WP/21/ 90 Measuring Digital Financial Inclusion in Emerging Market and Developing Economies: A New Index by Purva Khera, Stephanie Ng, Sumiko Ogawa and Ratna Sahay IMF Working Papers describe research in progress by the author(s) and are published to elicit comments and to encourage debate. The views expressed in IMF Working Papers are those of the author(s) and do not necessarily represent the views of the IMF, its Executive Board, or IMF management.2 2021 Internati ona l Moneta ry Fund WP/21/90 IMF Working Paper Monetary and Capital Markets Department Measuring Digital Financial Inclusion in Emerging Market and Developing Economies: A New Index Prepared by Purva Khera, Stephanie Ng, Sumiko Ogawa, and Ratna Sahay 1 Authorized for distribution by Ulric Eriksson von Allmen March 2021 Abstract Adoption of technology in the financial services industry (i.e. fintech) has been accelerating in recent years. To systematically and comprehensively assess the extent and progress over time in financial inclusion enabled by technology, we develop a novel digital financial inclusion index. This index is based on payments data covering 52 developing countries for 2014 and 2017, taking into account both access and usage dimentions of digital financial services (DFSs). This index is then combined with the traditional measures of financial inclusion in the literature and aggregated into an overall index of financial inlusion. There are two key findings: first, the adoption of fintech has been a key driver of financial inclusion. Second, there is wide variation across countries and regions, with the greatest progress recorded in Africa and Asia and the Pacific regions. This index should offer a useful analytical tool for researchers and policy makers. JEL Classification Numbers: C38, G10, G20, O30 Keywords: Fintech; digital financial services; financial inclusion Authors E-Mail Address: pkheraimf; sngimf; sogawaimf; rsahayimf 1 The auth o rs wi sh to thank To bias Ad ria n, Martin ihk, Ulric Eriksson von Allmen, Amina Lahreche, Kimberly Beaton, a nd Majid Bazarbash f or their guida nce, support and im mensely helpf ul f eedback and commen ts . Wed al so l i ke to than k Itai Ag ur, Pa tricia Alo n so-Gamo, Pelin Berkmen, Nicolas Blancher, Era Dabla -Norris, Bid ish a Das, Jen nif er Ellio t, Fed erico Grin b erg, Dirk Ja n Gro llem an, Frederic Lambert, Elena Lukoianova, Inutu Lukonga, Jennife r Moyo, Maria Soledad Martinez Peria, Nicolas Racine , Celine Roc hon, Kazuko Shi ron o, Amad ou Sy, Jose Torres, Tomohiro Tsuruga, Hec tor Carcel Vilanova for the ir helpful commen ts ; an d parti ci p an ts of vari ou s semi n ars an d con fere nc es i n whi ch we presen te d the key fi ndi n g s. IMF Working Papers describe research in progress by the author(s) and are published to elicit comments and to encourage debate. The views expressed in IMF Working Papers are those of the author(s) and do not necessarily represent the views of the IMF, its Executive Board, or IMF management.2 Table of Contents Abstract _ 2 I. Introduction_ 3 II. Literature Revi ew _ 5 III. Methodology _ 6 IV. Indicators of Financial Inclusion: Stylized Facts _ 9 V. Findings _ 11 A. Traditional Financial Inclusion Index _ 11 B. Digital Financial Inclusion Index _ 12 C. Comprehensive Financial Inclusion Index _ 13 VI. Conclusion _ 16 References _ 18 Appendix I: Composition of various financial inclusion indices_ 21 Appendix II: Construction of the Financial Inclusion Index _ 223 I. INTRODUCTION The rapid adoption of digital technology in finance offers a large potential to increase financial inclusion, namely, access to and usage of financial services by a wide section of the population. Digital financial services (DFSs), enabled by fintech (technological innovation in the financial sector), can help overcome the often-cited obstacles in accessing traditional financial services such as cost, geographical barriers, and information asymmetry. Recognizing this potential, the United Nations, Sustainable Development Goals include targets both on traditional and digital inclusion measures (Target 8.10). There are several anecdotal evidence, including country-based case studies (Jack and Suri, 2011; 2014; Tarazi and Breloff, 2010) and regional studies (Sy et. al., 2019; Berkmen et. al., 2019; Loukoianova et al, 2019, Lukonga, 2018, and Blancher et al., 2019), that show how fintech is increasing access to financial services, especially for those previously unbanked or underserved. Existing literature primarily focuses on financial inclusion facilitated by financial institutions such as banks, i.e. traditional financial inclusion. This is measured by indicators related to access to and/or usage of traditional financial services, such as the number of bank account per capita and ATM per capita, or combining these indicators into a composite index (Beck, Demirguc-Kunt and Martinez Peria, 2007, Honohan, 2008). Some recent studies have quantified the degree of digital financial inclusion by looking at relevant indicators, such as mobile money accounts and financial transactions using mobile phone (Sy et al., 2019; Loukoianova et al., 2019; Camara and Tuesta, 2017). However, these measures capture a single aspect of digital financial inclusion at a time, and do not present a comprehensive picture of combining multiple aspects including access and usage. This paper aims to fill this gap in the existing literature by incorporating both measures of access to and usage of DFSs into the measurement of financial inclusion. The key contribution of this paper is the construction of a digital financial inclusion index, covering 52 emerging markets and developing economics (EMDEs) for 2014 and 2017. The index is composed of indicators related to access to and usage of financial services provided through fintech, taking advantage of the new and expanded data coverage of the World Bank Global Findex Database and IMFs Financial Access Survey (FAS) data series on mobile money and other means of online financial services. We then construct a comprehensive aggregate index of financial inclusion, combining digital financial inclusion index with an index of financial inclusion through traditional financial institutions such as banks (traditional financial inclusion index), similar to those in existing literature. We construct a total of seven indices, which capture the degree of financial inclusion in each country through various dimensions. A three-stage principal component analysis (PCA), a commonly used objective weighting methodology in the literature is employed to determine the weight on each indicator. The first stage computes the access and usage sub-indices. Access is primarily captured by indicators related to availability of means to access payment s services (e.g., number of bank branches and ATM, and accesibility to the internet and mobile phone). Usage focuses on demand-side elements, such as account ownership and making/receiving payments through these accounts. The second stage then combines these sub-indices into separate indices for traditional and digital financial inclusion, and a weighted4 combination of these forms an aggregate measure of overall financial inclusion at the third stage. The aggregate views help assess the overall advance in financial inclusion, whereas t he granular view along the usage and access and digital and traditional aspects helps understand the drivers of changes and are helpful to inform policymakers in developing appropriate measures. Our indices exclusively focus on the payment aspects of financial inclusion. 2 This reflects the fact that payments are often the first step and the gateway to gaining access to financial services, while other aspects of financial services, such as credit and insurance, tend to come later with financial development and deepening. Moreover, cross-country comparable data on other measures of financial inclusion (credit, savings, insurance) are still not available. Our financial inclusion indices have several advantages over past measures. First, it provides a more comprehensive picture of financial inclusion by incorporating the digital channel. Second, instead of relying on a single indicator, such as mobile money account ownership, combining data from a variety of sources allows us to capture DFSs contribution to financial inclusion from a multidimentional perspective. Third, it distinguishes between digital and traditional financial inclusion, which allow for more granular understanding of the relative contribution of digitization versus traditional services in impacting financial inclusion in recent years. These indices were originally developed in the context of the IMF Monetary and Capital Markets Departmental Paper “The Promise of Fintech: Financial Inclusion in the Post COVID-19 Era”. This paper presents and explains the methodology that underpins them. Our new measure indicates that fintech has had a significantly positive impact on financial inclusion in payments. Digital financial inclusion increased between 2014 and 2017 across all countries, even where traditional financial inclusion was stalling or declining. Most countries saw an increase in both the access and usage dimensions. However, there are noticeable regional differences, with countries in Africa and Asia and the Pacific in the lead. While our new index should offer a useful analytical tool for researchers and policy makers, the analyses in this paper has limitations that the user should be aware of. These are primarily driven by data limitationsthe lack of granular and long time-series data on indicators related to digital financial inclusion. First, the size of the sample of countries is relatively small (52 EMDEs) and excludes advanced economies, due to data availability. Second, the databases used for the construction of the index do not differentiate between the providers of DFSs. In other words, the digital financial inclusion index would capture services provided by fintech companies as well as banks (such as mobile banking) including in partnerships with DFSs. Similarly, the databases do not provide granular information on the range of financial services a user has access to (e.g., only banks, only DFS, or both). This limits the understanding of whether fintech is broadening financial inclusion, or providing alternative means of access to those already financially included. 2 The i ndex speci fi cal l y foc us es on th e pay me nt s as pe ct of fi n anc i al i ncl usi on faci l i tate d by di g i tal me ans , an d does not cov er wi der topi cs suc h as cross-b ord e r pay me n ts, and the i mp act of cent ral ban k di g i tal currenc y.5 The remainder of the paper is organized as follows: Section II presents literature review; Section III lays out the methodology; Section IV discusses underlying data and stylized facts; Section V presents the index and findings; and Section VI concludes. II. LITERATURE REVIEW Existing measures of financial inclusion in the literature focus on financial services primarily provided by banks (Appendix I). Initial studies relied on single measures of financial inclusion by using different banking-service indicators such as: the number of branches and/or ATMs per adult population, and bank accounts per capita (e.g., Beck, Demirg-Kunt, and Martinez Peria, 2007; Honohan, 2008). But Sarma (2008) points out that the use of an individual indicator to assess the extent and impact of financial inclusion can be misleading. More recent studies have constructed more comprehensives measures of financial inclusion that combine different dimensions of financial inclusion, taking into account various aspects of access and usage by household and firms (Amidi, Massara, and Mialou, 2014; Dabla - Norris et al, 2015; Camara and Tuesta, 2017). These composite measures are constructed typically using two parametric approachesprincipal component analysis (PCA) and common factor analysis. 3 The papers generally find improved access over the last ten years. However, women, the poor, the young, and rural population are found to be disproportionall y excluded (Demirguc-Kunt, Klapper and Singer, 2013; Aslan et al., 2017). These measures, however, do not fully capture the contribution from the increasingly important role of technology in financial services. There is an increasing adoption of technology in financial services (i.e., fintech)with mobile money operators and other fintech companies newly entering the financial sectorat varying pace across geographical regions and countries. 4 At the same time, financial institutions are starting to adopt technology in delivering financial services. While the latter may be partially reflected in the traditional measures of financial inclusion, improved access and usage of financial services enabled by fintech companies are yet to be fully captured and quantified. Therefore, incorporating financial inclusion through fintech could present a more comprehensive, and potentially a very different, understanding of the progress across time and country. Some recent studies quantify the degree of digital financial inclusion by looking at relevant indicators, such as mobile money accounts and financial transactions using mobile phone (Sy et al., 2019; Loukoianova et al., 2019; Camara and Tuesta, 2017). However, these measures capture a single aspect of digital financial inclusion at a time, and do not present a comprehensive picture of combining multiple aspects including access and usage. 3 See Appendi x II on the PCA used i n thi s pap e r. 4 See Box 1 for the defi n i ti on of terms used i n thi s pap er.6 Box 1. Definition of Terms 1 Fintech: The technology-e n ab l ed i nno v at i on i n fi nan c i al servi ce s that cou l d resul t i n new busi nes s mo de l s, applications , proce sses or products with an associated mate rial effec t on the provision of financial services. Fin a n cial in clu sio n: Fin an cial in clu sio n is co mmonly defined as the “acc ess to and use of for mal financial services.” It captures a range of financial services (notably transac tions , savings, credit, and insurance ) for i ndi vi dual s an d fi rms (Sah ay et al ., 201 5b ). Digital financi al incl usion or fin tech-enabled financial inclusion: We use the two terms interchange ably in the paper. Digital acc ess to and usage of formal financial serv