Intellectual Property protection and enforcement in explaining the level of Inward Foreign Direct Investment flows

(1)

Master Thesis

Intellectual Property protection and enforcement in explaining the level of

Inward Foreign Direct Investment flows

MSc in Public Administration; Economics & Governance

Leiden University

Academic Year 2019-2020

Author: Estelle van der Velden (s2660350)

Supervisor: Dr. P. van Wijck

Date: June 9

th

2020

Word count: 23.174

(2)

2

Abstract

More countries produce and consume products that result from creative activity and innovation, but the current era of globalization has created the production and/or sale of unauthorized copies to supply the newly generated demand. Intellectual property institutions can determine the ability of firms to protect their intellectual assets against infringement and to exploit their assets in a host country. But in how far does the granted level of IP protection and enforcement function as a locational determinant for foreign direct investment?

This research aims to provide insight into the gap between the levels of intellectual property protection and enforcement strength within countries and to show the importance of reducing this gap in order for countries to attract foreign direct investment. Great rights on paper, measured by protection strength level, does not automatically mean that the rights are easily enforced. In general, the level of protection strength has been internationally harmonised to a certain level whereas the level of enforcement strength falls behind in most countries.

A panel dataset of 41 WTO member countries over 10 years was constructed. The results of one-way fixed effects regressions support the expectation that higher levels of IP enforcement strength lead to higher levels of IFDI flows. An impact assessment of a simulated harmonisation has been conducted to show that there should be more focus on harmonising the levels of IP protection and enforcement strength within WTO member countries instead of between them in order to encourage FDI.

(3)

3

Chapter 1 Introduction

Protection and enforcement of intellectual property (IP) became a major issue of multilateral negotiations during the Uruguay Round of the General Agreement on Tariffs and Trade (Nunnenkamp & Spatz, 2003). Forthcoming from the Uruguay Round, the World Trade Organization (WTO) established the Agreement on Trade-Related Intellectual Property Rights (TRIPS) that contains a set of minimum standards for IP protection and enforcement (WIPO, 1994; WTO, 2020a; 2020b;).

The role of the strength of IP protection and enforcement in attracting or discouraging foreign direct investment (FDI) has also received increasing attention in the literature. This increased attention can be explained as follows: IP laws and institutions can determine the ability of firms to protect their intellectual assets against IP infringement and to exploit their IP assets in a host country (Papageorgiadis, Xu & Alexiou, 2019). The volume of trade in goods protected by IP is becoming increasingly significant as more countries produce and consume products that result from creative activity and innovation (Seyoum, 2006). Furthermore, the globalization of markets has created (opportunities for) the production and/or sale of unauthorized copies to supply the newly generated demand.

Multinational enterprises (MNEs) have multiple options regarding the means by which they can serve foreign markets. Firms can export to a particular country or region or may decide to undertake FDI, which requires selecting the country to invest in (Maskus, 2005). It has long been recognized that a host country’s IP system has an important effect on the strategies and decision-taking of multinational enterprises, because it shapes their ability to appropriate returns on their investments in research and development (R&D), and innovation (Maskus, 2000; Papageorgiadis, Cross & Alexiou, 2014).

The OLI paradigm by Dunning (2008; 2001; 2000) claims that the extent and geography of foreign production and investment undertaken by MNEs is determined by the interaction of two sets of interdependent variables: ownership and locational advantages. These determine the locational attractiveness of countries as recipients of FDI. Ownership advantages suggest that the greater the competitive advantages of the investing firms, relative to those of the firms located in the host country, the higher the extent of foreign production. However, ownership advantages do not explain why the foreign presence should be established through FDI rather than exports. Therefore the locational advantages have to be taken into account since they determine whether the ownership advantages will result in FDI (Dunning, 2008). These advantages arise due to institutional differences such as the levels of strength of IP protection and enforcement in host countries, but also due to other host country characteristics forming an overall pro-competitive business environment such as countries’ level of trade openness and their levels of gross domestic product (GDP).

It can be suggested that foreign investors seek locations where primarily the institutional environment facilitates the development of their global firm-specific advantages (Bevan, Estrin & Meyer, 2004; Dunning & Zhang, 2008). The expected relation between IP protection and enforcement and levels of IFDI flows comes from the possibility of uncompensated R&D spill overs that affect firms’ investment decisions (Fan, Gillan & Yu, 2013).

Intellectual property rights play an important role in the processes of undertaking FDI, but their importance depends on how they are protected and enforced. Intellectual property (IP) refers to creations in the mind. Intellectual property protection refers to the protection of granted rights to creators of knowledge to prevent others from using their created assets and to use that right to negotiate payment in return for others using them (WTO, 2020a). This study is specifically interested in the effect of the strength of IP protection and enforcement on whether countries are being chosen as recipients of FDI. Therefore we focus on inward FDI flows that record the value of cross-border direct investment transactions received by the reporting economy during a year measured in US dollars (World Bank,

(7)

7 2020). It represents transactions that increase the investment that foreign investors have in enterprises of a specific industry located in the reporting economy.

Although the TRIPS Agreement is aimed at harmonising countries levels of IP protection and enforcement strength through the establishment of minimum standards, there still exist large differences in countries’ levels of IP protection and enforcement strength. Not only are there differences expected between the WTO member countries, also within countries it can be suggested that there exists a gap between the level of IP protection strength on the one hand and the level of IP enforcement strength on the other hand, especially since TRIPS does not include specific obligations on the practical enforcement of the IP laws (OECD, 2008). This study examines whether high levels of IP protection strength and IP enforcement strength respectively increase the locational advantages of a host country and therefore result in a higher level of IFDI flows.

1.1 Research question and goals

1.1.1 Research question

Former empirical research that examined the effect of IP protection on the level of FDI has led us to believe that the level of IP protection strength only captures the granted IP rights on paper. However, well protected IP rights on paper do not automatically mean that they are easily enforced. Therefore this study will include the component of practical enforcement of the laws concerning intellectual property protection. Moreover, it is examined whether other host country characteristics – GDP level and level of trade openness – have an effect on the level of IFDI flows. These host country characteristics may contribute to in how far it is perceived by MNEs that the economic system and regulatory environment of a host country enable them to exploit their ownership advantages profitably and thus are likely to be chosen as location for FDI.

As the case description of Chapter 2 will make clear, there is an observed divergence between a country’s statutory strength of IP protection (Law on the books) and the actual level of enforcement (Law in practice). Therefore it can be questioned whether MNEs base their decision to undertake FDI to a higher degree on the strength of IP enforcement than on the strength of IP protection. This thesis will shed light on the expected difference between the level of IP protection strength and the level of IP enforcement strength in explaining the level of IFDI flows. The focus of this thesis is on finding out whether and in how far the observed divergence in countries’ levels of IP protection and enforcement strength causes the observed divergence in countries’ level of IFDI flows. This is the reason why this thesis aims to answer the following research question:

What are the effects of the levels of Intellectual Property (IP) protection and enforcement strength of host countries on their levels of Inward Foreign Direct Investment (IFDI) flows? In order to answer this research question a panel dataset will be constructed. The use of panel data makes it possible to analyse the differences between countries, as well as differences over time. The research question will be answered by performing one-way fixed effects regression. One of the advantages of fixed effects regression is that it is able to control for all time-invariant country characteristics and therefore can control for stable country-specific omitted effects. To improve the precision of the estimates, the regressions control for two other possible factors influencing the level of IFDI flows: GDP level and level of trade openness.

(8)

8 Building on the prior literature, this thesis seeks to provide new insights into the role of IP protection and enforcement as a locational determinant for FDI. It serves four main goals:

1. To identify the relative importance of the level of IP protection strength in explaining the level of IFDI flows;

2. To identify the relative importance of the level of IP enforcement strength in explaining the level of IFDI flows;

3. To formulate some conclusions with respect to the interacting effect of the levels of IP protection and enforcement strength on the level of IFDI;

4. To conduct an impact assessment of the potential benefits of a harmonisation of countries’ levels of IP protection and enforcement strength.

This thesis makes use of indexes that measure the levels of IP protection and enforcement strength specifically for patent rights. However, since the indexes that are measuring the key independent variables have been applied by former empirical research to multiple categories of IP rights, this research will be focused on the protection and enforcement of IP in general.

1.2 Social and academic relevance

Identifying the size of the gap between the granted IP rights on paper, captured by the level of IP protection strength, and the actual enforcement of these granted rights, captured by the level of IP enforcement strength, will help policymakers at the national and international levels to design and implement effective policies that strengthen national IP systems. By increasing the level of IP enforcement strength, it is suggested that countries will enhance economic benefits, such as greater levels of foreign direct investment (Papageorgiadis, Wang & Magkonis, 2019). In fact, low levels of enforcement strength are commonly identified as an impediment to international trade and investment (Papageorgiadis et al., 2014). Therefore, a proper understanding of the relationship between IP protection and enforcement and IFDI flows will help firms and governments devise appropriate IP policy to encourage the growth and expansion of FDI (Seyoum, 2006). Moreover, industries with above-average use of IP, especially patents, are already known to make a greater contribution to GDP and external trade (EPO, 2017). This study shows that further improvement is however possible. The current international fragmentation between countries’ levels of IP protection and enforcement strength creates limitations that hinder cross-border trade and investment in IP- and technology-intensive industries.

Empirical analyses have not yet consistently confirmed the positive relationship between the strength of IP systems and IFDI (Seyoum, 2006). This is partly due to different operationalizations of the strength of national IP systems. Whereas most empirical research focuses on the statutory IP protection of countries that captures only the protection of intellectual property on paper, it can be suggested that its positive effect on IFDI flows is overestimated. The academic relevance of this thesis is formed by the additional focus on the enforcement of rights protecting intellectual property.

1.3 Outline

The next Chapter provides insight into the observed incongruence between countries’ strength of IP protection, existing of minimally the obliged standards of the TRIPS Agreement, and the strength of the enforcement of these laws in practice. Chapter 3 will give an overview of theories explaining the mechanisms through which IP protection and enforcement strength are expected to influence the level

(9)

9 of IFDI flows. Chapter 4 describes the techniques that will lead to answering the research question, and explains the construction of the panel dataset. Chapter 5 will present the results from the empirical analysis, while Chapter 6 will answer the research question.

(10)

10

Chapter 2 Case-description: Gap between Law on the books and Law

in practice

2.1 Introduction

The WTO’s TRIPS Agreement, negotiated during the 1986-94 Uruguay Round and came into force on 1 January 1995, forms the most comprehensive multilateral agreement on IP (WTO, 2020a). The Preamble to the Agreement describes that there was need for new rules and disciplines in order ‘to reduce distortions and impediments to international trade, and taking account of the need to promote effective and adequate protection of intellectual property rights, and to ensure that measures and procedures to enforce intellectual property rights do not themselves become barriers to legitimate trade’ (WIPO, 1994). The need for harmonisation of IP protection and enforcement has increased because the strength of countries’ IP protection and enforcement systems affects international investments by influencing the level and extent to which MNEs consider investing in and transferring advanced technological assets to a host country.

2.2 Law on the books: International IP legislation

Intellectual property (IP) refers to creations in the mind (WTO, 2020a). Intellectual property rights refer to the granted rights to creators of knowledge to prevent others from using their created assets and to use that right to negotiate payment in return for others using them. As Hayek (1997) pointed out: ‘’…in the case of the property of intangible assets, once made (literary works or discoveries) can be easily reproduced and unlimited, so it will only be through the law they will become scarce and their production will be encouraged.’’

IP rights take a number of forms: patents, copyright, trademarks, industrial designs, geographical indications and trade secrets (WIPO, 2020). The distinction between protection and enforcement of IP rights is made since well protected rights on paper does not automatically mean that they are easily enforced.

The WTO’s TRIPS Agreement is an attempt to narrow the gaps in the way that IP rights are protected and enforced around the world, and to establish common international rules guiding these principles. It has created minimum standards of protection and enforcement that each member country has to apply to the intellectual property held by nationals of fellow WTO members. However, since the Agreement only covers minimum standards of protection which should be provided by member states, IP protection and especially the enforcement remains national in scope in the sense that WTO members are able to tailor their national approaches. Part I of the Agreement covers the general provisions and basic principles (WIPO, 1994). The basic principles of the Agreement exist of the following three: national treatment, most-favoured-nation and balanced protection (WTO, 2020a). National treatment refers to treating foreign nationals no less favourably than one’s own nationals. Most-favoured-nation holds no discrimination among nationals of trading partners. Moreover, the Agreement has an additional general objective which states that IP protection should contribute to technical innovation and the transfer of technology.

Another part of the Agreement covers enforcement obligations and procedures to make sure the protection of IP rights to be meaningful (WIPO, 1994). These obligations are specifically aimed at the procedures for enforcement and do not cover the minimal required strength of enforcement.

(11)

11 Whereas the TRIPS Agreement has harmonised to a great extent the minimum standards of statutory protection for patents offered by countries (Law on the books), the same cannot be said for enforcement levels (Law in practice). This led to levels of IP enforcement strength that are often inadequate and too far behind on the level of IP protection strength.

2.3 Incongruence Law on the books and Law in practice

Effective national IP systems exist of two components: (a) the strength of Law on the books, and (b) the strength of the enforcement of Law in practice (Papageorgiadis & McDonald, 2019). Whereas the patent law protection in developed and developing WTO member countries became more harmonized as a result of the entering of the TRIPS Agreement, large differences remain in the strength of enforcement. The Agreement set clear requirements for the inclusion of specific enforcement procedures in the legislative frameworks of countries, but it did not set obligations on how effectively IP law should be enforced in practice (WTO, 2020b). Therefore, although the TRIPS Agreement provoked strong IP systems in countries, these laws may not be enforced in practice by public enforcement agents. The statutory level of a country’s protection of IP does not indicate a country’s adherence and practical enforcement (Sweet & Maggio, 2015). As a result there are still wide differences in the levels of IP enforcement strength between countries (Papageorgiadis et al., 2014).

The post-TRIPS era is characterized by converged formal IP protection regimes of countries since governments have sought to improve and harmonize book-law protection levels to align them to international standards and norms as part of their obligations since they signed the TRIPS Agreement (Papageorgiadis et al., 2014). WTO member countries classified as developed countries were required to adopt all TRIPS legal provisions by the 1st_{of January 1996; developing and transition countries by} the year 2000; and least developed countries have been granted an extension to comply with TRIPS by the 1st_{of July 2021 (WTO, 2013). Moreover, countries are free in the level of strength of IP protection} they apply at their national level as long as they apply the minimum standards. Therefore, the strength of statutory protection can be expected to still highly influence MNEs activity. Moreover, problems associated with the enforcement of book-law in practice means that MNEs continue to experience difficulties in upholding their patent rights around the world and countering the threat of illegal imitation and infringement (OECD, 2008). Since it seems plausible that IP protection and enforcement play a crucial role in determining in which host countries MNEs decide to invest in, this study distinguishes between the strength of IP protection and the strength of IP enforcement to be able to say something about the influence of patent system strength on the inward level FDI flows of countries.

We can highlight two new types of IP systems that emerged after the implementation of the TRIPS Agreement (Papageorgiadis & McDonald, 2019). Before the implementation of the Agreement, countries were thought to fall either in the category of countries with low qualities on both Law on the books and Law in practice, or in the category of high qualities on both indicators. The first new category relates to countries offering IP systems with high quality Law on the books but low quality Law in practice. These are countries that fully implemented TRIPS, but the effectiveness of enforcement by third party institutional actors is weak. The second category indicates countries which have poor quality Law on the books but high quality Law in practice. This category refers to countries where there is effective enforcement of IP rights, but who do not comply with the TRIPS IP Law on the books requirements or have not adopted other significant IP Law provisions.

(12)

12

Figure 1: The four post-TRIPS contextual categories of IP systems

(Source: Papageorgiadis & McDonald, 2019)

In countries where both the quality of Law on the books and the quality of Law in practice are low, firms are likely to face significant infringement of their IP. Firms in such locations face high transaction costs and risk when seeking to determine what legal protection exists for their IP and experience difficulties in enforcing IP. In countries where both the strength of protection and enforcement are high, firms are ensured effective protection and enforcement of their IP. Countries characterized by both strong IP protection and enforcement indicate locations where MNEs are likely to face low transaction costs and risks associated with the transfer of IP rich assets in the context of establishing and protecting their IP (Papageorgiadis & McDonald, 2019). Countries characterized by high IP Law on the books but low IP Law in practice provide for easy establishment of IPRs of firms, but lead to high costs and risk for firms that attempt to enforce their rights. The last group of countries having systems with low IP Law on the books and high IP Law in practice are not common. A reason for countries to not adhere to IP treaties and strengthen their IP legal systems is to protect domestic interests.

However, this incongruence between Law on the books and Law in practice is also visible at the supranational level of the EU. In November 2017, the European Commission (EC) communicated a Guidance on certain aspects of Directive 2004/48/EC on the enforcement of intellectual property rights (EC, 2020). Whereas the Directive achieved the objective of approximating the legislative systems of the member states for the civil enforcement of IP, the measures, procedures and remedies laid down in the Directive are not implemented and applied in a uniform manner among the member states (EC, 2017). This can be explained by the Directive providing for minimum harmonisation leading to a lack of uniform interpretation of the provisions (EC, 2016). Moreover, already existing differences in national civil law proceeding and judicial traditions increase these disparities. The Guidance serves in this study only as an indicator for the existence of a gap at the supranational level between nations in their strength of IP protection and enforcement.

(13)

13

2.4 Measurement Law on the books and Law in practice

This chapter emphasizes the existence of incongruence or gaps between WTO members’ national levels of statutory IP protection – Law on the books – and their actual adherence to these laws– Law in practice. However, existing empirical research on the relation between countries’ IP system strength and their levels of IFDI flows focuses primarily on the Law on the books aspect whereby no distinction is made between a host country’s level of protection strength and its level of enforcement strength. Voigt (2013) emphasizes the importance for indices measuring institutions to distinguish between how an institution is specified in the Law on the books and how agents in administrative systems implement and enforce the Law in practice. IP Law on the books can be understood as the availability of IP law conferring property rights and formal legal process for enforcement as these appear in the status of a national jurisdiction. IP Law in practice refers to the effectiveness with which third party governmental institutional actors in a country enforce allocated property rights (Papageorgiadis & McDonald, 2019). In most countries, IP treaties changed the context of these two institutional underpinnings of IP systems. However, most existing literature uses one single index to quantify the quality of a country’s IP system (Papageorgiadis & McDonald, 2019). However, as Figure 1 already pointed out, the post-TRIPS era is characterized by two new categories of IP systems whereby the divergence between a country’s quality on IP Law on the books and IP Law in practice is emphasized. By using only one index as a measure of the quality of a country’s IP system, you implicitly assume that both Law on the books and Law in practice are embedded in the selected index. The concern with this approach is that this single IP index does not take into account the different categories on the quality of Law on the books and the quality of Law in practice; it captures either the quality of Law on the books or the quality of Law in practice. The ability to compare the effects of both a country’s quality on IP protection – Law on the books – and the effects of a country’s quality on IP enforcement – Law in practice – is created by making use of two separate indices, one measuring the level of IP protection strength and one measuring the level of IP enforcement strength. Examination of the joint effect of protection and enforcement strength would require an interaction between the two separate IP indices whereby both aspects of the IP institutional system – protection and enforcement – are captured.

The Ginarte and Park (1997) index is used as a measure of a host country’s level of IP protection strength. As Park and Lippoldt (2008) pointed out, the Ginarte and Park index was ‘designed to provide an indicator of the strength of patent protection, not the quality of patent systems’ (p.761). This becomes clear for example from the fact that the index ascribes value to the signing of the Paris Convention for Protection of Industrial Property, but this does not indicate a country’s adherence. However, the Ginarte and Park index is effective for measuring the level of statutory IP protection strength in host countries and is therefore used to measure the strength of the Law on the books aspect, hereafter called ‘IP protection strength’.

The index created by Papageorgiadis, Cross and Alexiou (2014) measures the strength of a national patent system and places particular emphasis on the effectiveness of enforcement practices. Whereas the TRIPS Agreement on the international level and the IPRED on supranational level have harmonized to a great extent the minimum standards of legal protection for patents offered by countries, the same cannot be said for enforcement levels. Therefore, the patent system strength index by Papageorgiadis, Cross and Alexiou (2014) is used in order to measure the strength of the Law in practice aspect, hereafter called ‘IP enforcement strength’.

(14)

14

2.5 Summary

The WTO’s TRIPS Agreement attempts to harmonise the way IP rights are protected and enforced around the world, and sets minimum standards of protection and enforcement that each member has to give to the intellectual property held by nationals of fellow WTO members. This statutory level of a country’s protection of IP – Law on the books – does not indicate a country’s adherence and practical enforcement – Law in practice (Sweet & Maggio, 2015). As a result there are still wide differences in the levels of IP enforcement strength between countries (Papageorgiadis et al., 2014). However, since signatories of the TRIPS Agreement are free in the level of strength of IP protection they apply at their national level, as long as they apply the minimum standards, the strength of statutory protection is expected to still highly influence MNEs activity.

This thesis is especially interested in the two new types of IP systems that emerged after the implementation of TRIPS and the effects of these types on the level to which host countries are attractive to foreign investors, measured by their level of IFDI flows (Papageorgiadis & McDonald, 2019). Before TRIPS, countries were thought to fall either in the category of countries with low qualities on both Law on the books and Law in practice, or in the category of high qualities on both indicators. The current era of globalization has resulted in firms choosing deliberately where to invest based on whether they believe their investments will be made worthwhile. Economic returns on investments are mainly determined by a country’s level of protection of a foreign investor’s IP rights and its level of enforcement in case of infringement caused by for example imitation of the protected goods. Therefore, this thesis argues that a host country’s level of IP protection and enforcement strength will have an effect on in how far firms are able to exploit their ownership advantages and thus on the attractiveness of countries as investment locations for foreign investors.

(15)

15

Chapter 3 Theoretical framework

3.1 Introduction

New technologies, products, and services are created by human intellectual efforts which result in intellectual assets, pieces of information that may have economic value if put into use in the marketplace (Maskus, 2000). To the extent that their ownership is recognized, such assets are called intellectual property. The market structures and the legal rights that permit their owners to control their use determine the economic returns on intellectual property. These legal rights are called intellectual property rights. Intellectual property (IP) rights can be divided into a number of groups: copyright, patents, trademarks, industrial designs, geographical indications and trade secrets (WIPO, 2020). The mechanisms by which IP rights operate vary across these functional areas and their importance differs across sectors (Maskus, 2000).

Intellectual property needs to be regulated because it leads to market failures. It shares the two most important characteristics of a public good (Maskus, 2000). First, intellectual property is non-rivalrous: one person’s use of it does not diminish another’s use. Second, intellectual property may be non-excludable in the sense that it may not be possible to prevent others from using the information without authorization. If an intellectual asset is potentially valuable but easily imitated or used by others, there will be free riding. In turn, there may be no incentive to incur the costs of creating intellectual property. This is not desirable for society and therefore society has an interest in providing defined IP rights. To motivate innovation, governments protect IP rights to ensure that inventors can profit from inventing. However, a too strengthened protection may limit the dissemination of new ideas. Therefore, a trade-off is set in place between on the one hand allocative efficiency referring to access to protected goods, and on the other hand dynamic efficiency referring to the incentives for innovation. This problem is partly addressed by for example allowing patents to expire after a given period. In this way, both the goals of allocative efficiency and dynamic efficiency are achieved. IP rights can be seen as having roughly two main goals: they should be accessible and provide incentives for innovation. The strength of the IP regime determines which goal predominates. Weak protection and enforcement serves the goal of accessibility but leads to insufficient incentives for innovation.

Inward foreign direct investment (IFDI) records the value of cross-border direct investment transactions received by the reporting economy during a year measured in US dollars (World Bank, 2020). It represents transactions that increase the investment that foreign investors have in enterprises of a specific industry located in the reporting economy.

More than half of global investors choose (R&D) investment locations based on three factors: ‘lower R&D costs, availability and quality of local R&D labour, and intellectual-property protection’ (A.T. Kearney, 2005). The attribute of intellectual property protection accounts for roughly 50 percent in determining (R&D) investment locations. Local market size is considered the most important locational determinant in 28 percent of the cases. Another important factor shaping the economic and institutional context is the openness to international trade which accounts for 25 percent of the most important attributes in determining R&D investment locations.

IP rights can be seen as having roughly two main goals: they should be accessible and provide incentives for innovation (Maskus, 2000). The levels of IP protection and enforcement strength determine which goal predominates. Weak property rights, characterized by low levels of protection and enforcement strength, serve the goal of accessibility but lead to insufficient incentives for innovation.

(16)

16 One important argument for low levels of IP protection and enforcement strength is monopolistic behaviour that high levels of protection and enforcement strength permit by deterring entry into an industry (Gould & Gruben, 1996). Technology-importing countries may prefer low levels of IP protection and enforcement strength as a form of strategic trade policy in an open economy. This allows for uncompensated imitation and copying of foreign products and technologies. This may increase the imitative and adaptive capabilities and this underlies the resistance to an increase in the level of protection and enforcement strength in many developing countries (Maskus, 2000).

High levels of protection and enforcement strength create incentives to create intellectual property but leads to insufficient access. An important argument for high levels of protection and enforcement strength is that inadequate levels of protection and enforcement strength limits incentives for trade and inward FDI (Mansfield, 1994).

Mansfield (1994) finds that in deciding whether a host country’s system of IP protection and enforcement is adequate, firms are interested in the answers to three broad questions: (1) Can the country’s laws protect their technology? (2) Is there an adequate legal infrastructure in the country? (3) Do the relevant government agencies in the country enforce the laws and provide equal treatment to foreign firms?

3.2 Locational determinants of IFDI: OLI paradigm

3.2.1. Introduction

Doubt can be placed on the effect of the level of IP protection strength on the level of IFDI flows since economies with relatively strong legal IP regimes could fail to adequately enforce these rights since enforcement is not always reflected in IP indices (OECD, 2008). Dunning (1994) emphasizes the need for governments to pay special attention to their regulatory environments in an era of globalization, because different levels of IP protection and enforcement strength across countries may affect where MNEs decide to locate. Factors that function as locational determinants of FDI are part of the OLI paradigm.

3.2.2. OLI paradigm

Prior to deciding how to enter, investors have to decide where to invest; the institutional framework of countries has a crucial influence on locational choice (Bevan et al., 2004). The concept of locational advantages captures properties of host locations that make them attractive to potential foreign direct investors. Whereas most research used to focus on factor endowments as locational advantages, MNEs base their location decisions increasingly on knowledge-based assets, infrastructure and institutions of the host country (Bevan et al., 2004; Dunning & Zhang, 2008). Hence, investors seek locations where the institutional environment facilitates the development of their global firm-specific advantages.

For more than two decades, the OLI paradigm has remained the dominant analytical framework for accommodating a variety of operationally testable economic theories of the determinants of foreign direct investment and the foreign activities of multinational enterprises (Dunning, 2000). This framework claims that the extent, geography and industrial composition of foreign production and investment undertaken by MNEs is determined by the interaction of three sets of interdependent variables. The first set, ownership (O) advantages, suggests that the greater the competitive advantages of the investing firms, relative to those of other firms – and particularly those located in the country in which they are seeking to make their investments – the more they are likely to be able to engage in, or

(17)

17 increase, their foreign production. These advantages refer to the ability of enterprises to acquire certain assets not available, or not available at such favourable terms, to another country’s enterprises. These ‘assets’ refer to mean resources and capabilities capable of generating a future income stream. These include not only tangible assets, such as natural endowments and production factors, but intangible assets such as technology and information. Ownership advantages do not explain why the foreign presence should be established through FDI rather than exports.

The geographical location of FDI is addressed by locational (L) advantages that arise due to differences in relative labour costs, geographical distance, corporate tax systems and intellectual property rights systems (Dunning, 2008; 2001; 2000). Low levels of IP protection and enforcement decrease locational advantages of a host country (Javorcik, 2004). Locational advantages are specific to a particular location in their origin and use, but available to all firms (Dunning, 2008). One can think of the cultural, legal, political, financial and institutional environment in which assets are deployed. Alternatively, particular enterprises of the home country may own the assets, but they can be used with other resources and capabilities in the home country or elsewhere. Such assets may take the form of a legally protected intellectual property right.

The third set of advantages, internalization (I) advantages, offers a framework for evaluating alternative ways in which firms may organize the creation and exploitation of their core competencies. Internalization advantages explain why a foreign firm prefers to retain full control over the production process instead of licensing its intangible assets to local firms. However, this research only focuses on the ownership and locational advantages as a context for the expected effects of the levels of protection and enforcement strength on the level of IFDI flows.

3.2.3 National competitiveness

The national competitiveness of host countries can be distinguished into resources, capabilities and markets that form the physical environment in which firms and other organizations create economic well-being; and the institutions which set the rules of the game for, and determine the cognition and motivation of, firms and other wealth creating entities (Dunning & Zhang, 2008). Research on the relationship between the main ingredients of the competitiveness of national economies and the value of IFDI flows by foreign companies led us to the following theoretical assumptions. First of all, it is expected that firms do consider countries in terms of their ability to offer the pro-business environment and institutions that they need to make their investments worthwhile.

Whereas most research primarily focuses on the availability and quality of resources, capabilities and markets as the key determinant of economic welfare, this study emphasizes the equally or even more important role of the national institutional environment for IP protection and enforcement. The institutions of national economies are divided into forms and areas of institutional influence. This study mainly focuses on the formal institutions which consist of constitutions, treaties, laws and regulations, and the enforcement mechanisms (Dunning & Zhang, 2008). The area of institutional influence that is subject to this study is the area of intellectual property protection.

3.3 Institutional environment

3.3.1. Level of IP protection strength

The current literature points out that in countries where the capacity for imitation is high, measured by the ratio of R&D to GDP, a strengthening of IP protection increases the flow of trade (Park & Lippoldt, 2008). The study by Glass and Saggi (2002) finds a possible mechanism for divergence between the

(18)

18 growth rate of IP protection and the growth rate of IFDI. The increased difficulty of imitation caused by an increased level of IP protection strength generates resource wasting and imitation disincentive effects that reduces IFDI. On the other hand, firms may be reluctant to transfer more advanced and interesting technology to economies where piracy is high resulting from inadequate levels of IP protection strength. The ability to protect IP rights determines whether or not the developing world can actually attract R&D investments (A.T. Kearney, 2005). This has to do with the possibility of uncompensated R&D spill overs that affect firms’ investment decisions (Fan et al., 2013). These spill overs refer to the leakage of proprietary information through imitation or theft and these are more common in countries with weak IP protection regimes. So when a firm’s ability to capture gains from investment is limited because of a low level of IP protection strength, it becomes less attractive to invest in innovation. Imitation and information leakage are especially prevalent in developing countries characterized by inadequate IP protection regimes. However, firms do not automatically relocate to countries characterized by higher levels IP protection strength, suggesting that other locational factors play an important role as well (Fan et al., 2013).

A study by Khoury and Peng (2010) proposes how the implementation of the Paris Convention on Industrial Property Rights leads to more inward FDI. This institutional reform has direct implications for how MNEs intellectual property is protected according to the host country’s institutional environment. In international business research, the institution-based view asserts that the strategies behind the MNEs foreign direct investment decisions are affected by the rules of the game – or, institutions in host countries (Dunning & Lundan, 2008; Dunning & Zhang, 2008). The Paris Convention required member countries to provide the same protection for foreign owners as these countries provide for their own invention owners (WIPO, 2009). Moreover, with respect to disputes, the Convention prescribes that member host governments must provide foreign MNEs the same legal resources available to their host country firms and nationals. This provides foreign MNEs with more certainty since many developing countries have relied on imitative research activities of foreign inventions as a means to maintain industrial competitiveness.

The theory behind the expected positive influence of high levels of IP protection strength on the level of IFDI flows is that a low level of protection strength reduces the exclusive right of the foreign investor to use its IP rights (Seyoum, 2006). Multinational enterprises pay attention to the protection of IP when doing cross-country business, because the legal protection of their intellectual assets secure them competitive advantages over their rivals (Yang, Sonmez & Bosworth, 2004). Countries with inadequate levels of protection strength make it possible for infringing goods to be exported to other countries with inadequate IP protection, thus nullifying the potential benefits that would accrue to the foreign investor from market size. As a result, competition on the external trade side is likely to increase, often leading to lower payments for imports and a reduced market share to foreign investors (Seyoum, 2006). Following the above reasoning, the first hypothesis of this thesis is formulated as:

H1: An increase in the level of IP protection strength in host countries leads to an increase in their level of IFDI flows.

However, this positive effect is likely to be weak among countries with already high levels of IP protection strength. To the extent that different levels of IP protection strength across nations act as a locational determinant of FDI, the harmonization of IP institutional systems will offset such advantages (Maskus, 2005). It will make countries who are increasing their levels of IP protection strength more attractive but reduce the relative attractiveness of countries with already high levels of IP protection strength.

(19)

19 However, as the case description of Chapter 2 already emphasized, the harmonisation that is present in the strength of IP protection between member countries of the WTO is for a large part still absent in the strength of their national enforcement systems. The current literature primarily focuses on the Law on the books element when they investigate effects of IP protection, but the key interest of this study is to show the expected different role of IP protection when the actual enforcement of the legislation is taken into account as well.

3.3.2. Level of IP enforcement strength

Most studies in the existing literature focus on the strength of the regulatory structure of IP institutions (Law on the books) and find a positive effect on inward FDI flows, especially on FDI flowing form developed countries to less developed OECD economies (Papageorgiadis, Xu and Alexiou, 2019). But the era after the international harmonisation of the regulatory structure of IP institutions is characterized by an increase in disparities between countries’ levels of IP enforcement strength.

Papageorgiadis, Cross and Alexiou (2014) indicate that the evolution of IP systems in countries is mostly captured by quantifiable measures. However, the ability of firms to extract value, or to appropriate returns from the IP rights they own also depends on the ability of government agents and institutional actors to enforce the IP laws in practice. To capture the strength of national IP institutional environments, there should be taken account of additional aspects of the IP system of countries within which IP rights are granted, infringed and enforced (Maskus, 2000; Papageorgiadis et al., 2014). The need for including enforcement strength increased after the TRIPS Agreement since the Agreement decreased differences in the level of statutory IP protection between WTO members.

The availability of a comprehensive legal framework is a necessary but not sufficient condition to defend against imitation and misappropriation (Papageorgiadis et al., 2014). Enforcement-related operations involve the existence, availability, effectiveness and quality of organizations and firms such as police forces and customs officials as well as private enforcement agencies who facilitate the effective use of the patent system by economic actors.

The relation between the actual enforcement of IP laws and the level of IFDI flows can be approached from the following theoretical viewpoint: MNEs seeking to appropriate value from their intellectual assets need to ensure that they adequately account for institutional differences associated with a host country. The IP system imposes transaction costs to firms in the form of search, administration and enforcement costs. Transaction costs arise mainly from interactions between the IP system and the strategies that IP rights owners undertake in order to successfully exploit their intellectual assets (Papageorgiadis et al., 2014).

There arise three specific transaction costs from the interactions of foreign firms with the IP system of host countries: servicing costs, property rights protection costs and monitoring costs. Servicing costs are those costs that arise from the transfer of IP rights to a contracting party in a host country. The focus is more on external servicing costs which are determined by (i) the quality of administration in terms of the efficiency, transparency and timeliness of agencies that deal with related matters and their systems and routines; (ii) the complexity, clarity and communication of IP-related regulations and procedures; and (iii) the quality of administrative decisions made by government agencies who deal with IP-related matters for and on behalf of foreign and domestic firms (Papageorgiadis et al., 2014). Servicing costs measure the quality of IP administration. The property rights protection costs refer to whether or not ownership rights are upheld by the IP system in the case of infringement cases and the general effectiveness of the enforcement agencies. These transaction costs measure the strength of judicial enforcement. Monitoring costs arise from the efforts undertaken to constrain the infringement upon the property rights of IP owners. These costs are incurred as the IP owner scans the marketplace to identify instances of infringement of its intellectual assets. These costs

(20)

20 do not only measure the effectiveness of police enforcement, but also the perceptions of IP owners about national IP protection and enforcement levels.

In general, in countries where the level of enforcement strength is lower, managers are likely to experience greater transaction costs as they engage with enforcement issues (Papageorgiadis et al., 2014). Managers of MNEs are able to use the transaction costs as an indicator of IP systems strength to identify those countries where the enforcement aspect of the IP system is ineffective or inadequate. The level of IP enforcement strength determines the ability of MNEs to appropriate returns on market entry and development and influence the costs of international expansion. One cause for the lacking strength of enforcement of IP laws is formed by enforcement procedures which operate under social protocols that are closed or hard to access by foreign firms. This increases the costs and risks associated with protecting IP for MNEs (Papageorgiadis & McDonald, 2019). On the other hand, efficient functioning of institutions suggests a positive influence on the location selection, mode of entry, and technology transfer of MNEs since strong institutions are associated with a stable and low transaction cost investment environment (Papageorgiadis, Xu & Alexiou, 2019). A higher level of IP enforcement strength can lower the levels of transaction costs of IP owners who engage with the IP institution, reduce uncertainties and therefore attract FDI. Following the above reasoning, the second hypothesis of this thesis is formulated as:

H2: An increase in the level of IP enforcement strength in host countries leads to an increase in their level of IFDI flows.

3.3.3 Levels of IP protection and enforcement strength

Most of the empirical research on IP protection pays limited attention to how nations enforce their laws. Although all WTO members who signed the TRIPS Agreement are required to comply to the minimum standards for IP protection, the distinguishing factor among nations will be how they enforce those laws (Ostergard, 2000). It is noted that unless laws are enforced, they are virtually non-existent in terms of the protection afforded to IP owners and investors.

Foreign direct investors view institutions as an important aspect of the locational advantages of a potential host country and have been found to influence the extent of FDI (Bevan et al., 2004; Dunning & Zhang, 2008). Norms on intellectual property enforcement determine whether a legal system does in fact exist as such (Seuba, 2017). Defining enforcement is the starting point of drawing up boundaries and identifying the material content of the legal framework for the enforcement of IP rights. This is underlined by the TRIPS agreement which states that ‘domestic enforcement requirements may affect the nature and scope of the substantive rights incorporated into TRIPS’.

However, often the costs associated with the protection of intellectual property and the eventual negative welfare effect from certain enforcement measures are overlooked. Whereas intellectual property enforcement requires significant public investment in all countries, developing countries that transition to stronger protection are subject to short-run costs that are not trivial (Seuba, 2017). Moreover, strengthening right holders’ position at the cost of competitor’s procedural rights may deter legal competition and restrict access to legal products. The former negative effect refers to IP enforcement discouraging legal activities in fear of the consequences of eventual lawsuits. The threat of having to defend yourself against an intellectual property suit could also have deterring effects on trade. The other negative effect holds that some enforcement statutes have strengthened right holders’ position in such a manner that new substantive rights have been granted.

Although the national differences in enforcement strength were relevant and therefore largely remained, an important step was made in reaching an international and comprehensive set of enforcement norms. However, the provisions of the Agreement afforded member states a great margin

(21)

21 of discretion whereby the members could decide on the implementation of most enforcement provisions (Seuba, 2017). In the late 1970s and early 1980s, industrialized nations sustained that inadequate enforcement was a trade barrier, since their exports were being substituted by locally manufactured infringing products (Seuba, 2017). Considering the afforded margin of discretion on the implementation of enforcement provisions, it can be expected that nations are characterized by a large gap between the level of their IP protection strength and the level of their IP enforcement strength. Although countries will be subject to gaps between their levels of IP protection and enforcement strength, interacting the two variables to see their joint effect is expected to show a positive effect on the outcome variable. The above reasoning results in the following hypothesis regarding the interacting effect of IP protection and enforcement strength to be tested:

H3: Interacting the level of IP enforcement strength with the level of IP protection strength will have a positive jointly effect on the level of IFDI flows.

Figure 2: Conceptual model

Maskus (2005) stated that it must be emphasized that high levels of IP protection and enforcement strength alone are insufficient for generating strong incentives for firms to invest in a country. If that were the case, then Brazil, China and other high-growth, large-market developing economies with inadequate IP systems would not have attracted nearly as much FDI if investment were solely dependent on the protection and enforcement of IP (A.T. Kearney, 2005; Maskus, 2005). Rather, the levels of protection and enforcement strength should be seen as important components of an overall pro-competitive business environment which enables foreign investors to exploit their ownership advantages (Dunning, 2000). Two of these additional locational advantages, next to IP protection and enforcement strength, are added to the conceptual framework of this thesis.

The first one is the level of GDP which does not only control for country size and level of economic development, it also forms an indicator for the level of consumer demand for higher quality and more differentiated products on the one hand and the technological capabilities of domestic producers on the other (Maskus, 2000; Papageorgiadis et al., 2014). Foreign direct investment is likely to replace direct exports of a good where the size of the host market is large (Maskus, 2005).

The second one is the level of trade openness which, together with the level of GDP per country, highly determines whether foreign investors are able to exploit their ownership advantages. Foreign

(22)

22 investors usually prefer to operate on competitive domestic markets which is more the case in countries with more extensive liberalization of domestic and international markets (Bevan et al., 2004). However, it can be expected that trade openness has a weak positive influence on the level of IFDI flows in developed countries since in general all developed countries have open markets so that this variable exhibits little variability among developed economies (Park & Lippoldt, 2008).

(23)

23

Chapter 4 Research design & data

4.1 Introduction

The previous Chapter described a number of factors as part of the institutional environment of host countries that might have an effect on the level of IFDI flows. The contribution of this study is formed by its additional focus on the strength of national IP enforcement systems next to countries’ national statutory IP protection. Regarding the joint effect of IP protection and enforcement strength on a host country’s level of IFDI flows the following hypothesis was formulated:

H3: Interacting the level of IP enforcement strength with the level of IP protection strength will have a positive jointly effect on the level of IFDI flows.

This hypothesis is based on the literature that led to the following two hypotheses regarding the separate effects of a host country’s level of IP protection and enforcement strength respectively

H1: An increase in the level of IP protection strength in host countries leads to an increase in their level of IFDI flows; and

H2: An increase in the level of IP enforcement strength in host countries leads to an increase in their level of IFDI flows.

Moreover, host countries’ GDP level will be added to the regression since this variable does not only form an important indicator of a host country’s pro-competitive business environment, but it does also allow us to make comparisons between countries. In addition to the GDP level, is the level of trade openness included in the regressions as well in order to improve the precision of the estimates of our key independent variables. Chapter 3 can be written as the following regression equation:

IFDI = 𝛽0 + 𝛽1 protection + 𝛽2 enforcement + 𝛽3 protection_enforcement + 𝛽4 GDP + 𝛽5 openness + 𝜀

Equation 1: Regression equation

The expectations that were formulated in Chapter 3, and how they relate to the expected treatment effects of the independent variables and control effects of the control variables, are summarized in the table below.

(24)

24 𝛽1 > 0 The higher the level of IP protection

strength,

the higher the level of IFDI flows will be.

𝛽2 > 0 The higher the level of IP enforcement strength,

the higher the level of IFDI flows will be.

𝛽3 > 0 Interacting the level of IP protection strength with the level of IP enforcement strength,

will derive a positive jointly effect on the level of IFDI flows.

𝛽4 > 0 The higher the level of GDP, the higher the level of IFDI flows will be. 𝛽5 > 0 The higher the level of trade openness, the higher the level of IFDI flows will be.

Table 1: The expected treatment and control effects

This Chapter will focus on the research design applied to find the treatment (and control) effects, the data that will be used and its sources.

4.2 Research Design

4.2.1 Panel data analysis

This study makes use of the panel design which observes different units over time. In panel designs, several units are observed at a number of points in time (Toshkov, 2016). When using panel data both cross-sectional/group effects and/or time-series/time effects can be found.

One of the main advantages of large-N designs is that it can identify and estimate weak heterogeneous relationships. The strategy chosen for causal inference is the conditioning approach in which I build a multivariate statistical regression model that includes the key independent variables, the dependent variable and control variables. Including control variables in the regression is done to prevent that the relationship between the key independent variables and the outcome variable from the data would be biased. Moreover, a great number of observations is used in order to rule out that empirical associations are due to chance and to estimate the relationships more precisely in presence of noisy data.

A regression analysis is characterized by holding constant the most obvious and important differences between the units of observation (Angrist & Pischke, 2014). The regression model in this context is an equation linking the treatment variables to the dependent variable while holding control variables fixed by including them in the model. We control for measurement error since the measurement error can bias the causal inference when it is correlated with the values of the outcome variable or the values of the main explanatory variable of interest or with the values of an important confounder (Toshkov, 2016). This leads to a slightly different regression equation:

IFDIit = 𝛽0 + 𝛽1 protectionit + 𝛽2 enforcementit + 𝛽3 protection_enforcementit + 𝛽4 GDPit + 𝛽5 opennessit + 𝜀it

The i subscript indicates the group dimension (countries) while the t subscript denotes the time dimension (years).

In panel data analysis the group and time effects can be fixed and/or random. Fixed effects assume that omitted effects are correlated with variables included in the regression. To capture the differences across

(25)

25 groups (countries), the model gives each group its own intercept. The main consideration in choosing for a fixed effects model is whether you assume there is unobserved heterogeneity among the cross sections. In this case, unobserved heterogeneity among the cross sections is assumed and therefore the fixed effects model is chosen. One reason for this is that I think that the country-effect could well be correlated with the other independent variables in the regressions. Another reason is that wrongly assuming away the possibility of unobserved heterogeneity leads to using a random effects model resulting in inconsistent estimates.

4.2.2 One-way fixed effects regression

Since different countries can be wildly different from each other for a variety of reasons, the assumption of unobserved heterogeneity across countries makes sense. When unobserved heterogeneity (i.e. some unobserved factor that affects the dependent variable and is correlated with some observed regression) is assumed, the fixed effects regression is consistent. The corresponding regression equation differs somewhat from the previous one, regarding the intercept:

IFDIit = 𝛽0i + 𝛽1 protectionit + 𝛽2 enforcementit + 𝛽3 protection_enforcementit + 𝛽4 GDPit + 𝛽5 opennessit + 𝜀it

Equation 2: One-way fixed effects regression equation

The fixed effects model assumes that each country is independent of all other countries, that each time period is independent of all the other (no autocorrelation), and that the variance of the error term is constant (no heteroskedasticity) (Moody, 2005). When performing one-way fixed effects regression, the statistical programme (Stata) generates dummy variables per country. These dummy variables are necessary to estimate the unobserved time-invariant country effects. While this technique fits a model where the parameters are constant, the intercept differs per group (country), which is reflected by 𝛽0i in the model. Although every country has its own starting point, reflected by the individual intercept, one-way fixed effects regression estimates conditional expectations that apply to all countries.

4.2.3 Adding control variables

The model includes control variables to improve the predictive power of the model and the precision of the estimates. The omitted variable bias is reduced, because if we omit such variables from the analysis the relationship between the independent and outcome variable estimated from the data would be biased (Toshkov, 2016). The control variables are chosen as to reflect important differences between years and countries of which the literature states they influence the level of IFDI. Especially control variables measuring aspects of a host country’s pro-competitive business environment are expected to reduce the significance of the estimators associated with the host country’s institutional environment designed for the protection and enforcement of IP.

4.2.4 Dealing with possible bias

Regression analysis can be characterized by two common biases: serial correlation and heteroskedasticity. Serially correlated data are persistent, meaning the values of variables for nearby periods are likely to be similar (Angrist & Pischke, 2014). Heteroskedasticity occurs when the variance of residuals is related to regressors. If we ignore serial correlation and heteroskedasticity and use the simple standard error formula, the estimates are likely to be misleading. Robust standard errors allow

Intellectual Property protection and enforcement in explaining the level of Inward Foreign Direct Investment flows

Master Thesis