Firm Size and Regional Linkages. A Typology of Manufacturing Establishments in Southern Spain

For Peer Review Only Firm Size and Regional Linkages. A Typology of Manufacturing Establishments in Southern Spain Journal: Regional Studies Manuscript ID: CRES-2005-0132.R2 Manuscript Type: Main Section JEL codes: D57 - Input–Output Analysis < D5 - General Equilibrium and Disequilibrium < D - Microeconomics, L25 - Firm Size and Performance < L2 - Firm Objectives, Organization, and Behavior < L - Industrial Organization, R15 - Econometric and Input–Output Models|Other Models < R1 - General Regional Economics < R - Urban, Rural, and Regional Economics Keywords: SME, firm size, input-output, large enterprises, linkages, microenterprises http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies For Peer Review Only 1 FIRM SIZE AND REGIONAL LINKAGES. A TYPOLOGY OF MANUFACTURING ESTABLISHMENTS IN SOUTHERN SPAIN Isidoro Romero Department of Applied Economics I. University of Seville (Spain). School of Economics and Business, Ramón y Cajal Avenue, 1. E-41018 Seville. [email protected] Francisco Javier Santos Department of Applied Economics I. University of Seville (Spain). School of Economics and Business, Ramón y Cajal Avenue, 1. E-41018 Seville. fjsanto[email protected] The authors are sincerely grateful to Erik Dietzenbacher and the anonymous referees for their helpful comments. Special thanks as well to the Institute of Statistics of Andalusia for supplying the data used in this study. Regions with different levels of development are characterized by different enterprise compositions according to firm size and regional productive linkages. Using these two variables and two additional ones -the technological level and the position in the value chain-, the composition of the industrial sector in any region can be studied. In this respect, a new theoretical enterprise typology is proposed in this paper as a powerful analytical tool. Furthermore, an empirical work is carried out using a data set from the survey done to estimate the regional input-output table of Andalusia, a backward region in southern Spain. Thus, the theoretical enterprise typology is applied to the industry in Page 1 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 2 Andalusia, so that some strengths and weaknesses of the regional economy are identified. Key words: input-output; forward and backward linkages; firm size; SME; large enterprises; microenterprises. JEC: L25, D57, R15. 1. INTRODUCTION Productive linkages between firms and sectors within a specific spatial area have always been considered a source of static and dynamic externalities which encourage economic growth. These linkages can be studied at different levels of aggregation, depending on which unit of analysis is chosen: the activity sector or the firm. In both cases, the use of input-output models and accounts represents a powerful methodological approach. In the first case, there is a vast literature conceived at meso-economic level, as a focus on sectors and inter-sectoral linkages. In this respect, since the seminal works of Leontief, economists have extensively employed input-output techniques to study the production structures of national or regional economies (RASMUSSEN, 1956; FESER and BERGMAN, 2000; DIETZENBACHER et al., 2005). In the second case, many authors have studied linkages from a microeconomic point of view, considering buyer-supplier relationships between firms located within a territory, generally a local area. The whole literature regarding industrial districts or local production systems could be included in this group (MARSHALL, 1890; BECATTINI, 1979). Also, at this level, inputoutput models have been developed to study physical or monetary flows within an industrial district (see, for instance, ALBINO et al., 2003). Finally, many Page 2 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 3 works consider the internal transactions among different departments or units within an individual firm such as intra-firm linkages and apply enterprise inputoutput models and accounts to its analysis (see LIN and POLENSKE, 1998 for a review). In this paper we propose a new analytical framework combining these levels of analysis to study the relationship between firm size and the spatial structure of linkages. In this respect, it is frequently believed that SMEs, in comparison to large enterprises, have stronger linkages with local or regional suppliers and clients. Nevertheless, the internationalization process of SMEs and the changes in large firms’ organization in recent decades might be modifying these traditional assumptions. In any case, the relationship between firm size and regional linkages is more complex than what these generalized statements show. In order to more thoroughly explore this issue, we propose an original enterprise typology based on the spatial structure of inter-firm linkages, which distinguishes different enterprise models with different roles within the regional productive system. This typology can be a powerful tool to identify the strengths and weaknesses of a regional economy. In addition, we carry out an empirical analysis, using a data set of 29 manufacturing sectors from the regional input-output table of Andalusia (IEA, 1999), a backward region in southern Spain. With respect to this, the novelty of the approach lies in the fact that we do not work with data for sectors, as inputoutput tables traditionally do, but with data disaggregated into groups of firms within each sector according to their size, in terms of number of employees. This disaggregated data allows us to illustrate our methodology, analysing the productive structure of industry in Andalusia. Page 3 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 4 The paper is organized as follows. Section 2 reviews the literature considering the importance of productive linkages and the impact of firm size on them. In Section 3, we propose our theoretical enterprise typology. In Section 4, we apply the enterprise typology proposed in Section 3 to the industrial sector in Andalusia. Finally, Section 5 concludes the paper. 2. FIRM BEHAVIOUR AND PRODUCTIVE LINKAGES. A LITERATURE REVIEW Many authors have stressed the importance of linkages between sectors and firms as a source of externalities which improve economic performance (MARSHALL, 1890; HIRSCHMAN, 1958; LEONTIEF, 1966). Furthermore, the impact of the linkages has been reconsidered more recently from several economic streams. 2.1. Rethinking the linkages Firstly, the endogenous growth models have emphasized the role of certain dynamic externalities –the technological spilloversas a source of increasing returns. In these growth models, though the production of individual firms shows constant returns to scale, there are socially increasing returns due to the external effects arising from the individual firms’ investment, especially in technological and human capital (ROMER 1986; ROMER, 1990). Thus, the diffusion of knowledge resulting from the firm’s investment increases the efficiency in the whole economy. Regarding this, considerable empirical evidence supports the hypothesis that these effects appear more frequently in diversified productive systems due to a cross-fertilization among sectors (GLAESER et al., 1992; HENDERSON, 1994). Hence, it is reasonable to Page 4 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 5 believe that productive linkages are a probable route for these external effects. In this way, technological spillovers have been studied, in one of the methodological alternatives developed, applying input-output tables (VERSPAGEN and DE LOO, 1999; DIETZENBACHER and LOS, 2002). Although productive linkages have always been present in spatial economics (see, for instance, STREIT, 1969), the ‘New Economic Geography’ models have recovered some former ideas of Marshall and Hirschman with a new perspective. In this sense, many authors have introduced classical backward and forward linkages in formalized models as relevant factors in the spatial dynamic of the productive activity (KRUGMAN, 1991; FUJITA et al., 1999). Thus, productive linkages are viewed as a type of externality that can lead to agglomeration economies when they are subject to a spatial constraint (PARR, 2002; PARR et. al., 2002).1 Finally, the economical impact of the externalities associated with the linkages has been broadly considered in the ‘Endogenous Development’ literature, which defends a development strategy from below based on the local assets and entrepreneurial initiatives (GAROFOLI, 1992; VÁZQUEZ, 2002). Regarding this, the literature about industrial districts, local systems of firms and clusters has stressed the role of linkages and cooperation between firms within an area (BECATTINI, 1979; PYKE et al., 1992; MARKUSEN, 1996; PORTER 1998; GUERRIERI et. al., 2001).This authors has also paid attention to the local production models organised around large enterprises, pointing out the importance of linkages between large firms and indigenous SMEs that act as their suppliers (VÁZQUEZ, 1999; RUTHERFORD, 2000). Page 5 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 6 2.2. Firm size, productive linkages and enterprise models In comparison to large enterprises, it is frequently supposed that SMEs show a higher level of territorial integration and maintain stronger linkages with local or regional suppliers and clients (FLORIO, 1996, CRONE and WATTS, 2000; GÖRG and RUANE, 2001). Nevertheless, you can also often find dynamic SMEs, which operate with external suppliers and export a high proportion of their production, actively participating in the globalization process (ACS and YEUNG, 1999; AUDRETSCH, 2003; ETEMAD and WRIGHT, 2003; JONES and DIMITRATOS, 2004). On the contrary, it is commonly believed that large enterprises frequently have more intense linkages with productive units located outside the region than with firms operating in the same area. The extreme case of this productive behaviour is the model of ‘industrial enclave’. The enclave industry operates in a backward territory developing specific phases of a complex production process. The intermediate inputs required are purchased externally and the output is used in subsequent manufacturing activities in plants located in other territories. Normally, this type of firm features a limited economic impact in the area in the short and in the long run, and it even might distort the resource allocation at local or regional level (DELGADO, 1981; BARROW and HALL, 1995; HARDY, 1998; O’HEARN, 2001). However, changes in the organization of large enterprises in recent decades have led them to give up the functional strategies, according to which the territory was seen as a mere physical base of the firms’ plants. On the contrary, large corporations in some sectors are adopting territorial strategies so that, nowadays, they often procure to integrate themselves into local and Page 6 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 7 regional environments. This transformation in large enterprises has stimulated the decentralization of non-nuclear activities towards SMEs and the intensification of the linkages with local SME clusters. Large corporations in some sectors thus tend to set up their establishments in dynamic areas where they can exploit external economies and benefit from high quality resources and the presence of specialized suppliers. This model of a large enterprise that is more integrated in the territory has become increasingly more important in recent decades, especially in some high-tech and knowledge-intensive sectors. This type of company participates more actively in the technical, social and economic interactions within the territory where they operate, and their forward or backward linkages with other firms within the area can be a source of static and dynamic externalities (PIORE and SABEL, 1984; BARROW and HALL, 1995; BELLANDI, 2001; VÁZQUEZ, 2002).2 To conclude, forward and backward linkages between firms and industries have important effects on the localization and efficiency of productive activity. Although multiple factors play a role, the spatial structure of linkages is finally determined by firms’ decisions. Since firm size conditions the firm’s strategic policy, a different composition of the productive system in terms of firm size will lead to different spatial patterns of linkages. Furthermore, since linkages represent economies external to the firm, there is no guarantee that the optimum for each enterprise would lead to an optimum for the whole productive system in a region. Consequently, it is interesting to analyse the regional structure of productive linkages and to check whether potential linkages are being well-exploited or not, considering the consequences for the regional economy in terms of efficiency. For this purpose, it would be Page 7 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 8 convenient to have at our disposal an exhaustive typology of enterprises as an analytical tool. With this in mind, on the one hand, we can find in the literature some examples of firm typologies which take into account the implications for regional development, though they are not strictly focused on productive linkages (see, for instance, Young et al., 1994, regarding multinational enterprises, or Cooke et. al., 2005, for SMEs). On the other hand, as we have previously pointed out, certain authors have identified some models of firms characterized by specific spatial patterns of linkages (for example, PIORE and SABEL, 1984; BARROW and HALL, 1995; HARDY, 1998; BELLANDI, 2001) However, they do not provide a systematic and global framework leading to a full enterprise typology. 3. FIRM SIZE AND SPATIAL PATTERNS OF LINKAGES. AN ENTERPRISE TYPOLOGY In this paper, we propose an enterprise typology which identifies diverse patterns of firm behaviour within a regional productive system. The typology is based on two criteria: the location of suppliers and the location of sales markets. According to these variables, we can differentiate seven types or models of enterprises: the “domestic” firm, the “exporting” SME, the “dependent” SME, the “extravert” SME, the large “propelling” firm, the large “enclave” firm, and the large “market-orientated” firm. This classification can be represented graphically on an axis system (Figure 1), considering the localization of the suppliers (regional or external) on the abscissas axe, and the geographic orientation of sales (regional or external) on the ordinates axe.3 Page 8 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 15 million inhabitants located in southern Spain and it is one of the most backward regions in the country and in the EU-15, such as its low GDP per capita, high unemployment rate and high rate of labour in agriculture show (Table 2). According to the size distribution of firms, Andalusia also has other important comparative weaknesses regarding the EU-15 and even Spain. Mainly, as can be seen in Table 3, the percentage of microentreprises is higher than the average of the EU-15 and Spain as a whole and, likewise, the percentage of large firms is lower. Moreover, main activities are low-technological in industry and low knowledge intensive in services in comparison to the EU-15 and Spain. INSERT TABLE 2 AROUND HERE INSERT TABLE 3 AROUND HERE In this section, the empirical objective will be to study the composition of the industrial sector in Andalusia applying the enterprise typology proposed in Section 3 to our data set. 4.1. Data and methodology The survey made to estimate the input-output table includes, among other variables, information about firms’ purchases and sales. Thus, we could obtain from the Andalusian Statistical Institute data about the value of the total intermediate inputs purchases, classified according to their regional or imported origin, and the value of regional production, classified according to its destination to the regional or external market. Unfortunately, this data does not refer to single establishments, since the Andalusian Statistical Institute is obliged to withhold this information. Hence, we use data aggregated by size Page 15 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 16 groups –regarding the number of employees- and the sector. This use of the input-output data, as was pointed out in the introduction, represents a novelty, since we do not work with data by sectors, as input-output tables traditionally do, but with data disaggregated into groups of firms within each sector. Specifically, we have data for four size groups within each sector, as follows: establishments from 0 to 5 employees; from 6 to 19 employees; from 20 to 49 employees and from 50 and more employees. Nevertheless, for simplicity’s sake, we will consider the two intermediate size-groups as a single one. Thus, we distinguish three types of firms according to their size: microenterprises (from 0 to 5 employees), SMEs (from 6 to 49 employees), and large firms (from 50 and more).4 The data are available for 29 manufacturing industries. Various alternative indicators have been proposed to measure the interindustry linkages. Recent overviews of the literature can be found in DIETZENBACHER and LAHR (2001), SÁNCHEZ-CHÓLIZ and DUARTE (2003) or CAI and LEUNG (2004). Although our data allow us to split each column of the input-output table into four, we cannot do the same with the rows.5 Consequently, we cannot obtain the Leontief inverse and calculate the indirect effects. Thus, we will only consider the direct effects calculating, for each size group within each sector, the value of two indicators representative of the backward and forward linkages: the ratio of Regional Supplying and ratio of Orientation towards the Regional Market. - The ratio of Regional Supplying (RS) represents, for each size group s in each purchasing sector j, the proportion of regional inputs purchase to total intermediate inputs purchase, as follows: Page 16 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 17 ∑ ∑ i ijs i r ijs js x x RS = (1) in which ijs x indicates the value of the output from sector i (including both regional and imported products6) purchased by the establishments in s size group within sector j; whereas r ijs x exclusively stands for the consumption of inputs produced within the region. This ratio reflects, for each sector, the strength of the regional backward linkages.7 This ratio can be interpreted as a weighted average of the RS values obtained in the individual establishments within each sector and size group, as follows: Let us define ∑ ≡ i ijsjs xIC and ∑ ≡ i r ijs r js xIC (2) and also, ∑ ≡ i ijskjsk xIC and ∑ ≡ i r ijsk r jsk xIC (3) where ijsk x indicates the value of the output from sector i (regional and imported) purchased by each k establishment in the s size group and sector j; whereas r ijs x exclusively stands for the consumption of regional inputs. Then, ∑ k jskjs ICIC ≡ and ∑ k r jsk r js ICIC ≡, (4) and, from (1), (2) and (4) ∑∑ ∑∑          ⋅=           ⋅=== k k jsk jsk jsk kjsk jsk k jsk r jsk k jsk k r jsk js r js js IC IC RS IC IC IC IC IC IC IC IC RS ∑ ∑ (5) Page 17 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 18 Notice that in (5) the values of RS for each firm k are weighted by the participation of the firm’s intermediate purchase in the total input purchase of size group s and sector j. - The ratio of Orientation towards the Regional Market (ORM) represents, for each size group within each sector, the proportion of its production which is sold in the regional market. It is defined as follows: -1= js js js X E ORM (6) in which Ejs stands for the exportation of sector j products by industrial establishments in the s size group, and Xjs for the value of the total production of the group. As you can notice, this indicator doesn’t make a distinction regarding the intermediate or final character of the production. This differentiation, which would enrich the analysis, can not be applied in practice. The estimation of the exportations in the regional input-output table doesn’t allow us to disaggregate the data in this way. Nevertheless, the ratio of Orientation towards the Regional Market approaches the measurement of the forward linkages within the region since it reflects the regional or external orientation of each sector regarding the localization of its markets. The value of this ratio can also be interpreted as a weighted average of the values obtained in the individual establishments, as follows: ( ) ∑ ∑ ∑ - - kjsk jsk k jsk jskjsk k jsk k jskjsk js jsjs js X X X EX X EX X EX ORM           ⋅ − === ∑ ∑ k k jsk jsk jsk X X ORM           =∑ (7) Page 18 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 19 where the weights are the shares of each firm in the production of s size group in sector j. Additionally, we have classified the 28 manufacturing industries according to the two factors considered in Section 3: the position within the value chains and the technological level. On the one hand, we have approached the firms’ situation within the value chains by using as an indicator the ratio of Intermediate Outputs, defined as follows: ii j r ij iEX x IO - = ∑ (8) The denominator indicates the value of the regional production which is commercialized within the regional market, and the numerator gives the part of this production that is used as an intermediate input by other sectors in the region. Consequently, a high value for this indicator means the production of the i sector is mainly applied in the regional market as an intermediate input, whereas a low value indicates it for the most part goes to the final demand. More specifically, we consider a sector to be “basic’’ when the IO ratio is higher than 0.7; “intermediate”, when the IO ratio is between 0.7 and 0.3; and “final”, when the IO ratio is up to 0.3.8 On the other hand, we have separated the manufacturing industries into three groups - high-technological, medium-technological and low-technological industriesconsidering the OECD criteria (OECD, 2005). In this way, we have classified the 28 manufacturing industries in the nine types which appear in Table 1 (the Appendix at the end of the paper shows the whole classification). Page 19 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 20 Next, we have calculated the ratio of Regional Supply and the ratio of Orientation towards the Regional Market for microentreprises, SMEs and large firms in each of the nine types of industries. Since the values of these ratios in each size-group can be interpreted as a weighted average of the ones obtained in the individual establishments -see (5) and (7) in Section 4.1-, we can identify the most representative model in each size-group within each sector. In order to do that, we have considered the value of 50% for both ratios as a threshold to distinguish the different models of firm behaviour.9 Thus, we represent in graphical terms -see Figure 1- the ratio of Regional Supplying on the abscissas axe, the ratio of Orientation towards the Regional Market on the ordinates axe and place the origin of coordinates on the point (50, 50). 4.2. Results The results are summarized in Table 4, which shows the predominant firm-class in each size-group and type of industry. INSERT TABLE 4 AROUND HERE Industrial microenterprises in Andalusia are mainly represented by the model of domestic firm, whereas the model of dependent SME is the dominant one among the small and medium industrial establishments, and the large enclave firm preponderates among large enterprises. These three models are the dominant ones within their own size groups for 5 of the 9 types of sectors. The results remain the same if you consider each size-group as a whole without distinguishing between industries (see Table 5). Page 20 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 21 INSERT TABLE 5 AROUND HERE Figure 2 shows the results for basic industries more in detail.10 Both in the low and high technological basic industries, SMEs and large firms fit into the model of extravert SMEs and enclave enterprises, respectively. This implies they do not have important backward and forward linkages with other firms in the region. Specifically, the absence of strong forward linkages poses a serious limitation for the regional economy, since we are considering industries situated at the beginning of the value chains. This is especially important in the basic high technological sector, which is only represented by the manufacture of basic chemicals. This industry has a considerable presence in the regional economy due to the availability of some mineral materials and the existence of oil refineries which provide them with essential inputs. However, the basic chemical industry doesn’t have relevant links with the final chemical industries in the region. In consequence, Andalusia is missing the opportunity of developing a real chemical cluster what would benefit the regional economy. INSERT FIGURE 2 AROUND HERE On the other hand, we can appreciate the existence of strong forward linkages in the case of the medium technological industries, characterized by the presence of dependent SMEs and large market-orientated firms which sell their production in the regional market. This group is made up of some industries that supply inputs to the construction sector. Consequently, in this case, a strong construction cluster exists in Andalusia with a capacity to act as Page 21 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 22 a driving force in the regional economy, as we have pointed out in other works (ROMERO, 2004). INSERT FIGURE 3 AROUND HERE Figure 3 shows the results for the intermediate sectors. In this case, important weaknesses are also detected since the large enclave firms and the dependent SMEs are the dominant type in medium and high technological intermediate industries. As a consequence, some potential forward and, especially, backward linkages have not been exploited. This poses a restriction to the capacity of these activities to propel the regional growth and to increase the productive efficiency through pecuniary and technological externalities. On the other hand, the regional economy shows some strength in intermediate low technological industries in which the dominant enterprise models are the large propelling firm and the domestic SMEs. This positive performance can be mainly attributed to some food industries, especially the olive oil manufacturing which is an important exporting activity in Andalusia. These industries also have important backward linkages with the region’s agriculture. We have studied these productive chains in the agro-food cluster in more detail in DIETZENBACHER et. al. (2005). INSERT FUGURE 4 AROUND HERE To end with, let us discuss the results for final industries (see Figure 4). Firstly, we will consider the high technological group, which, in this case, is Page 22 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 23 formed only by the manufacturing of other chemical products. In this respect, as has been previously said, no relevant backward linkages exist with the basic chemical industry and, consequently, you mainly find dependent SMEs and large enclave firms. Secondly, the medium technological group, which is represented only by the manufacturing of motor vehicles, is characterized by the presence of dependent SMEs and large domestic firms for which Andalusia is the most important sales market. Again, that is not really good news for the regional economy. As these industries are situated at the end of the value chains, there are not many possible forward linkages to develop within the regional economy in these sectors. Thus, a higher exporting ratio in this sector, especially regarding the large enterprises, would have a positive effect on the regional economy so that the potential market for these activities would increase. This consideration is also applicable to the low technological industries, dominated by the presence of domestic SMEs and large marketorientated firms, which in addition do not have important backward linkages with intermediate and basic industries within the regional system. The insufficient presence of exporting SMEs and large propelling firms in these industries, especially in those with a higher technological level, constitutes a weakness of the regional economy. This fact could be attributed to the limitation that size represents for SMEs, but also to the low level of entrepreneurial quality in regionally-owned firms (GUZMÁN and SANTOS, 2001). 5. SUMMARY AND CONCLUSIONS In this paper, we have introduced a theoretical enterprise typology based on the firm size and the firms’ regional linkages which distinguishes seven types Page 23 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 24 or models of enterprises: the “domestic” firm, the “exporting” SME, the “dependent” SME, the “extravert” SME, the large “propelling” firm, the large “enclave” firm, and the large “market-orientated” firm. This typology can be a powerful tool to analyse the composition of the industrial sector in any region and to detect the strengths and weaknesses of a regional economy. Furthermore, we suggest the hypothesis that the industrial sector fits into different patterns of composition, according to our typology, in regions with different levels of development. Regarding this, we have applied our typology in this paper to study industry in Andalusia, a backward region in southern Spain. Thus, we have observed the predominance of dependent SMEs and large enclave firms, and the scarce presence of large propelling firms, exporting SMEs and extravert SMEs in the regional industry. Both groups of facts reveal the weakness of the regional industry. In addition to this, we have noticed an excessive orientation towards the internal market in SMEs and large firms operating in final industries, which limits the growth potential of the regional economy. Furthermore, the fragility of the regional industry is accentuated in the high-tech industries. Large firms in these industries behave as enclaves so that the regional economy doesn’t benefit from the technological spillovers they could originate. This situation can be explained by the insufficient development of these high-tech industries in Andalusia, which leads to the absence of efficient regional and competitive regional suppliers. From this perspective, the case of Andalusia is an example of a more general pattern that would characterize backward regions. On the other hand, according to our hypotheses, the competitive and powerful economies might show an important presence of exporting and Page 24 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 31 KEEBLE, D. (1997) Small firms, innovation and regional development in Britain in the 1990s, Regional Studies, 31 (3), 281-293. KRUGMAN, P. (1991) Increasing returns and economic geography, Journal of Political Economy, 99, 483-499. LIN, X. and POLENSKE, K. R. (1998) Input-output modelling of production processes for business management, Structural Change and Economic Dynamics, 9, 205-226. LEONTIEF, W. (1966) Input-Output Economics, Oxford University Press, New York. MARKUSEN, A. (1996) Sticky places in slippery spaces: a typology of industrial districts, Economic Geography, 72, 293-313. MARSHALL, A. (1890) Principles of Economics, Macmillan, London. OECD (2005) OECD Science, Technology and Industry Scoreboard, OECD. OBSERVATORY OF EUROPEAN SMEs. (2002) SMEs in Europe including a first glance at EU candidate countries. Nº 2. European Commission, Luxemburg. O’HEARN, D. (2001) The Atlantic Economy: Britain, the US and Ireland, Manchester University Press, Manchester. PARR, J. B. (2002) Missing elements in the analysis of agglomeration economies, International Regional Science Review, 25, 151-168. PARR, J. B.; HEWINGS, G. J. D.; JUNGYUL, S. and NAZARA, S. (2002) Agglomeration and trade: Some additional perspectives, Regional Studies, 36 (6), 675-684. PIORE, M. J. and SABEL, C. F. (1984) The Second Industrial Divide – Possibilities for Prosperity, Basic Books, New York. Page 31 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 32 PORTER, M. E. (2003) The economic performance of regions, Regional Studies, 37 (6-7), 549–578. PORTER, M. E. (1998) On Competition, Harvard Business School Publishing, Boston. PYKE, F.; BECATTINI, G. and SENGENBERGER, W. (eds) (1992) Los Distritos Industriales y las Pequeñas Empresas. Distritos Industriales y Cooperación Interempresarial en Italia, Ministerio de Trabajo, Madrid. RASMUSSEN, P. N. (1956) Studies in Intersectoral Relations, North-Holland, Amsterdam. ROMER, P. M. (1990) Endogenous technological change, Journal of Political Economy, 98 (5), 71-102. ROMER, P. M. (1986) Increasing returns and long-run growth, Journal of Political Economy, 94 (5), 1002-1037. ROMERO, I. (2004) Articulación intersectorial y complejos productivos. Un análisis del sistema productivo andaluz, Revista de Estudios Andaluces, 25, 11- 32. ROMERO, I. and SANTOS, F. J. (2006) Dimensión empresarial, encadenamientos productivos y calidad del tejido empresarial. Un análisis de la industria en Andalucía, Revista de Estudios Regionales, 75 (forthcoming). RUTHERFORD, T. (2000) Re-embedding Japanese investment and the restructuring buyer-supplier relations in the Canadian automotive components industry during the 1990s, Regional Studies, 34 (8), 739-751. SANTOS, F. J. (2004) Convergencia, desarrollo y empresarialidad en el proceso de globalización económica, Revista de Economía Mundial, 10/11, 171-202. Page 32 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 33 SÁNCHEZ-CHÓLIZ, J. and DUARTE, R. (2003) Production chains and linkage measures, Economic Systems Research, 15, 481-494. STREIT, M. E. (1969) Spatial associations and economic linkages between industries, Journal of Regional Science, 9 (2), 177-188. VÁZQUEZ, A. (2002) Endogenous Development. Networking, Innovation, Institutions and Cities. Routledge, London. VÁZQUEZ, A. (1999) Inward investment and endogenous development. The convergence of the strategies of large firms and territories?, Entrepreuneurship and Regional Development, 11, 79-93. VERSPAGEN, B. and DE LOO, I. (1999) Technology spillovers between sectors and over time, Technological Forecasting and Social Change, 60, 215- 235. YOUNG, S.; HOOD, N. and PETERS, E. (1994) Multinational Enterprises and Regional Economic Development, Regional Studies, 28 (7), 657-677. INSERT APPENDIX AFTER THE REFERENCIES Page 33 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 34 Appendix. Classification of manufacturing industries Technological level Low technological industries Medium technological industries High technological industries Basic industries Tanning and dressing of leather; manufacture of luggage, handbags, saddlery, harness and footwear Manufacture of wines and alcohol production Manufacture of wood and of products of wood and cork, except furniture Manufacture of paper and paper products Manufacture of rubber and plastics products Manufacture of cement, lime and plaster Manufacture of clay and ceramic products Manufacture of glass and glass products and other non-metallic mineral products Manufacture of basic chemicals Intermediate industries Manufacture of vegetable and animal oils and fats Manufacture of other food products Textile manufacturing Publishing, printing and reproduction of recorded media Metallurgy Manufacture of fabricated metal products Building and repairing of ships and boats Collection, purification and distribution of water Manufacture of machinery and equipment Manufacture of electrical machinery and apparatus Manufacture of radio, television and communication equipment and apparatus Manufacture of medical, precision and optical instruments, watches and clocks Position within the value chains Final industries Production and processing of meat products Processing and tinning of fish and vegetables Manufacture of dairy products Manufacture of bakery products Manufacture of wearing apparel; dressing and dyeing of fur Manufacture of furniture Manufacture of motor vehicles, trailers and semi-trailers Manufacture of other chemical products Source: Elaborated from IEA and OECD. Page 34 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 35 Figure 1. Enterprise typology according to the spatial patterns of productive linkages External sales market Regional suppliers • Exporting SME Large propelling firm Domestic firms (large, small and medium-sized and microentreprises) Large enclave firm Large market-orientated firm Dependent SME Regional sales market External suppliers externos Extravert SME Page 35 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 36 Figure 2. Basic industries: firm size, regional linkages and technological level Large firms MT Micro HT SMEs HT SMEs MT Micro MT Large firms LT Large firms HT SMEs LT Micro LT 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 ORM RS Low technological ind. (LT) High technological ind. (HT) Med ium technological ind. (MT) Page 36 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 37 Figure 3. Intermediate industries: firm size, regional linkages and technological level Large firms MT Micro HT SMEs HT SMEs MT Micro MT Large firms LT Large firms HT SMEs LT Micro LT 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 ORM RS Low technological ind. (LT) High technological ind. (HT) Medium technological ind. (MT) Page 37 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 38 Figure 4. Final industries: firm size, regional linkages and technological level Large firms MT Micro HT SMEs HT SMEs MT Micro MT Large firms LT Large firms HT SMEs LT Micro LT 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 ORM RS Low technological ind. (LT) High technological ind. (HT) Medium technological ind. (MT) Page 38 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 39 Table 1. Sector typology considering main potential externalities Technological level Low technological industries Medium technological industries High technological industries Basic industries Strong push effects Strong push effects Forward technological spillovers Strong push effects Strong forward technological spillovers Intermediate industries Push and pull effects Push and pull effects Technological spillovers Push and pull effects Strong technological spillovers Position within the value chains Final industries Strong pull effects Strong pull effects Backward technological spillovers Strong pull effects Strong backward technological spillovers Page 39 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For Peer Review Only 40 Table 2. Main regional indicators of Andalusia in comparison with the EU-15 and Spain Distribution of labor by sector (%) 2002 GDP p.c. (PPS 2001) EU-15=100 Unemployment Rate 2002 Agric. Indus. Servic. EU-15 100.0 7.8 4.0 28.2 67.7 Spain 84.2 11.4 5.9 31.2 62.9 Andalusia 64.3 19.6 10.3 25.6 64.1 Source : EUROSTAT Page 40 of 43 http://mc.manuscriptcentral.com/cres Email: [email protected] Regional Studies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60