Keywords

5.1 Introduction

The study of keiretsu has interested many scholars and researchers by offering an approach to understanding the complexity of the dynamics that allowed Japan to reach economic prosperity in an adverse environment and to determine which corporate governance characteristics are useful.

In the traditional keiretsu system, participating suppliers and buyers work closely, forming a unique association. This association results in improvements for all the parties involved by acquiring a deeper understanding of the requirements and needs of consumers and adapting their production to meet the customers’ demands. However, keiretsu is more than modifying production practices: it comprises a process of deep trust, collaboration, goodwill, and knowledge exchange among participants; therefore, the buyers usually maintain long relationships with key suppliers.

The term Keiretsu is defined as clusters of independently managed firms intertwined by governance mechanisms such as president’s clubs, partial cross-ownership, and personnel exchanges (Lincoln and Shimotani 2010). Grabowiecki (2012) identifies that the essence of Japanese corporations is the membership of business groups. Keiretsu is a form of organization of a group of firms, where together, they scheme, invest, trade, and cooperate toward their collective self-preservation (Miwa and Ramseyer 2006). In addition, a financial institution participates in the group, supplying various services: the keiretsu group receives financial advice from the bank, credits, and qualified personnel incorporated into their management. By doing so, the keiretsu group shields its stock from foreign or outsider buyers, maintaining the group’s control. Keiretsu typically conduct their business within the firms belonging to the group, keeping finance and trade-related matters internally. Keiretsu controlled the Japanese economy until the first half of the 1990s: Mitsui, Mitsubishi, Sumitomo, Fuji, the Daiichi Kangyo, and Sanwa were the most important groups.

Keiretsu derives from an older form of organization, the zaibatsu. Zaibatsu is an organizational form where multiple firms that participate in different industries are controlled by a family or holding company. The structure of zaibatsu could be either similar to pyramidal business groups or follow a hierarchical system that allowed a small number of Japanese families to gain control over large sectors of the country’s economy by organizing multi-tiered subsidiary businesses (Miyajima and Kawamoto 2010). Lincoln and Shimotani (2010) identify that Mitsui, Mitsubishi, and Sumitomo were the most prominent prewar zaibatsu and that, by 1937, the “big three” controlled about 12% of the total corporate capital in Japan. By the war’s end, the amount had grown to 23%.

The cold war broke out in Europe in 1948, and communism spread throughout Asia and Europe. The United States then viewed Japan as a bridge connecting the communist countries and the United States. The latter stopped the dissolution of zaibatsu and revived zaibatsu groupings to strengthen the economy of Japan (Morikawa 1992). The firms, however, created a network of dispersed companies around banks that allowed them to hold shares of other companies, facilitating the financial linkage within the group, and were called keiretsu. This dispersion (occasioned by the attempt of the government of the United States to eradicate zaibatsu) permeated and defined the new form of organization. For instance, keiretsu is interpreted as the incorporation or composition of “previously individual elements into a whole,” suggesting that companies are joined together loosely (Kensy 2001). Some keiretsu developed from old zaibatsu, but new groups also emerged (Wozniak 2009).

Keiretsu’s popularity peaked during the decade of the 1980s, not only in the manufacturing industry, with the companies associating closely with their buyers but also in business schools, where they sparked an interest in understanding the traditional keiretsu system. In the meantime, another trend rapidly gained popularity in the Western world: the manufacturers now focused on the prices due to the rise of the emerging economies offering low-cost production due to their low wages. It is considered that keiretsu died when Japanese producers started to cut costs in their manufacturing processes. According to Aoki and Lennerfors (2013a), it was later restored and reinvented by companies in the Japanese automotive industry by globalizing their supplier ties, opening to new markets, and becoming more aware of their costs.

The new keiretsu survived a crisis that affected the Japanese automotive industry during the decade of the 1990s when their competitors started programs aimed to cut costs drastically and turned to mega-suppliers. At the same time, the Japanese economy entered a period of stagnation followed by the burst of the economic bubble, forcing the Japanese automakers to allow foreign investors into their otherwise closed keiretsu groups to cut costs and maintain their competitiveness in the automotive industry (Aoki and Lennerfors 2013a).

On the other hand, Mexico is a recipient economy of FDI, for which a few statistical studies have been conducted regarding location factors. Some studies identify low wages, the size of the Mexican market, proximity to the northern countries, and the signing of the NAFTA trade agreement as reasons for firms’ location in the country (Blomström and Kokko 1997; Love and Lage-Hidalgo 2000). However, empirical studies that seek to determine location factors and regional distribution within the country are even scarcer (Jordaan 2012).

For the case of Mexico, the previous literature has focused on macro and regional-level factors in explaining the distribution of FDI. Lichtensztejn (2014), at a national level, finds that FDI flows are positively correlated with GDP levels and openness to free trade while it is negatively correlated with FDI restrictions and crime levels. Jordaan (2009), on the other hand, using state-level data, finds that demand, wages, schooling, labor quality, regional market location, and agglomeration economies explain the regional distribution of FDI.

A significant limitation in the previous literature is that they fail to analyze the specific case of Japanese FDI in Mexico using disaggregated data and analyzing certain factors that seem to influence the location decision of Japanese investors: Public policies, agglomeration economies, and keiretsu-type organization preference. The study is organized as follows: the discussion of prior theoretical and empirical research on the keiretsu as a factor for FDI location follows. The empirical model and data sources are presented in the third section. Results and conclusions are shown in Sects. 5.4 and 5.5, respectively.

5.2 Literature Review

In terms of the role of the keiretsu in the location decision of Japanese investment, studies have found that this type of organization facilitates agglomeration and creates positive externalities for the participating firms. Also, this seems to be more robust for the automotive industry (Head and Ries 1996; Smith and Florida 1994; Mayer and Mucchielli 1998; Huallacháin and Reid 1997; Belderbos and Carree 2002).

Belderbos and Carree (2002) find that keiretsu plant agglomeration is present in the automotive and electronics industry in the United States, and this type of organization extends to developing countries. The study identifies leading “core” firms and other member firms of the group. It confirms an impact from agglomeration effects and other regional factors that differ depending on market orientation and firm size.

According to Hackett and Srinivasan (1998), the keiretsu influence on the location decision of firms is related to the costs involved in searching for and switching to suppliers of parts and components that are not part of the keiretsu. The latter incentivizes group member firms to replicate existing supplier relationships abroad. Belderbos and Carree (2002) highlight that this is observed in the vertical type of keiretsu, characterized by an intensive inter-flow of information among firms, where the core firm assists member firms in relocating to the host country. In this case, the study finds that a “core” firm locates abroad, searching for appropriate locations, and the member firms follow the core firm’s lead. Head et al. (1995) similarly highlight the importance of keiretsu in the location decision of Japanese firms.

The empirical literature highlights the difficulties in testing the keiretsu hypothesis. Japanese firms do not publicly inform about their affiliation to a keiretsu; the information available is for the case of firms established in Japan and not generalizable to affiliate firms abroad. Chavarro-Jiménez and Guzmán-Anaya (2018) attempt to test the keiretsu hypothesis for Japanese firms in Mexico but fail to identify the supplier affiliation to a keiretsu group and suggest this analysis for future empirical work.

Kito et al. (2014) study the Toyota supplier network and identify critical structural features using measures from Social Network Analysis and network science. The study identifies key firms and reveals the overall heterogeneity composition of the network. The study also shows the presence of constituent sub-networks where their structure depends on factors such as product categorization, geographical closeness, and business alignment. Although the study reveals essential aspects in forming production networks, it focuses the keiretsu analysis and discussion solely on the Toyota case.

It is important to point out that the relationship between the Japanese car assembler and its suppliers is organized in the “vertical type” of keiretsu, where a company interacts with its suppliers in the tier directly below them. However, some authors question the keiretsu structure as a Japanese production system and suggest that this is only particular to certain firms (Miwa and Ramseyer 2006). Interestingly, authors such as Nishiguchi (1994) and Nobeoka (1997) argue that the keiretsu is a more complicated system of interlinked Keiretsu pyramids where suppliers are tied to more than one automotive assembler which results in an “alpine” structure. These claims are supported by Kito et al. (2014), showing the complexity and intertwined nature of keiretsu networks.

Another body of literature has focused on the economic performance of firms that constitute the keiretsu. Ahmadjian (1997) argues that automotive suppliers that link a large proportion of their output to a single customer perform similarly to more independent ones. The economic benefits of belonging to a keiretsu group seem related to shields against exogenous shocks and redistribution of profits between high and low-performing firms. The redistribution appears to be related to encouraging suppliers to make customer-specific investments. The authors mention that this is also associated with Japan’s cultural, social, and institutional context. On the other hand, Sambharya and Banerji (2006) note that lower levels of supplier-client dependency in automotive firms positively affected supplier-firm performance. Furthermore, the keiretsu affiliation was not significant in supplier-firm performance.

In Japan, business groups are called keiretsu, a unique network organization that gathers or, more accurately, “cluster” industrial and financial corporations. The definition of keiretsu in economic literature often stems from categorization designed for American or European organizations to define their Japanese counterparts. However, the latter may lead to erroneous evaluations and conclusions.

Empirical studies on the benefits or detriments of keiretsu have been conducted, with mixed conclusions. For instance, Calder (1989) affirmed that keiretsu was a key factor for Japan to achieve fast industrial development and economic growth since the beginning of the 1950s. The role of keiretsu after the second world war was crucial to Japan’s economy: according to Grabowiecki (2012), keiretsu was a critical factor in achieving economic and infrastructure growth in addition to closing the productivity and technological gap between Japan and other developed nations; Kensy (2001) found that the nine largest trading companies are responsible for more than 50% of the countries total exports; of all the groups, only Mitsui, Mitsubishi, Sumitomo, Fuyo, DKB and, Sanwa cluster about 650 firms that employ over 2 million people, controlling 20% of corporate profits and more than 2% of world production during the same period.

For some authors, keeping all the trade and financial activities inside the group results in more efficient firms (Miwa and Ramseyer 2006). The long-term orientation of the groups allows for data accumulation and better decision-making processes that significantly reduce their risks (Womack et al. 1990). The collaboration goes beyond trade: production and research of keiretsu members represent an advantage against their competitors that are not affiliated with any group. In addition, the strategy followed by Japanese companies belonging to a keiretsu group positions those firms in the adaptability field, unparalleled.

Another benefit related to keiretsu in its vertical form comes from the close relationship between the network members: when companies work and cooperate closely, technology transfers occur. According to Wachowska (2012), knowledge spillovers happen naturally because suppliers are integrated with development projects for larger companies. The author finds positive and significant effects on productivity and innovation for both participants in the network.

On the other hand, Paul Samuelson identifies keiretsu as simply “cheating” because the groups follow the structure of large oligopolies (Samuelson 2000). Furthermore, Bader (1994) mentions that keiretsu allegedly incur in international “predatory pricing schemes,” pushing competitors out of markets and allowing keiretsu to form monopolies and profit from higher prices. Keiretsu groups have also been accused of unfair business practices and, by working almost exclusively with network members, have effectively blocked foreign competitors from accessing Japan’s market (Wachowska 2012). Lincoln and Gerlach (2004) find that the horizontal keiretsu effects are not significant and that the big six groups behave as scheming oligopolists that extract monopoly rents from companies that do not belong to the keiretsu group. Caves and Uekusa (1976) argue that the Keiretsu group’s ratio of prices—opportunity costs is very efficient for the group members; however, outsiders are subject to profit maximization by the dominance of the keiretsu members across different markets. Kimino et al. (2012) conducted a study to evaluate the effects of keiretsu on inward FDI in Japan and examine horizontal and vertical networks to determine how they influence the decision of foreign companies to enter the market in Japan. The authors do not find evidence of keiretsu as an element in discouraging FDI to Japan.

The classification most frequently used for keiretsu is horizontal or kinyu and vertical or shihon. Horizontal keiretsu refers to non-hierarchical groups of prominent companies that originated and derived from the original zaibatsu, which are linked to a shared central bank due to its credit associations that allows it to benefit other members of the group with preferential treatment. For instance, Mitsui, Mitsubishi, Sumitomo, and Yasuda are considered horizontal keiretsu because they enclose important banks.

Vertical keiretsu works with major manufacturers instead: they are networks of subsidiaries that operate within larger corporations and are subordinate to them by long-term production–distribution relationships and capital (for example, Matsushita, Toyota, Toshiba, and Hitachi). As Grabowiecki (2012) points out, the vertical–horizontal division does not entirely reflect the complexity of the relationships conducted in both types of networks.

Therefore, this keiretsu type is a group of manufacturers or supply chains consists of suppliers, distributors, and subcontractors that follow a vertical division of labor gathered around industrial firms. In turn, the suppliers participate actively in the design and development of products and processes; thus, it is recognized as a model of manufacturing best practice (Lincoln and Gerlach 2004). Research suggests that the close cooperative ties that characterize vertical keiretsu allow knowledge transfer to affiliated companies.

Both forms of keiretsu have evolved and even survived globalization and economic changes. Dow et al. (2011) studied the evolution of vertical and horizontal keiretsu, measuring the stability of the networks through economic and competitive phases across time. In the case of vertical keiretsu, economic recessions translated into weak linkages; on the other hand, horizontal keiretsu ties were reinforced, showing more stability in times of crisis.

5.2.1 Keiretsu in the Automotive Industry

In the automotive industry, the structure followed by keiretsu is pyramidal. It comprises suppliers organized as follows: the first-tier suppliers provide services/products to the manufacturer, the second tier works for the first tier, and the third for the second. The assembler occupies the top of the pyramid, and the levels below comprise hundreds or sometimes thousands of suppliers (firms).

The studies that have empirically analyzed the keiretsu in the automotive industry are Shimokawa (1985), where the author finds different types of keiretsu type of organizations within the Japanese automotive industry. The author explains the origin of vertical integration in the sector: as a result of the initial low technological level in Japan, limited and dated processes of manufacturing (casting, plating, smelting, among others) by one producer prompted to create a system that would guarantee the on-time supply of quality parts from specialized producers. As a form of organization, keiretsu varies within the Japanese automotive industry in terms of flexibility or rigidity of the principle of market competition. It means that, within the keiretsu, the free entry or exit of producers to the supply chain is limited for the primary parts producers and less rigid for secondary or tertiary part producers. As a result, the author finds that keiretsu works as an intermediary between companies and markets.

Solis (2003) analyzes the relations of keiretsu in three countries: Japan, Mexico, and the United States. In the case of the automotive industry, keiretsu relations are positive. However, this is not the case for the electronics industry, which suffered sluggishness regardless of the keiretsu type of organization. It means that even under the organizational principles of keiretsu, the differences between industries can influence the outcome of the sectors. Dyer (1996) compares the case of the automotive industry in Japan versus the United States and finds that keiretsu alliances translate into advantages in the auto industry by seizing the benefits of inter-firm asset co-specialization. Still, those alliances may result in liabilities for other sectors where the networks and coordination are less complex.

Liker and Choi (2004) compare keiretsu relations for Toyota and Honda companies, contrasting the ties between these companies and their suppliers compared to the relationships that US companies in the automotive industry have with their suppliers. The authors found that keiretsu positively impacts its suppliers, which translates into quality improvements in production and reduces the delays and production times of the inputs. Ahmadjian and Lincoln (2001) argue that keiretsu governance forms are evolving from a network type to arms-length and top-to-down management. The authors analyze the changes in the administration of Toyota with two of their suppliers: Daihatsu and Denso. The supply network implemented for both cases has a hierarchical form of management; however, in the case of Daihatsu, Toyota adopted its transactions, whereas its long-term partner Denso reduced its dependence by increasing Toyota’s know-how on the production of electronic components. Aoki and Lennerfors (2013b) analyze the Toyota companies established in different countries and find that the company manages to establish relations of keiretsu type outside Japan. In addition, in a subsequent study, the authors analyze the vertical kind of keiretsu for 1991–2011 of Toyota, Nissan, and Honda, assuming the hypothetical disappearance of this form of organization. They found that far from disappearing, keiretsu has remained and strengthened even further, evolving toward hybrid and new forms of governance.

Finally, Kato et al. (2016) analyze the benefits of belonging to keiretsu for the case of Tier 1 and Tier 2 suppliers in the automotive industry. According to their study, the Tier 1 suppliers belonging to keiretsu have few benefits to being integrated into these production chains; this is similar for Tier 2 suppliers. For them, belonging to keiretsu does not translate into competitiveness improvement. However, independent Tier 2 suppliers have the propensity to become more competitive.

5.3 Empirical Model

The importance of FDI location determinants was recognized in the 1960s, especially in locating US companies in developed nations, which is considered a macro perspective of FDI determinant analysis. In the following decade, however, the emphasis switched to a micro perspective, focusing on understanding the reasons at the firm level why companies chose to establish their production in foreign locations instead of exporting their products to those destinations.

Orthodox economic theory cannot be used to examine FDI or MNEs since it assumes market structures with perfect competition, which means that in perfect competition, companies do not have the market strength or attributes that enable MNEs to prosper. As a result, market imperfections provide the ideal setting for MNEs to gain ownership advantages and exploit foreign manufacturing through them.

Theories that explain FDI behavior have been developed since the 1960s. The first approaches to explain FDI were based on the Heckscher-Ohlin and MacDougall (1960) and Kemp (1964) models (Assunção et al. 2011). In these models, FDI was motivated by a combination of low labor costs and exchange risks (favorable circumstances in some developing foreign markets), where higher profits could be made.

The increase in FDI at the world level prompted research on the determinants that explain this type of investment. As a result, in the empirical and theoretical literature, there is a vast catalog of determinants that try to illustrate the direct investment of multinational companies in specific areas. The eclectic or OLI (Ownership, Location, Internalization) paradigm stands out among the models discussed, with an institutional approach and the New Theory of TradeFootnote 1 model.

Following the OLI paradigm, the FDI determinants associated with the location dimension are infrastructure, human capital, economic stability, and production costs. The OLI paradigm of Dunning encompasses both internalization theory and traditional trade theories (Dunning 2002). It systematizes the benefits for foreign enterprises, tying them to the chosen entry options (Faeth 2009). The Dunning model establishes that there will be advantages to selecting the FDI as long as the factors of ownership advantage (O), location advantage (L), and internationalization advantage (I) are met simultaneously. The value of a company possessing assets such as cutting-edge technology, exclusive production techniques, patents, management skills, and other assets that can bring profits in the future is referred to as ownership advantage (Dunning and Lundan 2008). Location is significant when a company benefits from its presence in a specific market by taking advantage of special tax regimes, lower manufacturing and transportation costs, market size, access to protected markets, and lower risk. Internationalizing activities, for example, helps eliminate market failures such as the imbalance of international resource allocation, lowering transaction costs, and reducing the danger of copying technologies. As a result, selecting a specific place depends on unique factors that favor it (Ietto-Gillies 2005).

Dunning’s eclectic paradigm significantly contributed to the literature by combining numerous complementary theories and establishing a collection of characteristics (ownership, location, and internalization) that affect multinational corporations’ operations.

The key to this perspective is the application of these variables to commerce, international production, and international production organization, implying that the same analytical framework may cover the three primary modes of internationalization (exports, FDI, and licensing) (Assunção et al. 2013). Under this perspective, the importance of the keiretsu can be associated with the ideas discussed by the eclectic paradigm.

5.4 Empirical Framework

5.4.1 Network Analysis

Social Network Analysis (SNA) allows the quantitative measurement and visualization of relationships between actors (individuals, groups, organizations) by modeling these relationships and analyzing the links between them. This methodology helps to spatially represent these actors within a network, where the distance between them represents their similarities or differences. Furthermore, the SNA considers the network as a whole; analyzing a single node or some nodes lacks interpretative meaning.

Haythornthwaite (1996) argues that SNA may be used to study the exchange of resources among actors (individuals, groups, or organizations) where patterns of resource exchange reveal themselves as social networks, with actors represented as nodes and the exchange relationships as connectors among nodes. Kim (2015) employs SNA for the Korean automotive industry and suggests using centrality measures to interpret the network results and identify key firms. The author concludes that SNA metrics can help understand and analyze the structure of an automotive supply network.

It is necessary to satisfy the following principles to use the SNA methodology: (1) balance: it describes the particular pattern of the symbols of the lines that make up the network; if all symbols are positive, then the network is balanced. If there is a single negative sign, the balance is lost, and the existence or permanence of that link in the network is jeopardized; (2) triad: the presence of fewer than three agents cannot be considered a network (Scott 2000). Furthermore, according to what was proposed by Salas Durazo and Murillo García (2014), for a group of organizations to be considered a network, it is necessary the existence of cooperation links. The latter means that the emergence of connections depends on the same agents. In addition, there must be a mutual agreement between them to create this link. The authors add that because agents seek to satisfy their goals, asymmetries, and centrality can be created in the decision-making process. Salas Durazo and Murillo García (2014) propose a methodology that allows the analysis of micro-interactions within organizations through the relationship of the SRA theory with the homophily theory. The above will enable them to explain the integration and cooperative work between the agents belonging to a network.

Identifying the role of keiretsu in forming production networks is a complicated task. Kito et al. (2014) employ SNA to study Toyota’s production network, identify key firms, and show the overall heterogeneity composition of the network. However, identifying the role of the keiretsu in determining the relationships within the network is difficult since the keiretsu is a more complex system of interconnected relationships where suppliers may be tied to more than one automotive assembler (Nishiguchi 1994; Nobeoka 1997). Tsujimoto et al. (2014) analyze the cooperation between Toyota and Panasonic from the network analysis perspective; the authors find that this cooperation results from adaptive behavior in response to the speed of change in the business environment in the automotive industry. Thus, the study by these authors focuses on the analysis of the networks explicitly formed by the individuals who participated in the collaborative research processes because the formation of this cooperation network was a key factor in the integration and development of new technologies, resulting in spillovers for both companies.

The work of Gerlach (1992) aims to study Japanese cooperation networks using the methodology of Social Network Analysis. The study employs the block model and determines that the financial hierarchy and the position of the companies in the network are essential to identify the agents’ relationships and that the companies’ membership in the economic groups is significant.

According to Salas Durazo and Murillo García (2014), the social network methodology can be used to analyze and identify the factors that trigger cooperation ties at the group and organizational levels. It is novel, then, to analyze from an organizational theory approach how Mexican and Japanese companies, which are the object of our study, are structured and linked to form cooperation networks that allow them to stock up on the necessary inputs for production. This type of analysis is novel in the keiretsu literature and for Japanese automotive supplier networks in Mexico. The present study employs SNA to reveal the composition of the automotive production networks and the keiretsu’s role in forming these networks.

5.5 Results

Data was gathered from the Toyo Keizai Directory 2020 and the National Registry of Foreign Investment from Mexico’s Secretary of Economy for the keiretsu analysis. In total, 122 Japanese automotive supplier firms were identified in Guanajuato. Afterward, the individual connections of each firm with other firms were identified through the Toyo Keizai directory (2014, 2015, 2016 and 2019) and by revising each firm’s website to identify the procurement to other Japanese companies and firms of different nationalities. Additionally, all the firms were georeferenced by confirming their address, identifying their coordinates in Google Maps, and converting them from WGS-84 projection to ITRF-2008 in ArcGIS. In total, 606 connections were discovered for the Japanese automotive firms in Guanajuato.

Finally, the data were analyzed using the Gephi 0.9.2 software. The software allows for different representations of the network. In the first instance, the network was represented under the random layout. The results are presented in Fig. 5.1.

Fig. 5.1
An interconnected network includes names of automobile manufacturers such as Honda, Toyota, Mazda, Ohashi, Ford, Mitsubishi Motors, Fukoku Mexicana, and Yorozu, among others.

Random Social Network Analysis of the Japanese automotive network in Guanajuato. Source: Author’s calculations using Toyo Keizai Shinpo Sha (2020) data and company website information. Notes: The figure was generated using the Geph 0.9.2 software

Full size image

The results elucidate that automotive assemblers are crucial to the formation of the network. Figure 5.1 shows the relationship between Japanese automotive suppliers in Guanajuato and clients located in Guanajuato and other states. The figure highlights the names of those firms with more connections. The Japanese automotive assemblers Toyota and Honda show the largest number of links to automotive suppliers with 46 and 39 connections, respectively, and to a lesser extent, the cases of Mazda (26 connections) and Nissan (25 connections). A key player in the industry is the firm U-shin Autoparts, which has 34 connections and supplies all automotive assemblers included in the analysis. This firm is also a supplier for Ford and General Motors.

A second network representation was under the “Fruchterman Reingold” layout, which stimulates nodes as mass particles and edges as strings. This layout is useful to represent and understand the figure’s topology, where topologically close nodes are placed in the same vicinity, and further connected nodes are placed far from each other. This type of representation facilitates the identification of disconnected nodes. The results in Fig. 5.2 highlight the closeness of the nodes among the Japanese automotive assemblers located in Guanajuato state (Toyota, Mazda, Honda). It also shows that some major Japanese automotive suppliers are topologically located outside of the network (U-Shin Autoparts, Fukoku Mexicana, Katolec, Showa Autoparts, NSK, KYB, Kinugawa, Hiruta). These results indicate that Japanese automotive suppliers are not necessarily linked to one keiretsu and may supply more than one automotive assembler.

Fig. 5.2
An interconnected network includes names of automobile manufacturers such as Honda, Toyota, Mazda, Nissan, U-SHIN Autoparts, Mitsubishi Motors, Fukoku Mexicana, Showa Autoparts, and Yorozu, among others.

Fruchterman Reingold Social Network Analysis of the Japanese automotive network in Guanajuato. Source: Author’s calculations using Toyo Keizai Shinpo Sha (2020) data and company website information. Notes: The figure was generated using the Geph 0.9.2 software

Full size image

The analysis continued with the representation of the network for the automotive assemblers. Figure 5.3 represents the SNA of the Japanese automotive assemblers and their linkages with Japanese automotive suppliers. The network is represented under the “label adjust” function in Gephi. This functionality is a special type of algorithm to avoid label overlapping. The nodes’ size represents the assembler’s importance in terms of the frequency of edges. Something to highlight is the importance of Toyota in the automotive network in Guanajuato compared to the other assemblers (Honda, Nissan, and Mazda). The fact that Toyota was established in Guanajuato in 2019, while Mazda and Honda were established in 2014, increases the relevance of this firm in the network. Toyota registers 46 connections, Honda 39, Mazda 26, and Nissan 25. Even though Nissan plants are located in Aguascalientes and Morelos, they are connected to the automotive suppliers in Guanajuato. The individual keiretsu networks are presented in the next section.

Fig. 5.3
An interconnected network includes names of automobile manufacturers and suppliers such as Honda, Toyota, Mazda, Nissan, U-SHIN Autoparts, Mitsui Kinzoku ACT, Tritech Autoparts, and Furukawa Automotive Systems, among others.

Label adjust Social Network Analysis of the Japanese Automotive assemblers and their linkages to Japanese automotive suppliers in Guanajuato. Source: Author’s calculations using Toyo Keizai Shinpo Sha (2020) data and company website information. Notes: The figure was generated using the Geph 0.9.2 software

Full size image

Figure 5.4 represents Honda’s supplier network in Guanajuato. Honda has two assembly plants in Mexico, one established in El Salto, Jalisco, in 1995 and a second plant in Celaya, Guanajuato, in 2014. Previous studies have noticed that Honda has not developed in a standard Keiretsu form and qualifies instead as a vertical Keiretsu. Honda has deep ties with the financial institutions of the Mitsubishi group but is not considered a member of the Mitsubishi group. The company is believed to rely on trust and goodwill rather than explicit contractual obligations where the development of suppliers is organized and encouraged (Aoki and Lennerfors 2013a). In 2012, Honda started to search for suppliers outside of its group to achieve production growth; however, this rapid expansion brought quality issues incentivizing the company to turn to its keiretsu once again.

Fig. 5.4
An interconnected network has Honda connected to Nippon, Nissen, Furukawa Automotive Systems, Ohashi Technica, Yorozu, U-SHIN Autoparts, and Daikyo Nishikawa, among others.

Random Social Network Analysis of Honda’s automotive network in Guanajuato. Source: Author’s calculations using Toyo Keizai Shinpo Sha (2020) data and company website information. Notes: The figure was generated using the Geph 0.9.2 software

Full size image

Analyzing the SNA of Honda’s suppliers, many supplying firms are shared with other Japanese assemblers. This type of production preference allows the firm to benefit from shared supplier development, and supplying firms are not tied solely to one client. Honda, however, does maintain exclusive supplier relationships. Firms such as Katolec, Hiruta, RDCM, TST Manufacturing, Viscotec, and K-Tech Industrial are identified as unique suppliers to Honda company.

For the case of Mazda, the SNA network is presented in Fig. 5.5. The company began operations in Mexico in 2014. Interestingly, Mazda built a supplier industrial park within the Salamanca Mazda complex. Supplier firms in the supplier industrial park include Y-Tec Keylex Mexico, Daikyo Nishikawa Mexicana, and Aki Seat Manufacturing Mexico. Something observed through the SNA is that although several suppliers are located within the Mazda-built industrial park, this does not limit their dealings with other Japanese companies. For example, Daikyo Nishikawa is also a supplier to Toyota, Nissan, and Honda, and Aki Seat Manufacturing to Toyota. Something to also notice is that Mazda, through a joint venture with Toyota, produced the Yaris-R model in its plant from 2012 to 2020. Some unique suppliers for Mazda include Keylex, located in Mazda’s supplier industrial park, and Kawada, Matsuju, and Tritech Autoparts, located in industrial parks within the region. This finding indicates that geographical closeness to an assembler does not limit the dealings of Japanese suppliers to other Japanese firms. Tagawa et al. (2012) analyze the organizational structure of Mazda’s Keiretsu using the “SNW model” to examine the structural importance of the organization’s network. The findings indicate that Mazda’s Keiretsu structure determines the corporate performance of the firms that conform to the network.

Fig. 5.5
An interconnected network has Mazda connected to Kasai, Hirotec, Y-TEC Corporation, Matsuju, U-SHIN Autoparts, Keylex, Yorozu, Daikyo Nishikawa, Showa Autoparts, and Kinugawa, among others.

Random Social Network Analysis of Mazda’s automotive network in Guanajuato. Source: Author’s calculations using Toyo Keizai Shinpo Sha (2020) data and company website information. Notes: The figure was generated using the Geph 0.9.2 software

Full size image

Nissan has a long history of presence in Mexico. The company began to produce automobiles in the country in 1966 under the company plant name CIVAC. Mexico was the first country to produce Nissan automobiles outside of Japan. The first Nissan plant was built in 1982 in Aguascalientes, and a second plant was built there in 2013. The company does not have a plant presence in Guanajuato state, but from the SNA analysis, it has dealings with firms located in this state.

Nissan has its Keiretsu influence in Aguascalientes state. However, from the SNA, it was noticed that the company does not have unique suppliers in the state; in other words, it shares suppliers with other Japanese assemblers located in Guanajuato. These results show that the Keiretsu organization does have some control over firms that are geographically closely related to the assembly plant. According to Aoki and Lennerfors (2013b), the evolution of Nissan’s Keiretsu was influenced by the arrival of Carlos Ghosn in 1999. His goal to revive the company was accompanied by a cost reduction plan where the multinational sold its holdings in most supplier firms appearing to shift toward a Western model. However, by 2004, Nissan shifted its position with a new purchasing policy that revived the keiretsu. The company invested in one large supplier firm and re-established goodwill and trust as central issues in dealings with vendors. Nissan has established collaboration programs with suppliers, where the firm’s engineers take up process-improvement projects in conjunction with the supplier firm’s engineers. This practice is also present in Honda and Mazda (Fig. 5.6).

Fig. 5.6
An interconnected network has Nissan connected to Showa Autoparts, Saga Tekkosho, Y-TEC Corporation, Nissen, Kasai, Kinugawa, S K-TECH, and N H K, among others.

Random Social Network Analysis of Nissan’s automotive network in Guanajuato. Source: Author’s calculations using Toyo Keizai Shinpo Sha (2020) data and company website information. Notes: The figure was generated using the Geph 0.9.2 software

Full size image

Figure 5.7 shows the Guanajuato automotive supplier network for Toyota assemblers. The company ranked ninth in the Global 500 list in 2015 and was considered the largest company in Japan, with 344,109 employees worldwide. The company started operations in its Guanajuato plant in 2019 with an established production capacity of 100,000 vehicles. The plant produces the Tacoma truck for the North American market. Analyzing the suppliers for Toyota, the company shares suppliers with other Japanese companies and includes suppliers unique to the company. For example, firms such as Fuji Oozx, Corporación Mitsuba, G.S.W., and Tsubakimoto Automotive are the sole suppliers of Toyota. The morphological and functional form of Toyota’s keiretsu supply network is argued to contribute to its competitive advantage (Richter 2000). Kito et al. (2014) claim that the Toyota supplier network follows a barrel-shaped tier structure, not a pyramidal one, as theoretically indicated in the keiretsu literature. This type of structure allows suppliers to locate in different parts of the world and permits them to supply to other clients resulting in a complex and woven network. The results from the SNA analysis indicate that this type of keiretsu network is also present in Mexico, where most firms are suppliers of multiple Japanese assemblers.

Fig. 5.7
An interconnected network has Toyota connected to Mitsui Kinzoku ACT, Imasen, Daikyo Nishikawa, Omega, Hino, Nippon, T T Assembly, Tsubakimoto Automotive, Hirotec, and Furukawa Automotive Systems, among others.

Random Social Network Analysis of Toyota’s automotive network in Guanajuato. Source: Author’s calculations using Toyo Keizai Shinpo Sha (2020) data and company website information. Notes: The figure was generated using the Geph 0.9.2 software

Full size image

5.6 Conclusions

The importance of the keiretsu in the conformation of Japanese supplier networks was conducted using Social Network Analysis (SNA) and data from various sources in Guanajuato. The database comprised 122 Japanese supplier firms with 606 connections identified from the Toyo Keizai Directory 2020 and information from each firm’s website. The results showed that automotive assemblers are key players in forming networks under the keiretsu system. Toyota and Honda register the highest number of connections, with 46 and 39, respectively. From the supplier side, the firm U-shin auto parts are connected to 34 firms in the supplier network in Guanajuato.

The SNA analysis was also conducted using the “Fruchterman Reingold” layout, where the findings indicate that Toyota, Mazda, and Honda are topologically close and share ties among them. Also, major Japanese automotive suppliers such as U-shin auto parts, Fukoku Mexicana, Katolec, Showa Autoparts, NSK, KYB, Kinugawa, and Hiruta are topologically located in the outskirts of the network, indicating that they are not necessarily tied to one keiretsu and may supply more than one automobile supplier.

The SNA analysis was conducted for each automotive assembler to observe the connections with Japanese suppliers. The main finding from this analysis was that Japanese assemblers share supplying firms with other assemblers and maintain exclusive relationships with a relatively low number of companies. This production preference permits firms to benefit from share supplier development, and suppliers are not tied to one client. The results are consistent with Nishiguchi (1994), Nobeoka (1997), and Kito et al. (2014). They claimed that keiretsu is seen as a more complicated system of interlinked Keiretsu pyramids or what was referred to as an “alpine structure” where suppliers are tied to more than one automotive assembler showing the complexity and intertwined nature among keiretsu networks. For example, Mazda built a supplier park within its manufacturing plant complex, and the supplier firms located there supply Mazda and other Japanese assemblers. Furthermore, Mazda produced Toyota automobiles in its plant through a joint venture from 2012 to 2020, highlighting the cooperation structure of the Japanese keiretsu system in Mexico.

The results have shown that Japanese suppliers in Guanajuato seem to benefit from this type of keiretsu organization, where goodwill and trust are centerpieces in their dealings with their clients. Previous literature has also shown that collaboration programs between assemblers and suppliers are present, and future research should analyze these types of programs in the case of Mexico.

The COVID-19 crisis had adverse effects across all industries, and the automotive industry was no exception. At the beginning of the pandemic, due to security measures and to avoid contagion, the assemblers stopped operations. Later, in the automotive industry, the disruption of the production chains was due to the paralysis of mobility and the shortage of domestic and international supplies. Consequently, many car producers were again forced to stop production and close their plants. On the demand side, a drop in demand for new vehicles was observed worldwide.

In addition to the challenges faced by the manufacturers, shortages of critical inputs like semiconductors, high turnover due to sickness absence, and the need to ensure safety and modify or adapt the production area to meet the demands of social distance also caused complications.

The Japanese auto industry experienced a drop in global sales caused by a fall in demand and production. Furthermore, there is uncertainty about future production costs caused by changes in the prices of raw materials, labor, energy, and capital. The latter impacts sales, production, and innovation and represents new technological challenges.

The technological advances of the units suffered setbacks due to the shortage mentioned above, and occasionally, some of the most modern units had to be assembled with parts with technology already exiting the industry. Consequently, car manufacturers have had to postpone the production of new models.

In response, some Japanese companies have initiated joint ventures to become more resilient and overcome the many challenges. The industry has adopted more flexible forms of production and has returned to the “just in time” (JIT) production model.

Concerning productive value chains, Japanese companies have faced high dependence on their parts and components with China, their leading trading partner in this area. One lesson that the problems with supply chains in the automotive industry, derived from the COVID pandemic, have taught us has been the search for solutions for the procurement of inputs. The proximity to the US market places Mexico in a privileged location that favors nearshoring as a possible solution.

Nearshoring is a strategy in which nearby countries with a similar time zone are chosen to transfer part of their production, in contrast to traditional offshoring, whose objective is the search for suppliers that allow saving production costs, usually allocated in Asia.

The pandemic severely impacted supply chains, highlighting how vulnerable offshoring is. Due to favorable circumstances like proximity to the United States, its leading trading partner, the T-MEC that facilitates trade, the performance of the manufacturing industry, and particularly the automotive sector, nearshoring represents an opportunity for Mexico.

There are 114 nearshoring investment projects from October to December 2021 in Mexico, of which 51 are from the United States, 23 are from China, seven are from South Korea, and four are from Japan. Strategies for attracting investments have been implemented to benefit from this trend. For the first time, the governments of Mexico and the United States are collaborating to benefit from nearshoring trends that are advantageous to both countries. The aim is to construct a cutting-edge border to increase competitiveness and attractiveness in the nearshoring process.

In this regard, the manufacturing industry is being encouraged to diversify the processing of its parts and components. As a result, Japanese automakers will have better protection in their supply chains by relying on multiple countries for inputs; this represents an excellent opportunity for Mexican suppliers to enter the production chains.

A trend in the automotive industry is the commitment to electric vehicles. For instance, producing an electric vehicle decreases the demand for components necessary for its operation. However, the shift to hybrid or electric units has not gained the traction it needs. Despite the environmental benefits associated with electric vehicles, the public has yet to gain enough confidence in the purchase, mainly because the technology for these vehicles is being developed at a slow pace.

The Keiretsu network analysis presented in this chapter is limited to the state of Guanajuato. Other states have a strong presence of Japanese automakers. Future studies can also include companies from different origins in an attempt to find similar patterns of suppliers’ network ties in the Mexican automotive industry.