Master Thesis
“The impact of national culture on SMEs’ participation in
Industry 4.0: A cross-country analysis”
under the Supervision of
Prof. Eelke de Jong
Name: Steffen Buchholz
Student number: s1028385
Specialization: Master in Economics – International Business
Word Count: 16177 words
2
Table of contents
1. Introduction ... 5
2. Theoretical background ... 7
2.1. The concept of Industry 4.0 ... 7
2.2. Implementations of Industry 4.0 ... 9
2.3. Small and Medium-sized Enterprises in Industry 4.0 ... 12
2.4. Challenges of Industry 4.0 for SMEs ... 13
2.5. National differences in SMEs’ attitudes towards Industry 4.0 ... 14
2.6. The consideration of national culture ... 16
2.7. The impact of national culture on SMEs ... 19
3. Research Objective and Research Hypotheses ... 21
4. Data and Research Model ... 24
3.1. Dependent Variable ... 25
3.2. Main Variables for Cultural Imprints ... 26
3.3. Control Variables for the Economic Environment ... 31
5. Research Sample and Methodology ... 33
5.1. Research sample ... 33
5.2. Research Methodology ... 34
6. Main Results and Discussion ... 36
6.1. Direct effect of national culture ... 39
6.2. Effect of control variables ... 41
7. Moderating effect of national culture ... 42
8. Cultural impact on advanced and developing countries ... 44
9. Robustness check ... 52
10. Concluding remarks ... 54
10.1. Practical implications ... 56
10.2. Limitations of the study and further research ... 57
11. Conclusion ... 58
References ... 59
Appendix 2 ... 72
3
List of Tables and Figures Tables
Table 1: Distribution of TC ... 26
Table 2: Correlation matrix of dependent and cultural variables ... 27
Table 3: Distribution of PS ... 32
Table 4: Correlation matrix of all variables ... 33
Table 5: Summary statistics of the entire sample ... 34
Table 6: Main estimation of national culture’s effect on SMEs’ participation in Industry 4.0 ... 37
Table 7: Estimation of interaction effects between cultural and control variables ... 43
Table 8: Summary statistics of advanced countries ... 47
Table 9: Summary statistics of developing countries ... 47
Table 10: National culture’s effect on the participation in Industry 4.0 by SMEs in advanced and developing countries ... 48
Table 11: National culture’s effect on the participation in Industry 4.0 by SMEs without countries scoring high/low on TC ... 53
Figures Figure 1: Nine components of Industry 4.0 ... 8
Figure 2: Relationship between SMEs' technical capacities and power distance ... 27
Figure 3: Relationship between SMEs' technical capacities and individualism... 28
Figure 4: Relationship between SMEs' technical capacities and masculinity ... 29
Figure 5: Relationship between SMEs' technical capacities and uncertainty avoidance ... 29
Figure 6: Relationship between SMEs' technical capacities and long-term orientation ………...….….30
4
List of abbreviations
GNI………Gross national income ICT………Information and communications technology IT……….Information technology MSME………...Micro, small and medium-sized enterprise R&D………...Research and development SME………..Small and medium-sized enterprise
5
Master Thesis:” The impact of national culture on SMEs’ participation in
Industry 4.0: A cross-country analysis”
1. Introduction
Digital progress has been established in nearly all parts of our life. Consequently, the digitization of industrial productions is an omnipresent issue of surveys, research, political discussions, and governmental initiatives. “Industry 4.0”, originally introduced by the German government (BMBF, 2020), describes the intelligent networking of machines and processes in industry by means of information and communication technology (Plattform Industrie 4.0, 2020). This terminology established itself as an expression for the global need of a fourth industrial revolution. Ever since, industrial revolutions played a major role in economic history and can be, so far, divided into three stages. The first industrial revolution implemented mechanical production lines, the second revolution introduced electrical and motoric drives, enabling assembly-line work, and the third revolution further promoted automatization through computers and IT (Matt & Rauch, 2020). Hence, Industry 4.0 continues this eternal evolution, whereby its realization is just as crucial as the before incurred changes to enable competitiveness, economic growth, and innovation. Many countries recognized the importance of Industry 4.0 and announced their plans for implementation. Germany served as a model for other countries with their introduction of “Industrie 4.0”, followed by China’s “Made in China 2025” initiative that is inspired by the German counterpart. But also many other countries launched initiatives, as Japan “Society 5.0”, or USA “Manufacturing USA”, to name some examples (Matt & Rauch, 2020).
Yet, there is a dichotomy regarding small and medium-sized enterprises (SMEs) and their role in the progress towards Industry 4.0. SMEs play a major role in this fourth industrial revolution, because of their flexible and innovative traits, but simultaneously often struggle to
6
implement it, due to less resources to assess potentials and risks (Matt & Rauch, 2020). So far, few research has been done on the explanation for different handlings of Industry 4.0 by SMEs. This is despite the fact that there are many potential factors influencing SMEs. For example, the level of information and communications technology (ICT) development, the availability of financial and labor resources, or public subsidies supporting innovation. However, the implementation of Industry 4.0 varies just as well across countries that have similar features with regard to the latter factors. Thus, the question may be to what extent variations are due to differences in nationality rather to differences in certain prerequisites. In other words, the implementation of Industry 4.0 may be often influenced by “soft” properties instead of hard economic facts. The best way to analyze these “soft” properties is to look at the national culture. In recent years, there was a growing awareness for cultural impacts on economic activities. Furthermore, various studies show that SMEs are particularly affected by the national culture and are especially uncertainty averse. Uncertainty aversion, is an important aspect of the examination of cultural determinants, as its severity varies strongly dependent on the country. As a result, I will examine the cultural causes for different attitudes of SMEs towards Industry 4.0.
My research question is consequently: “What is the impact of the national culture on SMEs’ participation in Industry 4.0?” Answers will be offered by an empirical cross-country analysis, comparing SMEs from 118 developing and developed countries over the whole world. My study contributes to the existing literature in two respects. It expands the research on SMEs’ decisive but torn role for Industry 4.0, as most studies are still done on large enterprises. In addition, it is to my knowledge the first study bridging national culture to the implementation of Industry 4.0, in general as well as for SMEs. The study is structured as follows: First, the concept and global implementation of Industry 4.0 is reflected and linked to SMEs. In particular, challenges and opportunities of Industry 4.0 for SMEs are named and the particular
7
role of SMEs in Industry 4.0 emphasized. Thereupon, the need for a cultural perspective on this issue is pointed out and the concept of national culture is defined. After providing this theoretical basis, the data, the research hypotheses, and the further methodological proceeding are explained. Building on these technical explanations, the main results are shown and discussed with the distinction between national culture’s direct and moderating effect. In addition, a differentiated analysis of developed and developing countries will be carried out. A robustness check will be performed to ensure the generalizability of the results. Lastly, the shortcomings of the study are noted and concluding remarks given, followed by a short conclusion.
2. Theoretical background
2.1. The concept of Industry 4.0
Since the “Working Group Industry 4.0” published their final report in 2013, Industry 4.0 is an omnipresent issue in the field of economics. Many debates on its advantages and challenges were carried out, but the general consensus is that the fourth industrial revolution will take place and that sooner or later all countries have to cope with it (Rüßmann, Lorenz, Gerbert, Waldner, & Justus, 2015; Matt & Rauch, 2020). Industry 4.0 in simple terms is the unification of known industrial processes and information and communication technology (Matt & Rauch, 2020). The idea of Industry 4.0 can be divided into the following nine components (Rüßmann, Lorenz, Gerbert, Waldner, & Justus, 2015):
8
Figure 1: Nine components of Industry 4.0
The implementation of Industry 4.0 is thought to provide a stable basis for future challenges due to increased efficiency, sustainability, and competitiveness. Furthermore, it will enable more customized products and ease the global supply chain (Matt & Rauch, 2020). However, this progress entails challenges for business at the same time. One greater problem is the increasing need for well qualified work force with mechatronic and software skills, as these two areas will be united more and more in the future. In addition, many enterprises shy away from the large sums they have to invest, to meet the requirements of the industry of the future (Rüßmann, Lorenz, Gerbert, Waldner, & Justus, 2015). These new challenges and opportunities of Industry 4.0 make a provision of the best possible conditions, for example through subsidies, by governments even more important.
The cloud The Industrial Internet of Things Additive manufacturin g Autonomous robots Augmented reality Horizontal and vertical system integration Big data and
analytics Simulation
Cybersecurity
9
2.2. Implementations of Industry 4.0
The perceived importance of Industry 4.0 becomes clear with a view to the programs introduced by the top manufacturing countries to promote a quick transformation towards a high-tech industry. Generally one can differentiate between two motives for first mover attempts. Developed countries are currently at a crossroad as they run into danger to forfeit their leading positions to developing countries. Therefore their motive for an implementation of Industry 4.0 could be described as an attempt to maintain the status quo. In contrast, some developing countries lag behind and have not even reached the current industrial standards. For these countries the upcoming industrial change is the opportunity to leap the present backlog. Furthermore, leading manufacturing countries are often not well developed regarding digital skills, while highly digitally skilled developing countries catch up in manufacturing aspects (Richter, 2020). Thus, the race for the most innovative and efficient industry will be exciting to observe and will have long-term consequences till the dawn of the next industrial revolution. At the following, I will mention some Industry 4.0 programs of different countries to illustrate the width as well as the differences of the approaches.
One example for such a government program is “Industrie 4.0”, Germany’s pioneering approach to intelligent manufacturing (Klitou, Conrads, Rasmussen, Probst, & Pedersen, 2017). It is a long-term strategic initiative and a core of Germany’s future economy with several funds (Klitou, Conrads, Rasmussen, Probst, & Pedersen, 2017). The German chancellor Merkel stresses the need for a quick transition to Industry 4.0 and rises concerns that otherwise other countries with digital expertise will take over leadership in industrial production (Schroeder, 2016). China’s “Made in China 2025” initiative can be seen as a reaction to “Industrie 4.0” (Kennedy, 2015). Although “Made in China 2025” is a broader formulated initiative with impacts on all economic sectors, its focus also lies on innovative manufacturing. The proclaimed final goal of the initiative is to become the global top industry by replacing its main
10
competitive advantage of cheap labor with a high-end industry (Matt & Rauch, 2020). The need for change is there, because the present advantage of low labor costs will lapse due to the new efficiency of Industry 4.0 (Zhang, 2018). Following these assumptions, one could make the assertion that the future of China’s economy is, one way or another, highly dependent on the future development of Industry 4.0. In 2014, “Manufacturing USA” was introduced by the US government, whereas the goal is “nothing less than U.S. global leadership in advanced manufacturing” (National Insitute of Standards and Technology, 2018, p. 1). This initiative established a platform with 14 institutes in the whole country, over 1900 member organizations, more than 475 R&D collaborations, and a fund that was risen to $1 billion per year (National Insitute of Standards and Technology, 2018; McCormack, 2012). Japan’s approach to Industry 4.0 is, in turn, construed as an entire change in the society. Consequently, the Japanese government named it “Society 5.0” and hope for better solutions for social as well as industrial matters through digitization, artificial intelligence, and robots (The Government of Japan, 2020). This technological change is one of the main goals of the Abenomics, which became a synonym for the policy introduced by the Japanese Prime Minister Abe Shinzo to revive the weakened Japanese economy (The Government of Japan, 2020). Germany, China, the US, and Japan are currently the four leading manufacturing industries in the world (Richter, 2020). However, the lack of digitization in these countries become clearer with a look to the IMD World Competitiveness Digital Ranking (2019). According to this report, the US is the only manufactural leader that is represented in the top four of the most digitalized countries (IMD, 2019). In contrast, developing countries as Singapore emerge. Singapore ranks second in global digital competitiveness (IMD, 2019) and also recognized the importance to connect this advantage with a more developed industry. “SGInnovate” is a state-owned organization that committed itself to foster skilled labor forces, to deepen and expand technological exchange and networks, and to invest in start-ups and business creations (SGInnovate, 2020). The final goal is “to build 'technology-intensive' products borne out of science research” (SGInnovate,
11
2020). According to SGInnovative (2020), Singapore spent $19 billion until now for research in advanced manufacturing, biomed sciences, urban solutions and sustainability. Furthermore, Singapore is ranked 5th in the Global Innovation Index 2018 (Cornell University, INSEAD, and
WIPO, 2018), while the Global Competitive Index (2019), which measures the global growth potential, even rates Singapore as the most competitive country at the moment (World Economic Forum, 2019). All these numbers underpin Singapore’s strong starting point for the future and show how emerging and developing countries can rise through Industry 4.0. But there are also developing countries that did not yet take the necessary steps towards a modernized industry. Such examples are India and Brazil, which are both among the ten largest economies in the world (World Population Review, 2020). However, the question might be how much longer this is the case, as both countries lack to advance the positive trend they performed for some years. This stagnation also gets obvious with a glance to their recent global competitiveness measures, where Brazil is ranked 71st and India 68th, lying behind countries as
Kazakhstan or Oman (World Economic Forum, 2019). India’s competitiveness was even downgraded by 10 ranks in comparison to 2018, which is the largest decrease of all countries. Brazil’s development is also not surprising, as the Brazilian government launched its Industry 4.0 chamber in 2019, years after other countries started their programs. The Brazilian Industry 4.0 chamber consist of more than 30 institutions, firms, and universities and published the “Brazilian agenda for Industry 4.0” (MCTIC, 2019). However, up to now the expectations of Brazilian firms directed to Industry 4.0, in particular by SMEs, are rather low (BSA Foundation, 2020). This cautious attitude is not only an issue in Brazil. Many countries experience that SMEs are in particular skeptical about the implementation of Industry 4.0 as the initial years of several government programs went by. Despite their particular significance for Industry 4.0 in the global economy, most governmental programs struggle so far to convince SMEs to participate. This problem is broadly recognized, which is why there are more and more studies
12
on the potentials and challenges of Industry 4.0 for SMEs (Orzes, Rauch, Bednar, & Poklemba, 2018; Matt & Rauch, 2020; European Union, 2018).
2.3. Small and Medium-sized Enterprises in Industry 4.0
First, it is important to clarify the term and the global role of SMEs. Most commonly, SMEs are defined as firms with fewer than 250 employees, while they are subdivided in medium-sized enterprises with 50-249 employees, small enterprises with 10-49 employees, and micro firms with fewer than 10 employees. This complies with the definition of the OECD (OECD, 2005). SMEs provide more than 50 percent of worldwide jobs and represent 90 percent of global businesses (The World Bank, 2020). SMEs are often ahead of large enterprises concerning robustness, flexibility, and adaptability which is why some studies even claim that SMEs are the most innovative and digitization-propulsive business sector (Del Giudice, Scuotto, Garcia-Perez, & Messeni Petruzzeli, 2019). One proof of this, was not least the financial crisis, which underpinned SMEs’ driving force for the global economy (Matt & Rauch, 2020). Due to this importance in global economy, many governments set the focus of their programs on SMEs. In theory, SMEs provide a good basis for Industry 4.0, as they tend to implement novel manufacturing elements quickly, due to high competitiveness pressure (Matt & Rauch, 2020) and loser, more flexible organizational structures (Deloitte, 2015). Thus, SMEs can make up for their disadvantage in knowledge and financial resources with their organizational flexibility and the possibility of a quick and broad mobilization of their employees (Moeuf, et al., 2020). Furthermore, several SMEs already demonstrated the successful implementation of digitized processes in the run-up to Industry 4.0, to profit from a unique selling point (PWC, 2015). Some experts even give the prognosis that SMEs could be the primary beneficiaries of an industrial change (Sommer, 2015; Matt & Rauch, 2020).
Beside the important role of SMEs for the worldwide implementation of Industry 4.0, there are also many advantages of the latter. By now, intelligent technologies are more
13
affordable and also more suitable for SMEs (Moeuf, et al., 2020). Thus, Industry 4.0 could lead to production costs and production lead time reductions, and improvements in the quality of products and the work force productivity (European Union, 2018). ICT adoption is associated with inclusive growth and the access to a broader market. This effect is particularly observable for smaller firms and is likely to lead to less inequalities in wages (Hallward-Driemeier & Nayyar, 2018). Furthermore, SMEs will be more capable to meet the increasing demand for more personalized products and services, because digital technologies enable more differentiation possibilities (European Union, 2018). Until now, SMEs often did not have the resources for the modification of products and services, because they mainly had to focus on economics of scale, while customization processes entailed high costs (Matt & Rauch, 2020). The access to the global value chain is another big advantage of Industry 4.0, as it will enable SMEs better possibilities to coordinate their processes globally (Hallward-Driemeier & Nayyar, 2018).
2.4. Challenges of Industry 4.0 for SMEs
Despite the opportunities generated by Industry 4.0, SMEs also face new challenges with the rise of a more modern industry. Several studies found that SMEs face more challenges regarding Industry 4.0 compared to large enterprises. Some state that the firm size is a crucial factor when it comes to engagement in Industry 4.0, whereby SMEs suffer a disadvantage (European Union, 2018; Matt & Rauch, 2020; Sommer, 2015). Furthermore, SMEs often miss business support, are more cautious to invest in new R&D, and often do not have access to work force with the right qualifications (Kleindienst & Ramsauer, 2016). In addition, SMEs will also face more and more difficulties to employ digital natives in the future, if they do not implement state-of-the-art technology. Many SMEs struggle to reorganize their structure according to the needs of Industry 4.0 and to build technological foundations (European Union, 2018). Cybersecurity and the reliability of new technologies is another main concern of SMEs over
14
the world, as the connection of important firm data with all processes will be vulnerable if no sufficient safeguards are implemented (McKinsey Digital, 2015; Sommer, 2015). The funds risen by some countries to enhance technological innovation often run into danger to support firms that are already the most progressive ones, and thus drive small innovative firms out of the market (Zenglein & Holzmann, 2019). Governmental regulations will also play an increasing role in Industry 4.0 with the introduction of standards (European Union, 2018). Consequently, SMEs, especially in highly regulated sectors as healthcare, will be forced to move with the times to meet the statutory provisions. Thus, for some SMEs the transformation to Industry 4.0 is not a matter of choice, but a condition to remain at the market. As SMEs have disadvantages in all these areas, their competitiveness is in danger and they might even be forced to cooperate with competitors with greater resources (European Union, 2018). All these concerns of SMEs can be subsumed under the perception of uncertainty. Whether the lack of skilled labor force, the difficulty of internal restructuring measures, cybersecurity concerns, or governmental regulations, all these issues imply the risk of higher expenditures and the uncertainty whether such investments will indeed pay off. As a result, SME-specific strategies and solutions are needed to enable a successful and clearly defined transformation of SMEs from Industry 3.0 to Industry 4.0 (Matt & Rauch, 2020).
2.5. National differences in SMEs’ attitudes towards Industry 4.0
. The severity of the above mentioned challenges affecting SMEs vary from country to country. One can generally say that all SMEs over the world are affected by Industry 4.0 and its uncertainties, however their attitudes towards it differ. This circumstance has several reasons, because SMEs’ handling of Industry 4.0 is influenced by the domestic economic and infrastructural state, the subsidies provided by the government, and the felt pressure for change.
In general, there are several forms of uncertainty arising with the emergence of new technologies and industries. Firms are uncertain about the effectiveness of new technologies
15
and their right implementation in existing processes and strategies (Allen & Gale, 1999). However, the severity of these uncertainties are highly dependent on the present financial circumstances in a country (De Jong E. , 2009). The financial system of countries is often distinguished between market-based and bank-based systems. In bank-based systems, as in Germany or Japan, banks play the major role in the allocation, mobilization, and provision of capital. Market-based systems, in contrast, are rather driven by financial markets, as securities markets. England and the United States are typical examples for such financial systems (Demirgüç-Kunt & Levine, 1999). Allen and Gale (1999) stress the advantage of market-based systems for the emergence of new technologies, because there is a wider range of possible investors, who can draw their own conclusions on the value of the technologies. In bank-based systems, in contrast, the intermediary will weigh the advantages and disadvantages of an investment, which is likely to result in a refusal of an investment due to uncertainty (Allen & Gale, 1999). Vitols (2004) confirms that bank-based systems are often weaker concerning appropriate support for R&D and the implementation of new technologies (Vitols, 2004). However, there is also contradictory literature stating that neither bank-based nor market-based characteristics of a financial system have a significant influence on industrial growth (Beck & Levine, 2002).
To illustrate the national differences in SMEs’ attitude towards Industry 4.0, the variety of main concerns of SMEs will be pointed out for different countries. German SMEs, for example, mainly expect Industry 4.0 to have an impact on large enterprises and have concerns regarding the protection of data ownership. However, they have comparatively few concerns regarding financial resources and the implementation of new business models (Müller & Voigt, 2018). Chinese SMEs, in contrast, expect a large impact by Industry 4.0 due to efficiency gains and social benefits, while they fear new competitors and miss financial resources (Müller & Voigt, 2018). In the USA the development of smart manufacturing for SMEs is rather slow.
16
American SMEs lack of an access to appropriate financial resources and know-how to implement new technologies. Furthermore, American SMEs miss the mediation of highly qualified labor forces by the introduced departments for innovation (Tantawi, Fidan, & Anwar, 2019). But the main concern of American SMEs is the lack of clear policies guiding the path to a successful transformation, because most concepts for Industry 4.0 are conducted by large private enterprises for their specific sector (Shaheen, 2015; Sabo, 2015). According to the Institution of Mechanical Engineers (2019), only 29 percent of the SMEs in the UK have a strategy for Industry 4.0. However, 26 percent state that this circumstance is due to a lack of appropriate infrastructure and 13 percent criticize too few financial resources. Furthermore nearly 80 percent of UK’s SMEs have problems to recruit skilled work force and only 16 percent plan to invest more than 8 percent of their turnover in R&D (Institution of Mechanical Engineers, 2019). In Japan the main problem is that very few SMEs have incorporated IT due to high costs and a lack of IT-specialists (Kinoshita, 2019). Despite the strong development of Singapore’s economy in terms of innovation and competitiveness in recent years, its SMEs are struggling to keep up. Many SMEs in Singapore still have to get used to the need for cooperation with competing firms and the willingness to learn from other more successful firms. As this new style of “open innovation” contradicts the common management style of Singapore’s SMEs, it will be a process for its leaders to build mutual trust and an open view (Chan, 2019).
2.6. The consideration of national culture
As can be seen, the concerns of SMEs about Industry 4.0 vary widely, independent of structural similarities or identical prerequisites. In point of fact, skepticism against unknown factors as Industry 4.0 is often strongly based on biases. Such national biases are in turn strongly dependent on the national culture. Thus, some causes for a muted development of SMEs towards Industry 4.0 may be rather due to cultural and social habits than to economic conditions. In addition, as stated multiple times in the previous sections, the attitude of SMEs
17
towards uncertainty and risk plays a major role in their participation in Industry 4.0. So far, the influence on SMEs’ handling of Industry 4.0 by cultural attitudes as uncertainty were no subject of research. Hence, the consideration of national culture could contribute to the puzzle of SMEs’ simultaneous importance and passivity for Industry 4.0.
The exploration of national cultures, of which the attitude towards uncertainty is a main aspect, was determined at an early stage as a main reason for competitive differences between countries (Porter, 1990) and is subject of many studies since. Additionally, the introduction of Hofstede’s cultural dimensions facilitated the comparison of different cultural imprints and its integration into models (Andrijauskienė & Dumčiuvienė, 2017). Hofstede’s cultural dimensions capture, inter alia, a measure for the degree of anxiousness towards uncertainty by a society, defined as uncertainty avoidance (UAI). The other five dimensions are: 1. power distance (PDI), 2. individualism vs. collectivism (IDV), 3. masculinity vs. femininity (MAS), 4. Long-term orientation vs. short-term orientation (LTO), and 5. Indulgence vs. restraint (IVR) (Hofstede G. , National Culture, 2020). PDI measures the handling of inequality in a society, while IDV indicates whether individuals in a society rather stress their sole individuality or their membership in a group, for example in the family. The reflection of MAS specifies whether a society is rather “tough” (masculine) or “tender” (feminine). In the context of business, LTO represents whether behavior is “normative” (long-term orientation) or “pragmatic” (short-term). IVR reflects whether a society and its satisfaction of needs is driven by rather lose or strict social norms (Hofstede G. , National Culture, 2020).
Hofstede’s dimensions are shown to be a good tool to examine the impact of culture on economic issues (Handoyo, 2018; Kaasa, 2017; Cox & Khan, 2017; De Jong E. , 2009). For example, plenty of studies prove that innovation is dependent on the national culture. Thus, national culture has an impact on the successful transformation to Industry 4.0, as the innovativeness of a country is a decisive aspect for technical advance and digitization.
18
Andrijauskienė and Dumčiuvienė (2017) find that power distance and uncertainty avoidance have a negative impact on innovation, while individualism and indulgence increase the innovativeness of a country. This is supported by Cox and Khan (2017) as well as Kaasa (2017), who verify the positive effect of indulgence and individualism. However, Cox and Khan (2017) stress the importance of femininity and long-term orientation, while Handoyo (2018) and Andrijauskienė and Dumčiuvienė (2017) claim that masculinity and long-term orientation have no effect. Additionally, Kaasa (2017) finds that individualism increases innovation, which is confirmed by Handoyo (2018). Similarly, Lee and Peterson (2000) elaborate that cultures with low power distance, low uncertainty avoidance, high masculinity, and high individualism are more likely to lead to proactive, innovative and risk-taking entrepreneurial spirits, and as a result to global competitiveness.
De Jong and Semenov (2004) link differences in financial systems to Hofstede’s cultural dimensions. They find that market-based systems are associated with lower uncertainty avoidance. Furthermore, a market-based system is more suitable for highly masculine and individualistic societies with low power distance. A bank-based system is rather compatible with societies scoring low on masculinity, individualism and uncertainty avoidance, and a high score on power distance (De Jong & Semenov, 2004). Moreover, uncertainty-accepting, market-based countries have an advantage in terms of more flexible labor markets, more open economies, and more innovation (De Jong E. , 2009). The latter three aspects are all decisive for a successful implementation of Industry 4.0 and support more risk taking. Generally speaking, the institutional conditions of a country are highly influenced by the national attitude towards uncertainty. De Jong (2009) expects innovative industries to play a bigger role in countries with low uncertainty avoidance and explains that uncertainty avoidance has also a negative effect on R&D. Following these considerations, the implementation of Industry 4.0 is supposed to be more advanced in market-based countries as the US or England.
19
This selection of studies illustrates well how different the findings on the individual cultural dimensions may turn out, dependent on the sample and the method used. Nevertheless, all the studies show that there is indeed an effect of national culture on economic issues as the innovativeness of a country.
2.7. The impact of national culture on SMEs
Few research has been done on the specific impact of national culture on SMEs, although SMEs are particularly affected by cultural elements. The surrounding norms and beliefs have a greater influence on SMEs’ practices compared to large or multinational enterprises, which consist of multiple and often diverse cultures (Fernández-Esquinas, Van Oostrom, & Pinto, 2017; Per, 2012; Seteroff & Campuzano, 2010; Tran & Jeppesen, 2018). Consequently, SMEs are economic entities for which the found effects of Hofstede’s cultural dimensions are even stronger. This is supported by an increasing number of studies (Graham, 2014; Fernández-Esquinas, Van Oostrom, & Pinto, 2017), showing that the cultural influence on SMEs’ innovativeness could either be an enhancing strategic asset, or an obstacle for innovation. Thus, analyses on the innovativeness made by researchers or governments have to take the cultural factor into account, to have significant impact (Bernat, Bruska, & Jasińska-Biliczak, 2017). However, innovation is not the only aspect, regarding SMEs, that is influenced by culture. National culture has an influence on the capital structure of SMEs (Fairbarin, Henry, & Tsalavoutas, 2015), suggesting that dependent on the culture SMEs might be less willing to invest money to implement Industry 4.0. The impact of culture on SMEs’ management practices is investigated by Graham (2014), determining that power distance is more an issue in SMEs, as their owners often have more influence and represent the entire business. This enlarged influence of owner-specific characteristics and aims often aggravate the lack of innovation (Brown, 1997). The business leader of a SME is decisive for the implementation of Industry 4.0, as the majority of the decisions are determined by him. In contrast to larger enterprises,
20
where decisions are carried out by committees, SMEs are often rather a one man show with centralized decision-making power (Moeuf, et al., 2020). Similarly, the entrepreneurship is affected by national culture (Linán & Fernandez-Serrano, 2014), which is crucial to maintain competitiveness in times of Industry 4.0. Furthermore, SMEs avoid long-term debts due to their distinct uncertainty avoiding characteristic. The same negative relation to long-term debt holds for the dimension of individualism, indicating the importance for SME leaders to maintain independent control. Power distance has also a negative effect on debt for SMEs, because SMEs in high power distance countries have less voice concerning debt facilities. In contrast, SME leaders in masculine countries are more willing to take on debts to gain growth (Mac an Bhaird & Lucey, 2014). With the emergence of new technologies, it is often inevitable to take out a loan, and hence to take on debts. Thus, SMEs’ attitude towards investments in new technologies through loan capital is strongly driven by their national culture. It is shown that national culture affects SMEs’ attitude towards risk and proactive actions (Acton, 2011), which are both important characteristics for the handling of Industry 4.0. Yeboah (2014), finds that SMEs in countries scoring high on masculinity and power distance are more risk-taking.
Besides the cultural impact, the attitude toward risk is another factor that steers SMEs’ decision-making (Cabral, Pontes, & Forte, 2016; Verreynne, Williams, Ritchie, Gronum, & Betts, 2019). First, it is important to make a clear distinction between uncertainty and risk. The importance of a differentiation between the cultural measurable uncertainty avoidance and risk avoidance is also stressed by Hofstede (2001, p. 148). Both notions are similar and therefore often equated. However, Knight (1921) stresses that risk differs from uncertainty in the respect that it applies to actually existing situations where the potential outcomes are known and the odds of success can be estimated. For (fundamental) uncertainty, as Knight puts it, it is not possible to measure the “unknown” of a situation. This is due to the fact that uncertainty covers situations that entail no probabilities and are often perceived subjectively, without factual
21
grounds (Knight, 1921). Thus, the main factor for the distinction of risk and uncertainty is the possibility to anticipate probabilities (Toma, Chitita, & Sarpe, 2012). Some studies found that decision-makers of SMEs are more risk-averse compared to the ones of large enterprises due to limited access to resources and less technical and managerial capabilities (Wüstermann, 2016; Brown, 1997). For this reason, the introduction of new technologies is also associated with much higher risk for SMEs than for large enterprises, because false decisions are more difficult to compensate. The level of risk aversion is in turn strongly influenced by cultural values (Breuer, Riesener, & Salzmann, 2014), which are intensified for SMEs. Consequently, risk and uncertainty are of particular significance for SMEs in two respects. In economic surroundings, however, uncertainty is often the dominant factor, as the agents are part of a complex system with many other influencing agents (Kastelle, 2013). This is also the case for governmental programs and the changes of Industry 4.0, as they imply primarily uncertainty. For this reason the focus of this study will not be on risk, but on the impact of uncertainty and national culture in general.
3. Research Objective and Research Hypotheses
After providing a theoretical background, the research objective and the research hypotheses will be elucidated. The objective of the study is to analyze the impact of national culture on SMEs’ participation in Industry 4.0 in consideration of the economic environment. The secondary objective is to examine the moderating impact of national culture on SMEs’ participation in Industry 4.0 through its influence on the economic environment. Thus, the focus of the study is not put on risk aversion, but on uncertainty avoidance and the other corresponding cultural dimensions of Hofstede. In the following, research hypotheses will be formulated with reference to the related literature presented above. It should be noted that the hypotheses are made under the assumption that all countries have the same economic structure, in the sense of SMEs’ relative contribution to the whole economy.
22
As indicated above, SMEs are strongly driven by their avoidance of uncertainty, which bars the way to progress. Further on, one can associate an emphasis on power distance of SME business leaders with a less open attitude towards changes and hence less activities towards Industry 4.0.
Hypothesis 1: SMEs have less technical capacities in countries with high uncertainty avoidance (H1.1) and power distance (H1.2), and hence participate less in Industry 4.0.
In contrast, cultures with high masculinity and individualism are more advancing and open towards novelties. Long-term orientation was found by various researchers to enhance innovativeness. An advancing, open and innovative attitude is likely to facilitate future-oriented investments.
Hypothesis 2: SMEs have more technical capacities in countries with high masculinity (H2.1), individualism (H2.2), and long-term orientation (H2.3), and hence participate more in Industry 4.0.
Furthermore, national culture may interact with the control variables and hence have a moderating influence on the technical capacities of SMEs. Countries scoring low on power distance and uncertainty avoidance are found to be more advanced in the adoption of ICT. In addition, more individualistic countries show higher ICT adoption rates. Masculinity is expected to increase the ICT adoption, whereas long-term orientation tends to hamper ICT adoption. However, the effect of both masculinity and long-term orientation are expected to be insignificant for the ICT adoption of countries (Erumban & De Jong, 2005; Sriwindono & Yahya, 2012).
Hypothesis 3: The higher the level of power distance (H3.1) and uncertainty avoidance (H3.2), the weaker the positive effect of ICT development on SMEs’ technical capacities and hence the smaller the participation in Industry 4.0. In contrast, the higher the level
23
of individualism (H3.3) the stronger the positive effect of ICT development on SMEs’ technical capacities and hence the greater the participation in Industry 4.0.
Individualism is expected to have a positive influence on the venture capital availability in a country, while uncertainty avoidance has a rather negative impact (Antonczyk & Salzmann, 2012). Similarly, power distance and masculinity are negatively associated with venture capital activities (Sarajuuri, 2018). Long-term orientation is positively related to venture capital (Gantenbein, Kind, & Volonté, 2019).
Hypothesis 4: The higher the level of power distance (H4.1), uncertainty avoidance (H4.2), and masculinity (H4.3), the weaker the positive effect of venture capital availability on SMEs’ technical capacities and hence the smaller the participation in Industry 4.0. In contrast, the higher the level of individualism (H4.4) and long-term orientation (H4.5), the stronger the positive effect of venture capital availability on SMEs’ technical capacities and hence the greater the participation in Industry 4.0.
The educational level of the workforce in a country may also be influenced by the national culture. Uncertainty avoidance and power distance, for example, have a negative effect on the education of workforce, whereas individual countries are more likely to have competitive workforce. Long-term-orientated countries are more likely to have well educated workforce, as future oriented behavior is more likely to lead to continuing education (Rensink, 2016). Moreover, more masculine countries are expected to be more competitive and entrepreneurial and hence have better educated workforce (Fayolle, 2007).
Hypothesis 5: The higher the level of power distance (H 5.1) and uncertainty avoidance (H5.2), the stronger the negative effect of inadequately educated workforce on SMEs’ technical capacities and hence the smaller the participation in Industry 4.0. In contrast, the higher the level of individualism (H5.3), long-term orientation (H5.4), and
24
masculinity (H5.5), the weaker the negative effect of inadequately educated workforce on SMEs’ technical capacities and hence the greater the participation in Industry 4.0.
There is few literature on possible effects of national culture on public support and state subsidies. Nevertheless, one would expect countries with high power distance to provide less subsidies, while the contrary applies to more individualistic countries (Kammas, Kazakis, & Sarantides, 2017). Furthermore, uncertainty-avoiding, masculine and long-term orientated countries are more likely to provide public support. Public support is especially important in highly uncertainty-averse cultures to decrease peoples fear towards investments. Long-term orientated countries put a bigger emphasis on the maintenance of their competitiveness and hence hustle further development. As mentioned before, more masculine cultures are more advancing, which increases the need for public subsidies.
Hypothesis 6: The higher the level of power distance, the weaker the positive effect of public support on SMEs’ technical capacities and hence the smaller the participation in Industry 4.0 (H6.1). In contrast, the higher the level of uncertainty avoidance, (H6.2), individualism (H6.3), masculinity (H6.4), and long-term orientation (H6.5), the stronger the positive effect of public support on SMEs’ technical capacities and hence have more technical capacities and hence the greater the participation in Industry 4.0.
Now that the hypotheses are developed, the utilized sample and the research method will be explained in the next section.
4. Data and Research Model
An empirical cross-country analysis will be conducted to examine the cultural impact on SMEs’ participation in Industry 4.0. Several country-level control variables will be added, taking economic conditions into account that influence SMEs’ handling of Industry 4.0. In addition, Hofstede’s cultural variables will be integrated to estimate the effects of the national
25
culture on SMEs’ participation in Industry 4.0. If possible, the data is gathered for the same years for each country, to ensure an analysis as comparative as possible. For individual cases that miss data for that year, the most recent data is used.
3.1. Dependent Variable
The dependent variable indicates the national participation of SMEs in Industry 4.0 and is represented by the technical capacities of SMEs in a country (TC). The technical capacities of SMEs imply quite some informative value for their participation in Industry 4.0, since they will only be able to stay innovative and competitive through technical progress. Hence, better/worse technical capacities will lead to a more developed/less developed implementation of Industry 4.0 by SMEs. This is also identified by Jones-Evans et al. (2018), who stress the importance of appropriate technological capacities as one main pillar for the open innovation of SMEs. Other authors also discuss the importance to measure SMEs’ innovation through their inputs, as for example their technological capacities. In contrast, the measurement of the outputs, i.e. products or revenues, would be inappropriate, because that only reflects SMEs’ performance, but not their ability for innovation (Saunila, 2017). Thus, the technical capacities of SMEs are an appropriate tool to estimate their ability to cope with Industry 4.0.
The indicators for TC are gathered from the World Bank (2016) and the italic written name complies with the official definition of the index. The index of TC classifies countries into five ordered categories, going from 0 to 4. The number of 0 indicates very poor technical capacities, 1 indicates poor technical capacities, 2 indicates fair technical capacities, 3 indicates good technical capacities, and 4 indicates excellent technical capacities. Hence, the variable of TC is a categorical variable that only consists of integer values between 0 and 4. The distribution of TC’s categories is shown in Table 1.
26
Table 1: Distribution of TC
TC Freq. Percent Cum.
Very poor 6 5.08 5.08 Poor 42 35.59 40.68 Fair 36 30.51 71.19 Good 27 22.88 94.07 Excellent 7 5.93 100.00 Total 118 100.00
3.2. Main Variables for Cultural Imprints
In addition, Hofstede’s dimensions of power distance (PDI), individualism vs. collectivism (IDV), masculinity vs. femininity (MAS), uncertainty avoidance (UAI), and long-term orientation vs. short-long-term orientation (LTO) will be utilized as main explanatory variables, indicating differences in national cultures. The dimension of Indulgence vs. restraint (IVR) is left out of the model, because it is not given for many countries and the addition would have reduced the sample substantially. Thus, it is supposed that a bigger sample size will give more meaningful results than the addition of the variable IVR.
The majority of the cultural dimension indexes are obtained from Hofstede’s official research website (Hofstede G. , 2015). Missing cases particularly affect African and Latin American countries. Values for missing African countries are derived from Van Pinxteren’s measure (Van Pinxteren, 2018), which complies with Hofstede’s measures. This was assured by comparing the values of countries as Ghana and Egypt that are given by both authors. If there are deviations between the two measures, they are marginal and do not distort the results.1
Hofstede’s cultural variables are limited to values between 0 and 100 for this study. In general, cultural variables with values above 50 are considered as high, whereas values below 50 are
27
considered as low. Table 2 shows the correlations between the dependent variable TC and Hofstede’s five cultural dimensions.
Table 2: Correlation matrix of dependent and cultural variables
TC pdi idv mas uai lto
TC 1 pdi -0.296** 1 idv 0.495*** -0.691*** 1 mas 0.150 0.063 0.097 1 uai 0.330*** 0.185* -0.147 -0.010 1 lto 0.423*** -0.134 0.324*** 0.025 0.096 1 * p < 0.05, ** p < 0.01, *** p < 0.001
As can be seen, Hofstede’s dimensions are a good measurement for cultural imprints, as they capture a lot of the cultural effects on SMEs’. This is also shown by other authors, using Hofstede’s dimensions to study the cultural impact on SMEs activities (Munyanyi, Chiromba, Magweva, Bizah, & Diza, 2018; Yeboah, 2014; Graham, 2014).
Below, scatterplots of TC and the cultural variables are given to illustrate the relationships between Hofstede’s dimensions and the technical capacities of SMEs.
AGO ALB ARE ARG ARM AUS AZE BAN BDI BEL BFA BHR BOL BOS BRA BRN BUL BWA CAN CHE CHI CIV CMR COG COL CRI CYP DEU DNK DOM DZA ECU EGY EST ETH FRA GAB GBR GEO GHA GIN GRC GTM HND HRV HTI IDN IRL ISR ITA JAM JOR JPN KAZ KEN KHM KOR KWT LBN LBR LBY LKA LTU LUX LVA MAR MDG MEX MKD MLI MLT MMR MNG MOZ MRT MUS MYS NAM NGA NIC NLD NOR NPL NZL OMN PAK PAN PERPOL PHL PRT PRY QAT ROURUS RWA SAU SEN SGP SLV SRB SVK SVN SWE SYR TCD THA TUN TUR TZA UGA UKR URY USA VEN VNM ZAM ZIM 0 1 2 3 4 T C 20 40 60 80 100 pdi
28
Figure 2 illustrates that in the chosen sample, countries with very poor technical capacities for SMEs (TC=0) have very high values for power distance. For these countries the lowest score on power distance (TC=0) is about 70. The other categories of TC have a more balanced distribution concerning power distance.
Figure 3: Relationship between SMEs' technical capacities and individualism
Figure 3 shows a quite balanced distribution of individualism for countries with the TC categories 1, 2, and 3. Nevertheless, one can observe that the values for individualism rise with increasing technical capacities of SMEs in the chosen sample. Noticeable is that all countries with very poor technical capacities for SMEs (TC=0) have small individualism values below 40. The countries with excellent technical capacities for SMEs (TC=4), in contrast, have rather high values, with only one country having a value lower than 40.
AGO ALB ARE ARG ARM AUS AZE BAN BDI BEL BFA BHR BOL BOS BRA BRN BUL BWA CAN CHE CHI CIV CMR COG COL CRI CYP DEU DNK DOM DZA ECU EGY EST ETH FRA GAB GBR GEO GHA GIN GRC GTM HND HRV HTI IDN IRL ISR ITA JAM JOR JPN KAZ KEN KHM KOR KWT LBN LBR LBY LKA LTU LUX LVA MAR MDG MEX MKD MLI MLT MMR MNG MOZMRT MUS MYS NAM NGA NIC NLD NOR NPL OMN NZL PAK PAN PER PRTPHL POL PRY QAT ROU RUS RWA SAU SEN SGP SLV SRB SVK SVN SWE SYR TCD THA TUN TUR TZA UGA UKR URY USA VEN VNMZIM ZAM 0 1 2 3 4 T C 0 20 40 60 80 100 idv
29
Figure 4 reports no conspicuities with regard to the relationship of masculinity and SMEs’ technical capacities. The only exception are the countries with very poor technical capacities for SMEs (TC=0), as all of them but one have very similar masculinity values. Of these countries all have masculinity values around 40 and 50, which is neither considerably high nor low. Only Sri Lanka (indicated as LKA) is noticeable lower on masculinity with a value around 10. AGO ALB ARE ARG ARM AUS AZE BAN BDI BEL BFA BHR BOL BOS BRA BRN BUL BWA CAN CHE CHI CIV CMR COG COL CRI CYP DEU DNK DOM
DZA EGY ECU
EST ETH FRA GAB GBR GEO GHA GIN GRC GTM HND HRV HTI IDN IRL ISR ITA JAM JOR JPN KAZ KEN KHM KORKWT LBN LBR LBY LKA LTU LUX LVA MAR MDG MEX MKD MLI MLT MMR MNG MOZ MRT MUS MYS NAM NGA NIC NLD NOR NPL OMNNZL PAK PAN PER PHLPOL PRT PRY QAT ROU RUS
RWASEN SAU
SGP SLVSRB SVK SVN SWE SYR TCD THA TUNTUR TZA UGA UKR URY USA VEN VNM ZAM ZIM 0 1 2 3 4 T C 0 20 40 60 80 100 mas 0 1 2 3 4 T C
Figure 4: Relationship between SMEs' technical capacities and masculinity
30
Figure 5 shows clearly that countries with SMEs with excellent technical capacities (TC=4) are high on uncertainty avoidance. With USA having the lowest value of around 50, the average value of uncertainty avoidance of this group is remarkably high. Countries with very poor technical capacities for SMEs (TC=0), by contrast, all have uncertainty avoidance values in the medium range between 40 and 60. The other categories of TC from 1 to 3 are distributed quite balanced. However, it is noticeable that the average value of uncertainty avoidance increases with rising categories of TC.
Figure 6: Relationship between SMEs' technical capacities and long-term orientation
Figure 6 indicates that countries with SMEs with very poor technical capacities are rather short-term orientated. With the exception of Indonesia (indicated as IDN), all of these countries have long-term orientation values lower than 50. Half of the countries even have very low values below 20. The other categories of TC are distributed quite balanced regarding the long-term orientation of the countries. Similarly to Figure 5, it is noticeable that the average long-term orientation of each group of countries increases with rising categories of TC. AGO ALB ARE ARG ARM AUS AZE BAN BDI BEL BFA BHR BOL BOS BRA BRN BUL BWA CAN CHE CHI CIV CMR COG COL CRI CYP DEU DNK DOM DZA ECU EGY EST ETH FRA GAB GBR GEO GHA GIN GRC GTM HND HRV HTI IDN IRL ISR ITA JAM JOR JPN KAZ KEN KHM KOR KWT LBN LBR LBY LKA LTU LUX LVA MAR MDG MEX MKD MLI MLT MMR MNG MOZMRT MUS MYS NAM NGA NIC NLD NOR NPL NZL OMN PAK PAN PERPHLPRT POL PRY QAT ROU RUS
RWA SEN SAU
SGP SLV SRB SVK SVN SWE SYR TCD THA TUN TUR TZA UGA UKR URY USA VEN VNM ZAM ZIM 0 1 2 3 4 T C 0 20 40 60 80 100 lto
31
From the figures above one can derive some information for the further analysis. The categories 1 to 3 of TC show similar picture in all figures and are always evenly distributed regarding the cultural values. However, this is not always the case for the categories 0 and 4 of TC. One can observe “cultural clustering” for countries with the TC category 0 in all figures. This is especially the case in Figure 2 and 3, where the average values of power distance and individualism are extreme for these countries. Burundi, Indonesia, Sri Lanka, Mozambique, Mauritania, and Chad make up this group of countries with very poor technical capacities for SMEs. Furthermore, countries with excellent technical capacities for SMEs (TC=4) show similar “cultural clustering”. These countries include Germany, France, Israel, Japan, Slovenia, and USA. With a look to Figure 5 it becomes clear that these forerunners all score high on uncertainty avoidance.
3.3. Control Variables for the Economic Environment
Four control variables will be integrated. These variables take into account the economic environment of each country and its preconditions, which influence SMEs’ participation in Industry 4.0. The importance to add control variables to prevent under- or overestimations of cultural effects in economic surroundings is stressed by other authors (De Jong E. , 2009; Mihet R. , 2013). One of these variables is the information and communications technology development (ICT) of countries, as the standardization of ICT is a prerequisite for SMEs to digitize their processes (Kilangi, 2012). The ICT development index is obtained from the International Telecommunication Union (ITU), which monitors the global development of ICT over time (ITU, 2017). This index comprises the access, the use, and the skills considering ICT as well as the dynamic of countries in their development. The availability of venture capital (VC) is another decisive factor for SMEs’ ability to afford technological adjustments. As mentioned above, in section 2.6., the lack of financial resources is a main concern of many SMEs. The index concerning the availability of venture capital is part of the Global Competitive
32
Index from the World Economic Forum and is obtained from the World Bank (World Economic Forum, 2019). A similar concern of SMEs are inadequately educated workforce (ACW) and therefore a lack of employees with sufficient know-how to handle new technologies. The World Economic Forum developed an index to measure the amount of inadequately educated workforce in a country. This index is applied in this study and obtained from the World Bank (World Economic Forum, 2017). Furthermore, the governmental support for an implementation of Industry 4.0 can give SMEs a great leap forward (KPMG, 2018). For this reason, the variable public support in moving upmarket and acquiring technologies for SMEs (PS) is integrated. The index published by the French ministry of economy to measure public support for SMEs is obtained from the World Bank (Ministère de L'Economie et des Finances, 2016). The distribution of PS’ categories is shown in Table 3 below.
Table 3: Distribution of PS
PS Freq. Percent Cum.
Non 34 28.81 28.81 Poor 26 22.03 50.85 Fair 38 32.20 83.05 Good 15 12.71 95.76 Excellent 5 4.24 100.00 Total 118 100.00
Similar variables are used by the annual SME Competitiveness Outlook of the International Trade Center (2019) to reflect the national economic environment of SMEs, which emphasizes their relevance. The italic written names of the variables comply with the official definitions of these indices. ICT, VC, and ACW are continuous variables, whereas PS is a categorical variable. This means that it contains different categories which range from 0 to 4. Thus, 0 indicates no public support, 1 indicates poor public support, 2 indicates fair public support, 3 indicates good public support, and 4 excellent public support.
33
Table 4 below shows the correlations between the dependent variable TC, the cultural variables, and the control variables.
Table 4: Correlation matrix of all variables
TC pdi idv mas uai lto ICT VC ACW PS
TC 1 pdi -0.296** 1 idv 0.495*** -0.691*** 1 mas 0.150 0.0631 0.097 1 uai 0.330*** 0.185* -0.147 -0.010 1 lto 0.423*** -0.134 0.324*** 0.025 0.096 1 ICT 0.675*** -0.382*** 0.591*** 0.044 0.239** 0.454*** 1 VC 0.366*** -0.196* 0.331*** 0.007 -0.140 0.227* 0.570*** 1 ACW -0.009 -0.044 0.097 -0.096 0.018 0.082 0.169 0.261** 1 PS 0.682*** -0.391*** 0.444*** 0.020 0.081 0.413*** 0.561*** 0.357*** 0.067 1 * p < 0.05, ** p < 0.01, *** p < 0.001
5. Research Sample and Methodology 5.1. Research sample
The final sample consists of 118 observations capturing both developed and developing countries. The sample decreased during the data collection for two reasons. First, the included variables did not contain information for all countries, which excluded several countries. Second, there were outliers in the sample that had to be eliminated from the model, as they had a too strong influence on the model and distorted the results. These extreme observations2 were
identified through Cook’s Distance and a leverage-versus-residual-squared plot. Cook’s distance is considered as a very useful tool to find outliers that have a too strong influence on
2 Extreme cases that were deleted from the model are: Iceland, Finland, Austria, Czech Republic, Iran, Moldova,
Hungary, Chile, India, South Africa, Bahrain, and Spain. When examining the data, it becomes clear that some of the outliers have similar values for uncertainty avoidance and TC. Chile (86 for uai and 0 for TC) and Hungary (82 for uai and 1 for TC), for example, are both high on uncertainty avoidance and low on TC, which does not comply with the relationship between uncertainty avoidance and TC indicated in Figure 5. However, to ensure that the outliers do not still contain relevant information, a regression with the addition of the outliers was run. The results of this regression are listed in Appendix 2. The results do not differ significantly from the results without outliers, especially regarding the coefficient of uncertainty avoidance. Thus, the exclusion of outliers did not distort the results.
34
the model (Cook, 2011). Observations with critical values, which are defined as values greater than four divided by the number of observations, are advisable to eliminate from the model. According to the Stata manual, a leverage-versus-residual-squared plot is “One of the most useful diagnostic graphs…” to examine outliers (Stata, 2020, p. 17). Thus, such a plot is used to check its distribution on similarities with the statistics of the Cook’s Distance test. Finally, individual cases that show a too strong and influential effect in both analytical tools are eliminated from the model. The summary statistics in Table 5 below show the distribution of the ultimately integrated variables.
Table 5: Summary statistics of the entire sample
count mean sd min max
TC 118 1.889 1.010 0 4 pdi 118 66.779 19.805 13 100 idv 118 35.508 20.844 6 91 mas 118 47.466 16.739 5 100 uai 118 64.466 21.711 4 100 lto 118 38.542 23.289 0 100 ICT 118 5.309 2.210 1.27 8.85 VC 118 2.969 .792 1.47 5.24 ACW 118 7.391 4.363 0 23.5 PS 118 1.415 1.157 0 4 N 118 5.2. Research Methodology
To examine whether there is an effect of the cultural variables on the dependent variable TC, multiple regressions will be carried out. As the dependent variable TC consists of five ordered ranks from 0 to 4, the estimations will be carried out by ordered logistic regressions. An ordered logistic model is the appropriate method for an ordinal dependent variable (Grilli & Rampichini, 2014). The regression analysis will be divided in two parts. The first regression will only be performed with the integration of the five cultural variables. The resulting coefficients, indicating the effect of all five cultural variables on TC, will be inspected. Then next, a second regression will be carried out with the cultural variables and the addition of the control variables for the economic environment. The variable of PS is integrated to the model
35
as a factor variable to do justice to the categorized information in it. The stepwise addition of control variables is useful to examine in how far the effect of the cultural main variables on TC change. It is imaginable that, for example, a strong initial effect of a cultural variable on TC will reduce with the addition of the control variables, as they take on some of the effects that were previously accounted to the cultural variable.
The examination of the results of this second regression will be differentiated in two effects. First, the direct effects of the cultural variables on TC will be inspected once more, to identify if there are changes after the addition of the control variables. Then, the moderating effects of the five cultural variables on TC through the control variables will be examined. It is expected that Hofstede’s dimensions do not only have an effect on TC, but also on VC, for example. Thus, the question would be to what extent VC is increased/decreased by the cultural variables, and how far this increases/decreases TC. From these findings, final conclusions will be derived whether there is generally a cultural impact on SMEs’ participation in Industry 4.0. Furthermore, the proposed hypotheses will be tested and either verified or refuted. Moreover, the sample will be divided to two homogenous groups with similar characteristics and analyzed. The regression results of both groups will be examined on similarities and deviations from the main regression of the entire sample. In this way similarity structures between countries with related features can be identified. An example for such a distinction is the clustering of western and non-western countries (De Jong E. , 2009). Based on this, more specific conclusions will be drawn on the cultural influence for advanced countries and developing countries. In addition, this analysis in groups can serve as a first robustness check. Hereafter, a second robustness check will be performed without the integration of countries scoring very high or very low on TC, as Figure 5 indicated that these countries could be very influential. In this way, one can detect whether the original results were due to the composition of the sample and change after an adjustment of it.
36
6. Main Results and Discussion
National culture certainly has an effect on SMEs’ participation in Industry 4.0. This can be seen below in Table 6, which shows the results of the empirical main estimation. Column (1) presents the estimated coefficients for the sole integration of the cultural variables. Columns (2) to (5) present the estimated coefficients of the cultural variables after the stepwise addition of the control variables. Column (6) reports the regression results after the stepwise elimination of all insignificant variables. The different results of the six models will be summarized briefly and their differences highlighted. Based on that, the direct effects of national culture on SMEs’ technical capacities will be discussed.
Column (1) indicates a strong influence of national culture on SMEs’ participation in Industry 4.0 if no further control variables are integrated. Individualism, uncertainty avoidance, and long-term orientation all have a highly significant positive effect on the technical capacities of SMEs. This is surprising, as uncertainty avoidance was expected to have a negative influence on TC. Following these first results, SMEs in countries with high uncertainty avoidance, individualism, and long-term orientation have better technical capacities.
The picture changes in column (2) with the addition of the variable ICT. Uncertainty avoidance and long-term orientation remain significant, whereby their coefficient decreases. Moreover, the coefficient of uncertainty avoidance remains positive. Furthermore, the effect of masculinity increases and gets significant. The control variable of ICT has a positive effect and is also highly significant. Different to the results of column (1), the results indicate that SMEs in countries with high uncertainty avoidance, masculinity, and long-term orientation have better technical capacities. Additionally, a good ICT infrastructure increases the technical capacities of SMEs, which is not surprisingly. The addition of the variable VC in column (3) strengthens the prior impressions. National culture’s effect on SMEs’ technical capacities even rises when
37
taking venture capital availability into account. Individualism becomes significant again and has a positive effect. Concerning the control variables, ICT remains significant.
Table 6: Main estimation of national culture’s effect on SMEs’ participation in Industry 4.0
(1) (2) (3) (4) (5) (6)
Effect of cultural variables
Stepwise addition of control variables for economic
environment insignificant Without variables VARIABLES TC TC TC TC TC TC pdi -0.002 -0.002 -0.005 -0.004 0.019 (0.016) (0.015) (0.014) (0.014) (0.014) idv 0.056*** 0.028 0.031* 0.032* 0.054*** 0.040*** (0.016) (0.018) (0.018) (0.019) (0.018) (0.013) mas 0.019 0.022** 0.023** 0.021** 0.021* 0.025** (0.012) (0.010) (0.010) (0.010) (0.011) (0.011) uai 0.045*** 0.033*** 0.039*** 0.041*** 0.054*** 0.053*** (0.011) (0.009) (0.009) (0.010) (0.010) (0.010) lto 0.024*** 0.017* 0.018* 0.018** 0.007 (0.009) (0.009) (0.009) (0.009) (0.009) ICT 0.537*** 0.399** 0.401** 0.329** 0.387*** (0.135) (0.156) (0.159) (0.162) (0.132) VC 0.498 0.649* 0.081 (0.340) (0.352) (0.499) ACW -0.089** -0.089* -0.080* (0.045) (0.046) (0.046) 1.PS 2.116*** 2.168*** (0.703) (0.686) 2.PS 2.091*** 2.144*** (0.773) (0.772) 3.PS 4.004*** 3.915*** (1.252) (1.122) 4.PS 9.607*** 9.433*** (1.525) (1.296) Observations 118 118 118 118 118 118
a Ordered logit model with TC as the dependent variable. Robust standard errors in parentheses *** p<0.01, ** p<0.05, *
p<0.1. Cut-points are not shown in this table as they are not that meaningful for the objective of this section and due to lack of space. The variable PS is subdivided in its categories „poor“ (1.PS), „fair“ (2.PS), „good“ (3.PS), and „excellent“ (4.PS).
Column (4) reports that all but one cultural variable have a significant effect after the addition of the variable ACW. Namely, individualism, masculinity, uncertainty avoidance, and long-term orientation have a positive influence on the technical capacities of SMEs. However, uncertainty avoidance captures the most significant and strongest effect. The three control variables ICT, VC, and ACW all significantly affect the technical capacities of SMEs. In contrast
