multinational enterprises highlights that the own R&D activity may be located abroad where other own production plants are already in place (Pearce and Singh 1992; Cantwell and Piscitello 2002), if the benefits from innovation are strongly incorporated into the production processes (Cantwell 1989). The global value chain fragmentation and the dispersion of R&D activity make it more difficult to preserve the control on core competencies and the integrity of firm knowledge, causing a “fragmentation of capabilities” (Blanc and Sierra 1999). Despite the improvement in managing communication among units thanks to the development in information and communication technologies (ICT), the complexity of technological learning and high level of tacit process knowledge maintain strong importance to face-to-face contact, as a communication technology, particularly when much of the information is not codifiable (Storper and Venable 2004). The interdependence between R&D and production is enhanced when inter-functional communication, sharing of non-standardised information, transfer of tacit knowledge and joint problem-solving are needed (Gray et al. 2015). Geographical proximity may be an alternative mechanism of coordination and control, both within and across activities. Ketokivi and Ali-Yrkkö (2009) examine the conditions under which R&D and production require physical co-location. The complexity of products and processes, the increasing rate of industry change and new product introduction (“clockspeed” industry), low maturity of production process and low degree of modularity are factors which may increase the need of proximity. Buciuni and Finotto (2016), through multi-case study analysis of six Italian firms in low-tech industries, suggest that controlling a few key development functions (such as prototype development) that require distinct manufacturing competences ensures the constant generation of innovation and maintains the control of innovative activities. The empirical studies investigating co-location between production and R&D are conducted mainly in two fields. First, some studies investigate the effect of co-location between R&D and production on firm’s performance. Gray et al. (2015) highlight that manufacturing plants co-located with R&D labs tend to have better conformance quality of products reducing the probability of obtaining an adverse inspection outcome, in pharmaceutical industry. Tecu (2013) shows that a chemical firm is more than twice as productive in R&D activities in a metropolitan area
150
D. Castellani and K. Lavoratori
where the firm is already present with production units than another firm without plants. In line with these findings, Adams and Jaffe (1996) show that the effects of parent firm R&D on plant-level productivity diminish with an increase in geographical and technological distance between R&D labs and production plants. Other studies show that firm’s existing operations have a positive effect on the establishment’s performance, in terms of the probability of survival (Shaver and Flyer 2000; Alcácer and Delgado 2016). Second, recent empirical studies on location choice investigate the extent to which R&D activities co-locate with other activities of the same firm. These studies underline that prior firm’s investment in a location increases the probability to choose that location for a new investment, both in the same function (e.g. among R&D labs) and between functions (e.g. R&D and production). In a study focused on biopharmaceutical industry in USA, Alcácer and Delgado (2016) show that the positive effect of intra-firm linkages on location varies between activities along the value chain, the need to co-locate R&D and manufacturing is greater than the need to co-locate R&D and sales, or sales and manufacturing. Another important result is that not considering the internal factors (“internal agglomeration”) tends to overestimate the effects of external ones on firm location decisions. Defever (2012), through a multi-country and multi-sector analysis, investigates the spatial organisation of value chain activities and the incentives for firms to locate their activities close to each other, analysing the location choice of non-European MNEs in 224 European regions over the period 1997–2002. He finds that prior service or production investments located in the 75-mile area around a location have a positive effect on the location decision of a new investment in that location. Belderbos et al. (2016) analyse the location choice of 1908 cross-border greenfield projects in R&D made by MNEs over the period 2003–2011 in a sample of 57 world global cities. Their results confirm the positive effect of prior core investment, in manufacturing or service activities, on the probability of “follow-up” R&D investment in the same location. Various contributions in the literature highlight that firms may be heterogeneous in terms of their need of internal proximity. Heterogeneity in the scientific content of R&D activities. The interdependence between production and R&D activities may be influenced by the scientific content of R&D investments. Some empirical studies, through a survey conducted on small samples of firms, highlight a stronger co-location between production and R&D in case of applied research, with the aim to adapt the product or to create a new product for the local market (Pearce and Singh 1992; Papanastassiou and Pearce 2009). Demand-driven R&D investments are more likely located close to the final market and the production plants. Instead, the probability to separate R&D activities increases when the research investment is basic or science-intensive, because the link with the production is relatively less important. Mariani (2002) analyses the geographical organisation in the European regions of research activities made by 799 Japanese MNEs, of which 236 were affiliates performing both R&D and production in the chosen region. She underlines that the possibility of separating R&D from production is influenced by the scientific content of research: the higher
9 Location of R&D Abroad …
151
the science-intensity of firm sector, the less important is the linkage with production. Heterogeneity across industries. In the last century, large manufacturing firms with “in-house” R&D labs were the dominant source of technical change, but the increasing standardisation of production through automation and developments in ICT reduced the costs of vertical disintegration, creating a growing modular production system (Sturgeon 2002). The complete disintegration of innovation and manufacturing has not yet been achieved. Some specialised firms need to maintain and develop technological competencies, thus the association between research and production activities remains a strategic resource (Pavitt 2003). The need for inter-functional proximity may vary across industries, because the interdependence between production and R&D could be stronger in some industries than in others. Geographical proximity is more important in industries like engineering, where the product properties are closely related to the production activities (Ivarsson et al. 2016), when the development of new technologies depends on familiarity with the product process and the innovation follows the manufacturing (Pisano and Shih 2009), and the production is an important source of ideas. A survey conducted by Florida (1997) shows that around 5% of US biotechnological firms consider manufacturing plants as a significant source of innovation, 17% in case of chemicals/materials and around 37% in automotive industries. The result of an empirical analysis conducted on 146 R&D units of 17 largest manufacturing Swedish MNEs reveals that, in engineering-intensive industries, firms are more likely to co-locate their R&D and production units in order to develop new technologies and products for the global and regional markets, not to locally adapt existing products and processes (Ivarsson et al. 2016). Determining when manufacturing is critical to innovation and when it could be located far away is difficult. Pisano and Shih (2012) introduce an industry classification based on the degree (high or low) of modularity and maturity of production process technology (“Modularity-Maturity Matrix”). In their conceptualisation, the proximity between production and R&D is not needed in sectors like consumer electronics and semiconductors, where the technology process may be more or less mature but with a high degree of modularity. Instead, in sectors with low degree of modularity, the risk of separating design and production is significant (e.g. chemicals and pharmaceuticals). These underline industrial specificities in terms of the co-location pattern. Heterogeneity in firm characteristics. The heterogeneity of preference towards proximity between production and R&D could be related also to firm characteristics. The size of the firm may have an important role in the need for co-location. Larger firms have structured and developed—organisational and managerial— capabilities, able to coordinate operations across distance and manage knowledge transfer across geographically dispersed units, while smaller firms may need co-location in order to coordinate activities due to the lack of structured organisational processes (Gray et al. 2015). Alcácer and Delgado (2016), separating small and large firms, find that intra-firm linkages matter for both subsamples, but the effect is higher for small firms. While in larger firms the innovative activity is more
152
D. Castellani and K. Lavoratori
formalised and performed into R&D labs, in smaller firms this is more informal and incremental and performed within the production plants. Thus, co-location of production and R&D may be stronger for the latter (Paci and Usai 1999). Finally, a high degree of internationalisation creates the experience for reducing the coordination problems and costs of decentralised R&D (Lall 1979; Pearce and Singh 1992). Alcácer and Delgado (2016) explore the role of geographical diversification of firms. The effects of internal linkages are higher for firms operating in few locations, geographically diversified firms have better managerial capabilities to disperse value chain activities across distant locations, reducing the potential effect of co-location.
9.3
Data
The empirical analysis relies on data on international investment projects from fDi Markets, a database produced by fDi Intelligence, a division of the Financial Times Ltd., which tracks cross-border greenfield investments across different industries and countries worldwide.1 The database contains about 143,000 investment projects referring to the period 2003–2014, in 184 different countries. For each one of these projects, fDi Markets reports information about the parent company name, home country and city, the industry and the business activity involved in the project, as well as the location of project destination (host country and city). The database contains projects in different types of business activity, such as Research and Development, manufacturing, business service, logistics, marketing and sales, headquarters, ICT, education and training, and technical support. This study focuses the analysis on 2580 new R&D greenfield investments made by 1316 MNEs over the period 2003–2014. In particular, 738 of 2580 (28.60%) identify projects in Research and Development (R&D) activity, while 1842 (71.40%) identify projects in Design, Development and Testing (DDT). It is expected that R&D-related projects represent the basic (or science-intensive) type of research, whereas DDT projects represent the applied/adaptive type of research, but a causal inspection of the description does not allow to gauge the difference between the two clearly. Therefore, R&D and DDT projects are considered jointly. Figure 9.1 shows the geographical distribution of DDT and R&D projects, by destination city. These investment projects take place in 110 global cities around the world. Exploiting the classification of Globalisation and World Cities Network (GaWC),2 it is possible to classify cities in terms of their degree of external connectedness. Based on Beaverstock et al. (1999), GaWC provides a taxonomy of around 300
1
For more details, see www.fdimarkets.com. For more details, see www.lboro.ac.uk/gawc.
2
9 Location of R&D Abroad …
153
Fig. 9.1 Geographical distribution of R&D and DDT FDIs, by destination area. Source Own compilation based on fDi Markets
cities, from the more global (Alpha and Beta) to those with intermediate degrees of global connections (Gamma), and those with sufficient signs of global connections. In this work, we follow Goerzen et al. (2013) and focus on the cities classified as Alpha and Beta.3 Table 9.1 reports the distribution of R&D/DDT investment projects and cities across geographical areas. While 50% of our sample of global cities is situated in Europe and North America, they receive only about 30% of all R&D-related projects. On the contrary, around 30 global cities in Asia (about 30% of the sample) attract around of 60% of the projects. Moreover, Table 9.2 shows the geographical distribution of R&D/DDT investment projects by origin area of MNEs, revealing that 46.32% (1195 out of 2580) comes from North America and 30.74% (793) from European countries, while around 17% from East and South Asia (448).
9.4 9.4.1
Empirical Methodology Location Choice Models
In order to provide a robust test of theoretical arguments, a location choice model is estimated, investigating the effect of external and internal factors on the location decision of new foreign investments in R&D activities. 3
Examples of global cities are: Beijing, Guangzhou, Shanghai, Shenzhen in China, Berlin, Frankfurt am Main, Munich, Stuttgart in Germany, Milan and Rome in Italy, Birmingham, Edinburgh, London, Manchester in the UK, Boston (MA), Chicago (IL), Los Angeles (CA), NYC (NY), Philadelphia (PA), San Diego (CA), San Francisco (CA) in the United States.
154
D. Castellani and K. Lavoratori
Table 9.1 Geographical distribution of R&D/DDT projects and global cities, by destination area No. of projects in R&D/DDT Freq. Percent. East Asia and Pacific 1236 EU 529 South Asia 284 North America 181 Latin America and Caribbean 146 Middle East and North Africa 93 Non-EU Europe 46 Central Asia 38 Sub-Saharan Africa 27 Total 2580 Source Own compilation based on fDi Markets
Table 9.2 Geographical distribution of R&D/DDT projects, by origin area of MNEs
47.91 20.5 11.01 7.02 5.66 3.6 1.78 1.47 1.05 100
No. of cities Freq.
Percent.
21 33 7 18 12 8 5 2 4 110
14.09 30 6.36 16.36 10.91 7.27 4.55 1.82 3.64 100
No. of projects in R&D/DDT Freq. Percent. North America 1195 EU 793 East Asia and Pacific 375 Non-EU Europe 97 South Asia 66 Middle East and North Africa 30 Latin America and Caribbean 12 Central Asia 7 Sub-Saharan Africa 5 Total 2580 Source Own compilation based on fDi Markets
46.32 30.74 14.53 3.76 2.56 1.16 0.47 0.27 0.19 100
The literature on firm location choice has mainly estimated Conditional Logit (CLM) and Nested Logit (NLM) models (Basile et al. 2009; Arauzo-Carod et al. 2010; Nielsen et al. 2017). A strong assumption of CLM (McFadden 1974) is the property of independence of irrelevant alternatives (IIA), according to which the relative odds ratio of choosing between two alternatives is independent of the characteristics of any other alternative in the choice set. Adding a new alternative or changing the characteristics of a third one cannot change the odds ratio between pairs of alternatives. This assumption may prove to be too restrictive in the location choice analysis, especially with a large location choice set as in this analysis (110 global cities).
9 Location of R&D Abroad …
155
Recent empirical studies on location choices of multinational firms introduce individual random effects estimating a Mixed Logit (MLM) model (Train 2003), in order to capture less restrictive substitution patterns between choices than the Conditional and Nested Logit models (Basile et al. 2008; Defever 2012; Belderbos et al. 2016) and to overcome the restrictive assumption of IIA, but also to use more flexible models taking into account the firms’ heterogeneity (Rasciute and Downward 2016). The random coefficients could be decomposed in their mean, the average effect common to all the firms, and their standard deviation, which captures the deviation from the average effect for each observation. A statistically significant standard deviation reveals the presence of heterogeneous preferences across firms. The CLM specification is a special case of MLM, where b is a fixed term that does not vary over firms. In this analysis, we estimate Conditional Logit regressions as the baseline model and use Mixed Logit estimates as a robustness check.
9.4.2
Model Specification
Dependent variable. The dependent variable is the location choice of a new R&D/ DDT investment project. The variable assumes value 1 if a given R&D/DDT project is located in the global city c* and value zero for all possible alternative cities c 6¼ c*. Intra-firm co-location. We introduce three variables measuring firm prior investments in a given activity and global city, and capturing the effect of the intra-firm linkages between activities as an internal driver on firm location decisions. First, we consider the cumulated investments of firm f in city c at time t − 1 in R&D/DDT activities to capture the effect of intra-firm and intra-functional co-location. Second, we consider the cumulated investments of firm f in city c at time t − 1 in production activity to capture the effect of intra-firm and inter-functional co-location, between R&D and production. Finally, we control for the investments of firm f in other activities (such as sales, marketing, headquarters, logistics and business services) in city c at time t − 1. External agglomerations. The cumulated number of foreign investments by all MNEs in the sample in city c (at time t − 1), in production, R&D and other activities, respectively, are used as measures of external agglomerations. The overall number of investments received by a city may be correlated with city characteristics and other external factors attracting firms to certain locations, so while on the one hand the effects of these variables may be overstated, we are confident that the lack of a number of city characteristics should not significantly affect the estimates of intra-firm co-location effects. Geographical distance to home city. A control for the geographical distance is included, calculating the ellipsoidal distance between the home city of the firm and the host cities. The variables definitions are reported in Table 9.3.
156
D. Castellani and K. Lavoratori
Table 9.3 Variable list and description Variables
Descriptions
Location
R&D location choice among 110 cities over the period 2003–2014; it assumes value 1 if the firm chooses the global city c; 0 otherwise Previous Firm cumulated investments in R&D/DDT, in the city c at R&D time (t − 1) Previous Firm cumulated investments in production, in the city c at prod. time (t − 1) Previous Firm cumulated investments in other activities (e.g. logistics, other sales and marketing, HQs) in the city c at time (t − 1) External agglomerations R&D Log of the cumulated number of R&D investments made by MNEs in the city c at time (t − 1) Prod. Log of the cumulated number of production investments made by MNEs in the city c at time (t − 1) Other Log of the cumulated number of investments made by MNEs in the city c at time (t − 1), in other activities Distance Geographical distance between firm home city and the potential host cities (in log)
9.5
Source
Type
fDi Markets
Firm-city
fDi Markets fDi Markets fDi Markets
Firm-city
fDi Markets fDi Markets fDi Markets fDi Markets
Firm-city Firm-city
City City City Firm-city
Results
This section presents the results of Conditional Logit model (CLM) and Mixed Logit model (MLM) regressions, which estimate the effect of external and internal factors on new R&D investment location choices made by multinational firms. Table 9.4 reports the results of two different specifications. Mod. 1 controls for prior investments in R&D and production at the firm-city level (measuring the role of intra-firm linkages) and both for prior investments in production and R&D activities at the overall city level (measuring the importance of external agglomerations). Mod. 2 also controls for prior investments in all other activities included in the fDi Market database (e.g. HQs, sales and marketing, logistics). In both specifications, the control for the geographical distance between home and destination city is included. Results show that the external agglomeration economies have a positive effect on R&D location choices. R&D tends to collocate close to other firms, supporting the idea that R&D may benefit more from agglomeration economies compared to other activities, such as manufacturing and sales (Alcácer 2006) and that this agglomeration force is higher in case of other R&D-related activities (“External Agglomeration-R&D”). Results also underline a positive effect of distance between home and the destination city, suggesting that in the case of R&D-related activities, the geographical
9 Location of R&D Abroad …
157
distance is not an obstacle, in order to acquire specific knowledge located far away from the home country (Castellani et al. 2013). Controlling for the other activities, results highlight a reduction in the role of external factors. Findings are consistent with the idea that Mod. 1 is not adequately capturing external factors. The overall amount of investments that a city attracts is probably correlated with a number of city characteristics that make them attractive for MNEs location. While there may be a number of additional city characteristics which may affect the location of R&D, we are confident that the specification in Mod. 2 can account for a large proportion of the variance explained by city characteristics. These first results confirm the positive role of external factors on R&D location choice. Furthermore, a firm’s prior investment in a global city has a positive and significant effect on the probability that an MNE will locate a new R&D investment in that city, both in an intra-functional and inter-functional perspective. This is consistent with the idea that internal linkages matter and they increase the probability of co-location, with a higher effect in case of co-location between production and R&D activities.4 This suggests that the argument of uncertainty reduction, economies of scale and scope and knowledge transfers across activities dominates over the redundancy argument. Interestingly, co-location with other business activities does not seem to occur to the same extent, and having previous investments in other business activities does not have a strong effect on the location of R&D. In the third column (Mod. 3), the model is estimated considering the R&D investments made during the period 2008–2014. Our data measure flows of investments, and it is not possible to know whether a firm had R&D or manufacturing activities in a given global city before 2003. In order to partially tackle this problem, we allow flows of investments to cumulate during the period 2003–2007, so that this measure should be better correlated with measures of the stock of MNE presence in a given city. The main results are confirmed. Finally, MLM is estimated in order to control for the IIA assumption. Results are reported in the last column (Mod. 4) of Table 9.4. The independent variables of intra-firm co-location are included as random parameters. The mean coefficients and the standard deviations are reported. Also in this case, the main results are confirmed. Indeed, our findings underline that even if the location choice set is relatively large, since it is composed by relatively homogeneous locations (global cities), the Conditional Logit model does not produce biased estimates and the IIA property is respected.
4
This conclusion is also confirmed by the coefficients calculated as odds ratio, which facilitate comparisons, available under request to the authors.
158
D. Castellani and K. Lavoratori
Table 9.4 Results of conditional and mixed logit models Dependent variable: foreign R&D location choice Mod. 1 Mod. 2 Mod. 3 Mod. 4 Mean Internal factors Previous R&D (firm-city) Previous prod (firm-city)
0.0948*** (0.0308) 0.2064*** (0.0612)
0.0813*** (0.0258) 0.1820*** (0.0650) 0.1306 (0.0850)
0.0851*** (0.0263) 0.1562*** (0.0601) 0.1514* (0.0828)
0.1132*** (0.0362) 0.2143*** (0.0719) 0.056 (0.0639)
0.7293*** (0.0281) 0.1232*** (0.0253)
0.6200*** (0.0353) 0.0620** (0.0264) 0.2781*** (0.0351) 0.0634*** (0.0217)
0.6235*** (0.0394) 0.0235 (0.0312) 0.2397*** (0.0417) 0.0975*** (0.0281)
0.6183*** (0.0301) 0.0608** (0.0246) 0.2776*** (0.0340) 0.0650*** (0.0219)
Previous other (firm-city) External factors External agglomerations R&D (log) (city) Prod. (log) (city) Other (log) (city) Distance (log) Standard deviation Previous R&D (firm-city) Previous prod (firm-city) Previous other (firm-city)
0.0454** (0.0212)
−0.0834 (0.1618) −0.5309 (0.4242) 0.7142*** (0.1240) 231,785 110 1316 2003–2014
No. obs 231,785 231,785 175,575 No. cities 110 110 110 No. MNEs 1316 1316 975 Period 2003–2014 2003–2014 2008–2014 Pseudo-R2 0.1556 0.1592 0.1499 Simulated log-L MXL −9667.103 Conditional vs. Mixed model CLM CLM CLM MLM Note The dependent variable is equal to 1 if firm f is set in city c and zero for all city different from c. Standard error in parenthesis. Asterisks denote confidence levels: *p < 0.10, **p < 0.05 and ***p < 0.01 Source Own research
9 Location of R&D Abroad …
9.6
159
Conclusion
This study examines the determinants of the R&D internationalisation process. In particular, it investigates the role of external and internal factors in the firm location decisions of R&D activities abroad, looking at intra-functional and inter-functional interdependences across activities along the value chain, especially between R&D and production. On the one hand, external factors (e.g. external agglomeration economies) lead firms to disperse their activities geographically in search for the best location (we observe co-location between firms in the same location). On the other hand, internal factors drive firms to concentrate their activities along the value chain close to each other in a same location, to preserve intra-firm linkages (thus leading to co-location of different activities within the same firm). The empirical literature on MNEs is focused mainly on the external factors driving the location decision of multinational enterprises’ R&D activities. Although the existence of internal linkages, which require co-location, is not new in the literature, large-scale empirical evidence is still been lacking. This study contributes to filling this gap. Recent empirical studies attribute a crucial role to global cities as preferred locations for MNEs activities thanks to their degree of global interconnectedness, a cosmopolitan environment and for the presence of high levels of advanced producer services, especially in the attraction of R&D-related activities (Goerzen et al. 2013; Castellani and Santangelo 2016; Belderbos et al. 2016). Following these insights, this study investigates the determinants of location decisions, using data on R&D foreign direct investments in 110 global cities worldwide. The results confirm the positive role of external agglomeration economies. Additionally, the findings suggest that previous R&D and production activities of the same MNEs increase the probability to locate R&D in a given global city. The effect of co-location is higher with production activity, confirming the role of intra-firm linkages between R&D and production. This study has several limitations. First, the analysis exploits an imprecise geographical boundary, the city. In further investigations, more well-defined administrative boundaries (e.g. metropolitan area or functional urban area) could be adopted. Second, there is clearly a trade-off between considering a comprehensive set of fine-grained locations and gathering a rich set of characteristics for these locations. This study has chosen the former route, and some of the city characteristics used should be able to capture a significant portion of the variance, but it is not possible to exclude that some omitted variables may affect the results. Third, by considering co-location at a very geographically disaggregated level of analysis, some actions going on at a slightly more aggregated level could be missed. A disaggregated unit of analysis should make it more difficult to detect co-location. This makes the results like a lower bound. For example, R&D may be located closer to city centres, while production may take place in the outskirts. In order to address this issue, further investigations could move to the metropolitan area level.
160
D. Castellani and K. Lavoratori
References Adams, J., & Jaffe, A. B. (1996). Bounding the effects of R&D: An investigation using matched establishment-data. The Rand Journal of Economics, 27(4), 700–721. Alcácer, J. (2006). Location choices across the value chain: how activity and capability influence collocation. Management Science, 52(10), 1457–1471. Alcácer, J., & Chung, W. (2007). Location strategies and knowledge spillovers. Management Science, 32(10), 1231–1241. Alcácer, J., & Delgado, M. (2016). Spatial organization of firms and location choices through the value chain. Management Science, published online in articles in advance, January 27, 2016. http://dx.doi.org/10.1287/mnsc.2015.2308. Arauzo-Carod, J. M., Liviano-Solis, D., & Manjón-Antolín, M. (2010). Empirical studies in industrial location: An assessment of their methods and results. Journal of Regional Science, 50(3), 685–711. Arauzo-Carod, J. M., & Viladecans-Marsal, E. (2013). Industrial location at the intra-metropolitan level: The role of agglomeration economies. Regional Studies, 43(4), 545–558. Basile, R., Castellani, D., & Zanfei, A. (2008). Location choices of multinational firms in Europe: The role of EU cohesion policy. Journal of International Economics, 74, 328–340. Basile, R., Castellani, D., & Zanfei, A. (2009). National boundaries and the location of multinational firms in Europe. Regional Science, 88(4), 733–748. Beaverstock, J. V., Smith, R. G., & Taylor, P. J. (1999). A roster of world cities. Cities, 16(6), 445–458. Belderbos, R., Du, H. S., & Goerzen, A. (2017). Global cities, connectivity, and the location choice of MNC regional headquarters. Journal of Management Studies, 54, 1271–1302. https:// doi.org/10.1111/joms.12290. Belderbos, R., Leten, B., & Suzuki, S. (2013). How global is R&D? Firm-level determinants of home-country bias in R&D. Journal of International Business Studies, 44(8), 765–786. Belderbos, R., Sleuwaegen, L., Somers, D., & De Backer, K. (2016). Where do locate innovative activities in global value chain. Does co-location matter? OECD Science, Technology and Industry Policy Papers, no 30. Paris: OECD Publishing. http://dx.doi.org/10.1787/ 5jlv8zmp86jg-en. Blanc, H., & Sierra, C. (1999). The internationalisation of R&D by multinationals: A trade-off between external and internal proximity. Cambridge Journal of Economics, 23(2), 187–206. Buciuni, G., & Finotto, V. (2016). Innovation in global value chains: Collocation of production and development in Italian low-tech industries. Regional Studies. https://doi.org/10.1080/ 00343404.2015.1115010. Cantwell, J. A. (1989). Technological innovation and multinational corporations. Oxford: Blackwell. Cantwell, J. A., & Piscitello, L. (2002). The location of technological activities of MNCs in European regions: The role of spillovers and local competencies. Journal of International Management, 8(1), 69–96. Cantwell, J. A., & Piscitello, L. (2005). Recent location of foreign-owned research and development activities by large multinational corporations in the European regions: The role of spillovers and externalities. Regional Studies, 39(1), 1–16. https://doi.org/10.1080/ 0034340052000320824. Castellani, D., Jimenez, A., & Zanfei, A. (2013). How remote are R&D labs? Distance factors and international innovative activities. Journal of International Business Studies, 44(7), 649–675. Castellani, D., & Santangelo, G. (2016). Quo vadis? Cities and the location of cross-border activities. In 42nd Annual EIBA Conference, Liabilities of Foreignness vs. the Value of Diversity. Defever, F. (2012). The spatial organization of multinational firms. Canadian Journal of Economics/Revue canadienne d’économique, 45(2), 672–697.
9 Location of R&D Abroad …
161
Du, H., Belderbos, R., & Slangen (2017) Liability of foreignness and multinational firms’ investments location across the value chain: Global cities versus local cities, mimeo. Duranton, G., & Puga, D. (2004). Micro-foundations of urban agglomeration economies. In J. V. Henderson & J. F. Thisse (Eds.), Handbook of regional and urban economics (1th ed., Vol. 4, pp. 2063–2117, Chapter 48). Elsevier. Florida, R. (1997). The globalization of R&D: Results of a survey of foreign-affiliated R&D laboratories in the USA. Research Policy, 26, 85–103. Glaeser, E., Kallal, H. D., Scheinkman, J. A., & Shleifer, A. (1992). Growth of cities. Journal of Political Economy, 100(6), 1126–1152. Goerzen, A., Asmussen, C. G., & Nielsen, B. (2013). Global cities and multinational enterprise location strategy. Journal of International Business Studies, 44(5), 427–450. Gray, J. V., Siemsen, E., & Vasudeva, G. (2015). Colocation still matters: Conformance quality and the interdependence of R&D and manufacturing in the pharmaceutical industry. Management Science, 61(11), 2760–2781. https://doi.org/10.1287/mnsc.2014.2104. Hannigan, T. J., Cano-Kollmann, M., & Mudambi, R. (2015). Thriving innovation amidst manufacturing decline: The Detroit auto cluster and the resilience of local knowledge production. Industrial and Corporate Change, 24(3), 613–634. Henderson, V., & Ono, Y. (2008). Where do manufacturing firms locate their headquarters? Journal of Urban Economics, 63, 431–450. Howells, J. (1984). The location of research and development: Come observations and evidence from Britain. Regional Studies, 18(1), 13–29. Iammarino, S., & McCann, P. (2013). Multinationals and economic geography. Location, technology and innovation. Edward Elgar Publishing. Ivarsson, I., Alvstam, G., & Vahlne, J. E. (2016). Global technology development by colocating R&D and manufacturing: The case of Swedish manufacturing MNEs. Industrial and Corporate Change. Advance Access published May 13, 2016. Jacobs, J. (1970). The economy of cities. New York, NY: Vintage. Jaffe, A., Trajtenberg, M., & Henderson, R. (1993). Geographic localisation of knowledge spillovers as evidenced by patent citations. Quarterly Journal of Economics, 108, 577–598. Ketokivi, M., & Ali-Yrkkö, J. (2009). Unbundling R&D and manufacturing: Postindustrial myth or economic reality? Review of Policy Research, 26(1–2), 35–54. Lall, S. (1979). The international allocation of research activity by US multinationals. Oxford Bulletin of Economics and Statistics, 41(4), 313–331. Lawson, C., & Lorenz, E. (1999). Collective learning, tacit knowledge and regional innovative capacity. Regional Studies, 33, 305–317. Mariani, M. (2002). Next to production or to technological clusters? The economics and management of R&D location. Journal of Management and Governance, 6(2), 131–152. Marshall, A. (1920). Principles of Economics. London: McMillan. McFadden, D. (1974). Conditional logit analysis of qualitative choice behaviour. In: P. Zarembka (Ed.), Frontiers in econometrics (pp. 105–142, Chapter 4). New York: Academic Press. Nielsen, B. B., Asmussen, C. G., & Weatherall, C. D. (2017). The location choice of foreign direct investments: Empirical evidence and methodological challenges. Journal of World Business, 52, 62–82. OECD. (2008). The internationalisation of business R&D: Evidence, impacts and implications. Paris, France: OECD. Paci, R., & Usai, S. (1999). Externalities, knowledge spillovers and the spatial distribution of innovation. GeoJournal, 49, 381–390. Papanastassiou, M., & Pearce, R. (2009). The strategic development of multinationals: Subsidiaries and innovation. Springer. Parr, J. B. (2002). Agglomeration economies: Ambiguities and confusions. Environment and Planning A, 34(4), 717–731. Pavitt, K. (2003). Specialization and system integration: Where manufacture and service still meet. In A. Prencipe, A. Davies, & M. Hobday (Eds.), The business of system integration. Oxford: Oxford University Press.
162
D. Castellani and K. Lavoratori
Pearce, R., & Singh, S. (1992). Globalizing research and development. London: Macmillan. Pisano, G. P., & Shih, W. C. (2009). Restoring American competitiveness. Harvard Business Review, 87(7–8). Pisano, G. P., & Shih, W. C. (2012). Does America really need manufacturing? Harvard Business Review, 90(3), 94–102. Rasciute, S., & Downward, P. (2016). Explaining variability in the investment location choices of MNEs: An exploration of country, industry and firm effects. International Business Review. Rawley, E., & Seamans, R. (2015). Intra-firm spillovers? The stock and flow effects of collocation. Columbia Business School Research Paper no 15–2. Available at SSRN: https://ssrn.com/ abstract=2544518 or http://dx.doi.org/10.2139/ssrn.2544518. Sassen, S. (1991). The global city: London, New York, Tokyo. Princeton UP: Princeton. Shaver, J. M., & Flyer, F. (2000). Agglomeration economies, firm heterogeneity, and foreign direct investment in the United States. Strategic Management Journal, 21(12), 1175–1193. Siedschlag, I., Smith, D., Turcue, C., & Zhang, X. (2013). What determines the location choice of R&D activities by multinational firms? Research Policy, 42(8), 1420–1430. Storper, M., & Venables, A. J. (2004). Buzz: face-to-face contact and the urban economy. Journal of Economic Geography, 4(4). Sturgeon, T. (2002). Modular production networks: A new American model of industrial organization. Industrial and Corporate Change, 11(3), 451–496. Tecu, I. (2013). The location of industrial innovation: Does manufacturing matter? US Census Bureau Center for Economic Studies Paper No. CES-WP-13-09. Available at SSRN: https:// ssrn.com/abstract=2233366 or http://dx.doi.org/10.2139/ssrn.2233366. Train, K. E. (2003). Discrete choice methods with simulation. Cambridge: Cambridge University Press. Warrant, F. (1991). Le deploiement mondial de la R&D industrielle: Facteur et garant de la globalisation de la technologie et de l’economie. Bruxelles. Commission of the European Communities, Fast, 4. WTO. (2008). Trade, the location of production and the industrial organization of firms, World Trade Report 2008—Trade in a globalizing World. Geneva, Switzerland: World Trade Organization (WTO). Zaheer, S. (1995). Overcoming the liability of foreignness. Academy of Management Journal, 38(2): 341–363.
Davide Castellani is a Professor of International Business and Strategy at the Henley Business School, University of Reading, UK. He holds a Ph.D. in Economics from the University of Ancona (Italy), and his research focuses on the determinants of the firms’ internationalisation choices, and their impact on innovation, technology and economic performances of firms, regions and countries. He has published in top-ranked international journals, such as Journal of International Business Studies, Journal of International Economics, Regional Studies, Research Policy, Review of World Economics and Small Business Economics. Katiuscia Lavoratori is a Teaching and Research Fellow at the Henley Business School, University of Reading, UK. She holds a Ph.D. in Economics from the University of Perugia (Italy), and her research lies at the intersection of international business and economic geography, with special reference to the economic analysis of location choices, the relation between agglomeration and productivity, industry 4.0 and international business activities.
Chapter 10
Delocalisation of Enterprises: Qualitative and Quantitative Effects on the Labour Market Ewa Małuszyńska
10.1
Introduction
Impelled by the ever more rapid changes on a global, national, regional and local markets, enterprises are under constant pressure to adjust and adapt to changing conditions and circumstances. This process, commonly referred to as corporate restructuring, plays out across many dimensions, including the geographical, which involves a change in preferred locations, ultimately prompting companies to move to other locations. Any type of restructuring may result in the loss of some or all jobs. One particularly sensitive issue is having entire or parts of companies move to another country ravaging the local labour force in the process. The problem was discussed mainly in the so-called old EU member states during its historically biggest enlargement round in 2004 (Radło 2013). Fears and anxieties over jobs which companies, and especially large transnationals, transferred to the “new” EU member states became commonplace in surveys and statements presented in economic, social and political forums. In this context, a number of terms, such as the migration, relocation and delocalisation of enterprises as well as offshoring and outsourcing, are used without offering their definitions or clarifying differences among them. Many of the terms are treated as interchangeable. Studies on job loss to enterprise relocation vary also in their temporal and spatial scopes. They additionally differ in the methods employed for collecting information, which make comparison across studies fairly difficult. For these reasons, it is vital to address the following three questions: 1. What is corporate delocalisation and how does it relate to enterprise relocations, offshoring and outsourcing?
E. Małuszyńska (&) Poznań University of Economics and Business, Poznań, Poland e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_10
163
164
E. Małuszyńska
2. What methods are available for estimating the impact of delocalisation abroad on employment? 3. How does delocalisation impact upon the labour market in the countries from which businesses are moving to other locations?
10.2
Delocalisation in Enterprise Restructuring
Business delocalisation and related outsourcing and offshoring tend to take place within the framework of corporate restructuring. They are triggered by economic, social and political transformations taking place in close and more distant business environments. All of the numerous definitions of restructuring agree that the phenomenon involves “a dramatic modification of enterprises aimed at improving the efficiency and effectiveness of their operations and ensuring their growth in step with changes in the external environment and in response to the internal needs of companies” (Rojek 2016, p. 140). A broader definition has been furnished by Singh who postulated that “restructuring is a series of rapid changes in a company’s asset, physical and organisational structures brought about by a range of transactions that result in the disposal of parts of an enterprise or the launch of new activities, changes in the equity structure caused by the incurrence of debt, changes in ownership and/or changes in an enterprise’s inner organisation” (Singh 1993 in Romanowska 2011, p. 208) All definitions emphasise that the process is complex and lengthy. Restructuring, which may be either objective or subjective, is an attempt to repair or enhance companies. The objective type of restructuring focuses on the technical and technological aspects, whereas subjective restructuring extends to legal, organisational, financial and economic improvements (Mozalewski 2010). Other authors distinguish between reparatory and developmental restructuring (Suszyński 1999). Reparatory restructuring concerns companies which are threatened with liquidation, mostly due to a bad situation of the company. It aims to improve the adverse economic conditions and restore the solvency of the company. Developmental restructuring is generally carried out in companies with good economic condition in order to increase the efficiency and effectiveness of the company and increase its competitiveness. It applies to activities of an innovative and developmental nature. Both may entail operational, financial and ownership modifications (Mozalewski 2010). Within its broad scope, operational modifications embody changes in both company operations and related resource base. One type of change undertaken to revamp a company’s labour force is outsourcing, a business model described in inter-enterprise links theory that involves relocating individual parts of value chains. A different approach to classifying corporate restructuring was adopted by the European Monitoring Centre on Change (EMCC) for the purpose of monitoring the labour market changes precipitated by restructuring. The following types of organisational restructuring have been identified (Eurofound 2013):
10
Delocalisation of Enterprises …
165
– Internal restructuring: a case of a company planning to cut jobs in a manner not linked to any of the other types of restructuring defined below. – Closure: a case of a company or an industrial site shutting down for economic reasons not directly connected to either relocation or outsourcing. – Bankruptcy: a case of a company going bankrupt for economic reasons not directly connected to either relocation or outsourcing. – Relocation: when the activity stays in the same company, but is relocated within the same country. – Outsourcing: a case of activities being subcontracted to another company in the same country. – Offshoring/delocalisation: when the activity is relocated or outsourced outside the country’s borders. – Merger/acquisition: a case of two companies merging or of an acquisition involving an internal restructuring programme aimed at streamlining an organisation by means of job cuts. – Business expansion: a case of a company extending its business activities and hiring new workers. The theoretical basis for the relocation along with the restructuring of enterprises are classic location theories, institutional and behavioural theory and corporate strategies (Brouwer et al. 2004; Mariotti 2005; Pellenbarg et al. 2002). Relocation is most often treated as a specific form of location in which one location is replaced by another (substitution). Relocation outside the country (analysed in this chapter) also appears in the theories of internationalisation of enterprises. These theories focus on the factors of a specific location of the business and the motives behind making decisions on changing it, such as growth of effectiveness and competitiveness of enterprises. The theoretical debate on impacts of delocalisation on employment shows that “in the short term, there are risks of job looses, but the gains from offshoring could lead to the creation of new jobs” (OECD 2007a, p. 35). The decline in employment occurs especially among people with lower qualifications while increased demand is for better qualified workers.
10.2.1 Outsourcing, Offshoring, Relocation and Delocalisation A major challenge, addressed differently by different authors, is to formulate a clear definition of such processes which are akin to restructuring as offshoring, outsourcing, relocation and delocalisation, and to capture the relations between them. The term outsourcing is at times used interchangeably with delocalisation and offshoring (Blinder 2005; Fessler 2006; Whalley 2006). In the prevailing approach, the notions of offshoring and outsourcing apply directly to manufacturing sites and to externalising production. Both processes are also defined as having to do with
166
E. Małuszyńska
trade, determining the directions of trade flows (enterprise supplies) (Blinder 2007; Sako 2005). The above shows that the knowledge of both concepts is needed to obtain complete information on whether a move is domestic or foreign and whether activities are being internalised or externalised. Based on the ownership of inputs as the primary criterion, followed by that of location (either at home or abroad), the terms provided in the literature as well as their synonyms can be classified as follows (Table 10.1). The common usage of outsourcing and offshoring is limited to foreign transfers of individual modules of the value chain. Under this view, offshoring is defined as the moving of activities abroad while retaining the ownership of means of production, whereas outsourcing refers to entrusting the performance of some of company activities to an independent enterprise in another country. The definition of delocalisation presents a considerably lesser challenge. Its prevailing and simplest definition perceives it as a transfer of a part of enterprise activities to another location. Other more precise definitions stress that delocalisation, also referred to as delocalisation sensu stricto, occurs when the transfer of business is associated with a complete or partial loss of the jobs that exist in the enterprise’s original location (OECD 2007a; The EU 2005). The definition provided by the European Commission also presents job destruction as an integral and most essential attribute of delocalisation (European 2006). Despite these seemingly satisfactory and unambiguous definitions of outsourcing, offshoring and delocalisation, references to the relationships between them reveal a number of significant differences including: 1. As integral parts of the process of breaking up production, the notions of offshoring and outsourcing are considered to be forms of delocalisation (Gierańczyk 2008; Radło 2005; ManVis Report 2005; Zorska 2008). 2. Offshoring and outsourcing are viewed as another form of the relocation of business activities other than delocalisation (Gierańczyk 2008; Ciesielska 2009). 3. As “partial or complete transfers of business activities in a form that is invisible or physical, i.e. in the form of capital flows or business activities (goods and services), that are either domestic or international”, relocations may take the form of either delocalisation or outsourcing (Gierańczyk 2008, p. 91). Here, delocalisation is defined as “a process that lastingly changes in locations of business activities, including the relocation of existing business facilities, the moving of activities to alternative locations, the closures of production in full or in part and its resumption at an alternative location” (Gierańczyk 2008, p. 92). Considering the above, I nevertheless uphold the view that: – delocalisation is a form of moving activities in which, as a prerequisite, either some or all jobs are lost. It is therefore a form of insourcing and outsourcing abroad, as well as that of domestic insourcing and outsourcing (see Table 10.1). This view is consistent with the suggestion that “delocalisation decisions form a pure variant of a broader outsourcing paradigm” and that outsourcing may occur
10
Delocalisation of Enterprises …
167
Table 10.1 Classification of sourcing Sourcing Ownership of inputs retained Insourcing/in-house sourcing Insourcing abroad – Offshoring – Offshoring defined narrowly – Cross-border outsourcing to own affiliates – Captive offshoring – Offshore outsourcing – Intra-firm offshoring – Offshoring in the strict sense –Offshoring/delocalisationa
Domestic insourcing – Relocationa
No ownership of inputs Outsourcing Outsourcing abroad Domestic outsourcing – Outsourcing – Outsourcing defined broadly – Offshoring defined broadly – International subcontracting – Offsourcing – Cross-border outsourcing – Outsourcing to non-affiliates – Subcontracting abroad –Offshoring/ delocalisationa
– Onshoring – Outsourcinga
Source Own compilation a Terms employed by EMCC
without an accompanying delocalisation (Mazzanti et al. 2011, p. 420), as well as with the term offshoring (inclusive of delocalisation) (OECD 2007b). The number of jobs lost through relocation is an inherent attribute as well as a measure of delocalisation. – relocation is synonymous with moving one’s business.
10.3
Measurement Methods
Measurement of actual relocations of entire enterprises or individual modules of their value chain and related employment changes, due to the difficulty of direct observation of this process, is not easy. Particular problems are faced in researching whole countries which are home to a great number of companies. Other than of direct observation, indirect methods are used. They rely on more easily accessible data on the relocations of activities and their effects. The information secured using such methods extends to changes in employment associated with trade flows and the relocation propensity of specific sectors (Odrobina 2009; Radło 2008; The EU 2005; Mankiw and Swagel 2006; Hatzichronoglou 2005).
168
E. Małuszyńska
10.3.1 Direct Methods Direct research relies on information sourced from enterprises, concerning either the past relocations of entire enterprise or parts of the value chains or such relocations envisioned in their strategies. In both cases, the biggest challenge is to secure representative samples of enterprises. An added difficulty that appears in the latter case is uncertainty about the degree to which a given company follows through on its strategy and, specifically, completes its announced delocalisation (The EU 2005). Direct research may well be facilitated by dedicated databases. However, such databases are relatively rare and the information they contain is often limited to relatively short periods, thus preventing the identification of trends in the investigated processes. One such repository was created in the European Monitoring Centre on Change established in 2001, based on publically available information on announced delocalisations made both domestically and internationally in the EU member states and Norway. As the research by the European Monitoring Centre on Change is limited to cases in which at least 100 jobs are lost and—in companies employing at least 250 workers—to cases where job cuts account for at least 10% of the workforce, the results may well be underestimated. However, research without such restrictions conducted in other countries supports the assertion that the extent of job destruction and creation as a result of relocation in Europe is fairly limited. As in other similar projects, the EMCC database comes with weaknesses which include a lack of representativeness of the sample and emphasis on medium and large enterprises which are consequently mainly industrial. Another issue is the limited number of sources used in certain countries. The advantages of the EMCC database are its timeliness, its identification of individual cases of restructuring based on publicly available information and its uniqueness as an EU-wide dataset of larger-scale restructuring events (EMCC 2016). Despite the above drawbacks of the EMCC database, “the picture of restructuring that emerges from this dataset is largely consistent with data coming from more representative sources such as the European Labour Force Survey” (Eurofound 2013, p. 7).
10.3.2 Indirect Methods In the cases where information on the relocation of enterprises and their parts is not directly available, an indirect measure of the impact of delocalisation on employment can be used. Such measures are linked to trade. A number of authors have taken keen notice of the links which exist between the locations of business and foreign trade. The best-known conclusions with this regard have been supplied by Ohlin, who stated that “international trade theory is nothing other than a science of international locations” (Ohlin in Jovanović 2003, p. 61). His assertion was
10
Delocalisation of Enterprises …
169
supported by Isard who noted that “locations cannot be explained without at the same time accounting for trade and trade cannot be explained without the simultaneous determination of locations (…) trade and locations are two sides of the same coin” (Isard in Jovanović 2003, p. 61). The relocation of production of indirect goods allows the replacements of domestic with foreign investment. As a result, the latter is treated as a “reflection” of the volume of relocations (The EU 2005). Where domestic and foreign investment cannot be estimated in separation, the measure of foreign investment is replaced with that of the volume of any such import or investment which does not entail electricity consumption. One of the methods relies on estimating the employment equivalent of import (OECD 2007a). The authors say: “When it is possible to quantify imports resulting from offshoring and the decline in the value of exports due to offshoring, it is important to relate these changes to corresponding employment. As regards imports, the equivalent in direct jobs implied by these import is obtained by formula: Lm it ¼
X t
Mit
Lit Yit
where: Lm is employment equivalent of imports in sector i for year t it Mit the flow of imports in sector i for year t Yit the output by value in sector i for year t In other words, this means calculating the number of jobs necessary if the imported goods and services were produced domestically. This equation concerns only direct jobs” (OECD 2007a, p. 49). An alternative method of estimating the impact of enterprise relocation abroad on employment is to calculate the demand for labour. The assumption made here is that once an enterprise becomes more competitive by relocating some of its activities abroad, new jobs are created. Information on the volume of inter- and intra-industry flows and on investment not involving electricity consumption is used to construct outsourcing index proposed by Feenstra and Hanson (Ekholm and Hakkala 2006; Hatzichronglou 2005; Olsen 2006; OECD 2007a). The proposed formula is as follows1: OIi ¼
Xpurchases of inputs j excluding energy by industry i Mj : total inputs excluding energy used by i Dj i
where
1
The measure of foreign investment is replaced with that of the volume of any such import or investment that does not entail electricity consumption, where domestic and foreign investment cannot be estimated in separation.
170
E. Małuszyńska
Mj import of goods or services j Dj domestic demand for goods or services j where Dj ¼ Yj Xj þ Mj including: Yj production of goods or services j Xj export of goods or services j Outsourcing index can be added into standard regression equation reflecting changed demand for labour (OECD 2007a, p. 51). DlnLit ¼ a0 þ a1 DlnWit þ a2 DlnOlsit þ a3 DlnOlgit þ bDlnxit þ cDlnYit þ dDt þ eit The term D corresponds to the fixed effects for year t which are common through all industries, such as changes concerning the cost of capital. From these, it could be observed: – the growth in wages W and faster growth in outsourcing Olsit iOlgit indices would have a negative effect on employment. – the growth in the price of other inputs xit would encourage firms to replace these inputs with labour. – the growth in production Yit would have a favourable effect on employment. The approach makes it possible to answer the question of whether import, viewed here as a measure of a delocalisation achieved by means of outsourcing, is complementary or substitutive to employment (Falk and Wolfmayr 2005).
10.3.3 Enterprises’ propensity to delocalise A particular method of estimating the extent of delocalisation on the basis of the potential number of jobs lost involves estimating the delocalisation-proneness of sectors or industries. In applying the method, the number of jobs at risk of being lost is estimated in view of the potential delocalisation expected to result from the existence in a given area of sectors and industries having a specific propensity to delocalise. A job is classified as delocalisation-prone if its nature, i.e. what and where needs to be done, allows one to make a transfer to another location at home or abroad, even if no such a transfer is made at a given time. The delocalisation-proneness of jobs in a given sector depends on the extent to which the features of a typical enterprise in a specific industry or sector influence the ease of choosing a new location and thus its propensity to move within either the short or
10
Delocalisation of Enterprises …
171
the medium term. The factors considered are an enterprise’s ties with natural resources, suppliers and a specific customer base; the amount of capital investment; the geographical concentration of enterprises; the reliance on information technology; the ability to transfer activity results by means of advanced ICT; the proportion of codified knowledge contained in products and services; the need to continue to liaise directly with the buyers (Blinder 2007; Blinder and Kruger 2009; Hatzichronoglou 2005; OECD 2007b; Welsum and van Reif 2006) Other criteria include the differentiation of wages across specific job types, low barriers to market entry in relevant industries and limited social benefits required in host countries. The jobs (occupations, functions) at the greatest risk of being relocated and those associated with the greatest willingness to move are the jobs which may well be performed elsewhere without jeopardising compliance with the above criteria beyond a specified tolerance margin. Therefore, the most delocalisation-prone industries are those which: – maintain very weak ties with natural resources, suppliers and a specific customer base; – require only limited capital outlays; – have a low geographical concentration; – are to a great extent reliant on the latest ITC technologies and codified knowledge; – do not require direct contact with the buyers of goods and/or services. Based on the last four of the above-mentioned criteria, 11% of the global private service sector employment was found to be at high risk of being delocalised within the next 5 years. The remaining enterprises (89%) will most likely rely on a different growth strategy if and when needed (Offshoring 2009; OECD 2007b; Relocation 2005). The delocalisation-proneness of individual industries and occupations is not constant. Such variability results from changes in the way specific functions are performed and can be attributed largely to ever faster technological advances. A case in point are educational services which require prior face-to-face contact between the provider and the customer, and which are currently rendered by means of e-learning, etc. An analysis of changes in the rankings of jobs potentially at risk of delocalisation in the EU15, the USA, Canada and Australia points to a definite positive correlation with reliance on advanced technologies (Blinder 2007; Welsum and van Reif 2006). Other than potentially susceptible jobs, there are also jobs which, due to their nature and the impact of additional external factors, are bound to be moved and, as such, can be considered as lost in their original locations. The difference between the number of jobs potentially relocated and the number of jobs that are likely to be destroyed depends on “cost pressures, competition for resources, size of the firm and the legal, regulatory, social and political environments in the receiving country”(OECD 2007b; p. 65).
172
E. Małuszyńska
The use of indirect methods to measure the impact of relocation on employment, as in the case of direct methods, is not easy. Critics of these methods emphasise, for example, that neither in the case of investments nor in that of trade flows undertaken on national scale, it is possible to determine which investments or trade flows or which parts thereof have resulted from delocalisation and which can be linked to export by companies with no outsourcing links with home-country enterprises. The same applies to trade flows between enterprises and their affiliates. Neither do delocalisation measures account for the transfers of the part of enterprises’ goods or services for which the relevant strategies envision no reimportation of goods to the domestic market but rather their sale directly in the third countries (The EU 2005; Fessler 2006; Falk and Wolfmayr 2005). It is also not easy to estimate the amount of job losses related to delocalisation-proneness of sectors or industries. The distinction between jobs that are or are not offshorable is subjective. So, the results of research on job losses are often different. For example, the share of jobs potentially exposed to relocation in 2002 was, according to various studies, 18.1 or 13.7% of the total number of employees in the USA (Małuszynska 2013, p. 93). Because of difficulties in using above-mentioned methods in order to show the scale of the influence of delocalisation on employment in different countries as and for a long time, the only existing database was used, which is the database of European Monitoring Centre on Change.
10.4
Impact of Delocalisation on Labour Market
Both positive and negative effects observed in the home and host countries are a consequence of causal relationships in socio-economic systems in which one of the elements at play is an enterprise involved in delocalisation (Odrobina 2009). The effects of delocalisation in the home and host countries are felt on the markets for products and production factors. They can also take the form of the so-called external effects (Navaretti and Venables 2004). An analysis of changes resulting from delocalisation on home-country labour markets focuses on two aspects: 1. changes in the level of employment, 2. changes in the structure of employee skills resulting from fluctuations in demand for skilled and unskilled labour.
10.4.1 Level of Employment Assessments of the impact of delocalisation on the home country’s labour market present a considerable challenge. Some or all jobs get lost not only in enterprises
10
Delocalisation of Enterprises …
173
which shift their operations, but also in their suppliers which lose all or parts of their prior market. A further possible consequence are declines in real wages among workers employed in the affected suppliers. Such declines erode consumption triggering lay-offs in further undertakings (Ghibutiu and Poladian 2009; Spaventa and Monni 2007). Other adverse consequences of job cuts include drops in domestic demand and the resulting economic slowdowns, as well as reductions in research and development budgets and in other activities of importance for the development countries. On the other hand, as delocalising enterprises become more effective and consequently also more competitive, they gain the capacity to develop more innovative and advanced production processes and, at least theoretically, in the medium or long term should create jobs, thus partially offsetting the negative impacts (OECD 2007a, p. 35). Employment declines among suppliers mainly affect steady regular supplies, most commonly of goods and services which do not require high technologies and are therefore easier to replace. On the other hand, existing suppliers may export goods and services to an affiliate of a company which has moved its operations abroad. In such an event, the required restructuring will enable the suppliers to create jobs as well to gain the ability to handle such exports. What is more, the concerned enterprises may grow faster than before as a result of, e.g., moving into new markets and using the experience and contacts established by a relocating customer. In certain cases, in order to remain in contact with existing buyers of goods and services, suppliers may choose to migrate, i.e. move all of their operations in the pursuit of the delocalising company. The result, however, is that further jobs are lost in the home country.
10.4.2 Changes in Employment in the EU Despite the fact that job destruction in the EU as a result of delocalisation abroad is seen as an important problem, the actual number of jobs effectively offshored is relatively limited (Eurofund 2017). Furthermore, negative effects on the labour market are offset to a certain extent by job creation. This supports the assertion that “There is much less evidence that globalisation (including delocalisation as one of its manifestations [E.M.]) has had an impact on employment than politicians, workers and unions seem to think” (De la Dehesa 2006, p. 70). Other studies, conducted not only in Europe, support the claim that the scale of job cuts was limited and that as certain jobs were lost, others were also created. For instance, a study by the US Bureau of Labour Statistics found that 11% of all resignations recorded in the USA in 2004 resulted from delocalisation. Roughly 30% of those were caused by delocalisation abroad. According to OECD, the number of jobs lost in 2005 as a result of offshoring amounted to 3.4% of all jobs in France and for 7.2% of all jobs in Germany (BERR 2009). Current studies reveal the complexity of the issue. The findings vary depending on the sources and destinations of business moves. In a study of seven EU
174
E. Małuszyńska
countries, Falk and Wolfmayr (2005) demonstrated that the delocalisation of production to lower-wage countries has a statistically significant adverse impact on employment in the home country. Nevertheless, the actual decline in employment was fairly negligible at 0.26% points per annum. On the contrary, the delocalisation of production to highly developed countries had a positive impact on employment. Here, imports from highly developed higher-wage countries on the one hand and employment in the home country on the other hand are mutually complementary rather than substitutive, unlike in the case of delocalising operations to low-wage countries. The above is supported by the observations, conducted since 2002 by the EMCC in the EU and Norway, of lay-offs planned in connection with the restructuring of operations, including corporate delocalisation. Between 2002 and 2017, the lay-offs announced in connection with all restructuring projects affected 6,459,020 jobs, i.e. 2.6% of the EU’s total employment at the time. However, at the time, plans were also made to create 3,419,930 jobs (Table 10.2). The numbers of jobs to be destroyed and created in connection with international relocations were 258,034 and 10,033 jobs, respectively. The jobs to be lost as a result of such relocations accounted for 4.0% of the total number of jobs lost to restructuring over the same period. There is a certain regularity in the changes in the number of jobs expected to be lost by delocalising abroad or in the share of such jobs in the total jobs lost through all forms of restructuring. That number falls within the range of 1752–39,197, while the share fluctuates between 0.56 and 7.0%. The trend peaked in 2006 and hit a low in 2011. During the period in question, the number of jobs to be lost by delocalisations abroad and through all forms of restructuring almost always significantly exceeded the number of jobs whose creation was planned. As a consequence, the net jobs destroyed through delocalisations abroad amounted to 248,001 which accounts for 8.2% of the net total jobs destroyed through restructuring. Most jobs planned to be destroyed were reported in 2005–2009, due presumably to the European Union’s enlargement in 2004 and a period of excellent economic performance. The decline that followed may have been precipitated by the crisis. Delocalisation is always associated with the substantial costs which companies incur in both the home and destination countries. The drop in the number of jobs to be lost seen after 2009 resulted from “normal corporate instincts in the conditions of economic uncertainty that have prevailed for much of the period. A second potential explanation is that peak delocalisation period may have already passed by the time of the global financial crisis. Those companies wishing to offshore may have already largely taken advantage of production transfer possibilities to China or eastern Europe before the crisis” (Eurofund 2016, p. 23). According to the ERM Annual Report 2013, the announced jobs lost through delocalisations abroad in the old EU member states (other than Greece) in 2003– 2013 accounted for 6% of the total announced jobs lost. Meanwhile, in the new member states (other than Cyprus), the share was 2.7%. The highest values were recorded in Denmark: 17.9% and Ireland: 17.4%, the lowest in Romania: 1.1% and Lithuania: 0.4%.
10
Delocalisation of Enterprises …
175
Table 10.2 Announced jobs destroyed and created as a result of restructuring in EU28 and Norway between 2002 and 30.11.2017 Year
Total restructuring Jobs lost Job gained
2002 310,996 2003 329,855 2004 385,646 2005 538,302 2006 561,129 2007 332,607 2008 532,247 2009 658,753 2010 430,416 2011 442,399 2012 426,639 2013 356,752 2014 283,981 2015 255,662 2016 350,127 2017 263,509 Total 6,459,020 Source Own compilation based on March 2018)
Delocalisation abroad Jobs lost Job gained
–
1752 – 3835 17,614 – 89,185 26,906 90 365,016 29,252 2450 505,544 39,197 6656 382,944 22,163 40 268,657 17,136 – 198,507 25,014 – 177,473 10,145 – 213,161 9929 100 198,200 14,082 20 169,946 10,163 507 211,630 11,798 120 293–158 7141 – 281,241 10,127 50 354,591 5615 – 3,419,930 258,034 10,033 www.eurofound.europa.eu/observatories/emcc (Accessed 14
Even greater differences between the new and the old EU member states are visible in the share of the jobs lost through delocalisation in the total reported job losses connected with restructuring in the manufacturing sector in 2003–2016 (Eurofund 2016). The share was 10.9% in the EU15, 6.7% in the EU13 and 9.9% in the EU28. The highest values were recorded in Ireland: 32.29% and Portugal: 26.0%, the lowest in Poland: 3.2% and Lithuania: 2.3%. The most jobs were lost due to both the old and the new member states moving their industrial production to the EU13. The corresponding values were 44.8 and 31.9%, respectively (Eurofund 2016). The negative impacts of delocalisation on employment were additionally acknowledged by Gőrg and Hanley (2004). According to Fessler (2006) and Geurts (2009), such adverse influences on employment are stronger where production rather than services is outsourced. Positive and negative impacts on the labour market may arise in the case of delocalising a function performed by a small number of workers, workers additionally performing another function, the shutting down of entire branches and depending on whether the relocation concerns a core or a support function. Some studies are available which demonstrate either no effects or small positive or negative effects of foreign investment on employment (BERR 2009; Barbe and Riker 2017).
176
E. Małuszyńska
10.4.3 Skills and Wages As seen in the majority of the conclusions drawn from theoretical and empirical analyses of production chain break-ups and enterprise internationalisation, delocalisation tends to affect the production processes that are the least effective at a given location or the most labour-intensive while not requiring advanced skills from the workers. As activities are moved to the countries with the lower labour costs, low-skilled jobs are destroyed and more skilled workers are employed. Under such circumstances, delocalisation differentiates the demand for labour by skill level affecting employment and wage structures in both home and host countries. The phenomenon is examined on the basis of data concerning changes in the number of unskilled and highly skilled workers and the wages offered to those two categories. The best-known authors dealing with these issues are Feenstra and Hanson (1996). They use the import of direct goods as a measure of outsourcing2to examine its correlation with the earnings of higher-skilled workers. In the simplest single-sector scenario, positive impact of outsourcing on increases in the demand for skilled labour (and their wages) is observed in both the home and host countries. In less developed host countries, skills considered to be low in the source country may turn out to be relatively advanced, causing an increase in skilled labour demand in both countries. The relative increase in demand for skilled labour observed in the USA (home country) in the 1980s, which was caused by delocalisation, was estimated at 31–51%. A number of other studies put the estimate within the range of 25–50% (Molnar et al. 2007). Also in the period 1996–2007, a number of facts pointed at a strong link between offshoring and skill upgrading in France (Carluccio et al. 2015). Another driver of relative increases in higher-skilled labour in both home and host countries are cuts in lower-skilled jobs. While in such circumstances, the home country tends to retain highest-skilled workers, the host country gains jobs for more skilled workers (Feenstra 2007).3 However, the relation between delocalisation and changes in worker skill structure is not unequivocal. Research shows mainly shifts in demand in favour of relatively low-skilled labour in the host country. The effects of delocalisation as a driver of labour upskilling in the home country have not always been clear. A case in point is a decline in demand for unskilled production workers in Italy, discovered at the level of both individual enterprises and entire industrial regions. Nevertheless, delocalisation did either nothing or very little for employment in more skilled positions (Antonietti and Antonioli 2011). A whole different set of results came out of the studies conducted in Sweden and covering the 1970–1994 and 1995–2000 periods (Blomström et al. 1997; Ekholm and Hakkala 2006). As it turned out, the delocalisation of production bears a strong 2
Here, outsourcing refers to relocating production to both unaffiliated companies and foreign affiliates. 3 In this context, a distinction is made between unskilled, medium-skilled and highest-skilled labour.
10
Delocalisation of Enterprises …
177
positive correlation with low-skilled workers, while showing no impact whatsoever on reducing the size of this worker category. As for higher-skilled workers, positive although weak correlations were only observed in 1970 and 1974. The observed rise in employment in this group reflects the need for a new organisation, new supervision and new support efforts. Partial clarification came from the main investment targets adopted by Swedish enterprises, i.e. predominantly highly developed countries and especially the USA and countries of Western Europe. Contrary to the general trend, the production processes moved out of the country were the ones which required higher-skilled workers. An only small proportion of Swedish companies chose to relocate their labour-intensive production to low-wage countries (Hansson 2001; Blomström et al. 1997). A similar trend was observed in the USA (Blinder and Kruger 2009). However, there were cases in which the delocalisation of production to less developed countries did not have a significant negative impact on the demand for low-skilled labour. Declines were seen, however, in demand for workers with secondary education (Ekholm and Hakkala 2006). Slaughter (2000), in his turn, found no link between delocalisation and the demand for either skilled or unskilled workers.
10.5
Conclusion
A review of the definitions of the terms relocation, delocalisation, offshoring and outsourcing as well as other related notions made it possible to structure the terminology in current use. Thus, delocalisation was accepted to denote a form of moving firm’s business, provided it is accompanied by a total or partial job loss. This makes it a form of insourcing and outsourcing abroad as well as domestic insourcing and outsourcing (see Table 10.1). The number of jobs lost through relocation is an inherent attribute and a measure of delocalisation. Meanwhile, relocation is synonymous with moving firm’s business. One key problem continues to be faced in measuring a feature that distinguishes delocalisation: the loss of jobs associated with moving one’s business. The reason why only few databases with such information are available results from difficulties of obtaining information about job losses from enterprises. The databases which are created vary in their spatial and temporal scopes and employ different methodologies for data collection. As a consequence, it is impossible to compare results and identify trends in the observed processes. One of the biggest such databases was established by the EMCC in 2002. As in other similar bases, this database comes with weaknesses which is the lack of representativeness of the sample that it relies on and its emphasis on medium and large enterprises which are consequently predominantly industrial. The advantages of the EMCC database are its timeliness, its identification of individual cases of restructuring based on publicly available information and its uniqueness as an EU-wide (includes data on all EU member states) dataset of large-scale restructuring events (EMCC 2016). Despite the above drawbacks of this database, members of research teams using it always underline its usefulness for
178
E. Małuszyńska
estimating the sizes, trends and effects of corporate delocalisation on the loss of jobs. The indirect methods of researching the changes in headcounts that result from delocalisations abroad include those that rely on estimating the outsourcing index and the employment equivalent of import, as well as delocalisation-proneness of enterprises. Both the research on delocalisation based on the data provided in the EMCC database and other studies are therefore in line with the job losses and job creation indicated in theory. However, observed decrease is relatively limited compared with the job destruction which results from other forms of enterprise restructuring. Between 2002 and 2017, the share of expected job closures as a result of delocalisation abroad was merely 4% of the total number of jobs expected to be lost as a result of all types of restructuring. A larger share of 6.5% was observed in the old member states from 2003 to 2013. The rate in the new EU member states was 2.7%. The most jobs were lost due to both the old and the new member states moving their industrial production to the EU13. The corresponding values were 44.8 and 31.9%, respectively (Eurofund 2016). The relation between delocalisation and changes in worker skill structure is not unequivocal. Research show mainly shifts in demand in favour of relatively low-skilled labour in the host country. The effects of delocalisation as a driver of labour upskilling in the home country, as it is expected in theory, have not always been clear. Estimates of the effects of job cuts associated with enterprise relocations may help avoid or at least minimise the related social costs. An equally valid reason for the research is to curb other adverse developments precipitated by actual delocalisation as well as, and frequently so, by even its mere announcement. Such developments include downward pressures on wages and working conditions felt by workers who, desperate to retain their jobs, may be willing to make extraordinary concessions. The phenomenon is referred to as reversed delocalisation. Despite the well-documented low impact of delocalisation on job cuts compared to other restructuring processes, delocalisation is still viewed negatively. This conclusion is supported by, among other things, the findings of a 2012 Eurostat study which found that 41% of the EU’s inhabitants associated globalisation, which was generally seen as a positive factor that enables and facilitates relocations, with threats to domestic jobs. Research on enterprise delocalisation is far from easy. The challenge lies not only in the lack of direct information but also in the indirect operation of many impacts, positive as well as negative. Another cause of the difficulties is the long lag in the manifestation of certain effects. More often than not, there is no single answer which allows one to draw general conclusions. The influence of delocalisation on the number of jobs available may be either positive or negative, depending on the adopted time horizon, the level of development achieved by the home and host countries, the choice of delocalisation destinations, the level of sophistication of the relocated production, the ownership and organisation structure of the concerned enterprises and many other factors. Variations are also seen in the impacts of delocalisation on the structure of worker skills.
10
Delocalisation of Enterprises …
179
References Antonietti, R., & Antonioli, D. (2011). The impact of production offshoring on the skill composition of manufacturing firms: Evidence from Italy. International Review of Applied Economics, 25(1), 87–105. Barbe, A., & Riker, D. (2017). The effects of offshoring on domestic workers: A review of the literature. Office of Economics Working Paper 2017-10-A. https://www.usitc.gov/publications/ 332/working_papers/ecwp-2017-10-a-10-03-17-as-pdf.pdf. Accessed November 1, 2017. BERR Economics Paper. (2009). The globalization of value chains and industrial transformation in the UK, London, 6. Blinder, A. S. (2005). Fear of offshoring. http://princeton.edu/-blinder/papers/05offshoringWP.pdf. Accessed May 10, 2013. Blinder, A. S. (2007). Offshoring big deal or business as usual? CEPS Working Paper, 149. Blinder, A. S., & Kruger, A. B. (2009). Alternative measures of offshorability, a survey approach. NBER Working Paper 15287. Blomström, M., Fors, G., & Lipsey, R. E. (1997). Foreign direct investment and employment: Home country experience in the United States and Sweden. NBER Working Paper, 6205. Brouwer, A. E., Mariotti, I., & Ommeren, J. N. (2004). The firm relocation decision: An empirical investigation. The Annales of Regional Science, 2(38). Carluccio, J., Cunat, A., Fadinger, H., & Fons-Rosen, C. (2015). Offshoring and unskilled labour demand: Evidence that trade matters. VOX CEPR’s Policy Portal. http://voxeu.org/article/ offshoring-and-unskilled-labour-demand-new-evidence. Accessed November 1, 2017. Ciesielska, D. (2009). Wpływ offshoringu na rozwój przedsiębiorstwa w świetle koncepcji zarządzania wartością firmy. Finanse, Finansowy Kwartalnik Internetowy. http://www.efinanse.com/artykuly_eng/105.pdf. Accessed October 10, 2012. De la Dehesa, G. (2006). Winners and losers in globalization. Blackwell Publishing. Ekholm, K., & Hakkala, K. (2006). Regional effects of offshoring on labor demand. Evidence from Sweden. http://www.etsg.org/ETSG2006/papers/Ekholm.pdf. Accessed March 10, 2013. EMCC. (2016). European monitoring centre on change. https://www.eurofound.europa.eu/pl/ observatories/european-monitoring-centre-on-change-emcc/european-restructuring-monitor. Accessed January 5, 2017. Eurofound. (2013). Monitoring and managing restructuring in the 21st century. Luxembourg: Publication Office of the EU. Eurofund. (2016). ERM annual report 2016: Globalisation slowdown? Recent evidence of offshoring and reshoring in Europe. Luxembourg: Publication Office of the EU. Eurofund. (2017). Living and working in Europe 2016. Luxembourg: Publication Office of the EU. European Parliament, DG Internal Policies of the Union. (2006). Relocation of EU industry. An overview of literature. PE 382.166. Falk, M., & Wolfmayr, Y. (2005). The impact of international outsourcing on employment: Empirical evidence from EU countries. Vienna: Austrian Institute of Economic Research, WIFO. Feenstra, R. C. (2007). Globalization and its impact on labor. wiiw Working Papers 44. Feenstra, R. C., & Hanson, G. H. (1996). Globalization, outsourcing and wage inequality. NBER Working Paper 5424. Fessler, P. (2006). Home country effects of offshoring: A critical survey on empirical literature. SFB International Tax Coordination Discussion, Paper 23. Geurts, K. (2009). When work move off. Effect of outsourcing on firm level employment growth. https://lirias.kuleuven.be/bitstream/123456789/247414/2/geurts_whenworkmovesoff.pdf. Accessed January 17, 2017. Ghibutiu, A., & Poladian, S. (2009). Global sourcing of services: How well are the new member states coping with challenges? Romanian Journal of Economic Forecasting, 10(1), 123–135. Gierańczyk, W. (2008). Problematyka definiowania zmian w tendencjach lokalizacyjnych przedsiębiorstw przemysłowych w dobie globalizacji. In Z. Zioło & T. Rachwał
180
E. Małuszyńska
Problematyka badawcza geografii przemysłu (no. 11, pp. 86–97). Warszawa-Kraków: Prace Komisji Geografii Przemysłu. Gőrg, H., & Hanley, A. (2004). Labour demand effects of international outsourcing. Evidence from plant level data. Research paper series: Globalization and labour market, Research Paper 37, The University of Nottingham. Hansson, P. (2001). Skill upgrading and production transfer within Swedish multinationals in the 1990s. CEPS Working Documents, 163. Hatzichronoglou, T. (2005). The impact of offshoring on employment: Measurement issues and implications. OECD. https://www.oecd.org/sti/sci-tech/35644921.pdf. Accessed September 9, 2016. Jovanović, M. (2003). Local vs. global location of firms and industries. Journal of Economic Integration, 18(1), 60–104. Mankiw, N. G., & Swagel, P. (2006). The politics and economics of offshore outsourcing. NBER Working Paper 12398. ManVis Report. (2005). Manufacturing visions. Policy summary and recommendations. Fraunhofer Institute for System an Innovation Research, 6. Małuszynska, E. (2013). Migracje i delokalizacja przedsiębiorstw. Poznan: Wydawnictwo Uni-wersytetu Ekonomicznego w Poznaniu. Mariotti, I. (2005). Firm relocation and regional policy. A focus on Italy, the Netherlands and the United Kingdom. Netherlands Geographical Studies, 331. Mazzanti, M., Montresor, S., & Pini, P. (2011). Outsourcing, delocalization and firm organization: Transaction costs versus industrial relations in a local production system of Emilia Romagna, entrepreneurship & regional development. An International Journal, 23(7–8), 419–447. https:// doi.org/10.1080/08985620903233986. Molnar, M., Pain, N., & Taglioni, D. (2007). The internationalization of production, international outsourcing and employment in the OECD. OECD Economics Department, Working Papers 561. Mozalewski, M. (2010). Metodyka przeprowadzania restrukturyzacji przedsiębiorstw. Ruch Prawniczy, Ekonomiczny i Socjologiczny, Rok LXXII, z. 2, 215–236. Navaretti, G. B., & Venables, A. J. (Eds.). (2004). Multinational firms in the world economy. Princeton University Press. Odrobina, A. (2009). Delokalizacja jako skutek globalizacji i integracji. In S. Miklaszewski & E. Molendowski (Eds.), Gospodarka światowa w warunkach globalizacji i regionalizacji rynków (pp. 81–103). Warszawa: Difin. OECD. (2007a). Offshoring and employment. Trends and impacts. Denver. OECD. (2007b). Staying competitive in the global economy. Moving up the value chain. Denver. Offshoring: An Elusive Phenomenon. (2009). National Academy of Public Administration, USA. https://www.bea.gov/papers/pdf/NAPAOff-ShoringJan06.pdf. Accessed December 17, 2016. Olsen, K. B. (2006). Productivity impacts of offshoring and outsourcing: A review. OECD Science, Technology and Industry Working Papers 2006/1, OECD Publishing. http://dx.doi. org/10.1787/685237388034. Accessed December 15, 2016. Pellenbarg, P. H., Wissen, L. J. G., & Dijk, J. (2002). Firm migration. In E. McCann, & P. Elgar (Eds.), Industrial location economics. Cheltenham, Northampton. Radło, M. J. (2005). Offshoring i outsourcing a gospodarka europejska. Czy należy bać się delokalizacji? In M. J. Radło (Ed.), Polska wobec polityki gospodarczej Unii Europejskiej (pp 54–73). Warszawa-Gdańsk: IbnGR. Radło, M. J. (2008). Offshoring i outsourcing w Unii Europejskiej a wzrost gospodarczy i zatrudnienie. In A. Szymaniak (Ed.), Globalizacja usług. Outsourcing, offshoring i shared services centres (pp. 217–235). Warszawa: Wydawnictwa Akademickie i Profesjonalne. Radło, M. J. (2013). Offshoring i outsourcing. Implikacje dla gospodarki i przedsiebiorstw. Oficyna Wydawnicza SGH w Warszawie.
10
Delocalisation of Enterprises …
181
Relocation of Economic Activities Abroad and European Economic Development, Report. (2005). Committee on economic affairs and development, PE, Doc. 10757, Council of Europe. http://assembly.coe.int/Mainf.asp?link=/Documents/WorkingDocs/Doc05/EDOC10757.htm. Accessed August 14, 2012. Rojek, T. (2016). Wykorzystanie narzędzi restrukturyzacji w procesie zarządzania efektywnością przedsiębiorstwa. Studia Oeconomica Posnanensia, 2(4), 135–155. Romanowska, M. (2011). Restrukturyzacja jako reakcja na zmiany. Zeszyty Naukowe/ Uniwersytet Szczeciński, 629, 207–214. Sako, M. (2005). Outsourcing and offshoring: Key trends and issues. Oxford Said Business School. Singh, H.(1993). Challenges in researching corporate restructuring. Journal of Management Studies, 30, 147–172. Slaughter, M. J. (2000). Production transfer within multinational enterprises and American wages. Journal of International Economics, 5, 449–472. Spaventa, A., & Monni, S. (2007). What next? How the internationalization process might lead to the dissolution of Veneto’s low-technology industrial districts. CREI Working Paper no 3. Suszyński, C. (1999). Restrukturyzacja przedsiębiorstw. Proces zarządzania zmianami. Warszawa: PWE. The EU Economy Review. (2005). Rising international economic integration. Opportunities and challenges, DG for Economic and Financial Affairs 6. Welsum, D., & van Reif, X. (2006). The share of employment potentially affected by offshoring— An empirical investigation. OECD, DSTI/ICCP/IE(2005)8/FINAL. Whalley, J. (2006). The new global division of labour—How is the world economy affected by outsourcing and offshoring? CESifo Forum, 7(3). Zorska, A. (2008). Outsourcing i przenoszenie usług w dobie globalizacji oraz informatyzacji. In A. Szymaniak (Ed.), Globalizacja usług. Outsourcing, Offshoring i Shared Services Centres (pp. 189–2016). Warszawa: Wydawnictwa Akademickie i Profesjonalne.
Ewa Małuszyńska is a professor of economics. She obtained M.Sc. in physical geography at Adam Mickiewicz University in Poznań and Ph.D. in economics at Poznań University of Economics and Business in Poland, where she works today. Since 1998, she is a chair of the Department of European Studies. In the 1996–2000, she was a Polish member of the Council of Representatives of the Association of the European Schools of Planning. Her research specialisation and interests focus on the economics of cities and regions, regional policy of the EU, spatial economy and business location and delocalisation. The most important publications include “Business expectations in the old and new EU Member States”, “Regional differentiation of the economic structural changes” and “Business migration and delocalisation”.
Chapter 11
Epilogue: Relocation Factors and Consequences—Towards a Multidisciplinary Research Agenda Magdalena Dej and Paweł Capik
Relocations of economic entities in their multitude of forms and varieties, spatial variations, conditionings, as well as the commonness of occurrence are both a complex and a fascinating research subject. At the same time, it has been relatively poorly explored so far, at least comprehensively, which is influenced by numerous factors, among which particularly important barriers include difficulties with obtaining data, incomparability of the data among countries, as well as the number of relocation types, which makes the univocal classification of the observed behaviours of enterprises really difficult. For this reason, this book is a unique work within the current literature, as its authors took on the task to refer comprehensively to relocation processes in order to take into consideration a wide range of phenomena and pursued diverse approaches in studying those. Consequently, the book offered a wide-ranging overview in terms of the perspective of approaching the phenomenon, the spatial scale and the focus of analysis. The research findings presented in it create the theoretical and empirical framework which will be useful for future researchers aiming at setting their work in relation to the existing literature on the problem of the relocation of business entities. It is important since in a sense relocations of firms as a research subject get lost in other related issues, particularly in the processes of the location of economic entities. They are often treated as a specific variety of these processes, which is manifested both in the theoretical and empirical layer of analysis. A significant indicator of a specific marginalisation of relocations in the research so far is the fact that the processes, in spite of their complexity and universality, still lack a separate set of theories. They are usually analysed, depending on the spatial M. Dej National Institute for Spatial Policy and Housing, Kraków, Poland e-mail:
[email protected] P. Capik (&) Keele Management School, Keele University, Keele, UK e-mail:
[email protected] © Springer International Publishing AG, part of Springer Nature 2019 P. Capik and M. Dej (eds.), Relocation of Economic Activity, https://doi.org/10.1007/978-3-319-92282-9_11
183
184
M. Dej and P. Capik
scale and the character of a relocation as well as its aspect, by embedding in the theoretical context related to the location of economic entities, GVC, GPN and other sets of theories. It all makes this publication fill an evident gap in existing literature by systematisation of problems connected with the relocations of firms and identification of current trends within this process, as well as further research directions. Its value consists of the following aspects: – it provides conclusions on the empirical grounds, which were formulated by the authors with regard to a very diverse group of countries and regions being on different stages of social and economic development, – it provides knowledge from research on different levels of spatial analyses, exploring relocations at the international, national, regional and local levels, – analyses phenomena occurring nowadays, but also refers to relocations from the historical perspective, – analyses relocations both from the perspective of business entities as well as the socio-economic and institutional environment within which they occur. As the summary of its content, below there are main conclusions from the research presented in individual chapters. These are both empirical, and quite often theoretical and methodological conclusions. Proportions among these three categories included in the papers of individual authors differed. In the first part of the book, devoted to Relocation of Firm and Industries, in the chapter opening it, Suwala formulates conclusions concerning historical and spatial progression of developmental trajectories of the aviation industry in Poland. He emphasises the significance of the dynamic character of the location of this branch of industry, paying attention to omnipresent location competition. The historical reconstruction of the paths in the Polish aviation industry provides conclusions both on the theoretical and empirical grounds. The author embeds his research in two complementary approaches—process- and path-based approaches, and justifies that only the eclectic application of both enables to assess how historical and geographical conditionings have been influencing the shape of the aviation industry and its location and relocation in Poland since 1912. The author analyses historical location decisions by referring to the existing location theories. And thus he points out that in the initial period (until 1912), the location of the plants was randomly scattered all over the country, exploiting a number of opportunities which he refers to as the ‘windows of locational opportunity’. At the same time, concentration was also observed in some locations, for example in Lublin or in Warsaw. The events of the following years were the reason for which the developmental path strengthened the latter location in particular. Further political decisions, as well as other factors (financial difficulties in other locations, external economies of scale) were the factors for which with time Warsaw became the heart of the Polish aviation industry with advanced production (e.g. production of motors, assembly) and its great variety (e.g. gliders, planes). At the same time, two events of a different character— the planned construction of the Central Industrial District and German invasion of
11
Epilogue: Relocation Factors and Consequences …
185
Poland totally shook the landscape of the Polish aviation and considerably destroyed it. The landscape started to restore in the new realities of the communist government in Poland and in spite of generally poor conditions for the development of this industry just after the end of the Second World War, the locations of aviation plants from that period turned out to be permanent and some of them still play an important role on the map of the Polish aviation industry. In the second chapter, Wójtowicz, analysing relocations in the automotive industry in Brazil and Mexico, focuses on the influence of global and regional factors on the development and changes in the arrangement of the automotive industry in those countries in relation to global value chains (GVCs). Relocations of the automotive industry firms in both countries, as well as the accompanying growth of employment and production, are the indicators of success of the transformation the industry went through in the analysed countries. The author identifies that the transformation was made up of three groups of factors. Firstly, the changes in the industrial policy of both countries including the liberalisation of trade and more pro-export orientation of industrial production. Secondly, global changes in the automotive industry manifested mainly through the growing competition between MNEs the delocalisation forcing the closing down of factories in developed economies and their relocation to countries such as Mexico or Brazil. Thirdly, the changes in MNEs’ development strategies manifested in the restructuring of production plants. Those three groups of factors, in the author’s view, brought about transformations in the spatial arrangement of the automotive industry in Brazil and Mexico. In both countries, changes in the global and regional conditions of the sector functioning caused the mass process of the relocation of manufacturing plants, primarily to regions perceived as more peripheral. On the theoretical grounds, relating those results to global value chains (GVCs), directions of those relocations were strongly determined by a different role which the analysed countries played in the regional production systems and strategies of MNEs concerning the regional patterns of the GVC organisation in the automotive industry. In the next chapter, Fu and Yang focus on the analysis of the Chinese market and reorientation of Chinese firms with regard to the markets for their products, which was related to the limitation of the demand for goods in developed markets as a result of the economic crisis. The authors, using global production networks (GPN) concept and analysing furniture firms formulate conclusions about the significance of the factor which this crisis was for the modernisation of this industry in China. At the same time, they point to various modernisation trajectories. The example of the furniture industry in Pearl River Delta leads to the conclusions about the significance of emerging markets for the transformations and modernisation of industry in consequence of the spatial reorientation of production. In this specific example, owing to the integration of furniture firms with Chinese domestic markets, the occurrence of modernisation to a great extent was determined by the fact that in spite of this reorientation, the entities were still under the control of leading international companies in global production networks. In the last chapter of the first part, Koster and Pellenbarg focus on the changing landscape of firms in the Netherlands and locational behaviours of small businesses,
186
M. Dej and P. Capik
which prevail more and more in the Dutch economy. With the growth of the importance of self-employment as a form of conducting business, decisions related to localisation of the firm depend on the owner—an individual, and members of his household, to an extent greater than believed in the past. The authors argue that it translates into the dynamics of firms, both in terms of localisation decisions taken and the results of these processes. The most important conclusion from the presented research is that in such a case the borders of firms become permeable, which means that in the case of the smallest business entities, often located in the entrepreneur’s place of residence, factors not related directly to the firm’s activity are a significant location argument. The variables such as family arrangements and the spouse's place of work are equally important as premises concerning purely conducting activity by a firm. Therefore, the location of a firm is a product of many equivalent factors. The authors draw attention to the fact that a significant change has taken place within that scope in comparison with the initial period of research into relocations, during which the development of an economy was determined by large, often multinational businesses. The attempts to understand their relocation decisions were based on the assumption of the search for the optimal location exclusively from the perspective of a business entity. At present, in the changing conditions, the analysis of the location optimisation in the case of small firms comprise a broader spectrum of factors, also including the preferences and needs of entrepreneurs’ households. Moreover, the chapter provides very interesting conclusions on relations between entrepreneurship, location of business entities and the living conditions in specific areas. It is particularly interesting especially in the case of rural areas, where more than a half of firms are based in the entrepreneur’s place of residence. In the light of current debates on population decline, the pressure on rural labour markets and sustaining services in low-density areas, presented conclusions may constitute a crucial argument for the formulation of policies for those areas with regard to the provision of services and technical infrastructure. The second part of the book devoted to factors and effects of relocation processes consists of five chapters analysing this issue on the examples of selected countries and regions. In the opening chapter, Dej, Jarczewski and Chlebicki present the scale and the main characteristics of relocation processes in five Polish metropolitan areas. The relocation processes carried out in the years 2001–2013 were taken into consideration, without micro-enterprises. The chapter delivers both empirical and methodological arguments. The methodological achievement of the study is the development of the authorial method of the identification of relocation processes of firms in the internal scale in Poland, in the situation of the lack of the accessibility of data enabling directly to indicate firms which relocate. The use of commercial company databases (BISNODE Polska, in this case) and the data of the National Court Register, as well as GIS software, in spite of their time-consuming character, allow to reliably define the scale of the phenomenon and its main characteristics, including also the distance to which entities relocate. Investigating relocations in Polish metropolitan areas enabled to define their scale, which turns out to be of a mass character—as many as one-third of firms located in those areas made at least one relocation, although a lot of those firms made such moves more often, in extreme
11
Epilogue: Relocation Factors and Consequences …
187
cases even six or seven times. Other findings are consistent with those which have already been described in the literature based on research carried out in other countries. Among others, relations between undertaking relocation and the firm’s age were proven. New firms practically do not relocate, and the time needed to take such a decision and the whole decision-making process lasts two–three years on average. Moreover, it was statistically proved that if a firm does not undertake relocation within the first five–ten years of its activity, in the following years the probability of such a change decreases significantly. In the studied sample, also observations from earlier works were confirmed, that bigger firms relocate more often than smaller ones, at the same time the latter relocate to a greater distance, and the most mobile are firms within the market services branches. This is related to lower engagement of fixed assets and lower sunk costs. The conducted research provides new knowledge on spatial-economic processes in Polish metropolitan areas, at the same time revealing subsequent areas of the lack of knowledge which require further in-depth research, for example relocation motivations. The chapter by Rossi concentrates on the analysis in the smaller spatial scale. The author focuses on the processes taking place in Ticino—a southern canton of Switzerland bordering Italy, and particularly on factors influencing the attractiveness of that location, both its strengths and weaknesses, from the perspective of relocating firms. Based on the information coming directly from entities which have relocated, the author confirms the conclusions already described in the existing literature about the particularly positive role of the stable institutional environment, as well as good transport accessibility and the availability of relatively cheap and adequate means of production. A unique location and the specificity of Canton Ticino on the Italian and Swiss border makes the region benefit from such a location and create attractive conditions for firms which choose this area as a new place to conduct their business from. Italian firms often relocate to Ticino using stable Swiss institutional facilities, low bureaucracy, high quality of life. On the other hand, the analysis of the weaknesses of this area indicates a number of negative factors when compared with the areas of the origin of relocated firms (mainly Italy). These are, in the first place, insufficient human capital, especially poor availability of medium- and highly qualified employees, as well as problems connected with congestion. Yet, in comparison with other Swiss cantons, the advantage of Ticino also includes lower costs of conducting business, due to lower wages and prices of real estate. Rossi’s chapter provides a detailed case study of the border area as an exceptional place from the point of view of firm relocation processes, which has a chance, with the properly devised policy on the local, regional and national level, to become particularly attractive for firms ready or determined to change location. At the same time, the author suggests the necessity to extend the research and apply econometric models in order to identify relations between push and pull factors and decisions of firms considering relocation. Also, the next chapter by Kapitsinis concentrates on a sort of competition for firms between two bordering countries, this time Greece and Bulgaria. The analyses of relocations in this case are conducted in the context of 2007 global economic crisis, which in a special way, more than any other UE country, affected Greek
188
M. Dej and P. Capik
economy, businesses and society. First, the research presented in this chapter provides interesting discussion of how the global economic crisis affects the weight of individual relocation factors in the case of SME firms. In spite of the fact that most of the current literature assumes specific determinants of business mobility, the research findings do not confirm this. The comparison of the role and significance of relocation factors before and after the economic crisis as for relocations between Greece and Bulgaria indicates their changeability in time. When compared to the period before the crisis, the importance of some factors attracting firms from Greece to Bulgaria decreased (e.g. costs of labour), the importance of others increased (e.g. access to external co-financing, the role of social trust). Also, the significance of some factors pushing firms away from Greece has increased (e.g. no demand for goods, high level of bureaucracy, legal barriers, quality of institutional environment). At the same time, some factors, both before and after the crisis, turn out to be unalterably important for the relocation of firms from Greece to Bulgaria (e.g. geographical proximity of Greece, low taxes and costs of labour, as well as a moderate level of bureaucracy in Bulgaria), which makes it possible to regard them crucial. However, a very important conclusion from the presented research and the contribution to the existing state of knowledge is concerned with the changeability of the weight of some factors. Unfortunately, the fact of the overlapping of two substantial events in time—the accession of Bulgaria to the EU and the world economic crisis (2007)—does not allow very distinct separation of the significance of each of them for the increased level of firm location. However, with a high level of probability we can indicate that the occurrence of both of them additionally strengthened the scale of relocations between Greece and Bulgaria. In the penultimate chapter of the volume, Castellani and Lavoratori concentrate on the global dimension of relocations, analysing the determinants of the internationalisation process of the R&D industry. Based on the research carried out in over a hundred global cities all over the world, the authors reach conclusions about the key significance of those centres as preferred ones for the location of R&D activity. In particular, the attention of the researchers focuses on the analysis of the significance of external and internal factors in taking decisions about relocation by entrepreneurs of this industry. In particular, the authors ponder over the links of activities in the whole supply chain from the perspective of firms’ relocation, and especially between manufacturing and research and development activity. As they conclude, two opposite forces operate here. External factors tend to disperse activity in search for the most beneficial location. On the other hand, internal forces have the concentrating influence, which results from benefits of proximate localisation of activities along the value chain and maintaining intra-firm linkages. And this is exactly in this field—in terms of drawing attention and emphasising the significance of internal factors for the processes of location and relocation of R&D activity— that the chapter contributes most to the existing state of knowledge, as the research into this phenomenon so far has concentrated mainly on the role of external factors. The final chapter of the book by Małuszyńska focuses on delocalisation of enterprises and the result of those processes, both quantitative and qualitative ones, for labour markets. The discussion is devoted to the phenomenon of corporate
11
Epilogue: Relocation Factors and Consequences …
189
delocalisation and links this issue to migration of firms, offshoring and outsourcing processes, methods of estimating delocalisation scale, and finally the consequences of delocalisation for labour markets of these countries and regions from which firms relocate to other places. The analysed problem has been discussed particularly lively and widely around the time of the biggest extension of the EU in its history, in 2004, when the old EU states faced the perspective of the mass loss of jobs due to the relocations of firms, particularly plants of the largest multinational enterprises. The author establishes that what differentiates delocalisation from other processes related to the transfer of business activities is the fact that the transfer is accompanied by the total or partial loss of jobs. Thus, the number of jobs lost in consequence of delocalisation is an inseparable quality and measure of this process. Małuszyńska’s chapter provides important methodological observations as it explores the availability of data, their sources and the usefulness of specific databases (including EMCC, EMCC 2016) to measure delocalisation processes. At the same time, it identifies a number of factors and barriers to conducting this type of research and the interpretation of its findings. What is apparently significant is the adopted time horizon, and the level of the economic development of the countries between which delocalisation takes place, as well as the organisational and ownership structure of the business. On the empirical grounds, the author's analysis proves that in comparison with other forms of enterprise restructuring, foreign delocalisation of firms exerts limited influence on the loss of jobs. In spite of this, as the author emphasises, referring to 2012 Eurostat study, they are still perceived negatively. When comparing the old and the new EU member states in the years 2003–2013, a bigger loss occurred in the first ones (6.5% compared to 2.7% in new member states). At the same time, due to the fact that delocalisation generates demand in the hosting country mainly for relatively low-qualified employees, it was not possible to observe a significant impact of delocalisation on improving qualifications on the labour market. Today, the world economy is in a very peculiar moment of its history. This very dynamic time is a period of transformations, crises, the moment of the occurrence of new leaders among the emerging economies, an increase in the significance of new markets and the dynamic development of new business models. All of these pose significant research challenges, as the hitherto prevailing theories and models appear only partially accurate and relevant. Just like in the case of other economic phenomena, also within the scope of relocations, there is a need to look at these processes in a fresh and creative way, and with an open mind try to identify, name and generalise the newly observed processes. A fresh look is necessary at the relocation processes in their contemporary form, their progression, determinants, directions, effects. The book we are handing over to the readers at least partially fulfils this assumption, identifying the most current trends within that scope. We hope it will also inspire future researchers who would like to deal with this issue.