Abstract One of the major issues on the state of income inequality is the effect of globalization through foreign direct investment (FDI). It is well known that FDI inflows create employment opportunities for unskilled labor intensive countries. Hence, during recessionary (expansionary) periods, FDI outflows should cause an increase in a developing (developed) country's unemployment rate, worsening income inequality. This study differs from the previous literature by employing the key variables FDI, trade volume, and GINI coefficient for a panel of three groups of countries (developed, developing, and miracle countries). We estimated panel cointegration coefficients via FM-OLS. Our results show that the effects of trade liberalization and FDI on income distribution differ for different country groups.
Keywords FDI * Income inequality * Globalization * Trade volume * Panel cointegration
JEL C10 * F01 * F15 * J30
Globalization has been the magical word used to define the recent episodes of growth and increase in global welfare. With declines in tariffs or creation of free trade areas since the 1990s, globalization has built a set of prospects for mainly large firms around the world with declines in tariffs or creation of free trade areas. Companies have discovered ways to reach out to unchartered markets around the world. Accordingly, companies have also learnt how to decrease their costs further because it became possible for corporations to operate in economies that offer inherent opportunities, such as very low-wage-labor, convenience in tax payments, and lower transaction costs. Hence, globalization has led to broader perspectives and opportunities for many firms, the bulk of which belong to advanced economies. Nevertheless, globalization is beneficial not only for the firms of industrialized countries but also for the ones in developing nations. As a matter of fact, further integration of the developed world would be a bounded one with only fractional benefits. Thus, the steps that the developing economies took through globalization have been significant in their achievement of sustainable development processes. Convincing the developing countries to be more liberalized has also been very crucial, due to creation of additional gains for developed countries.
The globalization-led-international competition has caused firms of developing countries to struggle for market share, get rid of government subsidies and/or protectionist policies, and learn how to stand tall among giant corporations of developed countries. Trade liberalization brought forward low-cost products, making the consumers of the developing economies better off. Through channels of globalization, such as foreign direct investment (FDI), the technological know-how has reached around the globe. Lipsey (2002) argues that the yields of FDI are the contribution to growth by export enhancing developments, increase in revenue for budget through taxation of foreign firms (i.e., Multinational Corporations), and new employment opportunities. (1)
Apart from the economic changes, the interactions among economic and social development is crucial for all governments in the long term. In this respect, income equality arguably could be termed as the one of the most important indicators for social justice. The importance of social justice comes from the fact that achieving sustainable economic growth is necessary but not enough in a globalized world. In order to ensure social cohesion or peace, that is sine qua non for a long run success, social justice should be achieved and maintained. Otherwise, social diversifications and upheavals, which result in social, economic, and politic chaos, are inevitable in the long term. Therefore, this study examines how income distribution is affected by the FDI flows and openness of a country, as proxies for globalization.
Considering previous studies, there is not any common view on the effects of globalization. The findings of time series analysis differ from panel data results. The inequality variable measurement becomes crucial, robustness becomes a major problem, and advanced econometric techniques produce challenging results. To illustrate, Figinia and Gorg (2006) employ an unbalanced panel of 103 countries and analyze the relationship between wage inequality and inward FDI via fixed effects panel technique. They find a nonlinear relationship for developing countries. Their empirical findings show that the coefficient of the relationship starts as positive but then turns out to be negative as the effect of inward FDI on wage inequality declines. However, Choi (2006), employing pooled GINI coefficient for 119 countries for the period of 1993-2002, finds that FDI flows, regardless of being inward or outward, lead to inequality. Therefore, even though these studies investigate a panel of countries, their results differ. Besides, country-specific versus panel studies may indicate controversial results. For instance; Tsai (1995) analyzes the relationship between FDI and income inequality and concludes that FDI inflows deteriorate the income equality in most of the miracle countries of East/Southeast Asia. On the other hand, Mah (2003) examines the impact of changes in trade values and FDI inflows on the GINI coefficient for Korea and find that neither FDI inflows nor trade openness has an influence on the GINI coefficient. Lastly, the choice of inequality variable (human capital, wages, income) may greatly affect the results as well as the interpretation of the main research question. To illustrate, Francois and Nelson (2003) argue that increased trade would seem to reduce wage inequality for the U.S. data from 1967-1997. Nevertheless, Lee (2006) employs data from the period 1951-1992 for 14 European countries by using income inequality rather than wage inequality and concludes that trade openness has no significant effect on income inequality.
With the motivation of the inconclusive researches on globalization, we employ FDI inflows, outflows, and trade openness (as proxies for globalization) in order to test the effects of globalization. The effects of globalization on income (in)equality are considered separately for developing, developed, and miracle countries because their different economic situations may exhibit various patterns overtime. Our results show that the increase in FDI inflows improves income equality (i.e., a lower GINI coefficient) in both developed and developing countries but deteriorates the income equality in miracle countries. In addition, we observe that FDI outflows affect the GINI coefficient negatively. We conclude that the components of globalization--FDI inflow, outflow and trade openness--have mixed effects on income inequality in different country groupings.
We contribute to the existing literature in two ways. First, we contribute to the existing debate on globalization by incorporating different country groups. Our results point important and significant differences among selected countries. Especially from the point of policy makers, our findings suggest to use openness and FDI flows in order to affect the income distribution based on the group of that country. Second, our panel data analysis considers the income inequality (i.e., GINI coefficient) as the dependent variable, rather than an endogenous variable. This new perception relies on the fact that some countries with subsistence level of income cannot undertake policies to change their income distribution immediately so that income inequality is the result, not an exogenous variable, for these countries.
The study is organized as follows: First, a brief literature survey is discussed. Then, we present the theoretical framework of the relationship between globalization and welfare. Afterwards, we introduce our methodology and empirical findings. Lastly, we conclude with some policy remarks.
There is an ongoing literature examining the relationship between globalization and inequality. Different studies consider various parameters (human capital, income, or wages) to examine the interactions with FDI flows and openness.
Human Capital, Wages, FDI Flows and Openness
Studies generally use human capital or wage as an inequality variable. Human capital refers to the level of "education, job training and health embodied in workers, which increase their productivity" (Salvatore 2004, p. 141). In essence, the education level shows the level of labor skill. Hence, the more skilled the labor is, the more wage the worker is paid. Theoretically, if the level of education is evenly distributed within a society, then the level of income will be relatively equally distributed."
Among others, one of the recent studies for developed countries is by Francois and Nelson (2003). Using U.S. data from 1967-1997, they argue that increased trade would seem to reduce wage inequality for a large economy. Taylor and Driffield (2005) use three-digit industry level data for U.K. manufacturing sectors for the period 1983-1992 and find that there is a link between relative wages. Moreover, they argue that FDI increases wage inequality but at a decreasing rate through time. Bjornstad and Skjerpen (2006) using 1972-1997 Norwegian data argue that trade openness reduces the number of employed with low education levels and increases wage inequality.
In a study for Mexico, a developing country, Feenstra and Hanson (1997) employ data for the period of 1975-1988 and find that the increase in FDI inflows increased the wages of the skilled labor, leading to wage inequality. Hence, they argue that FDI disfavors human capital equality. Gopinath and Chen (2003), employing 11 developing countries within the period of 1970-1992, find that FDI flows into developing countries widens the skilled-unskilled wage gap. Moreover, Zhu and Trefler (2004) attribute the positive effect of international trade on inequality in developing countries to the role of technological catch-up.
Growth and Income Inequality
Different than many other papers focusing on the relationship between the growth and income inequality, this paper considers only some proxies of globalization (FDI flows and trade openness) and income inequality, not growth. Nevertheless, studies investigating the link between growth and income inequality also prove to us the importance and validity of choosing the income inequality rather than human capital or wages.
Following the classification of Garcia-Penalosa and Turnovsky (2006), there are two lines of the empirical literature on the relationship between growth and income inequality (but both of the lines provide inconclusive findings). The first line considers growth and income inequality as interdependent processes, so studies, such as Lundberg and Squire (2003), aim to explore the factors inducing these processes. The second line tests the U-shaped relationship between these variables. Some of the papers find negative contribution of income inequality to growth (Alesina and Rodrik 1994; Persson and Tabellini 1994; Perotti 1996), whereas other studies indicate positive or ambiguous results (Li and Zou 1998; Forbes 2000; Barro 2000). Especially, Barro (2000) distinguishes developed countries from developing ones by reporting negative relationship for poorer countries and positive for richer countries. If for a specific country higher growth rates bring more FDI flows and/or more openness, then our results may be linked to Barro (2000).
Income Inequality, FDI Flows and Openness
In one of the early studies for miracle countries, Tsai (1995) analyzes the relationship between FDI and income inequality and concludes that only East/Southeast Asian countries have a positive relationship. In other words, FDI inflows deteriorate the income equality in most of the miracle countries. Mah (2003) using annual data for the period 1975-1995, examines the impact of changes in trade values and FDI inflows on the GINI coefficient for Korea and find that neither FDI inflows nor trade openness has an influence on the GINI coefficient. Zhang and Zhang (2003) argue that foreign trade and FDI inflows in China are important factors in explaining the worsening of the regional inequality using annual data for 1978-1998. Sato and Fukushige (2009) analyze the same data as Mah (2003). They find that FDI inflows deteriorate and trade openness improves income inequality. They attribute the positive effect of FDI inflows on the GINI coefficient to the case that openness in capital markets benefits only specific industries or firms. However, trade openness increases the chance of welfare for the whole society, so the effect is negative. In another study, Lee (2006) employs 1951-1992 data period for 14 European countries and concludes that FDI inflows increase income inequality but trade openness has no significant effect. Also, Meschi and Vivarelli (2009) employing 1980-1999 data for 65 developing countries find that trade with developed countries worsens income distribution.
Recently, many studies have concentrated on a large panel of countries, rather than focusing on certain classifications. Among others, Figinia and Gorg (2006), using an unbalanced panel of 103 countries and separating them as OECD and non-OECD countries for the period of 1980-2002, analyze the relationship between wage inequality and inward FDI via fixed effects panel technique. They find a nonlinear relationship for non-OECD countries (i.e., developing countries). Their empirical findings show that the coefficient of the relationship starts as positive but then turns out to be negative as the effect of inward FDI on wage inequality declines. For the OECD countries, however, they find a negative relationship. Choi (2006), employing pooled GINI coefficients for 119 countries for the period of 1993-2002, finds that FDI flows, regardless of being inward or outward, lead to income inequality. However, he uses regional dummies and considers all the countries in a panel, disregarding the country differences. Finally, Basu and Guariglia (2007) examine the interactions between FDI (inflows) and human capital inequality, growth, and the share of agriculture in gross domestic product (GDP), using annual data for 119 developing countries for the period 1970-1999. They find a positive relationship between FDI and human capital inequality and FDI and growth and a negative relationship between FDI and the share of agriculture in GDP in the recipient country.
In general, the FDI inflows decrease the GINI coefficient (i.e., income inequality becomes smaller) in developed countries but increase (i.e., income inequality worsen) in developing countries before 1990s, and have a positive effect (i.e., increase in GINI coefficient) in many miracle countries. Trade openness has a negative effect on the GINI coefficient (i.e., income inequality becomes smaller) in developed countries and a positive effect in both developing and miracle countries.
The Stolper and Samuelson (1941) theorem shows that openness in trade will benefit the abundant factor of a country. (3) Theoretically for developing countries, (4) the income of the unskilled labor force should rise as trade is liberalized, leading to a better level for income equality. This result should be reversed as income distribution deteriorates in skilled-labor abundant, developed countries with a wage increase. However, empirical findings usually contradict this nicely built theoretical framework. Moreover, advances in globalization lead to various new theories. For example, Borjas and Ramey (1994) base their analysis on imperfect competition (5) and argue that as competition increases in concentrated sectors due to trade openness, the wages of the labor will decline. If the labor in those concentrated sectors is unskilled, trade leads to deterioration in income distribution. On the other hand, Spilimbergo et al. (1999) argue that more liberal governments tend to implement more liberal trade policies and ignore re-distributional effects. Thus, trade openness affects income inequality in a positive manner, keeping the factor endowments constant. Besides, Francois and Nelson (2003) propose that trade liberalization may theoretically lead to improvement in income distribution through specialization. Hence, they argue that increasing division of labor causes increasing returns to scale (6) with labor having higher marginal productivity.
There are also several other theories or observations to explain the effect of globalization on income distribution. The mechanisms may differ from country to country or from country group to country group. In the following sections, we provide other theoretical framework specifics for country groups: developing, developed, and miracle countries.
Effects of FDI Inflows and Trade Openness on Income Inequality in Developing and Miracle Countries
FDI mainly outflows from developed countries and has several advantages for both the inflowing and out flowing country. Developed countries outsource to miracle countries-which have large populations and low wage labor-on the grounds that this will lower their costs of production. Depending on these inflows as their main source of growth, miracle countries try to attract more FDI through providing many incentives, such as prevention of labor unions or pressuring labor wages, creating special economic zones, decreasing taxes, etc. Therefore, FDI turns out to be income deteriorating in miracle countries. Recently, FDI flows from developed countries have been directed towards developing countries in Latin America and Eastern Europe, countries that have been less attractive. Benefiting from their competitiveness (or capability in undercutting prices), FDI flows led to significant market shares for large corporations of developed countries. Nevertheless, this helped to improve the income distribution (i.e., lower GINI coefficient) in developing countries with a rise in wages. On the other hand, an increase in openness to trade leads to an increase in the level of competition. Technologically developing countries are unable to compete with developed countries. Moreover, the abundant factor of production is less skilled labor. Thus, cost reductions are obtained through wages of the less skilled labor. So, trade openness by competition deteriorates income equality (i.e., higher GINI coefficient) for developing countries. This process may also be the case for miracle countries.
Effects of FDI Inflows, Outflows and Trade Openness on Income Inequality in Developed Countries
The effect of FDI inflows to developed countries, just as in developing countries, creates employment opportunities for less skilled labor. FDI inflows contribute to growth rates and tax revenues. In developed countries, the social state mechanism (i.e., government budget) functions very well. The safety nets, such as direct and indirect transfers and community support systems, are directed towards low income groups in order to increase their prosperity. Hence, the more the tax revenues are, the better the social state mechanism is. Thus, we can expect an improvement in income equality (i.e., the lower GINI coefficient) in the developed countries with higher levels of the closing of domestic firms, which usually relocate to developing countries. Accordingly, less skilled labor in developed countries lose their jobs and income distribution deteriorates. If the social state mechanism works well, then the unemployed arising from outsourcing can be kept in the work force. Some examples of safety nets that are used by the social state mechanism include job training, job creation and replacement, and microcredit. Moreover, job losses may stimulate less skilled labor to increase their skills (human capital) or become better qualified and, accordingly, be able to earn higher wages. Thus, there should be a positive indirect effect of FDI outflows on income equality. On the other hand, domestic firms have the opportunity to increase their market share through trade liberalization. High income groups (capital owners) could gain more from trade openness compared to lower income groups in developed countries. However, this might hamper the effectiveness of the social state policies. Therefore, the effects of a change in trade openness on income distribution are rather ambiguous in industrialized countries from a theoretical standpoint.
The methodology part includes four sections. First, we introduce the panel unit root tests. Second, we briefly explain the panel cointegration tests. Third, we introduce our panel coefficient estimation method. Finally, we discuss our empirical findings.
Panel Unit Root Tests
In order to apply a panel cointegration technique, nonstationary panel data has to be at the same integration order. The integration order of variables is determined by panel unit root tests. Panel unit root tests differ in allowance for cross sectional dependency. Cross sectional dependency can occur if there are one (or more) common casc(s) among the cross sectional data. (7) Panel unit root tests that do not allow for cross sectional dependency are referred to as first generation panel unit root tests. Accordingly, the tests that allow for cross sectional dependency arc second generation panel unit root tests. We use Im et al. (2003) test (known as IPS, a first generation test) and Pesaran's OPS (as a second generation test) to check the stationarity. (8)
Im et al. (2003) test is based on the conventional Augmented Dickey Fuller (ADF) unit root test (Dickey and Fuller 1979, 1981). The IPS test has the null hypothesis that a fraction of the series in the panel is non-stationary. On the other hand, to model cross-sectional dependency, Pesaran (2007) proposed a cross-sectional Augmented Dickey Fuller (CADF) test. This test has unit root in the null hypothesis. The individual CADF t-statistics for each cross sectional unit are averaged and thereby t-bar statistics are used in the panel unit root test.9 The test statistic is called cross sectional augmented IPS (CIPS) test. (10)
Panel Cointegration Tests
Among others, we employ Westerlund and Edgerton (2007) panel cointegration test (WE). The WE test is a panel bootstrap cointegration test with the null hypothesis of cointegration. The test is based on McCoskey and Kao (1998) Lagrange Multiplier test. (11) Westerlund and Edgerton (2007) use bootstrap property to handle the dependency between and within cross sectional units. Moreover, the test is shown to have local power in small samples.
Panel Coefficient Estimation (Fully Modified Ordinary Least Squares FM-OLS)
To examine the direction and magnitude of the proposed relationship, we employ the FM-OLS group mean estimator (Pedroni 2000). The group mean panel FM-OLS estimator provides a test of a common value for the cointegrating vector under the null hypothesis, while under the alternative hypothesis the cointegrating vector does not need to be common. The FM-OLS group mean estimator is the average value of the individual FM-OLS estimates.
Data and Empirical Findings
We have four variables: GINI coefficient, FDI inflows, FDI outflows, and trade openness. As a proxy for income inequality we use the GINI coefficient, which is equal to the area between the Lorenz curve and the 45 degree line. The Lorenz curve shows the distribution of income to households as percentages. The GINI coefficient lies between zero (0) and one (1). Zero shows perfect equality and one shows the perfect inequality. Hence, as the GINI coefficient goes to zero, the income is distributed more evenly and equally. The GINI coefficient variable for each nation is household and income based, covering the whole nation. FDI inflows and outflows are used as percentages of GDP. Openness is calculated as dividing the sum of imports and exports by GDP.
For developed countries, using Organization for Economic Cooperation and Development (OECD) statistics, we employ the countries with the highest FDI outflows and inflows for which the GINI coefficient data are available. These countries are France, Germany, Netherlands, UK, and U.S. For developing countries, using OECD and WDI (World Development Indicators) statistics, we employ the highest FDI inflowing developing countries which are economically similar and the GINI coefficient data are available. These countries are Argentina, Brazil, Czech Republic, Hungary, and Poland. For miracles, we employ countries for which GINI data are available. (12) These are China, India, Korea, Malaysia, Singapore, and Thailand. GINI coefficient data are employed from WIID2 (The UNU-WIDER World Income Inequality Database), US Census Bureau (for USA data), Eurostat and other official statistical departments. Trade openness data are obtained from WDI and OECD. For developed countries, we employ annual data for 1995-2007, for developing countries annual data for 1995-2006 and for miracles annual data for 1990-2005 and 1995-2005 periods.
First, we check the stationarity of the panel variables. The results of both first and second generation panel unit root tests are in Table 1. We find that all the variables have a unit root except FDI outflow in developed countries with IPS test and FDI inflow in developing countries with CIPS test when a constant in included. However, these variables have a unit root whenever a constant and a trend are included. Therefore, we conclude that all our variables have a unit root.
Table 1 Panel unit root tests for developing countries Tests IPS CIPS Variables Constant Constant Constant Constant and Trend and Trend Developing GINI -1.073 -2.108 -1.656 -1.666 FDI Inflows -2.048 -1.653 -1.435 -1.002 Openness -0.214 -1.971 -1.680 -1.255 Developed GINI 1.440 -1.671 -1.301 -1.925 FDI Inflows -1.515 -1.609 -1.470 -2.664 FDI Outflows -2.441 (a) -2.281 -2.136 -2.774 Openness -0.545 -1.835 -1.151 -1.195 Miracles GINI -1.015 -2.026 -1.590 -2.423 FDI Inflows -1.988 -2.682 -1.197 -2.464 Openness -0.041 -1.875 -1.972 -1.430 For developing countries, the critical values for the IPS test arc 2.266 and -2.636 for the constant case, -2.958 and -3.378 for the constant and trend case at 5% and 10% levels, respectively. For developed countries, the critical values for the IPS test are -2.252 and -2.612 for the constant case, -2.936 and -3.336 for the constant and trend case at 5% and 10% levels, respectively. For miracles, the critical values for the IPS test arc -2.210 and -2.540 for the constant case, -2.870 and -3.210 for the constant and trend case at 5% and 10% levels, respectively. For developing and developed countries, the critical values for Pcsaran's CIPS test are -2.460 and -2.430 for the constant ease; -3.134 and -3.066, for the constant and trend case at 5% and 1% levels, respectively. For miracles, the critical values for Pesaran's CIPS test arc -2.490 for the constant case and 3.202 for the constant and trend case at 5% and 1% levels, respectively. (a) denotes significance at 5% level. Max lag value is selected as 1
After ensuring the prerequisite for cointegration analysis, we apply Westerlund and Edgerton (2007) panel bootstrap cointegration test to check whether our variables are cointegrated. In Table 2, we have the p-values for the constant and constant and trend cases for the panel bootstrap cointegration test. Since the p-values are larger than 5%, we fail to reject the null of cointegration for developing, developed, and miracle countries. Thus, we find that there is a long run relationship between globalization and income inequality for all country groups.
Table 2 Westerlund and Edgerton (2007) panel bootstrap cointegration test Regressions Bootstrap P-value Constant Constant and Trend Developing 0.975 0.964 Developed 0.240 0.180 Miracles (1990 2005) 0.624 0.812 Miracles (1995-2005) 0.738 0.922 As sieve estimation, cither Yule-Walker or OLS is used to ensure invertibility. The null hypothesis is cointegration between the variables
In Table 3, FM-OLS estimations are summarized. For developing countries, the FDI inflows contribute to income equality (i.e., lower GINI coefficient) through the creation of new employment. However, the openness variable has a deteriorating effect on income equality (i.e., higher GINI coefficient) as the competition in the global arena entails a decline in wages of the less skilled labor.
Table 3 Panel fully modified ordinary least squares test GINI Panel Group FM-OLS Results FDI Inflows FDI Outflows Openness Developing -0.09 (a) (-30.63) - 0.02 (a) (-161.20) Developed -0.10 (a) (-17.96) -0.14 (a) (-25.68) 0.01 (a) (-50.28) Miracles 0.27 (a) (-5.58) - -0.10 (a) (47.13) (1990-2005) Miracles 0.15 (a) (-9.06) - -0.07 (a) (-87.63) (1995-2005) The values in the brackets are the t-statistics. N = 5, 7 = 11 for developing countries, N = 3, T = 13 for developed countries. N = 5, T = 16 for miracle countries for 1990- 2005 and N = 5, T = 11 for miracle countries for 1995 2005. (a) denotes significance at 5% level. Constant and trend case is used and maximum lag value is selected as 1
For developed countries, FDI inflows have a negative effect on the GINI coefficient. This finding supports the view that the creation of employment and the increase in the tax revenues enhances the strength of the social state contribution to income distribution. FDI outflows also have a negative effect on GINI coefficients. We relate this improvement in income distribution to the changes in the economic structure of the developed countries. If developed countries are reluctant to employ domestic, less skilled labor (and fulfill the needs for unskilled labor by means of developing or miracle countries), then the less skilled labor could be eager to become more qualified (and thus, be capable to earn higher wages compared to their former state). The trade openness is found to have a negative impact on GINI coefficients which can be attributed to the theoretical mechanism underlying the division of labor and marginal productivity.
Finally, FDI inflows into miracle countries affect GINI coefficient in a positive way. The result is not surprising because these countries repress the wages of less skilled labor in order to attract FDI. On the other hand, trade openness has a significant and negative effect on the GINI coefficent. We may attribute this negative effect to the increasing chance of more welfare for the whole society (Sato and Fukushige 2009).
Our empirical findings are in direct contrast with the theoretical frameworks introduced before. First, our results do not confirm Stolper-Samuelson (1941) theorem as we obtain a positive sign for the effect of trade liberalization on inequality for developing countries. Second, Francois and Nelson (2003) state that high division of labor should increase trade liberalization, leading to a rise in marginal productivity of labor and a better wage distribution. However, we contradict this mechanism by obtaining a positive sign for developed countries. Third, Borjas and Ramey (1994) theory is also not consistent with our findings because unskilled labor in East/Southeast Asian countries are concentrated in industries with high competition. In light of these outcomes, we could argue that the theoretical framework must be developed further in order to explain the empirical findings, which can differ frequently, even within decades.
This study tries to capture the effects of globalization on income inequality. We employ trade openness, FDI inflows, and FDI outflows as these variables have been major products of globalization. Indeed, the recent boom episode has underlined the significance of both FDI inflows and outflows. These variables are crucial because they constitute a huge portion of economic activity in different income level countries. Moreover, the immense rise in the level of trade through improving technology in the last two decades has confirmed trade openness as an important determinant of income per capita around the world.
Our estimation results show that the increase in FDI inflows improves income equality in both developed and developing countries, whereas we find that in miracle countries income distribution deteriorates. This is probably due to the different levels of economic development. Theoretically, FDI outflows are disruptive on income distribution owing to the fact that there is a threat of job loss for the less skilled labor. However, we observe that FDI outflows affect GINI coefficients negatively. We attribute this to the changing economic structure of world economies, the efforts of low income groups of countries to be oriented with this change, and the social state mechanism.
Overall, we observe that the components of globalization-that is FDI inflow, outflow, and trade openness-have rather mixed effects on income inequality in different country groupings. Thus, it is not possible to argue for a unique effect of globalization on GINI coefficients because the transmission mechanisms do not operate in the same way. This is possibly due to the fact that the economic and social infrastructures in these countries are quite different.
Acknowledgments We would like to thank Pmar Deniz for her excellent research assistance.
Alesina, A., & Rodrik, D. (1994). Distributive politics and economic growth. Quarterly Journal of Economics, 109, 465-490.
Barro, R. J. (2000). Inequality and growth in a panel of countries. Journal of Economic Growth, 5, 5 32.
Basu, P., & Guariglia, A. (2007). Foreign direct investment, inequality and growth. Journal of Macroeconomics, 29, 824 839.
Bjomstad, R., & Skjerpen, T. (2006). Trade and inequality in wages and unemployment. Economic Modelling. 23, 20 44.
Borjas, G., & Ramey, V. (1994). The relationship between wage inequality and international trade. In J. H. Bergstrand et al. (Eds.), The changing distribution of income in an open U.S. economy (pp. 217-241). Amsterdam: North-Holland.
Breitung, J., & Pesaran, M. H. (2008). Unit roots and cointegration in panels. In L. Matyas & P. Sevestre (Eds.), The econometrics of panel data: Fundamentals and recent developments in theory and practice (pp. 279-322). Dordrecht: Kluwer Academic Publishers.
Choi, C. (2006). Docs foreign direct investment affect domestic income inequality? Applied Economics Letters, 13, 811-814.
Dickey, D. A., & Fuller, W. A. (1979). Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association, 74, 427-431.
Dickey, D. A., & Fuller, W. A. (1981). Likelihood ratio tests for autoregressive time series with a unit root. Econometrica, 49, 1057-1072.
Feenstra, R. C., & Hanson, G. H. (1997). Foreign direct investment and relative wages: evidence from Mexico's Maquiladoras. Journal of International Economics, 42, 371-394.
Figinia, P., & Gorg, H. (2006). Does foreign direct investment affect wage inequality? An empirical investigation. Institute for International Integration Studies, HIS Discussion Paper No. 186. Retrieved from: http://email@example.com/attachmcnts/FlginiGoerg_IZA_2336.pdf (Accessed 10.05.2009).
Forbes, K. (2000). A reassessment of the relationship between inequality and growth. The American Economic Review, 90, 869-887.
Francois, J., & Nelson, D. (2003). Globalization and relative wages: Some theory and evidence. GEP Research Paper 03/15, University of Nottingham. Retrieved from: http://www.gep.org.uk/shared/shared_levpublications/ResearchPapers/2003/03 15.pdf (Accessed 06.09.2009).
Garcia-Penalosa, C., & Turnovsky, S. J. (2006). Growth and income inequality: a canonical model. Economic Theory, 28, 25-49.
Gopinath, M., & Chen, W. (2003). Foreign direct investment and wages: a cross-country analysis. Journal of International Trade and Economic Development, 12(3), 285 309.
Im, K. S., Pesaran, M. H., & Shin, Y. (2003). Testing for unit roots in heterogeneous panels. Journal of Econometrics, 115, 53-74.
Lee, J. E. (2006). Docs globalization matter to income distribution in Asia? Applied Economics Letters, 13, 851-855.
Li, H. Y., & Zou, H. F. (1998). Income inequality is not harmful to growth: theory and evidence. Review of Development Economics, 2, 318-334.
Lipsey, R. E. (2002). Home and host country effects of FDI. Paper for ISIT Conference on Challenges to Globalization, Lidingo, Sweden. Retrieved from: http://www.cepr.Org/meets/wkcn/2/2316/papers/lipsey.pdf (Accessed 14.05.2009).
Lundberg, M., & Squire, L. (2003). The simultaneous evolution of growth and inequality. The Economic Journal, 113(2003), 326-344.
Mah, J. S. (2003). A note on globalization and income distribution: the case of Korea, 1975-1995. Journal of Asian Economics, 14, 157-164.
Marshall, A. (1890). Principles of economics. London: Macmillan Publishers.
McCoskey, S., & Kao, C. (1998). A residual-based test of the null of cointegration in panel data. Econometric Reviews, 17(1), 57-84.
Meschi, E., & Vivarelli, M. (2009). Trade and income inequality in developing countries. World Development, 37(2), 287-302.
Pedroni, P. (2000). Fully modified OLS for heterogeneous cointegrated panels. In B. Baltagi (Ed.), Nonstationary panels, panel cointegration, and dynamic panels, advances in econometrics, vol 15 (pp. 93-130). Amsterdam: JAI Press.
Perotti, R. (1996). Growth, income distribution, and democracy: what the data say. Journal of Economic Growth, 1, 149-187.
Persson, T., & Tabellini, G. (1994). Is inequality harmful for growth? The American Economic Review, 84, 600-621.
Pesaran, M. H. (2007). A simple panel unit root test in the presence of cross-section dependence. Journal of Applied Econometrics, 22, 265-312.
Salvatore, D. (2004). International economics (8th ed.). England: Wiley.
Sato, S., & Fukushige, M. (2009). Globalization and economic inequality in the short and long run: the case of South Korea 1975-1995. Journal of Asian Economics. 20, 62-68.
Spilimbergo, A., Londono, J. L., & Szekely, M. (1999). Income distribution, factor endowments, and trade openness. Journal of Development Economics, 59, 11-101.
Stolper, W., & Samuelson, P. (1941). Protection and real wages. Review of Economic Studies, 9, 58-73.
Taylor, K., & Driffield, N. (2005). Wage inequality and the role of multinationals: evidence from UK panel data. Labor Economics, 12, 223-249.
Tsai, P.-L. (1995). Foreign direct investment and income inequality: further evidence. World Development, 23, 469-483.
Westerlund, J., & Edgerton, D. L. (2007). A panel bootstrap cointegration test. Economics Letters, 97, 185 190.
Zhang, X., & Zhang, K. H. (2003). How docs globalization affect regional inequality within a developing country? Evidence from China. Journal of Development Studies, 39, 47-67.
Zhu, S. C, & Trefler, D. (2004). Trade and inequality in developing countries: a general equilibrium analysis. Journal of International Economics, 65(1), 21-48.
(1) There is fierce competition for more FDI through creating lower tax possibilities-in Special Economic Zones like China-and preventing labor unions to sustain the low-cost labor incentives.
(2) Thus, we also survey studies that analyze the relationship between human capital inequality and globalization.
(3) Under the Heckscher-Ohlin Theorem, nations tend to export goods whose production requires relatively abundant and cheap factors.
(4) It is well known that labor is less skilled in developing countries.
(5) Contrary to the Hcckscher-Ohlin Theorem.
(6) Francois and Nelson (2003) refer to Marshall (1890, Book 4, Ch.13, p.2) as, "... while the part which nature plays in production shows a tendency to diminishing return, the part which man plays shows a tendency to increasing return."
(7) These common casc(s) arc referred as "common factors" in the related literature such as Pesaran (2007).
(8) There are several first generation and second generation panel unit root tests. However, (8) we prefer to use the ones that have been very common in the recent literature. Breitung and Pesaran (2008) is an excellent survey of nonstationary panel data analysis.
(9) Pesaran (2007) used a different technique to test for the stationarity of the panel data. Instead of basing the test on the deviations from the common factors, he calculated the ADF regression with the cross sectional averages of lagged levels and first-differences of the individual series.
(10) Pesaran (2007) also proposed a truncated version of the test in order to avoid undue effects caused by small sample size.
(11) McCoskey and Kao (1998) test is shown to work poorly in small sample cases and docs not allow to for cross sectional dependency.
(12) Since the amount is negligible, we do not employ FDI outflow data for developing or miracle countries.
Department of Economics. Marmara University. Goztepe Kampusu, Kuyubasi, Kadikoy.
Istanbul 34722, Turkey
Department of Business Administration. Middle East Technical University, Sogutozu, Ankara 06531, Turkey
DOI 10.1007/s 11294-010-9281-0…