Printer Friendly

Did income inequality benefit or hinder economic growth in Europe?


Inequality does not necessarily mean poverty. In fact, inequality may increase even when an economy performs well or improves the situation of disadvantaged people. There are two main perspectives. The first is aimed at reducing social exclusion, without increasing the gap between social classes. The second is led by the Organization for Economic Cooperation and Development (OECD, 2015), stating that inequality promotes a negative and statistically significant effect on growth which, in turn, creates inefficiencies in the distribution of factors of production (i.e. land, labor, working capital, financial capital, technological progress). Throughout the long period of strong Keynesian influence (1945-1972), issues relating to inequality were not specifically studied. The dominant line of thinking assumed that the higher the growth, the fewer the problems relating to development, which included inequality. However, with the arrival of the new globalization, things have changed (Dewhurst and Mutis-Gaitan, 1995). Throughout the nineties, several economists researched a wide range of areas, such as the nature of GDP per capita convergence and the distribution of income (in terms of purchasing power parity), analyzing what appeared to be the major factors of economic growth and of the convergence of economies, for example, when the poorer economies, in terms of per capita incomes, tend to grow at faster rates than richer economies (Berumen and Perez-Megino, 2015).

One of the initial theories claimed that the distribution of income implicitly involves problems of inequality or, in other words, that salary differences make the rich richer and the poor poorer. In addition, the unequal distribution of wealth has been the result of the inequality of income over time, which would explain the role played by income inequality in the process of economic growth and why different countries (or possibly the same country at different moments in time) grow at different rates. The answer to the questions of why salary differences make the rich richer and the poor poorer, and what role is played by income inequality can be found firstly in certain Asian countries, such as South Korea (Lee et al., 2012) and Singapore (Tat Hui and Toh, 2014).

Southeast Asian economies experienced high economic growth from 1975 to 2009, which resulted in a substantial reduction in income inequality (World Bank, 1993; World Bank and IMF, 2015). On the other hand, during the same period the majority of Latin American economies had serious problems with income inequality and suffered severe monetary and financial crises in the nineteen eighties and nineties. If these two groups of countries are taken as the only references, it could be considered that there is a negative relationship between income inequality and economic growth. Nevertheless, there have several highly developed countries in which this relationship has not been the case. If we look at the United States and Sweden, we could reach the conclusion that there is a positive relationship between the two variables, given that the United States grew faster than Sweden, in spite of the higher inequality of income (Perez-Megino and Berumen, 2015). A final conclusion on the underlying causation between both economic variables is therefore not possible and, in fact, absolutely opposite conclusions could be reached. There can be positive or negative relationships, as shown in specialized literature and summarized in Table 1.

Barro (2000) highlighted the dynamic nature that exists between both variables. Table 2 shows how the inequality of income affects growth and how it also depends on the degree of a country's economic development. By analyzing panel data, Barro (Idem, page 32) stated that income inequality delays economic growth in developing countries (e.g. with a GDP per capita lower than 2,070 US dollars), but promotes it in developed countries (e.g. with a GDP per capita in excess of 2,070 USD).

Piketty and Saez (2001) embarked on a revolutionary line of research aimed at studying inequality of both income and wealth and took on the task of building an original database which, in itself, is a major contribution to the study of inequality, also completely in line with the definitions and units of measurement recommended by the United Nations.

Based on the above-mentioned elements, this study is aimed at responding to the question: does income inequality benefit or hinder economic growth? The following sections review the literature relating to the purpose of the study (section 2); explain the database and methodology used for the analysis (section 3); specify the econometrics model and its features (section 4); present the results obtained (section 5); and, finally, section 6 discusses the results and provides conclusions.

Theoretical Framework

Globalization and its Consequences

The dominant model of globalization in the terms described in the Washington Consensus (Williamson, 1990) has generated a justified debate on its pros and cons. Of all the questionnaires that could be discussed, two stand out for their importance: has the success of globalization been at the expense of greater inequality? And, if so, how has inequality affected economic growth in a particular country or region?

The most critical authors claim that the current economic globalization model has increased inequality between countries and in regions of the same countries (e.g., Firebaugh, 2003; Wade, 2004). At the other extreme are those who believe that thanks to globalization, millions of families have overcome poverty thresholds and inequality has decreased (e.g., Dollar and Kraay, 2002). The basic issue is: can both critics and supporters be right, or partly right? The answer is yes. Such contradictory conclusions can, in fact, be related to different methodologies (Mills, 2009). Indeed, depending on the analysis and sample used, conclusions can be reached like those of Milanovic (2005), in line with Dollar and Kraay (2002), or those of Sala-i-Martin (2006) who, after studying 138 countries from 1979 to 2000, found that inequality between countries had decreased considerably. On the other hand, Piketty and Saez (2006), Grabka and Westermeir (2014), amongst others, claim that inequality has increased in recent decades and, particularly, in different regions of the same country.

In the past, it was claimed that the role of government was crucial for both economic growth and globalization, as well as in the fight against inequality. A government may opt for globalization and therefore international trade, or adopt a self-sufficiency policy and obviously, in both cases, there will be consequences. Similarly, a government can choose to fight inequality and assume the costs, however high, or promote economic growth, even if it implies a high price to pay with respect to inequality (Deaton, 2015: 23). In any case, the majority of countries have opted for globalization, which began in the eighties under the framework of the mandatory provisions of the so-called Washington Consensus. One of the most evident consequences has been the global mobility of labor resulting from the demand for workers. For more than three decades, the number of workers willing to compete for jobs at both a national and international level has increased, which has allowed companies to enjoy a greater range of possibilities, as they can now hire workers with different skills at different costs, or resort to well-known processes of displacement and outsourcing, amongst other benefits to company competitiveness, production centres, regions, countries and groups of countries (e.g., Berumen, 2005; 2006; Berumen and Llamazares Redondo, 2007). This has not only meant a radical change in the way of doing business, but also in the distribution of income. To the extent that labor costs are cheaper in less developed countries, companies have decided to move their production facilities. The increase in highly qualified labor in developed countries and in the demand for unqualified workers in developing countries is therefore not surprising. Obviously, this trend in global demand for labor increases inequality, mainly but not only due to the differences in the distribution of income.

In recent years there has been an increase in demand for labor in developing countries and, as a result, an increase in the salaries paid for less qualified work and a reduction in demand for low qualified workers. However, in developed countries, the situation is quite different, given that the increase in demand for labor has increased the salaries of highly qualified workers. Accordingly, in the case of workers with a certain degree of qualification, globalization has reduce the inequality gap between countries; however, at the same time, it has increased the gap between regions of the same countries.

Growth of the Principal Indicators of Inequality

According to Piketty and Zucman (2014), over the last four decades, the wealth-income ratio in developed countries has grown constantly: from 200-300 percent in 1970 to 400-600 percent in 2010. The fact that the [beta] ratio, as defined using the Harrod-Domar-Solow formula shows growth during this time implies that the role played by wealth in society (and more specifically of inherited wealth) has been strengthened to the extent that the mechanisms that gave rise to the inequality rates in the last century could even be repeated. This has been precisely why authors such as Piketty (2011), Stiglitz (2012), Piketty and Saez (2013) have highlighted the need to design progressive taxation systems. From 1914 to 1945, several developed countries experimented with a significant decrease in the percentage of income and wealth concentrated in the upper social classes, especially in the years of the Great Depression (1929-1933). This gave rise to a substantial decrease in inequality in these economies large number of countries adopted and implemented Welfare Systems. However, over the last thirty years, the percentage of income concentrated in the upper social classes has increased consecutively (as shown in Table 3 [see following page], in the top 1.5 percent and top 10 percent). Accordingly, the absence of more efficient progressive taxation has resulted in the accumulation of income in fewer hands. Inherited income has played a significant role in society, as opposed to systems based on merit and personal effort, which is the cornerstone of capitalism (Stiglitz, 2015: 88).

As shown in Table 3, the speed at which these rates have increased varies in each case. The countries in which growth has been the highest are the United Kingdom, Ireland and Germany. In fact, the United Kingdom introduced strict neoliberal policies under the rule of Margaret Thatcher during the nineteen eighties and Ireland has the most lenient and less progressive taxation system in the entire European Union. The Gini Index is a commonly used indicator in specialized literature; however, in this study it was not particularly useful, as complete series for long periods of time were not available. In fact, in this research, it was not possible to gather data for the Gini Index from the World Bank, OECD and Eurostat databases from 1975 to 2009, which highlighted the need to continue improving the statistical systems available to researchers.

Justification of the Variables

Variables were chosen in three large groups: (i) measures of income, such as GDP per capita in US dollars or GDP per capita adjusted to Purchasing Power Parity; (ii) measures of inequality, such as the Pareto-Lorenz Coefficient, the percentage of income concentrated in the upper part of the population, around 1.5 and 10 percent; and (iii) control variables, in other words, variables that can positively or negatively affect regression, such as: the inflation rate, years of schooling, the extent to which a country is open to overseas trade, net rate of immigration, the real interest rate, final household consumption expressed as a percentage of GDP and the unemployment rate, amongst others. As indicators of income, we chose the GNP growth rate per capita and the GNP per capita in current US dollars. As measures of inequality, we chose the percentage of income concentrated in the upper part of the population of 1.5 percent and 10 percent and the Pareto-Lorenz Coefficient. As claimed by Wade (2004), when research is interested in studying issues related to the theory of why countries grow, they should be treated with the same relative importance (in other words, given the same weighting, irrespective of the population). This is how to overcome certain methodology problems that arise from weighting based on population, especially bias. If, for example, countries like China and India were included in the study and weighted according to population, the results would be different from those of a similarly study that ignores population.

Banerjee and Duflo (2003), Wade (2004) and Voitchovsky (2005) also recommend including certain control variables, such as:

* The extent to which a country is open to foreign trade, to determine whether more open countries have higher growth rates, as claimed by the World Bank (2002).

* Migration levels, given that the fastest way a poor person can improve his/her situation is to move to a wealthier country. A country with a self-sufficient trade regime, but with a high level of migration is, to all effects, a globalizing country, even if the country does not include itself in this category (Wade, 2003; 2004).

* Economic growth is inversely related to the inflation rate which, in the study, was included as an indicator of macroeconomic stability.

* Economic growth is directly related to the level of education of human capital. The number of years of compulsory schooling was included.

* Finally, end consumption in each economy was included as a percentage of GDP, a variable closely related to public spending and private consumption (both included) as well as the level of savings.

Data and Methodology

A variety of sources were consulted with the aim of including all the variables referred to in the specialized literature. To construct the complete time series for the period analyzed, we used the most recent and well-known databases, namely: the World Bank Database, OECD Statistics and the World Top Incomes Database. The World Bank database was used for the following variables: GDP growth percentage, GDP per capita in current US dollars, final household consumption as a percentage of GDP, final consumption in the economy as a percentage of GDP, inflation rate (measured according to the GDP deflator) and the percentage of GDP represented by trade. The World Top Incomes Database provided the historical series of data relating to the percentage of income concentrated in the upper part of the population, 1.5 percent and 10 percent. Finally, we consulted OECD statistics for the number of years of compulsory secondary education and the schooling rate. In all, the sample contained 442 observations (n=442) for the period ranging from 1975 to 2009.

With regard to methodology, several works acknowledge the difficulties faced by researchers in finding suitable variables to correctly analyze a sample. After this problem is resolved, the method of analyzing the database constructed must be chosen. For this study, we opted for multiple linear regression models which, as a generalisation of simple regression models, have more than one explanatory variable, which allows the study of the relation between a dependent variable and a number (to be determined) of independent variables (Maddala, 1992). This choice is justified by the fact that these models are a useful technique for modeling a large amount of phenomena (Marill, 2004). When the set of data fulfils the required theoretical assumptions, these techniques offer a model that, in general, can be accurately resolved using the estimates of the explanatory variable coefficients. The model used took into consideration observations by Barro (2000), Forbes (2000) and Bond et al. (2001) and lags (delayed values) were introduced to correctly model the delayed effect of the difference in income on inequality. The sample of data was thus reduced to a total of 237 observations. The model is structured as follows:

[] - [y.sub.i,t-1] = [alpha] y[p.sub.i,t-1] + [beta][] + [[mu]] (1)

where t and t-1 are observations separated by a difference of 7 years. i denotes the country in question. [] is the annual GDP growth rate PIB per capita and [] is the Napierian Logarithm of the GDP per capita. The [] vector may contain current or delayed values of several explanatory variables, including control variables: (i) certain measures of inequality, accounted for in t-1; (ii) the rate of schooling as an indicator of human capital; (iii) the degree to which an economy is open to foreign trade, as an indicator of the acceptance of globalization; (iv) final consumption in the economy as a percentage of PIB to show the aggregate consumption of households and the public sector; and (v) the inflation rate as a measure of macroeconomic stability. Finally, [] refers to error, which is normally included in this type of model.

Although specialized literature normally calls them growth models, the type of neoclassic model they are based on explains the level of the stationary state of income in the long term. In this context, any alteration of an explanatory variable will displace the long-term level of income (stationary state) and affect the growth rate only during the convergence with the new balance (1). As such, a permanent change in income inequality will influence the short-term growth rate, which is conditioned by the level of income, but will not have permanent consequences on the growth rate once the new stationary state is reached. Nevertheless, as pointed out by Barro (2000), given that it can take a long time to reach a new balance, the short-term effects on growth can be long-lasting and significant. According to theoretical recommendations, given that a change in income inequality will probably take from 5 to 10 years to become evident in an economy (Barro, 2000; Banerjee and Duflo, 2003; Voitchovsky, 2005) and due to the limitations of the database created, it was decided to use lags of 7 years because, as mentioned above, it significantly reduced the number of observations.

Explanatory Model and Technique

The following equation considers the first difference in GDP growth rates per capita as an endogenous variable and explains the econometric model chosen to respond to the question raised in the introduction.



[y.sub.i,t] - [y.sub.i,t-1] represent the difference between the GDP per capita annual growth rates and where there are 7 years' difference between t and t-1. Accordingly, i represents the country in question.

[Log_PIBpc..sub.i,t-1] is the GDP per capita logarithm for the period t-1.

[Cons.sub.i,t-1] represents the final aggregate household and public sector consumption as a percentage of GDP, in the period t-1.

[Top1.sub.i,t-1] expresses the percentage of income concentrated in the top 1 percent of the population in the period t-1.

[Top10.sub.i,t-1] expresses the percentage of income concentrated in the top 10 percent of the population in the period t-1.

[Apert.sub.i,t] represents the degree to which the economy is open to foreign trade, measured as a percentage of GDP in the period t.

[Escol.sub.i,t] is the rate of schooling in the period t.

[Infl.sub.i,t] is the domestic economy inflation rate in t.

a, also known as the model constant, is the value of the dependent variable when all the independent variables are zero.

[a.sub.i] is the independent variable coefficient.

[] is error.

The estimation technique used was that of Ordinary Least Squares (OLS). This estimation technique is useful provided the 10 basic assumptions described by Gujarati (2004) are fulfilled and the considerations pointed out by Maddala (1992) on the model error are taken into account. It was therefore possible to obtain unbiased estimators and, with a minimum variance, calculate the confidence ranges and perform the significance test.


The following equation shows the estimated model:

[y.sub.i,t] - [y.sub.i,t-1] = -41.685 + 4.536Log_PIBpc + 0.342Cons - 0.451Top1 + 0.15Top10 + 0.024Apert + 0.005Escol - 0.08Infl (3)

Table 4 shows the model estimates using the OLS technique. In effect, it was estimated that, on average, for each unit of increase in the GDP per capita logarithm and with all other variables constant, the annual economic growth rate increased by 4.536 percentage units. It should be pointed out that for the Napierian logarithm to increase 1 unit, the magnitude of the GDP per capita increase in absolute terms must be exceptional. As expected, the sign of the regression coefficient implies a positive correlation between GDP per capita and economic growth. In addition, it was estimated that, on average, for each percentage unit increase in consumption as a percentage of GDP and all other variables remaining constant, the annual economic growth rate increased 0.342 percentage units. When the percentage of income concentrated in the top 1 percent of the population increases by 1 unit and with all other variables constant, the economic growth rate decreases by 0.451 percentage units. On the contrary, if the percentage of income concentrated in 10 percent of the population increases by 1 unit, economic growth increases at a rate of 0.15 percent.

If foreign trade is granted more importance in the economy (in other words, increases its relevance with regard to GDP by one per cent and with all other variables remaining constant), economic growth increases at a rate of 0.024 percent. The next control measure is that of schooling: when schooling increases by 1 unit, it is expected on average that the economic growth rate will increase by 0.005 percent. Finally, if the inflation rate increases by 1 per cent, it is estimated on average that the economic growth rate will decrease by 0.08 percent (Table 4). All the variables included in the model, with the exception of the measure of macroeconomic stability (inflation), are significant. When a variable is not significant with respect to the t statistic, it can be eliminated from the model, without a risk of specification error by excluding a relevant variable. The low level of significance of the inflation variable can also be checked by interpreting the confidence intervals calculated at 95 percent, also shown in Table 4, which indicate how the variable could be zero and have a zero effect on economic growth.

The reliability of the model (Table 5) is estimated at 20.30 percent, which means that the variables considered explain 20.30 percent of the variations in economic growth, whereas the remaining 79.70 percent depend on other variables (such as technological stock, quality and quantity of available human capital and current legislation, amongst others). In addition, Table 6 performs the variance analysis (ANOVA), showing the degrees of freedom, size of the sample and overall meaning of the model using the F statistic and p-value. The analysis is, in fact, statistically significant for the sample ranging from 1975 to 2009. The multicollinearity analysis, which determines whether there is no linear relationship between the explanatory variables, is shown in Tables 4 and 5. The model is free of multicollinearity, because it does not show excessively high values for R Squared and the values of the t statistic for each variable are significant, in addition to the fact that no high correlations are observed between the independent variables (Tables 6 and 7). Similarly, when the regressors are individually analyzed, no multicollinearity is observed: both the VIF (Variance Inflating Factor) and TOL (Tolerance Factor) show suitable values.

The fact that multicollinearity is not present in the analysis could be due, amongst other factors, to: (i) the correct study of specialized literature providing initial information; (ii) the use of time series and transversal sections, as well as the correct processing of data; (iii) the elimination of variables that were not significant, such as inflation, which made it possible to prove that there were no specification errors; (iv) the work of the variables in the right units, as in the cases of GDP per capita and the first difference in the economic growth rate; and (v) the use of a suitable sized sample, which avoided the problem of micronumerosity. To ensure that that error was distributed normally, a histogram of residuals is provided (Graph 2) showing that average error is practically zero (-8.63E-15) and standard deviation close to one (0.985), thus confirming unbiased estimates.


Finally, one of the fundamental hypotheses of regression models is that random shock or error must have a constant variance in all observations. If this hypothesis is not fulfilled, the OLS estimator is no longer efficient, meaning that it is preferable to use other methods of estimation, such as weighted least squares, for example. To check for heterocedasticity, a range of techniques are also available. Graph 3 shows the residuals, indicating that there are no patterns of any specific behavior (examining residuals is a key part of all statistical modeling, because it can explain whether the assumptions are reasonable and the choice of model has been appropriate). Furthermore, Graph 4 shows the regression adjustment line. Finally, the White test was used to conclude that heterocedasticity is not present in the model, given that a p-value substantially higher than 0.05 is obtained.



Discussion of Results and Conclusions

The results reached show that in the countries analyzed, non-extreme inequality can encourage the process of economic growth and, on the contrary, that the higher the inequality, the lower the economic growth. Using regression coefficients for the top 1 percent and 10 percent, it can be observed that the lower the concentration of wealth, the higher the economic growth rates. It should be noted that the coefficient for the top 1 percent is -0.451 and 0.15 for the top 10 percent; in other words, whereas the concentration of income in the top 1 percent negatively affects economic growth, the concentration of income in the top 10 percent positively affects growth. This could possibly be explained by the fact that this percentage of society is more likely to reinvest and is therefore more willing to put cash back into economic circulation. Indeed, the results confirm the conclusions reached by Barro (2000), although with certain qualifications. Barro (2000) states that income inequality delays growth in developing countries (GDP per capita lower than 2,070 US dollars), whereas it encourages growth in developed countries (GDP per capita above 2,070 US dollars). This research, which only focused on developed countries, shows that extreme inequality, measured as wealth concentrated in the top 1 percent of the population, negatively affects economic growth, whereas a certain degree of inequality in lower levels of the distribution (top 10 percent of the population) tends to encourage economic growth. Secondly, as claimed in conventional theory, the more qualified the human capital is, the higher the productivity and, therefore, the stronger the elements that lead to economic growth. Finally, it can be said that the degree to which a country is open to foreign trade, or the acceptance of globalization, positively influences economic growth. Some of the characteristic variables of current globalization, such as intense changes in demand for labor, have heavily influenced inequality amongst countries (in line with the studies carried out by Firebaugh, 2003; Wade, 2004; Piketty and Saez, 2006; OECD, 2008), to the extent that the Kuznets Curve (1955) tends to take on a U-shape, instead of a bell shape (or inverted U), as claimed at the time by Simon Kuznets.

Piketty and Zucman (2014) took on the task of analyzing why the wealth-income ratio increased from 200-300 percent in the nineteen seventies to its current level of 400-600 percent (according to the Pareto-Lorenz coefficient, the results can be confirmed). This research has confirmed the vital role played by wealth, and therefore inherited wealth, in modern day society and how it has been effectively strengthened. Growing inequality in income and wealth can lead to poorer classes resorting to criminal activity, revolts and mass disturbances (Hibbs, 1973; Venieris and Gupta, 1986; Gupta, 1990). The use of violence in society can become a threat to the stability of political institutions and even private property (Barro, 2000). In fact, such unrest can also become a part of society as a result of high levels of inequality which, in the end, can also affect economic productivity and therefore undermine economic growth (Stiglitz, 2003). It is not surprising that to deal with these social problems, authors such as Piketty, Saez, Zucman et al make constant reference to the need to design progressive taxation systems which, in their opinion, are one of the political tools that least distort the economy and, at the same time, are an element able to control the degree of inequality that is acceptable in society, by distributing the profits of economic growth.

With the aim of furthering understanding of the process and the effect of inequality on economic growth, this study carried out an empirical analysis of 12 European countries, all of them with developed economies, from 1975 to 2009. The results of the analysis are significant. In short: (i) there are different reasons why countries grow at different rates, which highlight inequality; (ii) inequality, when not at an extreme level, has promoted economic growth in developed countries; and (iii) one of the most emblematic elements of the capitalist system is meritocracy and it appears that the greater the inequality, the more difficult it is for poorer people to improve their situation.

Irrespective of the results obtained, the debate on inequality is still open. In fact, these conclusions can in no way be extrapolated to emerging and developing economies and even less so to lagging economies, because the majority cannot guarantee sufficient or reliable data. The categories and indicators assumed to be unquestionable in more developed economies are not necessarily applicable to others, given that a high percentage of their productive force exists in the vast unknown of the underground economy, outside the control of official statistics. Piketty (2014) proposes the development of progressive taxation systems, expressly created to reduce the role played by wealth in society (and more specifically that of inherited wealth). This would require contrasting a series of proposals using the well-known Laffer Curve and will be the subject of future research.


Acemoglu, D. (1998). "Why Do New Technologies Complement Skills? Directed Technical Change and Wage Inequality." Quarterly Journal of Economics, 113 (4), 1055-1089.

Aghion, P. and Howitt, P. (1998). Endogenous Growth Theory. MIT Press: Cambridge, MA.

Alesina, A. and Perotti, R. (1996). "Income Distribution, Political Instability and Investment." European Economic Review, 81 (5), 1170-1189.

Alesina, A. and Rodrik, D. (1994). "Distribution Politics and Economic Growth." Quarterly Journal of Economics, 109, 465-490.

Amos, O. M. (1988). "Unbalanced Regional Growth and Regional Income Inequality in the Latter Stages of Development." Regional Science and Urban Economics, 18 (4), 549-566.

Banerjee, A. and Duflo, E. (2003). "Inequality and Growth: What can the data say?." Journal of Economic Growth, 8 (3), 267-299.

Barro, R.J. (2000). "Inequality and Growth in a Panel of Countries." Journal of Economic Growth, 5 (1), 5-32.

Benabou, R. (1996). "Inequality and Growth." NBER Macroeconomics Annual, 11-73.

Berumen, S.A. (2005). "An Approach to Local and Regional Competitiveness." Cuadernos de Administracion, 18 (29), 13-32.

Berumen, S.A. (2006). "Una Aproximacion a los Indicadores de la Competitividad Local y Factores de la Production." Cuadernos de Administracion, 19 (31), 145-163.

Berumen, S.A. and Llamazares Redondo, F. (2007). "La Utilidad de los Metodos de Decision Multicriterio (como el AHP) en un Entorno de Competitividad Creciente." Cuadernos de Administracion, 20 (34), 65-87.

Berumen, S.A. and Perez-Megino, L.P. (2015). "El papel de la desigualdad de ingresos en el proceso de crecimiento en Europa." Discussion Paper WP-09/15, Instituto Universitario de Analisis Economico y Social, Universidad de Alcala.

Bond, S.; Hoeffler, A. and Temple, J. (2001). "GMM Estimation of Empirical Growth Models." Discussion Paper No. 01/525, University of Bristol.

Deaton, A. (2015). The Great Escape: Health, Wealth, and the Origins of Inequality. New Jersey: Princeton University Press.

Dewhurst, J. and Mutis-Gaitan, H. (1995). "Varying speeds of regional GDP per capita convergence in the European Union, 1981-91", in H. Armstrong and R.W. Vickerman (Eds.), Convergence and Divergence among European Regions. Pion Limited: London, 23-39.

Dollar, D. and Kraay, A. (2002). "Spreading the Wealth." Foreign Affairs, 81, 120-133.

Domar, E. (1947). "Expansion and Employment." American Economic Review, 37, 34-55.

Firebaugh, G. (2003). The New Geography of Global Income Inequality. Cambridge, MA: Harvard University Press.

Forbes, K. (2000). "A Reassessment of Relationship between Inequality and Growth." American Economic Review, 90 (4), 869-886.

Grabka, M.M. and Westermeier, C. (2014). "Persistently High Wealth Inequality in Germany." DIWEconomic Bulletin, 6, 3-16.

Gujarati, D.N. (2004). Basic Econometrics (4th Edition). McGraw Hill: New York.

Gupta, D. (1990). The Economics of Political Violence. Praeger: New York.

Harrod, H. (1939). "An Essay in Dynamic Theory." Economic Journal, 49 (193), 14-33.

Helpman, E. (2004). The Mystery of Economic Growth. Belknap: Harvard.

Hibbs, D. (1973). Mass Political Violence: A Cross-Sectional Analysis. Wiley: New York.

Kuznets, S. (1955). "Economic Growth and Economic Inequality." American Economic Review, 45 (1), 1-28.

Lee, Hyun-Hoon; Lee, Minsoo and Park, Donghyun (2012). "Growth Policy and Inequality in Developing Asia: Lesson from Korea", Working Papers DP-2012-12, Economic Research Institute for ASEAN and East Asia (ERIA).

Maddala, G. S. (1992). Introduction to Econometrics (2th Edition). McMillan: New York.

Marill, K.A. (2004). Advanced Statistics: Linear Regression-part II: Multiple Linear Regression. Academic Emergency Medicine, 11 (1), 94-102.

Milanovic, B. (2005). Worlds Apart: Measuring International and Global Inequality. Princeton, New Jersey: Princeton University Press.

Mills, M. (2009). "Globalization and Inequality." European Sociological Review, 25 (1), 1-8.

Murphy, K.M.; Shleifer, A. and Vishny, R.W. (1989). "Income Distribution, Market Size, and Industrialization." Quarterly Journal of Economics, 104 (3), 537-564.

OECD (2008). Growing Unequal? Income Distribution and Poverty in the OECD Countries. OECD: Paris. ISBN: 978-92-64-04419-7.

OECD (2015). In It Together. Why Less Inequality Benefits All. OECD: Paris. ISBN: 978-92-64-23266-2. Available in: http://www.keepeek. com/Digital-Asset-Management/oecd/employment/in-it-together-why-less-inequality-benefits-all_9789264235120-en#page1

Okun, A.M. (1975). Equality and Efficiency: The Big Trade-Off. Brookings Institution: Washington, DC.

Panizza, U. (2002). "Income Inequality and Economic Growth: Evidence from American Data." Journal of Economic Growth, 7, 25-41.

Pareto, V. (1896). Cours d'Economie Politique. Droz: Geneve.

Perez-Megino, L.P. and Berumen, S.A. (2015). "Globalization and Inequalities in Developed Economies: Kuznets was not Right." Neumann Business Review, 1 (2), 23-36.

Piketty, T. (2011). "On the Long-Run Evolution of Inheritance: France 1820-2050." Quarterly Journal of Economics, 126 (3), 1071-1131.

Piketty, T. (2014). Capital in the Twenty First Century, The Belknap Press: Cambridge Mass.

Piketty, T. and Saez, E. (2013). "A Theory of Optimal Inheritance Taxation." Econometrica, 81, 1851-1886.

Piketty, T and Saez, E. (2001). Income Inequality in the United States, 1913-1998. NBER Working Paper No. 8467.

Piketty, T. and Saez, E. (2006). The Evolution of Top Incomes: A Historical and International Perspective. NBER Working Paper No. 11955.

Piketty, T and Zucman, G. (2014). "Capital is Back: Wealth-Income Ratios in Rich Countries, 1700-2010." Quarterly Journal of Economics, 129 (3), 1255-1310.

Sala-i-Martin, X. (2006). "The World Distribution of Income: Falling Poverty and the Convergence Period." The Quarterly Journal of Economics, 121, 351-397.

Shin, I.; Kim, H. and Yamamura, E. (2009). Technological Progress and the Future of Kuznets Curve's. MPRA Paper No. 18866.

Solow, R. M. (1994). "Perspectives on Growth Theory." Journal of Economic Perspectives, 8 (11), 45-54.

Stiglitz, J.E. (2003). Globalization and Its Discontents. W. W. Norton, New York.

Stiglitz, J.E. (2012). "Macroeconomic Fluctuations, Inequality, and Human Development." Journal of Human Development and Capabilities: A Multi-Disciplinary Journal for People-Centered Development, 13 (1), 31-58.

Stiglitz, J.E. (2015). The Great Divide. W. W. Norton, New York.

Sukiassyan, G. (2007). "Inequality and Growth: What Does the Transition Economy Data Say?." Journal of Comparative Economics, 35 (1), 35-56.

Tat Hui, W. and Toh, R. (2014). "Growth with equity in Singapore: challenges and prospects", ILO Working Papers 485017, International Labor Organization.

Venieris Y. and Gupta, D. (1986). "Income Distribution and Sociopolitical Instability as Determinants of Savings: A Cross Sectional Model." Journal of Political Economy, 94, 873-883.

Voitchovsky, S. (2005). Does the Profile of Income Inequality Matter for Economic Growth? Distinguishing Between the Effects of Inequality in Different Parts of the Income Distribution. Journal of Economic Growth, 10, 273-296.

Wade, R.H. (2004). "Is Globalization Reducing Poverty and Inequality?" World Development, 32, 567-589.

Wade. R.H. (2003). Governing the Market. Princeton University Press: Princeton, NJ.

Weil, D.N. (2005). Economic Growth. Addison-Wesley.

Williamson, J. (1990): "What Washington Means by Policy Reform", in J. Williamson (ed.), Latin American Adjustment: How Much Has Happened?, Institute for International Economics, Washington DC.

World Bank (1993). The East Asian Miracle: Economic Growth and Public Policy. Oxford University Press.

World Bank (2002). Globalization, Growth, and Poverty: Building an Inclusive World Economy. Oxford University Press: New York.

World Bank, y IMF (2015). Ending Poverty and Sharing Prosperity. Global Monitoring Report 2014/2015.

Table 4: Regression coefficients

Model   B            Unstandardized       Std.           t
                     Coefficients         Coefficients

                     Std. Error   Beta

1       (Constant)   -41.685      7.922                  -5.262
        Log_PIBpc    4.536        1.220   .331           3.719
        Cons         .342         .061    .475           5.571
        1 Topi       -.451        .127    -.236          -3.561
        Top 10       .150         .057    .195           2.627
        Apert        .024         .008    .210           3.080
        Infl         .005         .084    .004           .057
        Escol        -.080        .028    -.206          -2.858

Model   B            Sig.    95.0%                      Correlations
                     Lower   Confidence
                     Bound   Interval for B

                             Upper Bound   Zero-order     Partial

1       (Constant)   .000      -57.295      -26.075
        Log_PIBpc    .000       2.133        6.940         -.084
        Cons         .000       .221          .462          .317
        1 Topi       .000       -.701        -.202         -.101
        Top 10       .009       .038          .263          .226
        Apert        .002       .009          .039          .002
        Infl         .954       -.161         .170          .017
        Escol        .005       -.135        -.025         -.059

Model   B            Correlations    Collinearity

                     Part     TOL    VIF

1       (Constant)
        Log_PIBpc    .239    .219    .439   2.276
        Cons         .345    .329    .478   2.090
        1 Topi       -.229   -.210   .795   1.257
        Top 10       .171    .155    .630   1.588
        Apert        .199    .182    .749   1.335
        Infl         .004    .003    .742   1.348
        Escol        -.186   -.169   .670   1.493

(a.) Dependent Variable: yityit1

Table 5: Summary of the model

Model      R         R      Adjusted      Std. Error
                   Square   R Square        of the

1       .451 (a)    .203      .179     3.019096502430499

Model              Change Statistics             Durbin-

        R Square     F      df1   df2   Sig. F
         Change    Change               Change

1         .203     8.349     7    229    .000     1.209

(a.) Predictors: (Constant), Log PIBpc, Top1, Top10,
Apert, Esco1, Infl

(b.) Dependent Variable: yityit1

Table 6: Variance analysis (ANOVA)

Model             Sum of    df     Mean      F       Sig.
                 Squares          Square

1   Regression   532.687     7    76.098   8.349   .000 (b)
    Residual     2087.322   229   9.115
    Total        2620.009   236

(a.) Dependent Variable: yityit1

(b.) Predictors: (Constant), Log_PIBpc, Top1, Top10,
Apert, Esco1, Infl

Table 7: Correlation grid

Model                           Escol        Top 10         Apert

1   Correlations   Escol        1.000
                   Top 10       -.059         1.000
                   Apert        -.344         -.037         1.000
                   Topi         .080          -.378         -.018
                   Infl         .197          .052          -.011
                   Cons         -.268         -.332         .413
                   Log_         -.385         .196          .206
    Covariances    Escol        .001
                   Top 10    -9.328E-005      .003
                   Apert     -7.452E-005   -1.637E-005   6.035E-005
                   Topi         .000          -.003      -1.819E-005
                   Infl         .000          .000       -7.457E-006
                   Cons         .000          -.001         .000
                   Log_         -.013         .014          .002

Model                        Top1    Infl    Cons    Log_ PIBpc

1   Correlations   Escol
                   Top 10
                   Topi      1.000
                   Infl      -.024   1.000
                   Cons      -.044   .048    1.000
                   Log_      -.292   .330    .485      1.000
    Covariances    Escol
                   Top 10
                   Topi      .016
                   Infl      .000    .007
                   Cons      .000    .000    .004
                   Log_      -.045   .034    .036      1.488

Table 8: Collinearity diagnosis

Model   Dim.   Eigenvalue   Condition   Variance Proportions

                                        (Constant)   Log_    Cons

1        1       7.390        1.000        .00        .00    .00
         2        .430        4.144        .00        .00    .00
         3        .112        8.113        .00        .00    .00
         4        .043       13.123        .00        .00    .00
         5        .016       21.755        .00        .01    .00
         6        .005       37.769        .02        .02    .03
         7        .003       50.345        .00        .17    .34
         8        .000       136.077       .98        .79    .63

Model   Dim.          Variance Proportions

               Top1   Top10    Apert   Infl   Escol

1        1     .00     .00      .00    .00     .00
         2     .00     .00      .01    .70     .00
         3     .04     .01      .66    .01     .00
         4     .79     .00      .05    .01     .01
         5     .06     .56      .06    .04     .04
         6     .00     .04      .05    .10     .88
         7     .08     .39      .05    .05     .02
         8     .04     .00      .12    .10     .06

(a.) Dependent Variable: yityit1

Table 9: Casewise Diagnostics (a)

Case       Std.         yityit1        Predicted Value
Number   Residual

45        3.445     11.5758938317498   1.1748233835969

Case          Residual

45       10.401070448152943

Table 10: Residual statistics

                           Minimum       Maximum       Mean

Predicted Value            -6.95296239   2.57271385    -.557704140
Std. Predicted Value       -4.257        2.084         .000
Standard Error of          .304          1.199         .536
  Predicted Value
Adjusted Predicted Value   -7.5538048    2.54497933    -.557064565
Residual                   -7.0952997    10.40107059   .000
Std. Residual              -2.350        3.445         .000
Stud. Residual             -2.414        3.477         .000
Deleted Residual           -7.4876952    10.59574317   -.000639574
Stud. Deleted Residual     -2.440        3.565         .000
Mahal. Distance            1.395         36.255        6.970
Cook's Distance            .000          .071          .005
Centered Leverage Value    .006          .154          .030

                           Std. Deviation   N

Predicted Value            1.50238105       237
Std. Predicted Value       1.000            237
Standard Error of          .142             237
  Predicted Value
Adjusted Predicted Value   1.51205974       237
Residual                   2.97398473       237
Std. Residual              .985             237
Stud. Residual             1.004            237
Deleted Residual           3.08723826       237
Stud. Deleted Residual     1.009            237
Mahal. Distance            4.770            237
Cook's Distance            .010             237
Centered Leverage Value    .020             237

Sergio A. Berumen

Department of Applied Economics I Faculty of Law and Social Sciences King Juan Carlos University, Madrid, Spain

(1) In the first period, a change in will affect the level of income of the measure, however, the long-term effect on the level of income in a stationary state will be determined by. See Barro (1999).
Table 1: Nature of the relationship between income
inequality and economic growth, depending on the approach

Relationship                          Authors

Positive         Okun (1975), Benabou (1996), Aghion and Howitt
                 (1998) and Forbes (2000), amongst others.

Negative         Murphy et al. (1989), Alesina and Rodrik (1994),
                 Alesina and Perotti (1996), Acemoglu (1998),
                 Panizza (2002), Helpman (2004) and Sukiassyan
                 (2007), amongst others.

Questionable/    Amos (1988), Barro (2000), Weil (2005) and Shin
Non-conclusive   et al. (2009), amongst others.

Source: own data

Table 2: Implications of the relationship between
economic growth and inequality

              Early stage of
              economic growth

              Policy    High growth   Inequality

Priority      Low tax   Can be        Can be
of growth     rates     achieved      achieved

Priority to   Low tax   Can be        Can be
correcting    rates     achieved      achieved

              Close to
              stationary growth

              Policy     High        Inequality
                         growth      Correction

Priority      Low tax    Can be      Cannot be
of growth     rates      achieved    achieved

Priority to   High tax   Cannot be   Can be
correcting    rates      achieved    achieved

Source: own data based on Barro (2000)

Table 3: Percentage of income concentrated
in the top 1%, 5% and 10%. 1975-2005

Country       Top %   1975    1980    1985    1990

Germany        1%     9.93    10.43   9.57    10.49
               5%     20.84   21.82   21.06   22.54
               10%    30.82   31.67   31.03   33.03

Denmark        1%      6.8    5.47    5.21    5.17
               5%     18.76   15.89   15.12   15.44
               10%    29.51   25.85   24.59   25.1

Spain          1%      --      7.4    7.55    8.36
               5%      --     20.12   22.03   23.17
               10%     --     31.61   33.72   35.35

France         1%     8.48    7.63     7.2    8.23
               5%     22.06   20.11   19.96   21.45
               10%    33.41   30.69   31.05   32.64

Ireland        1%     5.96    6.65    6.27    6.64
               5%      --      --      --      --
               10%    28.62   31.5    31.28   31.05

Italy          1%     7.24     6.9    6.81    7.78
               5%     20.04   17.72   17.5    19.69
               10%    31.2    27.17   26.83   29.5

Norway         1%     5.41     4.6    4.45    4.28
               5%     17.5    15.06   14.21   13.47
               10%    29.03   25.26   23.64   22.19

The            1%     6.12    5.88    5.92    5.56
Netherlands    5%     17.4    17.52    18     17.33
               10%    27.47   28.26   29.1    28.2

Portugal       1%     7.49    4.32     6.1    7.21
               5%     20.12   12.49   16.89   20.7
               10%    30.71   18.77   24.76   31.19

United         1%      6.1    6.34     7.4     9.8
Kingdom        5%     17.4    18.42   20.75   24.43
               10%    27.82   29.32   32.65   36.9

Sweden         1%     5.29    4.05    4.12    4.38
               5%     16.14   13.44   13.35   13.73
               10%    26.38   22.73   22.33   22.75

Switzerland    1%     8.79     8.4     903     9m
               5%     20.47   20.04   20.64   20.35
               10%    30.29   29.88   30.35   30.23

Country       Top %   1995    2000    2005    2009

Germany        1%     8.84    10.99   11.95   14.42
               5%     20.84   24.03   25.89   28.93
               10%    31.4    35.21   37.58   40.02

Denmark        1%     5.03    5.73    5.78    5.44
               5%     15.14   16.18   16.17   15.84
               10%    24.58   25.67   25.66   25.44

Spain          1%     7.88    8.65     8.8    8.52
               5%     21.71   22.2    22.24   21.54
               10%    33.37   33.45   33.32   32.46

France         1%      7.7    8.29    8.73     808
               5%     20.93   21.65   21.88   21.44
               10%    32.41   33.05   32.89   32.69

Ireland        1%     8.19    10.32   11.6    10.5
               5%      --      --      --      --
               10%    35.33   33.87   36.43   36.13

Italy          1%     8.13     9m     9.35    9.38
               5%     20.58   22.56   22.78   23.17
               10%    30.57   32.94   33.19   33.87

Norway         1%     7.36    10.31   16.49   7.11
               5%     17.83   21.36   28.13   17.95
               10%    26.86   30.45   37.06   26.95

The            1%     5.37    5.61    6.81    6.43
Netherlands    5%     17.32   17.21   19.34   19.07
               10%    28.45   28.02   30.69   30.56

Portugal       1%     8.41     9m     9.77     --
               5%     23.84   24.58   26.01    --
               10%    35.38   36.13   38.25    --

United         1%     10.75   12.67   14.25   15.42
Kingdom        5%     25.8    27.04   29.57   T999
               10%    28.51   38.43   41.62   41.53

Sweden         1%     5.25    5.97    6.28    6.72
               5%     15.54   17.12   17.33   17.97
               10%    24.93   26.72   26.96   27.93

Switzerland    1%     8.48    10.42   9.84    10.54
               5%     19.99   22.44   21.84   22.9
               10%    29.94   32.32   31.88   33.15
COPYRIGHT 2016 Council for Social and Economic Studies
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Berumen, Sergio A.
Publication:The Journal of Social, Political and Economic Studies
Article Type:Report
Geographic Code:4E
Date:Jun 22, 2016
Previous Article:Economic openness, power resources and income inequality in the American States.
Next Article:Resources and opportunity for change: democracy, labor and the welfare effort in Latin America.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |