Market coordination and training networks in Australia

Introduction

Australia, in common with other liberal market economies, has struggled to develop an adequate supply of workers with intermediate-level vocational skills. Recurrent skills shortages and employer underinvestment in vocational training have been persistent problems in Australia with poor completion rates of vocational training being evident and employers demonstrating a strong preference for non-formal, informal and on-the-job forms of skill development over formal vocational education (ABS, 2003, 2009; NCVER, 2011).

Governments in more liberal market economies, including the Australian government, have sought to address this failure to develop appropriate workforce skills through supply-side measures to stimulate training and through reforms to Vocational Education and Training (VET) systems to increase employer engagement. In the Australian case, government has funded an expansion of training places to expand the supply of training, whilst also reforming the provision of training to establish an employer-led VET system. Training markets have been created to provide for greater employer choice in the selection of training providers and, at the same time, curricula and training delivery have been deregulated so that firms can customize the content, delivery methods and assessment of training to suit their individual needs (Cooney, 2010; Wheelahan and Carter, 2001).

Despite the use of supply-side measures for over two decades, employer underinvestment in vocational training persists in Australia and this lends some support to analyses of skill development that focus upon the demand deficiencies endemic to liberal market economies (Finegold, 1996; Finegold and Soskice, 1988). In liberal market economies where employers have the option to train or not to train, underinvestment is driven by employer concerns over the poaching of skilled labour and the desire to control the costs of training that may not be incurred by competitor firms. Vocational training in such systems tends to be more firm specific and job related, as firms seek to limit the amount of transferable training that is provided. This creates a collective action problem for employers who may be unable to generate a supply of skilled labour with the requisite transferable skills (Hall and Soskice, 2001).

This undersupply of vocational skills in liberal market systems stands in contrast to the situation in more coordinated market economies where occupational standards and vocational training are more highly regulated by government and employers are obliged to join associations that control the flow of vocational training funds. Social partner institutions at national, sectorial and firm levels are responsible for reaching agreement about the structure of vocational qualifications and the offer of training, with the latter partly funded by employers through compulsory levies (Le Deist and Winterton, 2012; Méhaut, 2005, 2007). In the more coordinated economies social institutions provide for forms of collective action by employers that are not evident in more liberal market economies (Crouch and Trigilia, 2001; Crouch et al., 1999).

In the search for ways to stimulate employer demand for training in more liberal market economies and improve employer coordination of the training effort, voluntary network forms of association have recently been promoted. Voluntary associations in networks, it is argued, provide a mechanism that can be used in liberal market economies to address the collective action problem of employers. Voluntary network arrangements lead to greater inter-firm cooperation, resource sharing and joint action to address labour market needs. Such cooperation can reduce the costs of vocational training and improve quality, to provide for the development of transferable skills. Voluntary association in networks may thus provide the benefits that more coordinated forms of cooperation generate, without the need for direct state intervention (Erickson and Jacoby, 2003; Grimshaw et al., 2010).

In order to be effective voluntary network arrangements need to develop dense, high quality linkages between firms; linkages that lead to the development of trust and reciprocity in the network. These dense linkages are the foundation for joint problem solving, thick information transfer, resource sharing and cooperation which can, in turn, support the development of joint action in the provision of training (Granovetter, 1985; Uzzi, 1996). The development of such voluntary networks is seen to be a feature of inter-firm relationships in the knowledge economy with a range of new business practices supporting the emergence of network linkages in an ‘associational’ economy (Cooke and Morgan, 2000).

The practical effect of this focus upon networks is that governments and employer groups in liberal market economies such as Australia, the US and the UK have shown an interest in the ways in which forms of voluntary network association might be leveraged to support a greater and better coordinated training effort on the part of employers (AiG, 2006; Skills Australia, 2011). The possibilities of greater cooperation by employers to secure the joint provision of training have been demonstrated by a small number of national case studies but questions remain about the extent to which voluntary participation in network arrangements can solve the collective action problem of employers (BIE, 1995; DfES, 2002; DfES and CBI, 2005; Erickson and Jacoby, 2003).

Voluntary training networks in a liberal market economy are coordinated through the market since the business practices underlying network formation are market-based ones, such as the development of buyer–supplier linkages in supply chains. What effect these practices have on training provision is unclear and whether new business practices can adequately substitute for coordination by social institutions is also unclear. Do market-based training networks exhibit dense linkages leading to joint problem solving and resource sharing or does coordination through the market restrict the extent of inter-firm linkages? Questions such as this have been little explored in the rush to embrace the potential of networks but there is some evidence that voluntary networks that are coordinated through the market do develop in ways that limit the development of network linkages. Studies of the role of power in market-based networks have shown that coordination through the market limits network development by enabling some firms to benefit disproportionately from network activities, whilst other studies have demonstrated that voluntary networks tend to develop thin linkages between firms that lead to little resource sharing and joint activity (Bishop, 2011; Kinnie et al., 2005).

The translation of the possibilities inherent in voluntary network arrangements into concrete actions by employers remains largely theoretical then and few studies have examined the actual extent of voluntary association in networks for the provision of vocational training. Whether network arrangements can overcome the barriers to employer demand for training in liberal market economies and solve collective action problems is unclear. This article begins an exploration of this question by examining the nature and extent of voluntary networking for the provision of vocational training in five Australian industries. The article outlines the key business practices that are linked to network training before going on to discuss the findings of the study and draw some conclusions about network training in liberal market economies.

The importance of networking

Networks are forms of social organization that are embedded in face-to-face social relationships. These relationships provide the context for the development of trust and reciprocity between individuals. Networks thus provide a normative basis for exchange that is independent of markets (Granovetter, 1985, 2005; Uzzi, 1996, 1997). This socially embedded exchange is effective for the transaction of exchanges that are hard to value in the market. In particular, networks facilitate the transfer of less tangible goods, such as expertise. Thick information exchange based upon experience – such as the transfer of tacit knowledge and skills – is usually effected through socially embedded networks and is often not readily available in the market (Powell, 1990). Cooperation leading to resource sharing and joint problem solving is, likewise, normatively sanctioned in networks through such socially embedded exchange and is not normally available as a good to be transacted in the market.

This social exchange of hard to value goods is dependent upon the development of dense linkages between network actors. These linkages are a feature of closed networks where network membership is restricted and this is a common feature of many market-based networks. Only those firms engaged in market exchanges with each other have access to the network (Uzzi, 1996). In this study we examine buyer–supplier relationships in supply chains and the development of business to business relationships, as examples of this form of network. More open networks permit the establishment of dense and thin linkages but these are usually externally coordinated networks with looser controls on network membership. In this study we examine localized learning networks and third party service networks as examples of this form of network (Powell and Grodal, 2005).

All of these networks are dependent upon the establishment of market transactions between network members for the development of the network. Such networks do not exist in isolation from markets but rather are a feature of new forms of market relationships based upon new business practices. Networking based upon new business practices has assumed a strategic importance as firms seek partners to achieve a range of business outcomes. Business partnerships may be formed to establish a long-term supply of goods and services essential to business operations, they may be formed to gain the knowledge and skills required to develop new products and new processes, or they may be formed on a short-term basis to open up new business opportunities. Cooperation in networks serves a variety of business needs and objectives but where training provision is concerned, there are four new business practices that are associated with the development of networks for joint employer action and we now turn to an examination of these practices.

Integrated supply chains

For large firms, the development of integrated or just-in-time supply is dependent upon the development of a network of close, cooperative relationships between themselves, first tier, second tier and third tier suppliers. Cooperation has become a feature of buyer–supplier relationships as the traditional arm’s length relationship between firms based solely on price in the market is supplemented by the development of closer relationships mediated by criteria specified in contracts. The emergence of markets where firms contract with only a small number of business partners has led to the development of ‘relational contracting’ between firms (Cousins and Crone, 2003). The networks created by these relationships (see Table 1) are relatively closed with admission to the network being controlled by the lead firm. Dense linkages are developed between firms in the network based upon repeated and exclusive transactions (Gulati, 1995). These linkages lead to joint problem solving and thick information transfer that includes the sharing of tacit and explicit knowledge about products, technology and production processes. These dense linkages are organized in ways (e.g. supplier associations) that limit the possibilities for spillover outside the network. Exclusive supplier relationships lead to the creation of internally stable networks that promote cooperation, trust and the transfer of knowledge and skills between network members but permit few and only thin linkages beyond the network (Lincoln et al., 1998; Owen-Smith and Powell, 2004).

OPEN IN VIEWER

Table 1.

Features of business networks.

Network	Open network	Closed network	Dense linkages	Thin linkages
Integrated supply chains		X	X
Business to business relationships		X	X
Localized learning networks	X		X	X
Third party service networks	X			X

These embedded contractual relations have resulted in a change in the character of firms in supply chains. Firms have had to move from developing stand-alone production systems to developing production systems that are integrated along the supply chain. This change arguably has an effect upon the provision of training, for the integration of business processes across firms entails not only the integration of technological systems but also of production practices. Training for the skills required to use integrated technological systems and training for common production practices thus become a feature of supply chains. The supplier development practices of some multinational firms provide examples of this form of cooperation (Dyer and Hatch, 2006; Dyer and Nobeoka, 2000; Lundquist, 1997; Sako, 1999). Common training in integrated supply chains may be supplied by the lead firm in the supply chain or through third party organizations. For the small and medium sized supplier firms down the supply chain such training is desirable because each individual firm alone may be unable to provide the training required to integrate operations. Shared training and skill development may thus come to assume a greater significance in integrated supply chains leading to the sharing of resources and the solution of collective action problems.

Methodology

In order to identify the extent of network training and the characteristics of training networks in Australia, a survey of firms in five industries was undertaken. Training, for the purposes of the survey, included all formal and non-formal instructional activities, whether accredited for recognized qualifications, accredited for licensing and certification, or non-accredited. Training, here defined, did not include informal learning at work or informal instruction on-the-job.

A draft questionnaire was developed and presented to an expert reference panel for comment. The questionnaire was then piloted with a small sample of firms to check that the questions were meaningful to respondents and produced reliable information. The final survey sample included firms in five significant Australian industries: clothing and footwear; engineering; information technology; scientific and medical equipment; and processed foods and beverages. These industries, in the Australian context, range from sunrise through to mature industries and include both low and high technology production processes.

The data were collected through a mail survey. Respondents had the option of completing an online version of the survey although few such responses were received. Three contacts with firms were attempted: an initial letter that explained the purpose of the survey; a reminder card for non-respondents; and a second reminder letter containing a new copy of the questionnaire.

The population for the survey (see Table 2) was the number of firms supplied by Dunn and Bradstreet from their database of Australian firms. Firms with fewer than five employees were excluded from the sample. The specification of the industries followed the Australian and New Zealand Standard Industrial Classification (ANZSIC) codes. These were converted to Standard Industrial Classification codes which in general were more specific than the ANZSIC codes. The engineering industry had substantially more firms than any of the other industries and thus a random sample of 750 engineering firms was selected for inclusion in the study. All firms from the other industries were included.

OPEN IN VIEWER

Table 2.

Number of firms in the population and the sample and response rates.

Industry	Pop’n n	Valid responses n	Response rate %
Clothing and footwear	728	113	19.8
Engineering	750	153	25.2
IT and telecommunications	701	100	17.6
Scientific and medical equipment	666	121	21.2
Processed food and beverages	604	111	22.1
Total	3449	598	21.0

Note: The response rate is based on a mailout adjusted for ‘out of scope’ firms among non-respondents.

Questionnaires were mailed to 3449 firms. The raw response rate was 19.1% of firms that were sent a questionnaire. The response rate is 21.0% if allowance is made for firms that would have been out-of-scope among non-respondents. The analyses of the data presented are weighted to compensate for the differential response rates across industries and the initial sub-sampling of engineering firms. The population to which they correspond is the initial population of 5240 firms in the Dunn and Bradstreet database of Australian firms. Respondents to the survey were also given the option of making further comment about the issues raised by returning their contact details. Follow-up telephone interviews were conducted with 63 respondents (see Table 3) who returned their details. These respondents came from each of the industries included in the survey and included firms with network training arrangements and those without such arrangements. Responses were analysed in two groups: those with and those without network training arrangements. Responses were manually coded to identify obstacles to networking (loss of intellectual property, loss of staff, lack of partners) and the benefits of networking (cost and quality of training, transfer of new knowledge and skill, development of business relationships).

OPEN IN VIEWER

Table 3.

Telephone interviewees by industry.

Industry	No. of respondents
Clothing and footwear	13
Engineering	18
IT and telecommunications	9
Scientific and medical equipment	16
Processed food and beverages	7
Total	63

The extent of network training

The survey results indicate that participation in training networks is a minor feature of the training provision of Australian firms and 82.4% of Australian firms do not engage in any network activity at all. Overall, 68.2% of firms provided employee training but did not participate in training networks and 14.2% of firms provided no employee training whatsoever. Only 17.6% of firms reported any participation in network training arrangements. This participation in networks for training did not mean that all of the firm’s training was delivered in conjunction with other firms, only that some of the training was so delivered. For many firms participating in network training (44.5%), such training formed only a small part of their total training effort. Few firms engaged in network training reported that this was a very significant (8.7%) part of their training provision.

The kind of training delivered through network arrangements was predominantly non-formal and reflects the preferences of Australian employers for training that is non-transferable. Most training was for the introduction of new technology, new products and new work practices (see Table 4). Such training can be considered as firm-specific training in the production processes of the firm and does not lead to the attainment of recognized qualifications. The use of networks for the delivery of formal vocational qualifications was limited. Most of the network delivered formal training was for basic vocational qualifications and entry-level training. Much of this training is mandated in industrial agreements and law. Very little formal training for advanced qualifications was provided through networks.

OPEN IN VIEWER

Table 4.

Type of training provided.

	Firms with no network training	Firms with some network training	% of network supplied training
Non-formal training
Training in new technology	59.7%	73.2%	47.0%
Training for new product introduction	45.0%	60.4%	45.5%
Training in new work techniques	50.4%	54.0%	41.5%
Training for occupational health and safety	55.9%	72.5%	34.2%
Formal training
Training for basic vocational qualifications	36.4%	52.2%	38.1%
Entry-level training	47.6%	59.1%	37.4%
Training for advanced qualifications	25.8%	42.3%	16.2%

One feature of firms engaged in network training, however, was that they provide higher levels of training generally than do firms with no network training arrangements. There is thus some suggestion that the network participants use networks to enhance their training effort, rather than to provide the majority of their training.

As might be expected in a liberal market economy where the provision of training is marked by employer voluntarism, stand-alone or non-network training dominated the approach to training. Some 68.2% of firms provided employee training but did not participate in training networks and had no training arrangements in place with other firms whatsoever. These firms were largely satisfied with this form of provision and had little interest in participation in networks. The current supply-side policies of the Australian government – deregulating provision and providing financial support to firms to purchase training services in the market – suit firms using non-network provision well, as one manager observed:

We have people who sell training to us, as a matter of fact we have one on-site today. They look at our requirements and deliver training to those. The arrangement works well for us. We have someone supply the training and the government pays for it. We are always open to improvement but at the moment we are not actively seeking extra training. (Manager food industry firm)

This satisfaction with arrangements for non-network provision was accompanied by a lack of interest in seeking out network training arrangements or potential partner firms. Some common barriers to collective action by employers in a voluntary system were identified as motivations not to participate in networks. The most significant concerns of firms were concerns about knowledge spillover in networks and the loss of intellectual property. Added to these were concerns about the potential for staff poaching to occur. One respondent observed that ‘When we train we look after our own’ and went on to observe that:

We do not like to train the opposition. We have our own techniques for things. We do not like to share how or why we do things, even the sequence of operations, it is all important to our advantage over the competition because basically we know why we beat the competition. (Manager engineering firm)

The survey found no significant difference among industries in the extent to which those firms that provide training are likely to participate in network training arrangements, so we now turn to an examination of the network arrangements to identify the key characteristics of network training.

The nature of training networks

The survey results indicate that firms participating in training networks participated in closed networks with few linkages beyond the network. The vast majority of firms participating in network training (72.4%) say that they received no external support for their network activities. The picture of network training that emerges is one of largely independent efforts of firms making their own informal arrangements with a small number of partner firms. Most network training arrangements are made through the informal liaison of human resources staff (46.1%) or by one of the partners taking responsibility for the delivery of the training (31.2%). When we examine the network linkages of those firms engaged in network training we find that these firms have dense linkages with a small number of partner firms. Firms engaged in network training do so with only between one (21.9%) and two or three (53.9%) partner firms. Most firms partner with the same firms regularly (54.8%) in reasonably long-standing arrangements of more than three years (48.9%).

These voluntary training networks are largely based upon the extension of business to business relationships. Firms that engage in network training do so to extend business linkages with partner firms (43.8%). The provision of training is often part of a package of business services provided as part of the business relationship and hence training is important for the development of that relationship. Some managers reported actively considering the training element of the package when seeking partners:

We usually source training through suppliers of equipment. We visit machine tool manufacturers. The servicing and setting of equipment is specialized knowledge and so we prefer to deal with the systems suppliers. We see the way that they have integrated components into their system, we learn what to do when the lights go out so we don’t have to call a service technician. (Manager engineering firm)

Whilst other managers spoke of the importance of providing appropriate training when supplying equipment, in order to cement business relationships:

We provide training to our customers and that is at various levels. Some have asked for training similar to that of our own services team but that frankly is often a waste of money. What they want is more user training at various levels if I can put it that way. So it is more a user type of training that is carried out customized to the databases and needs of our customers. (Manager IT firm)

The importance of network training as part of a package of business equipment and services was underlined by the regression analysis of the survey data. Table 5 shows the six characteristics of firms that have statistically significant relationships with the existence of network training arrangements, after controlling for other firm characteristics and given that the firm provides some training. This analysis suggests that in addition to the usual drivers of a firm’s training effort – firm size, emphasis upon employee skills, competitive environment – a cooperative relationship with suppliers of capital equipment is significant. The provision of network training would thus seem to be associated with the use of new technology in particular. Training comes as part of the package of services supplied with technology and is an important dimension of the relationship between vendor and customer. The vendor supplies training to extend the customer relationship and the customer is able to source specific training for specific technologies at lower cost than if it were independently sourced. One manager explained the mix of business considerations and skill development needs that drove the provision of network training:

We do external training for coal miners. We run training courses for them to use our mine safety equipment. If we have provided a system we train for that system so it is on a system by system basis, training for specific systems. … It is an extremely important part of the business. We do six monthly testing and certification but the customer needs to maintain the equipment between visits. The training is aimed at ensuring the customer can maintain the equipment 24/7. (Manager scientific equipment firm)

OPEN IN VIEWER

Table 5.

Observed and adjusted rates for participation in network training arrangements by selected firm characteristics.

Relationships		Participation in cooperative training arrangements
		Distribution	Observed mean	Adjusted mean
		%	%	%
Number of employees	p =		(0.00)	(0.00)
5–19		51.4	13.5	13.4
20–500		48.6	26.2	26.3
Firm made a pre-tax profit	p =		(0.03)	(0.05)
Yes		79.7	22.5	22.3
No		20.3	12.6	13.4
Firm’s relationships with other firms outside the industry are:	p =		(0.08)	(0.02)
Highly cooperative		40.4	24.5	26.3
Arm’s length		59.6	17.9	16.6
Mainly sells capital equipment to other firms	p =		(0.05)	(0.04)
Yes		25.2	26.5	26.8
No		74.8	18.2	18.1
Level of competition for main product	p =		(0.03)	(0.02)
Intense or strong		73.5	22.1	22.3
Moderate, limited or some		26.5	13.7	13.2
Workforce skills a competitive advantage	p =		(0.00)	(0.02)
Yes		5.6	42.8	29.6
No		94.4	18.8	19.6

Notes: Values for the distribution and observed mean exclude missing data. p values for observed means correspond to log-likelihood chi-squares. Adjusted means are based on 513 firms that provided some training and derived from a logistic regression equation with imputation of missing values. The adjusted means are derived from logits by setting the overall mean and distribution and solving logits for the category means. The p-values for the adjusted values correspond to Wald chi-squares from the logistic regression and are adjusted for the finite population correction. All values are weighted.

These results suggest that business to business relationships were the basis for the development of training networks. They further suggest that there was little stimulus to the development of network arrangements from the existence of integrated supply chains, localized learning networks or the activities of network brokers and third party agents. In the case of supply chains, the percentage of firms with network training arrangements that sell their products as an input to other firms’ production processes (20.8%) differs little from the participation in network training arrangements of firms that sell directly to wholesalers, retailers or consumers (19.5%). Only firms that sell their product or service as capital equipment (26.5%) were more likely to engage in network training activities. In the case of localized learning networks or the activities of network brokers and third party agents, we recall that the great majority of firms received no external support for the development of their network training activities.

Network training is then largely undertaken to extend business to business relationships and to address specific training needs within the firm. The need to up-skill employees for the introduction of new technology, new products and new techniques are among the more usual drivers of training within firms and these seem to be the drivers of network participation (Bartel and Sicherman, 1998; OECD, 1999).

Discussion

Voluntary network training based upon business relationships has been thought to have the potential to address some of the problems besetting employer demand for training in liberal market economies. This form of provision has been thought to lead to greater inter-firm cooperation, resource sharing and collective action to provide for the development of transferable employee skills that address the labour market needs of employers. There is limited evidence, however, to support the optimism surrounding the impact of market-based networks. Reliance on market mechanisms leads to a preference for closed networks where training is organized informally among network members. These networks supply limited amounts of training, most of which is non-formal.

Network training is a limited phenomenon among Australian firms, with fewer than one in five Australian firms engaged in any network training. Most firms use network training to supplement their training effort rather than provide the bulk of their training. These are firms that already have a substantial training effort and use their network connections to source specialized training for the development of skills in the application of new technology to improve production processes. Network training is thus largely non-formal training (that is training that does not lead to a formal vocational qualification) leading to the development of technology-specific, process-specific and hence firm-specific skills. The training that is provided through networks is thus consistent with that provided generally by employers in liberal market systems.

This non-formal training is supplied through closed networks associated with the development of business to business relationships. These relationships reduce the uncertainty in the application of new technology by providing for the transfer of expertise along with the purchase of equipment. This transfer of knowledge and skill is important as firms seek to valorize expertise in the application of new technology to gain a competitive advantage. The exploitation of new technology in new production processes and new products is designed to produce a unique and inimitable advantage and one that is not available to non-network members.

Examined from this perspective, the preference for dense inter-firm linkages in closed networks that provide firm-specific skills permits the exclusive use of technology and expertise. By limiting knowledge spillover within and beyond the network and by restricting the recognition of knowledge and skills acquired by employees and hence the ability of employees to transfer knowledge and skills through movement to another employer, firms hope to restrict the dissemination of expertise. By restricting the dissemination of expertise, firms hope to gain exclusive rights to the exploitation of new technology. Our interview data support this inference by underlining the ways in which managers view training as the creation of a firm-specific competition good. The manager who observed that: ‘when we train we look after our own’ was stating a widespread preference for firm-specific investments in knowledge and skill that are designed to develop proprietary expertise.

These results suggest that rather than being a form of social organization that sits between market and hierarchy (Powell, 1990), voluntary networks based upon new business practices are shaped by markets. Voluntary networks emerge from this study as a form of organization that is constrained by market relationships and one that is used within market coordinated economies to mobilize goods not readily obtainable through the market. Goods such as the acquisition of expertise are necessary for the valorization of new production processes and new products in the market but these goods are more readily obtained through social networks. Networks, in this perspective, act to resocialize market-based relationships and address the limitations of arm’s length relationships. By reintroducing social relationships into business relationships, closed networks permit the exchange of goods that cannot be readily defined in contractual arrangements. This resocialization of market relationships, however, occurs within the constraints of exclusivity and this restricts network development.

The voluntary networks found here are closed networks that do not permit the development of thin linkages within or beyond the network. Few linkages are evident between firms and social institutions. Network training is largely developed informally among network participants and there is little involvement of the formal VET system or social institutions in this form of provision. The training delivered through training networks would seem to provide little basis for the further engagement of employers in the Australian VET system to raise employer demand for training. Voluntary participation in networks is not incompatible with employer voluntarism in training but is rather consistent with it.

The voluntary networks in this study provide few substitutes for the beneficial constraints that are introduced by institutional coordination (Streeck, 1989, 1997). The provision of transferable skills in more coordinated economies provides for the labour market needs of employers but does so in ways that do not restrict the ability of employers to develop firm-specific expertise. Transferable training is based upon a body of formal knowledge and skill that is recognized by the social partners. Training in this formal knowledge does not restrict the ability of firms to develop firm-specific expertise in tacit knowledge and skill but rather provides a common stock of knowledge and skill upon which expertise can be built. This enabling view of institutional coordination emphasizes the adaptability of institutions as a counterpoint to the flexibility of the market in providing for the transfer of knowledge and skill (Casper, 2000; Sydow and Staber, 2002).

Social institutions thus provide more than a corrective for market failure in training; they enable the transfer of knowledge and skill by providing a common stock of knowledge and skill upon which expertise can be built. This knowledge ‘commons’ is what is lacking in market coordinated economies with their focus on exclusive rights. The development of closed network relationships provides no substitute for a commons but rather provides a normative basis for the exclusive use of expertise.

Few new business practices that are said to enhance coordinated action by firms are strongly linked to network training. Whilst there is some evidence that firms in integrated supply chains do engage in network training, this provision is not significantly different from the network training undertaken by firms that do not participate in supply chains. The use of new business practices would thus seem to sit comfortably with employer voluntarism in training as there is little evidence of new business practices leading to an increased demand for training. Network training thus seems to mirror the existing features of more voluntarist systems where there is a greater focus upon firm-specific training as a way of securing employer investment in training and avoiding the loss of that investment through knowledge spillover, employee turnover or poaching.

Given the limited scope of voluntary networking evident here, any possibility for expanding collective action by employers through networks would seem to turn on the use of government backed, supply-side measures to encourage the formation of third party service networks. These networks generate thin linkages that do not challenge exclusivity in the application of expertise and so pose fewer obstacles to participation in the network by employers. Government incentives for the provision of network training and government support for network building could increase knowledge and awareness amongst employers about the availability of network training and could promote the benefits of such training. An extension of government supply-side policies to encompass the funding of agents to develop inter-firm networks may be successful in matchmaking firms that might otherwise not engage in network training. Such networks could mobilize shared local resources for the provision of formal training leading to the attainment of formal vocational qualifications but they are unlikely to lead to the development of dense linkages between firms that would stimulate further cooperation.

The development of state-funded service networks would seem to be inherently unstable and prone to employer opportunism, however. In the absence of other dimensions to business relationships, collective action comes to be dependent upon incentives that enable employers to capture the benefits of training. Employers may thus be prone to moving in and out of service network arrangements as the perceived benefits of that training change. Australian employers are, by and large, satisfied with the results of their current training arrangements and there seems to be limited motivation to participate in network training activities.

Conclusion

Market-based relationships shape voluntary training networks in particular ways. They lead to the development of norms of exclusivity in the transfer of knowledge and skills through the restricted dissemination of expertise. Such norms are consistent with the development of closed networks where dense linkages are established with few partner firms. These networks provide little scope for the engagement of employers with the vocational training system and offer few incentives to raise the employer demand for training. Voluntary networks provide few substitutes for the beneficial constraints that are introduced by institutional coordination as they do not lead to the creation of a common stock of knowledge and skills. Voluntary network training thus provides little basis for collective action by employers to address labour market needs through the provision of transferable training leading to the attainment of vocational qualifications.

Few new business practices thought to develop network training are strongly associated with such training. The extension of business to business relationships is the key practice and here network training assists in the dissemination of tacit knowledge and skill associated with the exploitation of new technology. The extension of business to business relationships acts to resocialize the market transactions between business partners and provide for goods that cannot be readily specified in contracts and are not readily available in traditional, arm’s length market relationships. These relationships do not provide the foundation for the development of forms of collective action by employers. Australian employers are largely satisfied with their existing stand-alone training and the spread of new business practices seems to have had a limited impact on employers’ approach to the provision of training at work, despite the increased attention given to these developments by governments and employer associations.