هنگامی که سیستم نوآوری ملی و انواع سرمایه داری با استدلالات شایستگی های کارفرما مواجهه می شود در انواع مهارت و دانش علمی مورد نیاز برای نوآوری و افزایش محصول

The literatures on ‘varieties of capitalism’ (VoC) and ‘national innovation systems’ (NIS) propose very similar arguments about how firms require different types of labour qualifications to pursue strategies of radical product innovation (RPI), incremental product innovation (IPI), and product imitation (PI) respectively. Despite their similar lines of reasoning, however, the VoC scholars are concerned with the skill profiles of a firm's entire workforce, whereas the NIS proponents focus on the knowledge base of scientists. Given that both literatures have developed without explicitly taking the arguments of the neighbouring discipline into account, it is thus unclear whether they explain the same, or different, phenomena. Furthermore, both literatures propose firm level arguments but test them on the basis of macro- rather than micro-level indicators. This paper therefore asks: first, does micro-level evidence support the VoC and NIS arguments that particular types of employee skills and knowledge backgrounds of scientists are needed for different competitive strategies? And, if so, do RPI, IPI, and PI firms need to employ scientists in combination with a workforce having the respective qualifications, or is it sufficient if scientists or employees alone are adequately qualified. Quantitative analyses indicate that a particular mix of scientific knowledge combined with employee skills facilitate RPI, IPI, and PI strategies. The article thus concludes that – despite their similar reasoning – the VoC and the NIS literatures indeed describe different phenomena, without being aware of the synergies created whenever adequate employee and scientific qualifications are hired together.

Agreement is broad amongst contributors to the competitiveness literature that firms require people with distinct qualifications in order to pursue different product-market strategies. While employees with ‘general’ or ‘multi-tasking’ skills are said to be needed for radical product innovation, workers with ‘firm-specific’ or ‘occupational specialization’ skills presumably facilitate incremental product innovation. Low qualified and, hence, inexpensive labour is claimed to be required for low cost production based on product imitation.3 Despite this general agreement, different strands of the competitiveness literature focus on diverse holders of qualifications. While the literature on ‘varieties of capitalism’ (VoC) proposes arguments about the qualifications of the overall labour force of a company,4 the literature on ‘national innovation systems’ (NIS) tends to focus on the knowledge base of a firm's scientists.5 More concretely, the VoC literature argues that radical product innovation (RPI) requires employees with general skills because they can adapt more easily to constantly changing supplier–producer relationships and market demands which, in turn, are characteristic of this product-market strategy. Specific skills are said to be necessary for incremental product innovation (IPI) because the in-depth knowledge of a company, of its market, its suppliers and customers enables employees to continuously improve products and production processes, and to adopt products to specific customer needs. Furthermore, employees with an in-depth understanding of how their firm operates are able to work autonomously and to take on responsibility. They know, for example, how to rectify mistakes that occur during the production process, which, in turn, contributes to maintaining a high level of product quality. Finally, product imitation (PI) is said to rely on employees with neither general nor specific but with low skills as their salary levels are reduced. Even though low-skilled employees cannot often rectify mistakes that occur during the production process without precise instructions from their superiors, this does not harm the pursuit of PI strategies, as product quality is less important than product costs.6 The NIS literature, on the other hand, illustrates how the employment of scientists with diverse knowledge backgrounds crucially enables to pursue RPI, IPI, and PI strategies. Scientists with heterogeneous knowledge are said to facilitate RPI as ‘it might take an enormous intellectual effort or an extremely creative mind, to identify a potential new combination’ (Lundvall, 1992b: 8; see also Johnson, 1992: 29). Scientists who have worked with colleagues from diverse universities, countries, and disciplines – while being rather autonomous from their supervising professor – are more likely to have the necessary, radically innovative potential due to their increased imaginative capacities. Scientists with a homogeneous knowledge base, on the contrary, are found to enable the pursuit of IPI strategies. Since they have worked within the same field of research and the same team for a long time, scientists with homogeneous knowledge have an in-depth understanding of the technological opportunities in this area and are used to cooperating, and to combining their insights, in order to develop incremental innovations. At the same time, they might be so familiar with one environment that they have difficulties to imagine entirely new realities and, thus, lack the creative capacities to come up with radically new ideas. Finally, PI firms do neither require scientists with a heterogeneous nor a homogeneous knowledge base as they imitate the inventions of their competitors. PI strategists thus benefit from not hiring scientists.7 Two features of these literatures are particularly noteworthy. First, both literatures do not test their arguments on the basis of micro-, that is, firm-level indicators. Instead, the NIS and VoC literatures start from the observation that the innovative performance and product market strategies of firms vary between countries and seem to be supported by national institutions, including research as well as education and training (E&T) systems. Based on data aggregated at the industry level, both literatures conclude that these institutional differences cause firms to embark on diverse innovation or product market strategies as they facilitate the availability of different factor types, including scientific knowledge and employee skills. With some very few exceptions,8 micro-level assessments of scientific knowledge and skill profiles are not provided.9 Second, even though they both propose similar lines of reasoning, it is unclear whether the NIS and the VoC arguments refer to the same or different phenomena, because the two literatures developed in parallel without explicitly taking the arguments of the neighbouring discipline into account. While the VoC scholars consider the education and training which employees receive,10 the NIS proponents are rather concerned with the career paths of scientists.11 Ultimately, though, the reasoning of both literatures rests on the insight that the increased exposure of people to new ideas – be it in the form of employees changing firms more regularly, be it in the form of scientists being more autonomous and performance oriented in their choice of research projects – is crucial for the emergence of radical innovations. But, do firms need to hire scientists with a particular knowledge profile in addition to a workforce with distinct qualifications in order to pursue RPI, IPI, and PI strategies respectively? Or is it sufficient if scientists alone have a particular knowledge base, given that they constitute that employment group with the key capacities for innovation? Or are scientists merely one group of the firm's entire workforces and, hence, require particular skill profiles rather than knowledge backgrounds? Consequently, this article has two aims. First, it analyses whether micro-level data confirms the NIS and VoC arguments on the importance of different qualification types for RPI, IPI, and PI strategies. Second, the article explores whether the VoC and the NIS literatures explain similar or different phenomena. To these ends, the article studies pharmaceutical firms – including biotech, traditional pharmaceutical, and generics firms – in Germany, Italy, and the UK. Pharmaceutical firms are particularly revealing cases to study as the scientifically established notion of a ‘new chemical entity’ allows the distinction between RPI, IPI, and PI strategies at the firm level. Furthermore, firms in different countries need to be studied so as to reveal whether possible differences in the labour qualifications employed by RPI, IPI, and PI firms result from the competitive strategies of these firms, or from the sheer availability of diverse qualifications due to the country's research and E&T systems. If the employee skills and scientific knowledge employed by RPI, IPI, and PI firms differ between these competitive strategies rather than between countries, we can conclude that firms cannot randomly hire people, but that RPI, IPI, and PI strategies require workforces with distinct qualification profiles. Germany, Italy, and the UK offer most comprehensive insights as these countries are said to have particularly characteristic E&T and research systems providing people with the required qualifications for RPI, IPI, and PI strategies. More precisely, the E&T and research systems of the UK are held to teach employees and scientists mostly qualifications which are required for RPI strategies, whereas Germany's E&T and research systems are found to provide people with the necessary qualifications for IPI strategies. The poorly developed E&T and research systems of Italy, in turn, are said to leave people with neither general nor specific and, hence, low skills, thereby facilitating the pursuit of PI strategies.12 To understand whether micro-level data confirms the VoC and NIS reasoning and to illustrate whether these literatures propose similar or different arguments, the article proceeds as follows. Section 2 uses micro-level indicators in order to identify firms that pursue RPI, IPI, and PI strategies in the UK, Germany, and Italy. Section 3 conceptualizes, operationalizes, and measures the skill types of employees who work for the RPI, IPI, and PI firms identified in Section 2. Analyzing interviews with the firms’ Human Resources managers, Section 3 furthermore presents the results of multinomial logistic regressions which show that RPI, IPI, and PI strategists indeed rely on employees with general, specific, and low skills respectively. Similarly, Section 4 conceptualizes, operationalizes, and measures the knowledge base of scientists working for these firms and illustrates with the help of binary logistic regressions that RPI and IPI strategies employ scientists with heterogeneous and homogenous knowledge respectively, whereas PI firms do typically not require scientific knowledge. Section 5 finally sheds light on the question whether firms need particular employee skills in addition to distinct scientific knowledge. Interestingly, binary logistic regressions not only show this to be the case, but also suggest that synergy effects are created whenever adequate employee and scientific qualifications are hired together. Section 6 therefore concludes that, despite their similar reasoning, the VoC and NIS arguments describe different empirical phenomena without realizing the competitiveness-enhancing potential of their combined reasoning.

This article has studied the compatibility of the VoC and NIS arguments about the labour qualifications that firms need to hire in order to pursue strategies of radical product innovation (RPI), incremental product innovation (IPI), and product imitation (PI) respectively. Given that both literature strands base their arguments on similar lines of reasoning, while studying different groups of employees, it is unclear whether firms need to hire only scientists with particular knowledge backgrounds, only employees with distinct skill profiles, or both scientists and employees with the respective qualifications. It is furthermore striking that both literatures base their arguments on macro-level analyses and, with some very few exceptions,40 have not tested their hypotheses on the basis of micro-level indicators.41 The aims of this article were thus twofold: first, to assess whether micro-level data collected at the firm level supports the arguments of the VoC and NIS literatures; and, second, to understand whether RPI, IPI, and PI firms need scientists alone, an entire workforce, or both scientists and a workforce with distinct qualifications. The results obtained from interviews with Human Resources managers in the UK, Germany, and Italy were straight-forward. Irrespective of the country in which firms are based, the latter require employees who have general skills for pursuing RPI strategies, employees with specific skills for IPI, and employees who do not hold particular skill profiles for PI strategies. Furthermore, RPI strategies are facilitated by scientists with a heterogeneous knowledge background, whereas IPI strategies benefit from scientists with homogeneous knowledge, while PI firms are best off if they do not at all hire scientists pursuing R&D activities. Micro-level data thus confirms the arguments of the VoC scholars on the one hand and the NIS literature on the other. But what about the joint applicability of these theories? Do VoC scholars ignore that scientists with adequate knowledge backgrounds are sufficient for pursuing RPI and IPI strategies; are they mislead in their reasoning because variations in the workforces’ qualifications ultimately result from variations in the firms’ scientific knowledge employed? Or, do NIS scholars ignore that all employees of a firm need to have undergone particular education and training in order to acquire the necessary skills, so that the knowledge backgrounds of scientists merely represent the skill profiles of the entire workforce? Or, are the VoC and NIS arguments compatible in that firms need both employees with distinct skill profiles in addition to scientists with particular knowledge backgrounds? Again, the interviews with HR managers provided insightful answers. RPI, IPI, and PI strategists benefit from a combination of scientific knowledge and skill profiles: RPI strategies are facilitated by employees with general qualifications and scientists with heterogeneous knowledge backgrounds, whereas IPI rests upon a workforce with specific skills and scientists with homogeneous knowledge. PI firms, in turn, are best off if they neither hire scientists nor invest in educating and training their workforces. Interestingly, the combination of these qualification types seems to have a multiplicative rather than an additive impact. This means that firms which combine the respective employee and scientific qualifications are exponentially better in being radically or incrementally innovative than they would be if they hired these qualification types in isolation. These findings have several noteworthy implications. First, they illustrate that the innovative potential of people is not concentrated in the scientists employed by a firm. New ideas can be generated by all employees, at all stages of the value chain. To stay innovative, firms would thus be misled to listen exclusively to the ideas proposed by their scientists. Second, and as a corollary of the first implication, scientists play a key role of innovation. This seems, however, less the case because scientists are active in R&D and, hence, in the knowledge-generating stages of the value chain. It rather results from the finding that scientists can be holders of two different types of qualifications. Depending on their previous work experience and the composition of the team within which they work, scientists can have homogeneous or heterogeneous knowledge backgrounds. However, being part of the firm's entire workforce, scientists can also receive training in general (i.e. industry-related) or in specific (rather firm-related) topics. Furthermore, firms can encourage scientists to stay and work for them for a long time or, rather, to move on to another employer more rapidly. Scientists do therefore not only hold peculiar types of knowledge but also have particular types of skills. Interestingly, these types of qualifications seem to be inherently different capacities held by the same person. In other words, the innovative capacities of scientists seem to stem from the exchange of ideas with their colleagues on the one hand, and from an in-depth knowledge of their firm, its organization, suppliers, customers, and production processes on the other. Third, and as a corollary of the second implication, interactions of adequately skilled employees with knowledgeable scientists seem to be yet another and particularly important source of innovation. Whenever scientists learn not only from their colleagues but also from employees throughout the firm, and whenever employees at lower value-chain stages learn from scientists, these interactions seemingly facilitate a cross-fertilization of ideas which translates into the superior capacity of a firm to be radically or incrementally innovative. As this is the case with virtually all research, the present findings and their interpretations should be taken with a grain of salt. Most importantly, these findings are based on the analyses of only one sector: pharmaceuticals. While the pharmaceutical sector is frequently studied by NIS researchers in general, and VoC scholars (most notably Steven Casper) in particular, the arguments of these literatures are proposed for the entire economy. This article, however, does not consider empirical evidence beyond the pharmaceutical sector, which implies that the present findings might not be generalizable to other industries. But given that empirical evidence lent support to the VoC and NIS arguments individually in Sections 3 and 4, there seems to be no reason to suspect that the findings on the compatibility of VoC and NIS arguments in Section 5 could not be generalized to other industries. It can thus be concluded that the NIS and VoC literatures describe different phenomena when they illustrate how a country's research system shapes the knowledge backgrounds of scientists on the one hand, and how the economy's education and training systems provide employees with different types of skills on the other. However, these findings also suggest that the NIS and the VoC literature ignore the synergy effects resulting from the complementarities of the research and E&T systems. When pursuing RPI, IPI, and PI strategies, firms should thus be aware of the compatibility and complementarities of the NIS and VoC arguments and seek both to hire scientists with adequate knowledge backgrounds and to train their workforce in the required skills as this dramatically increases their innovative potentia.