The Effect of National Culture and Formal
Institutions on Innovation
Student: Danielle Henstra
Student number: S2586797
Email: e.d.henstra@student.rug.nl
Supervisor: prof. dr. S. Beugelsdijk
Co-assessor: dr. I. Maris-de Bresser
Course: Master Thesis IB&M (2
ndsemester)
Final Version Master Thesis
June, 12
th2015
3
The Effect of National Culture and Formal
Institutions on Innovation
Danielle Henstra
Faculty of Economics and Business | University of Groningen | Groningen, The Netherlands
Abstract
National culture and formal institutions are considered essential for innovation. However, previous research considered both forces separately, little research has attempted to include both forces simultaneously with innovation. Therefore, this paper examines the effect of national culture and formal institutions on innovation. This study also examines whether the effect of formal institutions on innovation is dependent on the cultural setting. Furthermore, previous research has often used Hofstede’s cultural dimensions to examine the relationship between national culture and innovation. This study uses Schwartz’s cultural framework as his framework is argued to capture more aspects of a culture (Imm Ng, Lee and Soutar, 2007) and considered ‘’superior’’ (Brett and Okumura, 1998, p. 500; Imm Ng, Lee and Soutar, 2007, p. 164) to Hofstede’s cultural framework. Furthermore, Schwartz suggested that his cultural framework also includes Hofstede’s dimensions. The article uses three main theories: theory of national innovation systems, value-belief-norm theory and the institutional theory. This study analyzes a sample of national culture, formal institutions and innovation from 30 countries between 2000 – 2010. The results are obtained by the statistical method least squares multiple regression analysis. The study found that Schwartz’s cultural dimensions do not have a significant direct effect on innovation. Whereas formal institutions – rule of law and intellectual property protection – correlate significant and positively with innovation. Finally, the results indicate that Schwartz’s cultural dimensions do not significantly moderate the relationship between formal institutions and innovation.
4
Table of Contents
1. Introduction ... 5
2. Literature ... 7
2.1. National innovation systems ... 7
2.2. Value-belief-norm theory ... 9
2.3. Institutional theory ... 10
3. Hypotheses ... 12
3.1. Culture and innovation ... 12
3.2. Formal institutions and innovation ... 17
3.3. Culture, formal institutions and innovation ... 19
3.4. Conceptual model ... 21 4. Methodology ... 21 4.1. Sample ... 21 4.2. Dependent variable ... 22 4.3. Independent variable ... 25 4.3.1. National culture ... 25 4.3.2. Formal institutions ... 25 4.4. Control variables ... 26 4.5. Data analysis ... 28
5. Data & Results ... 28
5.1. Descriptive statistics ... 29
5.2. Correlation matrix ... 29
5.3. Culture and innovation ... 31
5.4. Formal institutions and innovation ... 33
5.5. Culture and formal institutions on innovation ... 35
5.6. Interaction culture and formal institution on innovation ... 35
6. Discussion & Conclusion ... 37
6.1. Limitations ... 38
6.2. Future research ... 39
References ... 40
Appendix A: Original data ... 46
5
1. Introduction
The concept of innovation has been examined from different perspectives, such as R&D, knowledge and management, diffusion, and technological capabilities (Kumar, 2014). Kumar (2014) argues that innovation is the foundation of growth, and prepares firms to react to environment changes. Innovation is manifested in many different aspects of organizations, such as technology, products, and management processes. Innovation is key for firms when developing and maintaining a competitive advantage. Innovation helps firms to build stronger brands, successful products and enhancing firm value (Kumar, 2014). Furthermore, innovation is also important for a firm’s strategic course of action. Jones and Davis (2000) argue that as multinational enterprises (MNEs) internationalize their R&D facilities become vital to the success of the firm. In this study the term innovation is interpreted rather broadly, as ‘’the multi-stage process whereby organizations transform ideas into new/improved products, service or processes, in order to advance, compete and differentiate themselves successfully in their marketplace’’ (Baregheh, Rowley and Sambrook, 2009, p. 1334). This definition is applied as Baregheh and his colleagues collected 60 definitions of innovation from various disciplinary literatures, and created one common clarified definition of innovation that can be applied across multiple disciplines.
In addition to the importance of innovation for firms, governments are also more aware of the importance of innovation. Westhead, Wright and McElwee (2011) argue that governments have focused on creating a favorable environment for innovative activity. Innovation is a prominent item on the agenda of both governments and firms. Therefore, it is relevant to comprehend the factors that drive innovation at the national level. Previous research has shown that innovation is associated with national income levels, human capital, industry structure, trade openness, and military expenses (Shane, 1992 and 1993; Taylor and Wilson, 2012; Barro and Lee, 2010, Yanikkaya, 2003, David, 2007, McNeill, 1982; Smith, 1985; Ruttan, 2006). Besides factors that foster innovation in a structural way, national culture should also be considered when looking at the degree of innovation of a country. Shane (1993) suggests that nations differ in their degree of innovation as they have different cultural values and therefore different national cultures.
6 activities towards the rules and legislation of this environment. Innovation involves firms to undergo risky behavior and uncertainty, so that these firms are often exposed to expropriation (Lumpkin and Dess, 1996). Governments can facilitate innovation by creating a favorable environment for innovation, for example through means of intellectual property protection.
National culture and formal institutions are considered essential for innovation. However, previous research considered both forces separately. Little research has attempted to include both forces and see the effect on innovation. This study aims to examine the effect of both forces – national culture and formal institutions – on innovation. This study is based on three theories – theory of national innovation systems, value-belief-norm theory and the institutional theory – which highlight that culture and formal institutions, can be considered sources of variation in innovation performance between nations. Furthermore, previous literature argues that culture is embedded in institutional routines and in formal institutions of a country (Hofstede, 1991; North, 1990; Yong and Zahra, 2012; Fukuyama, 1995). Therefore, the study also aims to explain that the influence of formal institutions on innovation is dependent on the cultural setting.
While previous research on both forces separately produced interesting results, this study contributes through several improvements. First, this study aims to examine the influence of both forces on innovation in the same research. Second, studies aiming to examine the effect of national culture on national innovation often use Hofstede’s cultural dimensions. Therefore, this study aims to measure culture using Schwartz’s cultural framework (Schwartz, 1992 and 1994). Schwartz’s cultural values are argued to capture more aspects of a culture (Imm Ng, Lee and Soutar, 2007) and considered ‘’superior’’ (Brett and Okumura, 1998, p. 500; Imm Ng, Lee and Soutar, 2007, p. 164) to Hofstede’s cultural framework. Furthermore, Schwartz (1994) suggested that his cultural value frameworks also includes Hofstede’s dimensions.
7
2. Literature
The theory of national innovation system indicates that culture and formal institutions can be considered sources of variation in innovation performance between nations. Furthermore, the value-belief-norm theory and the institutional theory are applied as part of the theoretical framework, where the value-belief-norm theory highlights the relevance of culture for innovation and the institutional theory focuses on the effect of formal institutions on innovation.
2.1. National innovation systems
According to the theory of national innovation systems (NIS) a nation’s culture and players in the institutional environment can be considered sources of differences and variation in innovation performance between nations (Brown and Ulijn, 2004; Huber, 2004; Mahlich and Pascha, 2011; Earl and Gault, 2006; Lundvall, 2010; Nelson, 1993; OECD, 1997 and 1999). Based on these studies it is considered relevant to examine the effect of national culture and formal institutions on innovation. Below the origin and concept of the NIS will be examined.
According to Nelson (1993) the decrease in growth in the 1970s, the rise of Japan as an economic and technological power, the decline of the United States, and the concern in the European Union of being behind, have led to the importance of supporting innovation by policy makers and creating a favorable environment to spur innovation. Furthermore, nations such as Korea and Taiwan, have enhanced their technical capabilities. This has broadened the range of nations with competitive players (Nelson, 1993). Nelson (1993) argues that such developments have led to the rise in current interest in NIS and their ‘’similarities and differences, and in the extent and manner that these differences explain variation in national economic performance and innovativeness’’ (Nelson, 1993, p. 3).
8 initiate, import, modify and diffuse technologies’’ (p. 1). Lundvall (1992) defined NIS as ‘’the elements and relationships that interact in the production, diffusion and use of new, economically useful knowledge and are either located within or rooted inside the borders of a nation state’’ (p. 12).
The concept became accepted by academics and politics through the work of Nelson (1993) as previous studies did not perform a comparative analysis across countries. In his book ‘’National Innovation Systems: A comparative analysis’’ Nelson (1993) compared the US system to 15 other NISs and was the first to perform a comparative analysis. Nelson (1993) argues that a NIS comprises of ‘’a set of institutions whose interactions determine the innovative performance of national firms’’ (Nelson, 1993, p. 4). Nelson (1993) identifies that the creation and diffusion of innovation is affected by the social and political actors inside a nation, such as education, universities, banks, state institutions and intellectual property protection etc. All of these actors are interacting and determine a country’s innovative performance (Nelson, 1993).
The NIS concept has been accepted rapidly among scholars and policy makers (Brown and Ulijn, 2004). The Organization for Economic Co-operation and Development (OECD) has absorbed the concept of NIS as an integral part of their analytical perspective (OECD, 1997 and 1999). The OECD adopts the definition developed by Metcalfe (1995) as this definition comprises the different components by Nelson, Freeman, Lundvall and others. Metcalfe (1995, p. 462) and the OECD (1997, p.10 and 1999, p. 24) define NIS as…
‘’that set of distinct institutions which jointly and individually contribute to the
development and diffusion of new technologies and which provides the framework within which governments form and implement policies to influence the innovation process. As such it is a system of interconnected institutions to create, store and transfer the knowledge, skills and artefacts which define new technologies.” (Metcalfe, 1995, p. 462; OECD, 1997, p. 10
and 1999, p.24).
Based on this definition the innovativeness of a nation depends not only on how individual institution, such as organizations, universities, governments and so on, perform in isolation, but on ‘’how they interact with each other as elements of a collective system of
knowledge and use, and on their interplay with social institutions, such as values norms and
9 in the institutional environment, such as the policy makers and government, but as well by social institutions, referring to culture. The OECD (1997 and 1999) is not the first to argue that factors relating to institutions and culture can be considered sources of differences and variation in innovation performance between nations (Brown and Ulijn, 2004; Huber, 2004; Mahlich and Pascha, 2011; Earl and Gault, 2006; Lundvall, 2010; Nelson, 1993).
In the previous section, the NIS theory indicates that a nation’s culture and players in the
institutional environment can be considered sources of differences and variation in innovation
performance between nations. In the following sections, the value-belief-norm theory (VBN) and the institutional theory are applied as part of the theoretical framework. The VBN theory further highlights the relevance of culture for innovation. Next to that, the institutional theory will add to the VBN theory by examining the effect of formal institutions on innovation, and also examines the interaction effect of formal and informal institutions.
2.2. Value-belief-norm theory
The VBN theory originated with Stern (2000), he incorporated the value theory and normactivation theory by Schwartz and the new ecological paradigm by Dunlap in the VBN theory (Jansson, Marrel and Nordlund, 2011). The VBN theory indicates that ‘’values and beliefs held by members of cultures influence the degree to which behaviors of individuals, groups and institutions within are enacted, and the degree to which they are viewed as legitimate, acceptable, and effective’’ (House, Hanges, Javidan, Dorfman and Gupta, 2004, p. 17). It is therefore argued that societies and individuals behave according to dominant values, beliefs and norms present in their societies. Societies are characterized by different dominant values, beliefs and norms and therefore resulting in differences between societies what is considered legitimate, acceptable and effective. These dominant values, beliefs and norms constitute the national culture of a society.
10 these pressures influence behavior and decision making of individuals and organizations. This reflects that culture affects the way people in a society behave, which sets boundaries to innovation. As Earl and Gault (2006) also indicate, culture and values provide boundaries to the innovation system.
Innovation scholars across several disciplines have argued that national culture affects innovation by influencing the ‘’preferences, expectations, and incentives of individuals across a society’’ (Taylor and Wilson, 2012, p. 235). Therefore, the differences in national culture are considered a source of differences and variation in innovation performance between nations.
2.3. Institutional theory
In recent years the institutional theory has gained considerable attention with institutional theorists, such as Scott and North, and has been applied to various international management fields (Mahlich and Pascha, 2011). The institutional theory highlights ‘’the role of institutions for individuals and organizational behavior and decision-making’’ (Mahlich and Pascha, 2011, p. 160). North (1991) defines institutions as ‘’humanly devised constraints that structure political, economic and social interaction, and consist both of informal constraints (sanctions, taboos, customs, traditions, and code of conduct), and formal rules (constitutions, laws, property rights)’’ (p. 97). The foundation of the theory is that individuals and organizations are embedded within an institutional environment, which will eventually guide their behavior and decision-making. Furthermore, the theory does also include the social construction of organizational behavior and recognizes the limits imposed by social constraints (Mahlich and Pascha, 2011). Societal expectations influence innovation as they impact the behavior and decision-making of individuals and organizations, which are considered drivers of innovation (Mahlich and Pascha, 2011). In line with previous studies, the institutional theory argues that institutions influence innovation as they set the ‘’rules of the game’’ and therefore influence the environment in which innovations takes place (Mahlich and Pascha, 2011).
11
pillar as ‘’existing laws and rules in a particular national environment that promote certain
types of behaviors and restrict others’’. The normative pillar consists of ‘’social norms, values, beliefs and assumptions that are socially shared and carried out by individuals’’ (Kostova, 1997, p. 180), it includes the social expectations and practices. The last pillar, the
cognitive pillar reflects the cognitive structures and symbolic systems shared among
individuals (shared knowledge) (Kostova, 1997), and includes the subjective constructed rules and implicit meanings which can influence the beliefs and actions of the society (Scott, 2014).
In a similar manner to Scott, North (1990) classifies the institutional environment as the interaction of formal and informal constraints, whereby the formal constraints contain the regulative pillar and the informal constraints contain the normative and cognitive pillars (Mahlich and Pascha, 2011). The formal constraints (e.g. regulative pillar) refer to formal institutions, where the informal constraints refer to culture (e.g. normative and cognitive pillars). The definition by North (1990) includes both formal and informal institutions, thus ‘’linking the written rules with social norms and other constraints imposed by the social value systems’’ (Leković, 2011, p. 359). This indicates that a clear distinction should be made between on the one hand formal institutions, and on the other hand culture. The classification of the institutional environment by North (1990) will be used as his work makes a clear distinction between formal and informal institutions, which is also the focus of this research.
12 The normative and cognitive pillar represent a nation’s culture (Mahlich and Pascha, 2011). Culture is embedded in institutional routines and in formal institutions of a country. Hofstede (1991) argues that formal institutions are based on culture, which results in differences between formal institutions across countries. Therefore, this study argues that the influence of formal institutions on national innovation depends on cultural setting.
The classification of the institutional environment by North (1990) will be used in this study as his work makes a clear distinction between formal and informal institutions, which is also the focus of this research. In the following sections the influence of national culture and formal institutions on innovation will be examined. Based on previous literature, assumptions will be made about the influence of national culture and formal institutions on innovation. Furthermore, the study will also aim to indicate interaction effects between national culture and formal institutions.
3. Hypotheses
Section two in this study indicated the relevance of national culture and formal institutions for innovation. In this section, the influence of national culture and formal institutions will be discussed, and hypotheses will be formulated linking national culture and formal institutions to innovation. First, the factors and characteristics of a society, which lead to higher innovation are identified, linking these to Schwartz’s cultural dimensions. Second, the relationship between formal institutions (rule of law and intellectual property protection) and innovation will be examined. Lastly, the interaction between culture and formal institutions on innovation will be reviewed.
3.1. Culture and innovation
13 integrity problems, national level data is generalized to individual behavior, and the ‘’correspondence between the items used to measure cultural dimensions and the conceptual definition of these dimensions is tenuous’’ (Steenkamp, 2001, p. 32).
This study aims to examine the relationship between culture and innovation using the cultural value framework by Schwartz (1992 and 1994). Schwartz first derived theoretically an exhaustive set of 56 values recognized across cultures (Schwartz, 1992 and 1994; Imm Ng, Lee and Soutar, 2007; Steenkamp, 2001). Schwartz defined human values as ‘’desirable goals, varying in importance that serve as guiding principle in people’s lives’’ (Schwartz, 1994, p. 88). After Schwartz identified the set of 56 values, he examined which of these values had an equivalent meaning across countries (Steenkamp, 2001; Imm Ng, Lee and Soutar, 2007), resulting in 45 useful values. Consequently, Schwartz’s used his Schwartz Value Survey (SVS) to survey schoolteachers and college students from 67 countries from 1988 and onward. Respondents were asked ‘’to rate the importance of each value served as a guiding principle in their lives’’ (Imm Ng, Lee and Soutar, 2007, p. 168). He took the mean scores of each of the 45 values, and used ‘’smallest-space analysis to identify a number of meaningful and interpretable dimensions along which national cultures differ’’ (Drogendijk and Slangen, 2006, p. 364). This resulted in seven dimensions along which cultures could be analyzed, namely: intellectual autonomy, affective autonomy, embeddedness, hierarchy, egalitarian commitment, mastery and harmony.
Schwartz’s value framework is chosen to analyze culture as Schwartz’s values may have the potential to explain greater cultural variation than Hofstede’s dimension, as they seem to capture more aspects of culture (Imm Ng, Lee and Soutar, 2007). Furthermore, Schwartz suggested that his cultural value frameworks also includes Hofstede’s dimensions (Schwartz, 1994), as it does not exclude prior research on national cultures but ‘’instead tends to build on it’’ (Dostal, 2004, p. 20). Imm Ng, Lee, Soutar (2007) and Steenkamp (2001) argue that Schwartz’s value framework offers several advantages compared to Hofstede’s dimensions.
14 comprehensive than Hofstede’s value dimensions (Steenkamp, 2001; Imm Ng, Lee and Soutar, 2007). Fourth, Schwartz’s samples comprise of more diverse regions, including socialist countries (Steenkamp, 2001; Imm Ng, Lee and Soutar, 2007). Finally, Schwartz’s value framework only included values that have the same meaning across societies (Steenkamp, 2001; Imm Ng, Lee and Soutar, 2007), while for Hofstede it remains questionable whether the values included have the same meaning in different countries (Steenkamp, 2001). Based on the arguments above, Schwartz’s cultural framework is considered ‘’superior’’ (Brett and Okumura1, 1998, p. 500; Imm Ng, Lee and Soutar, 2007, p. 164) compared to Hofstede’s cultural framework, therefore this study applies Schwartz’s work on culture.
In the following sections the relationship between national culture and innovation will be examined using Schwartz’s cultural framework.
National culture and innovation: Schwartz’s cultural framework
Previous research has attempted to determine the relationship between national culture and innovation (Shane, 1992 and 1993; Hofstede, 1980; Taylor and Wilson, 2012; Jones and Davis, 2000; Everdingen and Waarts, 2003) and has looked at which societies are more likely to innovate than others. This section indicates which factors and characteristics of a society are likely to be associated with higher innovation. The identified factors and characteristics will be linked to Schwartz’s cultural dimensions and hypotheses will be formulated indicating the relationship between national culture and innovation. In total Schwartz identified seven cultural dimensions – intellectual autonomy, affective autonomy, embeddedness, hierarchy, egalitarian commitment, mastery and harmony. In this study only the cultural dimensions which can be linked to the identified factors and characteristics will be used.
Previous research has identified that societies with higher innovation are predominantly associated with concepts of freedom, autonomy, and independence (Jones and Davis, 2000; Shane, 1992 and 1993; Taylor and Wilson, 2012; Everdingen and Waarts, 2003), are
individualistic in nature (Shane, 1992 and 1993; Hofstede, 1980; Taylor and Wilson, 2012),
are characterized by diversity, tolerance, open-mindedness, creativity and non-conformity
15 (Taylor and Wilson, 2012), have an outward-looking view (Shane, 1992 and 1993; Hofstede, 1980) and information circulates across different levels of society and stresses the importance of social networks (Shane, 1992 and 1993; Hofstede, 1980) and finally societies associated with acceptance of personal wealth and rewards through recognition and financial remuneration (Shane, 1992 and 1993).
The factors and characteristics identified above can be linked to Schwartz’s cultural dimension autonomy, as concepts such as freedom, independence, diversity, creativity, and open-mindedness are central in autonomy societies. There are two types of autonomy:
intellectual autonomy and affective autonomy. Intellectual autonomy emphasizes individuals
as independent and autonomous beings, who are allowed to pursue their own intellectual ideas and thoughts (Schwartz, 1999). Intellectual autonomy societies value curiosity, broadmindedness, creativity and freedom of action and thought (Schwartz, 2006 and 2009). Affective autonomy emphasizes individuals as independent and autonomous beings, who are allowed to pursue their own stimulation and hedonism feelings, interests and desires (Schwartz, 1999). Affective autonomy societies value pleasure, self-indulgence, and an exciting, enjoying, and varied life. Both intellectual and affective autonomy societies accept personal wealth and rewards generated by actively pursuing positive experiences (Schwartz, 2006 and 2009). Based on the factors and characteristics identified above, it is assumed that intellectual autonomy and affective autonomy societies will foster higher innovation, which leads to the following hypotheses:
Hypothesis 1a: Intellectual autonomy societies will be more innovative than less intellectual
autonomy societies.
Hypothesis 1b: Affective autonomy societies will be more innovative than less affective
autonomy societies.
Furthermore, previous research has also identified that societies with higher innovation are characterized by minimized hierarchy/non-hierarchical structures, equality of prestige,
rewards and social power (Shane 1992 and 1993; Everdingen and Waarts, 2003; Jones and
16
narrow work roles (Shane 1992 and 1993; Everdingen and Waarts, 2003; Jones and Davis,
2000; Taylor and Wilson, 2012) and finally societies with higher innovation accept change in
the distribution of power (Shane 1992 and 1993; Jones and Davis, 2000; Taylor and Wilson,
2012).
These factors and characteristics identified above can be linked to Schwartz’s cultural dimension hierarchy, as less hierarchical societies are characterized by equality of rewards and social power, decentralized decision-making, free communication and acceptance of redistribution of power. A hierarchical society places emphasis on the legitimacy of unequal distribution of roles, resource allocation and power (Schwartz, 1999). Hierarchical societies value social power, wealth, authority, humility and influence over people and events (Schwartz, 2006 and 2009). It is assumed that societies that are characterized by higher levels of hierarchy will foster lower levels of innovation, which leads to the following hypothesis:
Hypothesis 2: Less hierarchical societies will be more innovative than more hierarchical
societies.
Thirdly, societies with higher innovation are also associated with emphasis on rewards,
recognition of performance and training and improvement of the individual (Shane, 1993;
Everdingen and Waarts, 2003), and place a greater emphasis on the task, levels of
achievement, as well as acceptance of some degree of conflict and competition (Jones and
Davis, 2003). These factors and characteristics can be linked to Schwartz’s cultural dimension
mastery, as mastery societies are characterized by concepts such as performance, achievement
and improvement. A mastery society emphasizes on getting ahead through active mastery of self-assertion and individuals’ rights to get ahead of other people (Schwartz, 1999). These societies value ambition, daring, competence, success, influence, choosing own goals, social recognition in the sense of respect and approval by others (Schwartz, 2006 and 2009). It is assumed that mastery societies will foster higher innovation, which leads to the following hypothesis:
Hypothesis 3: Societies that score higher on mastery will be more innovative than societies
that score lower on mastery.
Lastly, societies characterized by uncertainty acceptance, tolerance for change and
ambiguity, and which embrace the risks associated with an uncertain future, are associated
17 Furthermore, societies characterized by acceptance of competition and colleague dissent and where formalized rules are rejected, are also associated with higher innovation (Jones and Davis, 2000). These factors and characteristics can be linked to Schwartz’s cultural dimension
harmony, as less harmony societies accept more the fact of changing the world through
self-assertion and exploiting of people and resources (Schwartz, 2006). Harmony emphasizes ‘’fitting into the world as it is, trying to understand and appreciate rather than to change, direct or to exploit’’ (Schwartz, 2006, p. 23). A harmony society values the protection of the environment, a world at peace, unity with nature, and a world of beauty (Schwartz, 2006 and 2009). It is assumed that less harmony societies will foster higher innovation, resulting in the following hypothesis:
Hypothesis 4: Societies that score lower on harmony will be more innovative than societies
that score higher on harmony.
3.2. Formal institutions and innovation
18 Innovation is associated with the development of new ideas by firms, whereby supporting new ideas and creativity are considered essential. Rules and regulation created by governments can mitigate for uncertainty for individuals and organizations, stable environments are likely to foster innovation (Mahlich and Pascha, 2011). As firms take steps to innovate they are faced with risk and uncertainty, as innovation requires commitment of knowledge, assets and resources (Baird and Thomas, 1985). Mahlich and Pascha (2011) argue that within the regulatory environment especially the protection of intellectual property rights are considered important.
The OECD (1997) indicates that intellectual property rights are ‘’granted by state authority for certain products of intellectual effort and ingenuity’’ (p. 5), and that ‘’these rights are the subject of specific laws (statues) enacted by parliaments or state authority’’ (p. 5). Intellectual property rights can be assigned through patents, copyrights and trademarks (OECD, 1997). The International Chamber of Commerce (ICC) indicates that intellectual property rights have two purposes. On the one hand, intellectual property rights enable people and firms to benefit from their innovations and creative work, and on the other hand they also prevent others from copying or unfairly gaining from the innovation (ICC, 2005).
For firms with innovative activity intellectual property protection is essential as these firms are faced with risks and venture into the unknown. Moreover, innovation is often associated with intangible knowledge, which needs to be safeguarded. If the environment of firms fails to protect against expropriation or intellectual property breaches, these firms will be less likely to pursue risky behavior or accept uncertainty. These conditions, of risk taking and accepting uncertainty, are required for innovation. In environments where the intellectual property protection is low, innovation can be easily imitated (Oxley, 1999), whereby the firm is likely to lose its invested resources, knowledge and assets (Baird and Thomas, 1985). Insufficient protection of intellectual property rights weakens incentives for an individual or organization to invest in innovations (Mahlich and Pascha, 2011). In short, intellectual property rights safeguards innovations, and at the same time provide incentives for firms to produce new inventions and creations, leading to higher innovation (ICC, 2005). Therefore, it is assumed that an environment characterized by good intellectual property protection will have higher innovation, resulting in the following hypothesis:
Hypothesis 5: Societies with higher levels of intellectual property protection will be more
19 Furthermore, intellectual property protection also requires adequate specification and enforcement of rules and legislation, or in other words also referred to rule of law. The rule of law captures ‘’perceptions of the extent to which agents have confidence and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence’’ (Kaufmann, Kraay and Zoido-Lobaton, 1999, p. 3). The rule of law allows for monitoring, enforcement and execution of formal policies and regulations by governments. Lack of monitoring, enforcement and execution of policies related to intellectual property protection, might result in leakage of valuable intellectual property for firms. Policies aiming to protect intellectual property should be supported by strong and high quality rule of law (Rigobon and Rodrik, 2005), in order for governments to be able to create a favorable environment that can facilitate high innovation activity. Therefore, it is assumed that societies associated with high levels of rule of law will have higher innovation, leading to the following hypothesis:
Hypothesis 6: Societies with higher levels of rule of law will be more innovative than
societies with lower levels of rule of law.
3.3. Culture, formal institutions and innovation
20 societies with different cultural values can produce different economic outcomes’’ (Yong and Zahra, 2012, p. 97).
As argued by Hofstede (1991), North (1990), Yong and Zahra (2012) and Fukuyama (1995) ‘’formal institutions are embedded in different cultural settings and societies have their unique cultural values that are relatively stable over time’’ (Yong and Zahra, 2012, p. 98). Therefore, it is expected that the effect of formal institution on innovation to vary across countries depending on the cultural values prevalent in each society. Recent research by Gorodnichenko and Roland (2010) suggested that formal institutions are in part determined by a nation’s culture. In their study they argue that cultures which are individualistic in nature exert a positive significant effect on formal institutions, and that individualistic cultures moderate the effects of bad formal institutions on innovation (Gorodnichenko and Roland, 2010). Gorodnichenko and Roland (2010) argue that individualistic cultures value personal freedom, achievement, and the social status associated with personal accomplishments, which can be linked to Schwartz’s cultural autonomy dimension. Therefore:
Hypothesis 7: Autonomy (affective- + intellectual autonomy) will positively influence the
effect of formal institutions on innovation.
Hypothesis 7a: Autonomy will positively influence the effect of intellectual property
protection on innovation
Hypothesis 7b: Autonomy will positively influence the effect of rule of law on innovation.
Figure 1 depicts the interaction between autonomy and formal institutions on innovation.
21
3.4. Conceptual model
Based on findings within the theoretical background and hypotheses, the conceptual model was created. The conceptual model shows the relationship between national culture and formal institutions on innovation, depicted in figure 2.
Figure 2: Conceptual model: Culture and formal institutions on innovation
4. Methodology
Below the path along which this study will be directed and the central elements, are described and clarified.
4.1. Sample
22 similar sample sizes, with Shane2 (1992 and 1993) using 33 countries and Efrat3 (2014) 35 countries. This article has a sample size of 30 due to lack of data available. This study will control for national income level and industry structure (Shane, 1993), as these economic variables have been found to influence innovation. For this study Schwartz’s cultural dimensions will be used which he collected between 1988 and 1992 (Schwartz, 1994). Culture tends to remain relatively stable across time and the norms and values tend to be long lasting, (Hofstede, 1991; Inglehart and Baker, 2000), and therefore the effect of discrepancy between time periods is assumed to be not substantial4. The time period 2000 – 2010 is chosen as years before 2000 lack data, and between 2000–2010 more data is available for the chosen 30 countries and variables. The data between 2000–2010 is averaged in order to capture the effect on innovation over a longer time period. The current study uses a variety of secondary sources, such as the World Bank, World Economic Forum and SVS.
4.2. Dependent variable
The dependent variable in this study is innovation at national/country level. Extant research on innovation has used variables such as research and development expenditures and the number of scientific and technical personnel, however these variables are only available for a small number of countries (Shane, 1993). The measurement of innovation at the national level is considered challenging, as measurements such as the proxies mentioned above lack data and are only available for a small number of countries (Shane, 1993).
Previous studies have used three measures to capture innovation, each referring to a different aspect of innovation: (1) patents (Jaffe and Traijtenberg, 2005; Hu, 2004; Shane, 1993; Acs, Anselin and Varga, 2002; Acs and Audretsch, 1988; Seltzer and Bentley, 1999; Taylor and Wilson, 2012; Efrat, 2014), (2) scientific and technical journal articles (Chen, Lin and Huang, 2007; Fagerberg, Srholec and Knell, 2005; Taylor and Wilson, 2012; Amsden and
2 Shane (1992) cited by 510 articles; Shane (1993) cited by 667 articles. 3 Efrat (2014) cited by 62 articles.
23 Mourshed, 1997) and (3) trademarks (Shane, 1993; Mendonça, Pereira and Godinho, 2004; Amara, Laundry and Traoré, 2008; Gotsch and Hipp, 2012). Instead of using a single indicator to measure innovation, this study uses an innovation index composed of patents, scientific and technical journal articles, and trademarks. All three measures were obtained from the World Development Indicator database developed by the World Bank. The index represents per capita values as it is scaled by population, this also allows to control for differences in population sizes. To test for robustness of results, a separate regression analysis will be run based on the individually identified indicators.
The first indicator to measure innovation is the number of patents. Patents were first used by Scherer (1965) and Schmookler (1966) to measure innovation (Taylor and Wilson, 2012). Taylor and Wilson (2012) argue that patents are ‘’the most commonly used quantitative measure of national innovation’’ (p. 238). The use of patents to measure innovation has been criticized, as patents can be affected by differences in national policy (Oxley, 1999). The number of patents do not indicate whether a patent is actually turned into a viable product (Shane, 1993). Furthermore, the number of patents do not take into account the quality of the innovation being patented (Taylor and Wilson, 2012). Despite the criticism on the use of patents, it appears that the use of patents has been generally accepted by the economics literature to be one of the most appropriate indicators to measure innovation (Hagedoorn and Cloodt, 2003). Therefore, this study uses patents as an indicator of innovation. In line with previous research (Jaffe and Traijtenberg, 2005; Hu, 2004; Shane, 1993; Acs, Anselin and Varga, 2002; Acs and Audretsch, 1988; Seltzer and Bentley, 1999; Taylor and Wilson, 2012; Efrat, 2014), the patent indicator is measured as the ‘’ratio between the total numbers of patents granted in a country and the country’s population’’ (Efrat, 2014, p. 15).
24 Mourshed, 1997, p. 359). Furthermore, as only the world’s most essential journals are selected, the publications are ensured to ‘’meet a minimal standard of international quality’’ (Amsden and Mourshed, 1997, p. 359), and therefore comparable across countries. In line with previous research, the scientific and technical journal indicator is measured as the ‘’ratio between the total numbers of scientific and technical journals published per country to the country’s population’’ (Efrat, 2014, p.15; Taylor and Wilson, 2012).
The third indicator to measure innovation is the number of trademarks (Shane, 1993; Mendonça, Pereira and Godinho, 2004; Amara, Laundry and Traoré, 2008; Gotsch and Hipp, 2012). Shane (1993) defines trademarks as ‘’words or devices that differentiate one company’s goods from those of another’’ (p. 64). Furthermore, Shane (1993) indicates that in order for a company to receive a trademark, it must show that the product or service creates distinctiveness in the eyes of the consumer. That is, ‘’a product must be an innovation to receive a trademark’’ (Shane, 1993, p. 64). The use of trademarks to measure innovation is not without limitations. First, not all innovations are registered as trademarks and a single innovation might lead to multiple trademarks (Shane, 1993). Second, trademarks are also subject to differences in national legislation (Shane, 1993). Despite the limitations of trademarks, previous research has used the number of trademarks as a measurement of innovation (Shane, 1993; Mendonça, Pereira and Godinho, 2004; Amara, Laundry and Traoré, 2008; Gotsch and Hipp, 2012). The number of trademarks are measured as the ratio between the total number of trademarks and the country’s population (Shane, 1993).
Following a large body of studies relating the three innovation indicators, the incomplete nature of the chosen proxies for innovation are acknowledged. However, these indicators are used as previous research has identified these factors as suitable proxies to measure innovation. Furthermore, some researchers might argue that the chosen innovation indicators only represent innovation output and neglect innovation input. Therefore, the Global Innovation Index (GII) – by Cornell University, INSEAF and the World Intellectual Property Organization (WIPO) – will be used as a robustness check5 to indicate whether the results might change when innovation input indictors are included. The GII makes a distinction between innovation as an input and output and ranks countries based on several indicators6.
5 Data only available as of 2007, therefore robustness check will include time period 2007-2010. 6
25
4.3. Independent variable
4.3.1. National culture
The first independent variable within this study is national culture. This study aims to examine the relationship between culture and innovation using SVS (Schwartz, 1992 and 1994). Schwartz identified with the SVS seven dimensions along which cultures can be analyzed, from which only intellectual autonomy, affective autonomy, hierarchy, mastery and harmony are considered relevant for innovation. The use of the SVS has also been criticized as the respondents of the SVS were school teachers and college students. It therefore remains unclear if the respondents provide a representative sample of each society (Dostal, 2004). However, despite the questions surrounding the respondents, Schwartz’s cultural values are still perceived ‘’superior’’ (Brett and Okumura, 1998, p. 500; Imm Ng, Lee and Soutar, 2007, p. 164) to other measures (Steenkamp, 2001; Imm Ng, Lee and Soutar, 2007; Schwartz, 1994; Dostal, 2004). That is, the SVS does not exclude prior research on national cultures, but ‘’instead it tends to build on it’’ (Dostal, 2004, p. 25). Schwartz reports the mean scores of the cultural dimensions for 38 countries and cultural groups7 (Schwartz, 1994, p. 112-115). The SVS includes countries with different population groups among them, ‘’geographic, cultural, linguistic, religious, age, gender and occupation groups’’ (Israel Social Science Data Center, 2014, p. 1). The data available for this study are collected between 1988 and 1992 from Schwartz’s book ‘’Beyond individualism/collectivism: New cultural dimensions of culture’’ (Schwartz, 1994, p. 112-115).
4.3.2. Formal institutions
The second independent variable for this study is formal institutions. Intellectual property protection and rule of law were identified as variables for formal institutions. To measure intellectual property protection data8 from the Global Competitiveness Index (GCI) developed by the World Economic Forum (WEF) are used, focusing on sub-index intellectual property protection. The intellectual property protection index includes anti-counterfeiting measures (Schwab and Sala-i-Martín, 2014). The GCI measures the ‘’set of institutions, policies, and factors that set the sustainable current and medium-term levels of economic prosperity’’ (Schwad and Sala-i-Martín, 2014, p. 4). Next, rule of law (Kaufmann, Kraay and
7 Total sample was reduced to 30 countries, as Schwartz split some countries into more groups, such as Estonia, Germany and Israel
8
26 Zoido-Lobaton, 1999) is measured using the World Wide Governance Indicators by the World Bank.
4.4. Control variables
Control variables are used in this research to determine the strength of the relationships between the variables national culture, formal institutions and innovation. In the analysis control variables are included aimed to control for country differences in the macro-environment. Previous literature identified several factors that influence innovation, such as
national income level (Shane, 1992 and 1993; Taylor and Wilson, 2012), industry structure
(Shane, 1993), human capital (Barro and Lee, 2010), trade openness (Taylor and Wilson, 2012; Yanikkaya, 2003; Department for Business Innovation and Skills, 2015; David, 2007), and military expenses (Taylor and Wilson, 2012; McNeill, 1982; Smith, 1985; Ruttan, 2006). National income level and industry structure will be the main control variables (Shane, 1993) due to restricted sample size. As the sample of this study only contains 30 countries, adding to many control variables might affect the reliability of the results. Therefore, the variables human capital, trade openness and military expenses will be used as a robustness check for the results.
First, Vernon (1966) argues that countries with relatively high national income are also the countries in which innovation is more likely to occur. Innovation requires highly skilled staff. Firms are more likely to find highly skilled staff in developed countries as these countries often have high quality universities and greater access to financial resources (Shane, 1993). Taylor and Wilson (2012) argue that a country’s level of economic development will affect innovation, as they indicate that ‘’innovators with more economic resources per capita are better able to transform inputs into new technology’’ (p. 239). This indicates that relatively high national income levels are assumed to have a positive effect on innovation9. Therefore, this study controls for the effect of national income and is operationalized as the
27 Gross Domestic Product (GDP) per capita generated by the World Bank World Development Indicators (Shane, 1992 and 1993; Taylor and Wilson, 2012).
Second, the sectoral structure of a country is also a determinant of the degree of national innovation. It is important to take into account the national differences in sectoral structure, as some industries are associated with higher innovation (Nelson and Winter, 1977). Industries that for example produce physical goods are associated with higher levels of innovation (Shane, 1993). Therefore, national differences in sectoral structure should be taken into account. This study focuses on the industry sector, as this sector is assumed to be highly associated with innovation. The variable sectoral structure was taken from the World Bank, and constructed as the total value added10 – in mining, manufacturing, construction, electricity, water and gas – as percentage of the GDP (Shane, 1993). This ratio shows the ‘’tendency of a nation to have an industrial structure composed of industries most likely to innovate’’ (Shane, 1993, p. 66).
Third, in relation to the quality of highly skilled staff is a country’s human capital pool. Human capital refers to a nation’s knowledge base, the knowledge and capacities of the population (Dakhli and De Clercq, 2004), which also includes the level of academic education and job-related training. Human capital is assumed to positively influence innovation, as knowledge is required for innovation. Human capital is measured as the mean year of schooling developed by Barro and Lee (2010). Barro and Lee developed a new data set of educational attainment in the world for 146 countries (Barro and Lee, 2010).
Fourth, Porter and Stern (2002) indicated that trade openness can also be a determinant for the degree of innovation. MNEs active in international trade face several pressures, such as standardizing or customizing to customer preferences. MNEs often innovate to be able to deal with these pressures. Therefore, it is assumed that trade openness will positively influence national innovation. Trade openness is commonly measured as exports plus imports as a share of GDP from the World Bank (Taylor and Wilson, 2012; Yanikkaya, 2003; Department for Business Innovation and Skills, 2015; David, 2007).
Finally, innovation scholars argue that military spending is considered a major source of technological progress (McNeill, 1982; Smith, 1985; Ruttan, 2006), and is therefore
10
28 associated with innovation. Military spending is operationalized as military spending of the gross domestic product in percentages from the World Bank (Taylor and Wilson, 2012).
4.5. Data analysis
The effect of national culture and formal institutions on innovation are examined by the statistical method least squares multiple regression analysis. This method is chosen as the aim of this study is to determine whether correlation between the multiple concepts exists. This research does not aim to describe causal mechanisms by which culture and formal institutions affect innovation. Statistical data alone are not sufficient to examine these mechanisms (Taylor and Wilson, 2012), therefore this study only tests for the presence of general correlation between the variables. Regression analysis helps to understand whether variation in the independent variable influences the dependent variable to change. In order to be able to execute regression analysis the normality condition needs to be fulfilled (Appendix B). Several models will be created containing different combinations of variables in order to determine the significance of the concepts. The control variables are included into the multiple regression analysis. Due to restricted sample size, the independent variables are included individually in the regression analysis with the control variables (Shane, 1993). In this way, the number of independent variables in the regression analysis never exceeds three11. Finally several robustness checks will be performed12 and assumptions for regression analysis are checked13.
5. Data & Results
This section will discuss the results relating the effect of national culture and formal institutions on innovation. First, the summary statistics of the variables are depicted, followed by a correlation matrix. Subsequently, the results of the effect of national culture and formal institution on innovation are discussed. Finally, the results relating interaction effects between national culture and formal institutions are presented.
11
Exception for the interaction model between culture and formal institutions, which includes interaction terms and control variables.
29
5.1. Descriptive statistics
The summary statistics of the dependent, independent and control variables are shown in table 1, including their sample size, minimum and maximum value, mean and standard deviation. The sample size of this study is 30 as the data for Schwartz’s cultural dimensions are not available for more countries. Therefore, the data for all the variables are gathered for 30 countries. The sample includes inhabitants from every continent14, speakers of 30 different languages, and supporters of twelve religions including atheists (Schwartz, 1994).
Table 1: Sample summary statistics, 2000-2010.
Descriptive Statistics
Variables Sample Minimum Maximum Mean Std. Deviation Innovation index* 30 0,0058 0,2049 0,089257 0,0529083 Affective autonomy 30 2,76 4,41 3,5083 0,41292 Intellectual autonomy 30 3,68 5,33 4,3168 0,41016 Hierarchy 30 1,69 3,70 2,5045 0,52151 Mastery 30 3,63 4,73 4,1184 0,24932 Harmony 30 3,21 4,80 4,0993 0,44622 Rule of law 30 2,45 99,41 74,2917 23,16238 Intellectual property protection 30 2,73 6,23 4,7283 1,13036 GDP per capita ($1000s)** 30 $475,88 $53.606,55 $21.436,9427 $14.939,62652 Sectoral structure*** 30 9,27 46,44 29,4990 7,54430 Note: * DV= Innovation index: mean of total patents, total scientific and technical journals and total trademark applications per capita (population) in percentages (2000-2010). **GDP per capita in current U.S dollars; *** Industry, value added (% of GDP) in percentages.
5.2. Correlation matrix
The correlation matrix shows the correlation between all independent, dependent and control variables. When analyzing the correlation matrix, one should pay attention to problems of multicollinearity, a ‘’condition wherein the independent variables are highly correlated’’ (Keller, 2012, p. 680). Correlation among variables which are higher than 0,80 or 0,90 indicate possible problems of multicollinearity (Kennedy, 2003). In sections following the correlation matrix, the variance inflation factor (VIF) will be used to assess whether problems of multicollinearity can be identified (Keller, 2012). A VIF higher than 10 indicates problems of multicollinearity (O’brien, 2007).
14
30 Table 2 shows the correlation between all independent, dependent and control variables.
Table 2: Correlation matrix (N=30)
Variable 1 2 3 4 5 6 7 8 9 10 1. Affective autonomy 1 2. Intellectual autonomy 0,65** 1 3. Hierarchy -0,18 -0,45** 1 4. Mastery 0,1 -0,2 0,42* 1 5. Harmony 0,16 0,53** -0,69** -0,4* 1 6. Rule of Law 0,32* 0,45** -0,44** -0,46** 0,23 1
7. Intellectual prop. prot 0,44** 0,45** -0,34* -0,26 0,08 0,88** 1
8. GDP per capita 0,44** 0,54** -0,36* -0,19 0,18 0,8** 0,86** 1
9. Sectoral structure -0,22 -0,09 0,21 0,07 0,03 -0,4* -0,32* -0,51** 1
10. Innovation index -0,01 0,07 -0,28 -0,22 0,15 0,47** 0,39* 0,23 -0,18 1 Note: DV= Innovation index: patents, scientific and technical journals and trademarks (2000-2010). Pearson correlation coefficient .* p<0.05, ** p<0.01 (one-tailed).
Correlation between the cultural variables exists, but these correlations are not higher than 0,80 or 0,90 and therefore are not problematic in the sample. Intellectual autonomy correlates positively with affective autonomy and harmony and negatively with hierarchy at the 1% level. Mastery correlates positively with hierarchy at the 5% level. Harmony correlates both negatively with hierarchy and mastery, with hierarchy at the 1% level and mastery at the 5% level. Next to the cultural variables, correlation exists between the formal institution variables and cultural variables. The correlations between these variables are not problematic as they do not exceed 0,80 or 0,90. The correlation matrix does not indicate significant correlation between the cultural variables and the innovation index. However, positive correlation between the formal institution variables and innovation exists, whereby rule of law and innovation correlate positively at the 1% level and intellectual property protection and innovation at the 5% level. Based on the correlation matrix, the relationship between the cultural variables and innovation is insignificant, and the relationship between formal institutions and innovation is positively significant.
31 protection and GDP per capita positively correlate at the 1% level, with correlation coefficient 0,86. Based on the correlation matrix these correlations described above indicate possible multicollinearity. However, the correlation matrix is not the only measure to detect multicollinearity, the variance inflation factor (VIF) will also be used to assess whether multicollinearity can be detected (Keller, 2012). VIF higher than 10 indicates problems of multicollinearity (O’brien, 2007). As possible multicollinearity exists between rule of law, intellectual property protection and GDP per capita, these variables are included individually in the regression analysis and simultaneously. In this way, the effects of the independent variables on innovation can be analyzed separately.
5.3. Culture and innovation
This section will focus on the results of the effect of culture on innovation. First, regression analysis is performed for each of Schwartz’s cultural dimensions individually, so the effect of each dimension can be analyzed separately on innovation. The following regression analysis includes economic control variables to see whether these variables cause the relationship between culture and innovation to change.
Table 3 shows the results for each of the cultural variables separately on innovation. Table 3: Results culture on innovation
Variable/model 1 2 3 4 5 Affective autonomy -0,0011 (0,0242) Intellectual autonomy 0,0094 (0,0243) Hierarchy -0,0285 (0,0184) Mastery -0,0467 (0,0391) Harmony 0,0173 (0,0222) Constant 0,0932 0,0487 0,1607 0,2817 0,0183 R Square 0,0001 0,0053 0,0790 0,0485 0,0213 N 30 30 30 30 30
32 The regression results indicate that the cultural dimensions do not show a significant relationship with innovation. In all regression models the coefficients are small and insignificant at the specified significance levels. Intellectual autonomy and hierarchy do have the same relationship with innovation as hypothesized. Although the results are insignificant, it is surprising that the relationship between affective autonomy, mastery and harmony and innovation is the opposite compared to what was expected. The beta-coefficient of affective autonomy and mastery are slightly negative, whereby affective autonomy has a beta-coefficient of -0,0011 and mastery -0,0467. The beta-beta-coefficient of harmony is slightly positive 0,0173. Within the table above, model 3 has the best explanatory power with R square of 7,90%, indicating that hierarchy explains around 7,90% of the variation in the dependent variable innovation. Model 3 seems to be the best model to explain variation caused in the dependent variable, even though the model does not show any significance. Based on the results provided in table 3, it is necessary to reject hypotheses 1 till 4.
Table 415 shows the results of the culture variables with economic control variables.
Table 4: Culture and control variables on innovation
Model 1 2 3 4 5 6 Variables Ctrl. Var Affective autonomy + Ctrl. Var. Intellectual autonomy + Ctrl. Var. Hierarchy + Ctrl. Var. Mastery + Ctrl. Var. Harmony Ctrl. + Var. Affective autonomy -0,015 (0,027) Intellectual autonomy -0,002 (-0,031) Hierarchy (0,020) -0,027 Mastery (-0,040) -0,041 Harmony 0,029 (0,024) GDP per capita 6,621E-07 (0,000) 8,412E-07 (0,000) 7,045E-07 (0,000) 3,309E-07 (0,000) 5,173E-07 (0,000) 4,358E-07 (0,000) Sectoral structure -0,001 (0,002) -0,001 (0,002) -0,001 (0,002) -0,001 (0,002) -0,001 (0,002) -0,001 (0,002) Constant 0,090 (0,056) 0,138 (0,104) 0,098 (0,122) 0,164 (0,078) 0,264 (0,179) -0,014 (0,106) R Square 0,06 0,071 0,06 0,124 0,097 0,109 N16 29 29 29 29 29 29
Note: DV= Innovation index: patents, scientific and technical journals and trademarks; Standard errors reported between parentheses (2000-2010). *p<0.05, ** p<0.01, *** p<0.1.
15 No problems of multicollinearity arise, as the VIF has a maximum value of 2,054. 16
33 The control variables are entered simultaneously with the cultural variables, to see whether the control variables cause the relationship between culture and innovation to change. The results still indicate that the cultural dimensions do not show a significant relationship with innovation, even when controlling for GDP per capita and industry structure. Model 1 includes only the control variables and model 2-4 include a cultural dimension and control variables. The control variables cause the beta-coefficient of intellectual autonomy to change from slightly positive to slightly negative. GDP per capita indicates in all the models a positive relationship with innovation, even though the beta-coefficients are insignificant. Sectoral structure shows a slightly negative relationship with innovation across all 5 models. The control variables do not cause the relationship between culture and innovation to change, the results remain insignificant. Based on the results of table 3 and 4 hypotheses 1-4 can be rejected, indicating that the cultural variables do not have a direct effect on innovation.
5.4. Formal institutions and innovation
This section presents the results for the relationship between formal institutions and innovation. First, regression analysis is performed for rule of a law and intellectual property protection individually, to be able to analyze the relationship between these variables and innovation separately. Thereafter, the regression analysis includes economic control variables to see whether these variables cause the relationship between formal institutions to change. Table 5 shows the results for the relationship between formal institutions and innovation.
Table 5: Formal institutions on innovation
Model 1 2 3 4 5
Variable Rule of law
Intellectual property protection Rule of law + Ctrl. Var. Intellectual property prot. + Ctrl. Var. Rule of law + Intell. Prop + Ctrl. Var Rule of law 0,001 (0,000)*** 0,002 (0,001)** 0,001 (0,001) Intellectual property protection 0,018 (0,008)* 0,039 (0,016)* 0,018 (0,0,21) GDP per capita -1,542E-06
(0,000) -2,145E-06 (0,000) -2,222E-06 (0,000) Sectoral structure -0,001 (0,001) -,002 (0,001) -,001 (0,001) Constant 0,010 (0,030) 0,003 (0,040) 0,006 (0,058) -,008 (0,066) -,016 (0,064) R Square 0,219 0,152 0,288 0,238 0,308 N 30 30 29 29 29
34 Model 1 shows that rule of law has a positive significant effect on innovation. The beta-coefficient of rule of law is 0,001 at the 10% level. In model 3 the control variable are entered into the model with rule of law, rule of law remains positive significant with beta-coefficient 0,002. In model 3 the level of significance of rule of law increases to the 1% level. Model 2 shows the relationship between intellectual property protection and innovation, whereby intellectual property protection has a positive significant effect on innovation. The beta-coefficient of intellectual property protection is 0,018 at the 5% level. Model 4 includes both intellectual property protection and control variables, whereby the beta-coefficient of intellectual property protection increases significantly to 0,039 at the 5% level. Model 5 includes both independent variables rule of law and intellectual property protection and the control variables. In model 5 both formal institution variables show a positive relationship with innovation as hypothesized. However, both variables and control variables are insignificant. Both rule of law and intellectual property protection show a positive and significant relationship with innovation, whereby intellectual property protection has a slightly stronger positive effect on innovation. Based on the results provided above hypotheses 5 and 6 can be confirmed, as rule of law (model 1 and 3) and intellectual property protection (model 2 and 4) both have a positive significant effect on innovation. The effect of the control variables are both negative and insignificant in model 3-5.
35
5.5. Culture and formal institutions on innovation
Previous sections 5.3 and 5.4 examined the effects of culture and formal institutions on innovation separately. This section focuses on both the effects of culture and formal institutions simultaneously on innovation. Each of Schwartz’s cultural dimensions are entered at the same time with intellectual property protection and rule of law. Table 6 shows the results for the extent to which both forces – national culture and formal institutions – influence innovation.
Table 617: Culture and formal institutions on innovation
Model 1 2 3 4 5
Variable Affective autonomy
Intellectual
autonomy Hierarchy Mastery Harmony -0,023 (0,025) -0,022 (0,025) -0,009 (0,020) 0,004 (0,044) 0,003 (0,022) Intellectual property
protection Yes Yes Yes Yes Yes Rule of law Yes Yes Yes Yes Yes Constant 0,080 (0,077) 0,097 (0,098) 0,044 (0,073) 0,002 (0,193) 0,003 (0,099) R Square 0,247 0,245 0,227 0,222 0,222
N 30 30 30 30 30
Note: DV= Innovation index: patents, scientific and technical journals and trademarks (2000-2010). Standard errors reported between parentheses. *p<0.05, ** p<0.01, *** p<0.1
The regression results indicate that the cultural dimensions and formal institutions do not show a significant relationship with innovation. In all regression models the coefficients are small and insignificant at the specified significance levels. Affective autonomy, intellectual autonomy and hierarchy show a negative relationship with formal institutions on innovation, whereas mastery and harmony show a positive relationship with formal institutions on innovation. The R square of the models is quite similar, whereby model 1 has the highest R square of 24,7%. The R squares above indicate that culture and formal institutions explain around 22,2 – 24,7% of variation in the dependent variable innovation.
5.6. Interaction culture and formal institution on innovation
This section shows the results for the interaction effect between culture and formal institutions on innovation. Previous research has identified that cultures that are individualistic in nature have a positive effect on the relationship between formal institutions
17
