The Impact of Financial Development on Poverty, Growth and Inequality in Ethiopia

Introduction

Back Ground of the Study

Seers (1969) stated that development occurs with the reduction in poverty, inequality and unemployment within a growing economy [1]. For poverty reduction, Beck et al. (2007) suggest that financial development boosts incomes of the poor and reduces inequality. Based on Ä?ihák et al. (2012), this financial development is defined as a combination of depth, access, and efficiency [2,3].

A large body of literature assessed the impact of financial development on economic growth, inequality and poverty reduction [4-8,59]. However, there is conflicting view on distribution of income and poverty reduction, financial development involves welfare improvements in the functions provided by the financial systems like pooling of savings, capital allocation to productive investments, investment monitoring, risk diversification and exchange of goods and services [9].

But Jalilian and Kirkpatrick (2001), stated that financial development can contribute to poverty reduction by directly creating access to credit, saving and insurance and the indirectly through economic growth and income distribution [10]. First, by improving opportunity for the poor to invest in new technologies, education and social issues. Second, to achieve a level of sustainable gains by the poor, the financial system can enable the poor to access financial services. This in turn enhances productivity. Thus, financial instruments play a significant role on poverty reduction.

In Ethiopia, the year 1991 is marked as a turning point for the whole economic policy change. It was the period in which a shift from a controlled to market friendly policy regime [11]. As relation to this change, in World Bank country profile of (2018),the population annual growth rate is declining for the last three decades and poverty head count ratio at national poverty lines (%age of population) is declining [12]. It was 44.2, 29.6 and 23.5 in 2000, 2010 and 2018 respectively. In relation to inequality, income share held by the lowest 20% is declining. These show that poverty and inequality remain an obstinate challenge.

Furthermore, Ethiopia has recorded growth over the last 15 years led to walfare improvement. PCI reached USD 985 in 2018/19. Poverty declined from 38.7% in 2004/05 to 22% in 2018/19. Investment to GDP ratio increased to 35.2% while domestic savings stood at 22.3%; industrial sector showed 12.6 % growth; secondary education enrolment shows improvements by 5.8%; total telecom subscribers reached 43.6 million; inflation 12.6% and SMEs development has reached total loan disbursed Birr 7.3 billion [13].

Problem Statement

The world we are living is dynamically changing and the time is puzzling with poverty, inequality, war, climate change and pandemic. Contrary to literatures’ expectations of the trend called “Africa rising”, Tita and Aziakpono (2017) stated that Africa’s growth story has never been pro-poor [14]. Subsequently, it has little impact on poverty reduction and the income inequality which remains stubbornly high, suggesting that strong growth has largely been enjoyed by a few elite and rich people. However, Appiah et al. (2020) argue that in considering poverty, one echoes bone-dry population and poor health care system but not access to financial service [15].

However, there are large bodies of literatures have shown the impact of financial development on enhancing growth, inequality and poverty reduction. But the question is whether the focus of development strategies should be on direct or indirect impacts. There are inconsistent prediction on empirical studies in the impact of financial development on growth-inequality-poverty triangle. The first prediction shows a positive linkage between financial development on growth and poverty reduction [7,16]. The second prediction is that due to inadequate distribution of the gain of growth, the positive effects are weakened by the presence of growing inequalities [7,11]. These studies show lack of access to finance is the main cause of the pervasive generation of income inequality and poverty.

Furthermore, the model developed by Greenwood and Jovanovic (1990) predicts a nonlinear relationship between financial development and income inequality that depends on the level of economic advancement [17]. At low levels of prosperity, the wealthy and politically connected can afford to access finance. As Kuznets’ (1955) hypothesis of inverted-U, only after a certain threshold of economic development, a larger proportion of society benefits from financial deepening [18]. Nevertheless, Banerjee and Newman (1993) and Galor and Zeira (1993) suggest a linear relationship between financial development and income inequality [22,23]. The theoretical assumption is that financial market imperfections may be especially binding on the poor.

On other hand, evidiences suggest that the existence of an adverse consequence of quality of institutions on inequality. Argue where institutional quality is strong, financial development reduces inequality. When institutions are weak, improving welfare or reducing inequality is difficult due to discriminatory lending and corruption which opens door for oligarchy and establish interests to sway finance away from the poor [9].

In Ethiopia, domestic credit market is dominated by lending to state owned enterprises. The financial repression used to manage monetary policy and financing of large projects. This led to macro-financial imbalances, fiscal dominance, pressures on inflation, a chronic shortage of foreign exchang and the lack of financial system advancement [24]. However, Kirkpatrick et al., (2001) argue in favor of regulated rather than liberal financial markets to avoid financial crises and its instrumentality in poverty reduction [10]. The financial liberalization has led to an increase in financial fragility in developing countries. Thus, strict adherence to prudential regulations will reduce the incidence of failure and political repercussions.

To have a comprehensive picture of financial development, testing its total effect simultaneously with the triangle of "growth- inequality-poverty" is the interest of this paper. Yet, two main issues considered as a research gap. First, the findings of existing literatures are conflicting and leaving the debate open for further research. Second, the existing literatures failed to show its multidimensional impact by simultaneously evaluating the direct and indirect effects. In this context, searching for the dynamics of financial development on the wellbeing of the poor which has sociopolitical and economic consequences has a great importance. Therefore, the findings of this study is hoped to shed light for the planers to strategically design an effective policy.

Research Question

Primary Research Question:

How does the financial system development impact poverty, growth and inequality in Ethiopia?

Secondary Research Questions:

How much statistically significant are the direct and indirect impacts of financial development on poverty reduction?

Is there a trickle-down effect between financial development and economic growth?

What is the role of a financial institution on improving the wellbeing of the poor?

Is there Kuznets’ inverted-U shape relationship between financial development and inequality?

Is there a long run cointegration between financial development and poverty reduction?

Research Objective

General Objective

The aim of this thesis is to investigate the impacts of financial development on economic growth, inequality and poverty reduction in Ethiopia.

Specific Objectives

• First, it makes use of theoretical framework to establish link between financial development, poverty, growth and inequality

• Second, it tests the direct and indirect impact of financial development on poverty

• Third, it examines the trickle-down effect of financial development and economic growth

• Forth, it evaluates the role of institutional quality on reduction of inequality

• Fifth, tests Kuznets’ inverted-U hypothesis of financial development and inequality

• Finally, it provides suggestions for policy makers to promote inclusive financial system development to enhance innovation and industrial development

Research Hypothesis

Based on the theoretical and empirical literatures, the following hypotheses are developed.

• H1: There is direct effect of financial development on poverty reduction through access to financial service

• H2: There is indirect effect of financial development that improves the wellbeing of the poor through growth and income distribution

• H3: There is trickle-down effect between financial development and economic growth

• H4: There is improvement of walfare where there is a strong financilal institutions

• H5: There is Kuznets’ inverted-U shape relationship between financial development and income inequality

• H6: There is a long run cointegration relationship between financial development and poverty reduction

Literature Review

This chapter presents the concept of financial development, poverty, growth and inequality. It further discusses how poverty, growth and inequality are affected by financial development.

Theoretical Literature Review

Financial Development and Poverty Reduction

Zhuang et al., (2009) claim that financial sector development can impact poverty through two channels which is direct and indirect. First, financial development can allow the poor to expand the access to financial services which is direct impact. Second, financial development contributes to poverty reduction through economic growth. This is indirect effect is called trickle-down theory. In this theory, economic growth trickles down to the poor through job creation, income distribution effect and other economic opportunities [25].

Source: Adapted from Claessens and Feijen (2006) [20]

Figure 2.1: Financial Sector Development and Poverty Reduction

According to, the direct link refers to how the benefits of financial development are transferred to reducing poverty and income inequality while the indirect link shows how financial deepening exerts a positive effect on economic growth, and how the gains from growth are channeled to the poor [2]. It may affect the poor households through two channels which are aggregate growth and changes in the distribution of income. First, the development of the financial system can ease the credit constraints. The implication of the increased access to credit to the poor population can spend extra money on consumption expense and improve their welfare. Second, the market competition will rise as new firms enter to the sector. Jalilian and Kirkpatrick, (2001) stated a well- developed financial system tends to facilitate the process of capital accumulation which increases the level of economic growth which in turn assist in reducing poverty. This will reinforce the productive assets of the poor and enhance productivity [10].

The Direct Effect of Financial Development

Most of the theoretical and empirical results of the previous studies show that financial development contribute to poverty reduction through direct channel [23]. There is impact of finance on the welfare of the poor who lack resource to finance the productive projects [26]. This in turn slows the aggregate growth of the economy by diverting away the flow of fund to lesser ranked projects. If financial system fail to function properly, it will lead to higher income inequality by giving the opportunity to wealthiest.

The sound argument in the case of the developing countries is that for poverty eradication strategy, the focus shall be given to the poor to finance the small and medium-sized enterprises (SMEs) and informal businesses. It is because SMEs are labor intensive and creates job for the poor household. Fields (2001), as the credit and financial market is underdeveloped, the gain supposed to be harvested by the poor will be lost. This will result in high poverty and inequality and also the growth of the economy could be slower than expected. The implication is that creating better opportunity for poor to access financial service and increased entrepreneur participation through productive projects in turn expands their income. Thus, allowing credit access to the poor has important policy implication to reduce the likelihood of sinking in poverty trap and by smoothening the consumption and managing shocks of health and other challenges.

Growth in financial system might not spontaneously result in greater access of finance and removal of credit constraint. Therefore, the government intervention in form of microfinance and SME credit programs. The absence of public sector inclusive policy, lenders prefer to give huge amount of loans to the few rich. Hence, the government shall support the pro-poor sectors and SME to help the poor segment. The main bottle neck for growth of SMEs is often access to finance. Hence, government shall support SMEs to adopt new technology, train and empower, and facilitate linkages between SMEs with the experienced large enterprises to adopt technology and diversify investment into new markets [25].

Furthermore, Kirkpatrick et al., (2001) stated micro-finance plays a significant role in directly linking finance and poverty reduction. It can alleviate poverty and the best practice technologies can be translated into poverty reduction by directly linking finance. Thus, providing effective micro-finance services to poor people is part of a poverty-reduction strategy [10].

The Indirect Effect of Financial Development

The indirect channel through which financial development backing poverty reduction is basically economic growth and income distribution. There are different alternative dimensions that show the impact. For instance, argue that through job creation to the poor households, wage differentials between skilled and unskilled labor, increased tax revenue that resulted from the economic growth that could be distributed to fiscal resources for the socioeconomic expenditure (i.e. education, health and other social protection programs. Moreover, Aghion and Bolton (1997) conclude that economic growth leads to the availability of more public funds through the capital accumulation and increases the income level of the poor [26].

As Kuznets (1955) prediction, financial development reduces inequality. Thus, the indirect effect of financial development can have impact on income distribution. Likewise, Datt and Ravallion (1992) have tried to answer the changes in poverty as a result of growth effect, changes in average income and the income distribution effect which is resulted from the shifts in Lorenz curve. It is concluded that growth effect explains the changes in poverty that had been observed. According to Fields (2001), the level of inequality and the rate of economic growth are determining the extent of impact of growth on poverty eradication [27]. Similarly, there is an agreement on the importance of growth for fighting and it shall not be understated. This doesn’t mean growth alone is sufficient for poverty alleviation although it is necessary condition. In order to harvest the benefits of growth, the poor households shall expand their asset base.

Financial Development and Economic Growth

The theoretical foundations of the argument on financial development and growth were established by Joseph Schumpeter is creative destruction process, financial intermediaries materialize technological innovation and contribute to economic development. It continues with McKinnon-Shaw financial development approach and has been expanded for more countries by King and Levine (1993) as a representative study in the empirical literature. They argue financial depth is at the core of growth, in other words, financial repression hinders economic growth.

Findings on recent litrature show that financial development leads to growth and it is growth is pro-poor [7]. It means that the trickle- down hypothesis is firmly supported. Kuznets (1955) hypothesis, in the case of inequitable growth in developing countries, there will be limited impact of growth in poverty reduction [18]. Also, Beck, et al., (2007) argue that depending on the level of inequality, economic growth could have a positive impact on poverty reduction but not necessarily improve the wellbeing of the poor [2].

Levine (2005) stated that financial development affects factors of production that determines economic growth [19]. These factors are the accumulation of physical and human capital and total factor productivity that result from improvements in functions provided by the financial systems like mobilization of saving and capital allocation to productive projects. Then, proper monitoring of these investments and risk diversification. Therefore, financial development improves information flow, encourage risk sharing and improve the capacity to absorb shocks and reduce the volatility of macroeconomics and finally, enhance distributional effect of income.

Research Methodology

Introduction

The objective of this paper is examining the impact of financial development on poverty, growth and inequality in Ethiopia. In measuring the impact of financial development on poverty reduction, the simultaneous effects of growth and inequality has been considered.

Research Design

By using simultaneous equations model, the study assesses the virtues of financial development in reducing poverty and consider the simultaneous effects of growth and inequality. In doing so, the study adopted explanatory approach by using annual time series data to realize the stated objective. It is a type of correlational research that involves observing variable over an extended period of time. Thus, the research design looks at a long term effect of financial development.

Source and Type of Data

The data used in this study is annual time series data for the time period of 1991 to 2019. Basically, the secondary data collected from World Bank, the World Development Indicators (WDI) database of 2019 and National Bank of Ethiopia.

Econometric Model Specification

To evaluate the impact of financial development on growth, inequality and poverty, this study used the simultaneous regression equations model. These equations are: poverty Equation (1), the growth equation (2) and inequality equation (3). To clarify the effects of financial development, the study first questioned the direct effects on the endogenous variables which are poverty, growth and inequality. Then, it focuses on the variables specific to each equation.

In order to deal with the simultaneous regression, based on the prediction of Datt and Ravailon (1992), the effects of the progress of financial development had been decomposed into two conflicting effects. These are: the growth effect and a disparity effect. Besides, Dhrifi (2015) noted on examining poverty reduction by taking into account simultaneous effect of financial development on inequality and growth [5,49]. Hence, introduction of the variable measuring financial development is taking the form of an exogenous shock. For instance, Ravallion (1997) used a model as a function of poverty which is a set of control variables that are normally employed as explaining poverty. In the equation of poverty: Gini coefficient, the GDP per capita growth; infrastructure variable and the growth of population are used. In the equation of inequality: the institutional variables that reflect how the distribution is made would be introduced. In addition, in the same equation the growth rate and its square to test the hypothesis of Kuznets are added [50].

The Kuznets curve (1955) tested in this model by considering the effect of growth of GDP per capita and through the introduction of the quadratic effect which is its square (GDPG)² [18]. The income growth influence inequality in the long run according to a non-linear or inverted "U" hypothesis. Adjusting, the shape of Kuznets curve, inequality increases in the early stages of economic development and decrease after reaching the maximum threshold but the situation is determined by the presence of quality institutions.

Furthermore, variables in simultaneous equation models are classified as endogenous and exogenous variables. Endogenous variables which are explained by the functioning of system and the values of which are determined by the simultaneous interaction of the relations in the model are endogenous. The simultaneous equations models is the endogenous variable in one equation becomes exogenous variable in another equation. This method allows us to correctly distinguish the effects of financial development passing through growth and rising inequality. Thus, in order to answer the main problem, financial development is the only explanatory variable common to all three equations are assumed. It is likely to affect simultaneously but the three endogenous variables affected in different ways. Accordingly, the overall association of this model is explained in the following way:

<img src="https://www.opastpublishers.com/scholarly-images/7799-697c2edca6b8e-the-impact-of-financial-development-on-poverty-growth-and-in.png" width="500" height="100">

Where Pov is poverty, measured by national head count ratio of the % of population living below poverty line. GDPg represent growth of GDP per capita. INQ is income inequality measured by the Gini index. FD is measured by domestic private credit as percentage of GDP. TEL is an indicator of quality of infrastructure measured by number of subscriber telephone lines per 100 inhabitants. PPg is growth rate of population. OPEN is trade openness measured by sum of import and export as % of GDP. INF is an indicator of inflation which is consumer price index. GoEx is government expenditure which is government final consumption. HK is the average years of secondary schooling in the total population which measures human capital and INST is an index measuring the institutional quality by summing up five different institutional variables.

In the simultaneous model, there is a problem of endogeneity (i.e. omitted variable in the model, measurement error in independent variables and endogenous error in relation to reverse causality in which X causes Y and Y causes X) as the variables interact simultaneously in the structural equation. Therefore, these could result in bias on OLS estimator and it could end up on the model inconsistency. Empirical studies on the inequality-development relationship could suffer from endogeneity and reverse-causality problems [9]. For instance, in Kumhof et al. (2015) the impact of higher inequality on bank lending may be caused by an increase in the income of the rich (by increasing loan supply), or it could be associated with a decrease in the income of the poor (by increasing loan demand).

As a result of endogeneity problem among the variable, the Three Stage Least Square (3SLS) model has been used. For solving endogeneity problems, the instrument variables introduced and the endogenous variable is among the explanatory variable. A resolution is given using the method of 3SLS and different econometric tests necessary to the model was made. For data analysis STATA 14.2 has been used to analyze the time series data.

Under the 3SLS estimation system of structural equations where some equations contain endogenous variables among the explanatory variable, a dependent variable has its usual interpretation as the left hand side variable in the equation with the associated disturbance term. All dependent variables that are explicitly taken to be endogenous to the system are treated as correlated with the disturbance in the system equation [51]. Since 3SLS is a full information estimation method, the parameters estimated are asymptotically more efficient than the 2SLS estimates. But 3SLS is quite vulnerable to model misspecifications. Thus, to minimize this error of the model specification, post estimation tests like endogenity test, weak instrument test, over identification tests have been used.

Furthermore, Greene (2012) demonstrates that among all the estimators of instrumental variables, 3SLS method estimator is asymptotically efficient. Basically, this model is adopted from the same methodology of Lundberg and Squire (2003) and Dhrifi (2015) to study the impact of financial development on the triangle of poverty-growth-inequality. Hence, in addition to a specific equation of poverty, the total effect of financial development is tested and an interaction with the triangle is identified. Furthermore, 3SLS method helps to differentiate the impact of financial development passing through growth and income inequality [5,51,52].

Definition of Variables, Measurement and Hypothesis

Source: Compiled by the author

                                                                       Table 3.1: Variable, Description and Source

Result and Discussion

Econometrics Model Results

Test for Unit Root

Based on the ADF test which is shown in Table 4.1 in Appendex, the data is stationary for all variables with p value less than 0.05 at unit root test made in three different ways: (1) an equation with intercept only, (2) equation with trend and intercept and (3) an equation with no trend and no intercept. The variables of human capital and government expenditure are not stationary at level but stationary at first difference or integrated of order one.

Johansen Cointegration Test

This study assumed that there is long run relationship among the independent variables and that infers cointegration. Irrespective of shocks in the short run which may affect movement in the individual series, it converges with time in the long run. In otherwords, these variables are moving in the same direction. This implies that by observing the movement of one variable, one could predict changes in other variables. Thus, the findings show that there is a long run association between financial development and poverty reduction.

The Regression Result on Effects of Financial Development

To clarify the impact of financial development on poverty reduction, first, analyzed the direct effects on the endogenous variables which are poverty, growth and inequality. Then, focusing on variables to each equation, regression results interpreted.

Direct Impacts of Financial Development on Poverty Reduction

The regression result of GDP growth on poverty reduction shows that there is a negative and highly significant relationship between growth rate of real GDP per capita and poverty headcount ratio at national poverty line. An increase in growth rate of 1% point results in a 0.2% point decrease in head count ratio of the population living below poverty line. Concerning the effect of inequality on the incidence of poverty, the test results shows inequality measured by the Gini index is negative and insignificant, confirming it’s not robust.

Financial development on poverty reduction is the main interest of this study. The test result is significant and the coefficient of beta is positive. A point increase in level of financial development results in 0.7 points increase in the population living below the national poverty line. It means that the headcount ratio of the people living below the poverty line increases as the domestic credit to private sector by banks (% of GDP) increases. This finding is consistant with existing literatures. Greenwood and Jovanovic (1990) stated that bank credits may be caught up by high unit costs of small loans for poor and small business [17]. The imperfections in the financial market that only favor the wealthiest with the lesser risk of default and ultimately, they grow. Therefore, it can be interpreted as market failed to allocate private credit, a requirement of collateral that excludes the poor and the lenders skewness in favor of rich to reduce transaction cost might be a reason to increase the head count ratio people living below the poverty line.

Unequivocally, the success and survival of small business and also employment, growth strategies, productivity and innovation largely depend on access to financial resources. SMEs have lower productivity and this is partly related with financial constraints. The productivity is an important issue because of the creative destruction process and the spillovers will be generated for large firms and economy-wide productivity [54]. Although SMEs serve as an engine of job creation and are seedbeds of developing entrepreneurial talent and innovation, they are constrained by the access to finance [55].

The effect of population growth on the poverty is significant and positive with very high coefficient of beta which can increase the poverty whereas the investment in infrastructure can reduce poverty. For instance, rural road quality and electrification rates also remain low in rural areas and that have a negative impact on poverty (Planing and Development Commission, 2018) [66].

Indirect Impacts of Financial Development on Poverty Reduction

Financial development may also impact poverty reduction through two possible strategic variables which are economic growth and income distribution 1).

Effect of Financial Development Through Economic Growth:

On examining the results of growth equation, the model has significant effect. The test results validates the pro-poor hypothesis and consistant with theoretical predictions. As expected, a % point increase in financial deepening leads to a 2.58% point growth of real GDP per capita. In other words, for every additional amount of private credit, the expected level economy grows by 2.58%. The result is consistent with the view “trickle-down” mechanism of Jalilian and Kirkpatrick (2001) and theoretical predictions advanced by King and Levine (1993) [10,28].

	Cons	-94.03097	58.3524	-1.61	0.107	-208.3996	20.33763
Inequality	GDPg	-.2109833	.0987176	-2.14	0.033	-.4044662	-.0175005
	FD	-.851981	.1028943	-8.28	0.000	-1.05365	-.6503118
	Inst	-7.444826	3.26526	-2.28	0.023	-13.84462	-1.045034
	GGDP2	-.0010087	.0120534	-0.08	0.933	-.0246331	.0226158
	Cons	40.27578	4.011109	10.04	0.000	32.41415	48.13741

Source: Author’s computation from Stata

                                                              Table 4.3: Long Run Regression Result of the 3SLS Model

In addition, there is a strong positive link between growth and human capital investment and trade openness in the process of reducing poverty. But government expenditure are negative and insignificant. The result is consistent with the findings of Barro (1997) [56]. Later, it’s supported by Beck et al. (2007) that trade openness reduces poverty while inflation is harmful to the poor [2].

Although Gini coefficient is positive and significant, it refer that increasing inequality by one unit enhances growth by 2.97%. Ravallion (1997) concluded that in the presence of high inequality, poverty may still rise irrespective of the high growth in developing countries [50]. Fields (1989) stated that inequality either increase or decrease with economic growth [57]. In the presence of increased inequality could end up on higher poverty just by eliminating the gain of economic growth. (Seven and Coskun 2016; Ho and Iyke 2018) argue that due to inadequate distribution of the gain of economic growth, the positive effects of financial development are weakened by the presence of growing inequalities [7]. Basically, lack of access to finance, the positive effects of finance and the rising inequality happen at a time and it results in the benefit of growth which is expected to be gained by the poor get offset. Similarly, Datt and Ravallion (1992) refer that change in poverty measures can be decomposed into growth and redistribution components [49].

2) Effect of Financial Development through Distribution Chunnel: Based on the results of the indirect effects, the distribution channel reveals that a expansion in financial service has a negative effect on the rate of poverty. An increase in one unit of the financial development indicator leads to 0.85% point decrease in Gini coefficient. This result refers that access to financial service enhances income distribution and ultimately, decreases poverty.

The test result of economic growth shows that a percentage increase in growth rate of GDP results in a 0.21% point decrease in inequality. It refers growth has a negative impact on income inequality since growth is often positively associated with higher tax revenue for socioeconomic expenditures (eg. for health and education), investments and access employments. Thus, it is concluded that growth trickle-down through distribution channel and reduces poverty.

In examining the institutional variable introduced to the model, a strong financial institution plays a negative coefficient as expected at 5% threshold level which validating the hypothesis. Indeed, it is evident from the result that improving the quality of institution by one standard deviation point leads to decrease in inequality rate by 7.44 percentage points. The institutional variables that reflect the quality of governance directly affects the interaction between economic agents like control of corruption or rule of law or government procedures. This indicates that to having an effective income distribution policy instrument can reduce poverty.

Furthermore, through the introduction of the effect of growth of real GDP per capita and the quadratic effect (GDPG)², Kuznets curve estimation result has been made and the test result shows that the coefficient of the growth rate of GDP per capita is significant and negative, whereas the coefficient of (GGDP)² is insignificant and negative. Thus, the inverted “U” shape of Kuznets curve that is explaining the average income inequality in the long run of the model is unverified. This finding refers that it is inconsistent with the dualistic theories that dictates that inequality and growth positively associated, then, stabilize at the peak to the turning point and at final phase of redistribution effect, the inequality decreases. Similarly, Batuo et al. (2010) find out no evidence supporting the hypothesis whereas Deininger and Squire (1996) stated that there is no systematic linkage between growth and aggregate inequality [45,58].

On other hand, the short run regression result has no significant change in p values nor in the coefficient of beta estimates and change in the sign. The result is more or less similar to long run results. The only difference in the short run result is the P value of Gini in direct channel of first equation and the P value of government expenditure in second equation. The p value of Gini has turned up to be significant in short run and the coefficient is positive.

Diagnostic Test for Validity

Autocorrelation Test: Duribin-Watson (DW) test statistic values in the regression result shown below for respective simultaneous equations. The DW d-statistic (6, 29) = 0 .9262703, (7, 29), α = 2.390341 and (5, 29), α = 0 .6795734 for equation #1, #2 and #3 of respectively. The DW table shows at 5% significance level the critical value to test autocorrelation of lower (dL) and upper (dU) is (1.050 and 1.841); (0.975 and 1.944); (0.900 and 2.052) for equation #1, #2, and #3 respectively. Therefore, it is concluded that there is no autocorrelation.

Multicollinearity Test:The correlation table 4.3 in Apendex shows that the correlation between independent variables indicates there is no series multicollinearity. It indicates that the VIF result is less than 10 and it indicates that there is no serial correlation in that model.

Breusch-Pagan/ Cook-Weisberg test for heteroscedasticity; Ho: Constant Variance
Variables:    Fitted   values    of Poverty, Pov	Variables:    Fitted   values    of Growth of GDP, GDPg	Variables:    Fitted   values    of Gini
Chi2 (1) = 0.11 Prob > chi2 =0.7434	Chi2 (1) = 7.58 Prob > chi2 =0.0059	Chi2 (1) = 0.16 Prob > chi2 =0.6855

Source: Author's computation from Stata

Table 4.4: Heteroscedasticity Test

The above results shows in the variables fitted values of 1st equation poverty and 3rd equation Gini, the p value is >0.05. In both cases, there is constant variance. Therefore, fail to reject Ho. But in the 2nd equation, p value is < 0.05 and it indicates that the model is heteroscedastic.

Diagnostic Tests for Endogeneity, Over Identification and Weak Instrument

Based on the ivregress test results of each simultaneous equations of the corresponding diagnostics, the decision is made for each test. For endogeneity test, the ivregress result shows that P value and the beta coefficients are significant. Furthermore, based on the Durbin and Wu-Hausman’s test result of the P value of endogeneity test of each equation is below 0.05. Therefore, the variables are endogenous and the instrumental variables are strongly correlated.

The test result of over identification of Sargan (score) for each equation is insignificant. It indicates the instruments are good and valid instrument and error terms are not correlated with instrumental variables. Moreover, weak instrument test shows minimum eigenvalue =35 which is high and greater than the critical value it means that it is possible to reject the null hypothesis that the instruments are weak. Besides to that, R square = 0.9 which is high and it indicates stronger instruments. Also, F statistics is significant both simultaneous equation #1 and #2. But for equation #3, it’s failed to reject the null hypothesis and there are cases where the literature on weak IV detection is still limited and open for the debate in econometrics.

Conclusion and Policy Implications

To understand full picture of the multidimensional effect of financial development, simultaneously evaluating through growth-inequality-poverty triangle is very critical and one step ahead of traditional framework of earlier literatures. Thus, this study explored to examine the impact of financial development on poverty, economic growth and inequality in Ethiopia. As a result of economic policy change in 1991, the time series data has been used for the time period of 1991 to 2019 by employing the simultaneous equation model. Then, theoretical framework developed to analyze the link between financial development and growth-inequality-poverty triangle with simultaneous effect.

The direct impact of financial development on poverty reduction through access to financial service increases the head count ratio people living below the poverty line. It indicates that financial development is failed to benefit the poor through the direct channel and not necessary helps the poor. Primarly, the wealthy, oligarchy and the politically connected reap benefit from the improvement of financial system. The imperfection of financial market, inefficiency of allocating credit, transaction cost, requirement of collateral and lenders preference to finance rich can be reasons to the increase in head count ratio of people living below the poverty line.

The indirect impact of financial development on poverty reduction findings support the existence of a positive and significant effect through growth and income distribution channel. This impacts positively on levels of job creation, capital accumulation, savings or technological innovations. The finding supports the theory of trickle-down and conclude that it is pro-growth.

A strong financial institution has a significant effect on reducing inequality while effective governance, rule of law, political stability and income distribution policy instrument can reduce poverty. On other hand, the findings of the inverted “U” shape of Kuznets curve that is explaining the average income and inequality in the long run of the model is unverified. Therefore, the theoretical and empirical prediction is not supported, however, Kuznets argues when the financial system develops, the credit constraint can be removed; the poor can borrow and reduce inequality. This refers that it is inconsistent with the dualistic theories.

In addition, there is an interesting finding on long-run cointegration association between financial development and poverty reduction. In general, therefore, it is concluded that financial development plays a significant role on poverty reduction and mainly, through the indirect channel of economic growth and income distribution both in short run and long run.

The government shall design a compulsory pro-poor public policy inquiring financial institutions to allocate certain percentage of loan to poor, SMEs and informal sectors. In this, the government shall give guarantee for financial institutions to minimize the risk of default as the poor lacks collateral. When the financial market not functioning efficiently, alterative policy option can play instrumental role to poverty reduction and architecture for sustainable development as it is creating opportunity for poor to send their children school, to manage health needs, to finance innovative and productive projects, to invest in human and physical capital. This in turn will help to promote inclusive financial system and create resilient growth trajectory that move from vicious to virtuous cycle and ultimately enhance innovation, productivity, entrepreneur and industrial development.

References

1. Seers,  Dudley  (1969). The  Meaning  of  Development.International Development Review 11)4):3-4
2. Beck, T., Demirgüç-Kunt, A., & Levine, R. (2007). Finance, inequality and the poor. Journal of economic growth, 12, 27- 49.
3. Cihák, M., Demirgüç-Kunt, A., Feyen, E., & Levine, R. (2012). Benchmarking financial systems around the world. World Bank policy research working paper, (6175).
4. Quartey, P. (2008). Financial sector development, savings mobilization and poverty reduction in Ghana. In Financial development, institutions, growth and poverty reduction (pp. 87-119). London: Palgrave Macmillan UK.
5. Dhrifi, A. (2015). Financial development and the" Growth- Inequality-Poverty" triangle. Journal of the Knowledge Economy, 6(4), 1163-1176.
6. Abosedra, S., Shahbaz, M., & Nawaz, K. (2016). Modeling causality between financial deepening and poverty reduction in Egypt. Social Indicators Research, 126, 955-969.
7. Ho, S. Y., & Njindan Iyke, B. (2018). Financial development,growth and poverty reduction: Evidence from Ghana.
8. Tekin, I., & Cin, M. F. (2020). Finance, Growth and Income Distribution. Finans Politik & Ekonomik Yorumlar, 57(653), 161-188.
9. Demirgüç-Kunt, A., & Levine, R. (2009). Finance and inequality: Theory and evidence. Annu. Rev. Financ. Econ., 1(1), 287-318.
10. Jalilian H., Kirkpatrick C. (2001). Financial Development and Poverty Reduction in Developing Countries. Finance and Development Research Programme. Working Paper Series
11. Alemayehu Geda, Abebe S. and Daniel Z. (2006). Finance and Poverty in Ethiopia. A Household Level Analysis. Research Paper No. 2006/51 United Nations University.
12. World Bank (2018). World Bank country profile of 2018
13. National Bank of Ethiopia (2018/19). Annual Report
14. Tita, A. F., & Aziakpono, M. J. (2017). The effect of financial inclusion on welfare in sub-Saharan Africa: Evidence from disaggregated data. Economic Research Southern Africa, Working Paper, 679.
15. Appiah, M., Frowne, D. I., & Tetteh, D. (2020). Re-examining the nexus between financial development and poverty reduction: evidence from emerging economies. Applied Economics Journal, 27(2), 125-144.
16. Honohan, P. (2004). Financial development, growth and poverty: how close are the links?. In Financial development and economic growth: Explaining the links (pp. 1-37). London: Palgrave Macmillan UK.
17. Greenwood, J., & Jovanovic, B. (1990). Financial development, growth, and the distribution of income. Journal of political Economy, 98(5, Part 1), 1076-1107.
18. Kuznets, S. (1955). Economic Growth and Income Inequality.American Economic Review, 45, 1–28
19. Levine, R. (2005). Finance and growth: theory and evidence.Handbook of economic growth, 1, 865-934.
20. Claessens, S., and E. Feijen. 2006. Financial Sector Development and the Millennium Development Goals. World Bank Working Paper No. 89, World Bank, Washington, DC.
21. McKinnon, R. I. (1973). Money and Capital in Economic Development. Washington, D.C.
22. Banerjee, A. V., & Newman, A. F. (1993). Occupational choice and the process of development. Journal of political economy, 101(2), 274-298.
23. Galor, O., & Zeira, J. (1993). Income distribution and macroeconomics. The review of economic studies, 60(1), 35- 52.
24. Demirguc-Kunt, A., Klapper, L., Singer, D., Ansar, S., & Hess, J. (2017). Measuring financial inclusion and the fintech revolution. The Global Findex Database.
25. Zhuang, J., Gunatilake, H. M., Niimi, Y., Khan, M. E., Jiang, Y., Hasan, R., ... & Huang, B. (2009). Financial sector development, economic growth, and poverty reduction: A literature review. Asian Development Bank Economics Working Paper Series, (173).
26. Aghion, P., & Bolton, P. (1997). A theory of trickle-down growth and development. The review of economic studies, 64(2), 151-172.
27. Fields, G. (2001). Distribution and Development: A New Look at the Developing World. NY: Russell Sage Foundation, and Cambridge, MA: MIT Press.
28. King, R. G., & Levine, R. (1993). Finance, entrepreneurshipand growth. Journal of Monetary economics, 32(3), 513-542.
29. Claessens, S., & Perotti, E. (2007). Finance and inequality: Channels and evidence. Journal of comparative Economics, 35(4), 748-773.
30. Shaw, E. S. (1973). Financial deepening in economic development.
31. Naceur, M. S. B. (2016). Financial Development, Inequality and Poverty. International Monetary Fund.
32. Bourguignon, F. (2004). The poverty-growth-inequality triangle (No. 125). Working paper.
33. Dollar, D., & Kraay, A. (2002). Growth is Good for the Poor.Journal of economic growth, 7, 195-225.
34. Ogbeide, E. N. O., & Agu, D. O. (2015). Poverty and income inequality in Nigeria: any causality?. Asian economic and financial review, 5(3), 439.
35. Ray, D. (1998). Development economics. Princeton University Press.
36. Abachi, N. (2007). Ethiopia MFI Country Scan. MicroNed.
37. World Bank (2014). Ethiopia Financial Sector Development:The Path to an Efficient Stable and Inclusive Financial Sector,World Bank.
38. Prosperity, S. (2016). Taking on Inequality. Washington, DC: World Bank.
39. World Bank (2017). World Development Indicators.
40. Mulatu, E., & Wamatu, J. (2020). Ethiopian Micro-FinanceLandscape Report.
41. Dewi, S., Majid, M. S. A., & Kassim, S. (2018). Dynamics of financial development, economic growth, and poverty alleviation: The Indonesian experience. South East European Journal of Economics and Business, 13(1), 17-30.
42. Jalilian, H., & Kirkpatrick, C. (2005). Does financial development contribute to poverty reduction?. Journal of development studies, 41(4), 636-656.
43. Uddin, G. S., Shahbaz, M., Arouri, M., & Teulon, F. (2014). Financial development and poverty reduction nexus: A cointegration and causality analysis in Bangladesh. Economic modelling, 36, 405-412.
44. Sehrawat, M., & Giri, A. K. (2016). Financial development and poverty reduction in India: an empirical investigation. International Journal of Social Economics, 43(2), 106-122.
45. Enowbi Batuo, M., Guidi, F., & Mlambo, K. (2010). Financial development and income inequality: Evidence from African Countries.
46. Giri, A. K. (2015). Financial development, poverty and rural-urban income inequality: evidence from South Asian countries.
47. Fowowe12, B., & Abidoye, B. (2012). A Quantitative Assessment of the Effect of Financial Development on Poverty in African Countries.
48. Bourguignon, F. (2003). The growth elasticity of poverty reduction: explaining heterogeneity across countries and time periods.
49. Datt, G., & Ravallion, M. (1992). Growth and redistribution components of changes in poverty measures: A decomposition with applications to Brazil and India in the 1980s. Journal of development economics, 38(2), 275-295.
50. Ravallion, M. (1997). Can high-inequality developing countries escape absolute poverty?. Economics letters, 56(1), 51-57.
51. Zellner, A. (1962). Three stage least squares: simultaneous estimation of simultaneous equtions. Econometrica, 30, 63- 68.
52. Green, H. (2012). Econometric Analysis. 7th InternationalEdition, Pearson, New York University.
53. Lundberg, M., & Squire, L. (2003). The simultaneous evolution of growth and inequality. The economic journal, 113(487), 326-344.
54. Bakhtiari, S., Breunig, R., Magnani, L., & Zhang, J. (2020). Financial constraints and small and medium enterprises: A review. Economic Record, 96(315), 506-523.
55. Ayyagari, M., Demirgüç-Kunt, A., & Maksimovic, V. (2008). How important are financing constraints? The role of finance in the business environment. The world bank economic review, 22(3), 483-516.
56. Barro, R. J. (1996). Determinants of economic growth: A cross-country empirical study.
57. Fields, G. S. (1989). Changes in poverty and inequality in developing countries. The World Bank Research Observer, 4(2), 167-185.
58. Deininger, K., & Squire, L. (1996). A new data set measuring income inequality. The World Bank Economic Review, 10(3), 565-591.
59. Abosedra, S., Shahbaz, M., & Nawaz, K. (2016). Modeling causality between financial deepening and poverty reduction in Egypt. Social Indicators Research, 126, 955-969.
60. Acemoglu, D., Johnson, S., & Robinson, J. A. (2005). Institutions as a fundamental cause of long-run growth. Handbook of economic growth, 1, 385-472.
61. Geda, A. (2006). The structure and performance of Ethiopia’s financial sector in the pre-and post-reform periods, with a special focus on banking. In Domestic Resource Mobilization and Financial Development (pp. 163-202). London: Palgrave Macmillan UK.
62. Bazillier, R., & Hericourt, J. (2017). The circular relationship between inequality, leverage, and financial crises. Journal of Economic Surveys, 31(2), 463-496.
63. Beck, T., Demirguc-Kunt, A., & Levine, R. (2004). Finance,inequality, and poverty: Cross-country evidence.
64. Jeanneney, S. G., & Kpodar, K. (2008). Financial development and poverty reduction: can there be a benefit without a cost?.
65. Kumhof, M., Rancière, R., & Winant, P. (2015). Inequality, leverage, and crises. American economic review, 105(3), 1217-1245.
66. Planning and Development Commission (2018). Poverty and Economic growth in Ethiopia: 1995/96-2015/16
67. Seven, U., & Coskun, Y. (2016). Does financial development reduce income inequality and poverty? Evidence from emerging countries. Emerging Markets Review, 26, 34-63.

Appendix

Table 4.1: Dickey-Fuller Test for Unit Root

Variables	p-value for Z(t)	Test Statistic	1% Critical Value	5% Critical value	10% Critical value
Pov	0.0485	-3.422	-4.352	-3.588	-3.233	Trend regress lags(0)
GDPg	0.0008	-4.652	-4.352	-3.588	-3.233	Trend regress lags(0)
Gini	0.0024	-3.082	-2.479	-1.706	-1.315	Drift reg log(0
FD	0.0154	-2.283	-2479	-1.706	-1.315	Drift reg log(0
Infl	0.0004	-4.827	-4.352	-3.588	-3.233	Trend regress lags(0)
Tele		3.347	-2.655	-1.950	-1.601	Noconstant reg lags(0)
Open	0.0090	-2.523	-2.479	-1.706	-1.315	Drift reg log(0)
GoEx*	0.0000	-5.360	-3.736	-2.994	-2.628	*At first difference
HK*	0.0289	-3.070	-3.736	-2.994	-2.628	*At first difference
PPg	0.0094	-2.520	-2.492	-1.711	-1.318	Drift reg log(0)
Inst	0.0115	-2.418	-2.479	-1.706	-1.315	Drift reg log(0)
GGDP2	0.0310	-3.587	-4.352	-3.588	-3.233	Trend regress lags(0)

            Source: Author, Results from Stata, Time Series Data

          Table 4.2: Short Run Regression Result of the 3SLS Model

Equation	Obs	Parms	RMSE	“R-Sq”	Chi2	P
Poverty	29	6	3.53e-09	1.0000	2.28e+18	0.0000
Growth	29	7	9.73e-08	1.0000	1.18e+17	0.0000
Inequality	29	5	2.009075	0.8807	227.70	0.0000

	Inst	-7.302711	3.021052	-2.42	0.016	-13.22386	-1.381559
	GGDP2	.0003811	.0116614	0.03	0.974	-.0224749	.0232371
	e3	-.0115952	.0141946	-0.82	0.414	-.0394162	.0162257
	Cons	40.36789	3.725613	10.84	0.000	33.06582	47.66996
Endogenous variables: Pov GDPg Gini
Exogenous Variables: FD Tele PPg e1 Infl Open GoEx HK e2 Inst GGDP2 e3

Source: Author’s computation from Stata

Table 4.4: Correlation Matrix from Stata Command for Each Equation of the Model

Test for collinearity of equation #1

	GDPg	Gini	FD	Tele	PPg
GDPg	1.0000
Gini	-0.4262	1.0000
FD	0.3026	-0.9073	1.0000
Tele	0.3211	-0.0654	-0.0987	1.0000
PPg	-0.5519	0.7801	-0.6521	-0.5927	1.0000

Test for collinearity of equation #2

	GDPg	Gini	FD	Infl	Open	GoEx	HK
GDPg	1.0000
Gini	-0.4262	1.000
FD	0.3026	-0.9073	1.0000
Infl	-0.0925	0.0924	-0.1209	1.0000
Open	0.5626	-0.7711	0.5871	0.2653	1.0000
GoEx	0.4884	-0.3762	0.2083	0.1913	0.4937	1.0000
HK	0.5131	-0.3656	0.1205	0.3798	0.5883	0.8956	1.0000

Test for collinearity of equation #3

	Gini	GDPg	FD	Inst	â?»GGDPâ?»^2
Gini	1.0000
GDPg	-0.4262	1.0000
FD	-0.9073	0.3026	1.0000
Inst	-0.7539	0.4478	0.6254	1.0000
GGDP2	0.3146	0.0081	-0.3616	-0.3672	1.0000

Source: Author’s computation from Stata