《世界銀行：全球金融科技活動的模式與形成因素（英文版）（109頁）.pdf》由會員分享，可在線閱讀，更多相關《世界銀行：全球金融科技活動的模式與形成因素（英文版）（109頁）.pdf（109頁珍藏版）》請在三個皮匠報告上搜索。

1、Finance,Competitiveness,and Innovation Global Practice,World Bank Financial Institutions Group,International Finance CorporationGlobal Patterns of Fintech Activity and Enabling FactorsFintech and the Future of Finance Flagship Technical NoteGlobal Patterns of Fintech Activity and Enabling FactorsFin

2、tech and the Future of Finance Flagship Technical Note 2022 International Bank for Reconstruction and Development/The World Bank1818 H Street NW,Washington DC 20433Telephone:202-473-1000;Internet:www.worldbank.orgThis work is a product of the staff of The World Bank with external contributions.The f

3、indings,interpretations,and conclusions expressed in this work do not necessarily reflect the views of The World Bank,its Board of Executive Directors,or the governments they represent.The World Bank does not guarantee the accuracy,completeness,or currency of the data included in this work and does

4、not assume responsibility for any errors,omissions,or discrepancies in the information,or liability with respect to the use of or failure to use the information,methods,processes,or conclusions set forth.The boundaries,colors,denominations,and other information shown on any map in this work do not i

5、mply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries.Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank,all of w

6、hich are specifically reserved.Rights and PermissionsThe material in this work is subject to copyright.Because The World Bank encourages dissemination of its knowledge,this work may be reproduced,in whole or in part,for noncommercial purposes as long as full attribution to this work is given.Any que

7、ries on rights and licenses,including subsidiary rights,should be addressed to World Bank Publications,The World Bank Group,1818 H Street NW,Washington,DC 20433,USA;fax:202-522-2625;e-mail:pubrightsworldbank.org.ContentsAuthors and ContributorsviiiAcronymsixExecutive Summary 11.Introduction32.Relate

8、d Literature103.Data and Methodology:General Considerations134.Measuring Fintech Activity Across Countries154.1 Equity Investments in Fintech Companies154.2 Usage of Fintech Credit164.3 Usage of Digital Payments174.4 Finance Mobile App Downloads194.5 Aggregate Fintech Activity Index235.Enablers of F

9、intech Activity315.1 Basic Foundations315.1.1 Infrastructure315.1.2 Demographics345.2 Financial Development355.3 Policy Environment375.3.1 Legal and Regulatory Frameworks376.Main Results416.1 Fintech Activity and Infrastructure516.2 Fintech Activity and Financial Development516.3 Fintech Activity an

10、d the Enabling Policy Environment537.Summary and Conclusions55Appendix 1:Structure of Fintech Activity and Enablers Indices58Appendix 2:Data Gaps and Limitations62Appendix 3:Financial App Downloads During the COVID-19 Pandemic64Appendix 4:Stringency of Banking Regulation and Fintech Activity71Append

11、ix 5:The Cost of Remittances and the Role of Digital Financial Service Providers75List of FiguresFigure 1.Worldwide Google Searches for“Fintech”4Figure 2A.Equity Investments in Fintech Companies21Figure 2B.Usage of Fintech Credit21Figure 2C.Usage of Digital Payments22Figure 2D.Finance App Downloads2

12、2Figure 3.Aggregate Fintech Activity Index24Figure 4.Fintech Activity Index by Income Groups25Figure 5.Correlations between the Sub-Components of the Aggregate Fintech Activity Index26Figure 6.Fintech Activity Index by Developing Regions28Figure 7.Correlations between the Sub-Components of the Aggre

13、gate Fintech Activity Index29Figure 8.Fintech Activity in Africa30Figure 9.Fintech Activity Index and Income per Capita30Figure 10.Drivers of Fintech Activity43Figure 11.Fintech Activity and Financial Infrastructure51Figure 12.Fintech Activity and the Consumer Protection Framework54Figure B1.1.World

14、wide Downloads of Financial Apps during the Pandemic65Figure B1.2.Growth in Financial App.Downloads and the Incidence of the Pandemic67Figure B2.1.Fintech Activity and Stringency in Banking Regulation72Appendix 6:Appendix Tables79Bibliography94List of TablesTable 1.Country Coverage by Income Group a

15、nd Geographical Region42Table 2.Regression Analysis:Aggregate Fintech Activity Index45Table 3A.Regression Analysis:Equity Investments in Fintech Companies46Table 3B.Regression Analysis:Usage of Fintech Credit47Table 3C.Regression Analysis:Usage of Digital Payments48Table 3D.Regression Analysis:Finan

16、cial App Downloads49Table 4.Regression Analysis:Bank vs.Non-Bank Financial App Downloads50Table B1.1.Financial App Downloads during the COVID-19 Pandemic69Table B2.1.Stringency of Banking Regulation73Table B3.1.Sending$200 across Borders76Table B3.2 Average Cost of Sending$200 and Components77Append

17、ix Table 1.Aggregate Fintech Activity Index79Appendix Table 2.Country Coverage for the Regression Analysis85Appendix Table 3.Regression Analysis with Expanded Sample of Countries:Aggregate Fintech Activity Index86Appendix Table 4A.Regression Analysis:Equity Investments in Fintech Companies87Appendix

18、 Table 4B.Alternative Regression Analysis:Usage of Fintech Credit88Appendix Table 4C.Alternative Regression Analysis:Usage of Digital Payments89Appendix Table 4D.Alternative Regression Analysis:Financial App Downloads90Appendix Table 5.Alternative Regression Analysis:Bank vs.Non-Bank Financial App D

19、own91List of BoxesBox 1.Digital Remittances19Box 2.Financial App Downloads during the COVID-19 Pandemic23Appendix Table 6.Regression Analysis:Usage of Digital Payments92Appendix Table 7.Aggregate Fintech Activity Index:Decomposing Banking System Develop93This note is part of a series of technical no

20、tes developed for the“Fintech and the Future of Finance”report,a joint effort by the World Bank and the International Financial Corporation(IFC).This project was led by Erik Feyen and Harish Natarajan(both World Bank)and Matthew Saal(IFC)under the overall guidance of Jean Pesme,Mahesh Uttamchandani,

21、and Anderson Caputo Silva(all World Bank)and Paulo de Bolle and Martin Holtmann(both IFC).Alfonso Garcia Mora(IFC,formerly World Bank)provided guidance at inception and during earlier stages of the report.The note was prepared by Tatiana Didier,Erik Feyen,Ruth Llovet Montanes,and Oya Ardic.The team

22、is grateful to Rosario Cisternas Vial and Daniel Diaz for able research assistance at later stages of the project.The team would like to acknowledge inputs from peer reviewers:Ulric Eriksson von Allmen,Jon Frost,Leora Klapper,Martha Licetti,Douglas Pearce,and Sandeep Singh.We also received very help

23、ful detailed comments from Harish Natarajan,Matthew Saal,and Jean Pesme.The team thanks Machimanda A.Deviah(World Bank)for editorial assistance,Maria Lopez(World Bank)and Sensical Design for design and layout,Elizabeth Price,Melissa Knutson and Nandita Roy(all World Bank)and Henry Pulizzi and Elena

24、Gox(both IFC)for communications support,and Michael Geller and Arpita Sarkar(both World Bank)for overall coordination.Authors and ContributorsixGlobal Patterns of Fintech Activity and Enabling FactorsAcronymsACHAutomated Clearing HouseAML/CFTAnti-money laundering and combating the financing of terro

25、rismB2PBusiness to personCDDCustomer Due DiligenceDESIDigital Economy and Society Index DFSDigital Financial ServicesEAPEast Asia and the Pacific ECAEurope and Central Asia e-KYCelectronic know-your-customerEMDEEmerging Market and Developing EconomiesGPSSGlobal Payment Systems SurveysICTInformation

26、and Communication TechnologyMENAMiddle East and North AfricaMSMRsMicro,Small,and Medium Retailers P2BPerson to businessPEPrivate EquityRPWRemittance Prices WeeklyCPMICommittee for Payment Market InfrastructuresSARSouth Asia RegionSSASub-Saharan AfricaVCVenture CapitalWEFWorld Economic Forum1Global P

27、atterns of Fintech Activity and Enabling FactorsThe objectives of this paper are to take stock of the available fintech-related data,to document patterns of fintech activity across the world,and to help identify enabling factors.Fintech has seen remarkable growth over the past few years and will lik

28、ely continue to shape the financial sector in terms of products,business models,and industrial organization.Yet,measurement of fintech activity is challenging,complicated by both the lack of a widely accepted definition,as well as important data limitations.This paper tackles this measurement challe

29、nge by leveraging a wide range of data sources and developing a novel,country-level index of fintech activity for 125 countries,covering the period 2014-2018.The index covers three dimensions of fintech activity:fintech firm creation and growth through the availability of early-stage equity financin

30、g;usage of fintech credit and digital paymentsnow the most commonly used digital financial services,especially in developing countries;and the usage of mobile distribution channels for financial services.1 The fintech activity index is positively correlated with countries overall level of economic d

31、evelopment.For instance,high-income countries generally rank higher than middle-and low-income countries not only in terms of the aggregate fintech index,but also along its four constituent dimensions.However,significant variation across both regions and income groups persists,suggesting that other

32、enabling factors matter.This paper then uses the index to systematically analyze the association between fintech activity and a wide range of economic and technological factors in a multi-variate regression setting.Specifically,the paper explores the role of three broad set of enabling factors:basic

33、 foundations,including information and communications technology(ICT)and financial infrastructures;financial sector development,distinguishing between the development of the banking system and capital markets;and the enabling policy environment,capturing the legal and regulatory frameworks for digit

34、al financial services.There are three key findings in this paper.First,the estimations show that fintech activity is positively associated with ICT and financial infrastructures,though the relevance of the latter varies across types of fintech services.Specifically,the evidence indicates that ICT pa

35、yments infrastructure plays a more important role in the usage of digital payment services,whereas the development of credit information systems,a financial infrastructure,is more relevant for the usage of digital lending services.Second,the analyses also show a robust negative association between f

36、intech activity and bank development,consistent with the view that digital financial services may have more opportunities to develop in countries where the under-and un-served share of the market is relatively large.Countries with more stringent overall banking regulations exhibit subdued fintech ac

37、tivity,suggesting that this is linked to a less permissive environment for innovation and fintech entrants.At the same time,there is a higher prevalence of bank app downloads in countries with more stringent banking regulations,suggesting in these cases that the digital transformation is driven by i

38、ncumbents.Importantly,the Executive Summary1.Nascent,but rapidly evolving digital financial products and services such as central bank digital currencies,crypto-assets,stablecoins,and decentralized finance(DeFi)are beyond the scope of the current version of our index.2Global Patterns of Fintech Acti

39、vity and Enabling Factorsestimations also show that fintech activity is positively correlated with capital market development.These correlations stem from the development of digital financial services by institutions other than banks,such as fintech companies.The positive association with capital ma

40、rket development suggests that a supportive funding environment for fintech firms,especially start-up equity financing,can play an important role.For example,the mobile app data show that downloads of non-banking apps are significantly positively related to the development of capital markets but neg

41、atively associated with banking system development.The opposite patterns are observed for bank app downloads.The analysis thus supports the idea that the distinction between incumbent banks and fintech companies is particularly important when exploring the potential drivers of fintech activity.Third

42、,the empirical results are consistent with a high-quality policy environment as a necessary,but insufficient condition for fintech development.Other factors need to be in place as well for fintech activity to flourish.The degree of fintech activity is consistently on the low end of the distribution

43、in countries scoring poorly on policy indices that capture the existence of legal and regulatory frameworks relevant for digital financial services.Whereas,it varies widely across countries scoring high on these indices.In fact,there are several countries that despite having a supportive enabling po

44、licy environment exhibit relatively low levels of fintech activity.Finally,regulation could have positive and stabilizing impact on fintech activity in the longer term.These benefits are not likely to be reflected in the analysis,given the relatively short time horizon.Regarding the role of sector-s

45、pecific legislation and regulations,our results show mixed patterns.While the existence of laws and regulations for e-money,digital IDs,and e-signatureselectronic know-your-customer(e-KYC)frameworkstend to be positively associated with fintech activity,the coefficient on consumer protection tends to

46、 be negative.The results,however,are not as forceful as those related to the other set of enabling factors and may reflect the complexities of policy interactions,pre-conditions,and tradeoffs at different levels of fintech development as well as measurement challenges.Moreover,it is important to rec

47、ognize that alternative policy combinations can promote innovation and foster fintech activity,with similar outcomes.Overall,the demands on the enabling environment will likely evolve as fintech activity develops.Finding the right balance between trade-offs at every stage of fintech development rema

48、ins essential to promote activity and innovation while keeping excessive risks in check.Finally,separate in-depth analyses documented in the appendices explore two additional topics:the impact of the pandemic on finance app downloads and the link between the digitization of remittances services and

49、remittance costs.On the former,the papers analysis of mobile app download trends indicates that the pandemic may have accelerated fintech adoption.Moreover,the evidence indicates that the strict social distancing practices,including government implemented containment measures such as lockdowns,quara

50、ntines,and travel restrictions required to mitigate the spread of the coronavirus,has amplified the use of digital financial services.On the latter,the results indicate that digital service providers may help lower the costs of cross-border remittances,a key financial service for households in many

51、EMDEs.Specifically,the analysis shows that remittances costs are lower in corridors with a higher prevalence of digital service providers.3Global Patterns of Fintech Activity and Enabling FactorsFintech has seen remarkable growth over the past few years and has transformed the provision of financial

52、 services along several dimensions(for example,IMF and World Bank 2019;Fintech Market Participants Survey2).3 For users,digital technologies enabled by the near-ubiquity of cell and smart phones offer an opportunity to access faster,more affordable,more tailored,more secure,and more convenient finan

53、cial services,especially for those that are under-or unserved.4 For both incumbents and new fintech financial service providers,the costs of offering customers digital products can be considerably lower than those using physical infrastructures and relying mostly on paper-and human-based business mo

54、dels.Moreover,while traditional financial services are predominantly built on face-to-face interactions with financial service providers-many relying on cash transactions-digital financial services enable remote,contactless,and cashless transactions.Further,big tech companies have expanded into fina

55、ncial services by leveraging their large platforms and troves of alternative customer data that can be analyzed for insight through big data analytics.These developments have enabled financial services providers to serve more customers and compete with a broader set of products and at lower prices,a

56、lthough competitive outcomes are not predetermined(Market Structure note).Fintech thus has the potential to boost financial inclusion and enable productivity gains across countries at different stages of development,though the associated risks need to be carefully managed(Pazarbasioglu et al.2020;IM

57、F and World Bank 2018;Financial Stability Board 2017;McKinsey 2016).5 Fintech adoption can also benefit governments by improving the efficiency,responsiveness,and transparency of government operations,including enabling the rapid disbursement of cash transfers,reducing leakage in public expenditure,

58、and increasing tax revenues.A worldwide Google Trend search of the word“fintech”showcases the growing interest on fintech,with a remarkable increase since early 2014(figure 1).Importantly,the COVID-19 pandemic has accelerated the shift toward fintech adoption,including by traditional financial insti

59、tutions.For example,we document a marked spike in worldwide financial mobile application(app)downloads during the peak months of the COVID-19 pandemic.61.Introduction2.Global Market Survey:Digital Technology and the Future of Finance(Fintech Market Participants Survey)by Erik Feyen,Harish Natarajan,

60、Guillermo Rabadan,Robert Paul Heffernan,Matthew Saal and Arpita Sarkar.3.Various definitions of fintech exist.We adopt the broad definition of fintech from International Monetary Fund and World Bank Group(2018)to describe advances in technology that have the potential to transform the provision of f

61、inancial services spurring the development of new business models,applications,processes,and products.Examples include e-money,peer-to-peer lending,credit scoring and decisioning,robo advisory services,and distributed ledger technology.4.For instance,the use of e-money rather than cash saves conside

62、rable travel time and cost,reduces the risk of theft,and boosts convenience.Digital wallets can be used for a wide range of payment transactions such as receiving remittances,wages,and government subsidies,making purchases,or paying utility bills and school fees.Moreover,as individuals and businesse

63、s make digital payments,they create an electronic record of their receipts/sales and expenditures,which enables financial service providers to assess their credit risk.5.Despite these potential benefits,the accelerating growth of digital financial services could pose various risks to the stability a

64、nd integrity of the financial system if regulation and supervision do not keep pace(for example,Financial Stability Board 2017;IMF and World Bank 2018).For instance,the development of digital lending is raising concerns about predatory lending practices(for example,high late payment fees).Focusing o

65、n digital credit origination,Chava and Paradkar(2018)suggest that entry of fintech credit platforms can leave consumers more indebted,with increased default probabilities.Some other risks include cybersecurity risks,data privacy risks(for example,unauthorized disclosure,misuse of personal data),iden

66、tity fraud,discrimination emerging from big data analytics(for example,imperfectly calibrated algorithms),and regulatory arbitrage,among others.Embracing the benefits while mitigating the risks of fintech calls for a proportionate,technology-neutral regulatory and supervisory response that focuses o

67、n activities and risks rather than entities,where appropriate(see Regulation and Supervision of Fintech:Considerations for EMDE Policymakers(Regulation note)by Tatiana Alonso Gispert,Pierre-Laurent Chatain,Karl Driessen,Danilo Palermo and Ariadne Plaitakis with contributions from Ana M.Carjaval and

68、Matei Dohotaru).6.The pandemic appears to also have changed the considerations for some central banks on central bank digital currencies(CBDCs).See for example Auer et al.(2020a)and Boar and Wehrli(2021).4Global Patterns of Fintech Activity and Enabling FactorsFigure 1.Worldwide Google Searches for“

69、Fintech”10090807060504030201002004-072005-012005-072006-012006-072007-012007-072008-012008-072009-012009-072010-012010-072011-012011-072012-012012-072013-012013-072014-012014-072015-012015-072016-012016-072017-012017-072018-012018-072019-012019-072020-012004-01Source:World Bank staff.This figure sho

70、ws an index of worldwide google searches for the word“fintech.”The index takes the value 100 at its peak month over the period January 2004 to March 2020,indicating the maximum popularity of the word.But this large global uptake in interest in fintech masks significant cross-country heterogeneity.Wh

71、ile fintech is currently being adopted all over the world,countries are at different stages of fintech adoption.Some countries in Africa have emerged as leaders in mobile money,whereas others in Asia have expanded beyond payments services,notably China,with some fintech companies offering a wide ran

72、ge of financial services with a global footprint(Financial Stability Board 2020).But even within geographical regions,important differences exist.The increasing digitalization of financial services in recent years and its acceleration during the COVID-19 pandemic call for a better characterization a

73、nd understanding of fintech development and its enabling factors.The aim of this paper is two-fold:to measure fintech activity across countries and to explore the factors that potentially enable fintech development and may help explain the substantial variation across countries.To do so,we leverage

74、a wide range of data sources.Building on existing work,the diagram below shows a framework to illustrate the guiding principles and hypotheses underlying the analysis in this paper(see CPMI and World Bank 2016 and 2020,hereafter referred to as“PAFI”;GSMA 2019a,b).Appendix 1 provides a detailed overv

75、iew of the structure and indicators in our fintech activity and enablers indices and appendix 2 highlights several important data limitations.First,we develop a comprehensive measure of fintech activity across countries,which captures three different dimensions of fintech development,comprising four

76、 pillars that reflect both supply and demand forces:7 7.Nascent,but rapidly evolving digital financial services which are often based on distributed ledger technology such as crypto-assets,stablecoins,central bank digital currencies,Decentralized Finance(DeFi),and tokenized assets are beyond the sco

77、pe of this paper.These innovations have the potential to disintermediate or change certain roles of traditional financial institutions.They could also enhance the disruptive effects of new fintech entrants by enlarging the space to compete with intermediation activities of traditional financial inst

78、itutions.For example,lower contracting and verification costs and reduced informational asymmetries could help contest incumbents.5Global Patterns of Fintech Activity and Enabling Factors Fintech firm creation and growth.We proxy this dimension with an indicator measuring:1.Equity investments in fin

79、tech companies;Usage of digital forms of common financial services.In particular,we focus on:2.Usage of fintech credit(facilitated by electronic(online)platforms);3.Usage of digital payment services by households and firms;Mobile distribution channels.Given the rapid adoption of mobile devices,we an

80、alyze:4.Downloads of finance mobile phone applications(apps).Second,we explore several enablers of fintech activity that may help explain differences across countries,grouping them into three main dimensions:1.Basic foundations.This dimension encompasses both information and communications technolog

81、y(ICT)infrastructure as well as financial infrastructure(for example,credit information systems).This dimension also covers the level of economic development and demographics.2.Financial sector development.This dimension captures financial depth,efficiency,and inclusion and distinguishes between the

82、 banking sector and capital markets.3.Enabling policy environment.This dimension covers the existence of sector-specific regulation and supervision related to digital financial services(for example,regulations related to e-money,e-KYC,and digital consumer protection).Given the limited(or sometimes n

83、onexistent)time series coverage for many indicators,our analysis focuses on the cross-section dimension over the 2014-2018 period.In addition to using a systematic regression approach to evaluate the association between fintech activity and enabling factors,we also separately explore in the appendic

84、es several topics in more depth:the impact of the pandemic on finance app downloads(appendix 3),the role of more stringent banking regulation on fintech activity(appendix 4),and the link between the digitalization of remittances services and remittance costs(appendix 5).6Global Patterns of Fintech A

85、ctivity and Enabling FactorsThis paper contributes to the literature in several ways.First,whereas the literature often focuses on analyzing specific aspects of fintech activity(for example,payments or credit),we construct a comprehensive and multi-dimensional measure of fintech activity for a large

86、 sample of countries by drawing both from(confidential)survey data(for example,the Global Findex database and the World Banks Global Payment Systems Survey)and novel data sources(for example,equity investments in fintech companies and mobile finance app downloads).Second,we systematically analyze th

87、e association between this novel,comprehensive index of fintech activity and a wider range of enabling factors than is typical in the literature,which allows us to better discern the differential roles of the enabling environment.Third,while some studies explore the role for fintech activity of coun

88、tries general regulatory and institutional environment,we focus on legal and regulatory data that pertain more specifically to digital financial services.Finally,in the appendices we also offer novel analysis regarding the impact of the pandemic on mobile finance app downloads and the association be

89、tween remittances costs and the presence of digital service providers.Our findings can be summarized as follows.Despite substantial variation in fintech activity across geographical regions and income groups,consistent patterns emerge from cross-country analyses.Our novel fintech activity index is p

90、ositively associated with a countrys overall degree of economic and institutional development,as proxied by GDP per capita.High-income countries generally have a higher level of fintech activity compared to middle-and low-income countries,not only in terms of the aggregate fintech index,but also alo

91、ng its four constituent pillars.These differences across income groups are particularly strong for equity investments in fintech companies,which are dominated by the worlds leading financial centers(for example,Hong Kong SAR,China,Singapore,United Kingdom)and offshore financial centers(for example,C

92、ayman Islands,Liechtenstein,Malta).They are less marked in fintech credit,which is characterized by a larger presence of low and middle-income countries at the top of the distribution(e.g.China,Kenya,Rwanda,Uganda).From a regional perspective,developing countries from East Asia and the Pacific(EAP),

93、Europe and Central Asia(ECA),and Sub-Saharan Africa(SSA)generally exhibit a higher level of fintech activity compared to other developing countries,but a few individual countries in these regions(such as China and Kenya)tend to push up the regional average.A Guiding Framework for Fintech ActivityPot

94、entialenablersEnabling policy environmentLegal®ulatoryframework for DFSFinancial sector developmentBanking systemsCapital marketsBasic foundationsICT infrastructureFinancial infrastructureEconomic developmentDemographics&demand factorsFintechactivityFintech firm creation&growthEquity investmentin

95、 fintech firmsMobile distributionchannelDownloads of finance appsUsage of common financial services in digital formFintechcreditDigitalpayments7Global Patterns of Fintech Activity and Enabling FactorsAlthough per capita income and geography proxy for many aspects of a countrys stage of economic and

96、institutional development,they do not explain all the cross-country variation in fintech activity observed in the data.To better understand the patterns of fintech development across countries,we conduct multi-variate regression analyses using a fractional logit model,a framework that is suitable fo

97、r dependent variables that lie between 0 and 1 such as the fintech activity indices developed in this paper.We regress our novel fintech activity index and each of its four components(i.e.equity investments in fintech companies,usage of fintech credit,usage of digital payments,downloads of finance a

98、pps)against the three blocks of enabling factors(basic foundations,including infrastructure,financial development,and the enabling policy environment).First,our estimations show that fintech activity is positively correlated with widespread access,usage,efficiency,and affordability of ICT infrastruc

99、ture.We also find that fintech activity is greater in countries with a more developed financial infrastructure,though the strength of this relation depends on the type of fintech service provided.Specifically,we find evidence indicating that payments infrastructure plays a more important role in the

100、 usage of digital payment services,whereas the development of credit information systems is more relevant for the usage of digital lending services.Second,our results show that the distinction between incumbent banks and fintech companies is particularly important when exploring the potential driver

101、s of fintech activity.We find a robust negative association between fintech activity and banking system development.Countries with deeper,more inclusive,and more competitive banking systems tend to exhibit lower levels of fintech activity.The indicator on banking development captures aspects related

102、 to the potential demand for new fintech providers as well as supply-side constraints,such as potential barriers to entry.On the former,digital financial services may have more opportunities to develop in countries with less developed and less competitive banking systems where the under-and unserved

103、 share of the market is relatively large and the cost of financial services tends to be higher.In these circumstances,there is more space for new fintech providers to offer such services more efficiently and at relatively lower cost.On the latter,incumbent institutions may display a stronger anti-co

104、mpetitive behavior in a more concentrated and less efficient banking environment in order to maintain the status quo,thus hindering the incentives for entry of fintech companies.The negative correlation found in this paper suggests that the effects associated with the potential demand in countries l

105、ess developed banking systems,on average,dominate possible supply-side constraints associated with anti-competitive behavior by incumbents.Furthermore,whereas the estimations show a negative correlation between fintech activity and banking system development,they also show that fintech activity is p

106、ositively correlated with capital market development.These correlations stem from the development of digital financial services by institutions other than banks,such as fintech companies.For example,the mobile app data show that downloads of banking apps are strongly positively associated with bank

107、development,whereas they are negatively related to the development of capital markets.The opposite patterns are observed for non-bank app downloads.Overall,the findings suggest that well-developed(public and private)equity and bond markets,which reflect at least in part a more favorable investment c

108、limate,could foster the emergence of fintech companies by providing the funding that they need to develop.8,9 Consistent with this interpretation,many fintech companies indeed cite funding constraints as a barrier to scale-up operations(see,for example,Sahay et al.2020).10Third,our findings are cons

109、istent with a high-quality policy environment as a necessary,but not sufficient condition for fintech development.The degree of fintech activity is consistently on the low end of the distribution in countries scoring poorly on policy indices that capture the existence of legal and regulatory framewo

110、rks relevant for digital financial services,whereas it varies widely across countries scoring high on these indices.In fact,there are several 8.Capital markets,especially private equity markets,may also bring other auxiliary benefitsthrough signaling effects(by providing certification to outside inv

111、estors as well as potential customers),risk-sharing arrangements(by reducing the risk to other investors),improvements in corporate governance,among others.9.More developed capital markets may also be associated with the emergence of a specific range of fintech companies providing investment advice,

112、investment products,and portfolio management.10.Financial intermediaries,such as banks,often find it difficult to evaluate novel activities and typically do not accept as collateral the types of intangible capital that compose a large part of the capital stock of innovative firms.8Global Patterns of

113、 Fintech Activity and Enabling Factorscountries that despite having a supportive enabling policy environment exhibit relatively low levels of fintech activity.Regarding the role of sector-specific rules and regulations,our results show mixed patterns.On the one hand,our results tend to indicate that

114、 existence of legislation related to e-KYC are positively correlated with fintech activity.Digital ID and e-signature allow the identification and verification of customers without physical interactions,thus facilitating the registration process of customers and reducing onboarding and transaction c

115、osts.If supported by enabling regulations,these frameworks can also help digital financial providers comply with(know your customer)KYC requirements,which aim to prevent the criminal use of financial services.E-money provisions also tend to be positively correlated with fintech activity,though the r

116、esults are relatively weak arguably reflecting opposite effects on different market participants.E-money provisions create legal certainty,predictability,and transparency by establishing the definition of e-money,types of entities that can issue e-money,and the risk management framework required to

117、issue e-money.11 Yet,certain restrictions related to institutions and activities embedded in such regulations may also stifle innovation and market entry.On the other hand,the analysis shows a lack of consistency in the empirical link between financial consumer protection legislation and fintech act

118、ivity,perhaps reflecting diverging effects on consumers and providers.From the consumers perspective,consumer protection measures(for example,disclosure requirements,data privacy laws,and dispute resolution mechanisms)help build the necessary trust and confidence in digital financial services to sup

119、port adoption and usage.But at the same time,for financial service providers,stricter consumer protection rules could translate into larger compliance costs and greater legal responsibilities that could discourage them from offering certain products and services,especially more innovative ones.These

120、 two opposing effects,which are hard to disentangle with the available data,may be one of the reasons why the results are not statistically significant.While the analysis in this paper explores indicators capturing best practices for legal and regulatory frameworks to enable fintech activity,in the

121、absence of relevant foundations,countries have likely resorted to alternative approaches in practice.Moreover,different combinations of policy choices adopted to open the markets for competition and foster fintech activity might yield similar results.Such combinations and qualitative improvements th

122、erein may not necessarily be fully captured in quantifying these reforms via multidimensional indices explored in this paper.Lastly,the relatively short time span of the data analyzed in this paper does not allow us to disentangle the relation of the enabling policy environment and fintech activity

123、at different time horizons.The enabling policy environment could have positive and stabilizing impact on fintech activity in the long term but a negative one in the short term.Finally,separate,in-depth analyses documented in the appendices offer three key additional findings.First,appendix 3 shows t

124、hat global downloads of top financial mobile phone applications increased by 41 percent from December 2019 to the pandemics first peak in April 2020.Moreover,we find some evidence indicating that the stricter social distancing practices(including government implemented containment measures such as l

125、ockdowns,quarantines,and travel restrictions)to mitigate the spread of the coronavirus has amplified the use of digital financial services(DFS).Second,in appendix 4,using our more comprehensive fintech activity index,we confirm and expand previous research that established that countries with more s

126、tringent banking regulations exhibited weaker fintech activity,potentially because such regulations leave less arbitrage opportunities for fintech entrants.At the same time,we also find a higher prevalence of bank app downloads in countries with more stringent banking regulations,suggesting in these

127、 cases that the digital transformation is driven by incumbents.Third,in appendix 5,we document that remittances costs are lower in corridors with a higher prevalence of digital service providers,suggesting that these actors provide meaningful competitive pressure to traditional providers.We also fin

128、d evidence that ICT and financial infrastructures as well as enabling policies play an important role to lower remittances costs.11.In a recent survey,many fintech companies stated that uncertainty or frequent changes in the regulatory environment are more of a constraint than a clear road map with

129、tighter regulation(IMF,2020b).9Global Patterns of Fintech Activity and Enabling FactorsThe remainder of the paper is structured as follows.Section 2 reviews the literature on fintech development and its enabling factors.Section 3 describes the data and methodology.Section 4 documents the patterns of

130、 fintech activity across countries,focusing on the four pillars of fintech activity discussed above(i.e.equity investments in fintech companies,usage of fintech credit,usage of digital payments,and downloads of finance apps).Section 5 discusses the potential enablers of fintech activity(ICT and fina

131、ncial infrastructure,financial sector development,and legal and regulatory frameworks),while outlining the proxies used to capture such factors.Section 6 documents the results emerging from the cross-country analyses of fintech activity and its potential drivers.Section 7 concludes.10Global Patterns

132、 of Fintech Activity and Enabling FactorsA nascent,but growing body of research,has started to measure fintech development and explore its drivers,though most studies focus on either the experience of one individual country,or one type of fintech service(such as fintech credit),or a narrow set of dr

133、ivers(such as the existence of regulatory enablers or the potential for regulatory arbitrage).We discuss below how our findings relate to this existing literature.Closely related to our paper is Sahay et al.(2020),who develop a new index to measure digital financial inclusion and explore a set of po

134、tential enabling factors.A number of existing studies aim at measuring the digital economy more broadly.Most of this work relies on the construction of indices that combine a mix of enabling indicators,usage indicators,and indicators on the impact of new technologies,with many indices covering only

135、a few aspects related to digital financial services and others focusing on the degree of digitization.For example,the World Economic Forum(WEF)developed the Networked Readiness Index,which measures the propensity for countries to exploit the opportunities offered by information and communications te

136、chnology(ICT).In its latest edition,the index covers technology(including ICT infrastructure),people(how different agents use technology),governance(including aspects of inclusion),and impact(on economy and quality of life).Along the same lines,the Digital Economy and Society Index(DESI),developed b

137、y the European Commission,is a composite index that aims to track Europes digital performance and the evolution of digital competitiveness in its member states.BBVAs DiGix Index has a narrower concept of digitization,focused on demand and supply along six dimensions:infrastructure,costs,regulation,c

138、ontents,households adoption,and enterprises adoption.Another example is the Huawei Global Connectivity Index,which ranks countries according to ICT investment,ICT maturity,and digital economic performance.Some other studies narrow their focus to digital financial activity specifically,including repo

139、rts by Accenture,CB Insights,KPMG International,and McKinsey,among others.These tend to use data on private equity investments(including venture capital)into fintech companies to discuss cross-country patterns of fintech development.Others explore the data on fintech credit intermediated through onl

140、ine platforms(Cambridge Center for Alternative Finance,various reports;IOSCO 2017;BIS and FSB 2017).Ernst&Young(2017,2019)constructed a fintech adoption index that draws on an online survey of digitally active consumers in 28 markets to understand whether they were actually using fintech services on

141、 a regular basis.Sahay et al.(2020)introduce a new index to measure digital financial inclusion in payments across 52 developing countries,which captures access to ICT infrastructure and usage of digital payments services.We contribute to the existing research by leveraging a wide range of data sour

142、ces and developing a novel country-level index of fintech activity.The index is more encompassing than those in the literature,covering three dimensions of fintech activity:fintech firm creation and growth through the availability of early-stage equity financing;usage of fintech credit and digital p

143、ayments,currently the most commonly used digital financial services-especially in developing countries;and the usage of mobile distribution channels for financial services.Moreover,we analyze both a composite index and individual indices capturing specific dimensions of fintech activity for a large

144、sample of countries.2.Related Literature11Global Patterns of Fintech Activity and Enabling FactorsThere is new but quickly developing research that explores the determinants of fintech adoption.12,13 Because of data constraints,the more developed strand of this literature is concentrated on analyses

145、 of fintech credit.This research has explored the role of financial sector development.Using cross-country data on the volume of lending facilitated by online platforms,several papers find that the size of a countrys fintech credit market is negatively related to the competitiveness of its banking s

146、ystem,the depth of the banking system,and the density of commercial banks branch network(Claessens et al.2018;Frost et al.2019;Rau 2019;Sahay et al.2020).Marketplace lending to consumers is found to be negatively related to financial depth,especially in low-income countries,whereas for the business

147、segment,marketplace lending is greater in countries with lower degree of financial efficiency(Bazarbash and Beaton 2020).Moreover,Haddad and Hornuf(2019)report a negative relationship between the soundness of the banking system and startup formations of fintech companies providing lending services.T

148、hese findings are further supported with micro-evidence from specific markets and platforms.14Our findings corroborate and expand those in the literature by providing evidence that the negative correlation between fintech credit and financial sector depth and efficiency holds across a broader set of

149、 fintech activities(including digital payments)and is robust to the inclusion of a number of additional controls typically not found in other studies,such as those related to the policy environment.Moreover,our results also highlight that it is important to distinguish between incumbent banks and fi

150、ntech companies.In fact,our results show a positive correlation between banking sector development and bank financial app downloads,indicating that the negative correlation between financial system development and fintech activity does not stem from the development of DFS by incumbent institutions.F

151、urthermore,the existing research does not consider the overall structure of financial systems,leaving unexplored the role of capital markets.15 Our paper takes a step in filling this gap.Our findings suggest that these markets,especially private and public equity markets,could be an important enable

152、r for fintech activity to the extent that they support the provision of funding that fintech companies need to develop.Regarding the role of infrastructure,some work has indicated that fintech development arguably depends on the availability of a supportive ICT infrastructure(for example,widespread

153、access to mobile phones,computer,internet),though this literature is somewhat scarce.Focusing on the volume of investments into fintech companies,Haddad and Hornuf(2019)find that countries witness more fintech startups when the latest technological infrastructure is readily available,and people have

154、 more mobile phone subscriptions.Existing studies tend to capture specific aspects of ICT infrastructure,such as access to internet or mobile phone ownership,but do not consider the affordability and efficiency of such infrastructure.For example,Sahay et al.(2020)find that better access to ICT infra

155、structure,measured by the availability of the internet and mobile phones,is associated with higher usage of digital financial services.Moreover,existing work focuses on analyzing the role of ICT infrastructure as an enabler of fintech activity,leaving aside the potential enabling role of financial i

156、nfrastructure.For instance,automated clearing houses might facilitate the provision of digital payments,whereas the coverage,scope,and accessibility of credit information might promote the development of fintech credit.Bazarbash and Beaton(2020)is an exception.They explore the relation between the d

157、epth of credit information systems and fintech credit.We thus expand this strand of the literature by analyzing a comprehensive set of indicators reflecting the development of ICT infrastructures as well as by exploring the potential role of the financial infrastructures.12.See for example Frost(202

158、0)for an overview.13.A number of indices in the literature aim at measuring a specific set of drivers for the digital economy,but these tend to mix different types of indicators,not all of them capturing just enabling factors.For instance,The EIU Inclusive Internet Index focuses on the enabling envi

159、ronment for the adoption and productive use of the internet.It comprises four main blocks of indicators:availability(capturing the quality and breadth of available infrastructure required for internet access),affordability,relevance(capturing the value of being connected),and readiness(measuring the

160、 capacity among users to take advantage of being online).Citi Digital Money Readiness Index has four sub-components that capture the institutional environment,financial and ICT infrastructures,digital money solutions from government and private sectors,and the enthusiasm from consumers and businesse

161、s.Two indices constructed by GSMA are arguably an exception to the extent that they focus only on a key set of enabling factors.Specifically,the GSMA Mobile Connectivity Index measures enablers of mobile internet adoption along four dimensions:infrastructure,affordability,consumer readiness,and cont

162、ent and services.The GSMA Mobile Money Regulatory Index captures a set of regulations which can enable the development of scalable mobile money businesses,drawing from indicators related to consumer protection,KYC,infrastructure and investment environment,among others.We explore the data from the GS

163、MA indices where the underlying data were publicly available and added unique value to our analysis on fintech enablers.14.See for example De Roure et al.(2019),Tang(2019),and Jagtiani and Lemieux(2019).15.An exception is Haddad and Hornuf(2019),who argue that new fintech companies tend to emerge mo

164、re frequently in countries with well-developed capital markets,proxying the level of capital market development with GDP per capita.This analysis does not directly measure the degree of capital market development,nor does it explore the role of private equity markets.Another exception is Cornelli et

165、 al.(2020),who use aggregate indicators of capital market development in estimations of fintech and big tech credit volumes.12Global Patterns of Fintech Activity and Enabling FactorsAnother potential set of factors influencing the degree of fintech activity explored in the literature relates to the

166、legal,regulatory,and supervisory environments.Using data on credit intermediated through digital platforms worldwide,Sahay et al.(2020)find that higher protection of legal rights is positively correlated with fintech credit.Similarly,Rau(2019)finds that the rule of law,control of corruption,and qual

167、ity of general regulation in general are positively associated with the volume of fintech credit.The author argues that P2P platforms are often not explicitly covered by financial regulations.While this enables platforms to reduce compliance costs,it also leads to an unlevel playing field vis-vis re

168、gulated entities and may create new risks,including to consumers,financial integrity,and financial stability.Similarly,Braggion et al.(2018)find that P2P lending in China rose in cities that tightened loan-to-value ratios,consistent with borrowers tapping P2P credit to circumvent these regulations.I

169、n the United States residential mortgage market,Buchak et al.(2018)estimate that the higher regulatory burden on traditional banks accounts for roughly 60 percent of the recent growth of shadow banks,including online fintech lenders.Further,a few cross-country studies find that more stringent bankin

170、g regulation deters fintech credit,suggesting that fintech regulation might be more permissive in jurisdictions where banking regulation is more liberal(Claessens et al.2018;Frost et al.2019;Bazarbash and Beaton 2020).Focusing on the volume of investments in fintech companies,Cumming and Schwienbach

171、er(2018)and Navaretti et al.(2017)provide some evidence that venture capital and private equity investments into fintech companies are relatively more common in countries with weaker regulatory enforcement.Overall,the existing literature on the role of the policy environment for fintech development

172、focuses either on measures reflecting broadly the state of the legal and regulatory landscape or on specific banking sector regulations.This paper,instead,systematically analyzes the potential role of legislation,which aims to capture aspects of the legal and regulatory frameworks more closely relat

173、ed to fintech development.Our findings thus expand and complement those in the literature.Although a stricter regulatory environment for the financial sector might increase compliance costs and therefore deter financial innovation,it could also foster fintech adoption by promoting a safe,sound,and e

174、fficient financial system.For instance,stronger consumer protection rules(for example,disclosure requirements,dispute resolution mechanisms)might encourage greater usage of digital financial services by building the necessary trust and confidence in fintech providers.Another example is the existence

175、 of e-money laws,which might provide legal certainty,predictability,and transparency regarding the issuance of e-money,thus facilitating the emergence of a new,innovative,and secure payment instrument.In addition,certain regulations could help create a supportive infrastructure for the provision of

176、fintech services,such as digital ID and e-signature legislation.13Global Patterns of Fintech Activity and Enabling Factors3.Data and Methodology:General ConsiderationsA fundamental challenge in studying fintech activity is that there is no standard,widely accepted definition of“fintech”and the speci

177、fic technologies,providers,or financial services or products the term encompasses.16 Hence,measuring fintech activity is not straightforward.Moreover,supply side data are scarce as several of the new fintech activities still lie outside the regulatory and supervisory perimeter,and are therefore not

178、subject to reporting requirements.From the demand side,cross-country surveys offer important insights,but they are relatively infrequent and have only recently begun to add questions related to digital financial services.In light of these data challenges,we tackle the measurement of fintech activity

179、 from several angles,focusing either on developments within different types of financial products(for example,digital payments,digital lending,digital remittances,among others)or on developments across financial service providers(contrasting traditional financial service providers with fintech compa

180、nies).17 In addition,we investigate the role of potential enablers of digital transformation in financial services to take hold.Specifically,we analyze whether fintech activity is related to basic foundational characteristics(for example,ICT and financial infrastructures,the degree of economic devel

181、opment,demographics),the development of the financial system(for example,banking sector,capital markets),and the policy environment(for example,legal and regulatory frameworks for digital financial services).To identify patterns of fintech activity and characterize its enabling factors,we leverage a

182、 wide range of data sources,including some novel data on mobile app downloads.In the following sections,we provide a detailed description of the different dimensions of fintech activity,its potential drivers,and the main data examined in this paper.Appendix 2 describes some of the data limitations.I

183、t is worth pointing out upfront that many of the data analyzed in this paper are based on confidential surveys(of households,firms,or national/regional government authorities)with limited or no time-series coveragefor example,this is the case for some elements of the World Bank Global Findex survey

184、or the World Bank Global Payment Systems Surveys(GPSS).The timing that different surveys were undertaken is typically not alignedfor example,some surveys reflect information from 2017,others in 2018,and so on.This precludes a causal analysis of“pre-conditions”that exploits the time series dimension

185、of the data.In addition,the effects of some of the potential fintech drivers may take years to materialize.Hence,our analysis focuses on the cross-sectional variation of the data.For some variables,the sample is quite comprehensive,covering over 180 countries.18 For others,however,the coverage is mu

186、ch more limited(in some cases,less than 50 countries).Moreover,once these different data are merged,the sample of countries for which there are complete records across the different data elements is reduced significantly.We thus face a tradeoff:cover as many 16.While there are a few broad definition

187、s of fintech(FSB 2017,IMF and World Bank 2018),they do not easily translate into a statistical measure of fintech activity.17.We do not focus on the direct measurement of developments in technologies that underpin digital financial services,such as distributed ledger technology,big data analytics,an

188、d machine learning.While these may contribute to the creation and delivery of digital financial services,we measure supply and demand closer to the product level.From an enabling perspective,we assume these technologies can cross borders and could be applied across countries,provided there is no reg

189、ulatory barrier.Barriers to adopting specific technologies as well as different business models would be one of the means by which the differences in policy environment would have an impact on fintech adoption.18.In this paper,we use the term“countries”to denote jurisdictions.14Global Patterns of Fi

190、ntech Activity and Enabling Factorsdifferent dimensions or drivers of fintech activity as possible and end up with a limited sample of countries,or forgo some comprehensiveness in the analysis to improve the cross-country coverage.In balancing this tradeoff,we focused on obtaining a comprehensive sa

191、mple of developing countries.As we discuss our results in the following sections,we will highlight the robustness of our findings regarding this tradeoff.Given the sample limitations imposed by cross-country analyses,we construct indices so as to reduce the large number of individual variables to fe

192、wer,compact indicators,a better suited approach for cross-sectional analysis.This approach also allows us to deal with the confidential nature of some of the data explored,giving us room to report broad patterns.First,we construct a novel index of fintech activity across countries,which is composed

193、of four sub-indices capturing different pillars of fintech development.Second,we develop aggregate indices for each potential driver of fintech activity in order to study,and subsequently explain,the cross-country heterogeneity in fintech activity.To construct these different indices and their sub-c

194、omponents,we start by re-scaling each individual indicator,adjusting for their different measurement units and ranges of variation.To do so,we apply the minimum-maximum transformation,which projects all indicators within a range between 0 and 100 using the following equation:where Ic is the transfor

195、med value of the indicator and xc represents the actual value of the indicator in country c.To mitigate the impact of extreme outliers,we set maxc(x)at the 95th percentile of the indicators distribution.19 For countries with indicator values above the upper bound(at most 10 countries),the actual ind

196、icator value is replaced with the upper bound resulting in an index score of 100.We then construct aggregate indices for both fintech activity and its enablers as simple averages across their sub-indices.20 In the absence of well-established priors and to promote transparency,we assign equal weights

197、 to all normalized indicators within each index.When constructing these aggregate indices,we also transform each individual indicator such that they all have the same orientationthat is,a higher score always represents a“positive”score.19.For robustness,we also constructed indices based on the 99th

198、percentile of each indicators distribution.In practice,an index constructed with the higher threshold would assign the maximum value of 100 to two countries or less.In comparison to the adopted 95th percentile threshold,the higher 99th threshold indices yield relatively stable rankings,but countries

199、 tend to be bundled at the bottom of the distribution,with fewer countries at the top.Such uneven distribution affects the relative weight of different sub-indices in the aggregate indices.The results are qualitatively similar to the reported ones and are available upon request.20.While other,more c

200、omplex aggregation methods exist(for example,principal components),we believe that simple averages are most transparent and therefore more suitable for the purpose of the paper and given the various data limitations and gaps.This approach also facilitates the replication and expansion of the indices

201、 if/when additional indicators become available.Ic=xc-minc(x)maxc(x)-minc(x),15Global Patterns of Fintech Activity and Enabling Factors4.Measuring Fintech Activity Across CountriesTo document the broad patterns of fintech activity across countries,we develop a novel aggregate index of fintech activi

202、ty,measured over the 2014-2018 period.The aggregate index aims at capturing aspects of both the supply of and the demand for digital financial services.In light of the data constraints discussed above,we focus on three dimensions of fintech activity that capture four pillars which have sufficient cr

203、oss-country coverage.Specifically,the index comprises:Fintech firm creation and growth.We proxy this dimension with an indicator measuring:1.Equity investments in fintech companies.Usage of digital forms of common financial services.In particular,we focus on:2.Usage of fintech credit(facilitated by

204、electronic(online)platforms);3.Usage of digital payment services by households and firms.Mobile distribution channels.Given the rapid adoption of mobile devices,we analyze:4.Downloads of finance smart phone applications(apps).As discussed below,these sub-indices have varying scopes of coverage acros

205、s types of financial services and across financial service providers.A composite index constructed from such distinct sub-components may not reflect well the dynamics within each sub-component.To mitigate the drawbacks of aggregation,we systematically present the results of the analysis for both the

206、 aggregate index and its sub-components.4.1 Equity Investments in Fintech CompaniesThis first pillar of the fintech activity index focuses on the creation and growth of fintech companies themselves,since most of them are still quite young at this stage.Specifically,we analyze equity financing into f

207、intech companies through venture capital(VC)and private equity(PE)investments.Equity financing is particularly important for young,innovative,high-tech companies,including those in the financial sector.Existing research argues that financial intermediaries,such as banks,often find it difficult to ev

208、aluate novel activities and typically do not accept as collateral the types of intangible capital that compose a large part of the capital stock of innovative firms(Himmelberg and Petersen 16Global Patterns of Fintech Activity and Enabling Factors1994;Hall 2002;Bougheas et al.2003;Brown et al.2009;H

209、all and Lerner 2010;Czarnitzki and Hottenrott 2011).In contrast,equity finance is perceived as better suited for funding innovative,riskier firms because,unlike debt,equity contracts do not accentuate problems of financial distress for firms.In compensation for the risk assumed,equity holders direct

210、ly benefit if/when the firm succeeds.Consistent with this view,empirical research provides some evidence that more developed equity(but not credit)markets support faster growth of innovative-intensive industries(Brown et al.2013;Hsu et al.2014;Brown et al.2017;Didier et al.2020).This financing compo

211、nent of fintech activity thus essentially captures the extent to which,in a given country,there is a vibrant financial market adequately supporting the emergence of fintech companies.It thus proxies for the supply side of fintech development.Our main data sources for this pillar are PitchBook and Cr

212、unchbase,both of which contain detailed transaction-level information on equity investments into fintech companies,covering a wide spectrum of fintech-related financial services.Our goal is to capture the overall degree of fintech activity;hence,we aggregate these investment flows over the 2014-18 p

213、eriod to obtain the stock of investments into fintech firms.21 In particular,we construct the following three indicators:i.Value of VC and PE investments in fintech firms,accumulated over the period 2014-18,measured as a percentage of GDP in 2018;ii.Number of fintech companies that received PE and V

214、C equity investments during 2014-18,scaled by total population in 2018;iii.Stock of fintech companies as of December 2018,scaled by total population.After rescaling these three indicators,we aggregate them into an index capturing equity investments in fintech companies.22 This index covers 157 count

215、ries from all geographical regions and across all income groups(figure 2,panel A).The data show that equity investments in fintech companies are particularly large in financial centers,such as Hong Kong SAR,China,Singapore,the United Kingdom,and the United States.Offshore financial centers also appe

216、ar at the top of the distribution,with countries such as Cayman Islands,Liechtenstein,and Malta among the top 10.Among developing countries,China,India,and South Africa make it to the top 35.It is also worth highlighting the high ranking of Kenya,Ghana,and the Philippines;despite their relatively lo

217、w level of financial and economic development,they rank 45th,54th,and 58th respectively.4.2 Usage of Fintech CreditThis second pillar of the fintech activity index aims at measuring usage of fintech credit.However,these data are not available for a large set of countries.Many of the fintech credit p

218、roviders are not regulated and therefore not subject to regular reporting requirements.In addition,the standard reporting requirements for more established,regulated financial institutions are not well-suited to separately capture their provision of digital financial services.For instance,it is hard

219、 to identify digitally originated loans extended by banks,or the extent to which their loan underwriting and processing has migrated toward newer technologies.In light of these data limitations,to construct this pillar,we focus on cross-country data on the value of fintech credit facilitated by elec

220、tronic(online)platforms that match borrowers with lenders(investors)and are not operated by commercial banks.This pillar thus covers a single type of financial transaction(fintech credit)across 21.Because these are flow measures,focusing on a shorter,more recent time period would not necessarily cap

221、ture the early stages of fintech development in some countries.Hence,we opted for a longer time series when constructing this sub-index.Moreover,investment flows can vary substantially from one year to the next.Aggregating the data over a longer period smooths out some of these fluctuations.As robus

222、tness,we also considered the 2018-19 period.The results are qualitatively similar to the ones reported here.22.For all three indicators,we exclude offshore financial centers(OFC)for the purposes of rescaling the data.We use the list of OFC as defined by the IMF at https:/www.imf.org/external/NP/ofca

223、/OFCA.aspx.17Global Patterns of Fintech Activity and Enabling Factorsa specific set of financial service providers(electronic platforms).These data were collected through annual surveys of electronic platforms worldwide,capturing both business and consumer credit.These surveys were compiled by the C

224、ambridge Centre for Alternative Finance(CCAF),with academics and industry partners including the World Bank.23While these platforms can vary significantly in design,they all use innovations to interact with customers and process large amounts of customer information.Depending on the specific country

225、 and the underlying funding and lending approaches,these platforms are referred to as peer-to-peer lenders,loan-based crowd-funders,or marketplace lenders.The CCAF surveys also include platforms that use their own balance sheet to intermediate borrowers and lenders.A final consideration regards the

226、crowdfunding location,as several surveyed platforms in the sample have borrowers in multiple countries.For these platforms that operate in more than one country,the volume of fintech credit is allocated based on individual borrowers location,not the lenders location.Our index of usage of fintech cre

227、dit is constructed based on a single indicator:the value of total new financing intermediated through electronic platforms,accumulated over the period 2014-17(measured as a percentage of GDP).24 The index covers 173 countries from all geographical regions and across all income groups(figure 2,panel

228、B).In contrast with the index on equity investments in fintech companies,the usage of fintech credit index is characterized by a larger presence of low and middle-income countries at the top of the distribution.For example,China and Rwanda feature among the 9 countries at the very top,with an index

229、of 100.It is important to point out that fintech credit volumes remain small relative to the volume of credit provided by the commercial banking sector.For example,the top-2 countries,China and the United States,had about US$350 billion(2.9 percent of GDP)and US$43 billion(0.22 percent of GDP)in tot

230、al volume of credit through online platforms in 2017,respectively.Moreover,as this usage of fintech credit index focuses on the actual provision of digital financial services within each country,it is not surprising that the ranking of some financial centers(especially offshore financial centers)is

231、significantly lower than those in the index of investments in fintech firms.For example,Hong Kong SAR,China and Luxembourg are ranked 94th and 149th in fintech credit,respectively.4.3 Digital PaymentsThe third pillar of the fintech activity index captures the usage of digital payments.Innovations in

232、 payments systems have been at the forefront of the rise of fintech in the global financial landscape,typically emerging and developing earlier than fintech innovations in other financial services.The most pervasive“new”technology in payments is arguably the mobile phone,especially the smartphone.25

233、 Simply having a mobile phone can potentially allow access to mobile money accounts and other text-or app-based financial accounts.Having access to the internet as well further expands the possibilities.In fact,digital payments and a new generation of financial services accessed through mobile phone

234、s and the internet has driven the recent progress in financial inclusion(Dermirguc-Kunt et al.2018).In addition,the usage of digital payments for international remittances purposes can also play a role in reducing costs and increasing transaction speedssee box 1 and appendix 5.To construct this pill

235、ar of the fintech activity index,we analyze two types of data regarding the usage of digital payments:(i)household surveys and(ii)firm surveys.For the former,we use data from the 2017 Global Findex database.This database is amongst the most analyzed data in the literature on financial inclusion.The

236、data are collected 23.In cases primary data were not available(or where there were discrepancies in reported data),secondary data was used(from public information,annual reports,and press releases).Rau(2019)provides a more detailed description of the data collection process.24.For robustness,we alte

237、rnatively considered an index comprising two indicators:value of fintech credit through these online platforms to:(a)individuals and(b)business.However,these two indicators have a more limited time series coverage,they are available only for the 2014-16 period.The results were qualitatively similar

238、to the ones reported in the paper and are available upon request.25.Using digital financial services through mobile phone does not necessarily require sophisticated devices.In Sub-Saharan Africa,relatively simple,text-based mobile phones have powered the spread of mobile money accounts.Similar servi

239、ces are available in other parts of the developing world.18Global Patterns of Fintech Activity and Enabling Factors26.See World Bank(2016)for a detailed description of these data.27.The Global Findex database contains a number of other indicators reflecting access to financial services.We focused on

240、 the indicators that more accurately reflect household choices regarding their usage of digital financial services.For example,there is an indicator measuring the share of the population that used the internet to buy something online.However,buying something online does not necessarily mean paying f

241、or it online.In many developing countries,people commonly pay cash on delivery for internet orders.Although the Global Findex database has indicators capturing whether online purchases were paid digitally or by cash on delivery,the sample of countries with information for these indicators is rather

242、small.Similarly,we also did not include indicators that measure whether certain transactions were conducted through a financial institution or using“an account”,as neither of these measures specifically capture transactions through digital means.Lastly,we also did not include an indicator measuring

243、the share of the population with mobile money accounts due to data coveragethe sample of countries gets reduced significantly if we were to do so.28.This indicator measures the share of the population that in the past 12 months:used mobile money or mobile phones to make a payment from an account;or

244、used the internet to pay bills or to buy something online;or payed bills,sent or received(domestic)remittances,received payments for agricultural products,received government transfers,received wages,or received a public sector pension directly through a mobile phone.through nationally representativ

245、e surveys of more than 150,000 adults in over 140 countries,conducted every three years since 2011.We focus the analysis on indicators capturing the usage of mobile phones and the internet to conduct financial transactions,which are only available in the last round of the Findex Surveys conducted in

246、 2017.These data thus cover the usage of a single type of financial product(digital payments)across a wide range of financial service providers.Regarding the firm survey data,we use the 2015 Merchants Payments database,which provides data on the usage and acceptance of electronic payments by formal

247、micro,small,and medium retailers(MSMRs).26 These data,however,cover all forms of electronic payments,from electronic fund transfers as well as direct debit or credit transfers,to card payments and to mobile money payment transactions to merchants.They are available at the aggregate level,comprising

248、all these different forms of electronic transactions and do not allow us to focus only on mobile money transactions.Given the different coverage of the two sources of data and our focus on newer digital payment methods,we alternatively constructed an index based only on the household data.We include

249、 the following indicators in our measure of the usage of digital payments:27i.Households Share of the adult population that used the internet or mobile phone to access a financial institution account;Share of the adult population that used the internet or mobile phone to check their account balance;

250、Share of the adult population that made or received digital payments in the past year through mobile phones or the internet;28 Share of the adult population that used the internet to pay bills;Share of the adult population that used a mobile phone to pay utility bills;ii.Firms Share of retail sales

251、transactions by MSMRs that are paid electronically(P2B);Share of(immediate)supplier payments by MSMRs that are paid electronically;Share of wages by MSMRs that are paid electronically(B2P).Our index on the usage of digital payments covers 139 countries(figure 2,panel C).High-income countries dominat

252、e the top rankings of the index.The highest-ranked developing countries are upper-middle income ones:the Russian Federation and Turkey in 34th and 37th place,respectively.Kenya,Mongolia,and Ghana stand out among lower middle-and low-income countriesat 50th,53th,and 65th positions,respectively.Geogra

253、phically,countries in Eastern Europe and Central Asia(ECA)and East Asia and the Pacific(EAP)rank on average higher than countries in South Asia(SA)and Middle East and Northern Africa(MENA).19Global Patterns of Fintech Activity and Enabling FactorsBox 1.Digital RemittancesRemittances are small-value,

254、cross-border,person-to-person transfers.29 They are an essential source of income for millions of families across EMDEs,many of whom are poor.In reducing the cost of remittances,policies and regulations and the resulting market conditions can play an important role.When regulators and policy makers

255、create the right enabling environment,remittance service providers can leverage new technologies for the benefit of migrants and their families back home,for example,in the form of cost and time savings.At the same time,the use of technology can help reduce the time it takes to transfer funds.This i

256、ncludes time spent for travel and wait times.In addition,innovative services,such as the use of mobile money for international remittance transfers,can lower fees for sending remittances,which is a large component of the cost.For example,GSMA(2016,2018)notes the use of mobile technology reduces the

257、cost of remittances in half.Based on the Remittance Prices Worldwide(RPW)database,the World Bank reports that as of Q4 2020,the global average cost for digital remittances was recorded at 5.11 percent of the total amount sent,while the global average for non-digital remittances was 6.99 percent.30Th

258、e speed of an international remittance transaction is another challenge for which innovative models can provide solutions.International remittances take longer to process,compared to domestic transfers,from end to end due to a variety of factors,including differences in daily cut-off times and closi

259、ng times in different jurisdictions,time required for reconciliations,dispute resolutions and AML/CFT checks to name a few.New technologies can offer innovative ways of overcoming lengthy procedures for these purposes.According to the RPW database,speed of an international remittance transaction var

260、ies by the type of the remittance service provider(RSP):while on average the speed of a transaction is 25 hours,it is close to 5 days(69 hours)for banks and less than one day(17 hours)for non-banks(see appendix 5).Furthermore,recent experiences from COVID-19 lockdowns showed the importance of digita

261、l payment instruments for international remittances.Overall,the analysis of data on costs and service availability throughout 2020 indicates that while the availability of cash-based services have gone down,their costs have increased.31 The results of an econometric analysis of the cost of sending r

262、emittances in relation to several corridor-specific,origin-specific and destination-specific characteristics reveal a number of interesting findings and supporting evidence for the availability of services offered by digital MTOs and the cost of remittances:a larger share of services offered by digi

263、tal MTOs is associated with lower average costs.These results are presented in appendix 5.29.See CPMI and the World Bank(2007).30.See World Bank(2020a).These costs are reported as the average costs of sending$200.A digital remittance must be sent via a payment instrument in an online or self-assiste

264、d manner,and received into a transaction account,i.e.bank account,transaction account maintained at a non-bank deposit taking institution(say a post office),mobile money or e-money account.31.See World Bank(2021).4.4 Finance Mobile App DownloadsThe development of fintech has not only meant new finan

265、cial products and services,but also new distribution channels such as the ability to access and use a wide range of financial services through mobile phone apps.Many of these finance apps can also work in environments with slow data connections and low storage capabilities,typical of low-end smartph

266、ones,thus making them suitable for a wide range of customers in developed and developing countries alike.In fact,the development of such digital customer-provider interactions has challenged the traditional business model of brick-and-mortar incumbent financial institutions,such as banks,by enabling

267、 competition from other providers and diminishing the role of incumbents as gatekeepers to the consumer.20Global Patterns of Fintech Activity and Enabling Factors32.The data cover unique new downloads at the account level,tracked by each users unique identifier in each platform considered.That is,ap

268、p downloads in different devices(for example,mobile phones or tablets),re-downloads(a case in each a user install,delete,and then install an app again),and app updates do not lead to“double-counting”in the data.It is worth emphasizing that these are flow data,they do not measure the existing install

269、ed base(a stock measure).The country-level aggregation of downloads is based on the country of registration for each users account.33.See the discussion in Section 4.1 for the reasons we construct an index aggregating data over a 5-year period.The fourth pillar of our fintech activity index focuses

270、on these mobile finance apps.The data come from Sensor Tower and cover new financial app downloads by unique users from the two largest mobile app stores,Google Play and Apple App Store.We include in the analysis the full range of apps classified as financial by the app stores,thus covering apps fro

271、m a wide set of financial institutions(both banks and non-banks)as well as apps providing a wide range of digital financial services.Individual app download data are aggregated at the country level.32 This novel measure of fintech activity reflects both demand and supply aspects.It captures a demand

272、 component to the extent that individuals choose to download new apps and a supply dimension as it reflects the extent to which there are new apps to be downloaded.The app download statistics do not directly capture usage of the downloaded finance apps.Rather,the download data should be interpreted

273、as the number of new(potential)users of individual apps.Nonetheless,downloading an app is a pre-condition for usage,and the two are likely to be strongly and positively correlated.This sub-index contains two country-level indicators:33i.The total number of finance app downloads,accumulated over 2014

274、-18,measured as a share of the population in 2018;ii.The total number of finance app downloads as a share of total app downloadsboth the numerator and denominator are accumulated over the 2014-18 period.Although similar,these two indicators capture different aspects of the app download data.When fin

275、ance app downloads are scaled by population,they proxy not only for the potential usage of finance apps,but also for the overall potential usage of apps more broadly.Our second indicator scales finance app downloads by total app downloads to provide a sense of the relative prevalence and importance

276、of finance apps.The correlation between finance and all app downloads is high at 0.85.That is,countries with larger overall volumes of app downloads will likely also display high volumes of finance app downloads.However,countries with relatively low overall app download volumes could still exhibit h

277、igh values on the share of financial app downloads.In box 2 and appendix 3,we explore higher frequency data to assess the impact of the pandemic on finance app downloads.Our measure on finance app downloads covers 84 countries(figure 2,panel D).Both developed and developing countries feature within

278、the top rankings of the index.Among the top-10 countries are some East Asian countries(China,Hong Kong SAR,China,Republic of Korea,and Singapore),Nordic countries(Norway and Sweden),and some in the Americas(Brazil and United States).Although data coverage for African countries is not as comprehensiv

279、e as for the other measures,Kenya does feature high in this index as the 11th highest ranked country.Despite the heterogeneity at the top of the distribution,there is a high positive correlation between the finance app download index and countries overall income level.High-income countries have on a

280、verage higher values than middle-income countries,which in turn,have on average higher values than low-income countries.Countries income level here may be proxying,at least in part,for smartphone penetration.That is,countries with a greater share of smartphone connections tend to have more app downl

281、oads.However,the correlation between smartphone penetration and income level is high(around 0.82)and the analysis does not allow us to disentangle these effects.Although there is an uneven coverage of countries across geographic regions,some patterns emerge.Countries in East Asia and the Pacific(EAP

282、)have the highest values in this sub-index,displaying high levels of both absolute and relative financial app downloads.The few Sub-Saharan African countries in the sample tend to have on average a higher ranking in the aggregate finance app download index than many other developing countries.Althou

283、gh these countries tend to have low absolute values of financial app downloads,arguably reflecting the low penetration of smartphones,the share of financial app downloads is relatively high,thus pushing up their rank in the aggregate index.21Global Patterns of Fintech Activity and Enabling FactorsFi

284、gure 2A.Equity Investments in Fintech CompaniesSource:World Bank staff.This figure shows the index of equity investments in fintech companies across countries.Countries in grey have missing data for the index.Index1000No dataIndex1000No dataFigure 2B.Usage of Fintech CreditSource:World Bank staff.Th

285、is figure shows the index of fintech credit across countries.Countries in grey have missing data for the index.22Global Patterns of Fintech Activity and Enabling FactorsFigure 2D.Finance App DownloadsSource:World Bank staff.This figure shows the index of finance app downloads across countries.Countr

286、ies in gray have missing data for the index.Figure 2C.Usage of Digital PaymentsSource:World Bank staff.This figure shows the index of usage of digital financial services across countries.This index mostly focuses on the usage of digital payments.Countries in grey have missing data for the index.Inde

287、x100.06.31No dataIndex1000No data23Global Patterns of Fintech Activity and Enabling FactorsBox 2.Financial App Downloads during the COVID-19 PandemicThe social distancing and other containment measures adopted on a global scale to mitigate the spread of the COVID-19 pandemic have stressed the benefi

288、ts of DFS.While traditional financial services are predominantly built on cash transactions and face-to-face interactions with financial service providers,digital financial services enable remote,contactless,and cashless payments and transactions.While many have speculated that the pandemic has the

289、potential to accelerate the adoption of DFS,there is little evidence to date that this has indeed been the case.The novel app download data,reflecting the stream of new users of financial apps,can provide some early insights into DFS adoption worldwide during the COVID-19 pandemic.The analysis in ap

290、pendix 3 shows that there has been a marked spike in worldwide financial app downloads,especially of non-bank financial apps,during the peak months of the COVID-19 pandemic.The increase is particularly marked for non-bank financial apps.Global downloads of non-bank financial apps increased 45 percen

291、t,from an average of about 7 million downloads per day during the last quarter of 2019(2019Q4)to over 10 million at its peak on April 15,2020 and around the peak of policy measures taken to constrain community mobility.While these aggregate trends mask large cross-country dispersion,the analysis als

292、o shows a robust positive correlation at the country level between the growth in downloads of top-100 financial apps since the outbreak of the pandemic and the severity of the impact of COVID-19,even after controlling for GDP per capita and demographic characteristics.Moreover,the estimations indica

293、te that the increase in financial app downloads was related to the stringency of community mobility policies or practices rather than the contagion of the disease itself in a given country.4.5 Aggregate Fintech Activity IndexThe aggregate fintech activity index is calculated as a simple average of a

294、ll four pillars described above and covers 74 countries.Given the relatively limited coverage of the finance app downloads component,we also consider an aggregate fintech activity index that excludes this pillar.The country coverage for this alternative index with three pillars increases to 125 coun

295、tries(figure 3 and appendix table 1).The two aggregate indices display a correlation of 0.98 with each other,statistically significant at 1 percent,indicating that they yield similar results.34 The aggregate fintech activity index reflects well its underlying data,displaying positive and statistical

296、ly significant correlations of 0.76 or higher with each of its pillars.34.For the rest of this paper,we will focus on maximizing cross-sectional coverage,hence preferring to use the index with 3 components.But we will discuss how the results change for the 4-component index.24Global Patterns of Fint

297、ech Activity and Enabling FactorsSource:World Bank staff.This figure shows the aggregate fintech activity index across countries.Panel A shows the aggregate index comprised of four sub-components(equity investments in fintech companies,fintech credit,usage of digital financial services,and finance a

298、pp downloads).Panel B shows the aggregate index comprised of three sub-components(equity investments in fintech companies,fintech credit,and usage of digital financial services).Countries in grey have missing data.Panel B.Index with 3 componentsFigure 3.Aggregate Fintech Activity IndexPanel A.Index

299、with 4 componentsIndexNo data87446IndexNo data8744625Global Patterns of Fintech Activity and Enabling FactorsOur aggregate fintech activity index reveals some robust patterns across income groups and geographical regions.First,fintech activity is positively correlated with countries per capita incom

300、e as high-income countries tend to rank higher than developing countries(figure 4).In fact,this dichotomy between high-income and developing countries is observed not only in the aggregate index,but also along all its four pillars(figure 5).The differences are particularly acute for the component on

301、 equity investments in fintech companies and the component on the usage of digital payments.Within developing countries,middle-income countries rank,on average,similarly to low-income countries on the aggregate fintech activity index,though there are some differences across the fintech index sub-com

302、ponents:they tend to rank higher on the usage of digital payments and equity investment pillars,while lagging behind low-income countries in terms of usage of fintech credit.Figure 4.Fintech Activity Index by Income GroupsA.Aggregate fintech activity indexSource:World Bank staff.This figure shows th

303、e aggregate fintech activity index(with 3 sub-components)across income groups.Panel A shows the average of the aggregate index per income group as well as its cross-country dispersion.Panel B shows the average value of each sub-component of the aggregate fintech activity index for each income group.

304、B.Sub-components of the aggregate fintech activity index80706050403020100Investments in fintech companiesFintech creditUsage of digital financial servicesHigh incomeMiddle incomeLow incomeAggregate fintech activity index(3-comp)10.90.80.70.60.50.40.30.20.10High incomeMiddle incomeLow income26Global

305、Patterns of Fintech Activity and Enabling Factors35.These patterns are consistent with the results in Cornelli et al.(2020).They provide some evidence that big tech tend to enter markets first offering payment services,and then afterwards,they introduce other financial services.These results are als

306、o in line with survey results in the EY fintech adoption index,which shows that payment services are the most widely known and adopted type of fintech service in a sample of 27 countries(EY,2017;EY,2019).The incidence of different forms of DFS is also markedly different depending on countries level

307、of economic development.There is a positive correlation across all pillars of the fintech activity index for high-income countries,and most correlations are statistically significant at the 1 percent level(figure 5).Moreover,the data are consistent with a pecking order across different types of DFS

308、for high-income countries,where usage of digital payments would emerge first,and would then be followed by usage of fintech credit.35 At least conceptually,payment and transaction patterns(for example,cash in,cash out)can help establish creditworthiness and spur fintech credit.The top-left scatter p

309、lot in figure 5 shows that high-income countries with relatively large volumes of fintech credit also display high usage of digital payments.However,high-income countries with greater usage of digital payments display varying degrees of usage of fintech credit.In other words,the evidence points to u

310、sage of digital payments as a pre-condition for usage of fintech credit for high-income countries.Empirically an analogous pattern is not observed for developing countries.For instance,there are countries with relatively high volumes of fintech credit,but relatively low usage of digital paymentsRwan

311、da and Colombia are some examples.In fact,there is no correlation between usage of fintech credit and usage of digital payments for low-and middle-income countries.It is important to notice that we measure fintech credit as lending through electronic platforms,it does not encompass all forms of digi

312、tal lending.Figure 5.Correlations between the Sub-Components of the Aggregate Fintech Activity IndexPanel A.Scatter plots by income groupsHigh-income countriesMiddle-and low-income countriesUsage of digital financial servicesFintech creditUsage of digital financial servicesFinance app.downloadsUsage

313、 of digital financial servicesEquity investments in fintech companiesFinancial app.DownloadsEquity investments in fintech companiesFinancial app.downloadsFintech creditFintech creditEquity investments in fintech companies27Global Patterns of Fintech Activity and Enabling Factors0.29910.209Cor(equity

314、 investments,usage of fintech credit)Panel B.Correlation matrix by income groupsCor(usage of fintech credit,usage of digital payments)0.4985*0.03350.5587*0.6789*0.5342*0.27340.5108*0.4461*0.37380.3627Cor(equity investments,usage of digital payments)Cor(usage of fintech credit,fin.app downloads)Cor(f

315、in.app downloads,usage of digital payments)Cor(equity investments,fin.app downloads)High-income countries Middle-and low-income CountriesSource:World Bank staff.This figure shows in panel A the scatter plots and in panel B the cross-country pairwise correlations of the four sub-components of the agg

316、regate fintech activity index.In panel A,high-income countries are shown in blue,middle-and low-income countries are shown in red,In panel B,*denotes statistical significance at 1 percent level.Figure 5 continuedSecond,along the geographical dimension,there is wide dispersion across developing count

317、ries(figures 6 and 7).This cross-country heterogeneity is observed at both the aggregate and the four pillars of the fintech activity index.Countries in ECA,EAP,and SSA tend to have greater fintech activity than other developing countries,but a few countries in these regions tend to push up the regi

318、onal average.Most notably,four SSA countries exhibit the highest level of fintech activity.Specifically,Kenya and Rwanda show not only the highest fintech activity in SSA,they are among the top-5 developing countries(figure 8).Analogously,not all EAP countries exhibit high levels of fintech activity

319、.While China stands out from all other developing countries,Lao PDR and Myanmar appear at the bottom-5 developing countries,along with Ethiopia,Morocco,and Pakistan.While per capita income proxies for many aspects of a countrys stage of economic and institutional development as well as some of its s

320、tructural characteristics,and thus can help us explain part of the cross-country heterogeneity observed in the data,our aggregate index suggests that other country-specific factors might be at work.For instance,some countries have significantly greater fintech activity than their level of economic d

321、evelopment would predict(figure 9).These“outlier”countries comprise both developing(for example,China,Georgia,Kenya,and Rwanda)and high-income countries(for example,Estonia,Latvia,and Lithuania).In the next section,we explore these issues by focusing on possible enablers of fintech activity.28Global

322、 Patterns of Fintech Activity and Enabling FactorsFigure 6.Fintech Activity Index by Developing RegionsSource:World Bank staff.This figure shows the aggregate fintech activity index(with 3 sub-components)across geographical regions for low-and middle-income countries.Panel A shows the average in the

323、 aggregate index per region as well as its cross-country dispersion.Panel B shows the average value of each sub-component of the aggregate fintech activity index for each geographical region.B.Fintech activity index by components35302520151050Investments in fintech companiesFintech creditUsage of di

324、gital financial servicesEAPECALACMENASASSAA.Aggregate fintech activity indexEAPECASSASAMENALAC0.70.60.50.40.30.20.10Aggregate fintech activity index(3-comp)29Global Patterns of Fintech Activity and Enabling FactorsFigure 7.Correlations between the Sub-Components of the Aggregate Fintech Activity Ind

325、exSource:World Bank staff.This figure shows the cross-country pairwise correlations of the four sub-components of the aggregate fintech activity index.Only middle-and low-income countries are included in the scatter plots.By geographic region for middle-and low-income countriesEAPECALACMENASASSAUsag

326、e of digital financial servicesFintech creditUsage of digital financial servicesFinance apps.downloadsUsage of digital financial servicesEquity investments in fintech companiesFinancial app.downloadsEquity investments in fintech companiesFintech creditEquity investments in fintech companiesFinancial

327、 app.downloadsFintech credit30Global Patterns of Fintech Activity and Enabling Factors37 countriesFigure 8.Fintech Activity in AfricaFigure 9.Fintech Activity Index and Income per CapitaSource:World Bank staff.This figure shows the aggregate fintech activity index(with 3 sub-components)across countr

328、ies with available data in Africa.Source:World Bank staff.This figure plots the aggregate fintech activity index(with 3 sub-components)against countries GNI per capita.The shaded blue areas mark the different income groups(high,upper-middle,lower-middle,and low income).The figure also shows an expon

329、ential trendline fitting the data.Countries marked in yellow are those for which the aggregate fintech activity index stands above one standard deviation from the trendline.Index41223GNI per capita(in logs)1159575553515-512111097865Aggregate fintech activity index(3-comp)USAGBRESTLVALTUCHNGEOKENRWAS

330、LEUGAMWIMOZCAFAFGNPLCOGDOMECUGABLBNTTOKWTSAUCRIPRTCZESVNESPKORMLTCYPGRCPOLHRVMUSURYVENCHLITAAUTAREDEUBELNORCHELUXHKGNLDSGPSWEFINNZLISRDNKIRLJPNBHRPANTGOMLIHTISENLSOGHAMNGZAFCOLIRNMDASLVJORPRYARMINDUKRZMBTZAKHMZWEAUSCANFRASVKARGTURLBNKAZBGRMEXROUMNETURMYSRUSBWABLRTHABIHPERNAMSRBPHLMKDALBTUNLKADZAIDNG

331、TMBOLEGYLAO MARHNDVNMCIVNICCMRBGDNGAMMR PAKBENETH31Global Patterns of Fintech Activity and Enabling FactorsIn this section we explore several economic and technological factors that might impact fintech development.Why do some countries exhibit greater fintech activity than others?In which dimension

332、s are these countries different?Identifying these drivers helps us not only understand the current patterns,but also assess how to foster fintech development.In the diagram above,we presented a simple framework to illustrate the guiding principles and hypothesis underlying the analysis in the rest o

333、f this paper.Specifically,we consider three dimensions of over-arching enabling factors of fintech activity:361.Basic foundations.This dimension encompasses both ICT infrastructure as well as financial infrastructure(for example,credit information systems).This dimension also covers the level of economic development and demographics.2.Financial sector development.This dimension captures financial