Printer Friendly

Foreign direct investment and local linkages: an empirical investigation.

Abstract and Key Results

* This paper investigates the quantity and quality of indirect, transactional and collaborative linkages between foreign affiliates and domestic firms based in New Zealand. Seven clusters of affiliates are differentiated by linkage formation behaviour.

* Only fourteen percent of affiliates actively engaged in a broad spectrum of linkages, including alliances. Thirty-nine percent appear limited to demonstration and competition effects. Researchers and policy makers should recognise FDI heterogeneity with regard to impact on domestic industries.

* The results find competitive influence, levels of competition, motives for investment, business activity, technology transfer, age and ownership form are significantly associated with linkage clusters.

Keywords

Foreign Direct Investment, Multinational Enterprises, Linkages, Spillovers, Competition, Suppliers, Alliances

Introduction

The extant literature suggests that inward foreign direct investment (FDI) can contribute to increased productivity of host country resources, skills, and knowledge accumulation by domestic firms, as well as production, exporting and technological capability (Blomstrom 1989, Dunning/Narula 1996, Kokko et al. 1996, Aitken et al. 1997). These effects can be attributed to spillovers from affiliates of foreign multinational enterprises (MNEs) to domestic firms of assets, technology and capabilities that accompany inward FDI (Dunning 1993). However, many studies, particularly those undertaken in developing or transitional country contexts, suggest that inward FDI does not necessarily result in beneficial spillovers (e.g. Aitken/ Harrison 1999, Bosco 2001, Gunther 2005). This finding has been attributed to the limited number and/or intensity of linkages formed by different types of foreign affiliates in different host economy contexts (McIntyre et al. 1996). Therefore, understanding the relationship between the nature of the affiliate and the nature of linkages is important for understanding how FDI functions to benefit the host economy. This is the focus of this paper.

A review of the extant literature reveals several limitations. Emphasis by researchers on spillovers at a macro or meso level fails to adequately consider the mechanisms for foreign affiliate-domestic firm spillovers that occur at the micro or firm level (ie. linkages). Where firm level linkages are the focus, empirical investigation is usually limited to econometric studies using panel data and input/output tables (Girma et al. 2004), backward (and occasionally forward) linkages (UNCTAD 2001 ), or single, case-study firms (Ivarsson/Alvstam 2004). Measurement difficulties mean larger samples tend to focus on the quantity or number of linkages, rather than their quality or potential for beneficial spillovers. Country or region specific research has focused on developing/transitional countries (De Mello 1997), but rather neglected those in Australasia, despite their reliance on inward FDI. This is particularly true in the case of New Zealand, which in the mid-1990s was the most highly dependent developed economy on FDI from abroad (UNCTAD 1999). These trends signal the importance of understanding the longer-term spillover impacts of FDI in developed economies.

The purpose of this paper, therefore, is to address these limitations by developing constructs that measure both the quantity and quality dimensions of a wide range of inter-firm non-equity linkages. This is the paper's first contribution. The second is to use the linkage constructs to differentiate the affiliates by their patterns of linkage formation. The focus here is on the combinations of different types of linkages formed by the affiliates, rather than individual types of linkages. To the best of the author's knowledge, this is the only study that has attempted to identify and empirically measure linkage combinations at the firm level. The third contribution is to analyse the relationship between linkage combinations and affiliate characteristics, in other words, to explore the relationships between the nature of inward FDI and the nature of linkage formation in a small, developed country context. The paper draws on a unique data set of over 500 foreign affiliates based in New Zealand.

Spillovers and Linkages

Multinational enterprises (MNEs) have long been recognised as having an impact on host economies by virtue of their unique firm-specific assets--many of which arise as a result of multinationality (Bellak 2004). The direct developmental impact through output, tax revenues, exports, R&D and employment generated by MNE affiliates is well-documented (for reviews see Dunning 1993, Lall 1993, UNCTAD 2005). Less well-documented is the indirect development impact on host economies that can be attributed to the spillovers to local industry as a result of the foreign affiliates' activities (Blomstrom/Kokko 2001).

FDI spillovers are externalities that arise when the entry or presence of foreign affiliates leads to improvements in the capability, productivity or efficiency of domestic firms (Blomstrom/Kokko 2001). Essentially, the affiliate fails to capture the full value from its activities which then 'spillover' to domestic firms (Eden et al. 1997). Evidence of both positive and negative spillovers in host economies is provided by numerous studies in different contexts and using different methods, but with mixed, and often inconclusive results (Gunther 2005, for a reviews see Blomstrom/Kokko 1998, Gorg/Strobl 2001).

However, there are several limitations of the current literature for our understanding of how the activities of foreign affiliates impact on domestic firms. A focus on changes to domestic productivity or output as a result of the presence of FDI, obscures the mechanisms or interactions at the firm or micro-level that give rise to these improvements (De Mello 1997). As Gorg/Strobl state, '[T]he approach adopted in the empirical literature therefore largely avoids the (arguably difficult to answer) question as to how productivity spillovers take place, but focuses on the simpler issue of whether the presence of MNCs affects productivity in domestic firms' (2001, p. F724). Studies using panel level data or input-output tables do not include perceptual measures that enable a much more in-depth view of the influence of managerial/firm/environmental characteristics on the occurrence of spillovers (Sun 1996, Liu et al. 2000, Bosco 2001, Driffield et al. 2002). Criticisms have been levelled at previous studies that treat FDI as a homogenous phenomenon, assessing overall impact on an industry or economy level, rather than differentiating between affiliate activities, types of foreign investment, and the unequal benefits that accrue to domestic firms at the micro or firm level (Williams 1997, Girma et al. 2004). Indepth case studies of companies or industries provide some resolution to the limitations of aggregate data studies by offering valuable insights as to the types of FDI, linkages and even the extent of upgrading of domestic firms (Barrow/Hall 1995, Brown 1998, Dunning 1998, Ivarsson/Alvstam 2004). But case studies fall short of providing definitive or representative evidence.

In order to understand how FDI spillovers take place, we need to examine the mechanisms through which such spillovers occur at the firm level. In effect, these mechanisms are the linkages that form between foreign and local firms. In this paper, we define linkages as non-equity based relationships between foreign affiliates and domestic firms that have the potential for spillovers. While methodology, terminology and research focus vary, the general consensus in the literature is that linkages can make an important contribution to domestic firms' technical, managerial and organisation capabilities (Dunning 1993, UNCTAD 2001).

Linkages may be competitive (horizontal), occurring through competition effects: transactional (vertical) linkages with domestic suppliers, agents or customers, licensees or franchisees; or collaborative with domestic partners for strategic, technological or managerial purposes. While much of the literature is devoted to measuring spillovers from horizontal linkages, there is growing evidence that vertical and collaborative linkages benefit local firms more by providing channels for inter-firm exchange of technology, knowledge and other resources (Girma et al. 2004, Gunther 2005). These linkages are discussed below.

Competitive linkages arise from MNE entry or FDI presence in an industry. Competition effects can prompt innovation, efficiency and technical improvements in domestic firms (Caves 1974, Bertschek 1995). Increased competition may reduce the profits of domestic firms and force less efficient ones out of business. It can also encourage the affiliate to transfer technology from the parent in order to compete locally thus increasing the likelihood of spillovers (Sjoholm 1997, Gorg/Strobl 2003). Over time FDI can improve resource allocation and productive efficiency of domestic sectors by reducing input costs, promoting exports and facilitating inter-regional and inter-sector flows of labour and capital (Sun 1998, Markusen/ Venables 1999).

Transactional linkages generate demand for, and add to the supply of, locally produced goods and services. In addition, foreign affiliates may contribute to raising the capabilities of domestic suppliers or customers by raising quality standards and efficiency of production, as well as providing assistance and resources relating to procurement, design, quality control, training, or market information (Lall 1980, Driffield et al. 2002). Transitional linkages may be backward, forward or contractual in nature.

Backward (upstream) linkages with suppliers and subcontractors are the focus of most research and are frequently measured by the extent of local sourcing by affiliates and the resulting impacts on demand, employment or export competitiveness (Grosse 1988, Fuentes et al. 1993, Phelps 1993, Barkley/McNamara 1994, Collis et al. 1994, Barrow/Hall 1995, Poon 1996, Hood/Taggart 1997, Ruane/Gorg 1997, Papanastassiou/Pearce 1999, Belderbos et al. 2001, UNCTAD 2001, 2002). Spillovers result from pressure on domestic firms to improve quality, reliability or cost competitiveness. Affiliates may make an active contribution to domestic firm development through assistance or resource transfer, although fewer studies extend their analysis to include such transfers (Halbach 1989, Wong 1992, Crone/Roper 2001, Giroud 2003, UNCTAD 2001, Scott-Kennel 2004, Ivarsson/Alvstam 2005).

Forward (downstream) linkages with customers and agents appear important but are addressed less frequently in empirical research (McAleese/McDonald 1978, Sun 1996, Girma et al. 2004). Notable exceptions include Dunning's (1998) study of US manufacturers in UK industry that created positive spillovers via both forward and backward linkages. The manufacturers provided strict quality demands, training, transfer of technical and manufacturing knowledge and assistance. Another study by Driffield et al. (2002) found that forward linkages contributed more to domestic spillovers than backward linkages, reinforcing the need for their inclusion in future studies.

Contractual linkages involve relationships with domestic franchisees and licensees, but few studies consider these relationships employed as a strategy in conjunction with FDI or in studies of FDI spillovers. However, there is some evidence of these relationships involving the on-going transfer of resources and assistance to domestic firms (Scott-Kennel/Enderwick 2004), so this study includes them for completeness.

Collaborative linkages involve affiliate-domestic firm alliances, technology sharing or development agreements, management contracts and other non-equity agreements involving on-going collaboration between firms. Research on collaborative relationships between foreign affiliates and domestic firms is yet to be incorporated into the mainstream research on spillovers, but can be found in the growing literature on clusters, networks and strategic alliances, where the focus is on organisational learning and competitiveness (Hakansson/Johanson 1993, Chen/ Chen 1998, Enright 2000). There is growing recognition, however, by those investigating inter-firm linkages, of the potential for mutual technological cooperation and development as well as inter-organisational learning (Ivarsson 2002, Chen et al. 2004, Bell/Marin 2004, Scott-Kennel/Enderwick 2004).

Another important consideration, but again, one rather neglected in previous studies, is the patterns or combinations of multiple types of linkages formed by affiliates in a host economy (see Chen et al. 2004, Gunther 2005). The majority of studies confine their empirical analysis to a single or dual type of linkage, typically local sourcing or backward linkages with suppliers, and on occasion, forward linkages with customers. These studies contribute to our understanding of the extent of each of these types of inter-firm linkages, but do not offer a complete picture of affiliates' linkage behaviour. Thus existing research does not differentiate between linkage patterns formed by the affiliates, an issue addressed by our first research proposition:

Research Proposition 1. Foreign affiliates can be differentiated according to their patterns of linkage formation, which are determined by the combination of competitive, transactional (forward, backward, contractual) and collaborative linkages formed with host country firms.

Affiliate Characteristics

Wider literature on the impact of MNEs on host country development suggests that the nature of the affiliate's strategy and activities in the host economy play a crucial role in determining the extent and pattern of linkage formation. However, empirical research in this area is not well-established and specific relationships between types of linkages and types of affiliates are still being explored.

For example, research suggests that market-seeking, efficiency-seeking (Driffield/ Noor 1999), and resource-seeking (Tavares/Young 2002) motives for investment encourage higher levels of input linkages, and a strategic asset-seeking investment motive encourages collaborative linkages (Ivarsson/Jonsson 2003). Transfer of technology from the foreign parent contributes to the competitive position of the affiliate in the host economy, but may reduce the willingness of affiliates to engage in direct linkages for fear of losing proprietary advantages (UNCTAD 2001). Conversely, companies in highly competitive industries may be more likely to engage in local innovation and form alliance and technology sharing agreements with domestic firms (Ivarsson 2002). Other studies suggest that spillovers might vary according to industry (Blomstrom/Kokko 1998), and are more likely to occur where there are competitive pressures, but less likely when the competitive strength of the affiliate outweighs that of local competitors. Studies of technology spillovers, for example, find that benefits to local industry are fewer where capability and technology gaps between foreign and domestic firms are more pronounced and competition is, as a consequence, more uneven (Kogut/Chang 1991, Kokko 1994, Perez 1997, Ozawa 1996, Kokko et al. 1996, De Mello 1997, Gorg/Strobl 2003, Blomstrom/Kokko 1998, 2001).

There is also some evidence to suggest that ownership form influences linkage formation. Belderbos et al. (2001) found that local firms acquired by foreign MNEs --rather than subsidiaries or branches established as greenfield sites by the MNE--were more integrated into the local economy. Age and affiliate autonomy have also been positively associated with linkages. Several studies have found that linkages (eg. local sourcing) tends to increase overtime as affiliates become more embedded in the local economy (McAleese/McDonald 1978, Driffield/Noor 1999, Giroud/ Mirza 2004, Ivarsson 2002). Greater autonomy given to the affiliate has been associated with higher levels of backward linkages (Williams 1997, UNCTAD 2001, Giroud/Mirza 2004). Other studies consider the relationship between size (UNCTAD 2001, Giroud 2003, Giroud/Mirza 2004) or country of origin (Tavares/Young 2002, Driffield/Noor 1999) and linkage formation--with conflicting results. The majority of these studies focus exclusively on manufacturing industries (Gorg/Strobl 2001).

Our second proposition considers the relationship between combinations of linkages formed by the affiliates and their characteristics.

Research Proposition 2. Groups of foreign affiliates forming different combinations of linkages can be differentiated in terms of their: motives for investment; main business activity (manufacturing, service, or sales); reliance on parent company; competitive position; competitive environment; ownership form; age; autonomy; size; country of origin; and industry.

Research Context

Much of the research on linkages and spillovers is centred on developing or transitional countries (Blomstrom 1989, Grosse 1988, Svetlicic/Rojec 1994, McIntyre et al. 1996, Kokko et al. 1996, Sun 1996, Khawar 1997, De Mello 1997, Figueroa 1998, Aitken/Harrison 1999, Lall 2001, Narula/Sadowski 2002, Giroud 2003) or large, developed economies or regions (Globerman 1979, Dunning 1998, Cantwell/ Piscitello 2002, Sambharya/Banerji 2006). Some studies consider how FDI-led upgrading might occur in developed countries where firms can compete on a more even footing, services and knowledge based industries predominate, and cooperative activities are common (Girma et al. 2001, 2004, Liu et al. 2000). But few focus on countries where due to the small size and openness of the domestic market, international competitiveness and business (ie. trade and FDI) is fundamental to economic progression. We propose that in such economies, linkages become all the more important (but see Ruane/Gorg 1997, Ivarsson/Jonsson 2003). Research has tended to over-emphasize 'greenfield' or new investment disproportionate to its share of total FDI in developed countries. Greenfield FDI has more easily measurable impacts such as capital, technology transfer and employment. In contrast, the impacts of 'brownfield' or existing FDI occur over time and are more difficult to measure (Enderwick 1998).

In order to better understand the role of FDI in such economies this study considers the New Zealand case. New Zealand is a small, but developed open economy, highly reliant on a high level of FDI for capitalisation. Upgrading of knowledge-based assets is a particularly important issue as it seeks to develop its own outward investing firms. Yet the relationship between the nature of FDI, the resources that accompany it and the outcomes of inter-firm linkages has received insufficient empirical attention. New Zealand agencies continue to seek ever-diminishing opportunities for 'greenfield' projects without understanding the impact of the investment that is already there, or the impact of acquisitionary investment on local industry.

Research Method

Data Collection

A survey instrument was used to collect firm-level data, and questionnaire design followed the Total/Tailored Design Method (Dillman 1978, 2000). Pre-testing was conducted in four stages, including a pilot test. The questionnaire was administered to the target population of foreign affiliates, defined as all firms with 25 percent or more foreign ownership operating in New Zealand. A list of these companies was unavailable to use as a sampling frame for the study, so a database was constructed by the author using a number of sources including domestic and international business directories, industry associations, and government departments. To the best of all publicly, and some privately held knowledge available, the database constructed for this research included all the New Zealand head offices of foreign-owned firms at the time of the survey. The sample consisted of the entire sampling frame, or known population of 1800 head offices. The survey questionnaire was sent to the Chief Executive Officer or Managing Director. A total of 764 responses were received, two indicated the respondents did not wish to complete the survey, and 246 indicated that the firm was either no longer 25 percent or more foreign owned, or had been closed down. Thus, the initial count of foreign affiliate head offices was revised to 1554. In total, 516 useable responses were received, giving a useable response rate of 33 percent.

Affiliate Profile

The survey found that the affiliates, on average, had been established in New Zealand for over 27 years and owned by their current foreign investor for almost 17. Sixty-three percent were branches or subsidiaries of a MNE, and 25 percent had been acquired by MNEs. Parent companies were based in North America (29 percent, with the U.S. accounting for 27 percent alone), Asia-Pacific (40 percent, with Australia accounting for 25 percent and Japan 10 percent) and Europe (27 percent with the U.K. 9 percent, Germany 5 percent, and France 4 percent). The remaining 4 percent of affiliates were based outside these regions. In terms of main business activity, manufacturing affiliates (including primary processors) made up 34 percent of the sample, service affiliates 31 percent, and sales offices 33 percent. Affiliates were represented all industries given in Table 1.

The number of affiliates in the sample by industry, and the size of the domestic sector by number of enterprises (as at February 2000) is given in brackets as follows; manufacturing (158 affiliates, 7.4 percent), wholesale trade (138, 5.9 percent), property and business services (53, 31.3 percent), finance and insurance (39, 2.4 percent), transport and storage (34, 3.8 percent), retail trade (25, 12.3 percent), construction (15, 12.9 percent), agriculture, forestry and fishing (11, 3.9 percent), communications services (9, 1.2 percent), cultural and recreational services (9, 3.4 percent), accommodation, cafes and restaurants (8, 3.4 percent), mining (6, 1.3 percent), electricity, gas and water supply (5, 0.5 percent), and other (6, 10.8 percent).

Measurement

The focus of previous research tends to be on the quantity, rather than the quality of domestic linkages, despite the latter being the critical issue for upgrading (Turok 1993, Barrow/Hall 1995, UNCTAD 1999). Most studies attempt to quantify the extent of local sourcing by foreign affiliates, but neither investigate the types of products being sourced, nor the extent to which the firms exchange resources and knowledge in the process of these transactions. This study incorporates the number of different types of inter-firm linkages, their degree of specialisation (on-going interactions between the firms to customise or use or sell products/services), and most importantly, the extent of affiliate-domestic firm resource transfer and assistance. In this way the study captures both the quantity and quality of the mechanisms by which spillovers might occur.

To assess competitive linkage formation by the affiliates (RP1), respondents were asked to indicate the extent to which their affiliate had influenced changes to the level of competition, the number of major and smaller players, and the level of competitiveness of other firms in their industries in the three years prior to the survey. Forward linkages were measured by the number of agents (wholesale/retail trade, marketing and distribution) and customers for specialised products or services that the affiliate had worked with in New Zealand in the twelve months prior to the survey, as well as the number and types of assistance or resources transferred to local agents/customers by the affiliate during this time. Backward linkage measures included the number of relationships with suppliers/subcontractors; the transfer of assistance/resources to these firms; and the extent of reliance on New Zealand firms (rather than offshore firms) for sourcing of specialised products/services over a twelve-month period. Contractual linkages were measured by affiliates' number of licensee and franchisee relationships in New Zealand, and the transfer of assistance/resources in the same period. Finally, collaborative linkages were measured by the number of relationships formed with local partners for strategic/functional alliances or technology sharing and development in the three years prior to the survey, as well as the types of assistance and resources transferred to local partners from the affiliates and vice versa (see Table 2).

In order to measure the combinations of linkages formed by the affiliates, factor analysis was used to reduce these items into five linkage constructs; competitive linkage factor, backward linkage factor, forward linkage factor, contractual linkage factor, and collaborative linkage factor. Table 2 shows the rotated factor matrix, which loaded fifteen measurement items onto the five factors (the 'number of types of assistance given to suppliers/subcontractors' did not load onto any factor, and was removed from the analysis). Measures of both quantity (ie. number of relationships) and quality (ie. resources transferred or local sourcing of specialised inputs) were included in each linkage factor.

The variables used to measure affiliate characteristics are given in Table 1. Motive for investment included the resource-, market-, efficiency- and strategic asset-seeking motives used frequently in the literature (Dunning 1993). The 'trade supportive' motive, where affiliates support the value added activities of the investing firm by providing sales outlets and distribution facilities, or acquiring better quality or cheaper imports for the intra-corporate network, is also included (Narula 1996). With regard to business activity, manufacturers (including primary processors), service providers and sales offices are included as binary variables. Sales offices are affiliates that do not undertake production of a good or delivery of a specific service in the host economy, but operate for the purposes of importing/exporting, marketing, distribution and administration. Reliance on the foreign parent focussed on technology transfer, using a three-item factor construct. Competitive position both locally and internationally, and competitive environment were measured using 5-point Likert scales. Ownership form was either a subsidiary established by an MNE or a formerly New Zealand-owned firm acquired by a foreign MNE (joint ventures in the sample were too few to include as a separate category). Other variables are described in Table 1.

Data Analysis

Using the factor scores for the five linkage factors in Table 2, cluster analysis was employed to classify the affiliates into groups or 'clusters' based on patterns of linkage formation. Cluster analysis allows observations to be grouped based on similarity, in this case similar patterns of linkage formation by the affiliates. The idea behind cluster analysis is to group observations such that the within-group similarity is maximised and the between-group similarity is minimised. K-means cluster analysis, a partitioning rather than hierarchical technique was chosen for this study. The advantage of k-means clustering is that it allows observations to move between clusters in the process of computing the solution. That enables a certain degree of flexibility desirable in the context of an exploratory study. After multiple iterations, a seven-cluster solution was considered to achieve the best balance between differentiating the affiliates and interpreting the data. The cluster solutions were verified by comparing the means for each cluster, using ANOVA, which showed F-values significant at the 0.0001-level. A multi-discriminant analysis (MDA) also indicated that the clusters were significantly discriminated by all predictor variables, as the F-tests of equality of group means were all significant at the 0.0001-level.

The relationships between linkage clusters and affiliate characteristics were assessed using ANOVA for variables based on Likert scales and ratio/continuous data, and cross-tabulations for variables based on categorical (nominal) data, (refer to Table 1 for variable measurement). Where the ANOVA (F-test) indicated significant differences in the means of at least two clusters for a particular variable, pair-wise comparisons of cluster means were undertaken using Tukey's post hoc comparisons (Kanji 1995). Where the cross-tabulations (chi-square) indicated significant differences in the distributions of the observations, comparisons between clusters were made by testing for differences in proportions, based on the normal distribution.

Results and Discussion

Patterns of Linkage Formation (RP1)

By comparing the cluster centre values in Table 3 we are able to evaluate the relative strength of the clusters for each type of linkage. The cluster centres derive from the factor scores that are standardised about the mean at zero. In other words, a very small value indicates that the score for that type of linkage for a particular cluster is very close to the mean value for all cases. A positive value indicates that the cluster has an above average score on that linkage factor, while a negative value indicates the opposite. By plotting these centres, Figure 1 shows that each of the clusters is characterised by a unique position in terms of its pattern of linkage formation, lending support to our first research proposition. Each cluster is described below.

FIGURE 1 OMITTED]

Cluster 1 (86 affiliates) is the only group with below average scores on all linkage factors, and in particular, competitive linkages and backward linkages with suppliers and subcontractors. While it appears that affiliates in this cluster require a local presence in New Zealand, most engage in minimal linkage activity with local firms and consequently have the least potential for spillovers. Given this pattern of linkage formation, cluster 1 is named the Enclave Operators.

Cluster 2 affiliates (116 affiliates) score well--above average on the competitive linkage factor. This is indicative of a strong competitive influence over other firms in their industries in terms of the level of competition, the number of major and smaller competitors and the competitiveness of other firms in their industries. However, the affiliates in this cluster, on average, engage in fewer transactional or collaborative linkages. Scores for backward linkages in particular, are well below average. This suggests that spillovers to domestic firms will be limited to the affiliates' presence in the New Zealand market as competitors. Thus we name Cluster 2, the Competitors.

Cluster 3 (45 affiliates) score the highest on the forward linkage factor, and above average on the backward linkage factor. This indicates the affiliates have close working relationships with customers of specialised goods and services, and/or rely on agents in New Zealand for marketing and distribution of their products, as well as sourcing specialised goods/services from local suppliers and subcontractors. Competitive linkages are slightly above average, but contractual and collaborative linkages below average relative to the other clusters. As the distinguishing feature of this cluster of affiliates is supply to local firms via forward linkages, we name cluster 3, the Suppliers.

Clusters 4 (105 affiliates) and 5 (93 affiliates) make up 38 percent of the sample. The clusters both score above average on the backward linkage factor indicating high levels of upstream buying activity, but below average on forward, contractual and collaborative linkages relative to other clusters. The two clusters can be differentiated by the competitive linkage factor. Cluster 4, which we call Buyers, scores well below average, whereas Cluster 5 scores well above average, so we call this the Competitive Buyers cluster. This suggests both clusters have the potential for spillovers via linkages with suppliers/subcontractors, but the Competitive Buyers also exert a strong competitive influence in local industry.

Cluster 6 (20 affiliates) is distinguished by having the highest score on the contractual linkage factor, indicating relationships and resource/assistance exchange with domestic licensees or franchisees. Scores on the collaborative, backward and competitive linkage factors were all above average, indicating a wide range of inter-firm relationships with the potential for spillovers. However, as the membership of this cluster is limited to just 20 firms (4 percent of the sample) these results should be interpreted with some caution. Given that one of the primary reasons for FDI is for the affiliate to exploit firm-specific assets offshore, it is not surprising that a combination of FDI and contractual licensing or franchising is not a common strategy employed by foreign investors in New Zealand (Dunning, 1993). We call this cluster the Contractors.

Cluster 7 (51 affiliates) scored the highest on the collaborative linkage factor relative to other clusters. Affiliates in this cluster also have an above average score on the forward linkage factor, slightly above average score on the competitive linkage factor, and below average score on the backward linkage factor. These findings suggest that the most potential for spillovers will occur through collaborative activity with local partners and relationships with local agents and customers. These linkages are likely to involve the exchange of firm-specific resources, knowledge and mutual learning based on inter-firm cooperation. Affiliates in cluster 7 are named the Collaborators.

Affiliate Characteristics (RP2)

Table 4 shows that different clusters of affiliates can be distinguished by their motives for investment, main business activity, reliance on technology transfer from their parent companies, competitive position and competitive environment, ownership form and age as suggested in RP2. The terms "Higher' and 'Lower' given in Table 4 refer to a significant difference from at least one other cluster mean (ANOVA) or a significant difference in proportions between clusters. For example, the average response in the Enclave Operators (1) and the Buyers (4) clusters were lower (p<0.05) on the 'strategic asset-seeking' motive, while the average response for the Collaborators (7) cluster was higher (p<0.05). Therefore, Enclave Operator and Buyer affiliates can be viewed as less likely to have invested in New Zealand for strategic asset-seeking purposes, and Collaborator affiliates more likely to have done so, relative to at least one other cluster. These results are discussed in detail below.

Different clusters of affiliates can not be significantly distinguished by the affiliates' autonomy, size, country of origin of the parent, or industry (these are listed in Table 1 as per RP2, but not Table 4 as they are insignificant). A cursory glance over these variables suggests that on average, Suppliers (cluster 3) have more autonomy over short-term decision making than other clusters. In terms of size, Enclave Operators (1) tend to employ fewer staff, while Suppliers (3) and Competitive Buyers (5) employ more. Contractors (6) cluster is skewed upward by four large firms which have between 1200-5700 employees and up to NZ$1billion in sales each. Competitors (2), Contractors (6) and Collaborators (7) have higher proportions of parent firms based in the U.S. or Canada, while Buyers (4) and Contractors (6) have proportionally more parent firms from Asia, and Suppliers (3) have more from Europe. The following generalizations can be regarding industry breakdown. Enclave operators (1) are primarily engaged in retail and wholesale sales of consumer or industrial goods, finance, insurance and logistics. Competitors (2) are involved in importing, exporting distribution, software and consultancy. A high proportion of Competitors and Suppliers (3) are producers of industrial goods. Buyers' (4) main industries included transport, construction, financial services and advertising sectors and manufacturing. Competitive buyers (5) are strongly represented in the hospitality, banking, construction, property services and shipping service sectors, and food, chemicals, wool and processing of other primary products in manufacturing sectors. Many of the affiliates in the small Contractors cluster (6) are in the motor vehicle sector. Collaborators (7) in service sectors are strongly represented in telecommunications, information technology, finance, and facilities management, while others were involved in high-technology areas of production such as pharmaceuticals and equipment.

Characteristics by Cluster

The respondents in the Enclave Operators (l) cluster, on average, report a lower likelihood (p<0.05) of operating in New Zealand for the purposes of seeking strategic assets, a lower reliance (p<0.10) on parent technology, and a lower level (p<0.01) of industry competition in the past three years. The proportion of Enclave Operators that are sales offices (p<0.05), branches or subsidiaries of a foreign MNE (p<0.01) is greater, but acquired (former) New Zealand firms fewer (p<0.01). On average, these affiliates are younger, in terms of both their establishment in New Zealand (p<0.01) and their tenure with their current foreign investors (p<0.05). The overall picture of this cluster suggests that Enclave Operators maintain branch operations in New Zealand primarily for the purposes of maintaining a local presence in fairly stable markets. Sales offices are primarily retailers and wholesalers, while service firms operate in location-bound areas such as finance, insurance and logistics. As these sectors account for approximately one quarter of New Zealand industry (by enterprise number), and affiliates are not actively involved in local production or linkage formation with domestic firms, the likelihood of spillovers seems remote.

The Competitors (2) cluster is also characterized by a greater proportion (p<0.01) of sales offices. On average, the affiliates in this cluster have been established (p<0.01 ) and owned (p<0.05) by current foreign owners for fewer years than at least one other cluster. Competitors are more likely to operate in New Zealand for trade supportive reasons (p<0.01), but are less likely to be service providers (p<0.05). These affiliates are more likely to be dominant competitors in New Zealand (p<0.01), and report higher than average change to the level of industry competition (p<0.05). The Competitors' local presence is likely to be supported by their higher than average reliance on parent technology (p<0.05). Industry statistics as well as our findings show that foreign affiliates dominate service industries such as software and consultancy, sales offices focus on importing, exporting and distribution of a wide range of goods, and manufacturers are frequently involved in the production of industrial goods. The competitive challenge posed by affiliates to domestic firms operating ill New Zealand's most competitive industries appears well-supported by parent company connections.

The Suppliers (3) cluster has a significantly higher proportion of manufacturers (p<0.01), but fewer service providers (p<0.01). These affiliates are more likely to occupy a strong competitive position in New Zealand (p<0.01), but not internationally (p<0.05). On average, they are older in terms of both their establishment (p<0.01) and current foreign ownership (p<0.01) than at least one other cluster. Although not significant, manufacture of industrial goods is predominant and indicative of the active and largely autonomous role these firms play in domestic industry through forward linkage formation with agents, distributors and customers. Just over 40 percent have European-based parent companies, which combined with higher autonomy and age is suggestive of the historical manufacturing investments made by multi-domestic companies prior to New Zealand reform of import restrictions in 1984.

Affiliates in the Buyers (4) cluster are, on average, significantly less likely to be strategic asset-seeking (p<0.05) or trade supportive (p<0.01). They also have a lower reliance on parent technology (p<0.01). Buyers are more likely to have weaker competitive positions in New Zealand (p<0.01) and internationally (p<0.05), and report significantly (p<0.05) lower average change to the competitive environment than at least one other cluster. They are more likely to be younger, both in terms of establishment (p<0.01) and ownership by current foreign parent (p<0.01). The proportion of sales offices is lower (p<0.05), but the proportion of acquired New Zealand firms is higher (p<0.10) in the Buyers cluster. The overall picture suggests that for resource and market-seeking reasons MNEs require local presence, and often will acquire a domestic firm to obtain one. This view is supported by the industry breakdown which finds service providers seeking to serve local markets in the transport, construction, financial services and advertising sectors, while primary processors and manufacturing firms appear to use New Zealand as a base for supplying inputs and finished goods (eg. equipment) to offshore markets and/or the wider corporate network. As these sectors account for approximately 30 percent of domestic industry and one-fifth of the affiliates are in this cluster, the potential for spillovers is considerable.

The Competitive Buyers (5) cluster has a significantly lower number of firms engaged solely in sales office (p<0.05) and trade supportive activities (p<0.01), but more service providers (p<0.05). There is a higher likelihood that Competitive Buyers are strong competitors (p<0.01) faced with an increasingly competitive environment in New Zealand (p<0.10). These affiliates have, on average, been established in New Zealand for longer (p<0.05), but not owned by their foreign owners for as long (p<0.01). There is a greater proportion of former New Zealand firms acquired by foreign MNEs (p<0.10), but the affiliates are significantly less reliant on their parent companies for technology (p<0.01). Previous studies show that domestic firms rely more heavily on local sources of inputs than foreign affiliates, and these findings suggest that such behaviour continues post-acquisition (Ruane/Gorg 1997). Service affiliates in this cluster appear to be involved primarily in hospitality, banking, construction, property services and shipping, and manufacturers are involved in the production of food, chemicals, wool and processing of other primary products. These are not only New Zealand's founding industries (many by the first foreign investors), but also account for a large proportion of domestic enterprises. Thus, this cluster appears to be well embedded in competitive New Zealand industries, and potential for spillovers is high.

The Contractors (6) cluster had a significantly higher proportion of sales offices (p<0.05), but fewer manufacturers (p<0.10) than at least one other cluster. Affiliates in this cluster are significantly (p<0.05) more likely to occupy a stronger competitive position in New Zealand. Contractors are, on average, older both in terms of local establishment (p<0.01) and current foreign ownership (p<0.01) than at least one other cluster. Many of the affiliates in this small cluster are involved in the sale of motor vehicles, oil or motor vehicle parts--sectors long associated with foreign involvement. Although their impact will be lessened because there are so few, the affiliates appear to be integrated into both global corporate networks and the local economy via relationships with local licensees or franchisees, rather than their own manufacturing activities.

The Collaborators (7) cluster has a higher than average strategic asset-seeking motive for investment (p<0.05), and there is a greater proportion of service providers (p<0.10). Collaborators are distinguished by being, on average, stronger competitors both in New Zealand (p<0.10), and internationally (p<0.05), and by reporting higher levels of industry competition (p<0.01). The number of years since establishment, and ownership by current foreign MNEs, are both significantly lower (p<0.05) than at least one other cluster. The industry breakdown reveals a large proportion of the affiliates in this cluster are involved in high-technology, global sectors (telecommunications, information technology, finance, and facilities management, pharmaceuticals and the sale and/or manufacturing of equipment). These findings suggest a new wave of investment by MNEs into services and manufacturing, two areas targeted for growth in New Zealand. Spillover potential is very high as some seek local innovation through acquisition, and many share their own competencies through collaborative and transactional linkages with domestic firms.

Conclusions

The objectives of this paper were twofold. The first was to differentiate affiliates based on their combinations of inter-firm linkages. This objective was addressed by incorporating items for five types of inter-firm linkage into a factor analysis, and then combining the constructs using cluster analysis to assess the different linkage patterns formed by the affiliates. Incorporation of a broad range of linkages as well as quality and quantity dimensions avoided underestimating the linkage effects associated with FDI. The study found that affiliates could be grouped into seven clusters; two limited to indirect effects through demonstration or competition, three involving primarily transactional linkages with suppliers, agents and customers, and two where affiliates, in conjunction with transactional linkages, formed relationships with contractual and/or collaborative partners. Transactional clusters accounted for 47 percent of the affiliates, and a further 39 percent were limited to indirect effects. This left a mere 14 percent actively engaged in a wider range of linkages, including collaboration with domestic partners.

It is clear from our analysis that FDI cannot be treated as a homogenous entity, as different types of affiliates can, via different linkage combinations, exert different impacts on domestic firms. We suggest that Collaborators present the most potential for positive spillovers, given the higher likelihood of reciprocal resource transfer and joint-development associated with alliances and on-going collaborative activity. Enclave Operators, on the other hand, have the least potential given that they form very few (if any) linkages. Competitors have the potential for both positive and negative spillovers depending on competition intensity and technology gaps between domestic and foreign firms (Blomstrom/Kokko 1998, 2001, Sjoholm 1997). Suppliers, Buyers and Contractors all have the potential to contribute to spillovers, not only through supply and demand, but also by close working relationships (and competition effects in the case of the Competitive Buyers cluster).

The second objective was to investigate what types of affiliates would be most likely to form different combinations of linkages. This objective was addressed by determining the relationships between linkage clusters and affiliate characteristics. Seven characteristics were significantly associated with the linkage clusters, and can be divided into those that can be influenced by government policy (the extent of competitive influence and levels of competition in the affiliates' industries); and those that are more firm-specific (motives for investment, business activity, reliance on technology transfer from the foreign parent, age and ownership form). Autonomy, country of origin, industry and size were not found to be significant predictors of overall linkage formation, echoing the mixed results of previous studies (Blomstrom/Kokko 1998, UNCTAD 2001, Giroud 2003, Tavares/Young 2002, Driffield/ Noor 1999). This could be attributed to a number of factors, including the New Zealand context relative to that of the developing or transitional countries so often the subjects of linkage studies. What these findings do seem suggest, however, is that degree of linkage (incorporating type, quantity and quality of linkages) is better able to be predicted by the affiliates' activities and industry competition, rather than specific industries, size or origin.

Enclave Operators and Competitors are more likely to be sales offices, undertaking trade and support activities as branches or subsidiaries of the foreign parent, whereas both Buyer clusters are less likely to be engaged in such activities. In contrast, Competitors seem to fit the traditional MNE profile by exploiting (rather than sharing) firm-specific assets in order to compete in competitive domestic industries. Age appears to be an important factor explaining why Enclave Operators and Competitors have fewer linkages than other clusters, with younger affiliates less embedded locally and still reliant on foreign sources of inputs. This idea is supported by the finding that Suppliers, Competitive Buyers and Contractors- clusters demonstrating multiple linkages--tend to be older and focussed on the domestic market. Both Buyer clusters are more likely to be former New Zealand-owned firms acquired by a foreign MNE but with a lower reliance on parent technology, suggesting they retain existing networks of linkages post-acquisition. Conversely, we find that collaborators are younger--but age coupled with competitive position/environment--seems to suggest dynamic affiliates operating as part of the wider MNE network in globally competitive industries. These firms are more likely to be service providers seeking strategic-assets, perhaps explaining their desire to form collaborative and forward linkages in the host economy.

Within the wider literature of endogenous FDI-led development, the results of this paper occupy a rather unique position. While our findings are by no means able to be generalised to all countries or regions, they do offer useful insights for small, developed economies seeking to attract FDI. The findings help address a gap in an extant literature focused predominantly on developing or large economy country contexts (Dunning 1993, Blomstrom/Kokko 1998). In the smaller, periphery economy, which is lacking certain resources such as capital and international expertise, linkages with foreign affiliates occupy an important developmental role (Van Den Bulcke/Verbeke 2001). The final section of this paper discusses policy implications and future research arising from this study.

Policy Implications and Future Research

FDI has been widely touted as the solution to economic growth and development. However, this study finds that the extent and types impact of FDI on domestic firms via linkages depends very much on the type of FDI. While actual spillovers can only be inferred from this study, it finds that almost 40 percent of the respondent affiliates do not form any transactional or collaborative linkages with domestic firms. Thus, any influence they may have can only be limited to competition or demonstration effects. Such linkages offer negligible opportunity for inter-firm cooperation, learning, or sharing of resources. On the other hand, 14 percent of firms demonstrated their involvement with a wide range of linkages--from competition effects to collaboration. Policy makers need to understand that FDI stocks and inward flows are not necessarily correlated with long-term beneficial effects on domestic industry, unless they can be attributed to the types of firms that will engage in linkages that subsequently force or encourage domestic firms to increase output, productivity and performance.

Attracting the right 'type' of investment based on affiliate characteristics is therefore crucial, albeit difficult, for linkage formation. It is important to signal that these characteristics are likely to be different for the small economy than developing, or large economies. Small, developed economies neither attract inward FDI on the basis of low-cost labour as countries such as China or India can, nor by virtue of critical mass or markets, as might the United States. Nor are the types of linkages sought likely to be the same. The existing levels of host country technological and structural development will determine what types of linkage will be most appropriate (Narula 1996). These results suggest that countries seeking to attract, retain and benefit from integration of inward FDI into the domestic economy need to identify the types of investments most likely to form the types of linkages best suited to the country's stage of development, and to ensure that barriers to those types of linkages are minimised. By identifying the characteristics of the affiliates in each linkage cluster, this research provides a starting point for making future FDI promotion policies more effective when targeting specific investors or certain types of investment (UNCTAD 2003).

Policy should also consider the competitiveness of domestic industry. Many studies have shown that a smaller technology gap between domestic and foreign firms increases the likelihood of linkages and spillovers occurring due to better absorptive capacity. Others have found that availability, geographic proximity and competency of local suppliers also increases the likelihood of local sourcing and inter-firm linkages (for a review see Blomstrom/Kokko 2001). Competition between firms has been found to contribute to a strong industry base and international competitiveness (Porter 1990). Our study lends support to these ideas, as Competitor, Competitive Buyer, and Collaborator clusters are more likely to be operating in competitive, dynamic industries and enclave operators are less likely. All clusters except Enclave Operators and Buyers are also more likely to hold competitive positions in New Zealand. Judging from the types of firms dominant in these clusters it is important that policy makers endeavour to sustain a competitive environment both local and global industries, in order to (rather paradoxically) encourage interfirm cooperation via linkages. A good example of this in New Zealand is the electronics cluster near Christchurch, where local and foreign firms occupy a dominant position locally but operate within a global industry. The development of this industry has been supported by openness to inward FDI (most recently in the form of acquisition), competition on a domestic (and global) front, and the facilitation of supporting business and educational infrastructure by government.

This study has also revealed several potential areas of future research. The first is an extension of the study to different country or regional contexts, to see how and why patterns of linkage formation differ and the suitability of different types of linkages in different contexts. This is particularly important for small, developed economies where more research is needed on linkages and spillovers. The second is to refine approaches to measuring autonomy, industry, size and country of origin to address inconsistencies and gaps in research findings to-date. The third is a study of linkages including data from domestic firms. This would help us understand the nature of linkages from the local firm's perspective, and to what extent developmental spillovers actually occur. Finally, a survey of national or regional investment promotion organisations could assess the extent to which they take long-term spillover benefits into account when evaluating the potential benefits of FDI, rather than just the immediate changes to technology, output and employment.

Acknowledgements

The author would like to acknowledge the assistance from Elizabeth Rose with regard to some of the statistical tests used in this paper, and also the helpful suggestions of two anonymous reviewers on earlier drafts.

References

Aitken, B., G./Hanson, H./Harrison, A. E., Spillovers, Foreign Investment and Export Behavior, Journal of International Economics, 43, 1/2, 1997, pp. 103-132.

Aitken, B. J./Harrison, A. E., Do Domestic Firms Benefit from Direct Foreign Investment'? Evidence from Venezuela, American Economic Review, 89, 3, 1999, pp. 605 618.

Barkley, D./McNamara, K., Local Input Linkages: A Comparison of Foreign-Owned and Domestic Manufacturers in Georgia and South Carolina, Regional Studies, 28, 7, 1994, pp. 725 737.

Barrow, M./Hall, M., The Impact of a Large Multinational Organization on a Small Local Economy, Regional Studies, 29, 7, 1995, pp. 635-653.

Belderbos, R./Capannelli, G./Fukao, K., Backward Vertical Linkages of Foreign Manufacturing Affiliates: Evidence from Japanese Multinationals, World Development, 29, 1, 2001, pp. 189-208.

Bell, M./Marin, A., Where Do Foreign Direct Investment-Related Technology Spillovers Come from in Emerging Economies? An Exploration in Argentina in the 1990s, The European Journal of Development Research, 16, 3, 2004, pp. 653-686.

Bellak, C., How Domestic and Foreign Firms Differ and Why Does it Matter?, Journal of Economic Surveys, 18, 4, 2004, pp. 483-514.

Bertschek, I., Product and Process Innovation as a Response to Increasing Imports and Foreign Direct Investment, The Journal of Industrial Economics, 43, 4, 1995, pp. 341-357.

Bosco, M. G., Does FDI Contribute to Technological Spillovers and Growth? A Panel Data Analysis of Hungarian Firms, Transnational Corporations, 10, 1, 2001, pp. 43-68.

Blomstrom, M., Foreign Investment and Spillovers: A Study of Technology Transfer to Mexico, London: Routledge 1989.

Blomstrom, M./Kokko, A., Multinational Corporations and Spillovers, Journal of Economic Surveys, 12, 3, 1998, pp. 247-278.

Blomstrom, M./Kokko, A., Foreign Direct Investment and Spillovers of Technology, International Journal of Technology Management, 22, 5/6, 2001, pp. 435 454.

Brown, R., Electronics Foreign Direct Investment in Singapore: A Study of Local Linkages in Winchester City, European Business Review, 98, 4, 1998, pp. 196-210.

Cantwell, J./Piscitello, L., The Location of Technological Activities of MNCs in European Regions: The Role of Spillovers and Local Competencies, Journal of International Management, 8, 1, 2002, pp. 69-96.

Caves, R. E., Multinational Firms, Competition and Productivity in Host-Country Markets, Economica, 41, 162, 1974, pp. 176-193.

Chen, H./Chen, T.-J., Foreign Direct Investment as a Strategic Linkage, Thunderbird International Business Review, 40, 1, 1998, pp. 13-31.

Chen, T.-J./Chen, H./Ku, Y.-H., Foreign Direct Investment and Local Linkages, Journal of International Business Studies, 35, 4, 2004, pp. 320-333.

Collis, C./Noon, D./Berkeley, N., Direct Investment from the EC: Recent Trends in the West Midlands Region and their Implications for Regional Development and Policy, European Business Review, 94, 2, 1994, pp. 14-19.

Crone, M./Roper, S., Local Learning from Multinational Plants: Knowledge Transfers in the Supply Chain, Regional Studies, 35, 6, 2001, pp. 535-48.

De Mello, L., R., Foreign Direct Investment in Developing Countries and Growth: A Selective Survey, The Journal of Development Studies, 34, 1, 1997, pp: 1-34.

Dillman, D. A., Mail and Telephone Surveys: The Total Design Method, New York: John Wiley & Sons 1978.

Dillman, D. A., Mail and Internet Surveys: The Tailored Design Method, New York: John Wiley & Sons 2000.

Driffield, N./Noor, A. H. M., Foreign Direct Investment and Local Input Linkages in Malaysia, Transnational Corporations, 8, 3, 1999, pp. 1-24.

Driffield, N./Munday, M./Roberts, A., Foreign Direct Investment, Transactional Linkages, and the Performance of the Domestic Sector, International Journal of the Economies of Business, 9, 3, 2002, pp. 335-351.

Dunning, J. H., Multinational Enterprises and the Global Economy, Wokingham: Addison-Wesley Publishing Company 1993.

Dunning, J. H., American Investment in British Manufacturing Industry, London, New York: Routledge 1998.

Dunning, J. H./Narula, R., Foreign Direct Investment and Governments: Catalysts for Economic Restructuring, London, New York: Routledge 1996.

Eden, L. E./Levitas, E./Martinez, R. J., The Production, Transfer and Spillover of Technology: Comparing Large and Small Multinationals as Technology Producers, Small Business Economics, 9, 1, 1997, pp. 53-66.

Enderwick, P., Foreign Direct Investment: The New Zealand Experience, Palmerston North: Dunmore Press 1998.

Enright, M. J., Regional Clusters and Multinational Enterprises: Independence, Dependence or Interdependence?, International Studies of Management and Organization, 30, 2, 2000, pp. 114-138.

Figueroa, A., Equity, Foreign Investment and International Competitiveness in Latin America, Quarterly Review of Economics and Finance, 38, 3, 1998, pp. 391 408.

Fuentes, N. A. F./Alegria, T./Brannon, J. T./James, D. D./Lucker, W., Local Sourcing and Indirect Employment: Multinational Enterprises in Northern Mexico, in Bailey, R/Parisotto, A./Renshaw, G. (eds.), Multinationals and Employment: The Global Economy of the 1990s, Geneva: International Labour Office (ILO) 1993.

Girma, S./H. Gorg, H./Pisu, M., The Role of Exporting and Linkages for Productivity Spillovers from FDI, Research Paper Series: Globalisation, Productivity and Technology, GEP Research Paper No. 30, Nottingham: University of Nottingham 2004.

Girma, S./Greenaway, D./Wakelin, K., Who Benefits from Foreign Direct Investment in the UK?, Scottish Journal of Political Economy, 48, 2, 2001, pp. 119-133.

Giroud, A., Transnational Corporations, Technology and Economic Development: Backward linkages and Knowledge Transfer in South-East Asia, Cheltenham, United Kingdom: Edward Elgar 2003.

Giroud, A./Mirza, H., Multinational Enterprises and Local Input Linkages in Southeast Asia, Paper Proceedings of the Academy of International Business Conference: Stockholm 2004.

Globerman, S., Foreign Direct Investment and 'Spillovef Efficiency Benefits in Canadian Manufacturing Industries, Canadian Journal of Economics, 12, 1, 1979, pp. 42-56.

Gorg, H./Strobl, E., Multinational Companies and Productivity Spillovers: A Meta-Analysis, Economic Journal, 111,475, 2001, pp. 723-739.

Gorg, H./Strobl, E., Multinational Companies, Technology Spillovers and Plant Survival, Scandinavian Journal of Economics, 105, 4, 2003, pp. 581-595.

Grosse, R.. The Economic Impact of Foreign Direct Investment: A Case Study of Venezuela, Management International Reviews; 28, 4, 1988, pp. 63 78.

Gunther, J., Technology Spillovers from Foreign Investors in Transition Economies, Economic and Business Review,for Central and South-Eastern Europe, 7, I, 2005, pp. 5-24.

Hakansson, L./Johanson, J., The Network as a Governance Structure, in Grabher, G. (ed.), The Em- bedded Firm: The Socio-Economics of Industrial Networks, London: Routledge 1993, pp. 35-51. Halbach, A. J., Multinational Enterprises and Subcontracting in the Third World: A Study of InterIndustrial Linkages, Geneva: International Labour Office (ILO) 1989.

Hood, N./Taggart, J. H., German Foreign Direct Investment in the UK and Ireland: Survey Evidence, Regional Studies, 31, 2, 1997, pp. 139 150.

Ivarsson, 1., Collective Technology Learning between Transnational Corporations and Local Business Partners: the Case of West Sweden, Environment and Planning A, 34, 10, 2002, pp. 1877-1897.

Ivarsson, I./Alvstam, C. G., International Technology Transfer through Local Business Linkages: The case of Volvo Trucks and their Domestic Suppliers in India, Oxford Development Studies, 32, 2, 2004, pp. 241-260.

Ivarsson, I./Alvstam, C. G., The Effect of Spatial Proximity on Technology Transfer from TNCs to Local Suppliers in Developing Countries: The Case of AB Volvo in Asia and Latin America, Eco,omit Geography, 81, 1, 2005, pp. 83-111.

Ivarsson, L/Jonsson, T., Local Technological Competence and Asset-seeking FDI: An Empirical Study of Manufacturing and Wholesale Affiliates in Sweden, International Business Review, 12, 3, 2003, pp. 369-386.

Kanji, G. K., Statistical Tests. London: Sage 1995.

Khawar, M., The Impact of Multinational Corporations on a Developing Country: A Trade Off in the Long Run?, The Journal of International Trade and Economic Development. 6, I, 1997, pp. 101-112.

Kogut, B./Chang. S. J., Technological Capabilities and Japanese Direct Investment in the United States, Review of Economies and Statistics.73, 3, 1991, pp. 401-413.

Kokko, A., Technology, Market Characteristics, and Spillovers, Journal of Development Economics, 43.2, 1994, pp. 279-293.

Kokko, A./Tansini, R./Zejan, M. C.. Local Technology Capability and Productivity Spillovers from FDI in the Uruguayan Manufacturing Sector, Journal o[ Development Studies, 32, 4, 1996, pp. 602-612.

Lall, S., Vertical Interfirm Linkages in LDCs: An Empirical Study, Oxford Bulletin of Economics and Statistics, 42, 3, 1980. pp. 203-226.

Lall, S. (eds.), Transnational Corporations and Economic Development, London, New York: Routledge 1993.

Lall, S., Competitiveness, Technology and Skills, Cheltenham: Edward Elgar 2001.

Liu, X./Siler, R/Wang, C./Wei, Y., Productivity Spillovers from Foreign Direct Investment: Evidence from UK Industry Level Panel Data, Journal of International Business Studies, 31,3, 2000, pp. 407-423.

Markusen, J. R./Venables, A. J., Foreign Direct Investment as a Catalyst [or Industrial Development, European Economic Review 43, 2. 1999, pp. 335-356.

McAleese, D./McDonald, D., Employment Growth and the Development of Linkages in ForeignOwned and Domestic Manufacturing Enterprises, Oxford Bulletin of Economics and Statistics. 40, 4, 1978, pp. 321-340.

McIntyre, J./Narula, R./Trevino, L., The Role of Export Processing Zones for Host Countries and Multinationals: A Mutually Beneficial Relationship?, The International Trade Journal, 10, 4, 1996, pp. 435-465.

Narula, R., Multinational Investment and Economic Structure: Globalisation and Competitiveness. London. New York: Routledge 1996.

Narula, R./Sadowski, B., Technological Catch-up and Strategic Technology Partnering in Developing Countries, International Journal o[ Technology Management. 23, 6, 2002. pp. 599-617.

Ozawa, T., Japan: The Macro-IDP, Meso-IDPs and the Technology Development Path (TDP), in Dunning, J. H./Narula, R. (eds.), Foreign Direct Investment and Governments, London, New York: Routledge 1996, pp. 423-441.

Papanastassiou, M./Pearce, R., Multinationals, Technology and National Competitiveness, Cheltenham: Edward Elgar 1999.

Perez, T., Multinational Enterprises and Technological Spillovers: An Evolutionary Model, Journal of Evolutionary Economics, 7, 2, 1997, pp. 169-192.

Phelps, N. A., Branch Plants and the Evolving Spatial Division of Labour: A Study of Material Linkage Change in the Northern Region, Regional Studies, 27, 2, 1993, pp. 87-101.

Poon, T. S.-C., Dependent Development: The Subcontracting Networks in the Tiger Economies, Human Resource Management Journal, 6, 4, 1996, pp. 38-49.

Porter, M. E., The Competitive Advantage of Nations, New York: Free Press 1990.

Ruane, E/G6rg, H., The Impact of Foreign Direct Investment on Sectoral Adjustment in the Irish Economy, National Institute Economic Review, 160, 2, 1997, pp. 76-86.

Sambharya, R./Banerji, K., The Effect of Keiretsu Affiliation and Resource Dependencies on Supplier Firm Performance in the Japanese Automobile Industry, Management International Review, 46, 1, 2006, pp. 7-37.

Scott-Kennel, J., Foreign Direct Investment: A Catalyst for Local Firm Development?, European Journal of Development Research, 16, 3, 2004, pp. 624-652.

Scott-Kennel, J./Enderwick, E, Inter-Firm Alliance and Network Relationships and the Eclectic Paradigm of International Production: An Exploratory Analysis of Quasi-Internalisation at the Subsidiary Level, International Business Review; 3, 4, 2004, pp. 425445.

Sjoholm, F., Technology Gaps, Competition and Spillovers from Direct Foreign Investment: Evidence from Establishment Data, Working Paper No. 38, European Institute of Japanese Studies 1997.

Sun, H., Direct Foreign Investment and Linkage Effects: The Experience of China, Asian Economies, 25, 1, 1996, pp. 5-28.

Sun, H., Macroeconomic Impact of Direct Foreign Investment in China: 1979-96, World Economy, 21, 5, 1998, pp. 675-694.

Svetlicic, M./Rojec, M., Foreign Direct Investment and the Transformation of Central European Economies, Management International Review, 34, 4, 1994, pp. 293-312.

Tavares, A. T./Young, S., Sourcing Patterns of Multinational Subsidiaries in Europe: Testing the Determinants, Proceedings of the UK Academy of International Business Conference, University of Central Lancashire 2002.

Turok, I., Inward Investment and Local Linkages: How Deeply Embedded is 'Silicon Glen'?, Regional Studies, 27, 5, 1993, pp. 401-417.

Van Den Bulcke, D./Verbeke, A. (eds.), Globalization and the Small Open Economy. New Horizons in International Business, Cheltenham: Edward Elgar 2001.

UNCTAD, World Investment Report 1999: Foreign Direct Investment and the Challenge of Development, New York, Geneva: United Nations 1999.

UNCTAD, World Investment Report 2001: Promoting Linkages, New York, Geneva: United Nations 2001.

UNCTAD, World Investment Report 2002: Transnational Corporations and Export Competitiveness, New York, Geneva: United Nations 2002.

UNCTAD, World Investment Report 2003: FDI Policies for Development: National and International perspectives, New York, Geneva: United Nations 2003.

UNCTAD, World Investment Report 2005: Transnational Corporations and the Internationalization of R&D, New York, Geneva: United Nations 2005.

Williams, D., Strategies of Multinational Enterprises and the Development of the Central and Eastern European Economies, European Business Review, 97, 3, 1997, pp. 134-138.

Wong, P.-K., Technological Development through Subcontracting Linkages: Evidence from Singapore, Scandinavian International Business Review, 1, 3, 1992, pp. 28-40.

Author

Dr. Joanna Scott-Kennel, Senior Lecturer International Business, Victoria University of Wellington, Wellington, New Zealand.

Manuscript received October 2004, revised October 2005, final revision received August 2006.
Table 1. Measurement of Foreign Affiliate Characteristics

Affiliate Characteristics Measurement

Motive for Investment
 1 Resource-seeking Survey questionnaire asked respondents
 2 Market-seeking to indicate the extent to which each of
 3 Efficiency-seeking these five motives were reasons for
 4 Strategic-asset seeking investment into New Zealand on a 5-point
 5 Trade supportive Likert scale where 1 = not at all,
 2 = minor, 3 = moderate, 4 = major or
 5 = only reason for investment.

Main Business Activity Where 1 = yes, 0 = no
 6 Service Provider --service provider
 7 Sales Office --local sales office or branch of parent
 8 Manufacturer MNE
 --manufacturer or primary processor

Reliance on Parent Respondents indicated their affiliate's
 9 Technology transfer reliance on transfers of 1. product
 technology, 2. production technology,
 and 3. R&D, from their parent MNE on
 5-point Likert scales where 1 = not at
 all, 2 = minor, 3 = moderate, 4 = major
 or 5 = total reliance. These three items
 were combined into a factor score for
 technology transfer.

Competitive Position Respondents indicated that their
10 ... in New Zealand affiliates were 1 = not a competitor,
11 ... internationally 2 = a minor competitor, 3 = an average
 sized competitor, 4 = a major competitor
 or 5 = the only competitor.

Competitive Environment Where 1 = major decrease, 2 = decrease,
12 Industry competition 3 = no change, 4 = increase or 5 = major
 increase in the level of competition in
 the affiliate's industry in New Zealand
 in the past three years.

Ownership Form Where 1 = yes, 0 = no
13 MNE subsidiary --Subsidiary, branch or affiliate
14 Acquired local firm established by a foreign MNE
 --Former locally-owned firm acquired by
 foreign MNE

Age
15 Age --Number of years since first
16 Foreign ownership establishment in New Zealand
 --Number of years owned by current
 foreign MNE

Autonomy
17 Operational (short-term) The management of the affiliate has 1 =
18 Strategic (long-term) none at all, 2 = minor, 3 = moderate,
 4 = major or 5 = total influence over
 short/long-term decision-making.

Size
19 Employees --Number of full-time equivalent staff
20 Sales (FTEs).
 --Total sales value last financial
 year, NZ$ million.

Country of Origin
21 USA/Canada Country of origin of the ultimate parent
22 Australia where 1 = USA/Canada, 2 = Australia,
23 Japan 3 = Japan, 4 = UK/Ireland, 5 = other
24 UK/Ireland Europe, or 6 = other Asia/Pacific.
25 Other Europe
26 Other Asia/Pacific

Industry
27 Accommodation,
 cafes & restaurants Industry classification based on
28 Agriculture, forestry Australia New Zealand Standard
 and fishing Industry Classification (ANZSIC).
29 Communication services
30 Construction
31 Electricity, gas & water
32 Finance & insurance
33 Manufacturing
34 Mining
35 Personal & other services
36 Property & business
 services
37 Retail trade
38 Transport & storage
39 Wholesale trade

Table 2. Linkage Constructs and Factor Analysis

 Rotated Factor Matrix

Linkage Constructs Factor 1 Factor 2 Factor 3
Items measured Competitive Forward Backward

Competitive Linkages
Affiliate's influence over the:
 1 level of competition between 0.752 0.069 0.096
 firms
 2 number of major competitors 0.859 -0.011 -0.056
 3 number of smaller competitors 0.795 0.091 -0.096
 4 competitiveness of other firms 0.846 0.007 0.044
 ... in their industry in New
 Zealand (where 1=none at all,
 2=minor, 3=moderate, 4=major or
 5=complete influence)

Forward Linkages
Number of:
 5 agent relationships 0.020 0.604 0.023
 6 customer relationships ... 0.039 0.773 0.168
 formed by the affiliate in
 New Zealand;
 7 types of assistance and 0.085 0.600 -0.127
 resources * transferred to
 agents/customers by the
 affiliate

Backward Linkages
 8 Number of supplier/subcontractor 0.004 0.303 0.550
 relationships formed by the
 affiliate in New Zealand
 Reliance by the affiliate on -0.009 -0.026 0.851
 other firms based in New Zealand
 for:
 9 specialised products 0.000 -0.016 0.786
10 specialised services
 (where 1=none at all, 2=minor,
 3=moderate, 4=major, or 5=total
 reliance).

Contractual Linkages
Number of:
11 licensee relationships 0.021 0.121 0.043
12 franchisee relationships ... 0.026 -0.054 0.080
 formed by the affiliate in
 New Zealand
13 types of assistance and -0.023 0.064 -0.059
 resources * transferred to
 licensees/franchisees by the
 affiliate

Collaborative Linkages
Number of:
14 collaborative agreements formed 0.015 0.093 0.101
 by the affiliate in New Zealand,
15 types of assistance and 0.009 0.109 0.018
 resources ** transferred to
 New Zealand partner

Initial Eigenvalues (summary) 2.661 1.311 2.307
Total variance explained (percent) 18.31 8.74 15.38

 Rotated Factor Matrix

Linkage Constructs Factor 4 Factor 5
Items measured Contractual Collaborative

Competitive Linkages
Affiliate's influence over the:
 1 level of competition between 0.069 .0082
 firms
 2 number of major competitors -0.022 -0.025
 3 number of smaller competitors -0.017 -0.053
 4 competitiveness of other firms -0.009 0.038
 ... in their industry in New
 Zealand (where 1=none at all,
 2=minor, 3=moderate, 4=major or
 5=complete influence)

Forward Linkages
Number of:
 5 agent relationships 0.103 0.062
 6 customer relationships ... 0.053 -0.013
 formed by the affiliate in
 New Zealand;
 7 types of assistance and -0.107 0.252
 resources * transferred to
 agents/customers by the
 affiliate

Backward Linkages
 8 Number of supplier/subcontractor 0.126 -0.073
 relationships formed by the
 affiliate in New Zealand
 Reliance by the affiliate on -0.122 0.097
 other firms based in New Zealand
 for:
 9 specialised products 0.089 0.067
10 specialised services
 (where 1=none at all, 2=minor,
 3=moderate, 4=major, or 5=total
 reliance).

Contractual Linkages
Number of:
11 licensee relationships 0.431 0.286
12 franchisee relationships ... 0.812 0.001
 formed by the affiliate in
 New Zealand
13 types of assistance and 0.746 0.054
 resources * transferred to
 licensees/franchisees by the
 affiliate

Collaborative Linkages
Number of:
14 collaborative agreements formed 0.065 0.873
 by the affiliate in New Zealand,
15 types of assistance and 0.148 0.887
 resources ** transferred to
 New Zealand partner

Initial Eigenvalues (summary) 1.207 1.593
Total variance explained (percent) 8.05 10.62

* Types of assistance and resources include: product component
(or service) specifications, product samples or prototypes, production
(service delivery) technology, equipment, technical assistance, testing
and quality control procedures, inventory or services systems
management, staff training, information about markets, suppliers,
contacts, assistance with acquiring inputs, managerial assistance,
trade or exporting assistance, financial assistance, and other.

** Types of assistance and resources include: product (service)
technology, production (service delivery) technology, research and
development, management practices/culture, marketing systems,
distribution systems, employment practices, human resources and skills,
training (or training systems), economies of scale or scope,
specialised inputs, information, experience and expertise, access to
markets, finance, and other.

Table 3. Final Linkage Cluster Centres

 Cluster Number

Linkage Factors 1 2 3

Competitive -0.921 0.819 0.096
Forward -0.128 -0.257 2.211
Backward -0.965 -0.753 0.304
Contractual -0.054 -0.077 -0.134
Collaborative -0.216 -0.369 -0.486
Number of affiliates 86 116 45
in cluster (n=516)
Percent of total 17 22 9
sample

 Cluster Number

Linkage Factors 4 5 6 7

Competitive -0.976 0.793 0.092 0.132
Forward -0.388 -0.339 -0.120 0.315
Backward 0.588 1.022 0.380 -0.153
Contractual -0.247 -0.216 4.366 -0.425
Collaborative -0.159 -0.238 0.610 2.156
Number of affiliates 105 93 20 51
in cluster (n=516)
Percent of total 20 18 4 10
sample
COPYRIGHT 2007 Gabler Verlag
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Scott-Kennel, Joanna
Publication:Management International Review
Article Type:Report
Geographic Code:8NEWZ
Date:Jan 1, 2007
Words:11632
Previous Article:Host country cultural influences on foreign direct investment.
Next Article:Foreign subsidiaries' learning from local environments: an empirical test.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters