Relationship Between Financial Inclusion, Banking Stability And Economic Growth- A Dynamic Panel Approach2021.Pdf

JEAS 05 2021 0084 proof 1 16 Relationship between financial inclusion, banking stability and economic growth a dynamic panel approach Richard Boachie, Godfred Aawaar and Daniel Domeher Accounting and[.]

Relationship between financial

inclusion, banking stability and

economic growth: a dynamic

panel approach Richard Boachie, Godfred Aawaar and Daniel Domeher

Accounting and Finance, Kwame Nkrumah University of Science and Technology,

Kumasi, Ghana

Abstract

Purpose – The purpose of this paper is to analyse the relationship between financial inclusion, banking

stability and economic growth in sub-Saharan African countries given the interconnectedness between them.

Globally, financial inclusion has gained recognition as a critical channel for promoting economic growth by

bringing a large proportion of the unbanked population into the formal financial system This cannot be

achieved exclusive of the banking sector.

Design/methodology/approach – This paper focussed on 18 countries in sub-Saharan Africa Data on

financial inclusion and the economy were obtained from the World Bank, and bank soundness indicators data

were also obtained from International Monetary Fund covering the 11-year period from 2008 through 2018.

Panel system generalised method of moments is employed for the regression analysis because it has the

capability to produce unbiased and consistent results even if there is endogeneity in the model.

Findings – The results show that economic growth drives banking stability and not vice versa; confirming a

unidirectional causality from gross domestic product to banking stability So, this study finds support for the

demand-following hypothesis The paper further observed that financial inclusion positively and significantly

influences the stability of banks and economic growth The study established that bank capital regulation

negatively influences banking stability in sub-Saharan African countries.

Research limitations/implications – This study does not capture the unique country-specific relationship.

Practical implications – The policy implication is that policymakers in sub-Saharan African countries

should focus on growth-enhancing policies that improve the level of financial inclusion The central banks in

sub-Saharan African countries should take advantage of the positive effect of financial inclusion to develop

regulatory frameworks and policies that make it attractive for banks to continue to expand their operations to

the unbanked.

Originality/value – This is, as far as the authors know, the explanation of the interconnection of financial

inclusion, banking stability and economic growth in sub-Saharan Africa.

Keywords Financial inclusion, Banking stability, Economic growth, GMM, Sub-saharan Africa

Paper type Research paper

Introduction

Globally, financial inclusion (FI) has gained recognition as a means of drawing the unbanked

population into the formal financial sector to promote economic growth The role of FI has

become increasingly crucial for every economy especially emerging, frontier and developing

countries in sub-Sahara Africa (SSA) Following the work ofAbor et al (2018), this study

defines FI or inclusive finance as the case where quality financial products are accessible to

citizens most conveniently and cost-effectively through established policies and regulatory

frameworks that safeguard the users at all times The financial system has made concerted

efforts to draw those excluded financially to have access to a wide range of services through

FI initiatives

The expansion of access to finance has three potential benefits First, FI neutralises the

barrier to socio-economic growth which can be hampered by financial exclusion (

empirically noted to have a devastating consequence on the economy because the financial

Banking stability and economic growth

The current issue and full text archive of this journal is available on Emerald Insight at:

https://www.emerald.com/insight/1026-4116.htm

Received 24 May 2021 Revised 19 July 2021 Accepted 26 July 2021

Journal of Economic and Administrative Sciences

© Emerald Publishing Limited

1026-4116

infrastructure may be obstructed even though it is a fundamental pillar for growth (Diamond

services conveniently in a cost-effective way (Kim, 2016) This enables the poor in society to safely keep their funds away from the norm of housing it, helping them escape the risks of economic shocks (Kim et al., 2018) Third, FI helps in achieving a multiplier effect on the economy through credit creation from the pooled savings from a large segment of the unbanked population leading to an increase in economic activities and employment (Koku,

2015;Balach et al., 2016)

However, FI also provides a pathway for banks to achieve inclusiveness especially when these banks are stable The interconnectedness between FI and banking stability thus becomes apparent When a greater number of people remain unbanked or excluded financially, predicting banking stability becomes difficult and on the other hand, achieving a higher level of FI becomes difficult if banks are unstable (Khan, 2011) An important question that arises is“does expanding access to formal financial services work in conjunction with policies aimed at improving bank stability or threaten bank stability?” These are two divergent schools of thought

The first school of thought argues that the FI–bank stability nexus benefits the economy

in two ways First, inclusive finance provides a more stable deposit base for banks Second, financial inclusiveness improves bank stability through process improvement in executing intermediary functions These benefits, according to the proponents, are consistent with the view that financial integration systems enhance bank stability (Khan, 2011;Ryan et al., 2014;

2018), and inclusive finance reduces the unnecessary risk taken by individual banks The other school of thought also proposed that getting a greater percentage of the population included in the financial system is negatively related to banking stability They adduced two major reasons to support their argument First, inclusive finance reduces the quality of your credit portfolio, because the provision of credit to low-income markets will be affected by a wide range of information asymmetries related to customers who have no credit history or collateral; and second, the expansion of the financial channels can increase the credit risk and, above all, the concentration and liquidity risk This position proposes that inclusive finance has a negative influence on banks (Khan, 2011;Allen et al., 2013) Empirically, studies on FI so far have typically focussed on identifying the determinants and measures of inclusive finance (Demirguc-Kunt and Klapper, 2012;Allen et al., 2016;

on the impact of FI either on income, income inequality or poverty reduction (Inoue and

established a positive impact of FI on economic growth (Inoue, 2019) and other studies support this position (Morgan and Pontines, 2014;Kim, 2016;Lenka and Sharma, 2017;Abor

et al., 2018;Sethi and Acharya, 2018;Kim et al., 2018;Tang et al., 2019;Sethi and Kumar, 2019;

Furthermore, the study ofAnarfo et al (2019)changed the narrative and constructed the

FI index using principal components analysis (PCA) on two dimensions of FI– access and usage However, penetration and mobile money deepening factors were not fully factored in This impacted the results to be weak in application to SSA countries This study provides a robust composite index using three dimensions of FI variables– penetration, access and usage as indicated inAppendix 1

Earlier, studies have established an interactive nexus of banking stability and FI (Morgan

the cross-country and sector level A few of them focussed on the banking sector and JEAS

employed single or multiple banking stability proxies not banking stability index hence,

creating a contextual gap The most adopted proxy for banking stability is the bank z-score

The challenge is that the bank z-score is used to measure the stability of individual banks

Adopting it for a panel of banks or countries is liable to produce bias and weak results This

paper offers a solution by developing a composite index for banking stability using four key

dimensions of banking soundness– capital adequacy, asset quality, earnings/profitability

and liquidity

Furthermore, these previous studies on FI acknowledge the role of the banking sector in

the FI agenda, yet little is known about how FI interacts with the soundness of banks and

their composite impact on the economy Given the importance of FI and the intermediation

role played by banks and the existing empirical gaps, this study enriched the existing related

literature by using the multidimensional nature of FI and banking stability to build a

composite index to study the interaction between FI and banking stability and the effect on

the growth of the economy Also, the study incorporated the moderating effect of bank

regulation on the relationship between FI and stability of banks in SSA countries which lacks

or requires more research for policy directions and decisions

Based on the aforementioned contributions to knowledge, this paper achieved three

objectives: (1) investigate the causal (lead–lag) relationship between banking stability and

economic growth in SSA countries; (2) analyse the relationship of FI, banking stability and

economic growth in SSA countries; (3) examine the moderation role of bank capital regulation

on the relationship between FI and bank stability in SSA countries

This study is significant for banking practitioners and inclusive finance policymakers

The study recommends that policy regulatory framework should focus on mobile money

deepening as major delivery channel for financial inclusiveness The central banks, other

regulatory bodies and consultants should carefully examine economic policy directions in

conjunction with bank regulations to ensure the parallel growth of the economy and the

financial system The governments in SSA countries through the central bank should rely on

the study findings in determining policy decisions towards reducing bank-level risk by

reducing information asymmetry needed to access banking services

The remainder of the paper is organised as follows:Section 2discusses the theories that

underpin the study and empirical literature review on the study variables.Section 3also

addresses the methodology and data sources while Section 4addresses the results and

discussions on the study variables The final Section 5 draws out the conclusions and

recommendations of the study

Literature review and theoretical background

FI cannot be achieved in a vacuum without the involvement of banking institutions The

banking sector also operates within an economic environment The two divergent views on

bank stability and economic growth factors have theoretical underpinning We used

demand-following and supply-leading hypotheses to explain the bases for our argument Again,

achieving inclusive finance means large unbanked populations with dynamic characteristics

are being brought into the fold which potentially affects bank level of risk We also used

information asymmetry theory to explain this

In financial resource allocation by banks through debt contracts, there exist situations

whereby one party holds more or better information than the other This presents a challenge

for banks to differentiate between good and bad borrowers Therefore, Akerlof (1970)

developed this theory to deal with the situation In a seminal contribution byRichard (2011),

asymmetric information results in two-dimensional problems: adverse selection and moral

hazard A moral hazard essentially involves behaving in a manner that puts the interest of the

other parties to a contract at risk Adverse selection, on the other hand, assumes that lenders

Banking stability and economic growth

find it difficult to distinguish between borrowers with different levels of risk, thus limiting the performance of credit agreements These two problems create an increase in non-performing loan levels because of a wrong decision which in turn affects the performance and stability of the banks To mitigate the effect of adverse selection and moral hazard, banks insert collateral requirements into their loan contracts that might not be available or sufficient The performance of an economy has something to do with banking or the broader financial system There are two divergent schools of thought regarding the leading influence

in the nexus between the financial system and economic growth This theory propounded by

the other way round This suggests that when growth is weak, projected future demands may decline and will of course affect or suppress investment below the actual limit or speed needed

to realise optimal economic growth The demand-following hypothesis further argues that economic growth drives financial infrastructure development and bank-branch facilities The demand-following framework places emphasis that credit financing and investment spending from domestic institutions generate its savings This theory captures the essence

of economic growth factors causing the generation of savings by the populace in the banking sector The hypothesis is relevant to this research because it captures the essence of economic growth factors causing the generation of its savings which are kept in the banks as deposits This hypothesis is the other school of thought in the financial system and growth nexus The proponentSchumpeter (1911)and supported byGurley and Shaw (1967),McKinnon

driving force of economic growth The key argument for supply-based assumptions is that increased credit or credit deepening is a decisive factor in economic growth The effective arbitration occurring in the banking sector influences the optimal allocation of resources within the economy (Hurlin and Venet, i2008) The supply-based model shows that the causal link goes from finance to economic growth and that there is no feedback on economic growth Improved financial intermediation leads to a reduction in asymmetric information, transaction and monitoring costs In addition, a well-functioning financial sector can enhance the creation of financial services and the accessibility of anticipated demand for real economy participation They posit that finance promotes economic growth and further state that the promotion of economic growth occurs in the early stage of economic development

FI as a strategy has the focus of drawing the unbanked population to access and use formal financial services In examining the relationship with economic growth in developing economies, the study ofInoue and Hamori (2016) focusses on SSA countries from 2004 through 2012 and finds that commercial bank branches have a positive relationship for the selected countries.Balach et al (2016)also supported the findings by focussing on 97 cross-section countries from 2004 through 2012 using two FI proxies to assess the impact on economic growth The study established a positive effect for both proxies and on selected countries being studied Several studies support these findings establishing a positive relationship (Sharma, 2016;Kim et al., 2018;Sethi and Kumar, 2019) Furthermore, the FI and banking stability relationship has empirical underpinning

For instance, the study on SSA byAnarfo et al (2019)investigated the dynamic link between FI and financial sector development which is moderated by economic growth The findings from the study indicate a reverse causality between financial sector development and FI (Anarfo et al., 2019) Again, the study posits that FI is a driver for financial sector development and vice versa In a recent development,Anarfo et al (2019)took into account the impact of the regulatory framework on financial stability and FI in SSA countries and found that financial regulation influences financial stability and FI Several studies find support for this position (Fratzscher et al., 2016;Fern
andez et al., 2016;Owen and Pereira,

2018) Some studies established a negative relationship between FI variables and bank stability (Koong et al., 2017;Le et al., 2019)

JEAS

Moreover, another strand of literature focussed on banking stability and economic growth

relationship between financial stability and economic growth in Africa and employed a

dynamic fixed-effect model approach, and the results reveal that financial stability impacts

positively on economic growth (Manu et al., 2011) In support of the findings,Jokipii and

nexus using the vector autoregression (VAR) methodology approach in 18 OECD countries

for a period spanning from 1980 through to 2008 Several studies find support for a positive

relationship (Puatwoe and Piabuo, 2017;Jayakumar et al., 2018) even though some studies

also find a negative relationship between banking stability and economic growth (Inoue,

2019;Alsamara et al., 2019)

The study tested the hypothesis to establish the relationship between banking stability

and economic growth (Sethi and Acharya, 2018) and to investigate the direction of their

relationship The results of the study will either confirm demand-following or supply-leading

hypotheses propounded by Robinson (1952) and Schumpeter (1911), respectively The

findings inform the SSA countries whether economic growth leads or lags in the relationship

with banking stability We hypothesise as below:

H1 There is a lead–lag relationship between banking stability and economic growth

Another strand of literature focussed on the direction of causality between banking stability

and economic growth (Jayakumar et al., 2018) Several studies find support for a positive

relationship (Puatwoe and Piabuo, 2017;Jayakumar et al., 2018) even though some studies

also find a “negative relationship between banking stability and economic growth” (e.g

between banking stability and economic growth We then hypothesise as below:

H2 There is a bi-causality between banking stability and economic growth

Furthermore, the literature established a relationship between FI and economic growth For

instance, the study ofAndrianaivo and Kpodar (2011)adopted the generalised method of

moments (GMM) approach in a panel of 44 African countries and confirmed the positive and

significant effect of financial inclusion on economic growth.Inoue and Hamori (2016)find that

commercial bank branches have a positive relationship with the economy of the selected

countries.Balach et al (2016)also supported the findings by focussing on 97 cross-section

countries from 2004 through 2012 Several studies find support for these findings (Lenka and

hypothesise as below:

H3 FI significantly influences economic growth

Regarding banking stability and economic growth, the study ofManu et al (2011)examines

the relationship between financial stability and economic growth in Africa and established

that financial stability impacts positively on economic growth (Manu et al., 2011) In support

of the findings,Jokipii and Monnin (2013) confirmed the positive significant influence of

banking stability on economic growth in 18 OECD countries The study ofJayakumar et al

then hypothesise as follows:

H4 Banking stability significantly affects economic growth

We observe in the literature the relationship between banking stability and FI The study of

FI The same study also posits that FI is a driver for financial sector development and vice

Banking stability and economic growth

versa Several studies find support for this position (Fratzscher et al., 2016;Fern
andez et al.,

H5 FI significantly affects banking stability

To avoid arbitrariness in the conduct of banking business, each economy has a regulatory framework to guide and protect stakeholders The strictness of the regulation and its effectiveness have implications for expanded access to finance and achieving banking stability In the spirit ofDev (2006), other literature affirms the positive effect of regulation on stability and FI (Chiwira et al., 2013;Fratzscher et al., 2016;Sethi and Acharya, 2018) In a recent development, Anarfo and Abor (2020) considered the impact of the regulatory framework on financial stability and FI in SSA countries and found that financial regulation influences financial stability and FI Based on this assumption, the hypothesis is tested: H6 Banking regulation significantly moderates the relationship between FI and banking stability

Research methodology

In estimating a panel data set, the GMM has been adjudged to have the ability to avoid biases according toLevine et al (2000) This method is applicable in different statistical analytical circumstances The GMM estimation technique can correct the problem of endogeneity The assumption further shows that when the number of periods (T) of available data may be small and the number of observations (N) may be large, the GMM can produce unbiased results

Yi ;t¼ ∝ þ Xi ;tβ þ ηiþ μi ;t where i¼ 1 n; t ¼ 1 T (1) Based on the assumption of non-serial correlation among the error terms,μi ;tand weakly exogenous explanatory variables,Arellano and Bond (1991)suggest the following moment conditions:

E

ðYi ;t−sÞΔμi ;t



¼ 0; i ¼ 1 n; s ≥ 2; t ¼ 3 T (2)

E

ðXi ;t−sÞΔμi ;t



¼ 0; i ¼ 1 n; s ≥ 2; t ¼ 3 T (3) The two equations above are a two-step GMM estimator recommended byArellano and Bond

The second equation adopts the residual derived from the first equation to run an estimate of the covariance matrix and variance analysis which relaxes the axiom of independent homoscedasticity When these moment conditions are applied in the GMM estimator, the result is efficient estimates Based on the estimation technique, the following general model is employed:

Yit¼ ∝ þ X1itβ1þ X2itβ2þ μi ;t where i¼ 1 n; t ¼ 1 T (4) where Y5 the variable we are trying to predict;α5 the intercept; β1andβ25 the slope (Bun and Windmeijer, 2010); X1, X25 the variables used to predict Y;μ5 the error term The intercept (α) is the value of the dependent variable when the independent variable is equal to zero while the slope of the regression line (β1andβ2) represents the rate of change in Y as X changes (Bun and

GDPit¼ α þ β1FIIitþ β2BSIitþμit (5) JEAS

FIIit¼ αðBSIitÞ þ βXitþμit (6) BSIit¼ αðFIIitÞ þ βXitþμit (7) GDP is gross domestic product

FII is financial inclusion index

BSI is banking stability index

The study draws data from several sources to investigate the relationship between FI,

banking stability and economic growth The FI data set is obtained from the Global Financial

Index (FINDEX) of the World Bank, and the banking stability data set is obtained from the

Global Financial Development Database (GFDD) of IMF Finally, the data for GDP, the proxy

for economic growth is compiled from World Bank economic data This paper targeted 18

countries in SSA for the study These countries have a mixture of income,

lower-middle-income and upper-lower-middle-income countries This suggests that all the characteristics

of countries in SSA are represented Our data cover from 2008 to 2018, thus giving us 198 as

our panel observations The review of the empirical literature shows that economic growth as

the dependent variable is mostly represented by gross domestic product (GDP annual growth

rate) Our independent variables used here are banking stability and FI A detailed

description is found inAppendix 1 We followed empirical work to define the variables used

in this study (such asGhosh, 2010;Sarma, 2012;Kocisov
a, 2014)

Principal component analysis (PCA)

The study adopted a panel PCA estimation technique to construct banking stability and FI

indices made up of eight selected measures of banking stability and FI The PCA results are

presented inAppendixes 2 and 3 The study adopted indexes for banking stability and FI to

provide a common measure for countries in the SSA region and to provide a benchmark or

reference point for future studies using similar variables for individual specific countries

(Ghosh, 2010) In line with this estimation techniques, the ith factor index can be specified for

FIIi¼ Wi1X1þ Wi2X2þ Wi3X3þ þ WipXp (8) where FIIiis the financial inclusion index; Wiis the weight (factor loading) of the parameter of

the factor score; X is the original figure of the respective components, and P is the number of

FI variables in the equation.Eqn (9)below specifies the eight proxies used in the composite

index construction The variables are as follows: Number of Mobile Money Accounts per

100,000 adults (NMMA), Number of Commercial Bank Accounts per 1,000 adults (BAPTA),

Mobile Banking, Registered Agent Outlets Per 100,000 Adults (MBRA), Domestic credit to the

private sector by banks % of GDP (DCTP), Bank branches per i100,000 adults (BBPA),

Automated teller machines (ATMs) per 100,000 adults (ATMPA), Borrowers from

commercial banks per 1,000 adults (BFCB) and Depositors with commercial banks per

1,000 adults (DBPA) The index is specified as follows:

FII¼

Z

ðNMMA; BAPTA; MBRA; DCTP; BBPA; ATMPA; BFCB; DBPAÞ (9)

BSIi¼ Wi1X1þ Wi2X2þ Wi3X3þ þ WipXp (10) where BSIi is the banking stability index; Wi represents the weight of factor score, X

represents the original figure a component and P represents the number of variables used for

the index construction Eight banking stability indicators used are; Regulatory Tier 1 capital

Banking stability and economic growth

to risk-weighted assets (T1CWA), Nonperforming loans net of provisions to capital (NPLPC), Nonperforming loans to total gross loans (NPLGL), Return on assets (ROA), Return on equity (ROE) and Interest margin to gross income (IMGI), Liquid assets to total assets (LATA) and Liquid assets to short-term liabilities (LASL) The index is specified as follows:

BSI¼

Z ðT1CWA; NPLPC; NPLGL; ROA; ROE; IMGI; LATA; LASLÞ (11)

Empirical results and discussion Descriptive statistics

The presentation inTable 1provides a summary of the statistical properties of the variables All the variables have positive mean values and close mean–median scores reflect the approximately normal distribution of the data All the variables are skewed which shows that they are asymmetrical When the skewness is close to 0, then the data is normally distributed

or otherwise FromTable 1, it can be seen that BSI and FII are positively skewed, and this shows that they are asymmetrical Further, GDP and MOD exhibit a negative skewness which implies that they have a long-left tail Again, kurtosis values according toTable 1show that the data is normally distributed except for FII which is deviated from 3

Panel unit root results

We assert that adopting a fitting econometric model for the study variables requires that the panel unit root test is conducted As a first step to the analysis and discussions of the study results, the available data used are analysed for their statistical properties The study employs the Augmented Dickey–Fuller (ADF) test as well as the Phillips–Perron test for individual unit root process andLevin et al (2002)(LLC) test for a common unit process for all the variables The authors assume that the“test performs well when N lies between 10 and 250 and when T lies between 5 and 250” (Baltagi et al., 2007) When the“T is very small, the test is undersized and has low power” This paper has N being 18 and T being 11 which

is suitable for our data set.“Besides, panel data approach gives options for estimation ranging from no trend and non-constant estimations with a constant and deterministic trend testing for similar effects” (Anarfo et al., 2019) thus provides a “greater level of flexibility in computing the coefficients” From our results on panel unit root presented in

Note(s): BSI – banking stability index; FII – financial inclusion index, GDP – gross domestic product, MOD – moderating variable (i.e regulatory capital risk-weighted average)

Table 1.

Results of descriptive

statistics on variables

JEAS

Selection and computation of lag order

The appropriate model is selected based on a technique reported in Table 3 According to

Andrews and Lu (2001), the selection criteria are based on three models and the overall coefficient

of determination, VAR selection at the first-order model is preferred The results presented in

adopted for the study was established as stable (see Appendix:Figure A1) The stability graph

showed that all the roots of AR characteristics were within the circle with no outlier

BSI @Level

PP – Fisher χ 2

FII @Level

GDP @Level

ADF – Fisher χ 2

MOD @Level

Note(s): ** Probabilities for Fisher tests are computed using an asymptotic χ 2 distribution All other tests

assume asymptotic normality

1 48.05719 93.57787* 0.069128* 3.003177* 3.267097* 3.095292*

Note(s): * Indicates lag order selected by the criterion

LR: sequential modified LR test statistic (each test at 5% level)

FPE: Final prediction error

AIC: Akaike information criterion

SC: Schwarz information criterion

HQ: Hannan –Quinn information criterion

Table 2 Panel unit root test

Table 3 VAR lag order selection criteria

Banking stability and economic growth

Discussion of regression results

In investigating the interaction between FI, banking stability and economic growth, the system GMM estimator was adopted (Wang and Lee, 2018) From the results shown in

been presented using VAR Granger Causality/Block Exogeneity Wald Tests The results show from the upper panel ofTable 4that banking stability is influenced by economic growth measured by GDP and not vice versa Thus, economic growth lead and banking stability lag This also supportsAlsamara et al (2019), whose study finds that growth in GDP leads to stability in the banking sector This result finds support for the demand-following hypothesis

The economic implication of this study is that when the economy is performing and indicators demonstrate strength, it creates its own savings capacity where the citizenry keeps the banks as deposits The booming economic activities result in surplus savings out of which the banks create assets to generate interest income Impliedly, the good financial performance

of banks is borne out of the performance of the economy Further, confidence by the population in the banking sector is heightened when the economy booms because the stable deposits by the banks increase banks’ liquidity strength Another detrimental cause of banking instability is the level of non-performing loans (NPLs) and a major cause is attributed

to the state of the economy Thus, a lower NPL is a single reflection of a sound banking system This study supports that growth-enhancing policies should be the optimal focus for economic policymakers to ensure the long-term stability of financial intermediaries Further inTable 5, we have presented the Panel Granger Causality tests between GDP and BSI to explain the direction of causality The p-values show that GDP Granger causes BSI and not vice versa This means there is a unidirectional relationship between GDP and BSI This study supports the findings ofZang and Kim (2007)which examine the causal link between financial sector development and economic growth in East Asian countries and established strong evidence that economic growth precedes financial development These findings are the test results forH2

Moreover, the GMM regression estimator is adopted to test the interactive relationship between the three studied variables The results of the GMM inTable 6have shown a positive significant relationship between the FII and economic growth (GDP) of 18 SSA countries with other variables held constant The coefficient of FII with GDP as the dependent variable

Dependent variable: BSI Excluded

Dependent variable: GDP Excluded

Table 4.

VAR Granger

causality tests

Table 5.

Panel Granger

causality tests

JEAS