5.2.1 The Integration of New Peripheries into Transnational Production Networks

Transnational production networks in the automotive industry are integrated through investment and trade flows with automotive industry commodities: raw materials, parts, components, preassembled modules, semi-finished and finished vehicles, flows of capital in the form of FDI, dividends and the transfer of profits, flows of labor and personnel, and flows of information, know-how and knowledge that allow for a fine-grained division of labor and increased regional specialization. The spatial dynamism of transnational production networks in the contemporary automotive industry is based on the investment strategies of core-based firms that are constantly looking for investment opportunities in peripheral areas in order to improve or maintain the rate of profit by lowering production costs, which are the total cost of production and delivering finished products to the market (Reference PavlínekPavlínek, 2018; Reference Pavlínek2020).

The transnational integration in the automotive industry has been extensively analyzed generally (e.g., Reference Carrillo, Lung and van TulderCarrillo et al., 2004) and in the context of the European automotive industry since the early 1990s (e.g., Reference JonesJones, 1993; Reference Freyssenet, Shimizu and VolpatoFreyssenet et al., 2003b). Chapters 3 and 4 of this book conceptualize the geographic expansion of automotive industry production networks into new geographic areas and the contemporaneous restructuring in the existing production regions by drawing on Harvey’s theory of uneven development and spatiotemporal fix (Reference HarveyHarvey, 1982; Reference Harvey2005b), which emphasizes the investment strategies of core-based automotive firms in peripheral lower-production-cost regions. Although core-based automotive firms use various strategies to ensure profitability (Reference Boyer and FreyssenetBoyer and Freyssenet, 2002), they always strive to minimize production costs by controlling the cost of factors of production. Firms can more easily control labor costs than the costs of other factors of production (Reference DickenDicken, 2015) through technological and organization innovations and through the location of production into areas with labor surplus and low labor costs (Reference HarveyHarvey, 1982). A sharp decrease in transportation costs by more than 90 percent in the twentieth century (Reference Glaeser and KohlhaseGlaeser and Kohlhase, 2004), because of new transportation technologies (Reference LevinsonLevinson, 2006) and logistical systems (Reference Kaneko and NojiriKaneko and Nojiri, 2008; Reference DanylukDanyluk, 2018), along with the lowering of trade barriers and deregulation of FDI, made it easier for firms to establish production in low-cost areas at the international scale. The potential for higher profits in such areas has been further enhanced by government policies of investment incentives, low corporate taxes and financing the construction of modern infrastructure that lower set-up sunk costs for investing firms and, therefore, lower their investment risk (Reference Clark and WrigleyClark and Wrigley, 1995; Reference JacobsJacobs, 2019; Reference PavlínekPavlínek, 2016; Reference Pavlínek2020).

Chapters 3 and 4 of this book demonstrate how both assembly firms and component suppliers are attracted to lower-cost peripheral locations by the potential of a higher rate of profit. However, along with Reference HarveyHarvey (1982), I have argued that spatiotemporal fixes in the form of the establishment of production in new low-cost areas are only a temporary solution to declining profitability. As more and more firms are exploiting a spatiotemporal fix by establishing production in the same or similar peripheral regions, an increased demand for labor exhausts labor surplus, leading to rising wages that undermine the rate of profit and future growth. Rising production costs and declining profits eventually force firms that are most dependent on low labor costs to look for new production areas with labor surplus and lower wages, which often leads to relocations of the most labor-intensive activities, such as the assembly of cable harnesses, from the existing integrated peripheral regions to previously unintegrated peripheries (e.g., Reference Aláez-Aller and Barneto-CarmonaAláez-Aller and Barneto-Carmona, 2008; Reference Lampón, Lago-Peñas and González-BenitoLampón et al., 2015; Reference Lampón, Lago-Peñas and Cabanelas2016; Reference PavlínekPavlínek, 2015a). These new peripheral areas thus become competitive in attracting new investments of core-based firms especially in labor-intensive and routine production compared to the more expensive core or existing integrated peripheries (e.g., Reference Frigant and LayanFrigant and Layan, 2009). The influx of profit-seeking investment capital into areas with a potential for a higher rate of profit results in economic growth in new low-cost peripheral regions. The outcome of this spatial investment behavior is the geographic expansion of production into new areas that are integrated into a transnational production network through capital, commodity, trade and technology linkages (hence the integrated peripheries), along with the economic growth bouncing from region to region (Reference HarveyHarvey, 1982).

These processes can be demonstrated in the European automotive industry, where, as argued in Chapter 4 of this book, the geographic expansion of the automotive industry into peripheral regions and the development of transnational production networks have been strongly related to state development policies (Reference WardWard, 1982; Reference OberhauserOberhauser, 1987; Reference PavlínekPavlínek, 2016), regional integration, the establishment and expansion of the common market in the European Union and regional free-trade agreements with non-European Union countries (Reference Hudson and SchampHudson and Schamp, 1995b; Reference LungLayan and Lung, 2004; Reference JacobsJacobs, 2019). Since the early 1960s, carmakers have actively lobbied for the geographic expansion of European regional integration that would give them opportunities to establish production in low-cost areas (Reference Layan, Humphrey, Lecler and SalernoLayan, 2000; Reference LungFreyssenet and Lung, 2004). This has led to the geographic expansion of the automotive industry from its established centers into new areas since the 1960s, as documented in Chapter 4 of this book.

5.2.2 Restructuring in Core Areas

The growth of production in newly integrated peripheries impacts the existing locations within a transnational production network. The automotive industry in core areas continues to be favored by several crucial factors that make it attractive for additional investment, including large internal and external scale economies, high accumulated and exit sunk costs, an accessibility to large markets, low transportation costs, high-quality labor force, the proximity of R&D facilities, highly developed infrastructure and high-quality institutions (Reference Bordenave and LungBordenave and Lung, 1996; Reference Clark and WrigleyClark and Wrigley, 1997; Reference Carrincazeaux, Lung and RalletCarrincazeaux et al., 2001; Reference Frigant and LungFrigant and Lung, 2002). Core areas might benefit from the expansion of production in integrated peripheries because the finer division of labor and increased regional specialization within the transnational production network increase the specialization of core regions in capital-intensive production, skill-intensive, high-value-added activities and strategic functions. At the same time, the high-volume assembly of small cars with weaker engines and labor-intensive production of generic components can be gradually relocated to the integrated periphery because of lower production costs and labor surplus (Reference JonesJones, 1993; Reference PavlínekPavlínek, 2002d; Reference Pavlínek2020; Reference LayanLayan, 2006; Reference Frigant and LayanFrigant and Layan, 2009; Reference Jürgens and KrzywdzinskiJürgens and Krzywdzinski, 2009a). German automotive firms led by Volkswagen have been particularly successful in such complementary specialization by setting uplow-cost production of small cars and/or low-volume production of special models in Spain since the late 1980s (Reference JacobsJacobs, 2019), and Portugal (Reference Ferrão, Vale, Hudson and SchampFerrão and Vale, 1995) and Eastern Europe (Reference PavlínekPavlínek, 2002d) since the early 1990s. It resulted in the more efficient territorial division of labor in automotive GPNs and, consequently, in improved competitiveness and higher corporate profits (Reference ChiappiniChiappini, 2012).

At the same time, existing core locations and older integrated peripheries, such as Belgium and Spain, may experience declining production and job loss due to the expansion of production in new integrated peripheries, especially in labor-intensive, low-value-added and less profitable production of generic components that does not require proximity to other firms. In extreme cases, this restructuring may lead to factory closures and relocations of production, especially of automotive components (Reference Frigant and LayanFrigant and Layan, 2009; Reference Lampón, Lago-Peñas and González-BenitoLampón et al., 2015; Reference JacobsJacobs, 2019; Reference PavlínekPavlínek, 2020). As we can see in Chapter 4 of this book, in Western Europe, between 2005 and 2016, large restructuring events, resulting in the creation or loss of at least 100 jobs or 10 percent or more of the labor force in automotive industry firms or factories employing at least 250 workers, led to 181 factory closures, 50 relocations and 35 partial relocations. Additionally, 529 firms experienced rationalization and job cuts, leading to 387,000 job losses altogether. At the same time, 133,000 jobs were created, resulting in the overall loss of 254,000 jobs. Some labor-intensive activities that for various reasons cannot be relocated continue to persist in core areas. In those cases, labor surplus can be imported from abroad and immigrant labor has been used for the expansion of existing plants in Western Europe for decades (Reference WardWard, 1982).

Overall, therefore, the integration of peripheral regions into transnational automotive industry GPNs triggers restructuring in core regions, semiperipheries and older integrated peripheries that results in a finer division of labor and greater regional specialization. As we can see, this continuous process of change has underlined the dynamic geography of the European automotive industry since the early 1960s. Based on the conceptual discussion, I will next explain a methodology that I will use to delimit the spatial hierarchy of the European automotive industry, before presenting empirical results of the analysis.

5.3.1 Positional Power

The positional power of countries estimates the average network position of firms in its territory (Reference MahutgaMahutga, 2014). It focuses on power asymmetries within GVCs/GPNs and considers the uneven economic power position of individual countries in transnational production networks based on international trade. The positional power of countries is calculated from national trade data in a particular industry. In the case of the automotive industry, we can measure country’s j’s producer driven power ( $P_J^P$ ) as follows.

$P_J^P={\displaystyle \sum _{i=1}^{\text{n}}\log (}{{\displaystyle X}}_{ji}/{{\displaystyle Y}}_i.+1)$

where X_ji is the value of automotive industry exports from country j to country i, Y_i is the total value of imports of the receiving country i and log is the base 10 logarithm. Country j has a high producer-driven power when it captures a large share of markets in many other countries through its exports, that is, these other countries depend on imports from country j. It has a low producer-driven power when it has a small number of such trade partners (countries).

Since the producer driven power is only based on exports, it ignores the buyer-driven power of large assembly firms and global tier-one suppliers in GPNs. It also underestimates the positional power of countries whose automotive industry is geared to large domestic markets rather than exports. Therefore, I have also calculated the buyer-driven power ( $P_J^B$ ) of country j as follows (Reference MahutgaMahutga, 2014).

$P_J^B={\displaystyle \sum _{i=1}^{\text{n}}\log (}{{\displaystyle Y}}_{ij}/{{\displaystyle X}}_i.+1)$

where Y_ij is the value of automotive industry imports imported by country j from country i, X_i is the total value of exports of the exporting country i and log is the base 10 logarithm. Country j has a high buyer-driven power when it has many trade partners (countries) from which it imports a high share of these countries’ total automotive industry exports, that is, these other countries depend on exports to country j. It has a low buyer-driven power when it has a small number of such trade partners.

The trade data were calculated for the product categories 870120–871690 of the HS6 product specification from the Eurostat ComExt database (Eurostat, 2020a). The positional power of a particular country in the automotive industry was then calculated as the average of its producer-driven and buyer-driven power for each year between 2003 and 2017 (Table 5.2). Since positional power does not measure the size of the automotive industry, countries with a larger output can have a smaller positional power than countries with a smaller output and vice versa.

5.3.2 Ownership and Control Power

Spatial systems based on the core–periphery structure are integrated through authority–dependency relationships, in which core areas dominate peripheral areas (Reference FriedmannFriedmann, 1967). Therefore, if we want to evaluate the power position of countries in such structures, we need to include a measure of power and control other than the one based on trade relations. We need to consider the uneven distribution of decision-making power among automotive industry firms, that is, who controls the industry and has the power to decide about the production and the distribution of its rewards. In other words, who controls who will produce what, where, for what price and how the benefits of production (e.g., profits) will be distributed within the GPN? These dominance and control relationships are very important proxies of the core and periphery position of countries (e.g., Reference FriedmannFriedmann, 1967; Reference LungLung, 2004; Reference FischerFischer, 2015).

Generally, core countries are those that control production in other countries through resident TNCs that directly own production facilities abroad in the case of the automotive industry. Indirectly, TNCs control production abroad also through setting the terms of trade with automotive products and through dominating captive local suppliers in peripheral regions (e.g., Reference Pavlínek and ŽížalováPavlínek and Žížalová, 2016; Reference PavlínekPavlínek, 2018). The decision-making power about the entire TNC and its GPN tends to be highly concentrated in the TNC headquarters in their home countries (e.g., Reference Pries, Wäcken, Covarrubias and Ramírez PerezPries and Wäcken, 2020).

Peripheral countries are those whose industry is predominantly controlled from abroad typically through the direct ownership of production facilities in the automotive industry by foreign TNCs. This capital dependency has strong implications for the strategic decision-making, technological, know-how and managerial dependency. Firm-level empirical evidence from the Eastern European automotive industry shows that the most important strategic decisions about foreign-owned factories are made by parent companies abroad in their TNC headquarters (Reference PavlínekPavlínek, 2016; Reference Pavlínek and ŽenkaPavlínek and Ženka, 2016).

Semiperipheral countries are positioned in-between; they control production in foreign (mostly peripheral) countries through TNCs based in semiperipheral countries and, at the same time, a significant share of their domestic industry is controlled through direct ownership from abroad, mostly from core countries. In Chapter 4 of this book, in terms of foreign ownership and control, I consider semiperipheral countries of the automotive industry as those that lack high-volume domestic assembly firms but have domestic “global suppliers” that invest in foreign countries (e.g., Britain, Canada, Sweden) (see also Reference Mordue and SweeneyMordue and Sweeney, 2020).

The positional power of countries has been therefore normalized by the index of foreign control (Reference PavlínekPavlínek, 2018), which calculates the relative importance of foreign-owned firms in the automotive industry in a given country. The index of foreign control has been calculated for each country and year between 2003 and 2017 as the average value of the share of foreign-controlled enterprises of five indicators in the manufacture of motor vehicles, trailers and semi-trailers (NACE 29 (2008–2017) and NACE 34 (2003–2007)) (Eurostat, 2020c): production value, value added at factor cost, gross investment in tangible goods, number of persons employed, and turnover or gross premiums written. A low degree of foreign control indicates a core position, while a high degree of foreign control indicates a periphery position in transnational production networks. The index of foreign control can vary between 0 and 1, with 1 indicating a total foreign control of the automotive industry and 0 indicating zero foreign control. The positional power of each country for each year was normalized by dividing it by the index of foreign control, which strengthened the relative position of countries with the low degree of foreign control of its automotive industry (e.g., Germany), while weakening it for countries with the high degree of foreign control (e.g., Slovakia).

5.3.3 Innovation Power

As discussed in the conceptual section of this chapter, the core areas of spatial systems are the prime zones of innovation activities while peripheral regions are typified by lower innovation activity (Reference FriedmannFriedmann, 1967; Reference LungLung, 2004; Reference Tödtling and TripplTödtling and Trippl, 2005; Reference Isaksen and TripplIsaksen and Trippl, 2017). In order to estimate the intensity of innovation activities in the automotive industry as a whole, the index of innovation was calculated from the share of total R&D personnel and researchers of persons employed and the share of business expenditure on R&D of the total value of production in the automotive industry (NACE 29 (2008–2017) and NACE 34 (2003–2007)) (Eurostat, 2020d). Both measures were normalized for each country and year using the following method. A country with the highest value was set to 1 and the values of all other countries were calculated in proportion to the strongest country. Therefore, the values for all countries and both variables fall between 0 and 1. In the next step, I calculated the average of these two normalized measures for each country and a particular year, which I call the index of innovation. The index of innovation thus measures the relative importance of innovation activities in the automotive industry of a given country. Next, I used the index of innovation to further normalize the positional power to arrive at the automotive industry power through multiplying the index of foreign control normalized positional power by the index of innovation, which lowers the index of foreign control normalized positional power by a greater degree for countries with a weak index of innovation than for countries with a strong index of innovation (Table 5.3).

5.3.4 Data Limitations

The 2003–2017 study period was selected because the data for the index of foreign control and innovation index is unavailable prior to 2003. The automotive industry product categories 870120–871690 of the HS6 product specification from the Eurostat ComExt database, which were used for the trade data, are not 100 percent compatible with the automotive industry product specification NACE Rev. 2 (NACE 29), which was used for the index of foreign control and the index of innovation for the 2008–2017 period.Footnote ¹ No trade data is available for Malta and Cyprus. Luxembourg, Greece and Croatia also had to be removed from the analysis due to data unavailability for the index of foreign control and the index of innovation. Luxembourg had the lowest average 2003–2017 positional power of all European Union countries and Greece and Croatia were positioned just above the second-lowest-ranked Ireland but below Bulgaria, which suggests periphery positions for these three countries. Since none of them is an important automotive producer, their removal should not affect the overall analysis. Because trade data are unavailable for Poland and Slovakia for 2003, I used their 2004 trade data for 2003. The data for the index of foreign control and innovation index are based on NACE 34 for the 2003–2007 period and NACE 29 for the 2008–2017 period.Footnote ² The 2003–2007 data for the index of foreign control and the index of innovation are unavailable for Ireland. I have used the average values of the 2008–2012 data for these two indicators to normalize the positional power of Ireland for the 2003–2007 period. In cases when one or two data values of the individual components used for the calculation of the index of foreign control for a particular country was not available for a particular year, I used the data for the closest available year as these values do not change dramatically from year to year. Belgium, Germany, Austria, Sweden and Britain provide the data of the share of total R&D personnel and researchers of persons employed only every other year. I have calculated the data for missing years as an average value of the previous and following years. Denmark, France and Britain did not provide the 2003–2006 data for R&D expenditures and I have used the 2007 values for these years instead.

5.3.5 Delimiting Spatial Categories

The K-means cluster analysis was applied on the descendent order of the natural logarithm of average automotive industry power values in order to delimit five clusters for the 2003–2017, 2003–2007, 2008–2012 and 2013–2017 periods. Five-year automotive industry power averages were used in order to minimize the effect of data limitations on annual fluctuations in automotive industry power. Five delimited clusters correspond with the spatial categories as follows: a higher-order core, lower-order core, semiperiphery, periphery and lower-order periphery (Table 5.4).

Drawing on the cluster analysis, I have evaluated changes in the position of countries during 2003–2017 as follows. First, I have used the clusters based on the 2003–2017 automotive industry power averages to determine positions of individual countries during the entire 2003–2017 period. Second, I have compared the 2003–2017 position of each country with its 2003–2007, 2008–2012 and 2013–2017 positions. If a country was classified in the same cluster during all three five-year periods as during the entire 2003–2017 period, I considered its relative position to be stable. If not, I considered its relative position to be unstable.

5.4.1 Core Countries

5.4.1.1 Stable Core

The cluster analysis based on the natural logarithm of the average 2003–2017 automotive industry power values classified five countries in the core of the European automotive industry: Germany, France, Italy, Sweden and Britain (Table 5.4, Figures 5.1 and 5.2). Germany, France and Italy were delimited in the stable core, with Germany being classified in a separate cluster corresponding with its higher-order core position. France and Italy represented a much weaker lower-order stable core. The stable core countries consistently kept their automotive industry power rank positions during 2003–2017 (Table 5.5).

Figure 5.1 Automotive industry power of selected European Union countries, 2003–2017

Source: author, based on data in Table 5.3.

Figure 5.2 The core, semiperiphery and periphery of the European automotive industry delimited by cluster analysis based on the natural logarithm of average values of automotive industry power during 2003–2007, 2008–2012, 2013–2017 and 2003–2017

Source: author.

Table 5.5 Change in the relative position of European Union countries between 2003–2007 and 2013–2017 according to automotive industry power

	Rank 2003–2017	Rank 2003–2007	Rank 2013–2017	Difference between 2003–2007 and 2013–2017
Germany	1	1	1	0
France	2	2	2	0
Italy	3	3	3	0
Sweden	4	4	4	0
Britain	5	5	5	0
Austria	6	6	7	−1
Netherlands	7	7	6	1
Belgium	8	10	8	2
Spain	9	9	9	0
Finland	10	8	10	−2
Slovenia	11	15	11	4
Czechia	12	11	12	−1
Denmark	13	13	13	0
Lithuania	14	14	15	−1
Poland	15	17	14	3
Hungary	16	19	16	3
Estonia	17	12	21	−9
Romania	18	16	20	−4
Portugal	19	20	19	1
Latvia	20	18	17	1
Slovakia	21	22	18	4
Ireland	22	21	22	−1
Bulgaria	23	23	23	0

Source: calculated by author from data available at Eurostat (2020a; 2020c; 2020d).

Germany dominated trade relations with all European countries during the entire period (i.e., had the highest value of the positional power every year) (Tables 5.2 and 5.6), had the lowest index of foreign control (Table 5.7) and the second-highest average level of the innovation index (Table 5.8). Germany’s dominant position of the higher-order core is reflected by its automotive industry power being on average 8.4 times higher than that of France and 12.5 times higher than that of Italy (Table 5.3, Figure 5.1).

Table 5.6 Change in the relative trade position of European Union countries between 2003–2007 and 2013–2017 according to positional power

	Rank 2003–2017	Rank 2003–2007	Rank 2013–2017	Difference between 2003–2007 and 2013–2017
Germany	1	1	1	0
France	2	2	2	0
Belgium	3	4	3	1
Britain	4	3	4	−1
Spain	5	5	5	0
Italy	6	6	6	0
Sweden	7	7	9	−2
Poland	8	10	8	2
Czechia	9	11	7	4
Netherlands	10	8	10	−2
Austria	11	9	11	−2
Slovakia	12	19	12	7
Hungary	13	12	13	−1
Lithuania	14	14	14	0
Slovenia	15	16	15	1
Romania	16	20	16	4
Latvia	17	15	17	−2
Finland	18	13	18	−5
Estonia	19	17	19	−2
Denmark	20	18	20	−2
Portugal	21	21	21	0
Bulgaria	22	22	22	0
Ireland	23	23	23	0

Source: calculated by author from data available at Eurostat (2020a).

Table 5.7 Index of foreign control in the European automotive industry by country, 2003–2017

	Average 2003–2017 (%)	Average 2003–2007 (%)	Average 2013–2017 (%)	Rank 2003–2017	Rank 2003–2007	Rank 2013–2017	Change in rank between 2003–2007 and 2013–2017
Germany	14.6	14.1	14.8	1	1	1	0
Italy	20.3	20.8	19.6	2	2	2	0
France	22.8	23.1	23.5	3	3	3	0
Finland	28.4	26.5	29.7	4	4	4	0
Denmark	33.5	34.9	33.5	5	5	5	0
Slovenia	53.7	45.3	63.3	6	6	6	0
Sweden	56.9	52.3	66.1	7	7	8	−1
Estonia	64.5	59.8	66.2	8	9	9	0
Netherlands	68.0	71.0	64.8	9	13	7	6
Lithuania	68.8	56.5	80.9	10	8	14	−6
Ireland	72.6a	65.7b	79.4	11	10	13	−3
Austria	77.3	72.8	79.4	12	15	12	3
Latvia	78.1	65.8	85.3	13	11	16	−5
Spain	78.4	71.1	86.1	14	14	18	−4
Portugal	79.3	80.4	79.3	15	17	11	6
Britain	80.0	76.7	82.9	16	16	15	1
Belgium	81.0	81.9	79.2	17	20	10	10
Romania	82.8	67.3	91.6	18	12	20	−8
Poland	83.6	80.8	85.6	19	18	17	1
Bulgaria	85.0	81.5	87.8	20	19	19	0
Czechia	91.8	91.2	92.0	21	21	21	0
Hungary	93.1	92.1	94.6	22	22	22	0
Slovakia	95.6	93.1	96.4	23	23	23	0

Notes: ^a 2008–2017 average, ^b 2008–2012 average.

Source: calculated by author from data available at Eurostat (2020c).

Table 5.8 Index of innovation in the European automotive industry by country, 2003–2017

	Average 2003–2017 (%)	Average 2003–2007 (%)	Average 2013–2017 (%)	Rank 2003–2017	Rank 2003–2007	Rank 2013–2017	Change in rank between 2003–2007 and 2013–2017
Sweden	97.6	97.9	98.9	1	1	1	0
Germany	88.4	85.2	87.6	2	2	2	0
Austria	62.0	54.2	64.0	3	3	3	0
Britain	50.8	38.5	62.0	4	7	4	3
France	47.7	48.1	41.8	5	4	7	−3
Italy	46.6	38.6	52.7	6	6	6	0
Netherlands	44.0	39.9	53.3	7	5	5	0
Finland	28.3	26.8	29.8	8	8	8	0
Slovenia	21.3	8.2	26.4	9	17	9	8
Portugal	18.8	13.4	17.1	10	14	14	0
Spain	18.7	14.4	20.8	11	13	11	2
Czechia	18.4	25.3	14.3	12	9	16	−7
Denmark	18.1	9.8	22.2	13	16	10	6
Ireland	16.0a	15.8b	16.2	14	11	15	−4
Lithuania	15.3	8.0	18.8	15	19	13	6
Belgium	14.9	10.3	19.6	16	15	12	3
Estonia	12.5	19.1	8.8	17	10	20	−10
Hungary	11.1	8.0	13.0	18	18	17	1
Romania	10.5	15.0	8.5	19	12	21	−9
Latvia	10.0	6.8	11.5	20	20	18	2
Poland	7.1	5.1	11.3	21	21	19	2
Slovakia	4.2	3.2	6.7	22	22	22	0
Bulgaria	1.1	0.0	3.2	23	23	23	0

Notes: ^a 2008-2017 average, ^b 2008-2012 average.

Source: calculated by author from data available at Eurostat (2020d), Statistics Sweden (2020).

The lower-order core position of France is based on its second-strongest positional power, the third-lowest degree of foreign control and the fifth-strongest innovation index. The relative position of France weakened between 2003 and 2017 due to the relative decline of the French automotive industry since the second half of the 2000s (Reference Pardi, Covarrubias and Ramírez PerezPardi, 2020). France’s relative position also worsened in automotive innovation due to the partial relocation of automotive R&D abroad. Renault Technology Romania was opened in 2007 and it has employed 2,300 engineers at three sites in Romania who, in addition to providing technical support for Renault’s factories in Eastern Europe, Turkey and North Africa, develop and test vehicles on the M0 platform, which was previously done in France (Reference Benadbdejlil, Lung and PiveteauBenadbdejlil et al., 2017). Similarly, the Kwid had been the first Renault model that was completely designed abroad (in India) instead of the corporate R&D center in France (Reference Midler, Jullien and LungMidler et al., 2017). Consequently, despite the fact that French automakers continue to conduct the most important automotive R&D in France, the R&D’s share of total business expenditures and employment has declined in France.

Italy’s automotive industry power has been the weakest of the three stable core countries because of Italy’s weaker average positional power compared not only to France but also Belgium, Britain and Spain. Its car production halved after 2000 (Reference Calabrese, Covarrubias and Ramírez PerezCalabrese, 2020), weakening its positional power (Table 5.2). At the same time, Italy’s index of foreign control and index of innovation are similar to those of France. The second-lowest index of foreign control therefore differentiates Italy from unstable core countries and is the basis of its stable lower-order core position (Table 5.7).

5.4.2 Semiperipheral Countries

The semiperiphery is an intermediate spatial zone that is geographically concentrated in Western Europe and is mainly distinguished by a high degree of foreign control, weaker positional power than Germany and France and variable strength of innovation activities (Tables 5.1–5.5, Figures 5.1 and 5.2).

5.4.3 Peripheral Countries

The cluster analysis delimited two clusters that are classified as the periphery and lower-order periphery. With the exception of Portugal and Ireland, the automotive industry periphery is located in Eastern Europe and is typified by the highest degree of foreign control, the lowest innovation index and mostly low positional power. Due to the rapid growth of the FDI-driven export-oriented automotive industry (e.g., Reference PavlínekPavlínek, 2017a), all Eastern European countries, with the exception of the Baltic countries, improved their positional power. However, the relative ranking of the most rapidly growing Eastern European countries worsened in innovation activities as the increase in production and trade was much faster than the increases in R&D expenditures and employment (Reference PavlínekPavlínek, 2012). The index of foreign control increased in all Eastern European countries but most in those with the largest and fastest-growing automotive industries. Eastern Europe thus recorded the highest degree of foreign control in the automotive industry, which underscores its peripheral position.