Open innovation and its relevance to industrial research and development: The case of high-technology small firms

Some general principles

At a general level, the principal of open innovation is of interest to academics engaged in the study of HTSFs. This interest has a number of dimensions. First, successful new product development, the most rewarding form of industrial innovation, is the activity that HTSFs are principally engaged in, mainly through in-house R&D, and holds the promise that greater openness might lead to better HTSF innovative performance should be considered by these firms. Second, although mainly dependent on internal R&D, HTSFs do have significant contacts with external sources of expertise (regarding for example, marketing, subcontracting, university consultancy, capital funding). In principle, these external HTSF relationships and their interaction with otherwise ‘closed’ HTSF innovation systems might be enhanced by delivering the benefits of greater openness to internal R&D activities. Third, because the technology developed by HTSFs is often their only asset of any significant potential financial value, the issues of openness is both a source of great potential benefit (through collaboration with other holders of relevant expertise noted above), and conversely a threat, in that greater openness might jeopardise the confidentiality – and by implication the value of their R&D assets. For example, if assets developed internally by an HTSF are appropriated by a collaborating firm during the sharing of technical information in joint product development programmes, this eventuality would prove difficult to take legal action against, given the relatively meagre financial resources of most HTSFs. Thus, the cost–benefit of greater openness must be considered.

The combined impact of the above three key HTSF characteristics that have relevance to open innovation will be used to construct a general argument that HTSFs are (and always have been) ‘open’ to a greater extent than perhaps Chesbrough would argue. However, in instances where such firms are closed in their innovative behaviour – for example, in performing R&D – it will be argued that this is because they have good reason to be so, mainly in order to maximise the quality of their internal R&D effort and to protect valuable internally generated intellectual property. The purpose of this article is to make a case that in terms of HTSFs at least, the open innovation approach is either irrelevant because it already happens in a measured way, or that when it is avoided it is justified for the reasons detailed below.

As discussed previously, many of the arguments presented by Chesbrough (2003a; 2011) on open innovation relate to large industrial and commercial enterprises. This focus is probably because openness is better handled by large firms, since they have the financial resources to range widely over the world’s national economies in search of collaborative partners and external technology that might complement their internal R&D efforts. Their access to greater financial resources means that such firms can survive instances where open innovation collaboration with another organisation fails, and the costs of such an engagement need to be written off. Moreover, the financial strength of large firms allows them to cope better with any legal disputes that might arise over the ownership of jointly developed intellectual property with another firm.

The situation for HTSFs, in instances when they seek to collaborate with other firms, is very different. Even when a strong case for grievance exists, such as when a collaborative relationship breaks down, HTSFs often cannot risk the possibility that expensive legal action undertaken over a protracted period might fail to achieve a satisfactory outcome (Faems, 2012; Macdonald and Lefang, 1998). (This problem will be explored further in the subsection on capital investment in HTSFs below.) Consequently, the acquisition of HTSFs by a larger counterpart, either from the same or an adjacent sector, may be acceptable on the basis that the HTSF technology involved is not lost when it is acquired by a large firm since the acquirer would have better resources with which to exploit such a new advance.

However, while a degree of HTSF acquisition is inevitable, a healthy surviving HTSF sector in any developed economy is important for at least two major reasons. First, it is now well-established that research in HTSFs can be highly efficient compared to large high-technology firms (Cooper, 1970; Oakey, 1984; Rothwell and Zegveld, 1982). The HTSF’s close-knit team of researchers, often led by a founding technical entrepreneur (or entrepreneurs), experience minimal organisational or physical distance between HTSF management and the objectives of R&D, mainly because in many cases the same individuals are performing both functions (Roberts, 1991). Thus, although HTSFs spend less than large firms in absolute terms, the cost–benefit of what they produce is often greater than large firms due to their high level of organisational and technical focus.

The arrangement of R&D in HTSFs is very different from the organisational structure of large firms where frequently the objectives of senior management, and those of an often physically separate R&D department, may diverge. Indeed, many large firms have acknowledged the benefits of R&D in HTSFs by setting up small subsidiary firms that are located on sites physically separate from the main company facilities, or by taking equity stakes in new independent HTSFs with a view to acquiring them in the future if the products under development are successful (e.g. large pharmaceuticals firms’ acquisition activities in the biotechnology industry; Faems, 2012; Oakey, 1993; Oakey et al., 1990). The primary role of R&D in HTSFs is to develop new product technologies, often at the leading edge of scientific knowledge. HTSF technical entrepreneurs typically emanate either from universities (i.e. through academic entrepreneurship), or from existing large high-technology firms (e.g. the profuse number of ‘Fairchildren’ firms that have ‘spun off’ from the Fairchild Corporation in Silicon Valley; Cardullo, 1999; Mason, 1979; Oakey, 2012).

Second, the new HTSF entrepreneur is often a technical entrepreneur who has combined business and technical skills, yet no vested interest in the current technological status quo often maintained by one or more large firms. This HTSF independence enables an individual technical entrepreneur (or a small group of entrepreneurs) to think ‘outside the conventional technology box’ and initiate radical change, often at the expense of large incumbent firms seeking to maintain a technological status quo by continuing outdated product technologies at the expense of the consumer (Christensen, 1997; Oakey, 2012). Due to this freshness of approach, in an important minority of cases, HTSF founders have performed the key function of keeping the world economic system honest by developing new products that large firms have been unable or unwilling to develop for the customer (due to monopolistic or oligopolistic market control). For example, the willingness of Apple, together with other new computer manufacturers, to develop and produce cheap desktop computers in the mid-1970s for personal use, which thwarted IBM’s reluctance to enter this market due to their dominance of mainframe computer production, is probably the best known example of this phenomenon.

In terms of open innovation, the scope for external interaction regarding R&D aimed at product development in HTSFs is restricted by two key features of HTSF R&D. First, many HTSFs are in the process of developing leading edge products based on state of the art technology, thus it is often difficult to find external collaborators who could assist in this process. Second, the internal development of new products with high sales potential and possible disruptive power must remain highly confidential in order to maximise the financial returns from such new developments and to reward external investors, should they be successful. However, this does not mean that other support services cannot be openly purchased from external sources (e.g. marketing advice component parts, subcontracting out production) – a process of external involvement that has always been commonplace among HTSFs, especially when located in or near a high-technology cluster.

Chesbrough’s approach to open innovation can be criticised in that in some instances, confidentially-based ‘closed’ innovation is strategically essential and is not a moribund approach to R&D, while in other instances, advocating that seeking external technical assistance from support services should be practised when appropriate by industrial firms is merely stating the obvious, and is a practice that so-called ‘closed innovation’ firms have used since the beginning of the industrial revolution. Chesbrough’s tendency to assume that obvious commercial practices are not occurring can be illustrated by a quote from one of his most recent works, in which he argues: ‘Instead of treating customers as passive consumers, many companies are now involving customers in the innovation process’ (2011: 21).

Here, Chesbrough makes the mistake of implying that industrial firms do not practise openness towards their customers. Empirical investigations of how HTSFs operate have established that feedback from customers on ideas for new products and improvements to the performance of existing products in operation, are acknowledged by most HTSF managements as the key benefits of employing sales staff to make customer visits in addition to actual selling activities (Oakey, 1984; 1991; 1995; 2012). As will be discussed below, any industrial firms that did not talk to their customers would not be in business for long.

High-technology small firms and collaborative research and development

Nonetheless, it might be postulated that although internal R&D and the products that ensue from HTSFs are rarely amenable to open innovation, collaborative R&D on product development might be a way of stimulating a limited form of open innovation between HTSFs through inter-firm R&D projects (Chesbrough et al., 2006; Inauen and Shenker-Wicki, 2011). There are certainly a number of substantial advantages that might ensue from jointly conducted R&D, including sharing the R&D risks of failure and the R&D costs of producing a winning product. However, research into the propensity for HTSFs to enter into collaborative R&D agreements suggests that such collaborations are rare (Oakey, 1984, 1995; Klein-Wollhuis, 1999). There are three main reasons for this lack of collaboration. First, the idea of collaborating with another firm outside the close-knit R&D environment of a single HTSF contradicts the basic principle that makes HTSF R&D effective (i.e. negligible organisational and physical distance). Second, although collaboration might be attractive from cost- and risk-sharing viewpoints, the problems associated with agreeing on the relative inputs to the R&D process, the technological direction that such joint research should take and, perhaps most importantly, dividing the benefits of R&D outputs when a new product has been developed, can cause disputes over who pays for what, who does what, and who gets what. Third, the need to maintain confidentiality during joint product development over sometimes substantial physical distance, and among a larger number of workers from different companies, often deters HTSF management from entering into collaborative R&D arrangements (Klein Woolhuis, 1999). This tendency to avoid collaboration explains why the concept of open innovation is not readily appropriate in most HTSFs for good strategic reasons.

Incubators and science parks

With regard to open innovation, incubators and science parks are considered together, since in the majority of physical contexts, frequently they are both found on or near the campuses of universities (Westhead et al., 2000). The general concept that guides these facilities is that nascent firms begin life in an incubator before any production is appropriate, and often before the formal founding of a business, in order to work on a business idea. Relevant on-site local services and management support are made available to incubated new enterprises, and it is expected that the informal environment of the incubator, with ample opportunities for new entrepreneurs to network with each other, will aid progress through an open approach to innovation. However, the actual experiences of incubated entrepreneurs, and the findings of academics who have studied them, suggest that sharing information with external parties in general among prospective entrepreneurs, and technical information relating to their new business idea in particular, is rarely forthcoming (Albert et al., 2002; Oakey, 2007b; Pittaway and Robertson, 2004; Stockport and Kakabadse, 1994). Indeed, regardless of whether the new business idea is of a high-technology nature, or a new slant on a ‘me too’ type of generic business, nascent entrepreneurs who have not yet launched their new firms are extremely concerned to preserve confidentiality. Clearly, the scenario they wish to avoid is that should they divulge their idea to an acquaintance within the incubator, it will be stolen before it could be exploited or protected. For prospective entrepreneurs, whether the business idea involves manufacturing or providing a service, much of the latent potential of an innovation is the novelty and/or surprise that might be lost through an open innovation approach. For this reason, the key concept of incubators is somewhat flawed due to its rather naive attitude regarding the extent to which the business environment within the incubator is open.

Moreover, as a similar belief in how industrial innovation occurs was a major driver for university science parks in the 1980s, similar problems have occurred when new firms either graduate to science parks from local incubators or are attracted to a science park from the local industrial environment. Here too, concern regarding confidentiality and the close-knit teamwork common inside HTSFs has tended to reduce the level of consultancy with university academics, often cited by science park literature in the 1980s as a major reason for science park location (e.g. Trinity College, 1983). In addition, it has been empirically noted that since new product developments of HTSFs are often leading edge in a specialised field of science, the idea that any university academic would be willing or easily able to provide detailed technical help over short time span that would be of use to HTSF R&D teams is simplistic in the extreme, and has been observed to be rare in Silicon Valley (Oakey, 1985a, 1995). Neither is this scepticism merely an isolated phenomenon, since it has been consistently supported by other research, which found that R&D collaboration between high-technology science park firms and the physical science departments of adjacent European universities generally was poor (Lindholm-Dahlstrand and Klofsten, 2002; Massey et al., 1992; Westhead and Cowling, 1995; Westhead et al., 2000).

A trend must be added to the above problems of continuing limited openness, both in the UK and USA, for universities increasingly to view their own R&D outputs from academic departments as intellectual property to be confidentially exploited in order to raise capital for the university. Both the Bayh-Dole Act of 1980 in the USA, and a series of government initiatives in the UK (DBIS, 1998; DIUS, 2008), have attempted to replace the public funding of universities through the commercial exploitation of the intellectual property that they produce. This has had a diminishing impact on the openness of publicly available scientific knowledge, particularly in the area of biomedicine (e.g. gene technology), in circumstances where the traditional principle that ‘one does not patent medical discoveries’ has been replaced by a ‘patent everything’ approach. This attitude has been especially contentious in the field of genetics where some scientists, who have added the last stone to a pyramid of knowledge generated by others, have sought to patent the whole pyramid. Perversely, this is a trend that has intensified at a time when the concept of open innovation has been gaining strength.

Therefore, the limited success of both incubators and clusters can be largely attributed to a general unwillingness on the part of academics and business founders to share openly their embryonic ideas or the fruits of expensive R&D. In these circumstances, any open innovation-based idea that the generators of modern high-technology are becoming more relaxed about sharing innovations in general, and winning product innovations in particular, is questionable.

Clusters

The concept of clustering or agglomerations is not new, and dates back to the early work of Alfred Weber (1929) on local labour advantages. It well understood that a degree of openness is a key part of the comparative advantages that clusters offer over non-clustered locations through various forms of local business-to-business relationships maintained during production in industrial districts (Hall, 1963; Marshall, 1920; Martin, 1966; Wise, 1949). Work by geographers on behavioural linkages (Taylor, 1971; Wood, 1969) and tacit knowledge by economists (e.g. Williamson, 1975) have long acknowledged the importance of informal information flows in all kinds of valuable business transactions that can be passed advertently or inadvertently between customers and suppliers inside agglomerations.

There are several ways in which the identification of high-technology clusters may have appeared to promote the concept of open innovation at the end of the last century (Oakey, 1985b). Experiences from Silicon Valley, perhaps the most famous high-technology cluster, often have been interpreted by academics as evidence that technical collaboration in the form of localised open innovation was a major cause of agglomeration advantage in this cluster (Cardullo, 1999; Saxenian, 1985). However, while various advantages do exist in this type of cluster (e.g. local venture capital, skilled labour, customers and specialist subcontracted suppliers), and where low-level information on the development of new product technologies is informally shared by locally socialising entrepreneurs, fundamentally Silicon Valley is a highly competitive business environment in which important information on new product specifications remain confidential and is highly prized (Oakey, 2007b).

Indeed, although skilled labour is a local advantage for HTSFs, it has a hard edge represented both by labour poaching and the loss of workers through spin-off, during which new product ideas often are stolen from previous large or small-firm employers by new spin-off technical entrepreneurs (Oakey, 1995; Roberts, 1991) – this is benignly termed ‘spillover’ by some academics. However, once spun-off (or overspilt), there is substantial evidence that there is rarely technical contact between spillover firms and their previous employer. Thus, proximity is better explained by ‘acorns not falling far from the tree’ rather than any continuing functional innovation collaborations with a spin-off firm’s previous employer (Oakey, 1984; Oakey et al., 1988, 2001). Similarly, while venture capital is available to HTSF entrepreneurs in Silicon Valley, investment in these firms is made under conditions where deals are struck on a hard-nosed formal basis, often involving the ejection of the founding entrepreneur. Finally, while a rich choice of supplier and subcontractors may be available, customers will seek the best service at the most competitive prices from such interactions (Oakey, 1985b).

Perhaps the best documented example of how important external advice and customer feedback can be obtained by HTSFs in a cluster is the example of Apple Computers. In the early 1970s, the marketing consultants Mike Markkula and Regis McKenna helped Apple refine the design and function of their new laptop computer by trialling this new product with potential customers. This is what Regis McKenna has termed a ‘market-driven approach’, rather than a ‘marketing-driven exercise’ in circumstances where applications needed to be found for their new and substantially unknown product (e.g. in education, the office and for personal use). From the very early days of Apple, Steve Wozniak and Steve Jobs recognised the key importance of good design and customer approval in achieving success. However, this instance was only a particularly effective example of what always had been the case: namely, that talking to customers and gaining free information on product performance has been a consistent and crucial part of any sales representative’s work (Oakey, 1991). These observations make the point that although actual R&D in HTSFs might be relatively closed, key support services are open and never have been restricted.

Indeed, the useful interactions discussed here describe what are no more or less than normal business behaviour patterns; regarding the extent to which they are examples of open innovation, they are not new and were apparent in the industrial districts observed by Marshall (1920) – which raises the previous point made as to whether the term ‘open innovation’ offers anything new to our understanding of industrial relationships. (This is a theme that will be returned to in the conclusion of this article.)

When attempting to explain how a cluster such as Silicon Valley operates, a key issue is the apparent conflict between collaboration and competition in a local area, where clearly they are both present. One general way of resolving this dilemma is to observe that firms in an agglomeration of the Silicon Valley type tend to collaborate vertically with local customer or supplier firms inside supply chains of operation. For example, a printed circuit board maker might collaborate with a computer manufacturer customer (often exchanging technical information), yet strongly compete horizontally with another local printed circuit board maker in their own technical area of specialisation, thus contributing to collaborative advantage and competition. A real-life example of this complexity is cited by Berlin (2010), who observes that Bob Noyce from Intel advised Steve Jobs when he was setting up Apple Computers with Steve Wosniak, and that in turn, Steve Jobs advised Larry Page and Sergey Grin when they were establishing Google. Here, Noyce and Jobs were prepared to advise these colleagues because they were not in direct competition. Indeed, Jobs was likely to purchase Noyce’s semiconductors when building his Apple computer, while Page and Grin would enhance the utility of Job’s computers through the founding of Google. None of these assisting individuals would have been likely to help a directly horizontal competitor (Oakey, 2012).

What this example means in terms of open innovation is that while an exchange of information can occur between entrepreneurs within an industrial environment, in high-technology industries in general, and HTSFs in particular, such interactions are limited both in terms of sectoral scope and in the quality of information that passes between individuals interacting in such an open manner. In particular, information exchanges tend to take the form of general advice on technical information, while key technical information on new product development is closely guarded and not exchanged.