Financial Development and Income Inequality: Evidence from African Countries in the Franc Zone

The aim of this article is to evaluate the effect of 
financial development on income inequality in a sample of African countries of 
the Franc Zone. Using data from Worldwide Governance indicators, UNESCO, COBAC, 
the Banking Commission of West African States and the World Bank, based on 
cylindrical dynamic panel whose instrumentalisation and stationarity of 
variables enabled us to use GMM in system, the results show that financial 
development through its components which are credit to the private sector, 
network development as well as the monetary mass significantly reduce income 
inequality among households. This result is robust by alternative or dual use 
of the components as well as when several control variables are integrated.


Introduction
Based on the developments observed in the financial sector in the last two decades, the World Bank and the International monetary fund made it a strategic tool for the achievement of the aim of reduction of income inequality between the rich and the poor. Among financial development indicators, credit to the private sector as a percentage of GDP increased from 28.09% to 46% in Sub-Sahara Africa between 2010 and 2015 and from 15.4% to 16.3% for countries of the Franc Zone [1]. Equally, during the same period the monetary mass in terms of percentage of GDP increased respectively from 24.7% to 28.9% in the Franc Zone and 44.7% to 45.5% for Sub-Saharan African Countries. In addition to this dynamic of financial development, there is persistent income inequality among populations. Despite the fact that several studies highlight the importance of financial development in the explanation of the evolution of income inequalities which is a phenomenon that can easily be observed in developing countries.
In this regard, studies carried out so far have lead to the conclusion that, the effects of financial development on income inequality are not uniform. For example [2] show that financial development is susceptible of modifying the income level of the population through a non-linear relation; [3] estimates that the integration of the ratio of financial depth among the control variables has a positive and significant impact on but relatively weak on income inequality. Whereas [4] followed by [5] show that the influence of financial development on income inequality mainly depends on the structure of the economy considered. As a result they integrate interaction variables between financial intensity and the size of the modern sector in their regression. By estimating in transversal cuts using data on 71 countries for the period 1960 to 1995, [6] show that financial development affects the convergence of economies through growth of productivity instead of the accumulation of capital. The study of [7] on the role of financial development in the explanation of the inequality in income for a panel of 98 countries on the period 1980-2006, reveals that the reduction of the differences in income more often from a stable macroeconomic environment than an intensive financial sector. Contrary to this conclusion, [8] or again [9] established the existence of a linear relationship between financial development and income inequality. In the same way, [10] in a study on the rural area of China lead to the conclusion that there is an inverse relationship between financial development and income inequality.
The non-uniform nature of the results highlights the fact that these studies have a lot of weaknesses relative to the absence of certain considerations such as: 1) the bring together of countries belonging to different sub regions, but characterised by the use of a common currency; 2) the geographic dimension of financial development through the variable density of the network that enables not only to increase the size of the sample but also to appreciate the dynamism of the banking sector since a strong density facilitates access to credit by economic agents and reduces the inequality in revenue among them; 3) financial development as a composite variable is measured by 3 indicators namely credit to the private sector, an increase in the monetary mass and the non integrated variable which is the density of the network; and 4) the method of econometric analysis since that consecrated to the method of generalised moments in systems that is not widely used produces robust results on the analysis of the impact of financial development on income inequality.
In this light, the objective of this study is to evaluate the impact of financial development on income inequality for the period 2000 to 2014, on a sample of 14 countries 1 belonging to the Franc zone which form a homogenous group of 1 These countries are respectively: Benin, Burkina Faso, Cameroon, Congo, Ivory Coast, Gabon, Equatorial Guinea, Guinea Bissau, Mali, Niger, the Central African Republic, Senegal, Chad and Togo. countries united since the colonial period by a monetary policy based on the use of the Franc CFA as a unique currency with different macroeconomic specificities.
The rest of the study is presented as follows; in Section 2 the empirical strategy is presented. Section 3 presents and describes the data. Whereas Section 4 is dedicated to the results and Section 5 concludes the study. MM . Low-income economies often resort to such policies to boost demand and favour the reduction in income inequality [13]. This can explain the differences in income levels among the populations. Finally, the density of the banking network ( The second category is the control variables which are also determining factors of income inequality. They include, the growth rate of the GDP ( , i t Y ); [14] developed an explicit relationship between economic growth and income inequality with a causality link such that the different phases of economic development determine the distribution of income. This hypothesis is explained by the fact that growth is beneficial to the poor at the stages of development of a traditional sector. Also, included are the demographic variable namely the population growth rate (    est rates tend to be high, households with high purchasing power can benefit from the increase in income from their investments. On the contrary, modest income households cannot save. Thus, they do not benefit from this opportunity and at the end, the differences in direct income increase given the growth differential of incomes linked to the returns on saving [18].

Empirical
Taking into consideration the size of the sample and the more or less available data as well as the fact that the impact of financial development on income inequalities can take some time, inequality would be measured as the variation of its indicator during the period 2000 to 2014. This leads to the specification of the model in the form: In Equation (1)

The Econometric Strategy
The estimation of the impact of financial development on income inequality has a problem of endogeneity that can come from the omission of pertinent variables, a bias of simultaneousness or even the presence of a measurement error in one or the other of the control variables. In fact, at a precise date the level of income inequality in a country or a region can be influenced by its level at a pre-2 Final Prediction Error. 3 Akaike's Information Criteria. 4 Hannan and Quinn Information Criterion. tests associated to it [20] and that justifies its robustness. This includes on one hand the test of Sargan or Hansen that enables to test the validity of retarded variables as instruments and on the other hand to test the self-correlation based on the null hypothesis of the absence of serial self-correlation of the 2 nd order.

Variables and Sources of Data
Two types of variables are taken into account in this study. There is the endogenous variable, the exogenous variables which are divided into two categories that is financial development variables and control variables. The data relative to all these variables are not from the same source. Moreover, they cover the period from 2000 to 2014. The choice of this period was dictated by the availability of data on the banking sector and the aspect related to the quality of institutions (Table 1).

Descriptive Statistics and the Correlation Structure
The descriptive statistics presented in  There is a negative correlation between the Gini index and the financial variables (Table 3)    sector and the monetary mass significantly reduce income inequality at a threshold of 1% whereas this reduction is 10% for the density of the network. The analysis of the robustness of these effects will be done in two ways.

Results
First, we take into account the alternative measures of the different variables that constitute the vector of financial variables. As such, according to the results of Table 5 regressions 1, 6 and 11 show that a better access to credit, a consistent monetary mass or a good level of network development significantly respectively reduce income inequality at a threshold of 1%. These results are robust when new control variables are introduced especially public expenses (regressions 3 and 12), human capital (regressions 4, 10 and 15). More specifically bank credit to the private sector impacts negatively and significantly the Gini index. [23] came out with such results that indicate that the expansion of credit to the private sector can stimulate the growth of income at the level of poor quintiles and consequently reduce income inequality. The illustration of the poor peasant of [24] who needs credit to invest reinforces the idea of a positive effect of credit to the private sector on the reduction of income inequalities. The more credit to the private sector increases, the higher the incomes of poor households who have invested. Thus, a reduction of income differences between the poor and the rich [25].
Moreover, as for the ratio of the monetary mass as a percentage of GDP, the results show that an increase in this ratio leads to a significant fall in the Gini index. This variable that represents the rate of monetisation of the economy or adduction of money in the economy translates the idea of a positive impact of the quantity of money in circulation in an economy on income inequality. In fact, an increase in the quantity of money available increases the speed of circulation of money this improve access to money by economic agents which facilitates the transactions of economic agents who can use the money to have access to health services, nutrition, education, … Thus, by improving the living conditions of citizens, financial development through an increase in the quantity of money in circulation leads to an increase in the income of the populations even the very poor and consequently to a reduction in the possible differences in income between these later [26].
Finally, as concerns the variable access to financial services which translates the density of the banking network, it appears that an increase in the number of tellers in banks reduces the average number of persons using the services of the same bank branch. This increases the average efficiency per teller and a better access to financial services by economic agents. Under these conditions, a better access of the population to bank services is translated by a fall in the density of the network and by an induced effect that is translated by a fall in the Gini index.
This result is similar to that of [11] who explained that an increase in the number of bank accounts for every one thousand adults reduces income inequalities.
That is why the efforts made by countries of the Franc zone since the year 2000 are appreciable. As such, the network density of the CEMAC sub-region moved from 151,520 inhabitants per branch in 2006 to 90,414 in 2014 [27]. In the UEMOA countries this network density was already estimated at 116,000 inhabitants in 2005 per teller [28]. This increase in financial penetration is an indicator of development of the financial sphere that improves the distribution of income among the populations concerned.
Secondly, the analysis of the robustness consists of taking into account couples of financial variables (Table 6). Even in this case, financial variables reduce income inequality. The results of the tests confirm these results. By using these income inequalities in African countries of the Franc zone during the period from 2000 to 2014. Recent theoretical studies have showed through different methodological approaches that financial development plays a primordial role in the reduction of income inequalities either by credit to the private sector or by an increase in the monetary mass. This article however investigates from a different dimension of financial development that integrates geographical aspects of the development of the financial system namely, the density of the banking network or the rate of penetration of bank branches in the economic territory.
Using the method of generalised moments in system our results suggest that the effects of financial development on income inequality are statistically significant and of real important economic contributions. The geographical increase in the number of bank tellers increases the average efficiency per teller and improves access of economic agents to financial services at a lower cost and leads to the development of new activities that create income for poor households. Equally, financial development increases the rate of monetisation of the economy and enables a better supply of bank credit to households and entrepreneurs. This offers better possibilities of raising income to economic agents with low income and reduces the income gap between the rich and the poor.