Home country institutions and firm performance: a perspective from textile emerging market firms and foreign multinationals

Home country institutions and firm

performance: a perspective from textile

emerging market firms and foreign

multinationals

Master’s Thesis

Date of submission: 26th _{of June 2020}

2

Abstract

It has been established in multiple studies that home country institutions affect firm performance. However, the effect on the firm performance of emerging market firms has received relatively less attention. This study investigates the relationship between home country institutional frameworks and the firm performance of emerging market firms. It is argued that an increase in the quality of the home country institutional framework has a negative effect on the firm performance of emerging market firms. As the institutional quality increases foreign Multinational Enterprises will have an advantage over domestic firms. Using a sample of 665 emerging market firms (emerging market firms, multinationals and foreign multinationals) from four emerging markets, this study shows that the performance of domestic firms is negatively affected by an increase in the political stability of the emerging markets, whereas the MNEs and EMNEs are positively influenced by this increase. Moreover, an increase in rule of law leads to a decrease in the performance of domestic firms and EMNEs and an increase in the performance of MNEs.

Key words: institutions, institution-based view, emerging markets, emerging market multinational, emerging market firm, firm performance

3 Table of contents Abstract ... 2 List of figures: ... 4 List of abbreviations: ... 5 1. Introduction ... 6 2. Literature Review ... 9 2.1 Institutions ... 9

2.2 Institutions and firm performance ... 10

2.2.1 Political stability and firm performance ... 12

2.2.2 Rule of law and firm performance ... 14

3. Methodology ... 17 3.1 Sample collection ... 17 3.2 Data collection ... 18 3.2.1 Dependent Variable ... 18 3.2.2 Independent Variables ... 19 3.2.3 Control Variables ... 19 3.3 Descriptive Statistics ... 21 3.4 Data analysis ... 25 4. Results ... 26

4.1 Regression and Chow test results ... 26

4.2 Hypothesis testing ... 29

4.2 Robustness checks ... 31

5. Discussion ... 33

5.1 Contributions ... 34

5.2 Limitations and Future Research ... 34

5.4 Conclusion ... 35

References ... 33

Appendices ... 42

Appendix A – Preliminary Checks ... 42

4 List of figures:

FIGURE 1 Third leg on the strategy tripod: institutions-based view, Peng et al., (2008) ... 11

FIGURE 2 Conceptual Model ... 16

FIGURE 3 Histogram of ROA distribution of MNEs ... 42

FIGURE 4 Distribution of residuals PV ... 43

FIGURE 5 Distribution of residuals RL ... 43

List of tables: TABLE 1 List of home countries and Domestic firms, EMNEs and MNEs ... 18

TABLE 2a Summary Statistics and Correlations for Domestic Firms ... 22

TABLE 2b Summary Statistics and Correlations for EMNEs ... 23

TABLE 2c Summary Statistics and Correlations for MNEs ... 24

TABLE 3 Regression Analysis – Firm Performance as Dependent Variable ... 27

TABLE 4 Chow Test for independent variables and type of firm ... 28

TABLE 5 Summary of hypotheses and findings ... 31

TABLE 6 Breusch-Pagan Test for Heteroskedasticity PV ... 44

TABLE 7 Breusch-Pagan Test for Heteroskedasticity RL ... 44

TABLE 8 Variance Inflation Factors ... 45

TABLE 9 Variance Inflation Factors without GDP ... 46

5 List of abbreviations:

EMs Emerging Markets

EMNEs Emerging Market Multinational Enterprises GDP Gross Domestic Product

IB International Business IT Institutional Theory MNEs Multinational Enterprises OLS Ordinary Least Square

PV Political Stability and Absence of Violence

RL Rule of Law

ROA Return on Assets

6

1. Introduction

According to Peng (2003), home country institutional pressures affect the strategic choices of firms. The underlying premise here is that the firms are embedded in the institutional environment of a country, and therefore influenced by the regulative, normative and cognitive features of the institutional environment (Busenitz, Gómez, & Spencer, 2000; Scott, 1995). These in turn lead to normative, coercive, and mimetic pressures which shape firm behavior and performance (Scott, 1995). Institutions and institutional quality differ significantly per country, which leads to different types of influences on the strategy and performance of firms (Ring et al., 2005)

Over the years, the influence of home country institutions on firm performance or strategy in the developed markets, where institutions are well-developed, has received a lot of attention (Chan et al., 2008; Cherchye and Verriest, 2016; Meyer & Peng, 2003; Peng, 2009, Scott, 1995). However, the institutional environment (or the lack of it) in emerging markets is likely to shape firms strategies and performance in a more profound manner since the institutions are still evolving in these countries (Hoskisson et al., 2000; Peng et al., 2008). An emerging market can be defined as a country that meets two conditions: rapid economic growth and a government implementing policies that favor economic liberalization (Arnold & Quelch, 1998).

7 (McGahan and Victer, 2010), the effect of individual dimensions of the institutional environment on the performance of different types of firms has been overlooked. This is important to know for several reasons. First, the institutional environment is a composite of many factors and not all dimensions of institutions affect all types of firms in the same manner. Some type of firms (e.g. domestic) may be better at profitably exploiting specific types of institutional conditions (e.g. corruption) than other firms. Second, there are inherent differences between local firms and their multinational counterparts in terms of how they are likely to respond to changing institutional conditions in emerging markets. Third, even within multinational firms, local firms which have internationalized (also known as emerging market multinationals or EMNEs) may respond differently than foreign owned multinational firms (MNEs) to changing institutional conditions. Based on the above information, the following research question can be formulated:

‘What is the effect of legal and political aspects of home country institutions on the firm performance of different type of firms?’

The emerging markets differ from developed markets in multiple ways. The most important differences include the institutional voids, weak enforcement and antitrust laws and rapid institutional changes. Therefore, the legal and political aspects will be used for measuring institutions. The main question will be answered using a sample of 665 textile manufacturing firms from four different emerging markets, namely Brazil, India, Mexico, Russia. Moreover, a further distinction is made by distinguishing between domestic firms, EMNEs and MNEs, which has not been done in previous research (Cherchye and Verriest, 2016; McGahan and Victer, 2010 and Makino et al., 2014). The quality of home country institutions will be measured using two variables of the World Governance Indicators (WGI), namely political stability and absence of violence (PV) and rule of law (RL). The dependent variable firm performance is captured by the variable Return on Assets (ROA).

8 shown to be equal for the EMNEs and domestic firms. This is also not in line with what was expected for the domestic firms and EMNEs.

9

2. Literature Review

2.1 Institutions

Institutional theory (IT) has provided a theoretical foundation for research regarding the emerging markets (Meyer & Peng 2016; Peng et al. 2008). It integrates two perspectives: the economist and sociology view. On the economist side, North (1990, p.3) defines institutions as: “the rules of the game within a society, or the devised constraints that shape human interaction. These rules include both formal (laws) and informal constraints (values)”. From a sociology point of view institutions are divided into three pillars: regulatory, normative, and cognitive (Scott, 1995). The regulatory dimension includes regulations, legal requirements, and rules a firm must abide to. Normative dimensions include the informal rules and procedures and are associated with the values in a society. A firm needs to conform to these rules and procedures in order to avoid social reciprocation. The last dimension includes the cultural rules and frameworks a firm needs to follow (Scott, 1995).

10 (Khanna and Palepu, 1998; McMillan, 2007). This leads to a lack of transparency, information asymmetries and higher enforcement and monitoring costs (Xu and Meyer, 2013), as well as less well-developed antitrust laws (Knack and Keefer, 1997) and contract enforcement laws (Tybout, 2000). Secondly, the government and in addition government owned or controlled firms are active players within the market economy, regulating market competition and preventing other countries from entering the market to protect the economy (Ghemawat and Khanna, 1998; Cherchye and Verriest, 2016). Furthermore, transactions are determined based on networks, which are shaped in part due to tradition but also a consequence of less efficient markets (Xu and Meyer, 2013). Lastly, the emerging markets are characterized by a high volatility of economic, political, and institutional policies (Hoskisson et al., 2000). This leads to more risk and uncertainty for companies, resulting in that it is harder to predict future events and strategic decisions needed (Xu and Meyer, 2013).

This study focuses in part on the legal institutions and in part on the political institutions. The reason is that the characteristics of emerging markets, as mentioned above, differ substantially from the developed markets concerning these aspects. Researchers often consider them as similar, because they have similar goals, and both guide the behavior of firms. However, the processes are quite different. Legal institutions include the system of rules that the government uses to guide behavior and create social order via laws, property rights and binding contracts (North, 1990; Davis, 2014). These laws are not flexible and can only be adapted in accordance with public procedures (Chintakanda and Tan, 2016), whereas political institutions consist of policies that are created, applied, and enforced by the governments and political parties to govern a country’s social and economic system (North, 1990). These policies are quite flexible and subject to changes in the government.

2.2 Institutions and firm performance

11 The relationship between institutions and firm performance can be shown using the strategy tripod. The institution-based view is often seen as the ‘third leg’ in the strategy tripod (see Figure 1.) which explains why firms make strategic choices. It consists of the industry competition view and the resource-based view and together they affect the strategies firms use (Ahlstrom et al., 2003; Peng et al., 2009).

FIGURE 1

Third leg on the strategy tripod: institutions-based view, Peng et al., (2008)

12 institutional systems that differ substantially from the systems before, which causes turbulence (Peng, 2003). This makes strategic planning harder for firms. However, two strategies have been identified by Peng (2003) that firms in emerging markets apply. The first is a network-based strategy network-based on interorganizational ties and relationships. It was found that when the institutional quality in a country is weak, resulting in less formal constraints, the firms often rely on informal constraints for facilitating economic exchanges (Peng and Heath. 1996). The second is a market-based strategy in which the focus lies on the resources and capabilities of the firm (Peng, 2003). Peng (2003), states that the choice between these two strategies can be determined by the phase of transition the institution is in and the level of institutional quality. 2.2.1 Political stability and firm performance

Political stability is an important factor to include while measuring the effect of institutions on firm performance. The main reason is that many emerging markets are or have been transitioning (Peng, 2003), which often results in abrupt political changes and instability. Political stability has an influence on ‘the rules of the game’ because it affects the government policies and the environment in which firms operate (Chintakanada and Tan, 2016). A low level of political stability has multiple consequences. First, the firms are not able to react to changes in policies, because they have less time to understand them better and the business environment is therefore less stable (Chintakanada and Tan, 2016). Furthermore, there is a high level of information asymmetry, which leads to less competition, since firms and especially foreign firms will not be able to enter the market, because of the lack of knowledge (Chintakanada and Tan, 2016). Lastly, it has been found that a lower level of political stability leads to less investments, which leads to a lower level of economic growth (Alesina and Perotti, 1993). The theory behind this is that political instability leads to increased uncertainty regarding property rights and economic policies. This in turn leads to the firms fearing opportunism.

13 foreign firms to enter the market (DiMaggio and Powell, 1983; Cherchye and Verriest, 2016), resulting in more competition. Since domestic firms are embedded within the environment and cannot reap benefits from investments in other countries, their firm performance might decrease (McGahan and Victer, 2010). Moreover, an increasing competition leads to less resources and high-skilled employees available for the domestic firms (Driffield et al., 2012), resulting in a lower productivity and therefore a lower firm performance. Hence:

H1a: ‘Domestic firms in emerging markets are likely to benefit less from changes in political stability’

EMNEs have been and are still part of the environment and are therefore used to the abrupt political changes and voids as well. However, in contrast to domestic firms, EMNEs are geographically diversified and consequently less affected by the home country markets. This has multiple reasons. First, since they are embedded in other countries as well, they will be able to rely on profits made abroad, when a higher political stability leads to more competition and a lower firm performance (Chan et al., 2008). Second, as a result of the broader scope, EMNEs might have formed and gained some firm specific advantages and assets that can be used as a shield against more competition. Lastly, as EMNEs are part of multiple environments that might differ in terms of the quality of the institutions, they might have learned how to deal with institutions where the quality is much higher, therefore their firm performance might not be as negatively influenced. In sum:

H1b: ‘EMNEs in emerging markets are likely to benefit less from changes in political stability, but less so than the domestic firms’

14

H1c: ‘MNEs in emerging markets are likely to benefit more from changes in political stability’

2.2.2 Rule of law and firm performance

Rule of Law (RL) represents the extent to which public authorities follow and have confidence in the rules set up by society. This has in particular to do with the quality of property rights, contract enforcement, the courts, the police, and the overall likelihood of crime (Kaufmann et al., 2010). The effective enforcement of rules is necessary to enable exchange by reducing the threat of opportunism and uncertainty (Aron, 2000). Thus, when rule of law is high, more transactions will occur, because firms will be able to trust on the effective enforcement of contracts, resulting in more competition, since it is more attractive for MNEs to enter the foreign market. Most emerging markets however, are characterized by weak institutions (Hoskisson et al., 2000) and institutional voids (Khanna and Palepu, 1998). Therefore, a lower level of rule of law is expected, where laws are poorly enforced. This results in less transactions because firms fear opportunistic behavior. Furthermore, more effort needs to be put in gaining and developing the necessary resources and assets needed to operate (Elango and Lahiri, 2014), which results in a decrease in firm performance. Moreover, a lower rule of law results in firms often relying on informal constraints for facilitating economic exchanges (Peng and Heath. 1996).

Applying the above disadvantages of rule of law to the different type of firms distinguished in this study, it can be said that each firm will experience these constraints differently. As mentioned before, the domestic firms are integrated within the environment and used to operate in a market with a lower level of rule of law (McGahan and Victer, 2010; Khanna and Palepu, 1998). They will have established a broad informal network to do business in, based on trust. The domestic firms share values and norms with the other domestic firms which makes it easier to trust one another and for transactions to occur (Jones et al., 1997). Furthermore, these informal networks might consist of people from judiciary, who they might be able to influence in order to get favorable decisions and rulings. Domestic firms will therefore be less able to influence the rule making, should the quality of rule of law increase. This again results in more competition, since MNEs will now be interested to enter the market as well. Together these arguments result in a lower firm performance for domestic firms. In sum:

15 EMNEs are used to these conditions as well, resulting in the managers being more flexible and able to operate outside a contractual relationship (de Soto, 2000). Furthermore, managers will be able to develop a better strategy to cope with these institutional voids and will choose their partners more carefully. Should the quality increase, then EMNEs have the advantage of being active in multiple countries, where the rule of law might be higher, they might therefore have gained experience as to how to operate in a better legal environment and be able to fall back on the profits made abroad. Hence:

H2b: ‘EMNEs in emerging markets are likely to benefit less from changes in rule of law, but less so than the domestic firms’

In contrast, MNEs are not familiar with unstable conditions and are accustomed to relying on the enforcement of contracts (Cuervo-Cazurra and Genc, 2008). They will experience more problems with establishing informal contracts, because of the unfamiliarity with this way of doing business, information asymmetry and the different values and customs. This results in more costly market transactions and less efficient transformation (Henisz and Zelner, 2005), which decreases their overall firm performance. The main cause for the increase in transaction costs is that the assets need to be protected from expropriation, by means of, for example the use of mobile assets and loans against political risks (Henisz, 2000). Transformation costs increase as a result of the greater risk of contracts not being enforced and insecure property rights. Firms, will use inferior technology to produce and this leads to less efficiency, decreasing firm performance (North, 1990). As mentioned before, a better rule of law leads to firms being able to rely on the effective enforcement of contracts, property rights and results in more protection (Khanna and Palepu, 1998; North, 1990). Countries with a better-developed rule of law will therefore provide a more favorable context for MNEs, which leads to a better firm performance (Henisz, 2000). In sum:

16 In figure 2, an overview of the proposed hypothesis can be found.

17

3. Methodology

3.1 Sample collection

The sample of this study is derived from four emerging markets, namely Brazil, India, Mexico, and Russia. The reason for choosing these specific countries is threefold. First, it is expected that the effect of home-country institutions on firm performance will vary between the emerging markets as well, due to the different levels of development. Therefore, a more diverse sample is chosen that consists of countries from Europe, Asia, and South America. Secondly, the countries have undergone some major institutional transition (Peng, 2003). It would therefore be interesting to see whether this influenced the firm performance. Lastly, Brouthers et al. (2016), found that when samples are comprised of only one home-country, the sample is vulnerable and researchers cannot be sure of what drives the results, this affects the generalizability of the sample.

Additionally, the sample will contain three types of emerging market firms: domestic firms, EMNEs, headquartered in the emerging country and foreign MNEs. In addition, these firms are all active in the textile manufacturing industry. The reasons for this decision is fourfold. First, most emerging markets are specialized in manufacturing (OECD, 2019) and it is the industry that is the most influenced by national regulations (Beverelli et al., 2017). Also, for example in India, the textile industry is one of the biggest industries (India Brand Equity Foundation, 2019), where huge investments of the government are made to increase the quality and productivity. The industry is also becoming one of the largest industries in the world for textile (McKinsey, 2019). Furthermore, an increase in demand is expected because of the rise in income levels (India Brand Equity Foundation, 2019). The other countries, Mexico, Brazil, and Russia are also known for their textiles and play a major part in the world textile industry (Economic Intelligence Unit, 2019). Furthermore, Cherchye and Verriest (2016) focus on the whole manufacturing sector, however not every industry is affected in quite the same way. Therefore, it would be interesting to focus on one specific industry. Lastly, it leads to a more homogeneous sample and to a better comparison between the countries in the sample.

18 firm without foreign subsidiaries, whereas an EMNE is defined as a domestic firm with subsidiaries in other countries (>1). The foreign multinational is found by looking at where the ultimate owner is headquartered. (4) data is collected based on the last year available ranging from 2014 to 2018. This left us with a sample of 665. How the sample is distributed per country and per type of firm can be viewed in table 1.

TABLE 1

List of home countries and Domestic firms, EMNEs and MNEs

Home Country Domestic EMNEs MNEs

Brazil 17 10 16 India 326 245 7 Mexico 11 4 1 Russia 12 10 6 Total 366 269 30 3.2 Data collection

All the data collected for this research includes secondary data. 3.2.1 Dependent Variable

The dependent variable of this research is Firm Performance. This variable is used quite often by researchers to measure whether the actions taken by firms, as a reaction on a change in the external or internal environment, are efficient and effective (Atan et al., 2018; Bhaumik et al., 2010; Makino et al., 2004; McGahan and Victer, 2010). Multiple proxies can be used to capture firm performance, for instance by using Tobin’s Q or return of equity (ROE). In this research firm performance is captured by the variable Return on Assets (ROA). This is an indicator to measure firm profitability as well as the efficiency, since it measures how assets are converted into income (Atan et al., 2018). Equation 1 shows how ROA is calculated. The information on net income and total assets is found using Orbis.

(1) 𝐹𝑖𝑟𝑚 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 = 𝑅𝑂𝐴 = 𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒

19 3.2.2 Independent Variables

The independent variables used in this research are Political stability and Absence of Violence and Rule of Law. The data for these variables will be gathered using the Worldwide Governance Indicators (WGI) database from the World Bank. The WGI is created by Kaufmann, Kraay and Mastruzzi (2010) and captures the quality of institutions in over 200 countries. Both variables vary on a -2.5 to a +2.5 points scale, with the lowest score indicating a very low political stability and rule of law. In other words, the lower the score, the higher the political instability and lower the rule of law, indicating a weaker institutional quality.

3.2.3 Control Variables

Five general firm characteristics are used in this research as control variables, namely Firm

Size, Firm Age, R&D Expense, Leverage, and Firm Growth. Furthermore, one country

characteristic is added: GDP per Capita.

Firm size is found to influence firm performance positively (Greve, 2008). This is since larger

firms have a larger resource pool, benefit from economies of scale, and usually have a higher market power (Lee, 2009). Therefore, firm size is included as a control variable in this study. This study uses a proxy of Orbis of Bureau van Dijk, in which the firms are classified into four categories, namely 1 for small firms, 2 for medium size firms, 3 for large firms and 4 for very large firms. The reason for the use of this proxy is the lack of data availability in the sample.

Firm age is included as a control variable because it was found that firm performance is

positively related to age (Coad et al., 2013). The reason is that more mature firms have a higher level of productivity, larger size, lower debt ratios and higher profits. However, it was also found that the firm performance could deteriorate with age, because they have lower expected growth (Coad et al., 2013). It is therefore important to control for age. This variable will be calculated by subtracting the year of incorporation from 2020 and will be measured in years.

GDP per Capita is another measure that influences firm performance (Goldszmidt, Brito, & de

Vasconcelos, 2011). Therefore, this is included as a control variable. The variable used is the GDP per Capita per country per year and is derived from the World Bank database.

R&D expense is the amount of money firms spend on the search for new technological

20

Leverage is the debt ratio. Luthfiah (2018) states that the operational activities of a company

cannot be financed with only funds from within the company, therefore funds from debt are needed. The leverage ratio measures the amount of debt that is used to finance company activities as opposed to own capital (Luthfiah, 2018). The more funds available, the more money will be available for a company to invest in its processes, which affect firm performance. The equation (2) shows what leverage is composed of.

(2) 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒 = 𝑇𝑜𝑡𝑎𝑙 𝐷𝑒𝑏𝑡

𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠

Firm growth represents market acceptance and firm success (Abor, 2005). It was suggested that

21 3.3 Descriptive Statistics

In table 2, the descriptive statistics and correlations for each type of firm are presented. The descriptive statistics provide an overview of the mean, the standard deviation, and the minimum and maximum scores of the different variables.

22

TABLE 2a

Summary Statistics and Correlations for Domestic Firms

Variables Mean Std. dev Min Max ROA PV RL Firm

Growth

R&D Firm age Leverage GDP Firm Size

23

TABLE 2b

Summary Statistics and Correlations for EMNEs

Variables Mean Std. dev Min Max ROA PV RL Firm

Growth

R&D Firm Age Leverage GDP Firm Size

24

TABLE 2c

Summary Statistics and Correlations for MNEs

Variables Mean Std. dev Min Max ROA PV RL Firm

Growth

R&D Firm Age Leverage GDP Firm Size

25 3.4 Data analysis

The dataset used in this study is collected from two databases: Orbis and the World Bank. The main statistical method used was a linear regression analysis because the data collected is cross-sectional and the hypotheses refer to the individual effect of the independent variables on firm performance. With help of Stata 16, this study estimates the relationship between firm performance and political stability and rule of law. Before, the data can be analyzed certain preliminary tests need to be executed to test for certain assumptions. The results of these tests can be found in appendix A.

All in all, it was found that the dependent variables are normally distributed. The residuals are non-normally distributed and heteroskedastic, and the independent variables rule of law and political stability and absence of violence are somewhat correlated, which is mainly caused by the control variable GDP. For the most part, it is not necessary to change the data due to the moderate sample size, which makes it normally distributed, however it is considered when performing the regression and interpreting the results.

The analysis is executed and the equations 2 and 3 below are used to test hypothesis 1 and hypothesis 2.

(2) 𝑌_𝑅𝑂𝐴 = 𝛽₀+ 𝛽₁𝑋𝑃𝑉+ 𝛽_𝑐𝑣𝑖𝑋𝑐𝑣𝑖 + ε

(3) 𝑌_𝑅𝑂𝐴= 𝛽₀+ 𝛽₁𝑋𝑅𝐿+ 𝛽_𝑐𝑣𝑖𝑋𝑐𝑣𝑖 + ε

The YROA is the return on assets (dependent variable), β is the slope of the parameters, with 𝛽₀ being the intercept. 𝑋_𝑃𝑉 is the independent variable Political Stability and Absence of Violence in the first equation. In the second the independent variable Rule of Law is represented by 𝑋𝑅𝐿. The cvi are the control variables included in this research, consisting of cv1 Firm Size; cv2 Firm Age; cv3 GDP per capita; cv4 R&D expense; cv5 Firm Growth and cv6 Leverage. The ε is the error term consisting of the residual errors.

26 In this research it is tested whether the effect of the independent variables is different for each type of firm. With use of the Chow test it can be established whether the independent variables are more significant for EMNEs, domestic firms or MNEs.

The Chow test is executed by first creating dummy variables for all the type of firms in one database. Next, an interaction term is created of the regressors PV and RL and the dummy variables. The model is then fit with the interaction and the dummy and a Chow test can be executed, using first regression and then the command .test.

4. Results

4.1 Regression and Chow test results

27 TABLE 3

Regression Analysis – Firm Performance as Dependent Variable

Domestic EMNEs MNEs

Model 1 Model 2 Model 3 Model 4 Model 1 Model 2 Model 3 Model 4 Model 1 Model 2 Model 3 Model 4

Control variables Constant -13.12 (6.977) -26.64 *20.94) -13.04 (6.980) -43.47 (31.93) -15.27 (-0.79) -9.531 (-0.52) -15.17 (-0.78) -5.153 (-0.30) 23.21 (38.74) 60.13 (30.67) 26.32 (39.21) 101.0** (20.26) Firm Growth -0.0052 (0.0086) -0.0055 (0.0087) -0.0052 (0.0086) -0.0058 (0.0091) 0.0077 (1.81) 0.0080 (1.85) 0.0078 (1.82) 0.0079 (1.84) -0.578 (3.859) -0.566 (3.693) -3.775 (6.950) -2.819 (2.730) R&D 0.0132** (0.0048) 0.0128 ** (0.0048) 0.0130 ** (0.0048) 0.0130 ** (0.0049) 0.0012** (2.85) 0.0013** (2.90) 0.0012** (2.79) 0.0012** (2.84) 0 0 0 0 Firm Age -0.0365 (0.0361) -0.306 (0.0370) -0,0336 (0.0369) -0.0331 (0.0370) 0.0047 (0.26 -0.0034 (-0.16) 0.0025 (0.12) -0.0002 (-0.01) -0.227 (0.166) -0.183 (0.101) -0.224 (0.170) -0.140 (0.0697) Leverage 0.762 (0.993) 0.769 (0.993) 0.768 (0.995) 0.756 (1.002) -6.424*** (-5.10) -6.494*** (-5.13) -6.425*** (-5.09) -6.551*** (-5.10) -3.313 (6.094) -0.566 (3.693) -3.775 (6.950) -2.819 (2.730) Firm Size 3.720* (1.638) 3.815* (1.648) 3.661* (1.636) 4.129* (1.769) 4.598 (0.94) 4.570 (0.94) 4.596 (0.93) 4.557 (0.95) 0.369 (4.064) -3.883 (5.516) 1.394 (8.145) 2.002 (2.395) Independent variables Political Stability and Absence of Violence

-13.64 (19.44) -30.34 (30.07) 5.678 (1.10) 10.46 (1.13) 57.85 (28.50) 85.58*** (6.355) Rule of Law 3.594 (9.358) -12.20 (19.85) -1.605 (-0.39) 7.083 (0.86) 19.17 (96.53) 148.2*** (18.62) N 138 138 138 138 ₂₀₁ 201 201 201 ₁₅ 15 15 15 R squared 0.056 0.059 0.056 0.061 0.196 0.199 0.196 0.201 0.121 0.642 0.128 0.946 Adj. R squared 0.020 0.016 0.013 0.010 _0.175 0.175 0.171 0.172 _0.319 0.386 0.496 0.892 F 11.00 11.02 11.04 11.06 _10.93 10.93 10.96 10.95 _21.67 14.79 23.07 6.211

* (significant at the 0.05 level)

** (significant at the 0.01 level)

28

TABLE 4

Chow Test for independent variables and type of firm

Domestic vs. EMNE EMNE vs. MNE Domestic vs MNE

PV RL PV RL PV RL

F 3.14 2.76 7.52 0.70 9.39 1.25

29 4.2 Hypothesis testing

First of all, it needs to be remarked that the number of observations is lower than the total sample for each type of firm. This is because some values are missing for the control variables. These are missing completely at random (MCAR), indicating that some power might be lost in the design, but this does not mean that the estimated parameters are biased. Second, while looking at the constant, it is interesting to see that for the MNEs all are positive, which might indicate that when all predictors are set to zero, the firm performance will be positive. In contrast, the constant for the domestic firms and EMNEs is always negative, indicating that a better political stability and absence of violence and rule of law leads to a negative firm performance. However, to be able to draw conclusions the model needs to be inspected further. Model 1 is the base model of this analysis. Only the control variables on the dependent variables are tested here. R&D is highly significant (at the 0.01 level) for the domestic firms and EMNEs. This indicates that R&D is highly related to performance of domestic firms and EMNEs. Firm size is also slightly significant (at the 0.05 level) for domestic firms, indicating that firm size has some effect in firm performance as well. Furthermore, for the EMNEs the control variable leverage is highly significant (at the 0.001 level), meaning that significant funds from debt influence the firm performance of EMNEs. Lastly, the control variables for the MNEs show no high significance.

Model 2, in this model the independent variable PV is added to the base model, in order to test hypothesis 1. Hypothesis 1 is divided in three sub hypotheses, namely: a higher political stability benefits the firm performance of domestic firms and EMNEs less, whereas a higher political stability leads to a positive performance of MNEs. The results show a negative effect for the domestic firms; however this result is not significant. The coefficient of the EMNEs and MNEs are positive, indicating that an increase in political stability leads to an increase in the firm performance of EMNEs and MNEs. Hence, for now, hypothesis 1a and 1c are supported and hypothesis 1b is rejected.

30 careful with these results. Nevertheless, hypotheses 2b and 2c are for now confirmed and hypothesis 2a is rejected.

In model 4 all the variables are added to the regression analyses. Model 4 shows again a negative effect of political stability on the performance of domestic firms, but contradictory to model 3 a decrease in the performance when rule of law increases. For the EMNEs, political stability is still found to have a positive effect on firm performance, whereas rule of law is now shown to have a positive effect on firm performance. For the MNEs the coefficients remain positive, however now they show a significance (at the 0.001 level).

The R-square is used in the regression to see how well the independent variables explain the variation in the dependent variables. There is looked at the adjusted R-square because multiple variables are included in the regression analysis. It can be seen that the value of the adjusted R-square is low, ranging from 0.028 to 0.134. This shows that the models explain around 2.8% to 13.4% of the total variation in firm performance.

31

TABLE 5

Summary of hypotheses and findings

No. Hypothesis Results

1a. Domestic firms in emerging markets are likely to benefit less from changes in political stability

Supported

1b. EMNEs in emerging markets are likely to benefit less from changes in political stability, but less so than the domestic firms

Not supported

1c. MNEs in emerging markets are likely to benefit more from changes in political stability

Supported

2a. Domestic firms in emerging markets are likely to benefit less from changes in rule of law

Not Supported

2b. EMNEs in emerging markets are likely to benefit less from changes in rule of law, but less so than the domestic firms

Not supported

2c. MNEs in emerging markets are likely to benefit more from changes in rule of law

Supported

4.2 Robustness checks

To be able to assure the quality of the results, a robustness check is carried out. According to Neumayer and Plümper (2017), robustness tests provide the researcher with stability of estimates to alternative plausible model specifications.

33

5. Discussion

A total sample of 665 different domestic firms, EMNEs and MNEs have been used to investigate whether the quality of home country institutions have a different impact on the firm performance of these three types of firms. With the use of two hypotheses regarding the political stability and rule of law, this research has aimed to shed light on the relationship between home country institutions and firm performance First, it was hypothesized that domestic firms and EMNEs are will benefit less by an increase in the political stability and absence of violence indicator. This was confirmed for the domestic firms. Due to the increasing quality of the institutions, MNEs will perceive less risk by entering that country and competition will increase (Cherchye and Verriest, 2016), resulting in a lower firm performance. The EMNEs are shown to be positively influenced by an increase in political stability. A reason could be that they will be able to do more business with internationals resulting in a higher firm performance. Furthermore, it was predicted that the effect would be more significant for domestic firms, which was not the case. Moreover, it was predicted that the firm performance of MNEs will increase as a result of a better political stability, since they are not used to operating in unstable environments and will be able to rely more on the enforcement of contracts. This was supported by the results.

Furthermore, hypothesis 2 reflects the effect of an increase in rule of law on the type of firm. The results show a positive effect for the domestic firms and MNEs and a negative effect for the EMNEs. This is not in line with hypothesis 2a, since it was expected that a better rule of law leads benefits domestic firms less, mainly because of the increasing competition and firms not being able to rely on the informal markets anymore. A final remark is that all results are shown to be insignificant, meaning that we need to be careful with the results. A way to reach higher significance would be to increase the sample size.

34 5.1 Contributions

This study contributes to previous literature in multiple ways. First, barely any research has focused explicitly on whether the legal and political aspects of a country influence the firm performance of domestic firms, EMNEs and MNEs differently. Four studies come close to this issue; Chacar and Vissa, 2005, McGahan and Victer (2010), Makino et al. (2004) and Cherchye and Verriest (2015). However, McGahan and Victer (2010) do not state in the results whether the relation between institutions and firm performance is positive or negative. Makino et al. (2004) only include Japanese multinational firms in their sample, which limits the generalizability of the results. Moreover, Chacar and Vissa (2005) have looked at the institutions as an aggregate on firm performance. Lastly, Cherchye and Verriest (2016) only look at the effect of formal institutions on firm performance. Second, all the above papers focus on the developed economies or a combination of developed and emerging markets. The focus of this research was only on the emerging markets and the effect of the home-country’s institution on the different types of firms. Third, while McGahan and Victer (2010) have made a distinction between MNEs and Domestic firms, the other papers have not included other firms to test the relationship. Fourth, this study has analyzed the effect of the individual dimensions of the institutional environment on performance instead of the aggregate. Fifth, in the paper of Cherchye and Verriest (2016) the whole manufacturing sector is used to determine how it is effected by the institutions, however in this research there was looked at only one industry, namely the textile industry. Lastly, extending the research above (McGahan and Victer, 2010; Makino et al., 2004; Cherchye and Verriest, 2015), more evidence is provided regarding the important role that the political and legal aspects have on the performance of domestic firms, emerging market multinationals (EMNEs) and foreign multinationals from the textile industry and emerging markets.

5.2 Limitations and Future Research

35 Therefore, future research could focus on other industry types to see its effect on the different firms. Fourth, some firms had to be removed from the databases because no data was available to calculate return on assets. Therefore, future research could use other databases. Fifth, we must also be cautious with interpreting the results, as the variables used to measure firm performance and home country institution may provide shortcomings. Firm performance could be measured in multiple ways other than using ROA, therefore future research could focus on other ways to measure this variable. Moreover, the WGI indicators are often used in research to measure home country institutions, however some researchers have doubted the reliability of these indicators (e.g. Knack, 2006). Therefore, other indicators should be included for further research. Furthermore, most importantly, the results show no significance. The reason could be the smaller sample size of this research, therefore further research might look at more firms from more countries, in order to achieve significant results. Lastly, only six control variables have been used in this research, whereas firm performance is influenced by many things. Future research could identify other control variables that affect the relationship.

5.4 Conclusion

36

References

Abor, J. (2005). The Effect of Capital Structure on Profitability: An Empirical Analysis of Listed Firms in Ghana. Journal of Risk Finance. 6(5): 438-445.

Ahlstrom, D., Young, M. N., Nair, A., Law, P. (2003). Managing the Institutional Environment: Challenges for Foreign Firms in Post-WTO China. SAM Advanced Management

Journal. 68(1): 41–49.

Alesina, A., Perotti, R. (1993). Income Distribution, Political Instability, and Investment.

European economic review. 40(6): 1203-1228.

Amorós, J., Masferrer, M. (2010). Dynamics of Entrepreneurship and Quality of Government Institutions: An Evaluation using Cross-Country Data. Fourth Global Entrepreneur

Monitor Research Conference, October 2010.

Arnold, D. J., Quelch, J. A. (1998). New Strategies in Emerging Economies. Sloan

Management Review. 40(1): 7-20.

Atan, R., Alam, M.M., Said, J., Zamri, M. (2018). The Impacts of Environmental, Social and Governance Factors on Firm Performance: Panel Study of Malaysian Companies.

Management of Environmental Quality: An international journal. 29 (2): 182-194.

Bhaumik, S.K., Dimova, R., Kumbhakar, S.C., Sun, K. (2012). Does Institutional Quality Affect Firm Performance? Insights from a Semi-Parametric Approach. William

Davidson institute working papers series. Wp 1029, William Davidson Institute at the

University of Michigan.

Beverelli, C., Fiorini, M., Hoekman, B. (2017). Services Trade Policy and Manufacturing Productivity: The Role of Institutions. Journal of international economics. 104(1): 166-182.

Brouthers, L., Marshall, V., Keigh, D. (2016). Solving the Single-Country Sample Problem in Cultural Distance Studies. Journal of International Business studies. 47 (4): 471-479. Busenitz, L. W., Gomez, C., & Spencer, J. W. (2000). Country Institutional Profiles: Unlocking

37 Chacar, A., & Vissa, B. (2005). Are Emerging Economies Less Efficient? Performance Persistence and the Impact of Business Group Affiliation. Strategic Management

Journal. 26(10): 933-946.

Chan, C.M., Isobe, T., Makino, S. (2008). Which Country Matters? Institutional Development and Foreign Affiliate Performance. Strategic management journal. 29(11): 1179-1205. Cherchye, L., Verriest, A. (2015). The Impact of Home-Country Institutions and Competition

on Firm Profitability. Journal of international business review. 25(4): 831-846. Chintakanada, A. (2016). The Effects of Home Country Political and Legal Institutions on Firm

Geographic Diversification Performance. Global Strategy Journal. 6(2): 105-123. Chow, G.C. (1960). Tests of Equality between Sets of Coefficients in Two Linear Regressions.

Econometrica. 28(1): 591-605.

Coad, A., Segarra, A., Teruel, M. (2013). Like Milk or Wine: Does Firm Performance Improve with Age. Structural change and economic dynamics. 24(1):173-189.

DiMaggio, P. J., & Powell, W. W. (1983). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. America Sociology Review. 48(2): 147–160.

Driffield, N. L., Mickiewicz, T., Temouri, Y. (2013). Institutional Reforms, Productivity and Profitability: From Rents to Competition? Journal of Comparative Economics. 41(2): 583–600.

Elango, B., Lahiri, S. (2014). Do Market-Supporting Institutional Characteristics explain Firm Performance? Evidence from Emerging Markets. Thunderbird international business

review. 56(2): 145-155.

Estrin, S., Prevezer, M. (2011). The Role of Informal Institutions in Corporate Governance: Brazil, Russia, India, and China Compared. Asia Pacific Journal of Management. 28(1): 41–67.

38 Fuller, W.A. (1991). Simple Estimators for the Mean of Skewed Populations. Statistic Sinica.

1(1): 137-58.

Ghemawat P, Khanna T. (1998). The Nature of Diversified Groups: A Research Design and Two Case Studies. Journal of Industrial Economics. 46(1): 35–62.

Goldszmidt, R. G. B., Brito, L. A. L., de Vasconcelos, F. C. (2011.) Country Effect on Firm Performance: A Multilevel Approach. Journal of Business Research. 64(3): 273- 279. Greenwood, R., & Hinings, C. (1996). Understanding Radical Organizational Change: Bringing Together the Old and New Institutionalism. Academy of Management Review. 21(4): 1022-1054.

Greve, H.R. (2008). A Behavioral Theory of Firm Growth: Sequential Attention to Size and Performance Goals. Academy of Management Journal. 51(3): 476-494.

Guo, B., Wang, J., Wei, S.X. (2018). R&D Spending, Strategic Position and Firm Performance.

Frontiers of business research. 12(1): 1-19.

Henisz, W.J. (2000). The Institutional Environment for Multinational Investment. Journal of

Law, Economics and Organization. 16(1): 334–364.

Henisz, W.J., Zelner, B.A. (2005). Legitimacy, Interest Group Pressures, and Change in Emergent Institutions: The Case of Foreign Investors and Host Country Governments.

Academy of Management Journal. 30(2): 361–382.

Hoskisson, R. E., Eden, L., Lau, C. M. and Wright, M. (2000). Strategy in Emerging Economies. Academy of Management Journal. 43(1): 249–67.

India Brand Equity Foundation (2019). Textile Industry & Market Growth in India. Retrieved from: < https://www.ibef.org/industry/textiles.aspx >

Jones, C., Hesterlv, W. S., Borgatti, S. P. (1997). A General Theory of Network Governance: Exchange Conditions and Social Mechanisms. Academy of Management Review. 22(4): 911-945.

39 Khan, A., Muttakin, M. B., Siddiqui, J. (2013). Corporate Governance and Corporate Social Responsibility Disclosures: Evidence from an Emerging Economy. Journal of Business

Ethics. 114(1): 207-223.

Khanna, T., Palepu, K. (1998). Institutional Voids and Policy Challenges in Emerging Markets.

The Brown Journal of World Affairs. 5(1): 71-78

Knack, S. (2006). Measuring Corruption in Eastern Europe and Central Asia: A Critique of the Cross-Country Indicators. Ser. Policy research working paper, 3968. World Bank, Development Research Group, Public Services Team.

Knack S, Keefer P. 1997. Does Social Capital have an Economic Pay-Off: A Cross-Country Investigation. Quarterly Journal of Economics. 14(1): 275–286.

Lee, J. (2009). Does Size Matter in Firm Performance? Evidence from US Public Firms.

International journal of the economics of business. 16(2):189-203.

Luo, Y., Tan, J.J. (1998). A Comparison of Multinational and Domestic Firms in an Emerging Market: A Strategic Choice Perspective. Journal of International Management 4(1): 21-40.

Luthfiah, A., A., Suherman. (2018). The Effects of Financial Performance Toward Firm Value With Ownership Structure As Moderating Variable. Journal of Business and

Behavioural Entrepreneurship. 2 (1): 18-27.

Makino, S., Isobe, T., & Chan, C. M. (2004). Does Country Matter? Strategic Management

Journal. 25(10): 1027–1043.

McGahan, A. M., Victer, R. (2010). How Much Does Home Country Matter to Corporate Profitability? Journal of International Business Studies. 41(1): 142–165.

McKinsey and Company (2019). The State of Fashion. Retrieved From: <https://www.mckinsey.com/~/media/McKinsey/Industries/Retail/Our%20Insights/Th e%20influence%20of%20woke%20consumers%20on%20fashion/The-State-of-Fashion-2019.ashx>

McMillan, J. (2007). Market institutions. In L. Blume & S. Durlauf (Eds.), The new Palgrave

40 Meyer, K. E. (2001). Institutions, Transaction Costs, and Entry Mode Choice in Eastern Europe.

Journal of International Business Studies. 32(1): 357–67.

Meyer, K.E., Peng, M.W. (2016). Theoretical Foundations of Emerging Economy Research.

Journal international business studies. 47(1): 3-22.

North, D. (1990). Institutions, Institutional Change and Economic Performance. New York: Cambridge University Press.

Neumayer, E., Plümper, T. (2017). Robustness Tests for Quantitative Research (Methodological tools in the social sciences). Cambridge University Press. Doi: 10.1017/9781108233590.

OECD (2019). Business Insights on Emerging Markets. Retrieved from: <https://www.oecd.org/dev/BI_2019.pdf>

Pek, J., Wong, O., Wong, ACM. (2018). How to Address Non-Normality: A Taxonomy of Approaches, Reviewed and Illustrated. Frontiers in psychology. 9:2104.

Pelle, S. (2007). Understanding Emerging Markets: Building Business BRIC by Brick. New Delhi: SAGE Publications Pvt. Ltd.

Peng M.W. (2003). Institutional Transitions and Strategic Choices. The academy of

management review. 28(2): 275-296.

Peng, M. W., Sun, S. L., Pinkham, B., Chen, H. (2009). The Institution-Based View as a Third Leg for a Strategy Tripod. Academy of Management Perspectives. 23(3): 63–81. Peng, M.W., Wang, D., Jiang, Y., 2008. An Institution-Based View of International Business

Strategy: A Focus on Emerging Economies. Journal of international business studies. 39 (5): 920–936.

Ring, P. S., Bigley, G., D'Aunno, T., & Khanna, T. 2005. Perspectives on How Governments Matter. Acadamy of management review. 30(2): 383-396.

Scott, W.R. (1995), Institutions and Organizations, Sage, Thousand Oaks, CA.

Stock, J.H., Watson, M.W. (2007). Introduction to Econometrics. 2nd ed., Boston: Pearson

41 Tybout J. 2000. Manufacturing Firms in Developing Countries: How Well Do They Do, And

Why? Journal of Economic Literature. 38(1): 11–44.

University of Utah (2020). Checking Normality of Residuals. Retrieved from: <https://campusguides.lib.utah.edu/c.php?g=160853&p=1054157>

World Bank. Worldwide Governance Indicators (WGI) Project. The World Bank, Washington DC, USA (2018).

Wright, M., Filatotchev, I., Hoskisson, R. E., and Peng, M. W. (2005). Strategy Research in Emerging Economies: Challenging the Conventional Wisdom. Journal of

Management Studies. 42(1): 1–33.

Wright, P., Kroll, M., Pray, B., Lado, A. (1995). Strategic Orientations, Competitive Advantage and Business Performance. Journal of Business Research. 33(1): 143-151.

Xu, D. and Shenkar, O. (2002). Institutional Distance and the Multinational Enterprise.

Academy of Management Review. 27(1): 608–18.

Young, M.N., Tsai, T., Wang, X., Liu, S., Ahlstrom, D. (2014). Strategy in Emerging Economies and the Theory of the Firm. Asia Pacific Journal of Management. 31(2): 331-354.

Zajac, E., Kraatz, M., Bresser, R. (2000). Modeling the Dynamics of Strategic Fit. Strategic

42

Appendices

Appendix A – Preliminary Checks

In order to use a linear regression, some preliminary checks must be done. In total four assumptions need to be met before executing a regression analysis.

First, in a linear regression the relationship between the dependent and independent variables need to be linear. Therefore, it is important to look for outliers. This linearity assumption is tested by creating scatter plots of the dependent variable with the independent variables for each type of firm used in this research. The scatterplots revealed that there is a linear, zero slope, which is sometimes slightly negative. Furthermore, some outliers where identified by using the command avplot after the regression for the independent variables. In order to correct for them a winsorization method was used. This means that the dataset is trimmed and that the outliers are replaced with the nearest good data (Fuller, 1991). This is corrected for the 1st and 99th percentile and is done in STATA using the winsor command. There is also looked at the skewness and kurtosis value to see whether the variables need to be converted to logarithms. Only the ROA for the MNEs was positively skewed (see figure 3).

FIGURE 3

Histogram of ROA distribution of MNEs

43 thereafter the predict command is used to create residuals. Then, the qnorm command is used to see whether the quantiles of the variable are the same as the quantiles of a normal distribution.

FIGURE 4

Distribution of residuals PV

DOMESTIC EMNEs MNEs

FIGURE 5

Distribution of residuals RL

DOMESTIC EMNES MNEs

44 Another assumption that needs to be tested for is homoscedasticity. Homoscedasticity means that the variance of the error term is constant of various values of the IVs. If this is not the case then the OLS assumption does not hold and the values are heteroscedastic, which means that the error term does not have a constant variance. Consequences are that the estimators are not efficient, the standard errors are biased and that the t-statistic and significance tests are unreliable (Stock and Watson, 2007). This is tested by using the Breusch-Pagan test, the following results were found:

TABLE 6

Breusch-Pagan Test for Heteroskedasticity PV

DOMESTIC EMNEs MNEs

Chi2(1) 6.91 75.98 2.05

Prob > chi2 0.0086 0.0000 0.1522

TABLE 7

Breusch-Pagan Test for Heteroskedasticity RL

DOMESTIC EMNEs MNEs

Chi2(1) 0.22 66.57 7.13

Prob > chi2 0.6364 0.0000 0.0076

The null hypothesis is that the residuals are homoscedastic. As can be seen in the figures, the null is rejected for all except domestic firms and MNEs, which means that the residuals are for most heteroscedastic (Stock and Watson, 2007). To correct for this in the analysis, robust is added to the regress command. This shows STATA that heteroskedasticity robust standard errors need to be used.

45

TABLE 8

Variance Inflation Factors

DOMESTIC EMNEs MNEs

Variable VIF VIF VIF

RL 9.63 33.38 101.78 GDP 8.63 36.46 80.29 PV 3.74 2.81 3.33 Firm Size 1.71 1.16 6.86 Leverage 1.11 1.04 2.20 Firm Age 1.05 1.20 1.06 Firm Growth 1.05 1.01 3.29 R&D 1.00 1.03 0 Mean VIF 3.46 9.82 28.40

VIF values for the independent variables and control variables

46

TABLE 9

Variance Inflation Factors without GDP

DOMESTIC EMNEs MNEs

Variable VIF VIF VIF

RL 3.73 2.67 3.31 PV 3.62 2.61 1.58 Firm Size 1.23 1.00 2.21 Leverage 1.08 1.04 1.81 Firm Age 1.05 1.20 1.04 Firm Growth 1.04 1.01 2.57 R&D 1.00 1.02 0 Mean VIF 1.82 1.50 2.09

VIF values for the independent variables and control variables

47 Appendix B – Robustness Checks

In this section the robustness checks can be viewed, that were performed to check the reliability of the results.

TABLE 10

Results of regression without control variables

Domestic EMNEs MNEs

Model 1 Model 2 Model 3 Model 1 Model 2 Model 3 Model 1 Model 2 Model 3

Independent variables Political Stability and Absence of Violence -66.68*** (12.41) -10.74 (10.48) 22.39** (6.846) -16.91 (9.501) 29.36 (16.18) 7.439 (22.07) Rule of Law -81.52*** (9.442) -87.88 (12.14) -26.31*** (4.477) -36.78*** (8.316) -46.24*** (11.53) -43.54** (15.70) Constant 62.61*** (11.63) -0.0433 (0.814) -9.837 (9.639) 19.80** (6.475) -1.389 (0.733) -17.02 (8.871) N 366 366 366 ₂₇₈ 278 278 ₃₀ 30 30 R squared 0.196 0.436 0.439 _0.065 0.154 0.166 _0.090 0.308 0.946 Adj. R squared 0.194 0.435 0.435 _0.061 0.151 0.160 _0.057 0.284 0.262 F 20.69 17.32 17.31 _12.97 12.34 12.27 _22.90 19.96 20.26

* (significant at the 0.05 level)

** (significant at the 0.01 level)