How Does an Interfirm or Co-Localized Training System Arise?

The subject of training carried out by, and offered to companies, is global and involves entities of all sizes in all industries and in all places on the planet where humans live and work. Not surprisingly, it is a topic that has resulted in an enormous research output: our bibliometric analysis produced 27,296 results. Of course, the research we have undertaken has by necessity imposed geographic and socioeconomic limits, among others. Some of the main concerns associated with this general field of research are on-line training in different forms, evaluation and effectiveness of training, human talent management (linked to issues such as job satisfaction and career paths), management development, leadership, and organizational culture (see Figure 1). Several of these themes are not addressed in this paper as they lie outside our focus of attention.

Source: www.webofscience.com

Figure 1 Training between companies: results of 27,296 documents 

The evolution of a local, sectoral, or regional training system often originates in particular firms which have identified weaknesses in the competence of employees and skills gaps which bar them from taking advantage of local resources and the equipment and techniques that are available to exploit them. At this juncture, there exist challenges which perhaps for participants who join the system later are less inhibitive: vaguely defined trajectories and learning curves, limited economies of scale and experience, constrained financial resources due to, among other reasons, a lack of precedent, and so on. In this sense, Schumpeterian patterns of training system development may occur (^{Karniouchina et al., 2013}), beginning with high uncertainty and low entry barriers, and the leadership of larger firms (^{Biais et al., 2015}), with the increasing participation of smaller firms exploiting opportunity conditions later on (^{Liedtka et al., 2017}).

There are bound to be firm-specific, idiosyncratic differences in terms of timeliness, comprehensiveness of utilization, and level of impact with regard to the shared training system evolved in the horizontal dimension among neighboring firms and firms operating within the same industry, and in the vertical dimension among firms that engage in complementary activities. We may see this, for example, in the repercussions such a spillover might have on the horizontal plane compared with its impact on the vertical dimension of the chain. A situation may also exist in which firms with a similar operational focus have very different current skill levels or gaps. After all, the level of participation of a firm that, in terms of capacity building, is relatively immature, depends on its position in what could be termed the skills development lifecycle (similar to comments made by experts such as ^{D’Este et al., 2013}, about the industry lifecycle).

As the firm matures in terms of its training commitments, there may be an increasing division of labor leading to a demand for additional training capabilities. Again, this will impact on internal provision, whether an interfirm training system exists, and on other institutions operating locally or regionally in the field. In this case, the firm or firms with the most advanced skills and training endowment may have more to give the weaker firm(s) without actually benefiting much from the relationship in such a process. As cooperation involves interfirm openness and willingness, the temptation might be to reduce or cease this interaction precisely because of this asymmetry and perhaps the anticompetitive nature of the activity. On the other hand, it is quite possible that, all parties being happy with the arrangement, the interaction continues to the extent that greater specialization and breadth of offer occur.

However, certain firms may consider it unwise to take this step into a more-specialized skills formation, electing instead to engage in enhancing organizational integration as an antidote to a fragmented group of now-skilled employees lacking a proper system of interdependence and feedback (^{Foss et al., 2015}). Once the process of integration has been completed, the system may have a greater capacity to change or expand into new areas in a timely way that is not detrimental to its structure, existing core training, or other competencies. It may also give new specialized providers the opportunity to emerge or for old ones to modify their operations in response to the changing circumstances demanded by what may be a major customer in the local skills development market (^{Sharma, 2014}).

Likewise, the firm may also do this because it has run up against the problem of the absence of competent external capabilities for which a substitute for internal resources does not exist. A very good reason to slow down the evolution of the training system would involve a change of policy away from directed skills formation towards targeting skilled people to perform tasks immediately, in preference to waiting for the entire duration of a training program, which may indeed turn out to be a failure in terms of competency needs in the end. Competency-creation is a long-term commitment often without a guarantee of satisfactory results, while the transaction costs and recognized current abilities of a skilled recruit are clear and immediate. On the other hand, should the firm have a well-developed organization and be confident of its skills-formation trajectory, and if both internal and/or external training capabilities have the required range of competence, then a higher level of specialization may occur. The general skills level of current and incoming employees, and a reduction in the importance attached to relatively basic, general training, would also be important issues in this respect.

For the training cluster to function well, each firm should analyze how its internal capability is evolving so that the subsequent exchange of training knowledge between firms and other stakeholders can change appropriately over time. This requires long-term discipline, an adequate communication capability, and the right people and procedures. Another way to approach this theme is to recognize that, in effect, each firm has an “absorptive capacity” (^{Cohen & Levinthal 1989, p. 569}; see also ^{Ashford & Hall, 2011}) that relates to its ability to identify, take on, and exploit knowledge and skills from both internal and external sources. So, again, while externally the local cluster training inputs and spillovers might be quite frequent and accessible, whether a firm can take advantage of this is a question related to its own personnel, management structure and main activities. In practical terms, the influence of the cluster organization itself on intrafirm learning, it appears, has not yet been well investigated.

Is there a positive impact on training performance derived from geographical proximity of neighboring firms and training providers?

We carried out a simple bibliometric analysis of this specific topic of interfirm training. We obtained the following results: that the main concepts (all very generic) connected with vocational training are industrial structure, capitalism, labor relations, production structure, and, of course, cooperation between companies (see Figure 2).

Source: www.webofscience.com

Figure 2 Interfirm training: results of 21 documents 

There may be a stock of skills-development knowledge in a particular locality, based on an agglomeration of firms, providers, and other related agents located close together (^{Malerba & Orsenigo, 2000}; ^{Halász, 2011}). More specific approaches to analyzing this theme have been undertaken by ^{Mukhopadhyay et al. (2014, on regional networks)}, ^{Biao et al. (2013, on learning regions)} and ^{Arif (2012, on collective efficiency studies)}. They are characterized by their emphasis on joint action by collaborators, based on mutual trust and supporting institutions, leading to competitive advantage. Though it is only mentioned at most as one element of the mix, the collective evolution of a local training system is a significant part of this development. However, some provisions should be mentioned right from the start. The fact that this closeness of agents exists—with similar intrafirm learning mechanisms (^{Ibrahim et al., 2019}), capacity building activities, organizational system and shared needs—does not necessarily mean that the resulting learning pool or system will be accessed if (i) the firm has limited capability to use it, (ii) it is not what the firm requires, or (iii) the firm itself has its own resources and internal system with which it is reasonably satisfied.

If, like organizational structures, skills are a relatively immobile resource, then the proximity of collaborators and providers, and their capabilities, are central to the development of the training offer to the firm itself (this runs parallel to the immobility of organizational structures, as discussed by ^{Breschi and Malerba, 2001, p. 817}; and ^{Sheffi, 2012}). It is therefore likely that a particular co-localization of dynamic, reasonably funded, and underskilled firms will facilitate a comparatively well-developed cluster of providers and even in-house training departments. Furthermore, as an alternative, it could also be the case that a lot of training can be transmitted in codified form over large geographical distances without interpersonal mediation and interfirm mobility of personnel (^{Shafaeddin, 2012}). After all, upstream firms are often located in a particular place not for reasons of proximity to skills and the institutes which furnish their means of development but because of the natural resources that are found there. This is a situation that contrasts with the one in which a downstream firm is devoted to the production of innovations and is located in a place where essential knowledge inputs are available (as described in studies such as that of ^{Tavassoli & Carbonara, 2014}).

At another level, related to what firms learn from each other, some sort of organized or irregular association—resulting in such benefits as an exchange of best practices (for example)—may play a key part in the internal learning process involving training emphases and general performance in a given firm. The dynamics for this could be interfirm knowledge exchange and skills catch-up.

What is the nature of the interfirm training cluster?

A cluster in this context is a situation among neighboring firms of mutual search or support for the provision of training. It is inevitably a very important research field: according to the bibliometric analysis that was carried out, we ended up with 15,216 results (see Figure 3). The main themes include community, intervention and implementation; employability, educational cluster, curriculum, self-regulation, teaching and learning strategies, and classification. Another search—on training and proximity—yielded 1,963 results (see Figure 4): the most interesting topics in the context of this current research are social cognition, deep learning and communication.

Source: www.webofscience.com

Figure 3 Training cluster: 15,216 document results 

Source: www.webofscience.com

Figure 4 Training and Proximity: Results of 1,963 documents 

A group of firms (even companies from the same industry) and other bodies, operating in the same region, might come together because the opportunities to innovate represent the optimum training system among themselves and because of the absence of strong competitive concerns. The commitment or embeddedness of each participant in this endeavor is a reflection of the perceived benefit that they have already experienced or expect, but also the intervention of individuals in the decision-making process, favoring participation in the network sometimes in the face of objections. This rests on their belief in the advantages of working together rather than in isolation. Likewise, the firm’s ability to avail of the local skills provision depends on good social links, effective communication, and often active sponsorship. Obviously, the greater the emphasis on face-to-face learning—involving classrooms, OTJ, simulators, etc.—and the lesser the use of digital training methods, means that the issue of geographical proximity and localization are fundamental in addressing the training model created and its development over the course of time.

In sum, once some degree of joint action happens in the geographical or sectoral agglomeration of companies, then firms are open to create collective advantages: decisions on training focus and level, “training the trainers,” better and more specialized providers, facilities, and infrastructure, access to information and exchange of experiences, openness to sources outside the cluster, and so on (^{Sheffi, 2012, pp. 209-236}). In an ideal situation, these collective economies managed by joint efforts may lead to a “collective efficiency” (^{Schmitz & Nadvi, 1999}; ^{Newman et al. 2016}) of the generation and diffusion of training knowledge, and its practical implementation.

In the case of strong interfirm relations—good examples would be the case of a major firm and supplier/subcontractor firms, or a cluster engaged in interconnected activities within a trade association (^{Bramoullé et al., 2014}; ^{Lundvall, 2016})—there certainly could exist an incremental learning process, by example or dictate, to develop the training activities among them, based on the clear understanding that the collective organization of human capital formation enhances the skills within a shared skills-base and labor market; indeed this is part of the thinking that also sees shared knowledge and norms as working to everyone’s benefit (^{Kapunda, 2017}). This effort can be accomplished if there already exists a high-trust culture and legal regulation (^{Shafaeddin, 2012}). There might exist a certain level of agreed, formal development of this situation or (perhaps harder to appreciate) this process may lead to an informal growth and routinized coordination of the shared training system—a phenomenon that could arise, for example, in the absence of a more supportive system in a developing country or a larger agglomeration of firms and providers.

The more non-competitive and non-rival the character of the relationship between firms, the better the opportunity of some degree of informal/random or formal/organized participation in the local training system. However, it would be naive to expect that this will happen inevitably, or achieve the best possible results without other interventions, since such issues as the combination of tacit spillover and geographical distance in the transmission of a training culture, and even the establishment of physically close skills operations may not be sufficient to bring this about (^{Breschi & Lissoni, 2001, pp. 979, 988}; ^{Ottaviano & Peri, 2011}; ^{Qian & Acs, 2013}). Spatial proximity is important but of equal significance are the interplay of training knowledge codification, labor market weaknesses and strengths, the economics of knowledge transmission (as described for example by Rallet & Torre, 2000, and ^{Taylor, 2009}), and enterprise strategies that consciously or inadvertently result in a collective participation in the local training system.

The cluster could evolve a selective/adoptive/innovative mechanism that creates the common training system by a process of selection, imitation, variation, and monitoring of identified solutions (^{Maskell, 2001, p. 930}; ^{Galbreath et al., 2014}). At the same time, this process cannot abandon an individual perspective: the training offer must be aligned with the self-defined objectives of each enterprise, facilitated or delimited as this may be by a specific institutional endowment. Thus, the activities undertaken by the firms in the group will define what is learned, while the individual firms themselves will dictate specialized focuses and how these will be learned (^{Lundvall, 2016}). What prevents inertia from setting in among a group of firms thus engaged in training is their specific in-house demands/needs and developments pushing for new training answers—the sometimes “competing visions” that, when well-managed, maintain the dynamism of the shared capacity-building project (^{Loasby, 2001}; ^{Piazza, 2010}). Another antidote to stagnation is the regular entry of new stakeholders into the network.

The external training environment may well change in a very dynamic way, evolving in response to market forces (e.g., a firm may now want management courses rather than programs for electricians), government support and structures, the participation of individual local trainers or agencies of training provision, and so on. Simply put, the situation may progress from the initial formation of an immature training system, to growth and stabilization of a more mature system, and end in similar stages of expansion/stability, replacement or decline. The changing patterns of agent dominance and dependency, training market entry/participation/departure, and business volubility will all have a say in this. Depending on capabilities and needs internal to the firm and the interfirm grouping, the “mature” scenario could be one of relative inertness in a configuration involving non-dynamic firm(s) and its agents, limited dynamism between a growing firm(s) and its collaborators, and innovative dynamism in the relationship of a booming firm(s) and its necessarily highly responsive agents (an interpretation partly modelled on ^{Coombs et al., 2003, pp. 1131-32}; and ^{Baglioni and Sinclair, 2018}). Regarding long-term, external ramifications, the latter could be seen as the one with the greatest potential to push the range and scale of the training provision. However, the question concerning its efficacy would have to be addressed as there is no guarantee that sheer advancement is a matter of real best practices and benefit maximization.

The training provision that arises in the context of one single industry dominating a region or cluster is bound to be specialized and therefore limited in scope. This occurs because of demand-led forces dictating what is offered both within firms and outside them. On the other hand, if a broader range of industries are present, the provision should have a greater variety. This consideration is no small matter: in terms of capacity building, the influence of a cluster or co-localized firms can be very powerful (as Maskell concludes, clusters are “the territorial configuration most likely to enhance learning processes”; ^{Maskell 2001, p. 922}). In this sense, localized learning involving skills enhancement can create an oasis of provision that may or may not enrich the local skills pool and its specific skills gaps. However, it is not clearly efficient to supply a lot of human capital formation (as governments and NGOs have done) in the hope that the commercial and industrial activities of the locality will then inevitably be upskilled and diversify (^{Draxler, 2014}).

A given cluster of firms engaged in training might be more dynamic if there exists among them a knowledge leader or gatekeeper, certainly by comparison to more haphazardly organized clusters. It is a fact that a situation of reasonably active collective training efficiency could arise, based on structured collaboration that facilitates and stimulates better performance of capacity building.

What is the role played by foreign firms and multinationals?

Over the years, a large number of multinational companies have developed a culture of training their current staff or new recruits to achieve different efficiencies, stability of standards, harmony and retention of staff, capacity for managed and timely response to change, among other concerns. At best, this will mean that companies have the openness, if not the active ability, to diagnose what training is required to expand existing operations or launch new ones. In turn, this means that their example of cultural and practical training, whether in-house or provided by outside or local training suppliers, might have a positive effect on the locality or even the region, as long as everything works well.

Through a bibliometric method it was discovered that a large number of publications have dealt with this general field, with our search producing 2,955 results (see Figure 5). Topics include: human resource management and intercultural management; certification, sustainability and corporate social responsibility; innovation; various aspects of virtual learning; and, of course, globalization and developing countries.

Source: www.webofscience.com

Figure 5 International companies and training: results of 2,955 documents 

It has been argued that, since linkages with local companies are often weak, the main impact of foreign firms on their local equivalents in terms of skills competency is achieved through internal human capital formation, enabling their own employees and trainees to operate new machines and participate in modern working practices, and from there indirectly influence other nearby firms (^{Blomström & Kokko, 2003}; ^{Makki & Somwaru, 2004}; ^{Goedhuys, 2007, pp. 287-288}). All this does not rule out the possibility that an improved training regime may appear locally, at least partly as a result of the close presence of a developed training system within a large and dynamic firm. No firm is totally self-sufficient when it comes to the creation of a training regime and its subsequent operation and evolution: interaction and collaboration take place with stakeholders ranging from trainers and government supervisors to local community leaders (^{Edquist, 2006}; ^{Lundvall, 2016}). There is also no denying the strong influence a firm (whether foreign or otherwise) might have on its suppliers and subcontractors (^{Javorcik, 2004}; ^{Malik, 2015}), in terms of its role as a model of the benefits of a solid training culture and skills-level expectations of its partners.

Similarly, there also exists the question of a skills development spillover between the foreign firm and its local neighbors, with some authors arguing that there seems to be little evidence that it occurs effectively, particularly if the gaps in organizational sophistication, processes, institutional policy, and technological advance are considerable. A spillover could be expected to be easier if these gaps were closer. In this regard, it is part of the brief of this study to examine if certain authors are correct to argue that there seems to be little evidence that this occurs effectively (^{Ibrahim et al., 2009}; ^{Buch et al., 2014}; ^{Chen et al., 2015}).

The provision of training may be path-dependent (^{Sood et al., 2012}) if the firm is a subsidiary of a block of enterprises (e.g., part of a multinational) and is thus endowed with a ready-made and previously proven system; if the local context has a reasonably well-developed training provision and much use is made of it; or if the internal training system is established with relatively fixed ideas concerning how provision should be created. But neither ^{Sood et al. (2012)} nor ^{Malerba (2002}) devote enough discussion to how certain firms and indeed sectors may be quite rigid in this regard (as a reflection of such considerations as the high costs and risks involved as well as training habituation), while others, particularly “evolving” sectors, may be more open to innovation and trial-and-error experimentation. Of course, a major factor behind the evolving nature of the training provision is the firm-wide dynamism of innovation that is permitted to take place (these remarks were developed from some conclusions made by ^{Coombs et al. (2003, p. 1126)} and ^{Baglioni and Sinclair (2018)}. In short, how the training is developed is a question of the given conditions and the challenges they present, the capabilities (of funds, analytical capacity, personnel, structure, confidence in training, etc.) inherent within the firm, the competence and the nature of the relationship between the participating agents (e.g., whether this is cooperative or competitive), and the regularly assessed success or failure of the endeavor.

A parent firm might find that a subsidiary has enacted successful training processes and routines: in fact, the smaller company has been a laboratory for experiments in human training, on a scale that is definable, scalable, and reasonably credible. However, replication might not be easy to achieve. Not only is it hard to put together a single operation and discrete pieces of knowledge on paper in a systematic/ unified (codified) way, but the parent company often fails to replicate successful experiments, not taking into account the importance of organizational knowledge and structure (^{Szulanski & Winter, 2002, pp. 62-63}).

In the context of this study, the participation of multinational corporations in making localized decisions (possibly over the heads of local management and training coordinators) is an important issue. We should keep in mind that inward foreign direct investment (FDI) is responsible for a very sizable part of gross fixed capital formation in less developed countries (LDCs) and in all developing countries; as such, it has a very strong influence on all types of learning and capability accumulation. In one sense, the perspective at the multinational level might lead the firm to view collaboration with other firms as a breach of self-interested policy, while at the local level, the view might be that there are few competitive reasons against collaborating in this and other areas, and many others in favor. This develops arguments presented by ^{Breschi and Malerba (2001, p. 822)}. The top-down intervention might then be classified as under-informed and biased. Of course, the practices involved in globalization can support and reinforce local ways of doing things, just as the activities and knowledge originating in and tested at the local level may feedback very positively into the “globalized” training knowledge pool, facilitated as it may be by very direct communication links as exist in a multinational enterprise (^{Breschi & Malerba, 2001}, speak along these lines, as do ^{Wheelahan & Moodie, 2016}).

Studies have been conducted that support the argument that foreign firms have stronger vertical links with other enterprises, and that they tend to invest more in human and physical capital (e.g., ^{Goedhuys, 2007, p. 281}). Local firms tend to substitute in-house alternatives and have less formal connectivity with other local firms (^{Newman et al., 2016}). The reason for the higher prevalence of training among foreign companies can be explained by the transfer of equipment and working practices (and its demands on skills) from the parent company, the existence of manuals and other sources of information, the institutional training policy; and the availability of a training culture, transferable and adaptable system, funds and highly mobile training personnel. The question here relates to the possible higher provision and quality of the training offer among foreign firms, versus the perhaps better training fit that a firm can achieve through being more locally embedded.

Of course, lacking a sense of local loyalty and adequate local knowledge, a foreign firm might calculate that expatriate labor complies with its pressing needs. However, the presence of skilled workers imported from outside and the desire to have new colleagues meet their standards in an effort to reduce productive inefficiencies and human resource weaknesses due to a lack of uniformity in capability, might be a good lesson to engage in skills development, not to mention the corporate policy demanding a uniform high-skills level, quality production and focus on comparatively more demanding foreign markets.

What is the role of training providers?

Naturally, it is not possible to address the general topic of this paper without talking about training providers. The number of publications on this topic confirms its importance (this research found 24,617 references), and there is also a great abundance of sub-topics related to the role of providers from different locations, training approaches, levels of competence, teaching practices and evaluation methods/techniques, types of clients and of students/trainees/graduates, and so on (see Figure 6). Again, it should be mentioned that a large part of the results of the bibliometric analysis is not directly related to the topics on which this study focuses. However, it does show the magnitude of interest in this field. Then, among the results, the elements that seem to be of greatest interest in the context of this work are simulation, evidence-based practice, competencies and curriculum, collaboration, implementation, review, quality improvement, attitudes, training for providers, quality and evaluation, and adult education.

Source: www.webofscience.com

Figure 6 Training providers: results from 24,617 documents 

If the training model is characterized by a high level of training demand and opportunity (such as would exist when a new mine is opened, for instance), it could be expected that new providers will appear, perhaps of untested competence and offering services not suited to the new consumers. A given level of instability would result if such factors as firm-level training leadership, clear communication of service needs, and provider quality assurance (imposed by customer-firms or by some government authority, for example) were not also in place. The firm or group of firms could help remedy this situation by market forces (as buyers they choose who gets their custom) or perhaps by direct intervention (e.g., advisory input for the provider).

The key element in the long term is persistence in this feedback or intervention, as providers will tend to be relatively passive if training knowledge input and formation path management through demand-pull forces are not very active. A contrary situation is one in which the firm or group of firms create their own in-house provision, independent of outside institutes (though sometimes with their participation); or decide to cancel or reduce their activities with external providers and resort to the type of in-company skills development that furnishes them with more control of their vocational and technical training (^{Arif, 2012}). Yet another scenario shows the possibility that a large firm or a group decides to establish a more or less autonomous institute that is directly and collectively funded and managed by them, perhaps with some involvement of a government ministry, semi-state organization, or NGO.

Let us look briefly at the competence level of the providers themselves, without getting involved at this juncture in their particular public or private nature. Some will have experience in the field of the firm; others will be experienced in other fields and perhaps, because of this experience, are ready to make the transition to respond to the requirements of their new customers; and lastly, others will have experience in a completely unrelated field and be unwilling or incapable of making the transition, or once the transition is made, be poorly prepared. Newly created training providers comprise yet another category and these should carry out their surveys and target their capabilities. One important point to be made about both established and new providers is that they may have to co-locate to sometimes very challenging places (e.g., mining firms). The absence of basic elements such as an existing infrastructure, combined with company guidance if not also active support, should inspire them to be innovative. This is a theme that needs to be better investigated in terms of location studies and agglomeration effects.

The emergence of more numerous providers, many of whom could be expected to be specialized and of higher relative quality than before, should make for wider provision that is easier to access, possibly enabling a more secure training investment and development experience, and incrementally improved training/ learning trajectory; there is even a chance that prices would go down (^{Caniels & Romijn, 2003, p. 1266}; ^{Van Long et al., 2014}). The specialization that may emerge could be due to two factors. The first one is obvious: employees have basic skills and now require intermediate and advanced training that is by definition more specialized. The other factor relates to firm-level idiosyncrasies: the training that corresponds to a particular enterprise may be, for that very reason, firm-specific, addressing core competences and their associated skills and routines (^{Leahy, 2012}, looks at this in terms of a firm’s competitive capacities and sustainable advantage). There is reason to assume that, given enough training market volume and continuity, and a lack of firm-level provision, a cluster of training providers would not come into existence to form a collective, preferred option for the firm cluster. However, this would depend on local business culture, a sense of comfort in group linkages, and the discipline not to give in too readily to temptation and disintegrate the training group. The momentum for this could come from a dominant firm or the cluster itself, or a government agency recognizing the scale advantages and perhaps the benefits to public agents participating in the provider group.

The practice already exists in industry, in general, for a few decades to outsource productive activities to reliable outside contractors. For reasons now of tradition and satisfaction with this generic activity, a firm might be emboldened to do the same when it comes to training. Likewise, smaller firms with currently limited in-house capabilities and without plans to change this in the near future, may also contract outside providers. Their inability to design and manage projects dealing with training systems, coinciding with a clear present need to develop core competences that resolve current weaknesses and facilitate effective business partnering with other firms of a higher skills level, makes this practice a good alternative. The downsizing of public provision in certain regions, and the ineptness of much of the rest, could also help in the growth of private provision led to some extent by related outsourcing.