“The introduction of Corporate Governance Codes and the effect on foreign investment: a study of Non-OECD countries”

(1)

“The introduction of Corporate Governance

Codes and the effect on foreign investment:

a study of Non-OECD countries”

Marieke van Delden - 1508075 Double Master Degree Program: International Financial Management

Rijksuniversiteit Groningen

Faculty of Management and Organization First Supervisor: Prof. Dr. C.L.M. Hermes

Second Supervisor: Dr. W. Westerman Landleven 5

9747 AD Groningen, the Netherlands Uppsala Universitet

Department of Economics PO Box 513

(2)

Abstract

An increasing number of countries adopted Corporate Governance Codes. Codes can be adopted as a private initiative. Another model is on a comply, or explain basis and is directly linked with the securities market. The adoption of codes is voluntary.

Developing countries need foreign capital to experience growth. In order to attract foreign capital, a proper CG model is needed. CG Codes are intended to compensate for deficiencies in a country’s corporate governance system. The question is whether CG codes help to attract foreign capital? This results in the following research question:

“Does the introduction of corporate governance codes in non-OECD countries lead to a change in foreign investment?”

All non- OECD countries that have adopted Corporate Governance principles were selected for this research. First, FDI data are observed from 1995 (first introduction of CG codes) up until and including 2006 on a quarterly basis. Of a total of 78 countries, 29 countries adopted a code. These 29 countries were spread over various income levels and geographical areas. This research is the first that relates the adoption of a GC code to the increase in FDI and FPI in developing countries. Overall, the results show that the introduction of a code shows a significant relationship towards FPI, but not to FDI. Confirmation of hypothesis 2 is found (The introduction of corporate governance codes lead to an increase in FPI).

(3)

1. Introduction

The decision to invest in a foreign market depends on many factors, one of which is a good corporate governance system. The Enron case is a classical example that stresses the importance of good corporate governance in a country (Fremond & Capaul, 2002). Corporate Governance Codes support a country in improving their CG system and are therefore of great importance to foreign investors. The adoption of Corporate Governance Codes may in particular be relevant to countries in which information is harder to obtain. In this light, the adoption of CG codes may be especially relevant to the decision to invest in developing countries.

There has been research on the relation between CG Codes and FDI. Fernández-Rodríguez et al (2004) show that the Spanish stock market reacted positively to firm announcement of (partial) compliance with the code for the period 1998-2000. Also other research finds that there is a positive relation between the adoption of a code and economic growth or positive stock reactions (Levine, 1999).

This paper focuses on the relation between the adoption of codes and the increase of foreign investment in Non-OECD countries. For this research, Non-OECD countries that have adopted a code and Non-OECD countries that have not adopted a code will be used.

“Does the introduction of corporate governance codes in a developing country lead to a change in foreign investment?”

(6)

but do not relate the finding directly to foreign investment. Others focus on the importance of good governance codes on a firm level (Gupta & Parua, 2006; Klapper & Love, 2002). This paper will contribute to empirical research since no emerging economies have been tested on the relation between foreign investment ( both FDI and FPI) and the adoption of a CG code in a cross country analysis.

(7)

2. Theoretical Background

This theoretical framework will elaborate on the definitions used for Corporate Governance. It will then explain when codes were first introduced and why they have spread around the world. It will clarify the relation to foreign investment and developing countries. Finally, information will be provided from prior research on corporate governance codes and the effect on foreign investment.

2.1 Corporate Governance

Corporate governance is still not defined in a proper way (Gupta & Parua, 2006). It can be defined narrowly as the relationship of a company to its shareholders or, more broadly, as its relationship to society (Gupta & Parua, 2006). Still, many other theorists add subtle differences to the definition of Corporate Governance (Tirole, 2001). An example of a more narrow perspective is the definition by Schleifer and Vishny (1997), who defined corporate governance as follows: “Corporate Governance deals with the ways in which suppliers of finance to corporations assure themselves of getting a return on their investment”. An example of a more broader perspective is written by Aoki (2000): “corporate governcance concerns the structure of rights and responsibilities among the parties with a stake in the firm”.

(8)

2.2 Corporate Governance Codes

Corporate Governance Codes are a set of best practices intended to compensate for deficiencies in a country’s corporate governance system regarding the protection of shareholder rights. (Aguilera & Cuervo Cazurra, 2004). Cuervo (2002) defines a CG code as “a set of norms that regulate behaviour and structure of the board of directors”. According to Thomsen (2001), codes are sets of recommendations on CG, regarding, amongst others, structure, organization and decision process of the board, dealing with executive pay, information disclosure, and investor relations.

Generally, a code has two objectives: improving board governance quality and increased accountability of companies to shareholders while maximizing shareholder or stakeholder value (Aguilera & Cuervo Cazurra, 2004). Codes can be adopted as a private initiative and are then voluntary; these are often referred to as moral codes (Wymeersch, 2006). Another model is on a comply, or explain basis. Companies may deviate from recommendations but need to explain why they do that (Thomsen, 2006). The codes are also called ‘soft laws’ and support a country in separation of ownership and control, increased transparency, elimination of fraud and increased foreign investment. (Berglöf & Pajuste, 2005; Aguilera & Cuervo Cazurra, 2004; Thomsen, 2001). The advantage of CG codes is that they assure better governance practices, without a revolution of a country’s governance structure. Apart from the efficiency reasons, codes are also issued for legitimacy reasons, which means that a country can prove that it is committed to accountability and integrity. Issuing codes for legitimacy reasons is understandable since countries with good corporate governance practices attract more foreign investment. (Aguilera & Cuervo Cazurra, 2004).

(9)

are not so much legal, but more financial and economical. Legal sanctions are complex, since it is difficult to assess to what extent a company really adheres to a code. Market pressure and fear for reputation and financial damage makes businesses to adhere to a code.

2.3 Spreading of Corporate Governance Codes

The first code of good governance was issued in the 1970s in the USA (Aguilera & Cuervo-Cazurra, 2004). The USA continued to issue codes and also other countries created codes of good governance. Especially after the issue of the Cadbury Committee Report in 1992, which was seen as a landmark, more codes of best practices evolved over time (Aguilera & Cuervo-Cazurra, 2004; Thomsen, 2001). The scandals that took place in the 1990’s contributed to the increase of attention for good corporate governance.

In the past years, there has been a trend towards convergence of guidelines to common principles, transparency and accountability across countries. This process leading to convergence is also known as ‘isomorphism’. (Aguilera & Cuervo-Cazurra, 2004). One of the factors leading to cross national isomorphism is global institutional pressure, or the emergence of common organizational practices over time (DiMaggio & Powell, 1983). Even though there is literature that supports the view of a general trend towards worldwide convergence of the adoption of codes, there are several factors - such as political pressure- that cause a delay in the convergence of corporate governance codes (Reid, 2003). In some countries, companies de-list after the introduction of the code. A general complaint is the increased costs of financial reporting, both annually and quarterly (Haroon, & Kozhich, 2006). In other cases, codes are implemented, however the external pressure to adopt codes is lacking exactly in those environments where it is needed most. (Thomsen, 2001). In summary, codes are adopted only when the benefits outweigh the cost.

(10)

codes are more likely to be issued in countries where foreign investors are present. An interpretation related to this thesis can be as follows: if a country needs foreign investment, it should adopt a code. Thus, by introducing CG codes, foreign investment will increase (Liljeblom & Löflund, 2005; Thomsen, 2006). This paper will investigate the effect of the adoption of corporate governance codes on foreign investment.

2.4 Foreign Investment

Private investment flows to developing countries grew from $20 billion to $200 billion per year between 1980s and 1997. Latin America and East Asia received most of that. (Rueda-Sabater, 2000). Nowadays, resources and capital can be sourced in global markets much more easily compared to many years ago (Gupta & Parua, 2006). On one hand, transition economies are opened as the result of political (Baniak et al, 2005). On the other hand, other factors may play a role in the increased openness of transition economies. This research aims to find the relation between the adoption of codes and foreign investment, such as FDI and FPI. The difference between the two types of foreign investment dealt with in the research is explained below.

(11)

Both FDI and FPI contribute to economical growth and are therefore especially important to emerging market economies (Zorska, 2005).

(12)

(13)

2.5 Hypothesis

The aim of the research is to test whether there is a relationship between the adoption of code and foreign investment.

As emerging markets are generally looking for growth, a proper legal system, including laws and enforcement are crucial (Allen, 2005). This research will, however, not focus on enforcement. In order to attract investment from abroad, countries have to gain confidence from foreign investors. One way to gain that confidence is to offer favourable (economical) conditions, where transparency and accountability to shareholders are existent. Fremond & Capaul (2002) argue that over time, countries with no code, will encounter difficulties in attracting foreign capital. The question that has not yet been answered in existing research is whether the adoption of a code results in increased foreign investment for developing countries. Developing countries are in this research NON- OECD countries.

The aim of this research is to analyse the relation between the adoption of CG principles and foreign investment flows. Since there is not much existing literature that includes a cross country analysis concerning the relation between CG codes and foreign investment for developing countries, I propose the hypothesis of this paper:

Hypothesis 1: The adoption of corporate governance codes leads to an increase in FDI in Non-OECD countries

Hypothesis 2: The adoption of corporate governance codes leads to an increase in FPI in Non-OECD countries

(14)

3. Methodology

The purpose of this paper is to find out if the foreign investment flows of a non-OECD country increase after the adoption of CG principles. This research includes non- OECD countries that have adopted a CG code as well as non-OECD countries that have not adopted a code.

3.1 Sample

The data basically consists of two groups. First, non-OECD countries that have adopted Corporate Governance principles were selected for this research. Second, all non-OECD countries that have not adopted a code were selected. For the first group, only Taiwan has been excluded due to the lack of data. Thus, OECD countries, such as Hungary or Turkey are not taken into consideration for this research.

The World Bank classifies countries based on income level and geographical area. Income level is determined according to GNI per capita (2006) and is categorized into various income groups1 accordingly (World Development Indicators, The World Bank Group). The low and middle income countries are referred to as developing economies. Geographical area refers to the continent the country is situated in2_{. Appendix 1 gives an}

overview of all the countries that adopted CG principles per region and income level (in bold) and the countries in the same income group and geographical area. The group Sub Saharan Africa (low income) is quite large compared to other groups. In order to be able to compare Kenya to non-code countries, I chose to only include its surrounding countries (Somalia, Ethiopia, Uganda, Tanzania and Sudan).

1_{High Income (>$11,116), Upper Middle Income OECD, Upper Middle Income Non- OECD} (>$3,596-11,116), Middle Income, Lower Middle Income ($906-$3,595), Low Income (<$905), Heavily Indebted Poor Countries, Least Developed Countries

(15)

The introduction of CG codes has only begun approximately during the late ninetees. Many developing countries adopted such codes only a few years ago. The goal of this research is to find out if there is a relationship between the adoption of a CG code and foreign investment. It’s not realistic to compare the average FDI and FPI rates of countries that adopted a code to countries that did adopt a code in one point of time. Therefore, each country in the sample consists of quarterly data on FDI and FPI and the moment of introduction of the code is to be known. Many countries that have adopted a CG code, do have this type of data available. However, some countries that have not adopted CG codes do not offer quarterly data related to investment. This research therefore consists of quarterly data only when they are available. In case where these are not available, the year is divided by four to get the results per quarter. The total database results in 12 years (1995-2006) times 4 (quarters) equals 48 measurements per country. In total, I included 29 countries that have adopted a code and 49 countries that did not adopt a code (78 countries; 3744 observations in total). Instead of selecting 29 countries, I chose to increase the number of countries that did not adopt a code to 49.

3.2 Variables

(16)

supported by many other scholars (Bhavan et.al 2011; Leitao & Faustino 2010; Leitao 2010; Hailu 2010; Schneier & Matei 2010; Mohamed & Sidiropoulos 2010). Trade openness is also used by many scholars to explain FDI ( Jadhav, 2012). Poshakwale & Chandra (2011) add that the quality of governance and good investor protection lead to an increase in FPI. For the purpose of this particular research, I use the most important Location determinants to explain foreign investment. As Dunning (2001) states, Location variables can be the same to explain a change in both FPI and FDI. Therefore, I use the same regression model to explain FDI and FPI.

FDI or FPI /GDP = α + β adopt CG code+ β GDP per capita + β openness to trade + β Corruption level

The question that remains is how to measure the corporate governance variable? The main question to be answered is: does the adoption of a CG code result in a change of foreign investment. This can be measured by testing whether the fact that a country adopted a code contributed to a higher foreign investment. Another way of testing the relation between a CG code and foreign investment can be to analyse whether or not countries show a higher foreign investment during the time that they adhered to a code. Or, finally, the higher the cumulative number of months since the adoption of a code, the higher foreign investment will be. If this is true, one can state that it takes time to attract foreign investors after adoption of a CG code. Since I wanted to test the corporate governance variable in these three ways, the regression model is now divided into three different ones:

• FDI or FPI /GDP = α + β adopt CG code (yes/no)+ β GDP per capita + β openness to trade + β Corruption level

• FDI or FPI /GDP = α + β adopt CG code per quarter+ β GDP per capita + β openness to trade + β Corruption level

(17)

3.2.1 Dependent variable: Foreign Investment

For all the countries in the sample (that have and have not accepted CG principles), the FDI and FPI rates are collected. Both FDI and FPI rates are divided by a country’s GDP. First, FDI data are observed from 1995 (first introduction of CG codes) up until and including 2006 on a quarterly basis. Most GDP data are available on a quarterly basis. Only for three countries, I have used the yearly average for the four quarters in this year.

3.2.2 Independent variable: corporate governance code

For this research, all non- OECD countries that have adopted Corporate Governance principles will be used, excluding Taiwan. The aim is to find out if the FDI and FPI flows of a non-OECD country change or even increase after the adoption of CG principles. Therefore, we need the moment - the month and year- of introduction of the CG codes in each particular country. This information can be found on the web site of the European Corporate Governance Institute3. Every country is 48 times entered in a SPSS database based on 12 years * 4 quarters per year (from 1995 u.i. 2006). I decided to use the variable in three ways:

• first, I created a variable named “Adopt Code”. I allocated a “yes” to all the countries that adopted a CG code regardless of the moment of introduction. For example, Banglades adopted a code in March 2004. This means I allocated a “yes” to all the quarters (Q1 1995 – Q4 2006). A country that did not adopt a code was given a “no”for all 48 quarters.

• Second, the variable “Adopt Code per Quarter” was inserted and given a ‘yes’ in the quarters that a country adopted a code and a ‘no’ in the quarters that the country did not (yet) adopt a code. For example: all quarters before and including 2004 Q1: “no”; 2004 Q2: “yes”; 2004 Q3: “yes”. I only stated a “yes” when all the months in the quarter represent the adoption of a code. Thus, when a code is

(18)

introduced in March 2004, Q1 2004 is “no”, while Q2 is “Yes”. The outcome of the regression model now shows whether the quarters with a “yes” show different investment figures compared than the quarters that show a “no”.

• Third, I expect that when a country adopts a code, foreign investment will not increase with an immediate effect, but rather with a delay. Therefore I added the variable “Months cumulative”, which represents the number of months since the introduction. For example, Bangladesh introduced the CG code in March 2004. Then, Q2 2004 would state “3” months since the introduction of the code, Q3 2004 “6” months and Q4 “9” months since introduction of the code. This way, we can predict whether the number of months since introduction of a code relates to higher investment.

Concluding, the adoption of a code will be measured in three different ways and could result in three different outcomes. 1. countries that adopted a code at some point in time, show higher foreign investment compared to countries that do not. 2. the quarters in which a country adheres to a code, shows higher foreign investment than the quarters during which a country did not adhere to a code. 3. The cumulative number of months leads to increased foreign investment.

Revisions and, or additions to a code are not taken into account in his research. 3.2.3 Control Variables

(19)

governance theories: corruption. The variables relate to either the OLI framework or Corporate Governance literature.

3.2.3.1 Income per Capita

One economic variable in explaining institutional quality is income per capita. (Islam, 2006). The literature shows that when institutional quality / a good corporate governance system is in place, there will be an increase in foreign investment. Since the late 1990’s, growth rates and income per capita became important in cross country analysis (Bénassy-Quéré et al, 2007; Balasubramanyam et al,1996). Market size is composed of income level per capita and absolute population size. (Firebaugh, 1983). By including this variable, market size is covered in the regression analysis.

3.2.3.2 Openness to trade

For this research, openness to trade is especially important for financial sector development. (Berglöf & Claessens, 2004.) Also other theorists argue that trade openness contributes to FDI inflows (Kinoshita & Campos, 2003; Bevan & Estrin, 2004). Openness to trade is calculated as the ratio of exports plus imports to GDP. (Islam, 2006).

3.2.3.3 Corruption

(20)

of uncertainty it creates. (Uhlenbruck et al, 2006). Chipalkati (2007) argues that low levels of corruption significantly draw portfolio equity capital. For this research, data is retrieved from the World Bank website4. The governance scores vary from -2.5 to +2.5. A higher score implies a better governance and lower level of corruption. For this researched I added 2.5 for all the scores, so that the new scores are between 0 and 5, where 0 reflects the weakest governance performance/ highest corruption.

(21)

4. Results

4.1 Descriptives

In order to test the hypothesis, various data have been collected. First, FDI and FPI data were needed to test the relationship between level of foreign investment and the fact whether or not a country has adopted a CG code. Second, the moment of adoption of a code was required to discover it if the longer a country has adopted a code, the more it affects the level of FDI or FPI.

In the sample for this research, a total of 78 countries were used. 29 Countries adopted a code, while 49 did not adopt a code. A comparison of the means was done.

Mean

Adopt_Code FDIIn/GDP FPIIN/GDP

Yes 0,041 0,013

No 0,064 0,007

Total 0,055 0,010

Table 1: Adopt vs not adopted code for FDI & FPI

If the foreign investment rates of countries that have adopted a code are compared with countries that did not adopt a code, it becomes clear that the countries that adopted a code show a higher portfolio rate and a lower direct investment rate compared to countries that did not adopt a code. Obviously, this first insight needs to be investigated further.

(22)

# countries per continent

adopted

code

Continent No Yes Total

East Asia & Pacific 5 7 12

Europe & Central Asia 10 9 19

Latin America 14 5 19

Middle East & North Africa 8 2 10

South Asia 3 4 7

Sub-Saharan Africa 9 2 11

Total 49 29 78

Table2: Sample divided in Continent

# countries per continent adopted code

Income level No Yes Total

High Income (Non-OECD) 8 5 13

Low Income 6 4 10

Lower Middle Income 20 10 30

Upper Middle Income 15 10 25

Total 49 29 78

Table 3: Sample divided in Income level

The resesearch tries to find a relationship between the adoption of a code and foreign investment. In order to run the analysis, a regression is done. Several variables have been collected and included in the SPSS database. In order to get a complete overview of all the variables used, the descriptive statistics are presented below.

Descriptive Statistics Mean Std. Deviation N FDIin_GDP 0,05 0,14 3066 FPIin_GDP 0,01 0,04 2941 Adopt_Code 0,37 0,48 3739 Adoptcode_Quarter 0,14 0,35 3739 # Months with code 5,1 0,25 3737

GDPperCapita 1.835 4.175 3683

Trade_Openness 3,30 0,3 3739

Corruption 2,35 0,02 3739

Table 4: Descriptives

(23)

of a country’s GDP. For the variable “adopt code”, two dummies are assigned. “0” Is the country has not adopted a code, whereas “1” stands for a code adopted. The sample consists of more countries that did not adopt a code, which leads to a mean of 0.37. On average, countries show 5 months since the adoption of a code. GDP per capita is US $ 1,835 per quarter on average in the sample. The average for trade openness is US $ 3.30. Corruption scores are retrieved from the Worldbank and are between -2.5 and + 2.5, where -2.5 is weakest control for governance and implies the level of corruption is high. For this researched I added 2.5 for all the scores, so that the new scores are between 0 and 5, where 0 reflects the weakest governance performance. The average score is 2.35.

4.2 Correlation

For a correlation, the Pearson test cannot be used since this assumes a normal distribution. Also the homogeneity of variance assumption has been violated. Therefore the Spearman test is used.

FDIin_GDP FPIin_GDP Adopt_Code Codequarter

Cumulativem onthswithcod

e GDPperCapita openesstrade Corruption ,109** -,050** ,095** -,003 _,109** ,531** -,001 ,116** ,528** ,996** -,100** ,082** ,019 ,047** ,048** ,388** ,007 ,059** ,068** ,069** -,105** ,128** ,134** ,092** ,050* ,054** ,554** ,122** openesstrade Corruption

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

Spearman's rho FDIin_GDP FPIin_GDP Adopt_Code Codequarter Cumulativemont hswithcode GDPperCapita Table 5: Correlations

(24)

correlation table shows that GDP per capita has a negative relationship with the variable FDI. Corruption is negatively related to the adoption of a GC code. Openness to trade has a strong positive relationship towards FDI.

Correlation is a proper way to find relations between a limited number of variables. However multiple variables may have an influence on the outcome. Therefore a regression analysis is done.

4.3 Regression

Before conducting a regression analysis, the data has to be tested on reliability. According to the general rules (Field, 2005), standardized residuals greater than 3, should lead to excluding the case from the sample. If they are not excluded from the sample, the average could be affected and is a cause for concern. An analysis of the outliers shows that none of the cases is greater than 3 and has no undue influence on the model. The VIF is a measure testing collinearity and indicates if there is a relationship among predictors. The VIF value should be between 0.2 and 10. SPSS did not find problems in the data set used for this research with regard to collinearity.

The regression tables below are used to answer the hypothesis whether the adoption of a code impacts foreign investment. The first three models show the relation towards FDI, while models four to six relate to FPI. Models 1 & 4 add four variables: adopt code, GDP per capita, openness to trade and corruption level, but do not include dummy variables at this stage. Models 2 & 5 include income dummies and models 3&6 include the continent dummies. All countries have 48 observations (12 years * 4 quarters).

(25)

In the regression analysis below, the level of foreign investment is compared between countries that did and did not adopt a CG code (please refer to ch 3.2.2 for a detailed explanation of the CG variable).

• FDI or FPI /GDP = α + β adopt CG code (yes/no)+ β GDP per capita + β openness to trade + β Corruption level

Table 6: Adoption of code (yes/ no) and foreign investment.

Dependent Variable FDI FPI

Independent Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Constant 24.497 93.790 11.157 -8.277 15.904 -5.412

Adopt Code (Yes, No) -0.034* -0.028 -0,019 0.057** 0.055* 0.079** (-2.031) (-1.699) (-1.055) (2.519) (2.411) (3.093) GDP Per Capita 0.018 -0.042 0.036 -0.030 -0.103** -0.034 (0,936) (-1.709) (1.691) (-1.143) (-3.089) (-1.171) Opennes to Trade 0.687*** 0.679*** 0.677*** 0.003 0,006 0,003 (40.215) (39.509) (39.085) (0.115) (0.290) (0.113) Corruption Level -0.07*** -0.141*** -0.057* 0.131*** 0.071* 0.126*** (-3.497) (-5.598) (-2.810) (5.017) (2.069) (4.705) Income Level Dummy vs High Income

Low Income -0.116*** -0.103*

(-3.508) (-2.297)

Lower Middle Income -0.192*** -0.215***

(-4.418) (-3.662)

Upper Middle Income -0,098* -0.160***

(-2.843) (-3.439)

Continent Dummy vs Latin America

Europe & Central Asia 0.052** -0.033

(2.611) (-1.197)

East Asia & Pacific -0.035 -0.021

(-1.776) (-0.770)

Middle East & North Africa -0.011 -0.026

(-0.549) (-0.943) South Asia -0.011 -0.081** (-0.570) (-3.108) Subsaharan Africa 0.084*** -0.013 (4.540) (-0.517) R² Change 0,003*** 0.006*** 0.011*** 0.013*** 0.008** 0.005 Notes:

Standardized Betas reported. Values between brackets are t-values Model 1-3, N=3066; Models 4-6, N=2941

(26)

The adoption of a code leads to a change in FDI

In this regression analysis, the fact whether or not a country adheres to a CG code is not an influential factor of FDI. GDP per capita also shows that there is no significant relationship towards the dependent variable. On the other hand, openness to trade and corruption show a very significant relationship. The model shows that high trade openness results in a higher FDI. Corruption shows a significant negative relationship towards the dependent variable. The dummy variables for income show significant relationships. As far as the continent dummy concerns, only Sub Saharan Africa and Europe & Central Asia are significant.

The adoption of a code leads to a change in FPI

The regression analysis shows proves that the adoption of a CG code have a significant and positive effect on FPI. Openness to trade does not seem to affect FPI significantly. Again, the level of corruption has a significant relationship. The income level dummy for all income levels show significant relationships. Concerning the continent dummy, only South Asia shows a significant relationship.

Concluding, the regression analysis in models 1 & 4 shows that the adoption of a code only has a significant and positive influence on the level of FPI, but not on the level of FDI. When the income level dummies are included (models 2 & 5), the statement remains true. All income levels show significant relationships. When the continent dummies are included in the regression analysis, it is remarkable that Subsaharan Africa and Europe & Central Asia show a significant relationship in model 3 (FDI), while it is South Asia that is significant in model 6 (FPI) .

4.3.2. Model outcome regression 2

(27)

to the quarters during which a country does not adhere to a code. (please refer to ch 3.2.2 for a more detailed explanation of the CG variable).

• FDI or FPI /GDP = α + β adopt CG code per quarter (yes/no)+ β GDP per capita + β openness to trade + β Corruption level

Table 7: Adoption of code per quarter (yes/ no) and foreign investment.

Independent Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Constant 23.736 93.285 11.144 -8.047 16.728 -5.586

Adopt Code (Yes, No; Defined per quarter) -0.030 -0.023 -0.013 0.075*** 0.075*** 0.085*** (-1.850) (-1.362) (-0.749) (3.331) (3.320) (3.531) GDP Per Capita 0.020 -0.041 0.035 -0.034 -0.108*** -0.034 (1.049) (-1.651) (1.665) (-1.288) (-3.236) (-1.178) Opennes to Trade 0.690*** 0.681*** 0.678*** 0.002 0.006 0.002 (40.523) (39.713) (39.311) (0.071) (0.255) (0.085) Corruption Level -0.073*** -0.143*** -0.059** 0.133*** 0.070 0.132*** (-3.673) (-5.747) (-2.953) (5.097) (2.044) (4.966) Income Level Dummy vs High Income

Low Income -0.117*** -0.106*

(-3.563) (-2.382)

(-4.414) (-3.751)

Upper Middle Income -0.099** -0.160**

(-2.874) (-3.462) Continent Dummy vs Latin America

Europe & Central Asia 0.050* -0.024

(2.523) (-0.880)

(-1.814) (-0.928)

(-0.446) (-1.116) South Asia -0.014 -0.070* (-0.785) (-2.793) Subsaharan Africa 0.085*** -0.014 (4.552) (-0.565) R² Change 0.004*** 0.006*** 0.011*** 0.013*** 0.008** 0.004 Notes:

(28)

The adoption of a code does not affect FDI significantly. For models 1-3, openness to trade has a very significant, positive relationship towards the dependent variable. GDP per capita does not seem to be related to FDI, but corruption level does show a significant negative relationship towards FDI. The income level dummies for high and low income countries show significant relationships, while most continent dummies show insignificant relationships.

The adoption of a code has a significant, positive effect on FPI. Also corruption level seems to be an influental factor of FPI. Unlike the models predicting FDI, openness to trade is not related to FPI. Again, GDP per capita shows no relationship towards FPI. Also for FPI, the income dummies show significant relationships. For the country dummies, only South Asia shows a significant relationship.

Concluding, the regression analysis in models 1 & 4 show that the adoption of a code has a significant and positive influence on the level of FPI, but not on FDI. The income level dummies “high income” and “low income” show significant relationships. The continent dummies show various results in FDI versus FPI.

4.3.3. Model outcome regression 3

(29)

• FDI or FPI /GDP = α + β adopt CG code since adoption of code+ β GDP per capita + β openness to trade + β Corruption level

Table 8: Adoption of code (# months) and foreign investment

Independent Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Constant 53.152 92.530 10.952 -7.591 18.325 -5.546

Adopt Code (Count Months) -0.031 -0.019 -0.005 0.074*** 0.079*** 0.087*** (-1.866) (-1.141) (-0.256) (3.311) (3.471) (3.533) GDP Per Capita 0.020 -0.041 0.034 -0.032 -0.109*** -0.030 (1.026) (-1.659) (1.621) (-1.223) (-3.265) (-1.043) Opennes to Trade 0.691*** 0.682*** 0.679*** 0.001 0.005 0.002 (40.563) (39.686) (39.340) (0.031) (0.231) (0.090) Corruption Level -0.073*** -0.143*** -0.059** 0.132*** 0.064 0.132*** (-3.654) (-5.708) (-2.962) (5.041) (1.860) (4.972) Income Level Dummy vs High Income

Low Income -0.117*** -0.111*

(-3.555) (-2.482)

(-4.373) (-3.910)

Upper Middle Income -0.099** -0.164***

(-2.864) (-3.530)

Continent Dummy vs Latin America

Europe & Central Asia 0.048* -0.016

(2.465) (-0.607)

(-1.948) (-1.101)

(-0.413) (-1.173) South Asia -0.016 -0.068** (-0.879) (-2.742) Subsaharan Africa 0.084*** -0.012 (4.526) (-0.460) R² Change 0.004*** 0.006*** 0.011*** 0.013*** 0.009*** 0.004 Notes:

(30)

In this regression analysis, the number of months since the adoption of a code is not an influential factor of FDI. Neither GDP per capita shows a significant relationship towards the dependent variable. Again, openness to trade shows a very significant, positive relationship towards the dependent variable, while corruption level shows a significant negative relationship towards the dependent variable. The dummy variables for income show significant relationships. As far as the continent dummy concerns, only Sub Saharan Africa and Europe & Central Asia are significant.

For FPI, the number of months has a positive and significant relationship towards FPI. In this regression analysis, openness to trade does not seem to affect FPI. Again, the level of corruption has a significant relationship in models 4 & 6. The income level dummy for all income levels show significant relationships. Concerning the continent dummy, only South Asia shows a significant relationship.

Concluding, the regression analysis in models 1 & 4 shows that the number of months since the adoption of a code only has a significant and positive influence on the level of FPI, but not on the level of FDI. When the income level dummies are included (models 2 & 5), the statement remains true: all income levels show significant relationships.

(31)

(32)

5. Discussion

This research is one of few that relates the adoption of a GC code to the change in FDI and FPI in non-OECD countries. By comparing means, table 1 shows that the countries that did adopt a code show a higher foreign portfolio investment rate than countries that did not adopt a code. In the correlation matrix, it becomes clear that the adoption of a code has a significant relationship towards FPI, but not FDI. In concert with the correlation outcome, the regression results show that the introduction of a code shows a significant relationship towards FPI, but not towards FDI as a dependent variable.

The results for the dependent variable “FDI” are in line with Nowak et al (2006) and Thomsen (2001) who argue that there is no effect of adopting a code on foreign direct investment. One of the control variables in the research was “corruption”. Corruption is a significant negative determinant of foreign direct investment. This means that a stronger governance against corruption results in less FDI. This contradicts with a previous study conducted by Hellman et al (2002), who find that corruption reduces FDI inflows. Wijeweera & Dollery (2009) prove that corruption is not significantly related towards FDI. Openness to trade is significant and positive relation towards FDI, which is supported by Jadhav (2012).

The adoption of a code has a significant influence on FPI in all models. Not many researchers have linked he adoption of a corporate governance code as such, with portfolio investment. Some have tried to relate corporate governance with portfolio flows. For example, Chipalkati et al (2007) state that good public governance is an important determinant for portfolio flows to EMCs. The results for the dependent variable “FPI” are also in line with a study conducted by Dahlquist et al (2003). They state that there is a close relation between Corporate Governance and portfolio held by investors.

(33)

governance models must be in place. Klapper & Love (2002) confirm this statement by arguing that corporate governance matter more in countries with weak legal systems. Often, these countries are also low income countries. Those countries are often found in the category “NON-OECD countries”. The regression analyses show that in the regression models where FDI is the dependent variable, low and lower middle income countries show significant results.

The results show that the adoption of a corporate governance code is significantly related to foreign portfolio investment. This means that portfolio investors invest in countries that adopted a code. As a result, these countries have a comparative advantage over countries that did not adopt a corporate governance code. It seems that if a similar country refuses to adopt a code, it is more likely to lag further behind in economic development compared to countries that do adopt a code. Therefore such countries are forced to adopt a code. DiMaggio and Powell (1983) state that such global institutional pressure leads to cross national isomorphism.

This study is rather new since a relating a relative ‘young’ theory as corporate governance codes is combined with the very classical theory of international trade (Smith, 1776). Also the “location” variables are used to explain the relation between the adoption of a code and foreign investment. Furthermore, this research combines FDI and FPI in one paper. Furthermore, the study is concentrated on NON-OECD countries. In summary, this research adds to current literature that the regression model used is a combination of determinants of FDI according to the eclectic paradigm by Dunning (GDP per capita and openness to trade) combined with governance determinants (corruption).

(34)

period of time, others adopt one code after another in a longer period of time (big bang versus gradual implementation).

Adopting a CG code is one step, but when the codes are not enforced in a proper manner, investors will stay away. The most important determinant of economic performance is how effectively agreements are enforced (North, 1991). Enforcement of CG codes distinguishes developed market economies from transition economies. (Berglöf & Claessens, 2004.)

Another implication of code compliance is that an institution is not easy to change and the cost and effort may eventually outweigh the benefits of adopting a code. It is important for countries to see other countries that benefit from code compliance. Code compliance is not a solution to all countries. Low developed countries cannot attract FDI and will lack behind more and more compared to other non – OECD countries. National frameworks often exist, but lack in practice. Voting rights are often not structured well, ownership is often concentrated, insider trading and the monitoring of this is a problem across these countries. I expect that when more countries adopt a code, an increase of leaders and laggards will evolve.

Limitations

The non – OECD countries have adopted CG principles for only a few years. This research therefore functions as an attempt to show whether there may exist a relation between the introduction of CG codes and an increase of FDI, especially for non- OECD countries.

(35)

Some variables have not been taken into account, but may be included in another research. Since the adoption of codes is often voluntary, there exists heterogeneity of organizational practices within national boundaries ( Aguilera & Jackson, 2003). This effect is neglected in a country comparison.

Future research

In a few years, when there is more data available, a similar research should be conducted. Chipalkati (2007) also states that future research could include an examination of the relationship between the level of corporate governance and higher levels of portfolio flows.

Disclosure is another important condition for foreign investors and should be added to the research. Transition economies often disclose only financial information. Publishing non financial information, such as company objectives, off balance sheet commitments, ownership structure, remuneration key figures, risk factors, employee and policy is new to transition economies. (Fremond & Capaul, 2002). A higher level of disclosure leads to improved corporate governance model and thus better opportunities to attract foreign investment, which in turn, leads to economical growth and development.

Another area for future research is relating the quality of corporate governance codes to a change in foreign investment. Thereby, I aim at refining and updates of the code during the years.

(36)

6. Conclusion

Code compliance has been related to foreign investment before, but often amongst developed countries. This research examines this relation for non-OECD countries. This led to the following research question: “Does the introduction of corporate governance codes in a developing country lead to a change in foreign investment?”

This research combines corporate governance problems related to the agency problem with foreign direct investment theories, such as Dunning’s OLI framework. This research did not only try to explain a change in foreign investment by the introduction of codes or other financial variables, but also non financial variables were used to explain a change in foreign investment (level of corruption). The data showed that the variables corruption did not have any effect on the models. Of a total sample of 78 countries, 29 countries adopted a code. These 29 countries were spread over various continents and income levels.

The research distinguishes foreign direct investment from foreign portfolio investment. Three types of analysis were done. The first was a comparison between countries that did, and countries that did not, comply with the code. The second was a comparison between countries that comply with the code during a certain period of time. The third analysis was based on the relation between the number of months after compliance and the change in foreign investment. The regression analysis shows a significant relation between the adoption of a code and FPI. It also shows there is no significant direct relationship between the adoption of a code and the change in FDI. Hypothesis 1 is therefore rejected, while hypothesis 2 is supported. Of all the variables used in the models, openness to trade as well as corruption seemed to be most important in predicting FDI.

(37)

(38)

References

Aguilera, R.V., Cuervo-Cazurra, A. 2004. Codes of good governance worldwide: what is the trigger? Organization Studies 25 (3): 415-443

Aguilera, RV., Jackson, G. 2003. The cross-national diversity of corporate governance: dimensions and determinants. Academy of Management Review. 28 (3): 447-465

Allen, F. 2005. Corporate governance in emerging economies. Oxford Review of

Economic Policy, 21 (2): 164

Aoki, M. 2000. Oxford and New York: Oxford University Press. Information, corporate governance, and institutional diversity: competitiveness in Japan, the USA and the transitional economies.

Balasubramanyam, V.N., Salisu, M., Sapsford, D. 1996. Foreign direct investment and growth in EP and IS countries. The Economic Journal, 106: 92-105

Banyan, A., Murkowski, J., Herczynski, J. 2005. On the determinants of foreign direct investment in transition economies. Problems of Economic Transition, 48: 6-28

Barrell, R., Pain, N. 1999. Domestic institutions, agglomerations and foreign direct investment in Europe. European Economic Review. 43: 925-934

Bekaert, G., Harvey, C.R. 2002. Research in emerging markets finance: looking to the future. Emerging Markets Review, 3 (4): 429-449

(39)

Berglöf, E. & S. Claessens. 2004. World bank policy research working paper 3409. Corporate governance and enforcement.

Berglöf, E., Pajuste, A. 2005. What do firms disclose and why? Enforcing corporate governance and transparency in Central and Eastern Europe. Oxford Review of

Economic Policy 21 (2): 178-197

Bevan, A.A., Estrin, S. 2004. The determinants of foreign direct investment into European transition economies. Journal of Comparative Economics. 32: 775-787

Bhavan, T., Xu, Changsheng., & Zhong, C. (2011). Determinants and Growth Effect of FDI in South Asian Economies:Evidence from a Panel Data Analysis, International

Business Research, 4(1)

Black, B.S., Hasung Jang, Woocham Kim. 2006. Does corporate governance predict firms’ market values? Evidence from Korea. Journal of Law, Economics and

Organization, 22 (2): 366-413

Billington, N. 1999. The location of foreign direct investment: an empirical analysis.

Applied Economics. 31: 65-76

Bjorvatn, K., Eckel, C. 2006. Policy competition for foreign direct investment between asymmetric countries. European Economic Review. 50: 1891-1907

Boddewyn, J. J. 1988. Political aspects of MNE theory. Journal of International

(40)

Brouthers, LE., Werner, S., Wilkinson, TJ. 1996. The aggregate impact of firms’ FDI strategies on the trade balances of host countries. Journal of International Business

Studies, 27 (2): 359-373

Buch, CM., Kokta, RM., Piazolo, D. 2003. Foreign direct investment in Europe: is there redirection from the South to the East? Journal of Comparative Economics 31 (2003): 94-109

Chipalkatti, N., Le, Q.V., Rishi, M. 2007. Portfolio flows to emerging capital markets: do corporate transparency and public governance matter? Business and Society Review, 112 (2): 227-249

Claessens, S. 2003. Corporate governance and development: review of the literatures and outstanding research issues. Proceedings of the global corporate governance forum

donors meeting, March 13, The Hague, the Netherlands.

Cuervo, A. 2002. Corporate Governance mechanisms: a plea for less code of good governance and more market control. Corporate Governance: an international review , 10 (2): 84-94

(41)

Evrensel, A.Y. Kutan, A.M. 2007. Are multinationals afraid of social violence in emerging markets? Journal of Economic Studies 34(1): 59-73

Fama, E., Jensen, M.C. 1983. Agency problems and residual claims. Journal of law And

Economics, 26 (2): 327-350

Fernández-Rodríguez, E., Gómez-Ansón, S., Cuervo-García, Á. 2004. The stock market reaction to the introduction of best practices codes by Spanish firms. Corporate

Governance: An International Review, 12 (1): 29-46

Field, A. 2005. Discovering Statistics using SPSS, 2nd

edition. Sage Publications,

Oxford, Great Britain.

Firebaugh, 1983. Scale economy or scale entropy? Country size and rate of economic growth 1950-1977. American Sociological Review, 48 (2): 257-269

Fremond, O & M. Capaul. 2002. The State of corporate governance: experience from country assessments. World bank policy research working paper.

Friedman, J., Gerlowski, D.A., Silberman, J. 1992. What attracts foreign multinational corporations? Evidence from branch plant location in the USA. Journal of Regional

Science, 32: 403-418

Gibson, M.S. 2003. Is corporate governance ineffective in emerging markets?

Journal of Financial and Quantitative Analysis, 38 (1): 231-250

Guerin, S.S. 2006. The role of geography in financial and economic integration: a comparative analysis of foreign direct investment, trade and portfolio investment flows.

(42)

Guillen, M.F. 2000. Corporate governance and globalization: is there convergence across countries? Advances in international comparative management, 13: 175-204

Gupta, A & Parua, A. 2006. 10th_{Indian Institute of capital markets conference paper.}

An inquiry into compliance of corporate governance codes by the private sector Indian companies.

Hailu, Z.A. (2010). Impact of Foreign Direct Investment on Trade of African Countries,

International Journal of Economics and Finance, 2,122-133

Haroon, H.H., Kozhich, V. 2006. Corporate governance in an emerging market: a perspective on Pakistan. Journal of Legal Technology Risk Management, 1 (1):22 Hart, O. 1995. Corporate Gov ernance: some theory and implications. The Economic

Journal, 105: 678-689

Hellman, J.S., Jones, G., Kaufmann, D. 2002. Far from home: do foreign investors import higher standards of governance in transition economics? Worldbank

Holsapple, EJ., Ozawa, T., Olienyk, J. 2006. A synthesis of Foreign “Direct” and “Portfolio” Investment in real estate. Journal of Real Estate Portfolio Management, 12 (1): 37-47

Islam, R. 2006. Does more transparency go along with better governance? Economics

and Politics, 18 (2): 121-167

(43)

Jensen, M.C., Meckling, W.H. 1976. Theory of the firm: managerial behaviour, agency cost and ownership structure. Journal of Financial Economics, 3 (4): 305-360

Kinoshita, Y., Campos, N.F. 2003. Why does FDI go where is goes? New evidence from the transition economies. International Trade and Transition Economics, Discussion Paper no. 3984.

Kirkpatrick, C., Parker, D., Zhang, YF. 2006. Foreign direct investment in infrastructure in developing countries: does regulation make a difference? Transnational Corporations, 15 (1): 143-171

Klapper, LF., Love, I. 2002. Corporate governance, investor protection, and performance in emerging markets. World Bank Policy Paper, 2818

Leitão, N. C. (2010). Foreign Direct Investment: The Canadian Experience. International

Journal of Economics and Finance, 2(4).

Leitão, N.C., & Faustino, H.C.(2010). Determinants of Foreign Direct Investment in Portugal, Journal of Applied Business and Economics, 11 (3) : 19-26

Levine, R. 1999. Law, finance and economic growth. Journal of Financial

Intermediation, 8 (1-2): 8-35

Liljeblom, E., Löflund, A. 2005. Determinants of international portfolio investment flows to a small market: empirical evidence. Journal of Multinational Financial

Management, 15: 211-233

Mohamed, S.E., & Sidiropoulos, M.G. (2010). Another look at the Determinants of Foreign Direct Investment in MENA Countries: An Empirical Investigation. Journal of

(44)

Morgan, R.E., Katsikeas, C.S. 1997. Theories of international trade, foreign direct investment and firm internationalization: a critique. Management Decision, 35 (1): 68-78

North, D.C. 1990. Institutions, institutional change, and economic performance. Cambridge: Cambridge University Press.

Nowak, E., Rott, R., Mahr, T.G. 2006. The (ir) relevance of disclosure of compliance with corporate governance codes- evidence from the german stock market. Swiss

Finance Institute Research Paper, Series 6 (11)

Poshakwale, S.S., Thapa, C. 2011. Investor protection and international equity portfolio investments. Global Finance Journal, 22 (2): 116-129

Reid, A.S. 2003. The internationalization of corporate governance codes of conduct.

Business Law Review, 24 (10): 233-238

Rueda-Sabater, E. 2000. Corporate governance: and the bargaining power of developing countries to attract foreign investment. Corporate Governance: An international review, 8 (2): 117-125

Smith, A. 1976. The wealth of nations.

Schneider, K.H., & Matei, I. (2010). Business Climate, Political Risk and FDI in Developing Countries:Evidence from Panel Data. International Journal of Economics

and Finance, 2(5): 54-65

(45)

Thomsen, S. 2001. Business ethics as corporate governance. European Journal of Law

and Economics, 11 (2): 153-164

Thomsen, S. 2006. The hidden meaning of codes: corporate governance and investor rent seeking. European Business Organization Law Review, 7 (4): 767-815

Schleifer, A., Vishny, R. 1997. A survey on corporate governance. The Journal Of

Finance, 52 (2): 737-783

Škuflić, L., Botrić, V. 2006. Foreign direct investment in Southeast European countries, the role of the service sector. Eastern European Economics. 44 (5): 72-90

Uhlenbruck, K., Rodriguez, P., Doh, J., Eden, L. 2006. The impact of corruption on entry strategy: evidence from telecommunication projects in emerging economies.

Organization Science. 17 (3): 402-414

Wijeweera, A., Dollery, B. 2009. Host country corruption level and Foreign Direct Investment inflows. International Journal of Trade and Global Markets, 2 (2): 168-178 Wymeersch, E. 2006. Working Paper. Financial Law Institute. Gent University. Corporate Governance Codes and their implementation.

(46)

Appendix 1 - Sample

Country Geographical Area Income

(47)

41 Macao, China East Asia & Pacific High Income (Non-OECD) 42 Brunei Darussalam East Asia & Pacific High Income (Non-OECD) 43 Malaysia East Asia & Pacific Upper Middle Income 44 China East Asia & Pacific Lower Middle Income 45 Indonesia East Asia & Pacific Lower Middle Income 46 Philippines East Asia & Pacific Lower Middle Income 47 Thailand East Asia & Pacific Lower Middle Income 48 Fiji East Asia & Pacific Lower Middle Income 49 Samoa East Asia & Pacific Lower Middle Income 50 Tonga East Asia & Pacific Lower Middle Income 51 Sri Lanka South Asia Lower Middle Income 52 Maldives South Asia Lower Middle Income

53 India South Asia Low Income

54 Pakistan South Asia Low Income

55 Bangladesh South Asia Low Income

56 Nepal South Asia Low Income

57 Bhutan South Asia Low Income

58 Saudi Arabia Middle East & North Africa High Income (Non-OECD) 59 Israel Middle East & North Africa High Income (Non-OECD) 60 United Arab Emirates Middle East & North Africa High Income (Non-OECD) 61 Kuwait Middle East & North Africa High Income (Non-OECD) 62 Qatar Middle East & North Africa High Income (Non-OECD) 63 Bahrain Middle East & North Africa High Income (Non-OECD) 64 Malta Middle East & North Africa High Income (Non-OECD) 65 Libya Middle East & North Africa Upper Middle Income 66 Lebanon Middle East & North Africa Upper Middle Income 67 Oman Middle East & North Africa Upper Middle Income 68 South Africa Sub-Saharan Africa Upper Middle Income 69 Botswana Sub-Saharan Africa Upper Middle Income 70 Gabon Sub-Saharan Africa Upper Middle Income 71 Mauritius Sub-Saharan Africa Upper Middle Income 72 Equatorial Guinea Sub-Saharan Africa Upper Middle Income 73 Seychelles Sub-Saharan Africa Upper Middle Income 74 Ethiopia Sub-Saharan Africa Low Income

75 Sudan Sub-Saharan Africa Low Income 76 Kenya Sub-Saharan Africa Low Income 77 Uganda Sub-Saharan Africa Low Income 78 Somalia Sub-Saharan Africa Low Income

(48)

Appendix 2 - Regression results

2.1 Model 1-3 Relation ‘Adopt code’ – ‘FDI’

R Square Change F Change df1 df2 Sig. F Change 1 ,082a ,007 ,006 128183 ,007 13,675 1 2029 ,000 2 ,087b ,008 ,007 128158 ,001 1,793 1 2028 ,181 3 _,675c _,455 _,454 ₉₄₉₇₆ _,448 _1665,561 ₁ ₂₀₂₇ _,000 4 ,677d ,458 ,457 94714 ,003 12,228 1 2026 ,000 ,665 Model Summarye Model R R Square Adjusted R

Square Std. Error of the Estimate

Change Statistics

Durbin-Watson

a. Predictors: (Constant), Adopt_Code

b. Predictors: (Constant), Adopt_Code, GDPperCapita

c. Predictors: (Constant), Adopt_Code, GDPperCapita, openesstrade

d. Predictors: (Constant), Adopt_Code, GDPperCapita, openesstrade, Corruption e. Dependent Variable: FDIin_GDP

Standardized Coefficients

B Std. Error Beta Zero-order Partial Part Tolerance VIF

(Constant) 62677 3728 16,811 ,000 Adopt_Code -21325 5767 -,082 -3,698 ,000 -,082 -,082 -,082 1,000 1,000 (Constant) 65449 4264 15,351 ,000 Adopt_Code -20938 5773 -,080 -3,627 ,000 -,082 -,080 -,080 ,997 1,003 GDPperCapita -2 1 -,030 -1,339 ,181 -,034 -,030 -,030 ,997 1,003 (Constant) 2076 3521 ,590 ,556 Adopt_Code -10569 4286 -,041 -2,466 ,014 -,082 -,055 -,040 ,994 1,006 GDPperCapita -1 1 -,018 -1,079 ,281 -,034 -,024 -,018 ,997 1,003 openesstrade ,670 40,811 ,000 ,673 ,672 ,669 ,996 1,004 (Constant) 24497 7310 3,351 ,001 Adopt_Code -8744 4306 -,034 -2,031 ,042 -,082 -,045 -,033 ,979 1,021 GDPperCapita 1 1 ,018 ,936 ,349 -,034 ,021 ,015 ,717 1,395 openesstrade ,687 40,215 ,000 ,673 ,666 ,657 ,915 1,093 Corruption -11913 3407 -,070 -3,497 ,000 ,093 -,077 -,057 ,670 1,492

(49)

R Square Change F Change df1 df2 Sig. F Change 1 ,082 ,007 ,006 128183 ,007 13,675 1 2029 ,000 2 ,087 ,008 ,007 128158 ,001 1,793 1 2028 ,181 3 ,092 ,008 ,007 128134 ,001 1,771 1 2027 ,183 4 ,164 ,027 ,025 126965 ,018 38,483 1 2026 ,000 5 ,230 ,053 ,050 125346 ,026 18,559 3 2023 ,000 ,397

d. Predictors: (Constant), Adopt_Code, GDPperCapita, openesstrade, Corruption

e. Predictors: (Constant), Adopt_Code, GDPperCapita, openesstrade, Corruption, uppermiddleincome, lowincome, lowermiddelincome f. Dependent Variable: FDIin_GDP

Model Summaryf

Model R R Square

Adjusted R

Change Statistics

Durbin-Watson

(Constant) 62677 3728 16,811 ,000 Adopt_Code -21325 5767 -,082 -3,698 ,000 -,082 -,082 -,082 1,000 1,000 (Constant) 65449 4264 15,351 ,000 Adopt_Code -20938 5773 -,080 -3,627 ,000 -,082 -,080 -,080 ,997 1,003 GDPperCapita -2 1 -,030 -1,339 ,181 -,034 -,030 -,030 ,997 1,003 (Constant) 2076 3521 ,590 ,556 Adopt_Code -10569 4286 -,041 -2,466 ,014 -,082 -,055 -,040 ,994 1,006 GDPperCapita -1 1 -,018 -1,079 ,281 -,034 -,024 -,018 ,997 1,003 openesstrade ,670 40,811 ,000 ,673 ,672 ,669 ,996 1,004 (Constant) 24497 7310 3,351 ,001 Adopt_Code -8744 4306 -,034 -2,031 ,042 -,082 -,045 -,033 ,979 1,021 GDPperCapita 1 1 ,018 ,936 ,349 -,034 ,021 ,015 ,717 1,395 openesstrade ,687 40,215 ,000 ,673 ,666 ,657 ,915 1,093 Corruption -11913 3407 -,070 -3,497 ,000 ,093 -,077 -,057 ,670 1,492 (Constant) 93790 17591 5,332 ,000 Adopt_Code -7388 4349 -,028 -1,699 ,089 -,082 -,038 -,028 ,951 1,052 GDPperCapita -3 2 -,042 -1,709 ,088 -,034 -,038 -,028 ,428 2,336 openesstrade ,679 39,509 ,000 ,673 ,660 ,643 ,896 1,115 Corruption -23986 4284 -,141 -5,598 ,000 ,093 -,124 -,091 ,420 2,383 lowincome -49104 13997 -,116 -3,508 ,000 -,093 -,078 -,057 ,244 4,101 lowermiddelincome -50353 11396 -,192 -4,418 ,000 -,090 -,098 -,072 ,140 7,158 uppermiddleincome -26514 9325 -,098 -2,843 ,005 ,113 -,063 -,046 ,223 4,491 a. Dependent Variable: FDIin_GDP

(50)

R Square Change F Change df1 df2 Sig. F Change 1 ,082 ,007 ,006 128183 ,007 13,675 1 2029 ,000 2 ,087 ,008 ,007 128158 ,001 1,793 1 2028 ,181 3 ,092 ,008 ,007 128134 ,001 1,771 1 2027 ,183 4 ,164 ,027 ,025 126965 ,018 38,483 1 2026 ,000 5 ,262 ,069 ,064 124375 ,042 18,051 5 2021 ,000 ,404 Model Summaryf Model R R Square Adjusted R

Change Statistics

Durbin-Watson

d. Predictors: (Constant), Adopt_Code, GDPperCapita, openesstrade, Corruption

e. Predictors: (Constant), Adopt_Code, GDPperCapita, openesstrade, Corruption, SubsaharanAfrica, Southasia, EuropeCentralAsia, middleeastnorthafgrica, EastAsiaPacific

f. Dependent Variable: FDIin_GDP

(Constant) 62677 3728 16,811 ,000 Adopt_Code -21325 5767 -,082 -3,698 ,000 -,082 -,082 -,082 1,000 1,000 (Constant) 65449 4264 15,351 ,000 Adopt_Code -20938 5773 -,080 -3,627 ,000 -,082 -,080 -,080 ,997 1,003 GDPperCapita -2 1 -,030 -1,339 ,181 -,034 -,030 -,030 ,997 1,003 (Constant) 2076 3521 ,590 ,556 Adopt_Code -10569 4286 -,041 -2,466 ,014 -,082 -,055 -,040 ,994 1,006 GDPperCapita -1 1 -,018 -1,079 ,281 -,034 -,024 -,018 ,997 1,003 openesstrade ,670 40,811 ,000 ,673 ,672 ,669 ,996 1,004 (Constant) 24497 7310 3,351 ,001 Adopt_Code -8744 4306 -,034 -2,031 ,042 -,082 -,045 -,033 ,979 1,021 GDPperCapita 1 1 ,018 ,936 ,349 -,034 ,021 ,015 ,717 1,395 openesstrade ,687 40,215 ,000 ,673 ,666 ,657 ,915 1,093 Corruption -11913 3407 -,070 -3,497 ,000 ,093 -,077 -,057 ,670 1,492 (Constant) 11157 8067 1,383 ,167 Adopt_Code -5035 4773 -,019 -1,055 ,292 -,082 -,023 -,017 ,783 1,277 GDPperCapita 2 1 ,036 1,691 ,091 -,034 ,038 ,027 ,592 1,689 openesstrade ,677 39,085 ,000 ,673 ,656 ,633 ,875 1,143 Corruption -9663 3439 -,057 -2,810 ,005 ,093 -,062 -,046 ,647 1,547 EuropeCentralAsia 14849 5686 ,052 2,611 ,009 ,016 ,058 ,042 ,663 1,508 EastAsiaPacific -13056 7350 -,035 -1,776 ,076 ,008 -,039 -,029 ,674 1,483 middleeastnorthafgrica -4727 8610 -,011 -,549 ,583 -,041 -,012 -,009 ,645 1,550 Southasia -5159 9054 -,011 -,570 ,569 -,098 -,013 -,009 ,751 1,331 SubsaharanAfrica 32730 7209 ,084 4,540 ,000 ,190 ,100 ,074 ,761 1,314 5

(51)

2.2 Model 4-6 Relation ‘Adopt code’ – ‘FPI’

R Square Change F Change df1 df2 Sig. F Change 1 ,068a _,005 _,004 ₄₂₂₃₉ _,005 _9,106 ₁ ₁₉₆₀ _,003 2 ,077b _,006 _,005 ₄₂₂₂₀ _,001 _2,710 ₁ ₁₉₅₉ _,100 3 ,078c _,006 _,004 ₄₂₂₃₁ _,000 _,039 ₁ ₁₉₅₈ _,844 4 _,137d _,019 _,017 ₄₁₉₇₂ _,013 _25,167 ₁ ₁₉₅₇ _,000 _1,554

d. Predictors: (Constant), Adopt_Code, GDPperCapita, openesstrade, Corruption e. Dependent Variable: FPIin_GDP

Model Summarye

Model R R Square

Adjusted R

Change Statistics

Durbin-Watson

(Constant) 7780 1252 6,214 ,000 Adopt_Code 5830 1932 ,068 3,018 ,003 ,068 ,068 ,068 1,000 1,000 (Constant) 6622 1436 4,612 ,000 Adopt_Code 5710 1933 ,067 2,955 ,003 ,068 ,067 ,067 ,999 1,001 GDPperCapita 1 ,037 1,646 ,100 ,040 ,037 ,037 ,999 1,001 (Constant) 6581 1451 4,535 ,000 Adopt_Code 5732 1936 ,067 2,960 ,003 ,068 ,067 ,067 ,995 1,005 GDPperCapita 1 ,037 1,645 ,100 ,040 ,037 ,037 ,999 1,001 openesstrade ,004 ,196 ,844 ,001 ,004 ,004 ,997 1,003 (Constant) -8277 3294 -2,513 ,012 Adopt_Code 4868 1932 ,057 2,519 ,012 ,068 ,057 ,056 ,987 1,013 GDPperCapita -1 1 -,030 -1,143 ,253 ,040 -,026 -,026 ,737 1,356 openesstrade ,003 ,115 ,909 ,001 ,003 ,003 ,996 1,004 Corruption 7332 1462 ,131 5,017 ,000 ,121 ,113 ,112 ,732 1,367 4

a. Dependent Variable: FPIin_GDP

“The introduction of Corporate Governance Codes and the effect on foreign investment: a study of Non-OECD countries”