Economic Policy Uncertainty and Innovation as an Investment in
Intangible Assets: The Role of Economic Freedom and
Corruption.
Florence Nieves Asensio: 11443456
August 17th, 2018, Final
MSc, in Business Administration – International Management Track Supervisor: Dr. Mashiho Mihalache
Statement of Originality
This document is written by Student Florence Nieves Asensio who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Table of Contents Abstract ... 4 1. Introduction ... 5 2. Literature Review ... 9 2.1. Innovation ... 9 2.2. Institutional Environment... 12
2.3. Economic policy uncertainty ... 14
2.4. Corruption ... 16
2.5. Economic Freedom ... 18
3. Conceptual Framework and Hypotheses ... 22
3.2. Economic Policy Uncertainty and Domestic Innovation Output ... 22
3.3. Moderating Effect of Corruption ... 24
3.4. Moderating Effect of Economic Freedom ... 26
4. Methodology ... 29
4.1. Sample and Data Collection ... 29
4.2. Measures ... 32
4.2.1. Independent Variable ... 32
Economic Policy Uncertainty ... 32
4.2.2. Dependent Variable ... 33 Innovation ... 33 4.2.3. Moderating Variables ... 35 Corruption ... 35 4.2.4. Control Variables ... 37 5. Results ... 39 5.1. Model Specification ... 39 5.2. Descriptive Statistics... 41 5.3. Findings ... 43 6. Discussion ... 46 6.1. Theoretical Implications ... 48 6.2. Managerial Implications ... 50
6.3. Limitations and Recommendations for Future Research ... 53
7. Conclusion ... 56
Abstract
Over the years and specifically since the recession of 2008/2009, there has been a global increase in economic policy uncertainty. This paper aims to further delve into this theoretically novel concept in order to further understand its economic implications, specifically in relation to a country’s innovative activities. We extend on research concerning a country’s innovative output, as it is crucial for a country’s economic growth, by examining the relationship between economic policy uncertainty and innovation, and how elements of country’s institutional environment (corruption and economic freedom) moderate this negative relationship. Motivated by previous papers, we examine panel data for 21 countries across 13 years (2003-2015), we find that corruption acts as an exacerbator and thus increases the negative relationship between economic policy uncertainty and innovation. Where a country’s level of economic freedom acts as buffer and decreases the negative relationship between policy uncertainty and innovation.
keywords: Innovation, Economic Policy Uncertainty, Corruption, Institutions, Economic Freedom
1. Introduction
Technological innovation is composed of the implementation of a new or significantly improved products (goods or services), processes, organizational method in business practice, work place organization or external relations (Zanello et al, 2016; GII, 2017). It is deemed to be a key driver for the development of economic growth and has the potential to generate a sustainable economic advantage in developing and developed countries, indicating its economic value (Chen et al., 2014; Lee, O ̈ zsomer, and Zhou, 2015). According to Rosenberg (2004) technological innovation is accountable for about 85% of economic growth in countries. Emphasizing the importance of technological innovation with regards to a nations long term economic growth. A concept which has been comprehensively established in the past (Solow, 1957).
In addition, there has also been an advancement in the development of basic indicators of innovation at a country level, which have further revealed the potential impact of innovation with regards to a country’s economic development (Zanello et al., 2016). Placing innovation at the core of government growth strategies, leading to innovation having more of a relevant place with policy makers (GII, 2017; Zanello et al., 2016). Thus, showcasing innovation and its relevance with regards to a country’s institutional environment and the policies set within (Fu and Gong, 2011; Urban, 2014; Zanello et al., 2016). In addition, the moderating effects of institutional enablers or inhibiters and its subsequent effect on a country’s innovation output has not been adequately studied (Wang and Libaers, 2015). Indicating a gap in the literature regarding innovation, which can be regarded as a significant omission, since it can offer an opportunity to gain new insights, expand theory and can thus increase an understanding in this field by incorporating more contextualised considerations (Bruton et al., 2008). Justifying this papers interest for delving into research that deals with elements of a country’s institutions and
policies in relation to innovation for developed and developing countries. Especially, since the variety between countries can lead to insights into the different effects the country’s institutions can have on their capacity for innovation, adding more contextualised considerations (Furman, Porter and Stern, 2002).
Researchers have also emphasized the importance of the institutional setting of a country since it has significant control over regulatory policies and scarce resources, thereby shaping a firms’ competitive environment and its capacity for innovation (North, 2005; Peng, Wang, and Jiang, 2008; Gao et al, 2010; Nee and Opper, 2012). Therefore, in order to assist the process of innovation, efficient and effective institutions and policies that guide incentives for innovations are needed (Fu and Gong, 2011). Where members of government determine the policies and how conducive these policies are to innovative activities and their corresponding certainty (Fu and Gong, 2011; DiRienzo Das, 2015; GII, 2017). Further accentuating the institutional setting of a country, as it deals with the quality, the implementation and the creditability of the government’s policy formulation and the commitment to such policies (Büthe and Milner, 2008; GII, 2017). Establishing the significance of policies within a countries institutional framework and thus showcasing a more specific gap in literature regarding innovation. As a result, this study contributes to the growing literature regarding innovation, through the examination of the effect of institutional environment and corresponding policies on innovative activities.
Moreover, the effect of a nations policies and their capacity to influence innovation has been understudied (DiRienzo and Das, 2015; Bhattacharya et al., 2017). And more specifically, the perception of the certainty of such policies in relation to innovation has been mostly unexplored, since data bases measuring political indices are scarce (Bhattacharya et al.,2017).
And as the EPU index has become more available for use, this article will contribute to the literature with regards to innovation with a specific focus on the effect of economic policy uncertainty (Nguyen et al., 2017; Perić and Sorić, 2018). Especially, since there have been many studies which have emphasized the negative consequences of economic policy uncertainty on a nations entrepreneurial environment (Rodrik, 1991; Julio and Yook, 2012; Nguyen et al., 2017). As a result, it is regarded as a significant deterrent to tangible investments and a country’s capacity for growth (Bloom et al., 2007; Julio and Yook., 2012; Gulen and Ion.,2016; Baker et al., 2016). However, it has been understudied in relation to innovation, which is a significant omission, as innovation is regarded as an investment in intangible assets, which in turn affects a country’s capacity for growth (Ayyagari, Demirgüç-Kunt and Maksimovic, 2014; Bhattacharya et al., 2017). Therefore, this paper contributes on the growing literature which deals and focuses on the negative consequences of economic policy uncertainty on firms’ investment decisions and its consequent effect.
Nevertheless, a study conducted by Bhattacharya et al. (2017), found a significant negative relationship between policy uncertainty and innovation, which to our knowledge is the only paper having investigated these concepts in relation to one another. This is possibly due to both innovation and policy uncertainty being relatively new concepts which are regarded as difficult to measure as no standard measurement has yet been developed (Nguyen et al., 2017; Bhattacharya et al., 2017). However, as of late the EPU index has become more available for use in cross and longitudinal studies (Nguyen et al., 2017; Perić and Sorić, 2018). This index is also regarded as a more direct, efficient and contextually appropriate measure of economic policy uncertainty (Pastor and Veronesi, 2012; Nguyen et al., 2017). As a result, this study aims to delve further into the aspect of economic policy uncertainty and innovation, by extending on the research conducted by Bhattacharya et al. (2017). Through the use of the EPU
index which is considered to more appropriate to measure economic policy uncertainty and by investigating how other aspects of a country’s institutional environment such as corruption and economic freedom influence this relationship (Nguyen et al., 2017).
Understanding of the effect of economic policy uncertainty is also deemed particularly relevant in practice, in light of the growing economic policy uncertainty on a global level, aggravated by the recession and other aspects political uncertainty as discussed under managerial implications section (Bloom, 2013; Nguyen et al., 2017; Fahey, 2016; Hilsenrath, 2018). This results in policy makers and business leaders having to rectify the negative impact it has had on innovative activities and thus develop policies which incentivise and promote innovation and thus a country’s economic growth (DiRienzo and Das, 2015; Euro Monitor, 2018). This has further added to this papers interest in understanding how aspects of a countries institutional environment, such as corruption and the level of economic freedom influences economic policy uncertainty and consequentially a country’s innovative capacity and thus their ability to grow economically.
In sum, this study intends to aid policy makers in developing more adequate policies in order to promote the economic development of their respective countries, because previously, the implementation of policies to favour innovation in countries may not have worked as policy makers expected (DiRienzo and Das, 2015; Zanello et al., 2016). Furthermore, it could further assist in the assessment and measurement of the performance and consequences of policies and thus enable policy makers to ensure future interventions are effective and efficient in order to promote economic development in their respective country (Niosi, 2010; Lenihan, 2011). This paper intends to achieve this by further reviewing and analysing the resultant effect of economic policy uncertainty on innovation, thus also theoretically contributing to literature.
The paper is constructed as follows; section 2 consists of the literature applicable to this topic, which defines the concept of innovation, followed by stating the importance of the institutional environment in relation to innovation and policies. After which the concept of economic policy uncertainty and its relation to innovation is discussed in detail. This is followed by defining economic freedom and corruption and how this relates to policy uncertainty. Section 3 deals with the development of the conceptual framework of how economic policy uncertainty should result in the decrease in a country’s innovation output, and how corruption and economic freedom act as moderators and interact with economic policy uncertainty. This is where the conceptual framework and hypotheses are developed. Section 4, is where the methodological strategy is presented, and the variables used are explained. Followed by the results in section 5, in this section the estimation model, descriptive statistics and findings are discussed. Section 6 entails the theoretical and managerial implications of the findings, including the future recommendations and limitations of this paper. Lastly, section 7 concludes the arguments presented in this paper.
2. Literature Review 2.1. Innovation
Due to the diversity of technological innovation, a broad definition of Innovation will be used (Zanello et al., 2016, GII, 2017). This definition states that innovation is composed of the implementation of a new or significantly improved products (goods or services), processes, organizational method in business practices, work place organization or external relations (Zanello et al, 2016; GII, 2017). This definition was originally elaborated upon from the Oslo Manual, which was developed by European communities and the organization for co-operation and economic development (OECD, 2005). Demonstrating how the meaning of innovation has changed in the past two decades as it used to be focused on research and development (R&D)
and technological product innovation which was mainly produced by manufacturing companies internally (GII, 2017). As a result, technological innovation would have to be radical and take place at the global knowledge frontier (GII, 2017).
This new definition, however, implies that innovation can take multiple forms, such as; product innovation, process innovation, marketing innovation, managerial and organizational innovation (Zanello et al, 2016). In addition, it is important to be cognizant of the different forms of innovation as each form has a different effect on diffusion patterns and on firm performance (Hall, 2011). Moreover, this definition infers that innovation does not have to be new to the market or to the world as a whole, as it can be new to the firm and still have an influence on its productivity (Tushman and Anderson, 1986). Therefore, innovation can be regarded as radical or imitative, which both have economic value despite their implementation process being different (Zanello et al, 2016). Showing that, the capability of innovation is also about the ability to exploit new technical combinations, emphasizing the concept of incremental innovation and innovation which excludes research (GII, 2017). Thus, the creation, adoption, assimilation, diversification of technologies all form part of the innovative process (Zanello et al, 2016).
In addition, innovation can be viewed as a special investment in long term intangible assets that ought to lead to the generation of profits in the future (Ayyagari, Demirgüç-Kunt and Maksimovic, 2014; Bhattacharya et al, 2017). It is viewed differently from investments in tangible assets, as investments related to innovations have a longer time horizon and a higher tail risk (Hsu, Tian, and Xu, 2014; Bhattacharya et al., 2017). Furthermore, investments in innovation activity has become more significant at the firm, country, and global level, indicating its increasing relevance (Hsu, Tian and Xu, 2014; GII, 2017). As a result, economic
factors that affect innovation are also different to those that affect regular investments and it would be economically advantageous to identify these in order to promote a country’s economic growth (Solow, 1957; Chen et al., 2014; Lee, O ̈ zsomer and Zhou, 2015) Bhattacharya et al., 2017).
Innovation is considered to be diverse and complex, as a country’s capacity for innovation is embedded in and constituted by the dynamics between political, legal, geographic and socioeconomic subsystems (DiRienzo and Das, 2015; Zanello et al., 2016; GII, 2017). It is thus a multidimensional concept, for which its production at a country level is influenced by many factors (Crespi, 2004). Therefore, it is crucial for local capabilities to identify the right technology, suitable transfer mechanisms and to absorb and make adaptations to local economic, technical and environmental conditions (Fu and Gong, 2011). Which, necessitates for appropriate institutions and policies to guide incentives and facilitate the process (Nelson, 1993, Furman, Porter and Stern, 2002; Fu and Gong, 2011; Bhattacharya et al., 2017). Emphasizing the importance of the institutional environment in its role to promote and affect innovation which in turn plays a crucial role for the industrialization and economic progress of countries (Furman, Porter and Stern, 2002; Zanello et al., 2016). Success is then dependent on substantial and well directed technological efforts, sufficient human, financial and absorptive capacity of firms and industries (Cohen and Levinthal, 1989; Keller, 1996).
There has been a scarcity of research conducted regarding innovative activities in countries and the corresponding factors that influence them despite its economic value (Fagerberg and Verpsagen, 2009; Martin, 2012). This is a missed opportunity, especially as there is a vast variety amongst countries, making it highly feasible to examine the various the factors that impact innovation (Furman, Porter and Stern, 2002). It would also be beneficial to create a
further understanding of this topic in practice and for policy makers, due to their ability to significantly influence innovative activities, in order to promote a country’s economic growth and thus create a sustainable economic advantage (Solow, 1957; Fu and Gong, 2011; Chen et al., 2014; Lee, O ̈ zsomer and Zhou, 2015; Bhattacharya et al., 2017). Signifying this particular topic’s importance and consequentially this papers interest and motivation. The following section will discuss the importance of institutional environment in more detail and how it effects a country’s innovative activities.
2.2. Institutional Environment
The institutional environment of a country is composed of political, social and legal rules that set the foundation for production and distribution to which organizations must adhere to in order to gain support and legitimacy (North, 1990; DiMaggio and Powell, 1991; Scott, 1994). Institutions are thus regarded as the rules of the game and determine the certainty of said rules (Stiglitz, 2006; Welter and Smallbone, 2011; Xavier et al., 2012). These rules can be formal or informal and determine the incentive and cost structure of entrepreneurial activity (Stiglitz, 2006; Welter and Smallbone, 2011; Xavier et al., 2012). Which can be further divided into regulatory, normative and cognitive categories (Scott, 1994; Kostova, 1997; North, 1990; Busenitz et al., 2000). Where institutional theory places an emphasis regarding changes in macroeconomic and regulatory policy and the uncertainty it can create, deemed as formal rules (Marcus, 1981). As a result, the paper will focus on the formal rules set by institutions and therefore the policies that members of a country set and their consequent effect on innovative activity.
The government in a country is regarded as one of the most prominent institutions, as they have a significant influence over regulatory policies and control over scarce resources and therefore
shapes a firms’ competitive environment (North, 2005; Peng, Wang and Jiang, 2008; Gao et al., 2010; Nee and Opper, 2012). Emphasizing the relevance of the government and its ability to influence entrepreneurial activities in countries. It is important to note however that there is a difference between the institutions in emerging economies and those in developed economies (Zoogah et al., 2015). Developed countries can to an extent rely on conventional and stable political and economic environments, whereas developing countries often cannot (Zanello et al., 2016). This variety between countries can lead to insights into the different effects institutions and specifically governments, can have on a country’s capacity for innovation (Furman, Porter and Stern, 2002).
Furthermore, many governments are placing innovation at the core of their growth strategies, due to its beneficial economic potential (GII, 2017). Leading to an emphasis on the necessity for appropriate institutions and policies in order to develop a conducive environment for entrepreneurial activity relating to innovation (Furman, Porter and Stern, 2002; Fu and Gong, 2011; Chen et al., 2014; Lee, O ̈ zsomer, and Zhou, 2015). Implying that efficient and effective government regulation and policy is essential to innovation, which can be achieved through ensuring the development of an institutional framework that attracts business and promotes long term growth (Wang and Libaers, 2015; GII, 2017). Highlighting the importance of institutions and policies, especially as their perceived quality can either strengthen or weaken innovative capacity of firms (DiRienzo and Das, 2015; Wang and Libaers, 2015).
Therefore, a country’s institutional environment composes the structure of economic incentives, which can either limit or expand the strategic choice set of firms and consequentially effect the development and introduction of innovations at a country level (Dimaggio and Powell, 1983; Tolbert and Zucker, 1983; Wang and Libaers, 2015). Where
changes within institutions and corresponding policies, that generate uncertainty are regarded as hampering the perception of institutions and policy, thus negatively affecting a firm strategic choice set (Laffont, 1989; Wang and Libaers, 2015; Bhattacharya et al., 2017). Therefore, due to the lack of research completed on the effects of regulation and more specifically economic policy uncertainty and its influence on innovative activity (Zanello et al., 2016; GII, 2017; Nguyen et al., 2017; Bhattacharya et al., 2017). This paper will study how an economy’s economies policy uncertainty level affects a countries innovation output. Consequentially the next section will expand upon the concept of economic policy uncertainty.
2.3. Economic Policy Uncertainty
Institutional theory emphasizes the importance of macro-economic and regulatory policy and the subsequent policy certainty or uncertainty that can be generated (Marcus, 1981). As previously stated, institutions have a significant impact on innovative activities, implying that policies generated within an institutional context have the ability to substantially influence innovation generation, as certain rules will be beneficial for others and vice-versa (North, 1990; Barbosa and Faria, 2011). The focus of this paper will as a result be on the perception of economic policies in countries and its effect on the output of innovation produced, where perception refers to the degree to which polices are regarded as uncertain (Nguyen et al., 2017). Especially, since the concept of economic policy uncertainty is regarded as being important but yet unexplored (Nguyen et al., 2017).
A policy is a “plan or course of action, of a government, political party, or business intended to influence and determine decisions, actions and other matters” (Farlex, 2018). It can be referred to as a guiding policy, course of action or even a procedure (Farlex, 2018). Policies for this paper will focus on policies generated by government or a political party, who develop
policies and regulations that can influence a country’s economic environment and consequently a countries innovation output (Pa ́ stor and Veronesi., 2012; Zanello et al., 2016; Nguyen et al., 2017). In addition, Policies refer to matters such as the share of government expenditure, the share of public investment in gross domestic product, intellectual property policy, tax policy, trade policy, domestic credit expansion or contraction (Freeman and Soete, 1997).
Businesses have to adapt to their policy environment in order to produce appropriate innovations, innovations are therefore affected differently in different policy environments and industries (Bloom 2009; Zanello et al., 2016; Nguyen et al., 2017). Where policy uncertainty in countries appear due to inconsistent formulation and application of regulatory and macro-economic policies or due to policies that are regarded as normative and do not persevere over time (Rodrik, 1991; Julio and Yook, 2012; Bhattacharya et al., 2017). As aforementioned, changes in the political environment that generate uncertainty influence the structure of economic incentives, thus affecting the strategic options of firms and consequently a countries’ innovation output (DiMaggio and Powell, 1983; Tolbert and Zucker, 1983; Wang and Libaers, 2015). Policy uncertainty therefore plays a fundamental part in a country’s institutional, political and economic environment and thus on the process of generating innovation (Wang and Libaers, 2015).
This paper will focus on policy uncertainty as it is considered to lead to a decline in investment expenditures, increasing real option values, delaying investment and thus having a negative relationship on firm level capital investment, indicating its overall negative effect on investments (Bloom et al., 2007; Julio and Yook., 2012; Gulen and Ion.,2016; Baker et al., 2016). Thus, this may even affect innovation negatively, as it is regarded as an investment (Ayyagari, Demirgüç-Kunt and Maksimovic, 2014; Bhattacharya et al, 2017). Bhattacharya et
al (2017) confirmed this notion through their findings which state that policy uncertainty has a negative impact on a country’s innovation quantity, quality and originality. In addition, they state that policy uncertainty is said to be amplified for radical innovations, innovation intensive industries and ethnically heterogeneous countries and is said to be less significant for incremental innovation (Bhattacharya et al., 2017). Lastly, given the rise of policy changes occurring globally, it has been speculated that this concept will increase in relevance, further justifying this topics use (Bloom, 2013; Nguyen et al., 2017; Fahey, 2016; Hilsenrath, 2018).
Due to policy uncertainty negatively affecting an economy’s incentive to invest and innovate it can lead to a recession and thus damage a country’s growth (Bernanke, 1983; Bloom., 2009; Bhattacharya et al., 2017). According to Bhattacharya et al (2017) it does not matter which policy is put in place, it is the political gridlock and resulting policy uncertainty that matters and affects businesses as they do not know which policies to adapt too, leading to adverse economic consequences. Further highlighting this topics relevance and motivation behind investigating this topic. Consequently, this paper will extend on aforegoing literature and in addition will investigate how corruption and the level of economic freedom influences policy uncertainty and its resulting impact on a country’s innovation generation. For which corruption will be discussed next in the following section.
2.4. Corruption
Corruption is known to hamper economic growth, weaken the rule of law and undermine the legitimacy of institutions (Mauro, 1995). Issues regarding corruption are seen as an age-old problem, considered as being as old as governments and even human existence itself and is inherent in developed and developing countries (Lipset and Lenz, 2000; Mauro, 1995). Corruption is thus able to endure despite the development and modernization of countries,
taking on new forms instead (Girling, 1997). Therefore, it has always been part of reality, but to different extents and with various consequences and is thus regarded to be global issue, since it can be a major and dominant obstacle in the process of economic development (Bardhan, 1997; Klitgaard, 1988). Consequently, economists and global financial institutions such as the world bank and IMF (2018) have spent significant time on this issue, as it is becoming a growing concern (Lučić, Radišić and Dobromirov, 2016).
Corruption has been studied as an issue regarding political, social, cultural and moral underdevelopment (Lučić, Radišić and Dobromirov, 2016). There are numerous definitions regarding corruption both in academic literature and in policy documents of developed organizations (Walton, 2014). These definitions range from broad terms such as the abuse of power and institutional decay to a more formal/ legal definition that relates to public office (Walton, 2014; Lučić, Radišić and Dobromirov, 2016). This paper will use the most widely applied definition in recent times which can be referred to as the public office definition, which holds for both academic literature and policy documents of developed organisations (Walton, 2014; Lučić, Radišić and Dobromirov, 2016). In this context corruption will be referred to as; the abuse of public office for private gain (Jain, 2001). The popularity of the definition is mainly as a result of enduring concerns with regards to state corruption (Rose-Ackerman, 1999; Klitgaard, 1991).
It is important to note however that this definition can be expressed in different ways but they all link corruption to public officials acting out in private regard according to the World Bank (2018), Transparency International (2018) and Jain (2001). To be more specific corruption then refers to an arrangement between two parties (supplier and demander) that involves an exchange (Jain, 2001). This exchange influences the allocation of resources directly or in the
future, while involving the use or abuse of public or collective responsibility for private gain, in a manner that contravenes the rules of the game (Macrae, 1982; Jain, 2001, DiRienzo and Das, 2015). Therefore, this paper will focus on corruption committed by government officials who use their power for private gain and how this can lead to economic policy uncertainty and influence innovative activities (DiRienzo and Das, 2015; López-Valcárcel, Jiménez and Perdiguero, 2017).
Consequently, corruption leads to the misallocation of public spending, as resources are being allocated towards the payment of bribes instead of the improvement of lives and the economic environment (Alam, 1995; Ades and Di Tella, 1999; Montes and Paschoal, 2016). Leading to inefficient governments with regards to the provision of public goods, services and policy making (Wei, 1999; Zhao et al., 2003; Cuervo-Cazurra., 2006; DiRienzo and Das, 2015; Montes and Paschoal, 2016). In summary, corruption affects the formulation, adoption and the credibility of governments commitment to policies and thus the certainty of such policies (DiRienzo and Das, 2015; Montes and Paschoal, 2016). Therefore, this paper will investigate how corruption influences economic policy uncertainty and consequentially a country’s innovation production. Lastly, the level of economic freedom will be discussed below.
2.5. Economic Freedom
Economic freedom characterizes the degree to which an economy is a market economy, which refers to the degree to which one enters into voluntary contracts within a framework of a stable and predictable rule of law that upholds contracts, consists of limited government intervention in the form of ownership, regulations and taxes and to the extent to which it protects private property (Kreft and Sobel, 2005; Berggren, 2003). Economic freedom is best measured using the economic freedom of the world index (EFW) as it is regarded to be one of the most
ambitious attempt to measure it (Berggren, 2003). Furthermore, this index will be used as it is one of the most used measures due to its value to researchers, since measurements are more precise, transparent and available across more time periods (Gwartney and Lawson, 2003).
The EFW is composed of five main elements, size of the government, legal system and property rights, sound money, freedom to trade internationally and regulation, which are regarded as crucial functions of economic progress and development (Farr, Lord, and Wolfenbarger, 1998; Gwartney, Lawson, and Holcombe 1999; Graeff and Mehlkop, 2002; Gwartney and Lawson, 2003; Cole, 2003; Powell, 2003). Economic freedom is deemed to be part of the institutional framework within an economy as it has an effect on the incentives of individuals to produce and effects the efficiency of resource allocation (Graeff and Mehlkop, 2002; Berggren., 2003). Therefore, an institutional arrangement that is in conflict with economic freedom through the restriction of trade, undermining of property rights, high tariffs and high transaction costs, produces uncertainty, and decreases the perception of the institutional quality and the incentives for rational individuals to partake in economic activities (Berggren, 2003; Graeff and Mehlkop, 2002; DiRienzo and Das, 2015).
There are numerous manners in which the indices in the economic freedom of the world report (EFW) can be perceived, but for the purpose of this paper it will be viewed as a measure of the quality of a country’s institutional and consequently policy environment (Berggren, 2003; Gwartney and Lawson, 2003; DiRienzo and Das, 2015). As it is an index which can measure the consistency of a nations policies and institutions (Gwartney and Lawson, 2003). Literature on economic growth which has focused on this perception of the index emphasised the importance of the rule of law, security of property rights, monetary and price stability, free trade and avoidance of excessive taxes and regulation, (Torstensson, 1994; Knack and Keefer,
1995; Barro, 1996; Barro and Sala-i-Martin, 1995). Where in an economically free society with high quality policies, the core functions of the government are considered to be private property, the enforcement of contracts and freedom to exchange property (De Haan and Sturn, 2000; Gwartney et al., 1996).
As a result, this study will focus on the 2nd and 4Th element of the aspects which measure EFW,
which consists freedom to trade internationally and legal structure and security of property rights. These two elements were chosen as policies regarding these aspects; seem most relevant to dealing with investments in innovation and are regarded as central components to measuring economic freedom and are in line with academic stance which views EFW as measure of a country’s policy quality (Nelson, 1993; Sakakibara and Porter, 2000; Jeffrey et al., 2001; Gwartney and Lawson, 2003).
Legal structure and security of property rights refers to the protection of persons and their rightfully acquired property, which is regarded as a fundamental element to economic freedom and innovation (Sakakibara and Porter, 2000; Gwartney and Lawson, 2003). This section consists of judicial independence, impartial courts, protection of property rights, military interference in rule of law and politics, integrity of the legal system, legal enforcement of contracts, regulatory restriction on sale of real property, reliability of the police and business costs of crime (Fraser Institute, 2018).This is relevant to innovation as business who lack confidence that contracts will be enforced, that newly developed products will be protected and fear of intervention, lack the incentive to produce and thus innovate (Klitgaard., 1998; Gwartney and Lawson, 2003; Dirienzio and Das., 2015).
impact international exchange which consists of; tariffs, quotas, administrative restraints, exchange rate and capital control (Gwartney and Lawson, 2003). For a country to rate highly on freedom to trade low tariffs, large trade sector, efficient administration of customs, a freely convertible currency and few capital controls are needed (Gwartney and Lawson, 2003). This is extremely relevant to innovation as free markets provide access to foreign markets for labour, capital and other inputs in an efficient manner, which allows innovative firms to obtain and combines resources, exit successful ventures and provides for learning opportunities (Jeffrey et al., 2001; Gwartney and Lawson, 2003). This enriches their combinative experiences, especially since it enables firms to blend internal and external knowledge, which can lead to superior levels of innovation (Cassiman and Veugelers, 2006; Tranekjer and Knudsen, 2012; Bradley and Klein, 2016).
These aspects are all proven to promote innovative activities of firms and should thus result in an increase in innovation produced by a country as it is said to strengthen conditions for innovation (Jeffrey et al., 2001; Kiriyama, 2012). Moreover, it has been stated that firms need some threshold of internationalisation and thus access a broad range of markets in order to benefit adequately from their innovations (Kafouros et al., 2008). However, this requires some assurance that information or assets are to be protected, through policies such as intellectual property rights (Dirienzio and Das., 2015). Further emphasising the importance of policies that promote freedom to trade and protect property rights (Dirienzio and Das., 2015). The following section links the concepts analysed above and develops a conceptual framework and corresponding hypotheses.
3. Conceptual Framework and Hypotheses
In this section, we develop our hypotheses on how policy uncertainty affects a country’s innovation output, and how certain country specific factors such as corruption and economic freedom influences this relationship.
Figure: 1 Conceptual Model
3.2. Economic Policy Uncertainty and Domestic Innovation Output
Economic policy uncertainty relates to the uncertain impact of policies set by government on the profitability of the private sector (Nguyen et al, 2017). Policy uncertainty is created by the perceived probability of policy reversal and inconsistency, which in turn affects how conducive a country’s entrepreneurial environment is, since these rules relate to how a firm may operate within a country (Rodrik, 1991; Julio and Yook, 2012; Bhattacharya et al., 2017). Literature on a nations capacity for innovation emphasises the role of the policy environment set by government, specifically relating to policies of intellectual property rights and trade (Nelson, 1993; Sakakibara and Porter, 2000). Highlighting that policies play an important role in influencing a country’s capacity to innovate (Furman, Porter and Stern, 2002)
As previously stated innovation can also be viewed as special investment in long term intangible assets (Ayyagari, Demirgüç-Kunt and Maksimovic, 2014; Bhattacharya et al., 2017). Policy uncertainty and its effect on real investment is therefore a relevant concern for investors and politicians, due to its known negative relationship to investments (Julio and Yook, 2012; Kang, Lee and Ratti, 2014; Gulen and Ion., 2016). Especially, since uncertainty about future government policy is currently abnormally high, hindering economic activity and growth (Bloom., 2013). Policy uncertainty thus plays a significant role in this context, as firms become cautious under conditions of uncertainty and will hold back investments as the value to wait increases, due to the increase in perceived risk (Chen and Funke, 2003; Bloom et al.,2007; Bloom, Draca, and Van Reenen, 2016). Leading to a decrease in investment in innovative activities at a country level, due to the uncertainty of policies.
This is specifically applicable with regards to innovation as there are sunk costs involved, given that innovation is the exploration of unknown approaches and methods which necessitates for significant investments in intangible assets (Holmstrom,1989; Aghion and Tirole, 1994; Manso 2011; Ferreira, Manso, and Silva, 2012). In addition, since investments in innovation is long term it is particularly susceptible to its environment (Bhattacharya et al., 2017). The value of the option to wait is therefore amplified under uncertain policy conditions, since the value of the success of innovative exploration depends on the institutional environment and its corresponding policies, which determine the profitability of this type of investment (Bernanke 1983; Dixit 1989; Furman, Porter and Stern, 2002; Bhattacharya et al., 2017).
As a result, policy uncertainty can be found to negatively impact a country’s innovation output as policy uncertainty leads to firms being unable to determine what policies they must adapt to and how sustainable the policies are that they have to adapt to (Bhattacharya et al., 2017). It
will cause investors to hold their option to invest in innovation due to the potential risks, indicating how policy uncertainty discourages an economy’s incentive to innovate (Bloom, Bond and Van Reenen., 2007; Bhattacharya et al., 2017). Policy analysis, therefore, demands an evaluation of how innovation output varies within countries that have different levels of economic policy uncertainty, in order to further develop an understanding of the influence of policies (Furman, Porter and Stern, 2002). This is what this study aims to do, therefore, leading the construction of Hypothesis 1, as stated below. Followed by a discussion of the moderating role of corruption on Hypothesis 1.
Hypothesis 1: All else being equal, economic policy uncertainty decreases a country’s innovation output
3.3. Moderating Effect of Corruption
There is a close connection between policy of economic activity and corruption (Djankov et al. 2002; Kaufmann et al., 2005). Corruption is said to have a negative impact on public policy and investments and thus is regarded as negatively correlated with a countries economic, regulatory and political environment (Wei, 1999; Zhao et al., 2003; Cuervo-Cazurra., 2006). This condition results from politicians being able to extract more bribes from the private sector by restricting economic activity through policies, at a level that is below socially optimal (Aidt and Dutta, 2005). As politicians have temporary monopoly rights to perform political favours and thus use their position to distort economic policy for their private gain (Aidt and Dutta, 2005; DiRienzio and Das, 2015). This distortion of policy can thus result in an increase of uncertainty regarding economic policies, as distortion is regarded as the “action of giving a misleading account or impression” (Oxford Dictionaries, 2018).
The potential personal gain derived from economically restricting firms through policies, indicates that corruption can lead to an increase in the level of regulation and thus policies, consequentially leading to a decrease in government efficiency, where efficiency refers to the absence of overregulation, uncertainty and wastefulness (Aidt and Dutta, 2005; Anokhin and Schulze, 2009). Corruption can lead to inefficient development of policies, indicating that it can result in an increase in policy uncertainty through inefficiency (DiRienzio and Das,.2015; Montes and Paschoal, 2016). In addition, corruption has also been said to directly lead to ambiguous or uncertain legalisation, which negatively affects trust in the legal system and increases transaction costs of conducting business (Hellman et al., 2000; Dirienzio and Das, 2015).
In these conditions, business struggle to adapt to their environment due to the economic policy uncertainty created by inefficient and ambiguous policies, which are not implemented for the collective good (Anokhin and Schulze, 2009; Dirienzio and Das, 2015; Bhattacharya et al.,2017). This results in a possible decrease in a countries domestic innovation output because of a disincentive to invest in innovation and perform innovative activities (Bhattacharya et al.,2017). Furthermore, it can also lead to the perception of governments being unable to enforce and protect laws, due to the distrust that can be created, further fostering uncertainty (Klitgaard, 1998; Rodriguez et al.,2006; Anokhin and Schulz, 2009). This further emphasizes the potential negative impact that corruption could have on innovation by aggravating economic policy uncertainty, because it can significantly reduce the underlying fundamentals which govern the principles of business, economic and political institutions (World Bank, 2008; Anokhin and Schultz, 2009; DiRienzio and Das,.2015).
Moreover, as previously mentioned corruption affects the credibility of governments commitment to policies (Montes and Paschoal, 2016). This is as a result of governments being able to alter the policy environment, since unilateral domestic policy choices can be easily changed, especially if it is at the expense of the private sector and for private gain by policy makers leading to government inefficiency (Büthe and Milner, 2008). And as policy uncertainty results from inconsistent formulation and application of regulatory and macro-economic policies and/or due to policies that are regarded as normative and do not persevere over time (Rodrik, 1991). One could assume that there would be a higher likelihood of governments having lower credibility regarding commitment to policies in an environment with corruption, which should further increase overall economic policy uncertainty and thus lead do a decrease in innovative activity. Corruption also directly leads to an increased level of uncertainty and transaction costs, which complicates a firm’s ability to capture opportunities from innovative activities in a commercially and sustainable manner (Luo, 2004; Anokhin and Schultz, 2009). Therefore, leading to the development of Hypothesis 2 as stated below, indicating the negative moderating influence corruption should have. For which the role of economic freedom as a moderator on the main relationship H1, will be discussed below.
Hypothesis 2: Corruption has a negative moderating influence on the relationship between a country’s domestic policy uncertainty and innovation output.
3.4. Moderating Effect of Economic Freedom
As mentioned previously, innovation can be regarded as a long-term investment (Bhattacharya et al.,2017). Where governments can alter the environment which investors/businesses face through policies and as investment in innovation increases growth, it is important to indicate towards investors, through policies, that their investments can prosper (Marcus, 1981; Büthe
and Milner, 2008; Ayyagari, Demirgüç-Kunt and Maksimovic, 2014). Therefore, it can be deemed crucial to investigate the quality and consistency of policies that governments set, and its resultant uncertainty, as it can influence the innovation produced in a country and consequentially hamper a nation’s economic development (Graeff and Mehlkop, 2002; Gwartney and Lawson, 2003; Bhattacharya et al.,2017).
Investigating a country’s level of economic freedom is a valid measure of a nations quality and consistency of policies as it promotes incentives for individuals to partake in innovative activities (Farr, Lord, and Wolfenbarger., 1998; Gwartney, Lawson, and Holcombe., 1999; Cole., 2003; Powell., 2003; Graeff and Mehlkop, 2002; Gwartney and Lawson, 2003; DiRienzo and Das, 2015; Heritage Foundation, 2018). In addition, economic policy uncertainty is said to be influenced by economic freedom, as an institutional arrangement that is in conflict with economic freedom is said to produce uncertainty (Graeff and Mehlkop, 2002). Implying that an institutional arrangement which is in line with economic freedom should result in a decrease in economic policy uncertainty, especially since economic freedom deals with policies that influence economic activity (Heritage Foundation, 2018). This demonstrates the connection between economic freedom and economic policy uncertainty, which implies that a higher level economic freedom should reduce economic policy uncertainty.
As aforementioned, policy uncertainty in countries appear due to inconsistent formulation and application of regulatory and macro-economic policies or due to policies that are regarded as normative and do not persevere over time (Rodrik, 1991; Julio and Yook, 2012; Bhattacharya et al., 2017). Therefore, a consistent increase in economic freedom over time, should reduce the perception of economic policy uncertainty, as it can act as a signal to investors and businesses that a country is committed to the policies they have set (Gwartney and Lawson,
2003; Heritage Foundation, 2018). Since a commitment to economic freedom portrays a stable and predictable rule of law (Berggren, 2003). Therefore, a manner in which to indicate that investments can prosper and reduce economic policy uncertainty could be through indicating a commitment to economic reforms such as economic freedom. (Graeff and Mehlkop, 2002; Büthe and Milner, 2008). As it provides favourable conditions for innovators, entrepreneurs and businesses to prosper (Graeff and Mehlkop, 2002; Cole., 2003; Powell., 2003; Büthe and Milner, 2008).
Moreover, an increase in economic freedom also signals an increase in the quality of policies over time further emphasizing how this could decrease uncertainty as policies that effect business activities are showing improvement (Cole., 2003; Powell., 2003; Gwartney and Lawson, 2003; DiRienzo and Das, 2015). Therefore, if policies do change but are in line with economic freedom, they should change in favour of economic activities and thus decrease uncertainty further and result in an improvement of a country’s innovative activity, as the environment will be conducive to such long-term investments (Cole., 2003; Powell., 2003; Gwartney and Lawson, 2003; Bhattacharya et al.,2017). Further indicating another reason as to why a higher level of economic freedom in a country should lead to a decrease in economic policy uncertainty.
Lastly, subtle government interventions that reduce profitability of investments have also become a key concern to investors, where economic freedom is related to less interventionist policies (Berggren, 2003; Büthe and Milner, 2008). Therefore, a commitment to economic freedom could further lead to a decrease in policy uncertainty as polices implemented are less interventionist and thus limit the role of the government, which should cause for less concern for businesses regarding the uncertainty of policies (Gwartney and Lawson, 2003; Büthe and
Milner, 2008). All these aspects have been the motivation and development behind the formation of Hypothesis 3 in this paper, stating the positive moderating influence economic freedom should have. Thereby concluding the development of the theoretical framework, which will be followed by the methodology.
Hypothesis 3: Economic Freedom has a positive moderating influence on the relationship between a country’s domestic policy uncertainty and innovation output.
4. Methodology
4.1. Sample and Data Collection
The objective of this paper is to study the impact of economic policy uncertainty on a country’s innovation output. More specifically, we explore how the extent of corruption and the level of economic freedom, measured in property rights and freedom to trade, within these countries influence the main relationship aforementioned. Control variables have been included in order to remove the effects on the variables included. The control variables being Foreign Direct Investment (FDI), Population and Gross Domestic Product (GDP), as they are said to influence innovation directly (Furman, et al., 2001; Lin and Lin, 2010). The sections below justify the reasoning for which the aforementioned variables were chosen.
This study will be conducted in a quantitative form for which the objective is explanatory. The aim is to establish causal relationships between the construct/ variables and to measure the relationships between the constructs. Secondary, databases will be used in order to explore the effect of policy uncertainty on innovation, for 21 countries across 13 years (2003-2015), resulting in total sample set of n = 268. The final sample is exhaustive of all country’s available in the economic policy uncertainty (EPU) index and includes as many years possible for which
respective countries have available data. The countries consist of developed and developing countries. The developing countries according the World Bank (2018) consisted of; Brazil, Chile, China, India, Korea, Mexico, Russia. The developed countries according to the World Bank (2018) consisted of Australia, Canada, France, Germany, Hong Kong, Ireland, Italy, Japan, Netherlands, Singapore, Spain, Sweden, UK and the US. As a result, the EPU index determines the sample set in this paper. The exact strategical method undertaken to access the statistical significance of the findings will be discussed in detail in the results section, under model specifications.
To substantiate the use of the limiting dataset, we will emphasise the importance of economic policy uncertainty and innovation and the corresponding database (EPU index). As aforementioned, economic policy uncertainty relates to the uncertain impact of policies set by government on the profitability of the private sector (Nguyen et al, 2017). Where innovative activity of firms can be regarded as a long-term investment in intangible assets (Bhattacharya et al.,2017). In addition, innovation, is also considered to be a key driver for the development of economic growth and can thus create for a sustainable competitive advantage in countries (Chen et al., 2014; Lee, O ̈ zsomer, and Zhou, 2015). Indicating its economic value at a country level and not only a firm level. Moreover, economic policies set by governments, and their corresponding uncertainty, can have a significant impact on firms’ innovative activities and consequently a countries economic development (Chen et al., 2014; Lee, O ̈ zsomer, and Zhou, 2015; Nguyen et al, 2017; Bhattacharya et al.,2017). This substantiates and underpins this papers interest in collecting data relating to policies set by governments, and the uncertainty of such policies, and how they in turn effect innovative activities by firms. And thus, potentially influence a country’s economic progress.
In addition, economic policy uncertainty is also regarded as being on the rise, in light of the late financial crisis and other current economic developments and it has been promulgated as responsible for the further slowing down of the world economy, substantiating this papers interest in this particular topic (Bloom, 2013; Nguyen et al., 2017; Fahey, 2016; Hilsenrath, 2018). This subject has also been understudied and databases measuring political indices are scarce, for which the economic policy uncertainty index will be used as a database in this study (Nguyen et al., 2017; Bhattacharya et al.,2017). This index however, while useful and who’s methodology has been used in numerous papers, has been underutilised as a database in prior research due to its limited dataset in the past, nevertheless is exponentially growing in size, indicating it’s growing relevance and use amongst researchers (Nguyen et al, 2017; Bhattacharya et al.,2017; Perić and Sorić, 2018). Making it viable for this paper to use this scarcely used but useful and ever-growing and recognized database.
Moreover, this database consists of emerging and developed markets, making it conducive with regards to having heterogeneity in the data set, and particularly useful for the moderators, as there will be variety with regards to corruption levels and degrees of economic freedom for the selected countries. All these aspects mentioned, validate this papers goal to understand the effect of policy uncertainty on innovation and factors which may moderate it. As a result, substantiating the reasoning for the use of this particular sample set of countries and years determined by the EPU index, which can be considered as limiting. The following section will delve into the justification behind the databases used to measure the variables present in this papers analysis.
4.2. Measures
4.2.1. Independent Variable Economic Policy Uncertainty
We will obtain policy uncertainty data as mentioned previously, from the economic policy uncertainty (EPU) index developed by Baker et al. (2016). As it is an objective indicator for the intensity of concerns about policy uncertainty at a country level (Nguyen et al. 2017). It explores the coverage of policy related uncertainty in leading newspapers for each respective country (baker et al, 2016). It is based on frequency counts in newspapers based on the words “uncertain”, “economic”, “economy” and other terms which are relevant to policies (Baker et al., 2016). This index is regarded as novel and provides information regarding the extent and nature of economic policy uncertainty (Nguyen et al, 2017). It is an index, which allows an objective observation of the amount of uncertainty leading up to and following the passage of economic policies for a cross section of 21countries for a period of 13 years (Baker et al., 2016; Nguyen et al, 2017). The focus will be on the overall level of EPU on a yearly basis as all other databases used in this study are measured yearly. As a result, monthly measures were averaged and transformed into yearly measures.
Other studies have used national elections to measure policy uncertainty as elections are said to be relevant to all policies; fiscal, monetary, trade, social security, industry regulation and taxation (Durnev; 2010; Julio and Cook, 2012; Bhattacharya et al.,2017). Therefore, it has been argued that this form is regarded as a reasonable proxy for overall policy uncertainty. However, policy uncertainty, when the measurement is based on elections, is regarded as being more of a significant indicator of political uncertainty, because it is based on uncertainty of government actions (Pastor and Veronesi, 2012). Whereas, economic policy uncertainty is regarded as a proxy which relates to the impact of policy uncertainty set by governments which influences
the environment of the private sector, making it more relevant in the context of our study with regards to innovation being seen as a long-term investment by firms (Pastor and Veronesi, 2012; Bhattacharya et al.,2017). Lastly, election years do not capture the heterogeneity in policy uncertainty that could occur between elections, which is regarded as a significant omission due to their infrequent occurrence, further justifying the use of the EPU index in this paper (Nguyen et al, 2017).
As a result, this study will use EPU developed by Baker et al. (2016) as it is regarded as most relevant with regards to the context of this study. In addition, this method is regarded as a good proxy to measure uncertainty as it measures all forms of uncertainty in an economy (Nguyen et al, 2017). This is especially relevant on the assumption of the media being able to gauge any uncertainty indicated by market outcomes, professional economists and political debates (Nguyen et al, 2017). Furthermore, this databases validity of use is increasing significantly due to its ever-expanding dataset, indicating its growth in use amongst researchers.
4.2.2. Dependent Variable Innovation
Data with regards to a country’s level of innovation produced will be derived from the world bank data base developed by the world intellectual property organization (WIPO). WIPO has developed a patent report based on worldwide patent activity (World Bank, 2018). These consist of world-wide patent applications filed through the patent cooperation treaty procedure or via a national patent office dealing with exclusive rights for an invention (World Bank, 2018). An invention referring to a new way of doing something or a new technical solution to a problem (World bank, 2018). These applications are measured on an annual basis. This paper specifically focuses on resident patent applications in an attempt to measure domestic
innovation output (Markatou and Vetsikas, 2015). This paper utilizes The World Bank’s data base as it has been widely used amongst many studies (DiRienzo and Das, 2015).
Some scholars however have questioned the validity of using the aforementioned method to quantify the level of innovation in a country as its focus is based on a narrow aspect of innovative activity, which could potentially exclude product modifications as well as process innovations or activities (kalantaridis and Pheby, 1999). In addition, other researches have stated that this measure is more appropriate for measuring inventions as opposed to innovation as many patented ideas do not become viable products (Shane, 1992). Despite these limitations it is still regarded as a significant aspect that measures the level of technological activity in a country (Dakhli and De Clercq, 2004). Because numerous crucial conditions need to be met in order to file patents, for example, the invention needs to be useful, novel and showcase an innovative step that is not regarded as common with regards to technical innovation (Evenson 1984).
Furthermore, the use of patenting as a measure of innovation has become a common measure with regards to literature in innovation, signifying the reasoning behind this paper use of patent data (Hsu, Tian, and Xu, 2014; Sapra et al, 2014; Bloom et al. 2016). Moreover, literature has indicated that patents have the ability to predict aggregate stock market value, firm profitability and economic growth which validates the use of patent data to evaluate the influence of economic policy uncertainty (Hsu, 2009; Hirshleifer, Hsu, and Li, 2013; Kogan et al., 2017). Lastly, unlike other measurements of innovation like the Global Innovation Index (GII), patent data has information dating back many years, making it more suitable for a longitudinal study.
4.2.3. Moderating Variables Corruption
A country’s level of corruption can be measured by the corruption perception index (CPI), which has been developed by Transparency International Berlin. This index ranks 180 countries on their perceived level of public sector corruption according to experts and business professionals (Transparency International, 2018). The scale ranges from 0-10, where zero indicates a high level of corruption and 10 the lowest level of corruption (Transparency International, 2018). A country is included in the CPI if there are at least 3 surveys available for the assessment of its level of corruption, which includes assessments of the extent of illegal behaviour in a country in general (Graeff and Mehlkop, 2002). More specifically it “measures the degree to which officials and politicians are believed to accept bribes, or illicit payments in public procurement, embezzle public funds, or commit offenses” (DiRienzo & Das, 2015). The latter definition is used in this paper.
The CPI is the most known corruption indicator that is used in academic literature, however it has also been criticised (Berg, 2001). Sampford et al. 2006 state that there are challenges present in the methodology relating to the construction of measures of a country corruption. This paper also notes that the scaling used to develop the CPI data is problematic as it is based on a 10-point discrete scale that is dependent on a range of variables which are mostly ordinal (Sampford et al. 2006). In addition, it is mentioned that the data shows an over-reliability over time, due to the data being dependent on a small number of surveys that do not vary significantly over the years (Sampford et al. 2006). It is further mentioned that CPI is primarily based on the assessment of experts and not based on direct measures of household and business experiences, implying that it does not measure all aspects of corruptions in a country (Jones et
al., 2007). Lastly, it has also been stated that the existing measures do not incorporate quantitative sources of data into the perception-based data sources (Sampford et al., 2006).
The criticisms on the methodology used in this index is universal as they have also been used to criticise other methodologies that measure corruption, however, it is used in this paper as it is regarded as a well-recognized and robust measure of corruption (Berg, 2001; Serra, 2006).
Economic Freedom
The indices of economic freedom that have been used recurrently by researchers are the economic freedom of the world index (EFW) developed by the Fraser institute, the index of economic freedom (IEF) developed by the heritage foundation and the wall street journal (Miller et al., 2014; Gwartney et al., 2017). The IEF covers more countries than the EFW, however, the IEF is said to be based on procedures and measurements which have been recognized as being less precise and less transparent than procedures committed by the EFW (Gwartney and Lawson, 2002). Moreover, it is also viewed as less valuable to researchers who wish to analyse the changes of economic freedom across time period (Gwartney and Lawson, 2002). Therefore, this study will use the EFW to measure the level of economic freedom, for a number of reasons; it covers a time span of 13 years, includes all countries required for this paper and all the procedures used are regarded as being more transparent (Gwartney and Lawson, 2002).
As mentioned previously the EFW measures the level of economic freedom in five areas covering different aspects of economic freedom (Gwartney et al., 2017). Within the five major areas there are 24 components in the index which themselves are made up of sub components, in total there are 42 distinct variables (Gwartney et al., 2017). These are all derived from third
party sources, such as the international country risk guide, the global competitiveness report and the world bank in order to create subjectivity and transparency (Gwartney et al., 2017). It is an index which is updated regularly, and each update takes into account the revision in the underlying data (Gwartney et al., 2017). Each component is based on a scale from 0-10 (Gwartney et al., 2017). Where higher ratings are indicative of institutions and polices more consistent with economic freedom (Graeff and Mehlkop, 2002).
This paper will use two areas of economic freedom most relevant to innovation but not highly correlated to corruption, namely, freedom to trade internationally, legal structure and security of property rights (Nelson, 1993; Furman, Porter and Stern, 2002). Which are also viewed as principal aspects to measuring EFW (Nelson, 1993; Sakakibara and Porter, 2000; Jeffrey et al., 2001; Gwartney and Lawson, 2002). These measures will be aggregated as a representation of relative economic freedom in this paper.
4.2.4. Control Variables GDP
Prior research has shown that a country’s level of GDP (Gross Domestic Product) can influence country’s level of innovation, therefore this variable needs to be controlled (Furman, et al., 2001). This paper will use GDP Per Capita, which is available for the respective countries using the World Bank Data base. The world bank has again been chosen due to its frequent use by researchers and its considered reliability (DiRienzo and Das, 2015). GDP directly affects innovation as it has the ability to take into account the ability of a country to translate knowledge stock into a state of realized development (Furman, et al., 2001). As a result, it is regarded as an aggregate control for a country’s technological capabilities (Furman, et al.,
2001). Where the mean value for this variable is 2.296, with a standard deviation of 3.653, signifying how countries vary with regards to the amount of GDP they produce (Table 1).
Population
As the sample set of chosen countries varies significantly in size, this paper will differentiate between the scale-based differences using their respective population (POP) levels as a control (Furman, et al., 2001). The total population of a country has been chosen as a control, since innovation is affected by the number of people in countries. Since, larger countries have been identified by incurring more extensive exchanges of various resources at multiple levels and consequentially they are said to generate more patents (Dakhli and De Clercq, 2017). Again, the world bank data base will be used, for the same reasons mentioned previously. Where the mean value is 188417288.304 with a standard deviation of 363762959.483 (Table1). Indicating the significant variety in population between countries in this data set, which was to be expected as we have countries such as Italy and China.
FDI
Foreign Direct Investment (FDI), has been chosen as a control variable, as it is known to accelerate the process of innovation at a country level (Lin and Lin, 2010). FDI can accelerate innovation and benefit innovative activity in the receiving country through spill overs such as; skilled labour, reverse engineering, turnovers, demonstration effects and supplier customer relationships (Kui-yin and Lin, 2004). FDI is the sum of equity capital, other long-term capital, reinvestment of earnings and short-term capital as indicated by the balance of payments (World Bank, 2018). Where this paper discusses the FDI net inflows; as a percentage of GDP and where the net inflow is measured it uses new investment less disinvestment (World Bank, 2018).
5. Results
5.1. Model Specification
The data was analysed using a fixed effects model with a maximum likelihood estimation, where the model controls for unobserved heterogeneity with regards to time-invariant variables, using SPSS (Longford, Bryk and Raudenbush, 1993; Lavie, Kang and Rosenkopf, 2011). This model was chosen after a Hausman test (1978) was conducted in STATA, which indicated that the fixed effects model was superior to using the random effects model (Prob>chi = 0,0002). This paper uses a hierarchical model using cross section time series regression with country fixed effects. Therefore, the reported model explains within country variation in innovation due the inclusion of fixed effects.
Due to the data being panel data, as we have repeated measures across 13 years for 21 subjects (countries), apprehensions arise with regards to autocorrelation of errors within cross sections, which could reduce standard errors and overstate levels of significance (Lavie, Kang and Rosenkopf, 2011). Therefore, we conducted a Durban Watson test through the standard OLS regression method, which indicated that our model has a form of autocorrelation (Durban Watson = 0,345) (Durbin and Watson, 1950; Bhargava, Franzini and Narendranathan, 1982). Autocorrelation was treated through the inclusion of first-order autoregressive structure with homogenous variance AR (1) in the tested models. Therefore, this co-variance structure was chosen as it best fits the model according to Akaike (1969) information criteria, with assumption being that errors are correlated across adjacent years.
A hierarchical analysis was performed, where 8 models were tested on the basis of creating stability and robustness in the analysis. Where Model 1 serves as the baseline model, which consist solely of control variables. Model 2 adds the independent variable and thus tests the
