Impact of institutions on financial inclusion in Africa

Łaszewski, Antoni Ludwik Article Impact of institutions on financial inclusion in Africa Financial Internet Quarterly Provided in Cooperation with: University of Information Technology and Management, Rzeszów Suggested Citation: Łaszewski, Antoni Ludwik (2024) : Impact of institutions on financial inclusion in Africa, Financial Internet Quarterly, ISSN 2719-3454, Sciendo, Warsaw, Vol. 20, Iss. 2, pp. 42-61, https://doi.org/10.2478/fiqf-2024-0011 This Version is available at: https://hdl.handle.net/10419/329872 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/3.0/ 10.2478/fiqf-2024-0011 Abstract Financial inclusion, for which the keystone is access to a bank account, is crucial to overcome the socioeconomic backwardness of African countries and to improve the African societies’ wellbeing. The study concentrates on this continent to better understand the nature of its development in terms of financial inclusion. The research aims to identify the institutions’ impact on financial inclusion in 35 African countries in the years 2010-2019. The analysis is based on a panel model with fixed individual effects. Novelty of the study rests in incorporation of four institutional variables: constraints on the executive, resolving insolvency, property rights, and WGI. The results showed a positive and statistically significant impact of resolving insolvency on financial inclusion (a measure covering people with financial institution accounts) across the entire sample. However, this relationship is especially visible in more developed countries, while constraints on the executive turned out to be crucial for low-income countries. Another novelty of the study is creation of an index of financial inclusion covering Mobile Money which was used to verify the obtained results. In this case, no positive impact of any institutional variable was identified which may mean that a favourable institutional environment is not required for the development of Mobile Money. JEL classification: E02, G21, K00 Keywords: Financial Inclusion, Banking Sector, Mobile Money, Institutions, Sustainable Development Received: 08.10.2023 Accepted: 13.05.2024 Cite this: Łaszewski, A.L. (2024). Impact of institutions on financial inclusion in Africa. Financial Internet Quarterly 20(2), pp. 42-61. © 2024 Antoni Ludwik Łaszewski, published by Sciendo. This work is licensed under the Creative Commons Attribution-NonCommercialNoDerivatives 3.0 License. 1 Narodowy Bank Polski and Faculty of Economic Sciences and Management, Nicolaus Copernicus University in Toruń, Poland, e-mails: antonilasze[email protected]m, ORCID: https://orcid.org/0000-0001-9706-1413. success achieved by M-Pesa founded in Kenya in 2007. At the same time, there is a high diversity of financial inclusion (measured as a percentage of the population with an account) across African countries, and the way in which financial services are delivered to society is also different. In some, especially poorer countries, a significant percentage of the population has only a Mobile Money account. In others, people use the services of financial institutions. which include: banks, credit unions, microfinance institutions, and post offices (Demirgüç-Kunt et al., 2022). Because of the benefits mentioned above, ensuring broad access to financial services is one of the goals of many development programs for poor countries. It is one of the Sustainable Development Goals, in particular, the tasks related to reducing transaction costs of remittances, empowering economic inclusion and ensuring access to financial services (goals 1.4, 8.3, 10.C) (United Nations, 2015). Also, the World Bank (2018) in its "Universal Financial Access 2020" initiative, strives to increase global financial inclusion by focusing on 25 priority countries, almost half of which (12 countries) are in Africa. The actions are concentrated on improving access to payment infrastructure and reaching disadvantaged populations (women, rural population), as well as institutional issues: creating a favourable regulatory environment and disbursing social benefits from the government into bank accounts. The article aims to assess how the institutional environment affects the level of financial inclusion in Africa. The approach is based on a new institutional economics framework, which confirmed that institutions strongly influence economic development (Kuncic, 2014). Institutions may be defined in various ways. North (1990, p. 3) defined them as “the rules of the game in a society or (…) the humanly devised constraints that shape human interaction.” Institutions must be enforced and commonly used (and can be formal as well as informal) thanks to which they structure and reduce uncertainty in human interaction. Hodgson (2006) pointed out that institutions are both rules (formal) and social norms (informal) which determine people's behaviour in certain situations (in case of X do Y, even if other options are available). Institutions can be defined also as an outcome of the game (e.g. political order and effective state), (Greif, 2006). Glaeser et al. (2004) indicated a mutual relationship between institutions and country development. It is possible, because countries with better education, create stronger institutions – people learn to dispute resolutions peacefully. On the other hand, democratic countries with a strong institutional framework, are more willing to invest more in education. Both relations boost economic development. Because of the different forms of financial services are provided in Africa, the article presents two different Financial inclusion, defined by the World Bank (2022a) as a situation in which "individuals and businesses have access to useful and affordable financial products and services that meet their needs – transactions, payments, savings, credit and insurance – delivered in a responsible and sustainable way" is one of the factors supporting sustainable economic development and improving the well-being of society. This is possible because financial services facilitate many everyday activities and create new opportunities for their users. Basic applications include safe saving, efficient receipt of government aid, simple and cheap transfers, including those between family members who emigrate for work and in the event of negative shocks (such as illness or unemployment), (Suri & Jack, 2011; Suri & Jack 2016). Having a bank account also enables credit assessment based on the data from the bank statement, which allows for better verification of the borrower (Ahmad et al., 2020; Chatterjee, 2020). This should result in lower interest rates and easier access to credit (Banerjee & Duflo, 2011). Flexible payment methods for access to water or electricity are also possible – in the instalment or PAYG (pay as you go) system. Thanks to the latter solution, the user does not have to bear the entire expenditure related to the installation of the infrastructure at once, but only pays for current usage (Ikeda & Liffiton, 2019). Access to financial services is also one of the factors reducing gender inequality in Africa (Xu et al., 2022). A current account is a keystone of financial inclusion and a gateway to other financial services (World Bank, 2019b). Although this study understands "financial inclusion" (mainly) as having a bank account – which enables subsequent use of other financial services – various definitions of this term occur. Financial inclusion may be defined also as access to other products, such as microcredit. In this case, it may have a negative impact on society, as the growth of microcredit (if not sustainable) leads to the expansion of low-productivity self-employment and increased competition (decreasing profits) in these sectors. This is not conducive to the development of countries, which is generally driven by increased productivity and innovative enterprises (engagement of capital and employment in these sectors). Over-indebtedness of borrowers (whole communities) may also occur, especially when microcredit is used for consumption rather than investment (Bateman et al., 2019). Regardless of the dynamic development of the financial sector in the last two decades, Africa remains a region with a high level of unbanked population (Demirgüç-Kunt et al., 2022). The situation is improving at a moderate pace, even despite the dynamic development of Mobile Money (MM), which began with the 2020; Sawadogo & Semedo, 2021) in which financial inclusion was measured by covering only the banking sector. The literature examining the determinants of the level of financial inclusion is quite extensive, focusing on institutional or socio-economic factors. The first type of research consists of studies examining the impact of institutions on the level of financial inclusion using econometric methods. Nkoa & Song, (2020) studied the impact of Worldwide Governance Indicators on the level of financial inclusion using a panel model with sys-GMM. The study also considered infrastructural and socio-economic variables. It showed the positive and significant impact of WGI on financial inclusion in Africa. Other studies also confirmed the positive impact of institutions and economic freedom (Chinoda & Kwenda, 2019), institutions (Law & Habibullah, 2009), the rule of law (Omar & Inaba, 2020) on the level of financial inclusion. A study presented by Anarfo et al. (2020) showed that excessive tightening of capital requirements may lead to a decrease in financial inclusion in Sub-Saharan African countries. This is the result of limiting the lending possibility of the financial sector, although the negative impact of the above-mentioned regulation does not affect a stable financial system. On the other hand, research by Sarma and Pais (2011) showed that a large share of foreign capital in the banking sector negatively affects financial inclusion. Because of asymmetries of information, foreign banks use the "cherry-picking" strategy and serve only wealthy and profitable customers. This is one of the effects of globalization and deregulation of global finance, in which banks are moving away from offering services to all customers in a given area (country) to serving the most profitable customers around the world (multinational banks). In the case of the credit market, this means focusing on long-term relationships with customers perceived as safe (even if they live in less "safe" countries) and neglecting other customers – who thus become excluded from the financial system (Dymski, 2005; 2009). Kebede et al. (2021) also observed a negative impact of foreign banks on the level of inclusion when examining only African countries. However, after a detailed analysis of the phenomenon, the authors indicated that this effect is only visible in countries with weak institutions (where banks are uncertain about creditworthiness assessment, so serve only the most trusted/profitable customers). In countries with a favourable institutional environment, foreign banks increase inclusion due to greater efficiency. Other studies examined the impact of the economic situation on financial inclusion, analysing it from the ways of measuring financial inclusion. The main one, reflecting the percentage of the population having an account in a financial institution; the additional one (used for a robustness check) also includes people having a Mobile Money account. Measure of impact of institutions on financial inclusion was calculated using a panel model covering 35 countries (belonging to all income groups in World Bank classification) in the years 2010-2019. This period is particularly interesting due to the dynamic development of the financial sector in Africa. The World Bank (2024a) studies conducted in 2011, 2014, 2017 show that the average percentage of the African population with an account more than doubled between the first and last survey – from 20.8% to 42.2%. However, this growth was not even. Among the 27 countries with a level of financial inclusion below 50% of the population in 2011 (only South Africa and Mauritius exceeded this level – due to their high initial position, they could not count on spectacular growth in the following years), there are countries that have financially included from 3% to 40% of their population within 6 years – the average increase in inclusion in this group was 22%, with a standard deviation of 10%. The large variation in the pace of development of the financial system in this period makes it particularly valuable to identify factors that support financial inclusion, as well as those that hinder it. The article puts forward the following hypotheses: H1: Institutions positively affect financial inclusion in Africa, H2: Development of Mobile Money reduces the impact of institutions on financial inclusion in comparison to analysis concentrated on the banking sector (and other financial institutions). The article extends current literature in two aspects. First, it presents a model with a larger and more diverse number of institutional variables. Previous studies incorporated mostly one or two institutional variables, concentrating on Worldwide Governance Indicators (WGI), (Ajide, 2017; Chinoda & Kwenda, 2019; Kebede et al., 2021; Nkoa & Song, 2020; Omar & Inaba, 2020). The methodology used in this research allows for a better estimation of which institutions influence the level of financial inclusion and makes it possible to indicate prioritised actions. Secondly, it presents two ways of constructing the index of financial inclusion, which can be built on an annual basis based on the available data. The first one highly correlates with the percentage of population having an account with a financial institution. The second includes individuals using Mobile Money, it is a novelty in comparison to earlier research (Ajide, 2017; Chatterjee, 2020; Ehigiamusoe et al., 2021; Evans & Adeoye, 2016; Kebede et al., 2021; Nkoa & Song, 2020; Omar & Inaba, (2020) observed a positive impact of institutional variables as rule of law and control of corruption, on equal distribution of income among 40 African countries. A relationship between GDP growth and institutions was identified by Wandeda et al. (2021). This study discovered that impact varies across the continent and is particularly visible in low income and West African countries. Michalopoulos and Papaioannou (2013) observed the impact of pre-colonial institutions on the current level of countries’ development in the studied continent. Frankel (2010) trying to explain the success of Mauritius, discovered institutions and lack of ethnic conflicts as a foundation of the country’s (relative) prosperity. Financial inclusion is determined by various factors. Because institutions are only one of them, socioeconomics and infrastructural measures were incorporated as control variables in the model (the selected variables reflect the factors most frequently studied in the literature, for which it was possible to find complete data). All selected variables are presented in Table 1. According to the literature, they cover the most important factors influencing financial inclusion. Creation of an index of financial inclusion will be presented in the next subsection. Selection of institutional variables to include in the model requires a choice between different approaches found in the literature. While some researchers use a narrow definition of institutions, recognizing them as rules and constraints shaping interactions between individuals, others include also "outcomes of the game" (Kowalewska et al., 2023). In the case of the latter approach, the possibility of identifying the impact of institutions on the development/income is limited because both measures will change in parallel, due to their mutual dependence. A similar problem applies to the detailedness of institutional measurement. On the one hand, it should be as precise as possible, referring to a specific permanent "rule", which provides an objective and thorough examination of a given institutional area. On the other hand, such an approach means that it is necessary to measure a very large number of variables in order to obtain an overall picture, which poses a risk of omission of important but difficult to measure factors and may not fully reflect people's opinions about the institutional environment (Individuals' decisions, for example regarding investments, depend on subjective opinion about the quality of the institutions – so they may impact economic outcomes), (Voigt, 2013). For this reason, aggregated institutional measures, even if they do not meet all the economic situation on financial inclusion, analysing it from the demand side (they identify factors that make a given person more likely to have an account). Research by Ehigiamusoe et al. (2021) indicated that GDP growth boosts financial development but only in high and middle-income countries. This means that after reaching a certain level of income, the market is large enough to stimulate the development of financial enterprises. In turn, the study conducted by Yangdol and Sarma (2019) showed a positive impact of GDP size on financial inclusion (measured as having an account in a financial institution) regardless of income. The impact of GDP on having an MM account is negative in poor countries (along with income growth, people replace MM with financial institutions). The study also found that policies should target the poor, the unemployed, the illiterate and women to increase inclusion, as they are less likely to have accounts. However, in some other studies, GDP was not identified as a key factor for the development of the banking sector (Cherif & Dreger, 2016), which was interpreted by engagement of banks in financing public debt, instead of the developing private sector – significant effect was instead observed in the case of the capital market development (Naceur et al., 2014). Naceur et al. (2015) described structural factors, such as population size and density, as well as informal economic size (People and entrepreneurs in the informal sector do not use financial institutions because they record transactions and require declaration of income/assets. However, development of the financial sector may negatively impact the size of the informal sector – when the benefits of access to financial services (e.g. obtaining a loan) exceed the costs of paid taxes (Blackburn et al., 2012; Njangang et al., 2020), which simultaneously with political factors impact financial inclusion. Their impact, as well as the size of GDP, is important because they affect the efficiency of financial sector operation. In the case of many people living in a small area, providing access to ATMs or banking branches is cheaper and the expected volume of loans/deposits is higher (the level of GDP has a positive impact on the demand for financial services), (Evans & Adeoye, 2016; Fowowe, 2014). It means that some countries may have a lower level of inclusion despite good policies and macroeconomic stability. Therefore, it is valuable to include structural factors while studying financial inclusion. Other types of research worth mentioning are those analysing impact of institutions on social wellbeing and economic development in Africa. The study performed by Sawadogo and Semedo (2021) found a negative relation between financial inclusion and income inequalities which occurs only in countries with strong institutions. On the other hand, Kunawotor et al. while the World Governance Indicators were used as a synthetic variable reflecting the general institutional environment in the surveyed countries. This indicator is quite popular and has been used in previous studies on similar topics (Kebede et al., 2021; Nkoa & Song, 2020; Omar & Inaba, 2020). above-mentioned requirements, provide a valuable approximation of the overall institutional level (Kowalewska et al., 2023). Therefore, two types of institutional measures were included in the study. Constraints on the executive, property rights, and resolving insolvency are objective measures which refer to a specific institution, Table 1: Variables used in the model Variable Abbreviation Description Source FINANCIAL INCLUSION - ENDOGENOUS VARIABLE Index of financial inclusion FIN Index created by calculation average of standardized variables of number of ATMs per 100,000 adults, number of bank branches per 100,000 adults, value of outstanding bank deposits to GDP (values from 0 to 1) (Deposits) FAS, International Monetary Fund (2021); (ATM, bank branches) WDI, World Bank (2022b) INSTITUTIONAL Constraints on the executive CONSTRAINTS Index created by calculation average of judicial and legislative constraints (values from 0 to 1) V-Dem, Coppedge et al. (2022) Resolving insolvency RESOLVING Score Resolving insolvency (cost, time, outcome; values from 0 to 100) Doing Business, World Bank (2021) Property rights PROPERTY Property rights (right to acquire, possess, inherit, and sell private property; values from 0 to 1) V-Dem, Coppedge et al. (2022) World Governance Indicators WGI Index created by calculation average of Voice and Accountability, Political Stability and Absence of Violence/Terrorism, Government Effectiveness, Regulatory Quality, Rule of Law, Control of Corruption (values from -2.5 to +2.5 ) WGI, World Bank (2022c) SOCIOECONOMIC Education EDUCATION Level of education (completion, enrolment, equality, quality, human resources involved; values from 0 to 100) 2020 IIAG, Mo Ibrahim Foundation (2020) GDP per capita GDP Gross domestic product per capita, PPP (constant 2017 international dollars) WDI, World Bank (2022b) Inflation INFLATION Inflation of consumer prices (annual; percentage) WDI, World Bank (2022b) Urbanization URBANIZATION Urban population (percentage of the total population) WDI, World Bank (2022b) INFRASTRUCTURAL Infrastructure INFRASTRUCTURE Level of infrastructure (digital communication, access to energy, transport network; values from 0 to 100) 2020 IIAG, Mo Ibrahim Foundation (2020) Source: Own preparation. fore, a positive correlation between them is understandable. Inflation (which negatively impacts economic development) has no positive and significant correlation with any other variables. On the other hand, urbanization may boost development, due to lower transport costs and better access to infrastructure in urban areas (Naceur et al., 2015) – thus positive and significant correlation with other variables occurs. Descriptive statistics of all variables (including those used in index construction) are presented in Table 2, while Table 3 shows correlation among them (because index of financial inclusion – FIN – will be used as a dependent variable in the model, it was added in the correlation matrix). Higher values of all variables, except inflation and urbanization, indicate higher country development or stronger institutions. There- does not count only people who have an account (and may not use it), but also considers ease of access (e.g., the ability to withdraw/deposit cash to the account) and real use (value of credit/deposit). Drawing from accessible data, the Index of Financial Inclusion (FIN) built in this study will include only two dimensions (Data on the account ownership are published by World Bank (2024a) in Findex database with a few-year interval. They will be used for index evaluation in in the following parts) availability (bank According to Sarma (2008) financial inclusion has three dimensions: penetration, availability, and usage. They may be measured by respectively: number of accounts per capita or percentage of population having an account (penetration); number of bank branches/ ATM/bank employees per capita (availability); volume of credit/deposit to GDP (usage). Taking into account all dimensions provides a full picture of the level of financial inclusion. This is possible because the index Table 2: Descriptive statistic of variables Variable Mean Standard deviation Minimum Maximum Number of ATMs per capita 14.7 17.4 0.4 90.0 Number of bank branches per capita 8.3 10.0 0.6 55.1 Value of bank deposits to GDP 35.8 28.6 4.9 169.3 CONSTRAINTS 0.6 0.3 0.1 0.9 RESOLVING 34.9 15.5 0.0 69.1 PROPERTY 0.7 0.2 0.1 0.9 WGI -0.5 0.5 -1.6 0.9 EDUCATION 52.3 14.7 23.2 84.3 GDP 6,357.0 6,852.0 959.0 37,571.0 INFLATION 5.9 14.4 -3.2 255.3 URBANIZATION 43.4 16.6 15.5 73.2 INFRASTRUCTURE 41.3 19.4 6.7 88.3 Note: score resolving insolvency is zero in case of “no practise” of resolving insolvency – zero insolvency cases over past five years. Disregarding these cases, the smallest value of this variable is 15.7 Source: Own preparation. Table 3: Correlation matrix of variables used in the model ATMs Bank branches Value of deposits FIN CONSTRAINTS RESOLVING PROPERTY WGI EDUCATION GDP INFLATION URBANIZATION INFRASTRUCTURE 1.00 0.79 0.58 0.90 0.44 0.21 0.10 0.67 0.60 0.64 -0.04 0.43 0.68 1.00 0.61 0.91 0.38 0.06 0.14 0.58 0.61 0.55 -0.08 0.41 0.61 1.00 0.82 0.35 0.30 0.31 0.57 0.63 0.43 -0.07 0.31 0.72 1.00 0.44 0.21 0.20 0.69 0.70 0.62 -0.07 0.44 0.76 1.00 0.23 0.39 0.67 0.41 0.05 -0.01 0.04 0.33 1.00 0.17 0.37 0.39 0.09 -0.08 -0.08 0.43 1.00 0.29 0.18 -0.04 -0.09 0.34 0.26 1.00 0.68 0.35 -0.10 0.21 0.53 1.00 0.45 -0.02 0.37 0.76 1.00 -0.05 0.56 0.55 -1.00 -0.05 -0.02 1.00 0.58 1.00 Note: Correlation coefficient higher or equal to 0.14 is significant at 1%; 0.11 at 5%; 0.9 at 10% (Number of observations: 350), (Obilor & Amadi, 2018) Source: Own preparation. ATMs Bank branches Value of deposits FIN CONSTRAINTS RESOLVING PROPERTY WGI EDUCATION GDP INFLATION URBANIZATION INFRASTRUCTURE ATMs Bank branches Value of deposits FIN CONSTRAINTS RESOLVING PROPERTY WGI EDUCATION GDP INFLATION URBANIZATION INFRASTRUCTURE (1) Where Xi is the actual value of the variable, mini and maxi are respectively the lowest and the highest value of the variable. Standardized values are <0;1>. In the next step, the arithmetic mean of the standardized values (bank branches, ATMs, value of bank deposits) was calculated to obtain the index. A similar approach is used, among others, by the United Nations Development Program (2007) to calculate the Human Development Index. It assumes the possibility of substitution between individual variables (in the case of this index, bank branches can be a substitute for ATMs). The result is the index of financial inclusion (FIN), whose values are in the <0;1> range. Since an account is the basis of financial inclusion and a gateway to use ATMs or make deposits, it is valuable to present the correlation between index FIN and account ownership 15+. This allows assessing the quality of the created index. However, a 100% correlation is not expected, as the accounts held may be used very rarely, due to the lack of appropriate infrastructure (which FIN reflects). Chart 1 shows the relationship between the percentage of the population having an account with a financial institution and the FIN rate in 2011, 2014, and 2017. The equation of the relationship between the variables is presented in Table 3. It was calculated as follows: (2) where “i” denotes country. The error term is signed by “ε”. branches and ATMs per capita), and usage (volume of bank deposits to GDP). Excluding countries with missing data (In three cases, the data lacked a single value: number of bank branches per 100,000 adults in Equatorial Guinea in 2016, outstanding deposits with commercial banks (% of GDP) in Malawi in 2014, and legislative constraints on the executive in Egypt in 2015. Due to the linear course of the phenomena, in these cases forecast (average value from adjacent years) of missing values was used) the financial inclusion index covers 35 African (Algeria, Angola, Benin, Botswana, Burkina Faso, Cabo Verde, Cameroon, Chad, Côte d'Ivoire, Egypt, Equatorial Guinea, Eswatini, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Lesotho, Madagascar, Malawi, Mali, Mauritius, Morocco, Mozambique, Niger, Nigeria, Rwanda, Senegal, Seychelles, South Africa, Togo, Tunisia, Uganda, Zambia, Zimbabwe) in 2010-2019. The selected variables are commonly used in the literature, however differences in selection of particular variables can be observed. Many of the researches include ATMs and bank branches per capita (Ajide, 2017; Anarfo et al., 2020; Kebede et al., 2021; Nkoa & Song, 2020; Sawadogo & Semedo, 2021), although frequently used, in some as an additional variable (Anarfo et al., 2020; Kebede et al., 2021; Nkoa & Song, 2020; Sawadogo & Semedo, 2021), in some as the only one (Evans & Adeoye, 2016; Kumar, 2013; Omar & Inaba, 2020) is number of bank accounts or depositors per capita. Studies where the development of the financial sector is measured only by volume of credit to private sector (% of GDP) also occur (Ehigiamusoe et al., 2021; Law & Habibullah, 2009). The first stage of the index calculation was the standardization of the variables, using Zero Unitarization Method (Formula 1) (Kukuła, 2000; Sarma, 2008). − =− ii i ii X min Smax min 01 .    = ++ i ii Account ownership in fin institution FIN Chart 1: Correlation between account ownership in financial institution (age 15+) and FIN Source: Own preparation. mating a model which includes individual effect due to the occurrence of heteroscedasticity in residuals (Mátyás & Sevestre, 2008). Choice ought to be made between fixed effect (FE) and random effect (RE) models. RE model is given as follows: (4) Parameter νit is the error term of country i at time t, including the error component and random individual effect. FE model is calculated according to the following formula: (5) Parameter ε is the error term of country i at time t, ui is the individual fixed effect of country i. Hausmann test, makes it possible to choose between the FE and RE models. No basis to reject the null hypothesis means that the RE estimator is most effective. In another way the FE model should be used (Maddala, 2001). The model was estimated as described in the previous section. Rejection of null hypothesis in Wald, Breusch–Pagan and Hausmann tests means that the model with fixed individual effect is appropriate for the collected data. Results of all regressions are presented in Table 5. The obtained high correlation proves that the index well reflects the level of financial inclusion in each country. The study does not include the percentage of the population with a mobile money account, as MM is less regulated and provides less security. According to Ahman et al. (2020) in 2017, only Kenyan law provided insurance of deposits in MM. At the same time, deposit insurances are operating in almost half of the analysed countries (International Association of Deposit Insurers, 2022; World Bank, 2019a), and are under development in another six (International Association of Deposit Insurers, 2021). This means that people who have MM accounts in these countries and use them to save, take on greater risk than people saving in banks. A panel model can be created in a few ways. The most popular are ordinary least squares method, fixed effect, and random effect. Choice of the best one is possible by analysis of data and results obtained using several methods (Maddala, 2001). At the beginning of the model construction, the ordinary least squares method (OLS) was used (this method requires homogenous units). Its equation is given as follows: (3) here ε it is an error term of country i at time t. Its verification can be performed by the Wald and Breusch–Pagan test. If results of the Wald test indicate the lack of object homogeneity, another model should be calculated (Spierdijk, 2022). Rejection of null hypothesis in the Breusch–Pagan test also suggests estiTable 4: Results of regression of FIN on account ownership in financial institution (age 15+) and correlation coefficient 2011 2014 2017 Constant 0.0964*** 0.1074** 0.1837*** (0.0288) (0.0391) (0.0335) FIN (Index of financial inclusion) 1.2160*** 1.2089*** 0.9475*** (0.1765) (0.2103) (0.1678) R2 0.6600 0.5800 0.5400 Pearson correlation coefficient 0.8200*** 0.7600*** 0.7300*** Note: *** means variable/correlation significant at 1%; ** at 5%; * at 10% Source: Own preparation. 0 1 2 3 4 5 6 7 8 9 it it it it it it it it it it it FIN CONSTRAINTS RESOLVING PROPERTY WGI EDUCATION GDP INFLATION URBANIZATION INFRASTRUCTURE           = + + + + + + + + + + 0 1 2 3 4 5 6 7 8 9 it it it it it it it it it it it FIN CONSTRAINTS RESOLVING PROPERTY WGI EDUCATION GDP INFLATION URBANIZATION INFRASTRUCTURE v           = + + + + + + + + + + 0 1 2 3 4 5 6 7 8 9 it it it it it it it it it it i it FIN CONSTRAINTS RESOLVING PROPERTY WGI EDUCATION GDP INFLATION URBANIZATION INFRASTRUCTURE u           = + + + + + + + + + ++ business) do not offer the entire range of banking services, do not grant loans (do not bear credit risk) but only deposit their clients' funds in a pooled bank account – however, a stable and trusted banking sector facilitates this activity (Lal & Sachdev, 2015). Interestingly, even though MM aims to provide access to financial services to people living in rural areas, the positive impact of urbanization was still observed, although at a lower level of significance. Positive and statistically significant impact of infrastructure, is consistent with the theory, because good access to electricity/internet/mobile phones creates favourable conditions for the development of MM. The conducted research provided an identification and assessment of institutional determinants of financial inclusion, namely constraints on the executive, resolving insolvency, property rights, and WGI. The main part of the study focused on inclusion by access to financial institutional services (banks, credit unions, microfinance institutions, and post offices). Econometric estimation indicated significant impact of resolving insolvency on financial inclusion when studying the entire continent. However, detailed analysis has shown that this effect is particularly visible in wealthier countries (upper-middle and high income), in which WGI also positively affects financial inclusion, while in lowincome countries, constraints on the executive were identified as key to financial inclusion. In lower-middle The presented data indicate that the proposed index accurately reflects the percentage of the population with any type of account. The fit is significantly better than the previously presented FIN index. Due to large data gaps, especially in the first years of the study, 12 countries (from the main sample) in 2013-2019 (Cameroon, Eswatini, Ghana, Kenya, Lesotho, Madagascar, Mozambique, Nigeria, Rwanda, Uganda, Zambia, Zimbabwe. All countries with full data on MM outlets were included. The choice of the years 2013-2019 is dictated by the desire to examine a period as long as possible, without excessively reducing the number of countries included. Previously in this subsection (Tables 7, 8 and Charts 2, 3, 4), all countries for which data were available for a given year were considered) were chosen to create the robustness check model. Table 6 presents the results both for FIMM and FIN (in case of FIN only countries appearing in the FIMM model were included) – models 6 and 7 respectively. Half of the countries covered by the models were classified as low income, half as lower-middle income. Thus, results on the latter regression (FIN as a dependent variable) were similar to those for lowincome countries – differences only occur in the control variables: GDP and infrastructure. However, the results for the FIMM model are different. No significant impact of any of the institutional variables was identified, as well as education, GDP per capita and inflation. This may mean that a favourable economic and institutional environment is not required for the development of MM. This may be because MM providers (as their core Chart 4: Correlation between account ownership (age 15+) and FIMM Note: the chart shows the countries that were covered by the Findex study in a specific year (6, 12 and 23 countries, respectively). Source: Own preparation. ed the Central Bank Digital Currency (CBDC). In 2021, Nigeria, as the first African country (Africa's most populous economy), launched its digital currency – eNaira. The possibility of issuing CBDCs is being investigated by another 16 countries on this continent (Atlantic Council, n.d.). Currencies issued by central banks are indicated as one of the opportunities to increase access to financial services (Central Bank of Nigeria, n.d; Foster et al., 2021). The potential success of these initiatives will require further research, as well as redefining the ways of measuring financial inclusion (it will need to include people using only CBDC as well as MM). For example, the impact of resolving insolvency, identified especially in upper-middle and high income countries, may lose its importance. It is because commercial banks, in principle, hold loans to enterprises and individuals in assets, while central banks have foreign exchange reserves, loans to commercial banks and government bonds (Bindseil, 2020). For this reason, the indicated positive impact of this institutional variable, affecting the quality of banks' assets, and thus their potential for the development, will probably decrease along with the adoption of CBDC. The author would like to thank M. Szczepaniak for the supervision of this research, as well as all valuable comments and suggestions. All remaining mistakes or inadequacies are the sole responsibility of the author. This work was supported by the Nicolaus Copernicus University in Toruń from “Excellence Initiative – Research University” programme. income countries, no positive impact of any of the examined institutional variables was observed. Thus, the first hypothesis (H1) is partially supported – impact was observed but not in all cases. In comparison to earlier studies, it allows for constructing more precise recommendations for countries with large “financially excluded” populations. Firstly, it is important to strengthen creditors’ protection and reduce the time of insolvency proceedings. This will result in a smaller number of non-preforming assets within the banking sector and boost its development. Secondly, strenuous efforts should be made to conduct a predictable policy with clearly defined power of governors, especially in the least developed countries. Another outcome of the study is that the results vary depending on how financial inclusion is measured. The article presents a measure of financial inclusion which covers people using only mobile money, which was used for a robustness check. Such a measure shows greater levels of inclusion in countries with GDP per capita below USD 5000, where a substantial part of the population uses MM as the only account. Incorporation of such a measure as an endogenous variable in the model, resulted in identifying a previously significant institutional variable (constraints on the executive) as insignificant – impact of none of the institutional variables was significant. It may mean that the development of mobile money does not require (very) favourable institutional conditions, which would call for rethinking the impact of institutions on financial inclusion. It supports the second hypothesis (H2). Additionally, the study included years in which none of the countries in the world had yet implementAhmad, A.H., Green, C. & Jiang, F. (2020). Mobile money, financial inclusion and development: A review with reference to African experience. Journal of Economic Surveys 34(4), 753-792. Ajide, K.B. (2017). Determinants of financial inclusion in Sub-Saharan Africa countries: Does institutional infrastructure matter? CBN Journal of Applied Statistics 8(2), 69-89. Retrieved from https://www.econstor.eu/ handle/10419/191705 (Accessed: 08.10.2023). Allen, F., Demirguc-Kunt, A., Klapper, L. & Peria, M. (2016). The foundations of financial inclusion: Understanding ownership and use of formal accounts. Journal of Financial Intermediation 27, 1-30. Anarfo, E., Abor, J. & Osei, K. (2020). Financial regulation and financial inclusion in Sub-Saharan Africa: Does financial stability play a moderating role? Research in International Business and Finance 51, 1-16. Atlantic Council. Central Bank Digital Currency tracker. Retrieved from https://www.atlanticcouncil.org/cbdctracker/ (Accessed: 08.10.2023). Banerjee, A.V. & Duflo, E. (2011). Poor economics. PublicAffairs, New York. Bateman, M., Blankenburg, S. & Kozul-Wright, R. (2019). The rise and fall of global microcredit: development, debt and disillusion. Routledge, Abingdon. Bindseil, U. (2020). Tiered CBDC and the financial system. European Central Bank, Frankfurt am Main. Blackburn, K., Bose, N. & Capasso, S. (2021). Tax evasion, the underground economy and financial development. Journal of Economic Behavior & Organization 83(2), 243-253. Central Bank of Nigeria. Desgn Paper for the eNaira. Retrieved from https://www.enaira.gov.ng/download/ eNaira_Design_Paper.pdf (Accessed: 08.10.2023). Chatterjee, A. (2020). Financial inclusion, information and communication technology diffusion, and economic growth: a panel data analysis. Information Technology for Development 26(3), 607-635. Cherif, M. & Dreger, C. (2016). Institutional Determinants of Financial Development in MENA countries. Review of Development Economics 20(3), 670–680. Chinoda, T. & Kwenda, F. (2019). The impact of institutional quality and governance on financial inclusion in Africa: A two-step system generalised method of moments approach. Journal of Economic and Financial Sciences 12(1), 1-9. Coppedge, M., Gerring, J., Knutsen, C. H., Lindberg, S. I., Teorell, J., Alizada, N., Altman, D., Bernhard, M., Cornell, A., Fish, M. S., Gastaldi, L., Gjerløw, H., Glynn, A., Grahn, S., Hicken, A., Hindle, G., Ilchenko, N., Kinzelbach, K., Krusell, J., Marquardt, K.L., McMann, K., Mechkova, V., Medzihorsky, J., Paxton, P., Pemstein, D., Pernes, J., Ryd´en, O., von Romer, J., Seim, B., Sigman, R., Svend-Erik, S., Staton, J., Sundstrom, A., Tzelgov, E., Wang, Y., Wig, T., Wilson, S. & Ziblatt, D. (2022). V-Dem Dataset v12. Varieties of Democracy (V-Dem) Project, University of Gothenborg. Demirgüç-Kunt, A., Klapper, L., Singer, D. & Ansar, S. (2022). The Global Findex Database 2021: Financial inclusion, digital payments, and resilience in the age of COVID-19. World Bank, Washington DC, Retrieved from https:// www.worldbank.org/en/publication/globalfindex/Report (Accessed: 08.10.2023). Dymski, G. (2005). Financial Globalization, Social Exclusion and Financial Crisis. International Review of Applied Economics 19(4), 439-457. Dymski, G. (2009). The global financial customer and the spatiality of exclusion after the ‘end of geography’. Cambridge Journal of Regions, Economy and Society 2(2), 267-285. Ehigiamusoe, K.U., Guptan, V. & Narayanan, S. (2021). Rethinking the impact of GDP on financial development: Evidence from heterogeneous panels. African Development Review 33(1), 1-13. Evans, O. & Adeoye, B. (2016). Determinants of financial inclusion in Africa: A dynamic panel data approach. University of Mauritius Research Journal 22, 310-336. Retrieved from https://mpra.ub.uni-muenchen.de/81326/ (Accessed: 08.10.2023). Food and Agriculture Organization of the United Nations. (2023). Global population density estimates, FGGD. Retrieved from https://data.apps.fao.org/catalog/dataset/6d78b970-5883-11db-b1aa-000d939bc5d8 (Accessed: 08.10.2023). Foster, K., Blakstad, S., Gazi, S. & Bos, M. (2021). Digital currencies and CBDC impacts on least developed countries (LDCs), Dialogue on Global Digital Finance Governance Paper Series, New York. Fowowe, B. (2014). Law and finance revisited: Evidence from African countries. South African Journal of Economics 82(2), 193-208. Frankel, J.A. (2010). Mauritius: African success story. Harvard University Working Papers, Cambridge. Glaeser, E.L., La Porta, R., Lopez-de-Silanes, F. & Shleifer, A. (2004). Do institutions cause growth? Journal of Economic Growth 9(3), 271-303. Greif, A. (2006). Institutions and the path to the modern economy: Lessons from medieval trade, Cambridge University Press, Cambridge. Herger, N., Hodler, R. & Lobsiger, M. (2008). What Determines Financial Development? Culture, Institutions or Trade. Review of World Economics 144, 558–587. Hodgson, G. M. (2006). What are institutions? Journal of Economic Issues 40(1), 1-25. Ikeda, J., & Liffiton, K. (2019). Fintech for the water sector: Advancing financial inclusion for more equitable access to water, World Bank, Washington. International Association of Deposit Insurers. (2021). Systems under development. Retrieved from https:// www.iadi.org/en/about-iadi/deposit-insurance-systems/systems-under-development/ (Accessed: 08.10.2023). International Association of Deposit Insurers. (2022). IADI Annual Survey. Retrieved from https://www.iadi.org/en/ research/data-warehouse/deposit-insurance-surveys/ (Accessed: 08.10.2023). International Monetary Fund. (2021). Financial Access Survey. Retrieved from https://data.imf.org/?sk=e5dcab7e-a5ca -4892-a6ea-598b5463a34c&sId=1496861011734 (Accessed: 08.10.2023). Kebede, J., Selvanathan, S. & Naranpanawa, A. (2021). Foreign bank presence, institutional quality, and financial inclusion: Evidence from Africa. Economic Modelling 102, 1-16. Kowalewska, A., Osińska, M. & Szczepaniak, M. (2023). Institutions in the development of Sub-Saharan African countries in 2004–2019. Economics and Law 22(1), 103–125. Kukuła, K. (2000). Metoda unityzacji zerowanej [Zero unitarization method]. Wydawnictwo Naukowe PWN, Warszawa. Kumar, N. (2013) Financial inclusion and its determinants: evidence from India. Journal of Financial Economic Policy 5(1), 4-19. Kunawotor, M.E., Bokpin, G.A. & Barnor, C. (2020). Drivers of income inequality in Africa: Does institutional quality matter? African Development Review 32(4), 718-729. Kuncic, A. (2014). Institutional quality dataset. Journal of Institutional Economics 10(1), 135–161. Lal, R. & Sachdev, R. (2015). Mobile Money services – Design and development for financial inclusion. (Harvard Business School Working Paper 15-083). Retrieved from https://www.hbs.edu/ris/Publication%20Files/15083_e7db671b-12b2-47e7-9692-31808ee92bf1.pdf (Accessed: 08.10.2023). Law, S.H. & Habibullah, M.S. (2009). The determinants of financial development: Institutions, openness and financial liberalisation. South African Journal of Economics 77(1), 45-58. Maddala, G.S. (2001). Introduction to econometrics. John Wiley & Sons, Chichester. Mátyás, L. & Sevestre, P. (2008). The econometrics of panel data, Springer, Berlin. Michalopoulos, S. & Papaioannou, E. (2013). Pre-colonial ethnic institutions and contemporary African development. Journal of the Econometric Society 81(1), 113-152. Mo Ibrahim Foundation (2020). Ibrahim Index of African Governance. Retrieved from https://iiag.online/ downloads.html (Accessed: 08.10.2023). Muthiora, B. (2015). Enabling mobile money policies in Kenya: Fostering a digital financial revolution. (GSMA Mobile Money for the Unbanked). Retrieved from http://t1.daumcdn.net/brunch/service/user/P5W/file/ mxAyGPsEeNdS1FAkt9vWxBJVHSg.pdf (Accessed: 08.10.2023). Naceur, S. B., Barajas, A. & Massara, A. (2015). Can Islamic Banking increase financial inclusion? (International Monetary Fund Working Paper No. 2015/031). Retrieved from https://www.imf.org/en/Publications/WP/ Issues/2016/12/31/Can-Islamic-Banking-Increase-Financial-Inclusion-42710 (Accessed: 08.10.2023). Naceur, S. B., Mondher, C. & Kandil, M. (2014). What drives the development of the MENA financial sector? Borsa Istanbul Review 14(4), 212-223. Njangang, H., Nembot, L.N. & Ngameni, J.P. (2020). Does financial development reduce the size of the informal economy in sub‐Saharan African countries? African Development Review 32(3), 375-391. Nkoa, B.O.N. & Song, J.S. (2020). Does institutional quality affect financial inclusion in Africa? A panel data analysis. Economic Systems 44(4), 1-13. North, D.C. (1990). Institutions, institutional change and economic performance. Cambridge University Press, Cambridge. Obilor, E.I. & Amadi, E.C. (2018). Test for significance of Pearson’s correlation coefficient. International Journal of Innovative Mathematics, Statistics & Energy Policies 6(1), 11-23. Omar, M.A. & Inaba, K. (2020). Does financial inclusion reduce poverty and income inequality in developing countries? A panel data analysis. Journal of Economic Structures 9(1), 1-25. Sarma, M. (2008). Index of Financial Inclusion. Indian Council for Research on International Economic Relations Working Paper No. 215. Retrieved from https://www.econstor.eu/handle/10419/176233 (Accessed: 08.10.2023). Sarma, M. & Pais, J. (2011). Financial inclusion and development. Journal of International Development 23(5), 613-628. Sawadogo, R. & Semedo, G. (2021). Financial inclusion, income inequality, and institutions in sub-Saharan Africa: Identifying cross-country inequality regimes. International Economics 167, 15-28. Spierdijk, L. (2022). Assessing the consistency of the fixed-effects estimator: a regression-based Wald test. Empirical Economics 64, 1599-1630. Suri, T. & Jack, W. (2011). Mobile money: The economics of M-PESA. National Bureau of Economic Research, Cambridge. Suri, T. & Jack, W. (2016). The long-run poverty and gender impacts of mobile money. Science 354(6317), 1288-1292. United Nations (2015). Transforming our world: the 2030 Agenda for Sustainable Development. Retrieved from https://sdgs.un.org/2030agenda (Accessed: 08.10.2023). United Nations Development Programme (2007). Measuring human development: A primer. New York. Retrieved from https://hdr.undp.org/content/measuring-human-development-primer (Accessed: 08.10.2023). Voigt, S. (2013). How (Not) to measure institutions. Journal of Institutional Economics 9(1), 1-26. Wandeda, D.O., Masai, W. & Nyandemo, S.M. (2021). Institutional quality and economic growth: evidence from SubSaharan Africa countries. African Journal of Economic Review 9(4), 106-125. World Bank (2018). UFA2020 overview: Universal Financial Access by 2020. Retrieved from https:// www.worldbank.org/en/topic/financialinclusion/brief/achieving-universal-financial-access-by-2020 (Accessed: 08.10.2023). World Bank (2019a). Deposit insurance dataset. Retrieved from https://datacatalog.worldbank.org/search/ dataset/0040209 (Accessed: 08.10.2023). World Bank (2019b). Fintech: The experience so far - executive summary. Retrieved from http:// documents.worldbank.org/curated/en/130201561082549144/Fintech-The-Experience-so-Far-ExecutiveSummary2022 (Accessed: 08.10.2023). World Bank (2021). Doing Business Data. Retrieved from https://archive.doingbusiness.org/en/data (Accessed: 08.10.2023). World Bank (2022a). Financial inclusion. Retrieved from https://www.worldbank.org/en/topic/financialinclusion/ overview (Accessed: 13.05.2024). World Bank. (2024a). The Global Findex Database 2021. Retrieved from https://www.worldbank.org/en/publication/ globalfindex/Data (Accessed: 13.05.2024). World Bank. (2024b). World Bank Country and Lending Groups. Retrieved from https://datahelpdesk.worldbank.org/ knowledgebase/articles/906519-world-bank-country-and-lending-groups (Accessed: 13.05.2024). Xu, S., Asiedu, M. & Effah, N.A.A. (2023). Inclusive finance, gender inequality, and sustainable economic growth in Africa. Journal of the Knowledge Economy 14(4), 4866-4902. Yangdol, R. & Sarma, M. (2019). Demand-side factors for financial inclusion: A cross-country empirical analysis. International Studies 56(2-3), 163-185. Zins, A. & Weillb, L. (2016). The determinants of financial inclusion in Africa. Review of Development Finance 6(1), 46-57.