University of Ghana http://ugspace.ug.edu.gh UNIVERSITY OF GHANA COLLEGE OF HUMANITIES FINANCIAL INCLUSION ON POVERTY ALLEVIATION IN GHANA: ASSESSING THE IMPACT OF MOBILE MONEY AND BANKING BY KWAKU ABOAGYE ACQUAAH (10161364) A LONG ESSAY SUBMITTED TO THE DEPARTMENT OF FINANCE, UNIVERSITY OF GHANA BUSINESS SCHOOL, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF MASTER OF SCIENCE (MSC) DEGREE IN DEVELOPMENT FINANCE JULY, 2019 University of Ghana http://ugspace.ug.edu.gh DECLARATION I hereby declare that apart from references to other people which have been duly cited, this project work is the result of my own work, and that it has neither in the whole nor in part been presented elsewhere. ................................................... ……….......................................... Kwaku Aboagye Acquaah Date (10161364) i University of Ghana http://ugspace.ug.edu.gh CERTIFICATION I hereby declare that the preparation and presentation of this project work was supervised by me in accordance with the guidelines on supervision on project work laid down by the University of Ghana. …………………………………… ……………………..……..…………….. Dr. Agyapomaa Gyeke-Dako Date (Supervisor) ii University of Ghana http://ugspace.ug.edu.gh DEDICATION This work is dedicated to the Almighty God for his unmerited favor towards me. I also dedicate it to my dear late father, Mr. Emmanuel Kwame Acquaah, who passed away on July 22, 2019. Your memories will forever live with us. ......................................................... ……………………………………….. Kwaku Aboagye Acquaah Date (10161364) iii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENT First and foremost, I will like to express my profound gratitude to God for His abundant grace that enabled me to successfully carry out this research project. This would not have been possible if not for the support I received from my late father, Mr. Emmanuel K. Acquaah, my mother Ms. Dina Abuyaa and my siblings Juliana and William. You always encouraged me to focus, be diligent and resolute in all I do. To my wife, Regina, your unflinching support, encouragement, patience and prayers, was immense. Your counsel in difficult times always pointed me to the right places. I also wish to acknowledge my supervisor, Dr. Agyapomaa Gyeke-Dako for her patience, encouragement and guidance over the course of this research work. To my friends, Aaron, Perry, Alberta, Mary-Ann and Debbie, I say thank you. You were an amazing support system for me during one of my difficult moments in life. To my dear friend, Angela Hlalele, thank you. You are an amazing friend with a beautiful heart. I do not take your friendship for granted. Thank you for the support always. Lastly, I will like to thank all my lecturers, colleagues at work and friends who contributed in one way or another towards the completion of this research work. iv University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION....................................................................................................................... i DEDICATION........................................................................................................................ iii ACKNOWLEDGEMENT ...................................................................................................... iv TABLE OF CONTENTS ........................................................................................................ v LIST OF TABLES ............................................................................................................... viii ABSTRACT ............................................................................................................................. ix CHAPTER ONE ...................................................................................................................... 1 INTRODUCTION.................................................................................................................... 1 1.1 Background of the Study .......................................................................................... 1 1.1.1 Poverty Alleviation............................................................................................. 1 1.1.2 Financial Inclusion ............................................................................................. 2 1.2 Problem Statement .................................................................................................... 2 1.3 Aim and Objectives ................................................................................................... 3 1.4 Research Question ..................................................................................................... 3 1.5 Scope of the study ...................................................................................................... 4 1.6 Significance of the Study........................................................................................... 4 1.7 Chapter Disposition .................................................................................................. 5 CHAPTER TWO ..................................................................................................................... 7 LITERRATURE REVIEW .................................................................................................... 7 2.1 Introduction ............................................................................................................... 7 2.2 Theoretical Framework ............................................................................................ 7 2.2.1 The Trickle-Down Theory ................................................................................. 8 v University of Ghana http://ugspace.ug.edu.gh 2.2.2 Economic Model of Mobile Money Services .................................................... 9 2.3 Review of Related Studies....................................................................................... 11 2.3.1 Financial Inclusion and Poverty Alleviation ................................................. 11 2.3.2 Mobile Money Penetration and Poverty Alleviation .................................... 15 2.4 Summary of Literature Review ............................................................................. 18 METHODOLOGY ................................................................................................................ 20 3.1 Introduction ............................................................................................................. 20 3.2 Research Design....................................................................................................... 20 3.3 Variables in the Study ............................................................................................. 21 3.4 Source of Data.......................................................................................................... 22 3.4.1 The Global Findex Database ........................................................................... 22 3.4.2 Inter-University Consortium for Political and Social Research (ICPSR) Database .......................................................................................................................... 23 3.5 Data Analysis ........................................................................................................... 24 3.5.1 Panel Data Models ........................................................................................... 24 3.5.2 Model Specification .......................................................................................... 26 3.6 Ethical Considerations ............................................................................................ 29 ANALYSIS, RESULTS AND DISCUSSION ...................................................................... 31 4.1 Introduction ............................................................................................................. 31 4.2 Demographic characteristics of participants ........................................................ 31 4.3 Descriptive Analyses ............................................................................................... 32 4.3.1 Descriptive Statistics ........................................................................................ 33 4.3.2 Correlation Analyses ....................................................................................... 35 4.4 Results of financial inclusion as determinants of poverty levels ......................... 37 vi University of Ghana http://ugspace.ug.edu.gh 4.4.1 Financial inclusion and savings ...................................................................... 37 4.4.2 Financial inclusion and access to finance....................................................... 41 4.4.3 Financial inclusion and cashless transactions ............................................... 44 4.5 Synthesis of the findings ......................................................................................... 48 4.6 Chapter Conclusion ................................................................................................ 50 SUMMARY, CONCLUSION AND RECOMMENDATIONS ......................................... 51 5.1 Introduction ............................................................................................................. 51 5.2 Summary of Findings .............................................................................................. 52 5.3 Implications of the Study ........................................................................................ 53 5.3.1 Practical implications of the findings ............................................................. 53 5.3.2 Theoretical implications of the findings ......................................................... 54 5.4 Limitations of the study .......................................................................................... 55 5.5 Recommendations for Future Research ................................................................ 56 5.6 General Conclusion ................................................................................................. 57 REFERENCES ....................................................................................................................... 58 vii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 1: Description of variables in the study ......................................................................... 28 Table 2: Socio-demographic characteristics of participants .................................................... 32 Table 3: Descriptive statistics of scores ................................................................................... 34 Table 4: Correlation matrix showing relationships among the variables ................................ 36 Table 5: Financial inclusion as determinant of savings ........................................................... 38 Table 6: Financial inclusion as determinant of access to finance ............................................ 42 Table 7: Financial inclusion as determinant of cashless transactions ...................................... 45 viii University of Ghana http://ugspace.ug.edu.gh ABSTRACT This study examined the impact of mobile money and banking services on poverty alleviation in Ghana. In recent years, mobile money and banking services are argued to increase financial inclusion, especially in developing economies. The penetration of these services is therefore also theorized to reduce poverty significantly. However, despite the increasing penetration of banking and mobile money services in Ghana, there are limited studies assessing how the two impact on poverty alleviation. There is therefore the need for more empirical research to be conducted to provide evidenced-based information to guide financial inclusion policy regulations and practice. The current study therefore empirically examining how mobile money and banking services have affected poverty reduction between 2004 and 2017. Poverty level were measures as savings culture, access to finance and cashless transactions. Secondary data from Global Financial Index Database (also called Global Findex) and the Inter-University Consortium for Political and Social Research (ICPSR) were used, involving a sample of 832 participants. Findings showed that access traditional banking improved savings culture and access to finance. Specifically, rural bank and microfinance increased savings and access to finance among low income individuals. Mobile money services increased cashless transactions among the individuals. Together, these findings suggest that financial inclusion strategies have differential effects on poverty level among low income individuals, which calls for more nuanced policy initiatives. Keywords: financial inclusion, mobile money, traditional banking, poverty reduction, Ghana ix University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.1 Background of the Study Poverty levels still remain the most critical developmental challenge facing many developing countries. In Sub-Saharan Africa, poverty levels are predicted to remain in double digits by 2030, absent significant shifts in policy (World Bank, 2013). Majority of people in Africa leave below the poverty line, which is measured in international terms as the minimum line of $1.90/day (World Bank, 2018). Ghana is no exception to the challenge of poverty being a development problem. Even though Ghana is seen as a middle-income country, poverty levels are still high, putting Ghana within the lower middle-income bracket. 1.1.1 Poverty Alleviation The need to reduce if not eliminate poverty is therefore critical to the advancement of improving living conditions in Ghana, and other developing countries. It is for this reason that the United Nations (UN) Sustainable Development Goals (SDGs) has No Poverty as its number one agenda. The SDG 1 aims to alleviate the suffering of forms and build the resilience of those individuals still living in extreme poverty, in particular, sub-Saharan Africa. The efforts made by the global community in addressing these limitations led to the financial inclusion drive. In 2015 for instance, The World Bank Group and public and private sector partners committed to making financial access universally accessible. This led to the establishment of the Universal Financial Access (UFA) 2020 Initiative. The goal of the UFA 2020 is to bring financial services to 1 billion people globally (World Bank, 2015). 1 University of Ghana http://ugspace.ug.edu.gh 1.1.2 Financial Inclusion Financial Inclusion is now recognized as an important part of the mainstream thinking on economic development. Financial Inclusion is defined as the making of financial services or activities readily and easily available to not just those at the top of the economic pyramid, but to low income households, family businesses and individuals (Williams, Adegoke & Dare, 2017). It encompasses driving ease of accessibility and availability of the formal financial services, such as bank deposits, credits, insurance, etc., for all participants in an economy (Demirguc-Kunt, Klapper & Singer, 2017). The fundamental aim of financial inclusion is to get the unbanked and underbanked individuals, better access to financial services in order to drive individual and economic growth (Park & Mercado, 2016). 1.2 Problem Statement Various countries in Africa have concentrated on increasing access to finance, especially among low income individuals (Clamara, Ximena & Tuesta, 2014). In Ghana for instance, there have been significant growth both in traditional banking services and increasing penetration of mobile money services. Banking institutions (both commercial and rural, as well as microfinance firms) have increasingly been opening branches in low income areas (Yakubu, Dinye, Buor, & Iddrisu, 2017). The telecommunication companies have also focused on expanding access to mobile money services to all individuals, including those in low income areas, even where there is limited access to traditional banking services (Dzokoto & Appiah, 2014; Koomson & Ibrahim, 2018). Financial inclusion in developing economies is actually argued to be driven for instance by mobile money accounts which exceeds 10% (World Bank, 2018). 2 University of Ghana http://ugspace.ug.edu.gh These developments notwithstanding, surveillance studies or research on financial inclusion and poverty levels in Ghana have not paralleled the enthusiasm for expanding access to finance among low income individuals. Majority of the studies conducted in Ghana have focused on examining factors that influence the uptake of financial inclusion services and innovations. There are therefore limited studies on examining the direct impact of financial inclusion (defined in terms of access to banking and mobile money services) on poverty alleviation. Also, a direct comparison of the of traditional banking expansion and mobile money penetration on indicators of poverty are missing in Ghana and the broader literature on Africa. This study therefore sought to assess the role financial inclusion is playing in reducing poverty levels in Ghana. 1.3 Aim and Objectives The study sought to empirically test the effect of financial inclusion on poverty levels in Ghana. The specific objectives of the study were to: i. Examine the effect of growth in traditional banking services on poverty levels in Ghana ii. Assess the impact of growth in mobile money services on poverty levels in Ghana iii. Examine the comparative impact of traditional banking and mobile money services on poverty levels in Ghana. 1.4 Research Question Based on the specific objectives above, the following research questions were answered: 3 University of Ghana http://ugspace.ug.edu.gh i. How has the growth in traditional banking services affected poverty reduction indicators in Ghana? ii. How has the growth in access to mobile money services affected poverty level indicators in Ghana? iii. What has been the comparative impact of traditional banking and mobile money services on poverty levels in Ghana? 1.5 Scope of the study The main variables in the study are financial inclusion and poverty reduction. Indicators of financial inclusion are the independent variables and indicators of poverty reduction are the dependent variables. For the purpose of this study, financial inclusion was defined as the expansion of traditional banking services (encompassing growth in activities of commercial banks, rural banks and microfinance institutions) and penetration of mobile banking services (encompassing growth in activities of mobile money operations and other electronic payment systems). Poverty levels were defined using three indicators: savings culture, access to finance and credits, and use of cashless transactions. 1.6 Significance of the Study The outcomes from the current study are significant to both practice and theory. As regards the practical purposes, the outcomes of the current study offer in-depth analysis and knowledge 4 University of Ghana http://ugspace.ug.edu.gh concerning the direct effects of financial inclusion efforts on poverty alleviation among low income individuals in Ghana. The findings also guide development agencies and relevant stakeholders on decisions regarding expanding access to finance among low income individuals in Ghana. These all therefore contribute to achieving the SDG 1 in Ghana. Concerning the theoretical significance, the findings from the study contribute wealth of knowledge to the literature on financial inclusion and poverty alleviation. This contributes empirical evidence towards the literature and to building robust theories that drive the financial inclusion and poverty alleviation agenda. 1.7 Chapter Disposition This study is presented in five main chapters. The details of each of the chapters are described below: Chapter One discusses the introduction to the study. The sections discussed in the chapter are: background of the study, statement of the problem, objectives and research questions; significance of the study, scope and limitations. Chapter Two focuses on the literature review to present information concerning the definition of terms and concepts, theoretical framework of the study and review of related empirical studies to situate within the context of financial inclusion and poverty alleviation. Chapter Three discusses research methodology of the study. The sections of the methodology discussed in the chapter include: research design, variables in the study, source of data, 5 University of Ghana http://ugspace.ug.edu.gh procedures for data gathering, ethical considerations, data analysis techniques and summary of the chapter. Chapter Four presents covered data analysis and discussion of findings. The subsections discussed in the chapter five include: demographic profiles of participants, descriptive statistics of scores, correlation analysis, model testing, synthesis of the findings and chapter conclusion. Chapter Five covers the summary, conclusion and recommendations. The sections discussed in this chapter include: summary of the research findings, implications of the study, suggestions for future research, conclusion and recommendations. 6 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERRATURE REVIEW 2.1 Introduction This chapter places the current study within the literature on poverty alleviation and to what extend it is influenced by financial inclusion. The chapter essentially highlights the theories and models used as a framework for the study, appraisal of other related studies and ends with a summary of the literature. 2.2 Theoretical Framework Theoretical framework plays significant roles in research. It guides the conduct of the entire research process. The fundamental unit of theoretical frameworks is a theory or a model (Sayed & Ghalib, 2015). A theory is defined as a system of ideas that are organized to explain a phenomenon (Lal, 2018). Within the context of research, theories shape how ideas and concepts are organized together to achieve the overarching aim or purpose of a study (Miled & Rejeb, 2015). In this sense, theories are formulated to help in explaining, predicting and understanding or even challenging assumptions in pre-existing knowledge in order to broaden the boundaries of knowledge (Lal, 2018). Theoretical framework is, therefore, created by synthesizing the arguments and assumptions of more than one theory or model to provide a broader context for guiding research process (Boukhatem 2016). According to Lal (2018), a theoretical framework acts as a structure that supports the entire research process. The current study is situated within the context of financial inclusion,poverty reduction and alleviation in developing countries. There is therefore the need 7 University of Ghana http://ugspace.ug.edu.gh to develop theoretical frameworks that allow to examine the dynamics in Ghana, which is a developing country itself. Two main models are drawn on in establishing the current study’s theoretical framework. The two are: the trickle-down theory and the economic model of mobile money services. The trickle- down theory provides broader context of how financial inclusion in its general sense trickles down to empower low income individuals and households economically. The economic model of mobile money explains how mobile money leads to economic empowerment. 2.2.1 The Trickle-Down Theory The relationship between poverty reduction and financial development has been explained using the trickle-down theory, proposed by Aghionand Bolton (1997). This theory is therefore to be used as the framework for this study to investigate how financial inclusion measures at the macroeconomic or structural-level come down to affect poverty levels at the individual level in the long run. The trickle-down theory was developed in economics and it is therefore sometimes referred to as the trickle-down economics (Inoue, 2018; Sethi & Acharya, 2018). From its roots in economics, trickle-down economics basically argued that structural-level economic policies such as taxes on business activities and incomes of wealthy people in the long run, trickles down to affect low income masses (Sethi & Acharya, 2018). Over the years, trickle-down economics began to be applied in different fields and different contexts. In its current state, the trickle-down theory has begun to be applied in financial inclusion and poverty reduction studies (Inoue, 2018; Rewilak, 2017). The trickle-down theory proposes that financial development contributes to poverty reduction in three main ways. First, financial 8 University of Ghana http://ugspace.ug.edu.gh development improves access of low-income individuals to formal finance by reducing credit constraints that impede the flow of capital to low income individuals (Boskov, 2018). Secondly, financial development enables low income individuals to access accumulated savings or borrow money to start microenterprises, which in turn leads to income generation (Boskov, 2018). Lastly, financial development may trickle down to the poor through its influence on economic growth (Burlando & Canidio, 2015). This theory therefore implies that there will be a direct positive link between financial inclusion and poverty reduction. The theory is therefore used as a framework to examine the impact of financial inclusion on poverty levels in Ghana between 2011 and 2017. 2.2.2 Economic Model of Mobile Money Services Operating mobile money businesses in Ghana, as in the case of all developing countries where there are mobile money services, is guided by broad institutional policies that define the legal framework, market structure and requirements for licensing as a mobile money operator. Within the context of the current study, the review of this model is limited to mobile money services as a conduit to financial inclusion. The model is therefore discussed in relation to the formal and informal elements of mobile money in encouraging financial inclusion. The market structure of mobile money transactions in Ghana is characterized by five key stakeholders: regulatory bodies, telecommunication operators, banks and merchants or mobile money operators and mobile money subscribers (Boateng, 2011). The regulatory bodies are the institutions (mainly Bank of Ghana) that regulate and license mobile money transactions to the telecommunication companies. The telecommunications companies are the four main operator 9 University of Ghana http://ugspace.ug.edu.gh institutions (MTN, Vodafone, Airtel and Tigo) that provide mobile money services to the general public. The telecommunication institutions engage the services of middlemen, referred to as merchants or mobile money operators who carry out the mobile money transactions to the general public. Recently, some network companies (e.g. MTN) have also engaged the services of some banks in the provision of mobile money services to the general public. The last stakeholder in the structure is the mobile money subscribers who receive the services. The present study will concentrate on the merchants or the mobile money operators in examining value creation and risk management in their daily operations. The regulatory body gives licensing or legal framework for business operations to the mobile network operators or the telecommunication companies to operator mobile money transactions. In Ghana, the Central Bank (i.e. Bank of Ghana) is responsible for the provision of the legal operating license to the mobile network operators to engage in mobile money transactions to the general the public. Currently, MTN Mobile Money, Tigo Cash, Vodafone Money and Airtel Money are licensed to be operated by MTN, Tigo, Vodafone and Airtel respectively (Boateng, 2011). The telecommunication companies in turn give licensing to merchants to operate mobile money transfer business to the general public. The mobile network operators give the license to registered small businesses to also broker mobile money services to the general public who cannot go to the offices of the telecom operators to transact business. The telecom companies have also signed business agreements with some banks to provide mobile money services to the public. 10 University of Ghana http://ugspace.ug.edu.gh 2.3 Review of Related Studies This section provides a critical review of empirical studies that have been conducted to examine how various financial inclusion mechanisms are linked to poverty reduction and alleviation in low and middle-income country contexts. Financial inclusion itself has different dimensions, which include access, usage and affordability (Demirguc-Kunt et al. 2017). The review is organized around two main thematic areas, which reflect the objectives of the study. The first theme discusses how financial inclusion in its broader terms is associated with poverty alleviation. The second theme narrows to mobile money penetration and poverty alleviation in particular. Within each of the themes, the studies are discussed within the context of how various dimensions of inclusion (access, usage and affordability) impact on poverty alleviation. 2.3.1 Financial Inclusion and Poverty Alleviation Research on financial inclusion and poverty reduction in general has received appreciable research attention in the last decade. This is evident in the various research project initiatives and a plethora of studies that have examined different elements of financial inclusion and poverty dynamics, especially in low- and middle-income countries. Overview of the literature shows that there are both individual and country level factors that affect financial inclusion (Clamara et al. 2014; Demirguc-Kunt et al. 2017). The structural level studies examine how for instance, putting microfinance structures in place help alleviate poverty. Samer, Majid, Rizal,Muhamad, Sarah-Halim and Rashid (2015) examined how microfinance reduces poverty levels among a convenient sample of 780 low-income households in Malaysia. They found that households who had access to microfinance services were able to earn a steady income overtime. Specifically, women who had access to 11 University of Ghana http://ugspace.ug.edu.gh microfinance services were able to obtain loans and other credit facilities to improve their trading activities compared to those who did not have access to similar microfinance services. Several other longitudinal studies have reported findings to the effect that structural-level financial inclusion development reduces poverty. For instance, Inoue (2018) analyzed longitudinal data to examine the interaction effect of remittances and financial development on poverty alleviation in developing countries. The study involved analyzing a panel data of 120 developing countries over a thirty-year period, between 1980 and 2013. Findings from the study showed that structural level financial development significantly increased the inflow of income to low income individuals to ameliorate their poverty levels. Remittances were also found to increase during the period, significantly bringing poverty levels down in majority of the countries. These findings are corroborated by a cross-sectional study by Miled and Reieb (2015), involving the contribution of 1132 microfinance institutions in 57 developing countries which found that poverty levels are lowest in countries with increasing microfinance services. Similar findings were also reported by Boukhatem (2016), in a similar longitudinal study that examined direct effect of structural level financial development on poverty levels for 67 low and middle-income countries between 1986 and 2012. The study also sought to identify fundamental channels through which financial development trickled down to affect poverty levels. Findings showed that proliferations of different financial services which aimed at low-income individuals and households contributed significantly in reducing poverty levels in all the countries studied. It was found further that countries that removed financial restrictions that constrained low-income individuals from accessing microfinance services, saw the highest reduction in poverty levels. Several other longitudinal studies have established a significant link between financial inclusion and poverty levels (Donou-Adonsoua, & Sylwester, 2016; Revilak, 2017). 12 University of Ghana http://ugspace.ug.edu.gh In Ghana for instance, Danquah, Quartey and Iddrisu (2017) examined the influence of financial services (defined as rural and community bank services) on poverty alleviation among low income rural households. They analyzed the Ghana Living Standard Survey (GLSS 6). They reported that access to rural and community bank services significantly improved the standard of living in rural areas and reduced poverty levels significantly. Similar findings have been reported by recent studies examining the impact of cooperative banks on poverty reduction (Lal, 2018) and cross-country evidence linking financial inclusion to poverty reduction (Sethi & Acharya, 2018). Some of the studies that examine the individual level factors mainly look at constraints and facilitating conditions that mediate low income individuals’ access and uptake of financial services (Kim, Yu & Hassan, 2018). These group of studies report that individuals who live in low-income communities have very limited access to financial services, especially from the traditional banking systems (Park & Mercado, 2016). This may or can be explained by traditional banks not finding it economically viable to building brick-and-mortar branches in low population dense areas with low economic activity. The situation is further worsened by a lack of other forms of financial services including savings accounts, loans and insurance products - even to those who have access to bank accounts (Williams et al. 2017). Furthermore, the exclusion of many from the financial system may be due to the inability of the system to address issues of lack of competition,and information asymmetry (Camara, Peña, & Tuesta, 2014). The studies that examine country-level factors focus on how improvement in financial inclusion impact on general poverty levels within an economy. These group of studies also show that financial inclusion of any kind can reduce poverty among poor individuals. The argument is that access to financial services in one or more of it forms will significantly impact on the economic 13 University of Ghana http://ugspace.ug.edu.gh inequality, human development and social advancement (Demirguc-Kunt et al. 2017; Kim et al. 2018; Park & Mercado, 2016). This therefore presupposes that the extent of penetration of financial services leaves a positive impact on an economy. The studies further show that the level of penetration of financial inclusion in Africa compared to developed economies is on a low but is significantly rising. Demirgüç-Kunt et al. (2017) for instance, makes it evident that 69% of adults own an account, either with a bank or with a mobile money provider. This varies considerably between developed economies at 94% and developing countries’ economies at 63%. The report thus shows that financial inclusion is on the rise globally with 1.2 billion adults obtaining an account since 2011, including 515 million since 2014 (Park & Mercado, 2016). Between 2014 and 2017, the share of adults who have an account with a financial institution or through a mobile money service rose globally by 7 percent. In developing economies, the share rose 9 percent during the same period. It is however instructive to highlight that there are significant variations in account ownerships among individuals in developing economies (World Bank, 2018). The studies also identify different categorized services that are in demand under financial inclusion. One is payment services, which involves mainly the making of, or transferring of money from rural to urban and urban to rural areas, actively eliminating the middleman and cost, thereby including even the unbanked (Williams, Adegoke & Dare, 2017). Another is savings services which is argued to be the second most demanded form of financial inclusion services that ranges from saving for a rainy day, investment, business opportunity, income smoothing and retirement mostly in developing countries where there are hardly any pension plans and retirement programs (Koomson & Ibrahim, 2018). There are also micro-credit services which 14 University of Ghana http://ugspace.ug.edu.gh make credit available to small and or family businesses that require capital, mostly seen in developing countries (Williams et al. 2017). Again, there is also insurance services which are innovative risk mitigation services offered in the bid to protect individuals or businesses from specific losses. They may come in the form of health insurance, life insurance, crop insurance, etc (Boskov, 2018). The findings from the various studies cohere around the idea that eradicating poverty in all its forms can be tackled from a multisectoral perspective. These include creating opportunities for good and decent jobs and secure livelihoods, supporting inclusive and sustainable business practices and promoting better financial policies that target individuals within low income brackets. 2.3.2 Mobile Money Penetration and Poverty Alleviation Empirical studies assessing the relationship between mobile money and poverty alleviation have broadly focused on the value creation of mobile money services. Because of this, the literature in the field of mobile money and small and medium enterprises was drawn on to explain the dynamics of how mobile money leads to economic empowerment. The current upsurge of interest in empirically examining the contribution of mobile money to the operations of small businesses has produced research in two directions; descriptive studies and inferential studies (e.g. Chale & Mbamba, 2014). The descriptive studies, mainly using open-ended survey qualitative methodology, mainly describe the contribution of mobile money to business performance from the narratives of business owners. The inferential studies, mainly testing hypotheses, have explored different factors or variables that influence the outcomes of mobile money in business performance (e.g. Mbogo, 2010). Most of these studies reported that mobile 15 University of Ghana http://ugspace.ug.edu.gh money provides some benefits to small businesses as well as poses challenges in the form of risks. Some studies have provided empirical evidence to show that mobile money services significantly reduce poverty levels. Ky, Rugemintwari and Sauviat (2018) for instance, have showed that mobile money services increases savings culture. In a study that examined the association between mobile money usage and savings behavior, Ky et al. (2018) collected data from a sample of 500 users of mobile money services in Burkina Faso. Findings from their study showed significant association between mobile money use and savings behavior among the participants. They therefore suggested that promoting mobile money use among economically disadvantaged groups such as women and low educated individuals can improve their economic conditions. The findings agree with what has been reported in a longitudinal study of mobile banking sand saving among low-income individuals in 93 developing countries between 2005 and 2011 (Asongu & Odhiambo, 2017). Asongu and Odhiambo (2017) reported that mobile banking increased savings among individuals, which led to improved financial well-being. Aker, Boumnijel, Mcclelland and Tierney (2016) also examined the effect of using mobile money services on delivering cash transfer programs in Niger. The study involved household survey of a sample of 1,152 program beneficiaries in 96 intervention rural communities. The findings showed that, generally, mobile money services improved cash transfers among the participants in the study communities. This suggests that mobile money empowers individuals in low-income communities to engage in vibrant and economically viable transactions to mobile money transfers. Following findings from similar studies, some researchers have argued that in many low-income communities in developing countries, non-banking financial inclusion serves more direct impact on poverty reduction than traditional financial services (Aker, Boumnijel, 16 University of Ghana http://ugspace.ug.edu.gh McClelland & Tierney, 2016; Asongu & Odhiambo, 2017; Kochar, 2018; Sayid & Ghalib, 2016). Sekabir and Oaim (2017). For instance, they have reported that money services boost agricultural marketing and increase off-farm income in farming communities in Uganda. Several studies have also reported that mobile money provides numerous benefits to small businesses. Most of these studies have been reported in other parts of Africa, with Kenya’s M- Pesa leading the research on value creation of mobile money to small businesses. For example, Simiyu and Oloko (2015) examined the influence of Mobile Money Transfer on SMEs Growth in Kisumu City in Kenya. Using both quantitative and qualitative methodologies, they sampled a large number of 13,350 legally registered small businesses in Kisumum Municipality. They reported that majority of the business owners (77%), responded on the affirmative, that mobile money services has had a positive effect on their sales. In a related study to determine how SMEs used mobile money to grow their businesses, Chale and Mbamba (2014) examined 90 small businesses in Kinondoni district in the Dares Salaam region of Tanzania and reported a significant relationship of usage of mobile money towards SMEs business growth. They found that many of the business owners who used mobile money in their business transactions reported increased sales, efficiency in purchase of stocks, receipt of payments and smooth transfer of money. Similar findings have been reported in other studies elsewhere. For instance, Ngaruiya, Bosire and Kamau (2014) also reported that among 120 small business owners that the majority of the SMEs use mobile phone to carry out their financial transactions. They found that mobile money transactions are frequently used by small businesses to save and receive money as well as to make payments. In Ghana, Frempong (2009) also reported that among 600 enterprises were 17 University of Ghana http://ugspace.ug.edu.gh positive about the impact of mobile telephones on their business operations in terms of ease of contact with customers and suppliers, reduced cost of transportation and profitability. Different studies have also investigated factors that influence small business owners to use mobile money transfers for their business transactions. Mbogo (2010) found among a sample of 409 micro business entrepreneurs in Nairobi Kenya, that factors such as perceived accessibility, low cost, security, convenience, satisfaction and support factors are all positively related with the behavioral intention of using mobile money for payments. Thus, small businesses owners were more likely to use mobile money in their business transactions if they perceived mobile money as accessible, convenient and has high level of financial security. Bampoe (2015) also investigated the adoption of mobile money in emerging markets. Specifically, the study sought to find out variables that make users of the mobile money services use the service frequently in Ghana. Bampoe (2015) advanced among a sample of 320 mobile money users in Accra, that there exists a positive relationship between trust and intention to adopt to mobile money services in Ghana. 2.4 Summary of Literature Review The studies also show that financial development efforts in the form of developing financial inclusion strategies that capture low-income individuals provide opportunities for alleviating poverty levels. The studies further show that traditional banking services (in the form of rural and community banking services) reduces poverty in urban settings while mobile money services have capacity to reduce poverty levels in rural and low-income communities. However, the evidence is inconsistent with regards to which approach exerts greater impact on reducing 18 University of Ghana http://ugspace.ug.edu.gh poverty levels. The current study therefore contributes empirical evidence to add critical insight into the financial inclusion and poverty alleviation agenda. 19 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE METHODOLOGY 3.1 Introduction This section presents the methodological processes and procedures that were undertaken in collecting and analyzing data for the current study. Specifically, the sub-sections covered under this chapter include: descriptions of the research design, sources of data for the study, databases used for the study, data analyses and ethical considerations. 3.2 Research Design The research design used for the study is quantitative method. Quantitative research methodology is broadly defined by Bryman and Bell (2015) as a method of research where the data for all the variables in the data are measured or assessed in quantitative terms. Thus, according to Saunders (2011), in research studies that use quantitative methods, after the data for the study has been collected, they are first translated into numbers before any form of analyses is done. Therefore, by using quantitative methods, numerical figures of financial inclusion indicators and poverty indicators were used to track financial inclusion impact on poverty alleviation in Ghana. The quantitative method helps to test the impact or effect of one variable on the other. In this study therefore, the quantitative method helped to empirically test the effect of financial inclusion on poverty in Ghana. The main source of data for the study was the use of secondary data. Secondary data has been argued to be of very important utility in business and finance research because it helps in researching into the past trends of individuals and organizations (Cotteleer & Wan, 2016). The 20 University of Ghana http://ugspace.ug.edu.gh argument is that because secondary data already exists, a researcher cannot manipulate or be biased in the data collection process as the researcher makes use of data that has been reported already by an institution and so it provides opportunity for cross-checking by other researchers (Moore, Salter, Stanley & Tamboukou, 2016). Secondary data uses ex post facto information which helps to understand the trends in the phenomenon under study and also tracks past performance of poverty and financial inclusion proxies, as well as factors that influenced the change (Cotteleer & Wan, 2016). 3.3 Variables in the Study For the purpose of the study, the following variables are operationalized as follows: i. Financial Inclusion: this is the independent variable in the study. It was measured using the penetration of two indicators: traditional banking services (which includes commercial banks, rural banks and microfinance) and mobile money services. Three indicators were therefore used as indicators of financial inclusion: a. Access to commercial bank account b. Access to rural bank or microfinance account c. Access to mobile money account ii. Poverty alleviation: this is the dependent variable in the study. It was measured using three indicators. These are: a. Formal savings 21 University of Ghana http://ugspace.ug.edu.gh b. Access to finance c. Cashless transactions These indicators are chosen as poverty levels because they have been used in other studies and reports as proxies for poverty levels (Asongu & Odhiambo, 2017; Danquah et al. 2017). Access to finance for instance is found to reduce poverty and encourage savings (Burlando & Canidio, 2015). Therefore, access to finance and savings are good proxies for measuring poverty levels (Burlando & Canidio, 2015). Cashless transaction too has been reported to be common among individuals with access to electronic systems of payment, such as mobile money (Aker et al., 2016; Boateng, 2011). Thus, individuals report of engagement in cashless transactions since getting access to mobile money is a good proxy for poverty levels (Aker et al., 2016; Chale & Mbamba, 2014). 3.4 Source of Data Secondary data was used for the study. The data for the study was obtained from two different databases. These are the Global Financial Index Database (also called Global Findex) and the Inter-University Consortium for Political and Social Research (ICPSR). Each of these databases assess financial inclusion and poverty levels, mainly in several developing countries. Each of the databases are briefly described below. 3.4.1 The Global Findex Database The data will be obtained from The Global Findex Database of The World Bank. The Global Findex database is the world’s most comprehensive data set on issues relating to individuals’ 22 University of Ghana http://ugspace.ug.edu.gh financial practices including how they save, make payments, borrow, and manage risk. It was established with funding from the Bill & Melinda Gates Foundation, and has being publishing every three years since 2011. The study used the latest Global Findex data which was gathered in 2017 (i.e. Global Findex Database 2017). The 2017 edition includes updated indicators on access to and use of formal and informal financial services. This database provided information on access to commercial bank accounts, access to mobile money accounts and poverty level indicators. 3.4.2 Inter-University Consortium for Political and Social Research (ICPSR) Database Data was also sourced from the Inter-University Consortium for Political and Social Research (ICPSR), which is the largest repository of social science data in the world. The ICPSR provides opportunities for secondary data research to guide primary data research. Specifically, the data from the “Poverty Assessment and a Comparative Study of Rural Microfinance Institutions and Government Programmes in Ghana” was used for the study. The programme and the data was launched in 2004, and since gathered data on various indicators of financial inclusion and poverty levels in low income communities in Ghana. The project was led by researchers from the University of Cape Coast, Ghana. Specifically, the 2014 data (ICPSR 35296) was used for the study. This databased provided information or data on how to gain access to rural banks, microfinance institutions, mobile money services and poverty levels among low income urban communities and rural areas, from three regions in Ghana, which were: Central, Greater Accra, and Eastern. 23 University of Ghana http://ugspace.ug.edu.gh 3.5 Data Analysis The data for the study was analyzed using regression analysis. Regression analysis is best suited for examining cause and effect and therefore it helped to assess the extent to which financial inclusion has reduced or impacted on poverty levels in Ghana. Descriptive statistics were first used to analyze or explore the data, using means and standard deviations. Secondly, inferential statistics was used to examine the impact of indicators of financial inclusion (i.e. access to commercial bank account, access to rural bank or microfinance account, and access to mobile money services) on indicators of poverty levels (i.e. savings, access to credit and cashless transactions). Pearson r was used to test the relationship between the financial inclusion indicators and the poverty level indicators. After testing the relationship between the variables, the data was then analyzed using panel data multiple regression. Panel data multiple regression models are suited for analyzing panel data that is collected from different units over a period of time (Baltagi, Song & Koh, 2003; Born, & Breitung, 2016; Jumono, Achsani, Hakim & Fidaus, 2016). In the current study, the data from the two databases spanned a period of 2004 and 2017. 3.5.1 Panel Data Models Panel data regression techniques were used to test the effect of the selected indicators of financial inclusion on the selected indicators of poverty levels with the aid of Stata version SE 13. Before analyzing the data, Stata was set to handle panel data using the Xtset. Panel data technique was used because it provides much more in-depth and detailed information on financial inclusion and poverty levels as it provides a mix of cross-sectional dynamics and longitudinal dynamics in the data (Moon & Weidner, 2015). Thus, in this case, individuals represented the entities or panels (i) and year represented the time variable (t). 24 University of Ghana http://ugspace.ug.edu.gh Clark and Linzer (2015), show that panel data can be estimated using different strategies including models such as random effects (RE) model, constant coefficient effect (CCE) model and fixed-effect (FE) model. The fixed-effect (FE) model was used to estimate the effect the financial inclusion on poverty alleviation in the current study. The FE model was used because the study was interested in analyzing the change in poverty levels, as have been impacted by different financial inclusion strategies overtime. The FE model helps to control for all time invariant characteristics of the individuals so that the net effect of financial inclusion and poverty levels can be assessed (Clark, & Linzer, 2015). For example, some socio-demographic characteristics such as gender is constant but others like educational levels, age, marital status etc. change overtime. All these differences can therefore affect the study. The FE model therefore helps to control for all these by removing the time- invariant characteristics. Another assumption of the FE model is that the time-invariant characteristics are unique to individual entities in the study and therefore the characteristics should not be correlated with other individual characteristics (Clark, & Linzer, 2015). To test this, the error term and constant should not be correlated with the other characteristics (Born & Breitung, 2016). This assumption is tested using the Hausman test. When Hausman test shows that the error terms and the constants are not correlated, then FE is suitable but if they are correlated, then FE is not suitable, which requires that random-effects (RE) be used in such circumstances (Born & Breitung, 2016; Cameron & Trivedi, 2013). The idea in the FE model is that if the unobserved variable does not vary over time (in the short term), then any changes in the dependent variable must be due to influences other than these fixed characteristics. 25 University of Ghana http://ugspace.ug.edu.gh 3.5.2 Model Specification The equation for FE entity model for n entity-specific intercepts is given as: Yit = β1X1it +…+ βkXkit + αi + µi+ eit equation 1 Where - Yit is the intercept or dependent variable (DV) - Xit represents one independent variable (IV), - β1 is the coefficient for that IV, - αi (i=1...k) is the unknown intercept for each entity (k entity-specific intercepts). - µi represents a trend term that allow for a shift of the intercept overtime - eit is the error term - i represents unit of observation - t represents time period Some researchers explain that it is prudent to control for time effects whenever unexpected variation or special events may affect the outcome. In the model specified, there is no trend term that tracks the shift in the intercept or the dependent variable overtime (Clark & Linzer, 2015; Moon & Weidner, 2015). Therefore, the effect of time is controlled for. The model equation for such a specification therefore reads as: Yit = β1X1it +…+ βkXkit + αi+µi+ eit equation 2 When applied in the current study, time effect was controlled for because the period under investigation (2004–2017) has seen some special events in the form of different economic 26 University of Ghana http://ugspace.ug.edu.gh changes which may have affected the outcome. For instance, MTN introduced the first mobile money service to the Ghanaian market six years ago in 2009 and the celebration of Mobile Money Month was launched two years later in 2011 and remains an annual event since. However, prior to that commercial banks, rural banks and microfinance institutions were already operating in Ghana. It is therefore important to control for time to account for these differences. Table 1 shows the descriptions of variables included in the model in the current study. 27 University of Ghana http://ugspace.ug.edu.gh Table 1: Description of variables in the study Variable Description Independent Variable: 3 Proxies for Financial inclusion Access to commercial bank This assesses whether the participants have a bank account (CBA) account with any of the commercial banks operating in Ghana. Access to rural bank account This assessed whether the participants had an account (RBA) with any rural bank operating in Ghana. Access to microfinance account This assessed whether the participants had an account (MFA) with any microfinance institutions operating in Ghana Access to mobile money services This assessed whether the participants had registered (MMS) with mobile money services with any of the telecommunication companies operating in Ghana. Dependent Variable: 3 indicators of Poverty levels Savings (S) This examined the rate of savings among the participants Access to finance (AF) This assessed participants’ access to or ability to obtain financial services such as credits and loans for personal or business use. Cashless transactions (CT) This assessed the rate at which participants received or made payment using cashless transaction platforms in Ghana. 28 University of Ghana http://ugspace.ug.edu.gh The empirical model used in the current study is given as: Savings it = β1(CBA)t+ β2(RBA)2t+ β3(MFA)3t + β4(MMS)4t+ eit equation 3 Finance it = β1(CBA)t+ β2(RBA)2t+ β3(MFA)3t + β4(MMS)4t+ eit equation 4 Cashless it = β1(CBA)t+ β2(RBA)2t+ β3(MFA)3t + β4(MMS)4t+ eit equation 5 Where: Savings it = rate of savings of participant i in year t Finance it = access to finance of participant i in year t Cashless it = cashless transaction of participant i in year t (CBA)t = Commercial bank account ownership of participant i in year t (RBA)2t = Rural bank account ownership of participant i in year t (MFA)3t = Microfinance account ownership of participant i in year t (MMS)4t = Mobile money services of participant i in year t 3.6 Ethical Considerations Even though secondary data was used, some ethical considerations were adhered to. Use of secondary data is argued to be part of archival business research. Archival research also requires high ethical considerations like any other kind of research. Cowton (1998) argues that the fact that ex post facto data on organization or individuals are used does not mean throwing ethics to the wind during the research process. 29 University of Ghana http://ugspace.ug.edu.gh Even though obtaining information from a website of the databases might not require explicit permission from the organization to do so, for ethical purposes, the data that was retrieved from the homepages of the databases were used solely for the purposes of this study and not for any other purpose. 30 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR ANALYSIS, RESULTS AND DISCUSSION 4.1 Introduction This chapter entails the findings that answer the research questions. For the purposes of an in- depth analysis and understanding, a panel data regression was used to model the data. The panel regression modelling helps to determine how and by what magnitude financial inclusion measures impact on poverty levels within the study period. The chapter progresses as follows: First, descriptive analysis is presented. Secondly, the models showing financial inclusion proxies as predictors of poverty levels are presented. Thirdly, findings of the study are discussed followed by a summary of the chapter. 4.2 Demographic characteristics of participants Data from eight hundred and thirty-two (832) participants were extracted for analysis. The socio- demographic profiles of the respondents are provided on Table 2 below. As shown on Table 2, more than half of the participants were female (52.3%), with the rest being males(47.7%). Their ages ranged between 20–59 years. In terms of education, majority of them (32.9%) had basic education. 31 University of Ghana http://ugspace.ug.edu.gh Table 2: Socio-demographic characteristics of participants Characteristic Categories Frequency Percentage Gender Male 397 47.7% Female 435 52.3% Age 29 – 29 years 88 10.6% 30 – 39 years 266 32.0% 40 – 49 years 388 46.6% 50 – 59 years 90 10.8% Educational level No education 121 14.5% Basic education 274 32.9% High school 212 25.5% Tertiary 135 16.2% 4.3 Descriptive Analyses This sub-section is made of two parts. The first part presents descriptive statistics of the scores on the various variables. Here, the number of observations, minimum and maximum scores, mean scores and standard deviations are provided. Values indicating the skewness and kurtosis are also provided to check the distribution of the data for statistical analysis. The second sub-section presents a correlation matrix for all the variables in the study. The correlation matrix checks the relationship between the variables in the study using Pearson r. This helps to provide a preliminary insight into the associations that exist among the variables. 32 University of Ghana http://ugspace.ug.edu.gh 4.3.1 Descriptive Statistics The descriptive statistics gives a summary of the scores of the three indicators of poverty levels and the four determinants of financial inclusion among the participants used in the current study. For financial inclusion, the participants were asked to indicate ‘yes or no’ as to whether or not they have a commercial bank account, a rural bank account, a microfinance account and a mobile money account. For poverty levels, the participants were asked to rate (on a scale of 1 – 10, where highest number indicate ease), the extent to which they engaged in savings, have access to finance and engaged in cashless payment systems. The results of the descriptive statistics are summarized in Table 2. 33 University of Ghana http://ugspace.ug.edu.gh Table 3: Descriptive statistics of scores Variable Indicators f (%) / M (SD) Financial inclusion proxies Commercial bank accounts Yes 344 (41.3%) No 488 (58.7%) Rural bank accounts Yes 214 (25.7%) No 618 (74.3%) Microfinance accounts Yes 93 (11.2%) No 739 (88.8%) Mobile money accounts Yes 698 (83.9%) No 134 (16.1%) Poverty levels proxies Savings 6.34 (2.45) Access to finance 4.98 (2.78) Cashless transactions 2.97 (3.34) As shown on Table 2, majority of the participants had access to mobile money accounts (83.9%), followed by access to commercial bank accounts (41.3%) and finally rural bank accounts (25.7%). Comparing the mean scores for the poverty level proxies, the participants rated savings higher (M = 6.34, SD = 2.45), compared to access to finance (M = 4.98, SD = 2.78) and cashless 34 University of Ghana http://ugspace.ug.edu.gh transactions (M = 2.97, SD = 3.34). The scores on the poverty level indicators suggest that savings is common among the participants, while access to finance and cashless transactions are still challenging to them. 4.3.2 Correlation Analyses A correlation analysis was conducted to examine the relationship between three poverty level indicators (savings, access to finance and cashless transactions) and the financial inclusion indicators (commercial bank accounts, rural bank accounts, microfinance accounts and mobile money accounts) using Pearson r statistical test. Correlation among two variables represents the kind of movement that exist between the variables. For example, a positive correlation means that the two variables move in the same direction. This means that the two variables either increase or decrease together. A negative correlation means that the two variables move in different directions. This means that as one variable increases, the other variable decreases and vice-versa. Correlations only indicate the relationship between the variables but do not necessarily imply a causal relationship or causation. For example, the fact that two variables move in the same or opposite directions does not in itself mean that one variable has a direct cause on the other. Table 3 provides the correlation matrix between the variables. 35 University of Ghana http://ugspace.ug.edu.gh Table 4: Correlation matrix showing relationships among the variables 1 2 3 4 5 6 7 1. Savings 1 2. Access to finance .358** 1 3. Cashless transactions .210** -.029 1 4. Commercial bank account .321** -.036 .259** 1 5. Rural bank account .222** .231** -.433** .107 1 6. Microfinance account .235* .329*** -.237** .170* .483** 1 7. Mobile money account .059 .181* .444*** .009 -.062 .104 1 *significant at p < .05; ** significant at p <.01; *** significant at p <.001 Table 3 shows that savings is positively correlated with commercial bank accounts (r = .321, p < .01), rural bank accounts (r = .222, p < .01) and microfinance accounts (r = .235, p < .01), but not mobile money accounts (r = .059, p > .05). Access to finance was found to be significantly related to rural bank accounts (r = .231, p < .01), positively related to microfinance accounts (r = .321, p < .01) and also positively with mobile money accounts (r = .181, p < .05). Cashless payments were positively related to commercial bank accounts (r = .259, p < .001) and microfinance accounts (r = .444, p < .001) but negatively with both rural bank accounts (r = - .433, p < .001) and microfinance accounts (r = -.237, p < .01). 36 University of Ghana http://ugspace.ug.edu.gh 4.4 Results of financial inclusion as determinants of poverty levels This sub-section of the chapter presents results on the predictors of profitability indicators among the commercial banks in the period under study (i.e. between 2004 and 2017). Based on the specific objectives of the study, three estimated models were formulated and tested. 4.4.1 Financial inclusion and savings The panel regression model of the financial inclusion as determinants of savings is presented on Table 4 below. 37 University of Ghana http://ugspace.ug.edu.gh Table 5: Financial inclusion as determinant of savings Predictors β S.E. t p Commercial bank accounts .237 .048 4.378 .000 Rural bank accounts .158 .058 4.28 .000 Microfinance accounts .139 .059 2.338 .020 Mobile money accounts .012 .048 1.431 .154 Constant (Intercept) 17.273 7.387 5.231 .000 Number of observations 119 R – square (R2) .367 Adjusted R2 .337 F 12.111 df (4, 828) p .000 Root MSE 8.002 In estimating the model for predicting savings, the financial inclusion indicators were dummy coded as (yes = 1, no = 0). Secondly, commercial bank accounts, microfinance accounts, rural bank accounts and mobile money accounts were inserted into the model as predictors of savings. As shown on Table 4, overall model was significant (F = 12.11, p < .001). All the predictors accounted for about 36.7% variance in savings (R2 = .367). When the individual determinants were examined, the findings showed that savings were positively predicted by commercial bank accounts, explaining about a 23.7% variance (β = .237, 38 University of Ghana http://ugspace.ug.edu.gh p < .001) to rural bank accounts, explaining about 11.6% variance (β = .116, p < .001) and microfinance accounts, explaining 13.9% variance (β = .139, p < .05). Mobile money accounts did not predict savings. Model 1: From Table 4, an estimated model is built to explain how financial inclusion affects a savings culture among the participants. Based on equation 3, the model is estimated as follows: From 5, it was predicted that: Savings it = β1(CBA)t+ β2(RBA)2t+ β3(MFA)3t + β4(MMS)4t+ eit equation 3 The findings however showed that Savings it = -.139(CBA)t + .116(RBA)2t + .237 (MFA)3t + 17.273 equation 3a From 3a, the estimated model 1 shows that when access to commercial bank accounts rose by 1 percent, there was a corresponding 13.9% increase in savings culture. Also, a unit increase in access to rural bank accounts was associated with an 11.6% improvement in savings culture. A unit increase in access to microfinance accounts, however, was associated with a 23.7% increase in savings culture. These results suggest that access to microfinance accounts had a more positive effect on saving culture compared to commercial and rural bank accounts. Discussion of financial inclusion and savings: The findings have shown that financial inclusion increases savings culture among the individuals. Specifically, access to commercial bank accounts, rural bank accounts and microfinance accounts significantly improve savings among them. Mobile money accounts, however, was found not to have any significant impact on savings. These findings show that 39 University of Ghana http://ugspace.ug.edu.gh access to the traditional system of banking accounts is a crucial way for reducing poverty levels among low income individuals in Ghana. The possible explanation for this is that in many rural banks and microfinance institutions for instance, there are mobile bankers who visit clients on a daily basis at their place of work to collect their savings deposits. These savings deposits are usually in small amounts, as can be afforded by the individuals. This ensures that the individuals are encouraged to make their savings contributions to their financial institution. This might explain why access to savings accounts in these three financial institutions increases savings. The findings are consistent with some previous studies (Clamara et al. 2014; Demirguc-Kunt et al. 2017; Miled & Reieb, 2015; Samer et al., 2015). Samer et al (2015), for instance, have reported that access to microfinance in deprived communities increases micro savings in Malysia. Miled and Reieb (2015), in analysis of 57 developing countries, have also reported that microfinance institutions encourage savings among low income households. Mobile money on the other hand had no effect on savings, possible reason being that among low income individuals, mobile services are mainly registered for sending and receiving money. Mobile money services are not usually used for savings among the majority of mobile money subscribers, especially individuals whose incomes are low. This might explain why mobile money did not have a significant effect on savings. This contradicts evidence reported in other developing countries that mobile money increases savings (Aker et al., 2016; Asongu & Odhiambo, 2017; Ky et al., 2018). Ky et al. (2018), for instance, have reported that in Burkina Faso,mobile money services increases saving among low income women. Asongu and 40 University of Ghana http://ugspace.ug.edu.gh Odhiambo (2017) have also reported that mobile banking increased savings among individuals, which led to improved financial well-being. In the current study however, access to mobile money services was found to have no significant effect on savings among low income individuals. The possible explanation for this is that, in Ghana, many people are still not used to mobile money as a platform for savings. The traditional forms of savings which involve depositing money with financial institutions is still dominant, especially among individuals with low income.. Therefore, they would rather use mobile money services for sending and receiving money (which are more dominant themes in the mobile money uptake drive and campaign in Ghana) rather than for savings. The other forms of services offered by mobile money platforms such as savings, insurance is not as popular compared to the idea of using mobile money for sending and receiving money, especially among low income individuals and households in Ghana. Perhaps the majority of individuals who use mobile money for savings are individuals with high levels of education and income, who appreciate these service features of mobile money platforms. 4.4.2 Financial inclusion and access to finance The panel regression model of the determinants of access to finance is presented on Table 5 below: 41 University of Ghana http://ugspace.ug.edu.gh Table 6: Financial inclusion as determinant of access to finance Predictors β S.E. t p Commercial bank accounts -.195 .032 -3.125 .000 Rural bank accounts .116 .858 6.155 .000 Microfinance accounts .126 .043 3.128 .002 Mobile money accounts -.012 .090 .866 .387 Constant (Intercept) 18.324 4.456 6.345 .000 Number of observations 119 R – square (R2) .437 Adjusted R2 .412 F 25.052 df (4, 828) p .000 Root MSE 7.813 As shown on Table 6, the overall model was significant (F = 25.052, p < .001). All the predictors accounted for about 43.7% variance in return on equity (R2 = .367). The findings showed that access to finance was negatively predicted by commercial bank accounts, explaining about 19.5% variance (β = -.195, p < .001), but was positively predicted by rural bank account, explaining about 11.6% variance (β = .116, p < .001) and microfinance account, explaining 12.6% variance (β = .12.6, p < .01). 42 University of Ghana http://ugspace.ug.edu.gh Model 2: From Table 6, an estimated model is built to explain how financial inclusion impact access to finance among the participants. From equation 5, it was predicted that: Access to finance it = β1(CBA)t+ β2(RBA)2t+ β3(MFA)3t + β4(MMS)4t+ eit equation 5 The findings however showed that Access to finance it = .116(RB)t + .126(MFA)2t - .195 (CBA)3t + 18.324 equation 5a Equation 5a therefore shows that access to rural bank account and microfinance account improves access to finance among the participants, while access to commercial bank account decreases access to finance. Mobile money account had no significant effect on access to finance. Discussion of financial inclusion and access to finance The findings have shown that access to rural bank accounts and microfinance accounts increases access to finance, while commercial bank accounts decreases access to finance. Mobile money accounts were found to have no impact on access to finance. Invariably, what this means is that low income individuals who have savings accounts with rural banks and microfinance institutions are more likely to have access to finance compared to low income individuals who have savings accounts with commercial banks. The possible explanation for this is based on how access to finance was measured. Access to finance looked at access to loans and credit facilities. Relatively, rural banks and microfinance institutions provides microloans to low income individuals easily compared to commercial banks. The requirements for accessing loans in rural banks and microfinance institutions are 43 University of Ghana http://ugspace.ug.edu.gh sometimes easily met by low income individuals, compared to the requirements for accessing loans in commercial banks. Therefore, individuals who have savings accounts with rural banks and microfinance institutions are more like to receive microloans, compared to low income individuals with savings accounts in commercial banks. The findings in this study support findings from some previous studies (Danquah et al., 2017; Williams et al. 2017), and contradict findings from other studies (Lal, 2018; Sethi & Acharya, 2018). Sethi and Acharya (2018) have for instance reported that access to saving accounts in corporate banks reduces poverty. Lal (2018) also says that having savings accounts with corporate banks improves access to capital. Evidence from this study shows that having saving accounts with corporate or commercial banks reduces access to finance for individuals with low incomes. The possible reason, as alluded to by Williams et al. (2017), is that commercial banks do not find individuals with low income to be economically attractive or viable for loans, due to the possibility of high levels of defaulting Danquah et al., (2017) have also reported that in Ghana microfinance institutions readily give loans to low income individuals.Therefore, in terms of improving access to finance, rural banks and microfinance institutions promote poverty reduction more among low income individuals, compared to commercial banks and mobile money services. 4.4.3 Financial inclusion and cashless transactions The panel regression model of how financial inclusion affect cashless transactions is provided on Table 7 below. 44 University of Ghana http://ugspace.ug.edu.gh Table 7: Financial inclusion as determinant of cashless transactions Predictors β S.E. t p Commercial bank accounts -.102 .041 -2.113 .031 Rural bank accounts -.193 .032 -3.231 .009 Microfinance accounts -.137 .042 -2.338 .032 Mobile money accounts .112 .021 2.866 .007 Constant (Intercept) 16.645 .089 5.438 .000 Number of observations 119 R – square (R2) .337 Adjusted R2 .322 F 15.231 df (4, 828) p .000 Root MSE 8.672 As shown on Table 7, the overall model was significant (F = 15.231, p < .001). All the predictors accounted for about 33.7% variance in return on equity (R2 = .337). The findings showed that cashless transactions was positively predicted by mobile money accounts, explaining 11.2% increase in cashless transactions (β = .112, p < .01), but negatively predicted by commercial bank accounts, explaining about 10.2% decrease (β = -.102, p < .05), rural bank accounts, explaining 45 University of Ghana http://ugspace.ug.edu.gh about 19.3% decrease (β = -.193, p < .01) and microfinance accounts, explaining 13.7% decrease (β = -.137, p < .01). Model 3: From Table 7, an estimated model is built to explain how financial inclusion impact on cashless transactions among the participants. Based on equation 5, the model is estimated as follows: From equation 5, it was predicted that: Cashless transactions it = β1(CBA)t+ β2(RBA)2t+ β3(MFA)3t + β4(MMS)4t+ eit equation 4 The findings however showed that Cashless transactions it = .112(MMS)4t -.102(CBA)t - .193(RBA)2t - .237 (MFA)3t + 16.645 equation 4a From 4a, the estimated model 3 shows only mobile money services increased cashless transactions among the participants. Commercial bank accounts, rural bank accounts and microfinance accounts were all found to be associated with reduced cashless transactions among the participants. Discussion of financial inclusion and cashless transactions Cashless transactions have been argued to constitute towards a key poverty reduction strategy, even among individuals and households with low income. Cashless payment services, which involves mainly the making of or transferring of money without the physical cash, reduces poverty by actively eliminating the middle man and cost, thereby including even the unbanked (Williams et al., 2017). It is therefore imperative to understand how financial inclusion strategies affect cashless transactions among low income individuals in Ghana. 46 University of Ghana http://ugspace.ug.edu.gh In the current study, the findings have shown that access to mobile money services and increased cashless transactions among individuals with low income. Access to traditional banking, such as commercial bank accounts, rural bank accounts and microfinance were found to reduce cashless transactions among the participants. This suggests that for individuals with low incomes, mobile money services have helped in encouraging their engagement in cashless financial transfers and transactions. The findings are in line with some previous studies (Aker et al., 2016; Asongu & Odhiambo, 2017; Chale & Mbamba, 2014; Ky et al., 2018). Aker et al. (2016), have reported that in Nigeria, mobile money services improved cash transfers among the participants in the study communities. Chale and Mbamba (2014) have also reported that mobile money services increased sales, efficiency in purchase of stocks, receipt of payments and smooth transfer of money among individuals who operate small businesses in Tanzania. In Ghana, Frempong (2009) has also reported that majority of the respondents were positive about the impact of mobile money services on their businesses in terms of ease of payment, contact with customers and suppliers, reduced cost of transportation and profitability. Contrary to what was expected, the traditional forms of banking however has been found to reduce cashless transactions in the current study. Having accounts with commercial banks, rural banks and microfinance institutions were found to reduce cashless transactions. There are several possible explanations for this. First, majority of rural and microfinance institutions do not provide financial transfer services. Therefore, individuals who have accounts with these institutions are less likely to engage in cash transfers and cashless transactions using rural banks or microfinance institutions. This might explain why having account with rural banks and microfinance institutions were found to be associated with reduced cashless transfers. 47 University of Ghana http://ugspace.ug.edu.gh Commercial banks in Ghana on the other hand provides several platforms for cash transfers and cashless transactions. However, a deeper look into it shows that the technologically sophisticated nature of these cash transfer platforms, and their associated charges mean that the services are better suited for individuals with both high income and high level of education. What this means is that individuals with low income and education, who maintain accounts with commercial banks in Ghana, are not able to use their platforms for online transactions and transfers. This might explain why having access to commercial bank accounts is associated with reduced cashless transactions among low income individuals in Ghana. 4.5 Synthesis of the findings Overall, findings from the current study provide empirical evidence to the effect that financial inclusion strategies alleviate poverty indicators among individuals with low incomes. However, there are different pathways through which this happens, depending on the particular financial inclusion strategies assessed and the indicators of poverty used. In the current study, financial inclusion was assessed as access to traditional systems of banking (i.e. access to savings accounts in commercial banks, rural banks and microfinance institutions) and access to mobile money services. Poverty indicators used were rate of savings, access to finance (in the form of micro loans and credits) and cashless transactions. The findings have shown that these financial inclusion strategies affect poverty reduction in different pathways. When it comes to rate of savings, traditional systems of banking have been shown to increase savings among individuals in low income brackets, while access to mobile money services had no significant effect. When it comes to access to finance, access to rural 48 University of Ghana http://ugspace.ug.edu.gh banks and microfinance institutions were found to increase access to microloans and credits to individuals with low income, compared to commercial banks. Here too, access to mobile money services did not have significant impact. However, when it comes to cashless transactions, access to mobile money services was found to increase cash transfers and cashless transactions among the individuals. Traditional systems of banking were found to significantly reduce cashless transactions among low income individuals. Suffice to say that, these financial inclusion strategies do not work in isolation. They work together in complex ways in affecting poverty levels of low-income individuals and households. For instance, an individual may have mobile money account together with other traditional banking accounts, such as a rural bank account, commercial bank account and microfinance account. In the same ways individuals may also have multiple traditional banking accounts, such as having savings accounts with commercial bank and rural bank simultaneously. However, it is imperative to understand that, these financial inclusion strategies may serve different purposes for individuals. For instance, as has been shown in the findings from the current study, low-income individuals use mobile money services for sending and receiving money; rural bank or microfinance accounts for savings and accessing microloans and credits,and commercial bank accounts for savings. It is imperative to understand the complex pathways through which financial inclusion strategies work to reduce poverty among low- income individuals and households, so that financial intervention policies and programs can be tailored to their needs. 49 University of Ghana http://ugspace.ug.edu.gh 4.6 Chapter Conclusion This chapter has presented findings on how financial inclusion impact poverty levels, especially among low income individuals in Ghana. The findings and the discussions have largely showed that rural bank accounts, microfinance accounts and mobile money services have a more positive effect on poverty reduction (in terms of increasing savings, improving access to finance and promoting cashless transactions) among low income individuals, compared to commercial bank accounts. 50 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE SUMMARY, CONCLUSION AND RECOMMENDATIONS 5.1 Introduction In Ghana, a significant number of individuals live in poverty. The situation is further compounded by limited access to finance. There is therefore a spiral and cyclical association between access to finance and poverty levels. Financial inclusion is therefore argued to be an important step in alleviating extreme poverty. There have there been upsurge policy and interests in Ghana (in line with global interest) to understanding the link between financial inclusion and poverty alleviation. These developments, notwithstanding surveillance studies or research on financial inclusion and poverty levels in Ghana, have not paralleled the enthusiasm for expanding access to finance among low income individuals. Most of the studies conducted in Ghana for instance, have focused on examining factors that influence the uptake of financial inclusion services and innovations. There is therefore the need for more empirical research to be conducted to provide evidenced- based information to guide financial inclusion policy regulations and practice. The current study therefore fills this gap in research, especially in Ghana by empirically examining how financial inclusion strategies or interventions have affected poverty reduction overtime. This is done mainly by analyzing time series panel data on access to traditional banking (i.e. commercial, rural and microfinance) and mobile banking services, and how these have impacted poverty levels (defined as savings culture, access to finance and cashless transactions) between 2004 and 2017. 51 University of Ghana http://ugspace.ug.edu.gh The final chapter incorporates all the findings from the study to offer a holistic understanding into how financial inclusion strategies impact on poverty reduction among low-income individuals in Ghana. The summary of the findings from the study are presented first. Thereafter, a discussion of the practical and theoretical implications of the findings. This is followed by a discussion of the limitations of the study and then recommendations are made for future studies and conclusion drawn. 5.2 Summary of Findings This section presents the summary of the findings from the study. Fundamentally, findings from the study have indicated that: i. Access to traditional banking in the form of commercial bank accounts, rural bank accounts and microfinance accounts significantly improved savings among low income individuals. Mobile money accounts however, were found to not have any significant impact on savings. These findings show that access to traditional system of banking accounts is a crucial way for reducing poverty levels among low income individuals in Ghana. ii. Access to traditional banking had complex effects on access to finance among low income individuals. Specifically, access to both rural bank accounts and microfinance accounts were found to significantly increase access to finance, while commercial bank accounts was found to decrease access to finance. Here too, mobile money accounts were found to have no impact on access to finance. This suggests that low income individuals who have savings accounts with rural banks and microfinance institutions are more likely 52 University of Ghana http://ugspace.ug.edu.gh to have access to finance compared to low income individuals who have savings accounts with commercial banks. iii. Access to mobile money services was found to significantly increase cashless transactions among individuals with low income. However, access to traditional banking (i.e. commercial, rural and microfinance accounts) were found to significantly reduce cashless transactions among the participants. This suggests that for individuals with low incomes, mobile money services have helped in encouraging their engagement in cashless financial transfers and transactions. 5.3 Implications of the Study The findings from the study have several implications, both for practice and theory. The practical implications of the study relate to poverty reduction efforts among low income individuals in Ghana. The theoretical implications relate to theory-building and broadening of the literature from the findings in this study. The practical and theoretical implications of the study are therefore discussed. 5.3.1 Practical implications of the findings Based on the findings in the current study, the following implications and recommendations are made for using financial inclusion as poverty reduction tool among low-income individuals in Ghana. i. Access to rural banking and microfinance services were found to improve both savings culture and access to finance for low income individuals in the current study. It is 53 University of Ghana http://ugspace.ug.edu.gh therefore recommended that the drive for access to traditional banking in low-income communities focuses on these two. ii. In recent times, there have been an increase in the collapse of microfinance institutions, and in many of the cases, low-income individuals lose their savings. The situation might discourage savings culture among them. It is therefore recommended that the Bank of Ghana takes interventions to bring sanity into the microfinance sector, since it plays a key role in financial inclusion and poverty reduction in Ghana. iii. Access to mobile money services was found to increase cashless transactions, mainly in the form of receiving money. It was observed that mobile money had no impact on savings culture. In Ghana, mobile money services have a savings element in them. However, based on findings from this study, it appears that the savings element of mobile money banking has not caught on with low-income individuals. It is recommended that this feature of mobile banking services is popularized in low income communities so that they can engage in cashless micro savings. 5.3.2 Theoretical implications of the findings Theoretically, the findings from the current study broaden the global literature and theory on financial inclusion and poverty reduction. The empirical evidence from the current study shows for instance, that as financial inclusion strategies move from planning and policy makers into the communities, the nature changes based on how the communities receive such interventions. From the trickle-down theory’s assumptions, it can be argued that the features of mobile money services eventually trickle down to communities to use. The findings in the study indicate that not all features trickle down. Among the low-income individuals in the context of this study for 54 University of Ghana http://ugspace.ug.edu.gh instance, while money transfer element of mobile money has trickled down, the savings element has not. There is the need therefore to revise current theory to adequately explain when, how and why the elements of macro-level financial inclusion strategies trickle down to communities. 5.4 Limitations of the study The current study has some limitations which should be taken into consideration when interpreting or applying the results for practice or policy purposes. These limitations are: i. First, the study examined the direct effect of financial inclusion strategies on poverty reduction among low income individuals. The study did not measure how the individuals responded to the various financial inclusion indicators that brought about the effect being observed in the current study. It is therefore imperative that the findings in the study are interpreted independent of how these individuals respond to financial inclusion strategies. ii. Secondly, poverty was operationalized as savings culture, access to finance and use of cashless transactions. These three indicators, even though they provide understanding of the financial context of poverty, do not necessarily capture all aspects of poverty. Definition of poverty is very complex, fluid and multifaceted, especially among low income individuals in a middle lower income country like Ghana. Therefore, the three indicators need to be in mind when discussing poverty within the context of this study. iii. Definition of poverty is also very subjective, in a sense that poverty may mean different things to different individuals. In the current study however, an objective approach to poverty was taken. The study was purely quantitative, with the use of secondary data. Despite the high utility of secondary data, there are still some limitations. For instance, 55 University of Ghana http://ugspace.ug.edu.gh the quantitative nature of the study does not provide in-depth understanding into the subjective experiences and complex mechanisms involved in linking the financial inclusion strategies to poverty reduction among low income individuals. Because of these limitations, findings from the current study should be interpreted and/or applied with caution. 5.5 Recommendations for Future Research Notwithstanding the limitations discussed, the study has provided useful insights, which can serve as entry point for understanding the complex association between financial inclusion strategies and poverty reduction. Based on the findings from the study, it is submitted that future studies ought to build on it to provide deeper understanding. The following recommendations are therefore made for consideration for future studies: i. Future studies should consider examining how low-income individuals respond to macro- level financial inclusion strategies or initiatives. This will provide deeper understanding of the mechanisms that mediate how different financial inclusion strategies affect poverty levels among these group of people. ii. There is the need to also use qualitative methods in exploring how financial inclusion strategies affect poverty levels in low income communities. Qualitative studies will provide deeper understanding into for instance, how individuals define poverty subjectively and what financial inclusion means to them. Qualitative methods such as in- depth interviews, focus group discussions and ethnography can be used to delve deeper into their subjective lives. Thus, while quantitative approaches provide understanding of 56 University of Ghana http://ugspace.ug.edu.gh breadth, qualitative studies can also supplement with understanding of depth so that a comprehensive, more encompassing and all-inclusive poverty reduction strategies can be implemented in low income communities. 5.6 General Conclusion The current study sought to explore how financial inclusion strategies impact on poverty alleviation among low-income individuals in Ghana. Financial inclusion was defined in terms of traditional banking (i.e. commercial banking, rural banking and microfinance) and access to mobile money services. Poverty alleviation was defined in terms of savings culture, access to finance and cashless payments. The findings showed that the three financial inclusion strategies impacted poverty alleviation, albeit in different ways. For savings culture, traditional systems of banking increased savings among individuals in low income brackets, while access to mobile money services had no significant effect. For access to finance, rural banks and microfinance services were found to increase access to microloans and credits to individuals with low income, compared to commercial banks. For cashless transactions, access to mobile money services was found to increase cash transfers and cashless transactions among the individuals. Traditional systems of banking were found to significantly reduce cashless transactions among low income individuals. Overall, findings show different pathways through which financial inclusion strategies alleviate poverty indicators, depending on the financial inclusion strategies assessed, and the indicators of poverty used. 57 University of Ghana http://ugspace.ug.edu.gh REFERENCES Aker, J. C., Boumnijel, R., McClelland, A., & Tierney, N. (2016). Payment mechanisms and antipoverty programs: Evidence from a mobile money cash transfer experiment in Niger. Economic Development and Cultural Change, 65(1), 1-37. Asongu, S. A., & Odhiambo, N. M. (2017). Mobile banking usage, quality of growth, inequality and poverty in developing countries. Information Development, 0266666917744006. Baltagi, B. H., Song, S. H., & Koh, W. (2003). Testing panel data regression models with spatial error correlation. Journal of econometrics, 117(1), 123-150. Bampoe, S. H. (2015). Mobile Money Adoption in Emerging Markets: a case of Ghana. Unpublished Master Thesis, University of Ghana Business School. Bell, E., Bryman, A., & Harley, B. (2018). Business research methods. Oxford university press. Boateng, R. (2011). Mobile Phones and Micro-trading Activities; Conceptualizing the link. Information, 13(5), 48-62. Born, B., & Breitung, J. (2016). Testing for Serial Correlation in Fixed-Effects Panel Data Models. Econometric Reviews, 35(7), 1290-1316. Boskov, T. (2018). Level Of Deepening Financial Infrastructure, Fintech Companies and Financial Inclusion: Theory and Evidence. IJIBM International Journal of Information, Business and Management, 10(4), 23-31. Boukhatem, J. (2016). Assessing the direct effect of financial development on poverty reduction in a panel of low-and middle-income countries. Research in International Business and Finance, 37, 214-230. 58 University of Ghana http://ugspace.ug.edu.gh Burlando, A., & Canidio, A. (2015). Financial inclusion of vulnerable households through Savings and Borrowing Groups: Theory and experimental evidence from Uganda. Universiy of Oregon Research, Eugene. Cameron, A. C., & Trivedi, P. K. (2013). Regression analysis of count data (Vol. 53). Cambridge university press. Chale, R. P. & Mbamba, U. (2014). The role of mobile money services on growth of small and medium enterprises in Tanzania: evidence from Kinondoni district in dares salaam region. Business Management Review pp 81-96. Clamara, N., Ximena, P., & David Tuesta. (2014). Factors that Matter for Financial : Evidence from Peru. BBVA Research. Clark, T. S., & Linzer, D. A. (2015). Should I use fixed or random effects?. Political Science Research and Methods, 3(2), 399-408. Cotteleer, M. J., & Wan, X. (2016). Does the Starting Point Matter? The Literature‐Driven and the Phenomenon‐Driven Approaches of Using Corporate Archival Data in Academic Research. Journal of Business Logistics, 37(1), 26-33. Cowton, C. J. (1998). The use of secondary data in business ethics research. Journal of Business Ethics, 17(4), 423-434. Danquah, M., Quartey, P., & Iddrisu, A. M. (2017). Access to Financial Services Via Rural and Community Banks and Poverty Reduction in Rural Households in Ghana. Journal of African Development, 19(2), 67-76. Demirguc-Kunt, A., Klapper, L., & Singer, D. (2017). Financial Inclusion and Economic Growth: A Review of Recent Empirical Evidence. Washington DC: The World Bank. 59 University of Ghana http://ugspace.ug.edu.gh Demirguc-Kunt, A., Klapper, L., & Singer, D. (2017). Financial Inclusion and Economic Growth: A Review of Recent Empirical Evidence. Washington DC: The World Bank. Demirguc-Kunt, A., Klapper, L., Singer, D., & Van Oudheusden, P. (2015). The Global Findex Database 2014. Measuring Financial Inclusion Around the World (Policy Research Working Paper No. 7255). Washington DC. Donou-Adonsou, F., & Sylwester, K. (2016). Financial development and poverty reduction in developing countries: New evidence from banks and microfinance institutions. Review of Development Finance, 6(1), 82-90. Dzokoto, V. A., & Appiah, E. (2014). Making sense of mobile money in urban Ghana: Personal, business, social and financial inclusion prospects. Institute for Money, Technology and Financial Inclusion (IMTFI). University of California, Irvine. Frempong, G. (2009)."Mobile telephone opportunities: the case of micro- and small enterprises in Ghana", info, Vol. 11 Iss 2 pp. 79 94 Inoue, T. (2018). Financial development, remittances, and poverty reduction: Empirical evidence from a macroeconomic viewpoint. Journal of Economics and Business, 96, 59-68. Kim, D.-W., Yu, J.-S., & Hassan, M. (2018). Financial inclusion and economic growth in OIC countries. Research in International Business and Finance, vol. 43(C), pages 1-14. Kochar, A. (2018). Branchless banking: Evaluating the doorstep delivery of financial services in rural India. Journal of Development Economics, 135, 160-175. Koomson, I., & Ibrahim, M. (2018). Financial Inclusion and Growth of Non-farm Enterprises in Ghana. In Financing Sustainable Development in Africa (pp. 369-396). Palgrave Macmillan, Cham. 60 University of Ghana http://ugspace.ug.edu.gh Ky, S., Rugemintwari, C., & Sauviat, A. (2017). Does Mobile Money Affect Saving Behaviour? Evidence from a Developing Country. Journal of African Economies, 27(3), 285-320. Lal, T. (2018). Impact of financial inclusion on poverty alleviation through cooperative banks. International Journal of Social Economics, 45(5), 808-828. Mbogo, M. (2010). The Impact of Mobile payments on the Success and Growth of Micro- Business: The case of M-Pesa in Kenya. The Journal of Language, technology and Entrepreneurship in Africa, 2(1), 182- 203. Miled, K. B. H., & Rejeb, J. E. B. (2015). Microfinance and poverty reduction: a review and synthesis of empirical evidence. Procedia-Social and Behavioral Sciences, 195, 705-712. Moon, H. R., & Weidner, M. (2015). Linear regression for panel with unknown number of factors as interactive fixed effects. Econometrica, 83(4), 1543-1579. Moore, N., Salter, A., Stanley, L., & Tamboukou, M. (2016). The Archive Project: archival research in the social sciences. Routledge. Ngaruiya, B., Bosire, M. & Kamau, M. S. (2014). Effect of Mobile Money Transactions on Financial Performance of Small and Medium Enterprises in Nakuru Central Business District. Research Journal of Finance and Accounting. Vol.5, No.12. pp 53- 59 Park, C. Y., & Mercado, R. V. (2016). Does Financial Inclusion Reduce Poverty and Income Inequality in Developing Asia?. In Financial Inclusion in Asia (pp. 61-92). Palgrave Macmillan, London. Rewilak, J. (2017). The role of financial development in poverty reduction. Review of development finance, 7(2), 169-176. 61 University of Ghana http://ugspace.ug.edu.gh Samer, S., Majid, I., Rizal, S., Muhamad, M. R., & Rashid, N. (2015). The impact of microfinance on poverty reduction: Empirical evidence from Malaysian perspective. Procedia-Social and Behavioral Sciences, 195, 721-728. Saunders, M. N. (2011). Research methods for business students, 5/e. Pearson Education India. Sayed, Y., & Ghalib, A. K. (2016). Lifting the Urban Poor Out of Poverty: Assessing the Role of Non‐banking Financial Institutions in India. Strategic Change, 25(5), 585-601. Sekabira, H., & Qaim, M. (2017). Mobile money, agricultural marketing, and off‐farm income in Uganda. Agricultural Economics, 48(5), 597-611. Sethi, D., & Acharya, D. (2018). Financial inclusion and economic growth linkage: some cross- country evidence. Journal of Financial Economic Policy. Simiyu, N. C. & Oloko, M. (2015). Mobile money transfer and the growth of small and medium sized enterprises in kenya a case of kisumu city, Kenya. International Journal of Economics, Commerce and Management. Vol. III, pp 1056-1081 Williams, H. T., Adegoke, A. J., & Dare, A. (2017). Role of Financial Inclusion in Economic Growth and Poverty Reduction in A Developing Economy. Internal Journal of Research in Economics and Social Sciences, 7(5), 265-271. World Bank. (2013). End Extreme Poverty 2030 Promote Shared Prosperity. Annual Report, Washington DC. World Bank. (2015, April 17). Retrieved September 30, 2018, from The World Bank: http://www.worldbank.org/en/news/press-release/2015/04/17/world-bank-group- coalition-partners-make-commitments-accelerate-universal-financial-access 62 University of Ghana http://ugspace.ug.edu.gh World Bank. (2018). Poverty and Shared Prosperity 2018: Piecing Together the Poverty Puzzle. Annual Report, Washington DC. World Bank. (2018, May 14). Retrieved September 29, 2018, from The World Bank Group: http://www.worldbank.org/en/region/afr/overview#1 Yakubu, I., Dinye, R., Buor, D., & Iddrisu, W. A. (2017). Discriminant Analysis of Demand- Side Roadblocks to Financial Inclusion in Northern Ghana. Journal of Mathematical Finance, 7(03), 718. 63