The Determinants Of Income Inequality: A System Panel Data Analysis

(1)

The Determinants Of Income Inequality: A System

Panel Data Analysis

A study on the Global Income Inequality

(2)

Abstract

Income inequality is one of the most ubiquitous challenges around the world. During the past few decades, many researches dedicated their efforts to examine how income is distributed. Even though many common understanding have been achieved, a lot of ambiguities still remain in the research field. The goal of this study is to find the determinants of income inequality; to confirm or rejects theories from precious works; and to distinguish the difference between different countries. Export of goods and service has a positive

relationship with income inequality. The extent of influence differs depending on country development level. To be specific, the effects is far greater in LDCs and developing countries, but less in developed countries. Foreign direct investment raises gap between income groups. The exact level of influence differs depending on country stage of

development, though the distinctions are not as great as other predictors. Inflation was found to be not significantly influencing income inequality in general. Developed countries are more liable to be affected by rise in inflation, while developing and LDCs are not. Lastly, the positive relationship between economic growth and income inequality is validated. On

country level, while growth largely increase income discrepancy in developed and developing countries, the effects is reduced considerably in least developed countries.

(3)

Preface

This master thesis concludes the last part of my Master International Business and

Management at the University of Groningen. After the 7 months of intensive researching and studying, I am honored to introduce my thesis about determinants of income inequality around the world.

I would like to highlight the importance of all the support and assistance I received form the various departments of University of Groningen. Their kindness, warmness, and supportive attitude were instrumental to the writing of this thesis.

Then I would like to give my whole heartily thanks to all my friends and families back at home. This report would not be made possible without their unconditional love and support. Thanks to my friends for giving countless valuable suggestion and help throughout the process. Certainly, I want to thank my parents for their incredible emotional and financial support during my studying in the Netherlands.

Last but not least, I am deeply grateful for the tireless support from my supervisors. I would like to show my gratitude to Dr. D.H.M. (Dirk) Akkermans for the enormous amount of guidance, patience and flexibility. I also want to thank dr. M.J. (Mariko) Klasing for Co-assessing my thesis. Thank you for the opportunity and support!

I hope you enjoy reading my master thesis!

(4)

2.3 Inflation ... 11 2.4 Economic Growth ... 13 2.5 Control variable ... 15 2.5.1 Education ... 15 2.5.2 Technology ... 15 3. Methodology ... 16 3.1 Empirical Mode ... 16 3.2 Data ... 16 3.3 Data Collection ... 17 3.3.1 Measurement ... 18 4. Empirical Analysis ... 20 4.1 Quality of Data ... 20 4.1.1 Summary ... 20 4.1.2 Normality ... 20 4.1.3 Correlation ... 21 4.1.4 Multicollinearity ... 21

4.1.5 Heteroskedasticity and outliers ... 22

4.2 Model Tests ... 22 4.2.1 Country Sub-groups ... 24 4.2.2 Endogeneity ... 25 5. Empirical Result ... 26 5.1 Model Analysis ... 26 5.2 Model Results ... 27 5.2.1 Export ... 27

5.5.2 Foreign Direct Investment ... 27

5.5.4 Inflation ... 28

5.5.5 Growth ... 28

5.5.6 Control Variables ... 28

5.3 Conclusion empirical results ... 29

6. Discussion ... 30

6.1 Export ... 30

6.2 Foreign Direct Investment ... 30

6.3 Inflation ... 31

6.4 Economic Growth ... 32

7 Limitations and Future Research ... 34

(5)

7.2 Future Research ... 34

8 Conclusion ... 36

Source: ... 38

Appendix ... 44

Appendix 1 – Multicollinearity Test ... 44

Appendix 2 - Scatterplot ... 45

(6)

1. Introduction

The world has been experiencing increased level of inequality in the most recent decades. The Human Development Report in 2014 shows that income inequality is increasing at an exponential rate. The distribution of wealth and income varies significantly over time and region. Some countries have seen a smoothed curve of inequality, others experience

unprecedented gap. This phenomenon suggests that finding the driving forces of inequality is important.

Piketty and his influential book published in 2014 have sparked debate in most recent inequality studies. As suggested by Piketty (2014), it is necessary to make distinguish between wealth inequality and income inequality. In his book, the definition of income is rather straightforward, income is understood as wage or salary earning through exchange of labor. Wealth is defined as tradable asset that generates financial return in a society at one given moment in time.

(7)

Figure 1 – Regions affected by income inequality

Source: Outlook on the Global Agenda 2014 by World Economic Forum

Income inequality is one of the most challenging socioeconomic questions in today’s world. Even though many scholars have dedicated their lifetime research to discover the cause of rising inequality, the true cause is still not clear and the problem still persists. In fact, Global Agenda 2014 predicted that income gap between and within countries will keep increasing for the coming years (figure 1).

This study aims to shed lights on the determinants of income inequality. The rest of the paper is structured as follows. Chapter 2 establishes the relationship between various potential determinants and income inequality. This process is done though reviewing existing and well-known literatures done by previous studies. Chapter 3 shows the construction of model, methodology and data set. Chapter 4 examines the data quality and various empirical models. Chapter 5 presents the results of empirical testing. Chapter 6 discusses the empirical results and links them with relevant theories from literature review. Chapter 7 proposes the

(8)

2. Literature Review

2.1 Export

That trade liberalization may affect labor wages is a result predicted from the well-known Heckscher Ohlin Samuelson (HOS) model of trade theory and the Stolper Samuelson (SS) theorem. Both theories imply that wage is positively affected by trade in labor surplus

economies and negatively affected by trade in capital surplus ones. A simple extension of the SS theorem implied that wage gap widens between skilled and unskilled worker within an economy. This is a consequence of increased trade liberalization around the world after the General Agreement on Tariffs and Trade (GATT) came into force (Cline, 1997).

Company pays a worker the average wage plus or minus premiums for the skill above or below certain level (Beladi and Batra 2004). Thus, a rise in wage inequality implies a slowed rising pace of earnings of the low skilled workers.

Early research and modeling of skilled and unskilled labor and structural unemployment extended HOS model to show trade liberalization increased inequality in the USA (Batra and Slottje, 1993). Empirical studies show that increased trade is in fact associated with enlarging wage inequality between unskilled and skilled labor in both developing and developed

countries (Harrison, McLaren, & McMillan, 2010).

Trade plays a special role in developing countries. In order to encourage trade, governments in developing economies grant export subsidies and foreign exchange retention to exporters (Goldberg & Pavcnik, 2004). However, trade licenses and quotas benefit the rich and the powerful much more than the poor, since the rich and powerful can easily obtain trade

privileges through established business connection, corruption and other measure (Yao, 2002; Wedeman, 2004; Gong, 2006). The powerful not only have a direct access to trade licenses and quotas, they can also make it harder for those less resourceful and less connection (Wang and You 2012).

(9)

on one hand and the underpaid labor force on the others (Harrison, McLaren, & McMillan, 2010). First, wages are set low due to limited labor force mobility between industry, low skill transferability, low education levels, as well as various other social and legal barriers

(Goldberg & Pavcnik, 2004, 2007).

Second, the creation of a high return and high profit export sector not only produce a small portion of high-income employees with high compensation, it further increases inequality by severe competition for unskilled jobs (Harrison, McLaren, & McMillan, 2010). Decreasing the bargaining power of labor and increase income gap even further (Giesecke, Heisig & Solga 2015). These tendencies are enhanced by the strategies most developing nations pursue.

Third, export-oriented firms are usually foreign-owned or financed (Robbins 1996). This leaves economies vulnerable to demand fluctuations in the global market, putting the labors in export sector even weaker position, contributes to persistency of ever increasing income inequality (Ferreira et al., 2007; Gonzaga et al., 2006). Lastly, the subordinate position of developing governments paired with foreign capital and influential transnational actors such as the IMF, decrease their ability to implement labor beneficial social and economic policies (McMichael 1996; Goldberg and Pavcnik 2004).

The nature of a nation’s participation in the global trading system also plays a role in income inequality (Arbache, Dickerson & Green 2004). Commodity concentration describe to the degree to which a nation’s export role is limited to a few commodities and raw material (Breau and Rigby, 2010). In comparison, nations with a more diversified value added product of exports have more choices in responding to changing demand in the global economy (Breau and Rigby, 2010). Companies that are attempting to produce high quality goods also have the tendency to hire highly skilled worker. These workers earns higher wage that contribute to increasing income inequality (Verhoogen 2008). Since developing nations are the main suppliers in the global commodity market (Knight and Song 2003), it is not hard to find that developing countries with high commodity concentration has been found to be vulnerable to downturn commodity price.

(10)

Hypothesis 1: A country’s level of export activity has a positive relationship with its level of income inequality.

Hypothesis 1a: Export activities have larger impact on developing countries and least developed countries income inequality than developed countries.

2.2 Foreign Direct Investment

Precious researches have found there is a strong relationship between investment dependency or transnational corporate penetration (TCP) and income inequality (Beer 1999). TCP

reinforces income inequality in developing nations by altering economic conditions in these countries (Bornschier and Chase-Dunn, 1985). Companies operating in developing countries seek to take advantage of the lower wage rate in these economies (Beladi et al., 2008; Anwar and Rice, 2009). However, the wages offered by the international companies are still well above the rate in developing countries (Eckel, 2003; Moore and Ranjan, 2005). This directly results in a gaping income discrepancy between residents who are employed by international firms and those who are not. Furthermore, Bornschier and Chase- Dunn hypothesized that the "comprador class" has no incentive to take redistributive policies. Therefore, government strategies are often designed to attract foreign capital investment by offering tax allowance, profit repatriation, along with lower labor costs. Strategies like these, in turn, hinder or prohibit of labor unions or strikes even (Marjit and Kar, 2005; Wang, Fang and Huang, 2009).

Many researchers had studied the clear contradictions between multinational corporation (MNCs) tendency to create income gap and their need for economic and political stability (Marjit et al., 2004; Chaudhuri and Yabuuchi, 2007; Anwar and Rice, 2009). as above, I described MNS's very presence tends to cause unrest between labor in different income group. This can leads to political instability that they often seek to avoid (Moore and Ranjan, 2005). In addition, the transfer of resources back to the home country through repatriation of profits limited the amount of capital available for investment and re-investment in domestic economy, in turns, further deepening the income inequality problem (Khan and Riskin 1998; Deardorff 2005; and Horgos 2009).

(11)

have better chance in collude with foreign investors in granting permits (Dobson and Ramlogan-Dobson 2010). In this way, the rich gain disproportionally from the rents generated by foreign direct investment (Andres and Ramlogan-Dobson 2011). However, some researchers had proposed that the flow of foreign capital into developing world reduce the high borrowing cost, and that increase the credit accessibility for the poor to set up their own businesses (Villegas-Sanchez 2009). Therefore, the end effect of foreign direct

investment on income distribution is not conclusive.

Finally, the governments in developing countries encouraged by the need to attract and retain highly mobile foreign capital, employ strategies and policies that decrease the bargaining power of labor and inhibit vertical mobility by the lower classes (Banga 2005; Anwar, 2006). These policies and strategies include guarantees of tax concessions, profit repatriation, and unfavorable labor laws (Taylor and Wyatt 1996; Glytsos 2002).

To sum up, according to the theories described above, there are mainly three reasons that foreign investment creates inequality (Beer 1999). First, foreign capital generates large sectorial and labor disparities in a country's economy, which results in the underutilization of indigenous labor. Second, Multinational Corporation create a disproportional share of

repatriate profit rather than reinvesting them again in local economy. Lastly, the governments are motivated to by necessity to implement strategies and policies that reduce labor

negotiation power and limit vertical mobility.

Based on the analysis above, the following hypotheses are developed:

Hypothesis 2: A country’s level of foreign direct investment inflow has a positive relationship with its level of income inequality.

Hypothesis 2a: Foreign Direct investment inflow have larger impact on developing countries and least developed countries income inequality than developed countries.

2.3 Inflation

(12)

"inflation-adjusted contract"- such as option and stock ownership. The first type of contract is mainly used by non-managerial labor compensation, while the second type of contract is used to construct managerial compensation (Bénabou 2005; Flynn & Donnelly 2012).

After clarifying the difference between nominal contract and inflation- adjusted contract, the relationship between Inflation and income inequality can be studied. Inflation reduces labor with nominal contract available resources for consumption through a loss in purchase power (Flynn & Donnelly 2012). It is reasonable to assume that these compensation characteristics exclude nominal contract wage labor at the bottom of the income scale, who are generally much more sensitive to real-wage fluctuations (Flynn & Donnelly 2012). On the other hand, inflation-adjusted contract workers receive compensation other than cash and are employed under a different wage regime than nominal contract (Flynn & Donnelly 2012). Top-level managers receive most of their compensation in share, options and noncash benefits, the market value of which is immune to inflation (Frydman and Jenter, 2010). Therefore, top mangers face little or no inflation distraction, since their marginal product of labor is

unchanged. Returns on assets owned by a wealthy insider might also be better protected from inflation (Frydman &, 2010). These workers have the option to invest in assets that are not affected or weakly affected by inflation or assets that has growth rate higher than inflation rate. In a period of consumer price fluctuation, whether the individual's income keeps pace with inflation obviously depends on the bargaining power of the individuals or the group to which they belongs (Cahuc, Saint-Martin, Zylberberg, 2001). This implies that the two groups have unequal leverage in wage negotiation and, in consequence, uneven pay rises. As a result, this make the adoption of policies-such as inflation more favorable to the high income group.

From political perspective, inflation is a regressive tax that affect the poor and middle class much harder because they hold more nominal assets as a portion of their total income, than the rich (Chiroleu-Assouline and Fodha 2014). This implies that the rich- who has many ways to avoid inflation tax- might prefer inflation to progressive taxes such as income tax or inheritance tax. Furthermore, in most developing countries where democratic institutions are non-existent or less effective, rich groups carry more political weight than others (Poterba, 2007). In other words, money talks. In a political system where corruption is high and

(13)

The effects of inflation can be summarized as follows. First, workers whose compensation is protected against price level fluctuation (inflation-adjusted contract) would increase their incomes relative to the unprotected group (nominal contract). Second, the value of incomes of both groups would fall, but the effect is more severe for labors with nominal contract. Third, while government policies like minimum wage can prevent low income group from falling into poverty those policies are insufficient to narrow down the inflation-generated income distribution gap, because the number of nominal contract worker is significant larger than the number inflation-adjusted contract worker. One would expect the effects of fiscal policy to be weakly correlated with inflation-related changes in income inequality.

Based on the analysis above, the following hypotheses are developed:

Hypothesis 3: A country’s level of inflation has a positive relationship with its level of income inequality.

Hypothesis 3a: Inflation have larger impact on developing countries and least developed countries income inequality than developed countries.

2.4 Economic Growth

Since the foundation work of Greenwood and Jovanovic (1990), the question of whether economic growth increases or decreases the income inequality has been subject to heavy debate. Greenwood and Jovanovic proposed the idea of an inverted U-shaped relationship between economic development and income inequality. The argument was based on the assumption that, as an country develops, its economic structure shifts from agricultural production to industrial production.

(14)

in the economy (Zhu and Trefler 2005). Therefore, at the beginning stage of economic development, there is a positive relationship between growth and income inequality. Latter on, workers keep shifting from the agricultural sector into the industrial sector, the reduced labor supply in the agricultural sector drives up wages in this sector (Lee and Saez 2012). As a result, Inequality falls with later stages of development, a negative relationship between economic growth and inequality occurs (Barro, 2000). Hence, based on this theory, an inverted U-shaped relationship between economic growth and income inequality was proposed.

However, the inverted-U shape theory is far from perfect. Many studies have observed a positive relationship between growth and income inequality (Galor and Zeira, 1993; Banerjee and Newman 1993). Although industrialization and urbanization create growth for economy as whole, the economic gain is disproportionately distributed among actors. Few studies has suggested that it is possible for income gap to increase as economic undergo further

development. Elbers and Lanjouw (2001) reported that increases in modern sector incomes tended to increase income inequality. Shahbaz (2010) also confirmed that income inequality appears to be positively and significantly associated with economic growth in both short-term and long-term. Those studies believe the real cause of income inequality is distribution. Another branch of studies have shown that economic development does not have any relation with the income distribution. Ravallion (2005) proposed that there was no correlation

between average income and economic growth. Das and Barua (1996) find no evidence of a trade-off between economic growth and income growth in India. Meanwhile, a study by Deininger and Squire (1998) also supported the theory that rising income inequality is not caused with economic growth.

To summarize, although the multitude of studies that have studied the effect of economic growth on income inequality, a common and definite answer cannot be concluded. Based on the analysis above, the following hypotheses are developed:

Hypothesis 4: A country’s level of economic growth has a positive relationship with its level of income inequality.

(15)

2.5 Control variable

2.5.1 Education

Another key process is access to education. Some theorists argue that education allows for the attainment of credentials and skills necessary for employment in the modern industrial sectors of the economy (Simpson 1990). This argument is derived from modernization perspective (Crenshaw 1992) in that the relationship is dependent on national economic growth and increasing internal sectorial complexity. Thus, a positive relationship exists between education and income inequality. Specifically, the more people have access to education, the more likely for income inequality to rise.

2.5.2 Technology

(16)

3. Methodology

3.1 Empirical Mode

The main focus of this paper is to study the determinants of income inequality. I run an OLS regression with income inequality as the dependent variable and export, FDI, inflation and growth as the key independent variables. The mode formula can be shown as follow: Inequalityct= α+ β0+ β1EXPit+ β2 FDIit+ β3INFit+β4GRWit+β5Xit+λ+ ε

The dependent variable “Inequalityct” is income inequality of country c in year t according to

Gini index. The independent variable “EXPit” stand for the export of goods and services,

available to country i in year t. "FDIit” stand for the inflow of foreign direct investment in

country i in year t. "INFit” stand for the Inflation in country i in year t. "GRWit” stand for the

GDP per capita growth in country i in year t. X stands for control variables that affect income distribution.

Two fixed effects are included in the mode: α is the country-specific effect, and λ is the year effect. Lastly, ε represent the error term that changes across countries and over time. To handle the potential endogeneity arising from control variables, I use a GMM panel estimation based on relevant explanatory variables.

3.2 Data

This paper examines the determinants of country's income inequality from 1996 to 2010. Based on Human development Index, the countries are categorized into three sub-groups, which are Developed countries, developing countries and Least Developed Countries (LDC).

--- Insert Table 1 Here ---

(17)

Philippines, Poland, Russia, South Africa, South Korea, Switzerland, Uganda, United Kingdom, United States, Yemen, Zambia and Zimbabwe (Table 1).

The 45 countries are composed by countries from different stages of development according to Human Development Index. Together these nations paint a comprehensive picture for the world's income inequality. The 45 countries comprise in total of more than 60 percent of the world’s population, account for more than 70 percent of world’s GDP and deposit more than half of the world’s natural resources over all continents.

3.3 Data Collection

I gathered data from several different data centers; these are the International Monetary Fund, International Trade Statistics Database, The World Bank, the United Nations Development Programme, the Standardized World Income Inequality Database, United Nations Conference on Trade and Development and UNESCO Institute for Statistics. For each country, I collect data about income inequality, export of goods and services, foreign direct investment net inflow, inflation, GDP per capita growth, labor force with secondary education and research and development expenditure and trade openness from the time period 1996-2010. The complete descriptions and sources of the data are presented in table 2.

I use Gini index from the Standardized World Income Inequality Database to measure the income inequality of a nation. Gini coefficient is a wide accepted measurement of the income distribution of a country’s labor force. The GINI index, which ranges between 0 and 1, define the gap between high-income group and low-income group. 0 means totally equal and 1 means totally unequal distribution of income.

While Gini index is a popular indicator of income inequality, it is not without limitations. First, income distribution measurement is not the same as wealth distribution measurement, which cannot be indicated by Gini. A poor nation could have the same level of Gini

(18)

only as accurate as the data produced by a country. This is a persistent problem for many countries, especially developing nations. Therefore, the index offers only a proxy estimation rather than exact reality. Furthermore, there is a negative correlation between GDP per capita and Gini coefficients, because low-income countries tend to have higher index numbers. 3.3.1 Measurement

Exports of goods and services (annual % growth)

Annual growth rate of exports of goods and services based on constant local currency. Aggregates are based on constant 2005 U.S. dollars. Exports of goods and services represent the value of all goods and other market services provided to the rest of the world. They include the value of merchandise, freight, insurance, transport, travel, royalties, license fees, and other services, such as communication, construction, financial, information, business, personal, and government services. They exclude compensation of employees and investment income (formerly called factor services) and transfer payments.

Foreign Direct Investment

Foreign direct investment are the net inflows of investment to acquire a lasting management interest (10 percent or more of voting stock) in an enterprise operating in an economy other than that of the investor. It is the sum of equity capital, reinvestment of earnings, other long-term capital, and short-long-term capital as shown in the balance of payments. This series shows net inflows (new investment inflows less disinvestment) in the reporting economy from foreign investors. Data are noted in current U.S. dollars.

Inflation

Inflation as measured by the consumer price index reflects the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services that may be fixed or changed at specified intervals, such as yearly. The Laspeyres formula is generally used.

Growth: GDP per capita growth (annual %)

(19)

added by all resident producers in the economy plus any product taxes and minus any

subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. Education: Labor force with secondary education (% of total)

Labor force with secondary education is the share of the total labor force that attained or completed secondary education as the highest level of education.

Technology: Research and development expenditure (% of GDP)

(20)

4. Empirical Analysis

4.1 Quality of Data

4.1.1 Summary

A total of 45 countries were included in this study, they are suitable representation of the major economies around the global and they all have reasonable amount of influence on the global economy. The countries are in different economic development stages and are vary in political and culture characteristics and therefore selection bias is avoided. The data are collected from the time period 1996 to 2010. The dataset covers a wide range of developing and developed economics, which I believe is reliable for panel data study. Table 3 presents the descriptive results of the data.

In total of 675 observations were included in this panel dataset. Though there are missing data in many of the independent variables and control variables, the remaining data information is still valuable for analysis.

4.1.2 Normality

Table 4 in appendix 3 shows the normality test result. The degrees of freedom are the consistent for all variables while P values are greater than 0.05. Since my sample is smaller than 2000, Shapiro-Wilk test was conducted to confirm the normality of the data. If the p value is greater than 0.05, the null hypothesis can be accepted and the normality can be assumed. Therefore, I conclude that distribution of data are normal and are acceptable for panel data study.

(21)

4.1.3 Correlation

After confirming the normality of the data, this study moves forward to examine the

correlation matrix for correlation between the independent variables. The purpose of this step is to facilitate understanding of results. Correlations between independent variables are not ideal, however a high correlation between the independent and dependent variable is desired for further analysis. As the correlation table show (table 5), there is no correlation greater than 0.7 in the data, indicating no strong relationship between independent variables. Therefore, I conclude the data does not suffer form high correlation. However, further tests need to be done to eliminate make sure there is no multicollonerity.

4.1.4 Multicollinearity

Before running the model tests that were used for find correlation, it is necessary to see if the regression analysis contains any multicollonerity risk. Multicollonerity refer to the situation where one independent variable is highly correlated to the others. This means that two collinear variables are providing the same information, and therefore cannot separate the explanatory power of individual variables (Neter et al., 2004). Therefore, I check for the normality of the residuals in order to confirm the absence of multicollinearity (Appendix 1). Initial correlation matrix shows potential (>0.5 or <-0.5) of multicollinearity; therefore variance inflation factor (VIF) tests were done for further analysis. VIF are below 10 across all level (Tables below), thus there is no risk of multicollinearity. I obtain an average variance inflation factor (VIF) of 2.6194 ranging from 1.124 to 4.655, well below the accepted limit of 10 (Neter et al., 2004). One of the concerns in this study is that the effect of FDI may over lap with export, which leads to multicollinearity between the two. However this issues is

(22)

4.1.5 Heteroskedasticity and outliers

Scatterplot of residuals (Appendix 2) are used to check heteroskedasticity before regression analysis. Scatterplots shows no erratic or non-consistent relationship. An added fit line on scatterplots confirms the consistent relationship between variables. Therefore it is safe to conclude that there is no heteroskedasticity in the data. Furthermore, scatterplot also confirm that there are no extreme outliers in the dataset.

4.2 Model Tests

The empirical analysis start with testing panel regression models for all independent variables that are hypothesized to contribute to income inequality. The first model is a simple panel with a common constant. The results are present in Table 6.

The dependent variable in all regression is the GINI coefficient, therefore a positive figure indicate an increase in income inequality. Only independent variables in the model

(regression 1) and control variables were added in the following models (regression 2), until a full model (regression 3) is reached. The last regression in table 6 indicates that the

(23)

The results show that inequality is positively affected by Export of goods and services. Therefore, the more active a country engage in export activities, the higher the income inequality. In terms of FDI, it is found that FDI have a greater impact on inequality than Export. The results are positive and statically significant, thus suggesting an increase occurs in inequality again. Inflation is not significant, and therefore is not included in other fixed-effect models. Growth has a positive relationship with inequality, and the results shows stronger correlation after control variables were included in regression 2 and 3. The results for control variables are consistent with common constant models. Education is not

significantly correlated with income inequality. Meanwhile, technology has a positive relation with inequality, meaning the more advanced a country in technology development, the greater the income gap. It is important to note that after testing for redundant fixed effects, the null hypothesis is rejected, which means the fixed effect models offer more meaningful results than common constant models.

The third panel regression method used is random effect model. The results for this model testing are presented in the second part of table 7. The same procedure was employed, where regression 4 includes all independent variables, and then put in each control in later

regression (regression 5 and 6). The results of full models (regression 6) share many similarity with the fixed effect model in terms of export, FDI and growth. In addition, inflation was significant at first (regression 4), but the results of latter regressions (5 and 6) show no significance again. With regard to the control variables, the results are rather similar with fixed- effect models. Education was found insignificant and removed in regression 5 and regression 6. On the other hand, Technology is statistically significant and has a positive impact on income inequality.

(24)

model in favor of the random effects model. The second reason is that the random effect model shows smaller standard errors and a higher R-sq.

4.2.1 Country Sub-groups

The next step is to reduce heterogeneity among countries. This time all specification was estimated by categorize countries sample to more homogeneous sub-groups. The sub-groups are divided as developed countries, developing countries and least developing countries. The results of these tests are presented in Table 8. There are two specifications for each sub-group, 1,3,5 are the independent only models and 2,4,6 are the full models. From the full models of each group, export increases income inequality for all three sub-groups. Therefore export of goods and service can be concluded to increase inequality in all major countries. It is important to take notice that export affects different country groups differently. To be specific, export has the most impact on developing countries, moderated effect on developed countries; and least effect on LDCs.

Regarding Foreign Direct Investment, the results are consistent across country groups. In general, FDI has a positive relationship with income inequality, meaning that the more FDI a country receive, the more likely inequality increase. Similar to export of goods and service, FDI also has the most influence on developing countries and reduced impact on developed and least developed countries.

In terms of inflation, the results are rather ambiguous among different sub-groups. While inflation has a small effect on developed countries, the developing countries and LDCs seems to be not affected by inflation at all. Overall, this study concludes that inflation does not affect income inequality very significantly.

(25)

groups. Therefore, this study believes economic growth is contributing to income inequality on a global scale.

4.2.2 Endogeneity

In order to test robustness, as well as to take into account potential endogeneity problems, a repeat process of country sub-groups estimation procedure was conducted again using GMM method. The results are shown in Table 9.

(26)

5. Empirical Result

In this section, the empirical results of models from previous section are discussed. In the first sub-section an explanation is provided regarding the outcome generated from statistical testing. The second section continues with the results of each independent variable of multiple regression analysis.

5.1 Model Analysis

In this study the adjusted Rsquare is used for testing the predictive power of different models. The adjusted R square is chosen over R square, because R square estimation are overly optimistic about the how well a model fits the population. Adjusted R square therefore decreases R square accordingly. Another advantage of adjusted R square is it is also useful for comparing models with different number of independent variable because it adjusts the model for increase expected in sample R square when more variables are put in (Neter et al., 2004). In the comparison of fixed-effect model and random-effect model, we can see that random-effect model has higher adjusted R square in most scenario. Therefore, random-effect model is better than fixed-effect in terms of predictive power.

Furthermore, according to Hausman test random-effect is preferred when P>0.05. As shown in table 6, the results of Hausman test is 0.07, which means random-effect model is the most suited test in this study. Therefore, combined with the model evaluation from last section, Random-effect model was chosen for the study and the results are presented in Table 10.

(27)

5.2 Model Results

After tests of data quality and conclude the most preferred model of testing, the independent variables and their explanatory power on the dependent variable is discussed in the following section.

5.2.1 Export

The first hypothesis predicts that a country’s export activities have a positive relationship with income inequality. The coefficients of export are positive and significant in all three models. It is also important to notice that export has the largest coefficient among all

independent variables, which means it contribute the most to income inequality in this study. This directly support the hypothesis that the more a country is engaged in export activities, the more likely for its income gap to widen. Therefore, hypothesis 1 is supported.

More evidence is provided in country sub-group test. The coefficient of export changed significantly for developed countries. This means while export still play a role in income inequality at these countries; its influence is much smaller than other independent variables. On the other hand, the export coefficients stay large and significant in developing countries and LDCs, suggesting that they are more vulnerable to export caused income discrepancy than developed countries. Thus, hypothesis 1a is supported as well.

5.5.2 Foreign Direct Investment

The second hypothesis predicts the relationship between foreign direct investment and income inequality. The coefficients of FDI are significant and positive as well. This confirms the hypothesis that the more FDI in the country, the more likely for income inequality to rise. Therefore, hypothesis 2 is supported.

(28)

5.5.4 Inflation

The third hypothesis predicts the influence of inflation on income inequality. Inflation was believed to have a positive relation with income inequality. However, the coefficient results show that inflation is insignificant in all models. This means there is no support for the proposed relation between inflation and income inequality. Therefore, hypothesis 3 is rejected.

The results of inflation from country sub-groups are more or less the same with the exception of developed countries. In both developing and LDCs, inflation remains insignificant and non-influential. However, the significant results of developed countries proposed something different. Inflation may have a greater impact on developed countries than on the others. This makes developed country more susceptible to rising inflation. Thus, hypothesis 3a is rejected as well, and the opposite is true.

5.5.5 Growth

The forth hypothesis predicts the effect of economic growth on income inequality. The coefficients of growth are rather consistent. All three models show significant and positive coefficients, which support the proposed relationship in hypothesis 3. The development of economic growth will further increase country’s income inequality. Thus, hypothesis 4 is supported.

Growth also appears to be a major origin of income inequality. All three sub-groups have significant and positive coefficient, though the extent of influence is different based on development. The developed countries are the most susceptible to economic growth; the developing counties are the second; and the LDCs are the least affected by a rather great margin. This indicates that economic growth plays a unique role at creating inequality depending on development level of country. Therefore, hypothesis 4a is rejected. 5.5.6 Control Variables

Education

(29)

significant. This could mean that education plays a more influential role when taking the effect of technology into consideration.

Technology

Technology is the last control variables, and it is included in model 3. The coefficient of technology is significant and positive. This support the relationship discovered in literature review, that when country moves forward in technological advancement, the income gap would be enlarged by it.

5.3 Conclusion empirical results

In this section I sum up the finding from empirical results. First of all, export of goods and service has a positive relationship with income inequality. This result is aligned with previous study (Batra and Slottje, 1993; Goldberg & Pavcnik, 2004; Yao, 2002; Wedeman, 2004; Gong, 2006; Wang and You 2012). The extent of influence differs depending on country development level. To be specific, the effects is far greater in LDCs and developing countries, but less in developed countries.

Secondly, foreign direct investment raises gap between income groups. This result confirms findings from other studies (Khan and Riskin 1998; Deardorff 2005; and Horgos 2009; Marjit et al., 2004; Chaudhuri and Yabuuchi, 2007; Anwar and Rice, 2009). The exact level of influence differs depending on country stage of development, though the distinctions are not as great as other predictors.

Thirdly, inflation was found to be not significantly influencing income inequality in general. This is in direct contract with believes of other scholars (Frydman and Jenter, 2010; Cahuc, Saint-Martin, Zylberberg, 2001; Chiroleu-Assouline and Fodha 2014; Poterba, 2007). On the other hand, developed countries are more liable to be affected by rise in inflation, while developing and LDCs are not.

Lastly, the positive relationship between economic growth and income inequality is

(30)

6. Discussion

The empirical results conclude form last chapter provided some interesting findings. In light of these empirical results I discuss these outcome in this chapter. The theoretical perspective and empirical links between numerous predictors and inequality presented in chapter 2 is very useful for coherent interpretation of the results described in last section.

6.1 Export

Starting with the role of export of goods and service, the empirical evidences shows an increase in income inequality from export in 45 countries. Many studies have offered several different explanations as how export affects income inequality. Dependency theorists believe trade between developed and developing countries have a tendency to create unequal

exchange, causing high level of equality between them (Porto, 2003).

First of all, Policy makers in these economies usually adopted low wage, along with social and legal barriers, to limit labor force mobility and low skill transferability.

Secondly, Export is capable of creating high profit and high return for people work in this sector. Workers in export business enjoy higher compensation and benefits than unskilled employee, which directly leads to income inequality (Harrison, McLaren, & McMillan, 2010). This is especially true for developing countries and least developed countries where wage in traditional industries are significantly lower than export workers (Giesecke, Heisig & Solga 2015).

Thirdly, commodity concentration can also offer an explanation to the widening income gap. LCDs and developing countries with high commodity orientated export have less high skilled worker, and less diversified value added products (Breau and Rigby, 2010). At the same time, high commodity concentration country are more sensitive to downturn price pressure, while value added product exporter can offset these pressure with non-commodity goods and services. This theory is confirmed in the empirical study; where export has a larger impact on developing countries and LCDs than developed countries.

6.2 Foreign Direct Investment

(31)

by foreign company are still higher than average in local economy (Eckel, 2003; Moore and Ranjan, 2005; Beladi et al., 2008; Anwar and Rice, 2009). Such arrangement directly leads to income gap within in the local country. Workers employed by foreign firms enjoy higher compensation than those who works in local sectors.

The second reason can be summarized as repatriation of profits. Such practice employed by cross-broader cooperation greatly reduces the amount of resource available for re-investment in local economy. Therefore, it is not surprise to see an ever-growing inequality between the developed and underdeveloped world (Khan and Riskin 1998; Deardorff 2005; and Horgos 2009).

A third argument can be made from the access to FDI permits. The rich and well connected are better position to obtain permission to enter highly profitable business arrangement with foreign investors (Andres and Ramlogan-Dobson 2011; Gupta et al., 2002). The powerful ones also have a more opportunities to attract foreign direct investment (Gyimah-Brempong and Muñoz de Camacho, 2006). Meanwhile, the opposite line of reasoning is not reflected in reality. Villegas-Sanchez (2009) proposed that foreign capital inflow into the developing world lower the borrowing cost, and thus making credit more accessible to the regular people. However, results from this study find that FDI increases inequality in countries of all

development stage. The benefits brought by FDI are largely distributed among themselves; and among the rich and powerful few with foreign business connection.

Lastly, policy makers in local country play an important role. Banga (2005) and Anwar (2006) suggest that developing country governments face problems from the highly mobile foreign capital. In order to attract and retain FDI, policy makers are encouraged to set favorable regulation for foreign investors. However, these policies limit the negotiation power of labor force and reduce mobility of workers. Furthermore, incentives like tax concessions and profit repatriation makes closing the income gap between the rich and poor much more difficult (Taylor and Wyatt 1996; Glytsos 2002).

6.3 Inflation

(32)

immune to the movement of inflation. Inflation-adjust contract includes stock options, shares, and non-cash benefits in employee’s compensation package. This type of contract is typically used to construct compensation for top management talent. Furthermore, assets owners are also shielded form inflation, since they have the choice to invest in assets that are not easily affected by inflation. They can also invest in asset with high growth rate to offset loses from inflation devaluation (Fryman and Jenter, 2010).

The second argument is that rich class prefers regressive tax against progressive tax. They also have more influence in policy setting, which means they regressive tax like inflation is implemented more often than income tax and inheritance tax. This is especially true for underdeveloped countries where democratic institution environment is too weak to protect the interest of the poor (Chiroleu-Assouline and Fodha 2014). Researchers also proposed that raise minimum wage is not effective in reducing inequality, because workers receive nominal contract is far greater than inflation-adjusted contract (Poterba, 2007).

Controversy to the theories mentioned above, results in this study find no correlation between inflation and income inequality. One possible reason is during the most recent decades the world has gone through astonishing economic development. In most countries people’s income has out grown if not on par with inflation increase, resulting relatively stable purchase power for nominal contract worker.

6.4 Economic Growth

The relationship between economic growth and inequality has been long debated by the research community. One school of scholars argues that as economic structure shift towards industrialization, income gap will first increase and then decrease. In another word, an inverted U-shaped relationship exists between growth and inequality.

At the beginning stage of industrialization, labors from traditional agricultural sector were freed form low-income manual activity. These labors then migrate into modern sector, which generally have higher return (Fan and Stark, 2008). This creates a gapping income

(33)

and eventually arrive equilibrium (Lee and Saez, 2012). Therefore, an inverted U-shaped relationship was concluded form this theory.

However, many researches are directly opposing the inverted U-shaped theory. Galor and Zeira (1993), as well as Banerjee and Newman (1993), argues that economic gain generated though urbanization is unevenly distributed among society. The owners of production facility gain significantly more economic benefit than the labors in production (Galor, 2000). Many other researchers also support the proposal that economic growth create larger gap between income groups due to unbalanced distribution (Elbers and Lanjouw, 2001; Shahbaz, 2010). While the debate between inverted U-shaped theorist and linear theorists is not settled, a third school of studies find that no relationship exists between growth and inequality (Ravallion, 2005; Das and Barua, 1996; Deininger and Squire, 1998).

(34)

7 Limitations and Future Research

7.1 Limitations

This study provides some important insight for the inequality research field. However, this study also has few limitations, some of which are suggested for future research. The first limitation is that this study focuses on income inequality. As proposed by piketty (2014), income inequality is very different from wealth inequality. Therefore the results cannot be generalized to wealth inequality and opportunity inequality.

Secondly, the number of country involved in this study is limited, though there are enough observations for a meaningful panel study. To conclude more comprehensive result for the whole world, more countries are needed for analysis.

Lastly, due to limited time and resources that can be spend on this research several variables are measured by the less preferred data. For example, economic growth is measured by GDP growth per capita. However literature review suggests industrialization should be used to describe growth, and it is positively correlated with income inequality (Galor, 2000, Zhu and Trefler 2005; Fan and Stark, 2008; Zhu and Trefler 2005).

7.2 Future Research

Suggestions for future research can be drawn from some of the limitations mentioned in the last section. First of all, more determinants are to be tested in order to confirm their

relationship with income inequality. Inequality is a large and complicated topic that has been studied by researchers for decades. The true cause of inequality goes well beyond the

(35)

(36)

8 Conclusion

Income inequality is one of the most ubiquitous challenges around the world. To assess the extent of inequality, economists developed a wide range of measure. The public and

researchers of inequality have cited those measure in their publications. During the past few decades, many researches dedicate their efforts to examine how income is distributed. Even though many common understanding have been achieved, a lot of ambiguities still remain in the research field.

The goal of this study is to find the determinants of income inequality; to confirm or rejects theories from precious works; and to distinguish the difference between different countries. In Chapter 1, the topic of income inequality is introduced and briefly discussed. Inequality has become a more and more pressing issue; it requires more attention from both researches and the public than ever before.

In Chapter 2, literatures from the past are reviewed. Based on these literatures, hypotheses regarding the relationship between export, foreign direct investment, inflation and growth are developed. This study also proposes that each determinant have different influence on

countries form different development stage.

Chapter 3 presents the regression model used for panel analysis and data collation. Gini coefficient is used to measure income inequality. Export of goods and service volume is used to measure export activity. Foreign direct investment inflow data measures FDI. Inflation is measured by country inflation fluctuation. Economic growth is quantified by GDP per capita. Education and technology are included as control variables.

In Chapter 4 the data quality is tested to rule out potential problems like multicollinearity and heteroskedasticity. Several models were compared and random-effect model is concluded as the best model. Country sub-groups analysis and GMM analysis are conducted to gain more insight, and to reduce the risk of endogeneity.

(37)

significant impact on inequality regardless the level of development. The relationship

between inflation and inequality is rejected, with the exception of developed country. Lastly, growth indeed contributes to the widening income gap. Though the inverted U-shaped relationship cannot be confirmed or rejected completely.

In Chapter 6 the empirical results are linked to theories form literature review. Discussion about each determinant concludes their relationship with income inequality. Furthermore, their explanatory mechanisms are reexamined. Chapter 7 mentions the limitations of the study and provides suggestions for future study.

In the end, inequality is not an issue developed over night. There is perhaps no simple

(38)

Source:

Adam, A., Kammas, P., & Lapatinas, A. (2015). Income inequality and the tax structure: Evidence from developed and developing countries. Journal of Comparative

Economics, 43(1), 138-154.

Andres, A. R., & Ramlogan-Dobson, C. (2011). Is corruption really bad for inequality? Evidence from Latin America. Journal of Development Studies, 47(7), 959-976.

Anwar, S., & Rice, J. (2009). Labour mobility and wage inequality in the presence of endogenous foreign investment. Economic Modelling, 26(6), 1135-1139.

Arbache, J. S., Dickerson, A., & Green, F. (2004). Trade Liberalisation and Wages in Developing Countries*. The Economic Journal, 114(493), F73-F96.

Banerjee, A. V., & Newman, A. F. (1993). Occupational choice and the process of development. Journal of political economy, 274-298.

Banga, R. (2005). Critical issues in India’s service-led growth. Indian Council for Research. Batra, R., & Slottje, D. J. (1993). Trade Policy and Poverty In the United States: Theory and Evidence, 1947‐1990*. Review of International Economics, 1(3), 189-208.

Beladi, H., & Batra, R. (2004). Traded and nontraded goods and real wages. Review of

Development Economics, 8(1), 1-14.

Beladi, H., Chaudhuri, S., & Yabuuchi, S. (2008). Can International Factor Mobility Reduce Wage Inequality in a Dual Economy?. Review of International Economics, 16(5), 893-903. Bénabou, R. (2005). Inequality, technology and the social contract. Handbook of economic

growth, 1, 1595-1638.

Berman, E., Bound, J., & Griliches, Z. (1994). Changes in the demand for skilled labor within US manufacturing: evidence from the annual survey of manufacturers. The Quarterly Journal

of Economics, 367-397.

(39)

Bourguignon, F., Ferreira, F. H., & Menendez, M. (2007). Inequality of opportunity in Brazil.

Review of Income and Wealth, 53(4), 585-618.

Breau, S., & Rigby, D. L. (2010). International trade and wage inequality in Canada. Journal

of Economic Geography, 10(1), 55-86.

Cahuc, P., Saint-Martin, A., & Zylberberg, A. (2001). The consequences of the minimum wage when other wages are bargained over. European Economic Review, 45(2), 337-352. Chiroleu-Assouline, M., & Fodha, M. (2014). From regressive pollution taxes to progressive environmental tax reforms. European Economic Review, 69, 126-142.

Christiaensen, L., & Todo, Y. (2014). Poverty reduction during the rural–urban transformation–the role of the missing middle. World Development, 63, 43-58. Cline, William R. Trade and income distribution. Peterson Institute, 1997.

Das, S. K., & Barua, A. (1996). Regional inequalities, economic growth and liberalisation: A study of the Indian economy. The Journal of Development Studies, 32(3), 364-390.

Deardorff, A. V. (2005). A trade theorist's take on skilled-labor outsourcing. International

Review of Economics & Finance, 14(3), 259-271.

Deininger, K., & Squire, L. (1998). New ways of looking at old issues: inequality and growth. Journal of development economics, 57(2), 259-287.

Dobson, S., & Ramlogan-Dobson, C. (2010). Is there a trade-off between income inequality and corruption? Evidence from Latin America. Economics letters, 107(2), 102-104.

Crenshaw, E. (1992). Cross-national determinants of income inequality: A replication and extension using ecological-evolutionary theory. Social Forces, 71(2), 339-363.

Eckel, C. (2003). Labor market adjustments to globalization: unemployment versus relative wages. The North American Journal of Economics and Finance, 14(2), 173-188.

Fan, C. S., & Stark, O. (2008). Rural-to-urban migration, human capital, and agglomeration.

(40)

Flynn, S. M., & Donnelly, M. (2012). Does labor contract completeness drive unionization? Experimental evidence. The Journal of Socio-Economics, 41(4), 445-454.

Frydman, C., & Jenter, D. (2010). CEO compensation (No. w16585). National Bureau of

Economic Research.

Galor, O. (2000). Income distribution and the process of development. European Economic

Review, 44(4), 706-712.

Galor, O., & Zeira, J. (1993). Income distribution and macroeconomics. The review of

economic studies, 60(1), 35-52.

Giesecke, J., Heisig, J. P., & Solga, H. (2015). Getting more unequal: Rising labor market inequalities among low-skilled men in West Germany. Research in Social Stratification and

Mobility, 39, 1-17.

Glytsos, N. P. (2002). The role of migrant remittances in development: evidence from Mediterranean countries. International migration, 40(1), 5-26.

Goldberg, P. K., & Pavcnik, N. (2004). Trade, inequality, and poverty: What do we know? Evidence from recent trade liberalization episodes in developing countries (No. w10593).

National Bureau of Economic Research.

Goldberg, P. K., & Pavcnik, N. (2004). Trade, inequality, and poverty: What do we know? Evidence from recent trade liberalization episodes in developing countries (No. w10593).

National Bureau of Economic Research.

Goldberg, P. K., & Pavcnik, N. (2007). Distributional effects of globalization in developing countries (No. w12885). National bureau of economic research.

Gong, T. (2006). Corruption and local governance: the double identity of Chinese local governments in market reform. The Pacific Review, 19(1), 85-102.

(41)

Gupta, S., Davoodi, H., & Alonso-Terme, R. (2002). Does corruption affect income inequality and poverty?. Economics of Governance, 3(1), 23-45.

Gyimah-Brempong, K., & de Gyimah-Brempong, S. M. (2006). Corruption, growth, and income distribution: Are there regional differences?. Economics of Governance, 7(3), 245-269.

Harrison, A., McLaren, J., & McMillan, M. S. (2010). Recent findings on trade and inequality (No. w16425). National Bureau of Economic Research.

Horgos, D. (2009). Labor market effects of international outsourcing: How measurement matters. International Review of Economics & Finance, 18(4), 611-623.

Janko, Z. (2008). Nominal wage contracts, labor adjustment costs and the business cycle.

Review of Economic Dynamics, 11(2), 434-448.

Khan, A. R., & Riskin, C. (1998). Income and inequality in China: Composition, distribution and growth of household income, 1988 to 1995. The China Quarterly, 154, 221-253.

Knight, J., & Song, L. (2003). Increasing urban wage inequality in China. Economics of

Transition, 11(4), 597-619.

Kutner, N. Neter (2004). Applied Linear Regression Models, 4.

Lee, D., & Saez, E. (2012). Optimal minimum wage policy in competitive labor markets.

Journal of Public Economics, 96(9), 739-749.

Marjit, S., & Kar, S. (2005). Emigration and wage inequality. Economics Letters, 88(1), 141-145.

Marjit, S., Beladi, H., & Chakrabarti, A. (2004). Trade and wage inequality in developing countries. Economic Inquiry, 42(2), 295-303.

(42)

Michaels, G. (2008). The effect of trade on the demand for skill: Evidence from the interstate highway system. The Review of Economics and Statistics, 90(4), 683-701.

Mohammed, A. (2014, January 1). Deeping income inequality. Retrieved August 10, 2015, from World Economic Forum: http://reports.weforum.org/outlook-global-agenda-2015/top-10-trends-of-2015/1-deepening-income-inequality/

Moore, M. P., & Ranjan, P. (2005). Globalisation vs Skill‐Biased Technological Change: Implications for Unemployment and Wage Inequality*. The Economic Journal, 115(503), 391-422.

Piketty, T. (2014). Capital in the 21st Century. Cambridge: Harvard Uni.

Porto, G. G. (2003). Trade reforms, market access and poverty in Argentina (Vol. 3135).

World Bank Publications.

Poterba, J. M. (2007). Income inequality and income taxation. Journal of Policy Modeling, 29(4), 623-633.

Prechel, H. (1985). The effects of exports, public debt, and development on income inequality. The Sociological Quarterly, 26(2), 213-234.

Ravallion, M. (2005). Inequality is Bad for the Poor. World Bank Policy Research Working Paper, (3677).

Reardon, T., Berdegué, J., & Escobar, G. (2001). Rural nonfarm employment and incomes in Latin America: Overview and policy implications. World development, 29(3), 395-409. Robbins, D. J. (1996). Evidence on trade and wages in the developing world.

Shah, A. (2006). Fiscal incentives for investment and innovation. Available at SSRN 896144. Shahbaz, M. (2010). Income inequality-economic growth and non-linearity: a case of

(43)

Simpson, M. (1990). Political rights and income inequality: A cross-national test. American

Sociological Review, 682-693.

Taylor, J. E., & Wyatt, T. J. (1996). The shadow value of migrant remittances, income and inequality in a household‐farm economy. The Journal of Development Studies, 32(6), 899-912.

Villegas-Sanchez, C. (2009). FDI spillovers and the role of local financial markets: evidence from Mexico. Manuscript, University of Houston.

Wang, M. C., Fang, C. R., & Huang, L. H. (2009). International knowledge spillovers and wage inequality in developing countries. Economic Modelling, 26(6), 1208-1214.

Wang, Y., & You, J. (2012). Corruption and firm growth: Evidence from China. China

Economic Review, 23(2), 415-433.

Wedeman, A. (2004). Great disorder under heaven: Endemic corruption and rapid growth in contemporary China. China Review, 1-32.

World Economic Forum. (2014). Outlook on the Global Agenda 2014. Switzerland: World

Economic Forum.

Yabuuchi, S., & Chaudhuri, S. (2007). International migration of labour and skilled–unskilled wage inequality in a developing economy. Economic Modelling, 24(1), 128-137.

Yao, S. (2002). Privilege and corruption: The problems of China's socialist market economy.

American Journal of Economics and Sociology, 61(1), 279-299.

(44)

Appendix

Appendix 1 – Multicollinearity Test

Coefficients

Model

Collinearity Statistics Tolerance VIF Export of goods and services 0.304 1.153 Foreign Direct Investment 0.869 1.124 Export of goods and services 0.336 1.625 Foreign Direct Investment 0.303 1.261

Inflation 0.913 2.326

Export of goods and services 0.182 2.116 Foreign Direct Investment 0.279 3.582

Growth 0.213 4.582

Growth 0.735 2.364

Education 0.892 2.143

Growth 0.432 3.119

Education 0.752 2.174

Technology 0.923 1.953

(45)

(46)

(47)

(48)

(49)

Appendix 3 -‐ Tables

Table 1 - Country list

Country lists

Developed Developing LDC Australia Argentina Cambodia

Belgium Brazil Chad

Canada Chile Colombia

Denmark Czech Republic Myanmar Finland Hungary Nepal

France India Uganda

Germany Indonesia Yemen

Iceland Iran Zambia

Ireland Israel Zimbabwe

Italy Lithuania

Japan Malaysia

Netherlands Mexico

New Zealand Pakistan Norway People's Republic of China Switzerland Philippines

United

Kingdom Poland

United States Russia

South Africa

(50)

Table 2 – Data Description

Concept Variable Source Code

Income

inequality GINI index (World Bank estimate) World Bank national accounts data, and OECD National Accounts data files, the United Nations Development Programme and the Standardized World Income Inequality Database

INE

Export Exports of goods and services (annual % growth)

International Monetary Fund, International Trade Statistics Database, World Bank national accounts data, and OECD National Accounts data files

EXP Foreign Direct Investment Foreign direct investment, net inflows (BoP, current US$)

International Monetary Fund, United Nations Conference on Trade and Development, World Bank national accounts data, and OECD National Accounts data files

FDI

Inflation Inflation, consumer

prices (annual %) International Monetary Fund, International Financial Statistics and data files. INF Growth GDP per capita

growth (annual %) World Bank national accounts data, and OECD National Accounts data files GWO Education Labor force with

secondary education (% of total)

UNESCO Institute for Statistics, World Development Indicators

EDU

Technology Research and development expenditure (% of GDP)

World Bank national accounts data, and OECD National Accounts data files

TEC

Table 3 - Descriptive

(51)

Table 4 – Normality Test

Shapiro-Wilk Statistic df Sig. Export 0.971 123 0.053 FDI 0.567 123 0.061 Inflation 0.704 123 0.058 Growth 0.972 123 0.066 Education 0.957 123 0.056 Technology 0.879 123 0.078 Income Inequality 0.937 123 0.079 * This is a lower bound of the true

significance

Table 5 - Correlation

Variables Income inequality

Export FDI Inflation Growth Education

Export 0.3502* FDI 0.4276* 0.027 Inflation 0.036* -0.017 0.2017 Growth 0.013 0.0786* 0.0563* 0.075 Education 0.036** -0.016 -0.0077 -0.06** 0.0329* Technology 0.065** 0.043 -0.0012* -0.095 0.0681 -0.035*

T statistics are calculated based on robust standard errors. Standard errors are in parentheses * Means statistical significance at the 10% level (2 tailed)

The Determinants Of Income Inequality: A System Panel Data Analysis