Global Experiences from Regulatory Sandboxes FINANCE, COMPETITIVENESS & INNOVATION GLOBAL PRACTICE Fintech Note | No. 8 © 2020 International Bank for Reconstruction and Development The World Bank Group 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org DISCLAIMER This work is a product of the staff of The World Bank with external contributions. The findings, 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 of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. RIGHTS AND PERMISSIONS The 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 queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Photo Credits: Freepik.com Table of Contents Acknowledgements III Abbreviations and Acronyms V Executive Summary VII 1. Introduction and Background 1 1.1 The Promise of Fintech in Emerging Markets 1 1.2 Observed Regulatory Responses to Fintech 2 2. The Rise and Evolution of Sandboxes 5 3. Global Experiences and Lessons Learned 9 3.1 Objectives and Context 11 3.1.1 Market Maturity & Demand 11 3.1.2 One or Many? 12 3.1.3 Cross-Border Sandboxes 14 3.2 Design Considerations 16 3.2.1 Feasibility Assessments 16 3.2.2 Interaction Between Legal Systems and Sandboxes 19 3.2.3 Resources and Governance of Sandboxes 20 3.2.4 Testing Durations 22 3.2.5 Thematic Sandboxes Promoting Specific Technologies or Products 22 3.2.6 Consumer Safeguards 24 3.3 The Impact So Far 25 3.3.1 Assisting Policy-Maker Decisions and Effecting Regulatory Change 26 3.3.2 Benefits for Regulatory Institutions 28 3.3.3 Enhancing Financial Inclusion 28 3.3.4 Assisting Private Sector Firms 31 3.3.5 Fostering Partnerships in the Market 34 3.3.6 Stimulating Market Competition 35 3.3.7 Enabling Fintech Market Development 37 TABLE OF CONTENTS I 4. Evaluating Sandbox Impacts and Remaining Agile 41 4.1 Country-Level Outcomes 43 4.2 Regulatory Outcomes 44 43 Firm and Cohort-Level Outcomes 45 43 Operational Outcomes 46 5. Concluding Notes 49 APPENDIX 1. Methodology, Definitions, and Data Sources Employed 51 APPENDIX2. Further detail on Innovation Hubs 53 APPENDIX 3. Database of Global Sandboxes 55 Glossary 63 Endnotes 65 LIST OF COUNTRY EXAMPLES Country Example: Estonia’s Fintech Market and Fintech-Driven Initiatives 11 Country Example: The Philippines — Scaling Up Using a Test-and-Learn Approach 12 Country Example: Complementarities of Thailand’s Multiple Regulatory Sandboxes 12 Country Example: The Case of Hong Kong 14 Country Example: The Inception of the Sandbox in India 14 Country Example: Regional Sandbox Under the Pacific Islands Regional Initiative (PIRI) 15 Country Example: Rwanda’s Simulation Exercise 16 Country Example: Australia’s Sandbox — A Process of Iteration and Fine-Tuning 18 Country Example: Morocco’s Bank Al Maghrib — Strategic Choices for Fintech Support 18 Country Example: Sandbox as Part of Mexico’s Holistic Fintech Approach 20 Country Example: Colombia — Introducing Supportive Fintech Regulation via a Sandbox 20 Country Example: Inside the Bank of Thailand’s Regulatory Sandbox 21 Country Example: Leveraging Intra-Government Expertise in India 21 Country Example: Bank Negara Malaysia’s Thematic eKYC Sandbox Track 23 Country Example: Blockchain Fintech Product Testing in Lithuania 24 Country Example: QR Codes Through the Bank of Thailand’s Thematic Sandbox 24 Country Example: Using the Sandbox Approach to Adopt New Regulations in Kenya 27 Country Example: Stress Testing Underwriting Algorithms in the United States’ CFPB Sandbox 27 Country Examples: Jordan, Bahrain, and Sierra Leone 28 Country Example: Australian Licensing Exemption Scheme 32 Country Example: Encouraging Fintech Partnerships in Brazil 34 Country Example: Increasing Demand for InsureTech in Singapore 36 Country Example: Accelerating Fintech Entrepreneurship in Jordan 36 Country Example: The United Kingdom and the Digital Sandbox 37 II GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Country Example: Inter-Regulator Coordination in South Africa to Support Growth of the Fintech Ecosystem 38 Country Example: Stimulating Demand for Sierra Leone’s Sandbox 38 Country Example: Leveraging Multiple Initiatives to Drive Fintech in Indonesia 39 Country Example: Sandbox as Part of a Coordinated Strategy to Expand Fintech Ecosystem in South Korea 39 Country Example: Sandbox Regulatory Outcomes in South Korea 43 Country Example: Evolving Sandboxes to Meet Industry Needs — The Case of Hong Kong 46 LIST OF BOXES Box 1. Global Financial Innovation Network — The Global Sandbox 15 Box 2. Testing the Feasibility of a Sandbox — Simulation Exercises 17 Box 3. Dubai’s Progressive Approach to Developing Regulation Through a Sandbox 27 Box 4. Leveraging Other Innovation Tools: Germany 29 Box 5. Fintech and National Financial Inclusion Strategies (NFIS) 30 Box 6. Reactions from Fintech Firms 32 Box 7. Early Successes from the ASEAN Financial Innovation Network 34 Box 8. The Use of Landscape Assessments for Fintech Market Development 37 Box 9. Remaining Agile — Using Evaluation Results to Adjust a Sandbox Over Time 42 Box 10. Sample Quantitative Indicators to Measure the Operations and Functioning of a Sandbox 46 Box 11. Evaluation — Review of the Regulatory Sandbox Framework 47 LIST OF FIGURES Figure 1.1. Sandboxes, Innovation Hubs, and Regtech Labs Around the World (April 2020)8 3 Figure 2.1. Rise of Sandboxes by Region 6 Figure 2.2. Number of Fintech Sandboxes Created Since 2016 7 Figure 2.3. Timeline of Fintech Sandbox Creation Since 2016 7 Figure 2.4. Global Fintech-Related Regulatory Sandboxes (Announced and Operational) 8 Figure 3.1. Investment growth in Thailand 13 Figure 3.2. Number of Fintech Sandboxes by Legal Jurisdiction 19 Figure 3.3. Variation in Sandbox Resource Structures 21 Figure 3.4. Variation in Testing Periods of Sandboxes 22 Figure 3.5. Sandbox Themes by Region 23 Figure 3.6. Safeguards Commonly Used in a Sandbox 25 Figure 3.7. Outcomes from Sandbox Participation 26 Figure 3.8. Fintech-related Sandboxes with Financial Inclusion Elements 30 Figure 3.9. Number of Firms Supported by Innovation Offices and Regulatory Sandboxes 32 Figure 3.10. Benefits of Fintech According to Regulators 35 TABLE OF CONTENTS III Figure 4.1. A Suggested Measurement Framework for a Regulatory Sandbox 42 Figure 4.2. Sample Stressors, Regulatory Implications, and Their Corresponding Measurement Metrics 45 LIST OF TABLES Table 3.1. Overview of Considerations Behind Sandbox Design and Implementation 10 Table 3.2. Countries with Multiple Fintech-Related Sandboxes and the Authorities Governing Them 13 IV GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Acknowledgments This paper is a product of the Financial Inclusion, Infrastructure & Access Global Unit in the World Bank Group’s Finance, Competitiveness & Innovation Global Practice and was funded by the multidonor grant facility- Financial Sector Reform and Strengthening Initiative (FIRST) Initiative. It was prepared by Sharmista Appaya (Senior Financial Sector Specialist & Task Team Leader), Helen Luskin Gradstein (Financial Sector Specialist), and Mahjabeen N. Haji (Consultant). The paper benefited from substantial input from Ivan Daniel Mortimer who was also a peer reviewer. The team would like to thank the regulators and firms who participated in interviews with the research team and have been named within the report. The team is grateful for the feedback and support received from other peer reviewers Onur Ozlu, Rinku Chandra and Cedric Mousset. Mahesh Uttamchandani provided overall guidance. Thanks to Susan Boulanger for editorial support and FPS and Aichin Jones for design and layout assistance. The findings, interpretations, and conclusions expressed in the paper and case studies are entirely those of the authors. They do not necessarily represent the views of the World Bank Group and its affiliated organizations or those of the Executive Directors of the World Bank or the governments they represent. ACKNOWLEDGMENTS V VI GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Abbreviations and Acronyms ADGM Abu Dhabi Global Market AE advanced economy AFI Alliance for Financial Inclusion AFIN ASEAN Financial Innovation Network AI artificial intelligence AML anti-money laundering API application programming interface ASEAN Association of Southeast Asian Nations ASIC Australian Securities and Investment Commission BCB Banco Central do Brasil (Central Bank of Brazil) BCBS Basel Committee for Banking Supervision BFA Bali Fintech Agenda BIS Bank for International Settlements BNM Bank Negara Malaysia BNR National Bank of Rwanda BOT Bank of Thailand BSP Bangko Sentral ng Pilipinas CBK Central Bank of Kenya CBN Central Bank of Nigeria CDD customer due diligence CFPB Consumer Financial Protection Bureau CFT combating the financing of terrorism CGAP Consultative Group to Assist the Poor CMA Capital Markets Authority (Kenya) CNBV Comisión Nacional Bancaria y de Valores, Mexico CPMI Committee on Payments and Market Infrastructures DFS digital financial services DLT distributed ledger technology eKYC electronic (digital) know your customer EMDE emerging markets and developing economies EU European Union FATF Financial Action Task Force ABBREVIATIONS AND ACRONYMS VII FCA Financial Conduct Authority FSB Financial Stability Board GFIN Global Financial Innovation Network GFS Global Fintech Survey HKMA Hong Kong Monetary Authority ICT information and communication technology IDB Inter-American Development Bank IFC International Finance Corporation IMF International Monetary Fund IRDAI Insurance Regulatory and Development Authority of India IT Information Technology JFSA Japan Financial Services Agency KYC know your customer LAC Latin America and the Caribbean MAS Monetary Authority of Singapore MENA Middle East North Africa MNO Mobile Network Operator MSME micro-, small-, and medium-sized enterprises NAL no-action letter NBFI non-bank financial institution NFIS national financial inclusion strategy NGO nongovernmental organization OIC Office of Insurance Commission (Thailand) OJK Otoritas Jasa Keuangan (Financial Services Authority of Indonesia) P2P peer-to-peer QR Quick Response RBI Reserve Bank of India RURA Rwanda Utilities Regulatory Authority Regtech Regulatory Technology SAMA Saudi Arabia Monetary Authority SAR Special Administrative Region (Hong Kong) SARB South African Reserve Bank SEBI Securities and Exchange Board of India SFC Securities and Futures Commission of Hong Kong Suptech supervisory technology UAE United Arab Emirates UFA Universal Financial Access UK aid UK aid agency for sustainable development UNCDF United Nations Capital Development Fund UNSGSA UN Secretary-General’s Special Advocate for Inclusive Finance for Development USAID United States Agency for International Development VIII A ROADMAP TO SUPTECH SOLUTIONS FOR LOW INCOME (IDA) COUNTRIES Executive Summary The demand for digital financial services has increased significantly in recent years. Fintech1 plays a key role in meeting this demand by leveraging technology to bring digital financial services to previously underserved populations. These technological innovations have been met with policy responses that have the potential to create new opportunities for fintech firms through targeted regulatory approaches while balancing the potential risks to consumers and firms. One such approach is the “regulatory sandbox,” which provides room for experimentation while guiding regulation toward embracing emerging technologies. Regulators globally have embraced the regulatory sandbox2 as a means of providing a dynamic, evidence-based regulatory environment to test emerging technologies. Using country case studies and analysis of operations and outcomes of fintech sandboxes globally, this report highlights the benefits, challenges, and lessons learned from the implementation experiences of 73 unique fintech sandboxes in 57 countries. The intention is to provide key insights for policy makers looking to establish a new fintech sandbox or to evaluate an existing one. The report details evolving concepts and key lessons for emerging markets and developing economies (EMDE), where 70 percent of the studied fintech sandboxes were created. The emerging trends and key findings have been structured using the themes of country context; sandbox design; and impact at the level of the institution, market, and individual firms. Country Level Objectives and Context Considerations • Maturity of existing fintech ecosystem: Implementing a sandbox in a nascent fintech market may not be cost efficient for regulators, as sandboxes are resource intensive and bear large opportunity costs. In markets with few fintech firms, a sandbox may divert attention away from other fintech initiatives or policy reforms and policy makers may not realize the full benefits in terms of identifying broader policy insights. Hence, where fintech activity is at an early stage, other fintech tools and innovation mechanisms might be more effective than a sandbox. • One or many? When are multiple sandboxes in a country appropriate? Multiple sandboxes are most prevalent where there are different regulators for different areas of financial services. With proper inter-agency coordination, multiple sandboxes have shown promise in generating an enabling legal and EXECUTIVE SUMMARY IX regulatory environment and supporting fintech • Testing durations: Testing is an important development. Potential challenges related to part of the sandbox process. While the testing additional bureaucratic processes for firms can be period varies with the type and objectives of the mitigated with effective inter-agency coordination individual sandbox, it is important that the testing to align objectives and provide clear messaging to period be time-bound to keep the process agile innovators. However, there may remain differences and to prevent underdeveloped or simply unviable in legal, regulatory, and supervisory practices business models from operating indefinitely. All and mandates that are difficult to align through jurisdictions require sandboxes to define limited coordination alone. testing periods, with testing durations ranging from two weeks to two years. • The role of cross-border sandboxes: If used effectively, cross-border sandboxes can allow • Thematic sandboxes promoting specific fintech firms to benefit from streamlined licensing technologies or products: Some evidence shows and reciprocal license arrangements, reducing the that well-defined, thematic sandboxes can be regulatory burden on firms looking to scale. They effective in encouraging particular technologies could also potentially be used to help reduce risks or products to come to market. Often, the success of regulatory arbitrage so fintech firms denied of thematic sandboxes depends on the availability licensing in one country are not granted access of supporting technology or financial sector to global markets via another jurisdiction. Other infrastructure. While most fintech sandboxes are advantages include support, collaboration, and geared toward general fintech innovations, some are harmonization between regulators on issues such geared specifically toward themes such as enhancing as anti-money laundering/combating financing of blockchain technology, technology innovations that terrorism (AML/CFT) compliance and remittances. support insurance or payment systems, and remote know your customer (KYC)/digital ID technologies. Sandbox Design Considerations • Measures for protecting consumers: Consumers • Interaction with the legal system: No definitive can face added risks if consumer protection measures relationship exists between the country’s legal are not properly implemented with participating system and the efficacy of a regulatory sandbox. firms. Most sandboxes have a component in their Common law countries, civil law countries, and framework that addresses consumer protection. countries operating under a hybrid system have However, both the type of consumer protection all established sandboxes, despite their differing measures mandated in the sandbox framework supervision roles and mandates, but no one system as well as the supervision capacity available for has achieved greater benefits. In some countries, oversight vary significantly. however, the regulators may have greater latitude within their mandate to implement the sandbox as Impact: The Evidence so Far well as to adjust regulation. • Assisting policy makers’ decisions and effecting • Resource intensiveness: Sandboxes are highly regulatory change: While early evidence suggests resource intensive, and different governance models that sandbox programs can result in regulatory have been adopted for running them. The two most change, interviews with some policy makers common approaches are the “hub-and-spoke”3 suggest that change can be attributed to the open model or the dedicated unit. No one set-up is ideal, engagement between regulators and innovators. however, and it is likely that not all the expertise Sandboxes are not necessarily uniquely positioned necessary to review applications and support firms to test all innovations, but they are useful where through testing will be available from a single set of empirical evidence is needed to support policy resources. development. They can be beneficial where regulatory requirements are unclear or missing or create barriers to entry disproportionate to the risks. X GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Sandboxes can also help to build the consensus • Strengthening competition: Policy makers have among different stakeholders needed to endorse or reported mixed results when assessing if a sandbox support broader regulatory change. has led to an increase in competition in their respective markets. While a sandbox can encourage • Benefits for regulatory institutions: Sandboxes competition and lower barriers for smaller firms offer value to policy makers looking to increase their to enter the market, it can also create an unequal understanding and capacity to facilitate and regulate playing field between firms admitted to the sandbox a range of fintech innovations, particularly where and those not admitted. existing policy frameworks can be tested against new technologies and business models. Sandboxes • Enabling fintech market development: When can also help to build internal capacity on different they operate within a strategic framework that fintech innovations and provide a structured process enables fintech and alongside a set of fintech- through which to strengthen dialogue and interaction driven initiatives, regulatory sandboxes can provide with the industry. valuable insights to policy makers and enable innovation. For fintech to thrive, a multi-dimensional • Financial inclusion: While some examples show approach must be adopted, including a gap analysis how sandboxes can be linked to financial inclusion of existing laws and regulations combined with an mandates and potentially encourage innovations open dialogue between regulators and the industry. that reduce barriers to inclusion, evidence is limited overall to suggest that a sandbox with an Taken together, the overall evidence from outcomes explicit financial inclusion objective can have observed from fintech sandboxes suggests that they a greater impact than a general fintech sandbox. have several benefits for regulators as well as for The limited time that these sandboxes have been the financial sector ecosystem as a whole. They can in operation could be part of the reason. However, provide an evidence base from which to make policy when sandboxes are implemented properly and decisions; influence future supervisory methodology; used to encourage consumer-focused products and help to define, create, or amend regulation; and, services, they can potentially impact the broader in some cases, support the regulator’s competition financial system. mandate. For firms, sandboxes have been shown to offer a faster route to market and a better understanding • Assisting private sector firms: While sandboxes of the regulatory environment, but in some cases, are often open to both regulated and unregulated sandboxes prolong regulatory uncertainty. From a firms, some fintechs have attributed the ability more macro perspective, the indirect benefits include to access markets to their participation within a spillover effects into the overall fintech ecosystem, sandbox. Moreover, some evidence shows that a spurring consumer-centric products, and signaling that sandbox has reduced time to market for some firms. the market is open to innovation. • Fostering partnerships in the market: Sandboxes At the same time, implementing a sandbox can pose can help attract and develop marketplace several risks, particularly when poorly considered partnerships, or even investors, either directly and implemented. It can potentially pose unexpected through the design of a sandbox or indirectly through burdens on the regulators and promote risks such as firms that gain legitimacy from the sandbox. Specific creating unlevel playing fields in the market. Countries design features that can encourage partnerships that find themselves with fewer resourced regulators include partnership requirements between a fintech or a less pervasive fintech market may find it more and a licensed firm for eligibility to participate in the challenging to replicate a sandbox approach, and a sandbox as well as close association with industry sandbox in such jurisdictions may be less appropriate. accelerators that can provide advice and mentorship For instance, some jurisdictions have operated a from more established players. sandbox in markets with little to no material fintech activity. As a result, few fintech companies applied, EXECUTIVE SUMMARY XI and even fewer entered the sandbox. In this scenario, innovation, or increasing financial inclusion. Despite budgetary, staff, and opportunity costs borne by the successes, implementing a sandbox is not always a regulators may have outweighed the benefits offered fitting solution for unlocking financial innovation. by the sandbox. Sandboxes are, however, a new regulatory instrument and have only been in operation for four years; hence, Before embarking on creating a regulatory sandbox, results are still developing. When properly designed authorities should step back and objectively and implemented, sandboxes can be useful tools that review the environment in which they operate, provide valuable insights into fintech, but they are specifically by considering their primary objective(s): not the only mechanism that policy makers can use to increasing competition, fostering an environment for foster financial innovation. XII GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 1. Introduction and Background 1.1 The Promise of Fintech in Emerging Markets The rapidly changing financial services industry, driven by new technology, is transforming the market. Through new business models and new players — some from outside the traditional financial sector — fintech challenges existing business practices and norms and leverages technology for consumer-centric products, offering alternatives to incumbent service providers and bringing financial services to previously underserved populations. These changes have increasingly led regulators to question whether their approaches to regulation and supervision continue to be adequate for the changing environment in financial services. The increased demand for digital financial services (DFS), and the policy responses that follow, can create new opportunities for fintech firms. This, coupled with the landscape of the COVID-19 emergency, presents an added challenge for both financial institutions globally and the people and firms that rely upon them.4 While not a “one-size-fits-all” solution, the regulatory sandbox can guide regulation toward embracing emerging technologies, as well as create a dynamic, evidence-based regulatory environment from which regulators can learn. Around the world, 57 countries currently operate 73 fintech sandboxes.5 This report collates evidence available from all known regulatory sandboxes globally and presents the lessons learned and key emerging themes using illustrated case studies with the aim of providing insights for emerging markets and developing economies that are considering or have started on a sandbox journey. The report aims to depict what makes a sandbox effective and what makes it less than effective, while providing a basis for comprehensive analysis of preliminary observable outcomes from the implementation experience of fintech sandboxes at the macroeconomic, market, and institutional levels. Appendix 3 contains what is currently the most comprehensive database of fintech regulatory sandboxes in existence globally, and it has been made downloadable so that readers can conduct their own analyses based on different variables.6 As context for the details on sandboxes covered in Section 2, Section 1.2 outlines other common responses to fintech adopted in various jurisdictions. Section 3 highlights the benefits, challenges, and lessons learned, illustrated using country case studies. Section 4 discusses how policy makers can measure the impacts of a sandbox, 1. INTRODUCTION AND BACKGROUND 1 including assessing its ongoing appropriateness and its generally include testing of products or services. broader impacts within the financial sector market and See Appendix 2 for more detail on hubs. jurisdiction as a whole. • Regulatory accelerators (also referred to as regtech labs): Accelerators are more inwardly 1.2 Observed Regulatory Responses to focused and enable partnership arrangements Fintech7 between fintech firms and government authorities Regulators have responded to fintech using a range of to innovate on shared technologies, allowing responses, each with its own benefits and limitations. the regulator to learn about new technologies While some fintech activities can be covered under while the innovator benefits from testing its existing regulatory frameworks, the majority of solution on a real-world use case. It should be jurisdictions are taking or plan to take additional noted that firms that partner with an institution regulatory measures — particularly in areas where the in an accelerator process most likely do not fall regulatory framework is either unclear or nonexistent within the regulatory perimeter due to conflict of — to respond to emerging fintech services. These interest issues. responses vary substantially in scope and scale and • Regulatory sandboxes: Regulatory sandboxes include new laws, innovation offices, regulatory are typically a virtual environment that enables sandboxes, and even reskilling to respond to the live testing of new products or services in a transforming environment. controlled and time-bound manner. Controlled These fintech-related regulatory tools and approaches experimentation in a live environment provides can be classified into four broad categories: a structured approach to promoting innovation and guiding interactions with firms while • Wait and see. In this approach, regulators observe allowing regulators good oversight of emerging and monitor innovation trends at arm’s length before financial products. Regulatory sandboxes are intervening where and when necessary. open to innovative business models, products, • Test and learn. In this market-driven approach, and processes, whether regulated, unregulated, regulators create custom frameworks for individual or slated for possible future regulation. Typically, business cases, allowing the business to function firms that apply to enter a regulatory sandbox in a ring-fenced, live environment (often with have already developed an offering and wish to dispensations, such as a “no-objection” or “no- test its viability in the market. action” letter). • Regulatory laws and reform. This approach • Innovation facilitators. Regulators using this entails introducing new laws or enhancements to approach put in place a framework and mechanisms existing laws or licenses in response to innovative to promote innovation and experimentation. A firms or business models. proactive, often regulator-driven approach, this These approaches are discussed at length in our category includes: publication “How Regulators Respond to Fintech: • Innovation hubs (also referred to as innovation Evaluating Different Approaches — Sandboxes offices or labs): Innovation hubs can take various and Beyond” (2020). In that report, we emphasize forms depending on the goals and mandate of the that no one “blanket approach” exists for enabling authority. Most often, a hub serves as a central and regulating fintech, as appropriate regulation contact point to streamline queries and provide depends on the jurisdictional context, including support, advice, and guidance to either regulated legal and regulatory frameworks, the complexity of or unregulated firms, helping them navigate the fintech market, and the availability of resources. the regulatory, supervisory, policy, or legal For more guidance on identifying the approach best environment. Support can be direct or indirect, suited to specific regulatory needs, please refer to via guidance to the market, and it does not the previous report. 2 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Here, we focus only on regulatory sandboxes and support, advice, guidance, and even, in some cases, the lessons learned from examining over 70 of them physical office space to either regulated or unregulated globally. The methodology, data sources, and definitions firms to help them identify opportunities for growth employed are covered in detail in Appendix 1. While and navigate the regulatory, supervisory, policy, or the authors acknowledge that sandboxes are not a legal environment. It should be kept in mind that universally suitable approach, regulatory sandboxes sandboxes are a relatively new phenomena and have have been shown to be effective and amenable to most only been in existence since 2016. As such, evidence business needs and to provide flexibility in terms of to draw definitive conclusions on outcomes is limited, resources and architecture. They are often seen as although it continues to grow. the first step along a regulatory journey, providing Figure 1.1. Sandboxes, Innovation Hubs, and Regtech Labs Around the World (April 2020)8 Cyprus Malta Israel Bermuda Jamaica Hong Barbados Bahrain Kong Brunei Rwanda Singapore Fiji Eswatini Mauritius Sandboxes, Innovation Hubs Sandboxes, RegTechs Labs Sandboxes, Innovation Hubs, RegTechs Labs Innovation Hubs, RegTechs Labs RegTech Labs Sandboxes Innovation Hubs Source: WBG Research. 1. INTRODUCTION AND BACKGROUND 3 4 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 2. The Rise and Evolution of Sandboxes Regulatory sandboxes9 have garnered a lot of attention over the past few years, and currently WBG research indicates that over 70 fintech-related sandboxes have been officially announced globally (Appendix 3). These sandboxes have been used by regulators around the world for a number of reasons, including to help assess and adapt a jurisdiction’s regulatory framework and to signal the regulator’s (or government’s) openness to innovation. A regulatory sandbox has the potential to meet several objectives, both regulatory and institutional. While regulatory objectives are most commonly limited to financial stability, integrity, consumer protection, inclusion, and, occasionally, competition, institutional objectives may be wider in scope, such as supporting the fintech ecosystem or encouraging engagement with the private sector. Sandboxes, however, are not a panacea for all challenges confronting regulatory and policy- making bodies faced with innovations in the financial services ecosystem. The sandbox concept is well grounded. It originated in the IT industry to refer to a segregated, isolated environment for testing products or software, thus mitigating risks before products were brought to market. Developers used IT sandboxes to execute suspicious code, launch stealth attacks, or check security software for vulnerabilities without risking harm to the host device or network. Sandboxes have also been used in the health industry to identify and experiment with innovative tests and services. For instance, Health Data Research UK, the United Kingdom’s national institute for health data services, used a sandbox environment to virtually test services and innovations for predictive early detection of neurodegenerative diseases, antidepressant treatment responses, or rare disease scanning, among other medical uses.10 While banks and payment ecosystems have often experimented with new products and services, sandboxes only made their way into financial sector regulation in 2012 with the introduction of Project Catalyst, launched by the Consumer Financial Protection Bureau (CFPB) in the United States, with the sole intention of promoting consumer-friendly innovation solutions.11 The term “regulatory sandbox,” however, was popularized by the U.K.’s Financial Conduct Authority (FCA) through its Project Innovate,12 which in 2016 first promoted the sandbox idea to support and enable the environment for fintechs. While advanced economies (AE) such as the United 2. THE RISE AND EVOLUTION OF SANDBOXES 5 Kingdom and Singapore have arguably been first Objectives for these sandboxes include improving movers in this space, EMDEs have not been far behind. cross-border regulatory harmonization and fintech firms’ ability to scale more rapidly on a regional or The objectives of a sandbox vary in practice. Sandboxes global basis.14 are usually classified into four types, based on their objectives: (i) policy-focused; (ii) product or innovation Since 2016, 73 fintech-related sandboxes have focused; (iii) thematic; and (iv) cross-border. These been announced globally. Of these, 52, or about 70 categories are not mutually exclusive, however.13 percent,15 ere initiated in EMDEs; the rest were created in AEs. Some countries have created more than one • Policy-focused sandboxes: These sandboxes fintech-related sandbox (see Table 3.2), reflecting use the sandbox process to evaluate particular the jurisdictional authorities’ different priorities and regulations or policies. resources. Interestingly, the highest number of fintech- • Innovation- or product-focused sandboxes: related sandboxes have been created in the East Asia These sandboxes encourage innovation by lowering and Pacific region, closely followed by Europe and the cost of entering the regulated marketplace, Central Asia. On the lower end, North America and allowing firms to test the market viability of new South Asia reported the fewest sandboxes, although business models. India and the United States have concentrations (see Figure 2.1 below). • Thematic sandboxes: Sandboxes of this type focus on a precise theme with the objective of accelerating More than half of all relevant sandboxes, or about 56 adoption of a specific policy or innovation or percent, were created between 2018 and 2019, and supporting development of a particular subsector about a fifth were created in the first half of 2020 or even of specific products aimed at particular alone, suggesting rapid growth around the world in population segments. the use of sandboxes to test fintech innovations and regulation. The increasing density of global fintech- • Cross-border sandboxes: Cross-border or multi- related sandboxes, particularly from mid-2018 through jurisdictional sandboxes support firms’ cross-border 2020, is illustrated in Figures 2.2 and 2.3. movement and operations while encouraging regulator cooperation and reducing arbitrage. Figure 2.1. Rise of Sandboxes by Region East Asia & Pacific Europe & Central Asia Africa Middle East & North Africa Latin America & Caribbean North America South Asia 0 2 4 6 8 10 12 14 16 18 20 2016 2017 2018 2019 2020 Source: WBG Research, Appendix 3. 6 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Figure 2.2. Number of Fintech Sandboxes Although not all sandboxes are created equal, at Created Since 2016 their core sandboxes are formal regulatory programs that react to the rapidly changing backdrop of 25 digital financial services. They provide a dynamic, 22 evidence-based regulatory environment for learning from, and evolving with, emerging technologies. 20 18 As this document illustrates, no one-size-fits-all # of FinTech Sandboxes approach exists for implementing a regulatory 15 14 sandbox. Several types of sandboxes have emerged 12 over the past five years, each with unique traits and 10 attributes, and Section 3 of this report unpacks the 6 impetus behind this proliferation and offers some of the insights that have emerged from their study, 5 particularly in EMDE contexts. 0 2016 2017 2018 2019 2020 Source: WBG Research, Appendix 3. Figure 2.3. Timeline of Fintech Sandbox Creation Since 2016 Singapore (MAS) Thailand (BoT) Indonesia (OJK) Georgia (NBG) Saudi Arabia Korea, South Thailand (BoT) UAE (Abu Dhabi) (ADGM) Indonesia (BoI) Nigeria (SEC) USA (Arizona State Regulators) (SAMA, SACMA) (FSC, FSS) Turkey (MIT) China (PBOC) Thailand (SEC) Serbia (NBS) Hong Kong Mozambique (BoM, FSDMoc) Singapore (MAS) Denmark (DFSA) Thailand (BoT) Switzerland Pakistan (SECP) Thailand (OIC) Brazil (BCB) (FINMA) Hungary (MNB) India (SEBI) Rwanda (BNR) USA Sri Lanka (CBSL) Malaysia (BNM) Brunei (AMBD) Bahrain (CBB) (Kentucky) Barbados (CBB, FSC) Bulgaria (MOF) UK (FCA) Lithuania (BOL) South Africa (IFWG) Australia (ASIC) Netherlands (AFM & DNB) Nigeria (CBN & NIBSS) Kyrgyzstan ( BKR) Jun-16 Jul-16 Sep-16 Oct-16 Dec-16 Feb-17 Mar-17 May-17 Jul-17 Aug-17 Oct-17 Dec-17 Jan-18 Mar-18 May-18 Jun-18 Aug-18 Sep-18 Nov-18 Jan-19 Feb-19 Apr-19 Jun-19 Jul-19 Sep-19 Nov-19 Dec-19 Feb-20 Apr-20 Isle of Man (DIoM) UAE (Abu Dhabi) (DFSA) Japan (FSA, Malta (MGA) India (RBI) Brazil (CVM) Jamaica (BOJ) Mauritius (EDB) Hong Kong (HKMA, SFC) Govt. of Japan) Taiwan (FSC) Bermuda (BMA) Tunisia (BCT) Spain (MoEB) Canada (CSA) Sierra Leone (BoSL) Fiji (RBF) Russia (CBRC) USA (BCFP) India (IRDA) Jordan (CBJ) Kazakhstan (AFSA) Eswatini (CBE) Mexico (CNBV, USA (Nevada DoBI) MoF, Banxico) Poland (PFSA/ KNF) Kenya (KCMA) Norway (MoF) USA (Utah) Colombia (SFC) Egypt (CBE) Kuwait (CBK) Source: WBG Research, Appendix 3. 2. THE RISE AND EVOLUTION OF SANDBOXES 7 Figure 2.4. Global Fintech-Related Regulatory Sandboxes (Announced and Operational) Operational Announced Source: WBG Research, Appendix 3. Note: Some countries with multiple sandboxes, such as Brazil, Nigeria, and the United States, have both announced and operational sandboxes. These countries are represented as having operational sandboxes. 8 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 3. Global Experience and Lessons Learned Many regulatory authorities have established sandboxes in hopes of developing of a vibrant, innovative financial sector. Some were also spurred by a desire to emulate country peers or to target specific inroads into the fintech sector. Sandboxes emerge from contexts unique to each country, and their benefits can be difficult to replicate elsewhere. While sandboxes can be a useful tool when used and set up appropriately, the rush to create them can run into snags when authorities strive to achieve multiple aims. This is especially true if all of the many factors that can contribute to the success and overall impact of a sandbox have not been fully considered. Consequently, although many sandboxes have achieved some degree of success, others have encountered bottlenecks and challenges, including the failure to attract firms to participate. This section explores common queries from policy makers and identifies themes emerging from sandboxes globally. We draw evidence from country experiences across different regions and income groups and provide insights from regulator and firm interviews and surveys. The benefits, challenges, and lessons highlighted in this section are intended to inform policy makers’ efforts to establish a new sandbox or to evaluate an existing one.16 To provide these insights, this section explores sandboxes from three perspectives: (i) Objectives and Context; (ii) Design considerations; and (iii) Impact from the evidence thus far. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 9 Table 3.1. Overview of Considerations Behind Sandbox Design and Implementation Selected Case Summary Findings & Insights Studies Objectives & Context • The differing market context and the objectives of authorities • Philippines Issues have led to diverse structures and aims of sandboxes; we • Estonia observed initiatives across advanced and developing markets. • Maturity of the existing But some local fintech activity is needed for effectiveness. • Kenya fintech ecosystem • Thailand • There are few examples of multiple and cross-border sandboxes • The role & need for inter- to date; but the need to improve domestic and cross-border • Hong Kong agency coordination cooperation is recognized. • GFIN • The role of cross-border • Some authorities have concluded that other means are available sandboxes to facilitate competition and they do not always have a strong • One or many: When are mandate to promote the fintech industry. multiple sandboxes in a country appropriate? Design Considerations • Authorities benefit from clearly defining their objectives and • Thailand identifying their legal and institutional constraints. • Mexico • Inter-relationship with the • The most prevalent aim of sandboxes is to promote • India legal system development of digital financial services innovation and act as a • Resource intensiveness signaling mechanism that the regulator is open to dialogue. • Australia • Testing durations • Legal systems do not appear to be an impediment to or • Lithuania • Focusing on themes significantly complicate the set-up and operation of sandboxes. • Rwanda • Measures for protecting • Sandboxes are resource intensive but can be effective for consumers providing an evidence base for initiating legal change and can enhance an authority’s knowledge and skills. • The testing periods range from 2 weeks to 2 years based on the risk appetite of the regulator. • The detailed scope and organization of sandboxes may need to evolve over time as they respond to changing market conditions. Impact => The Evidence • Evidence is insufficient to demonstrate a direct causal relation • Malaysia So Far between a sandbox created with an intention of innovation and • South Korea one with wider financial inclusion goals. • Inter-relationship with the • Brazil • Regulatory sandboxes are one mechanism within a broader legal system • Jordan array of interventions that can help to promote an orderly • Resource intensiveness development of fintech in pursuit of greater dynamic efficiency • Testing durations and innovation. • Focusing on themes • Several initiatives have made pragmatic contributions to promoting entry of new firms and addressing specific challenges, • Measures for protecting such as around eKYC. consumers • It is still too early to draw definitive conclusions about the overall efficiency and opportunity costs of sandboxes as mechanisms for achieving regulatory change and stimulating innovation and competition. 10 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 3.1 Objectives and Context Additionally, a fintech market study by the Kenya Capital Markets Authority (KCMA) suggested that 3.1.1 Market Maturity & Demand an innovation office would be an effective way of resolving the regulatory questions of fintech start-ups For a sandbox to function effectively, it must meet looking for guidance. Currently, no single regulatory an existing market demand. In general, the local source is available to supply fintech companies with ecosystem must already have a functioning and clarification on regulations, hindering innovation and mature entrepreneurial environment, including raising the risk of regulatory uncertainty. Based on some local fintechs, although a market can also consultations with stakeholders, KCMA is considering proactively expand by inviting foreign fintechs into a one-stop-shop regulatory helpline combined with a the sandbox. Since most sandboxes only admit firms regulatory sandbox to encourage fintech innovation in with a viable and tested product, less mature markets the country.21 may have fewer fintechs (if any) that are eligible to enter the sandbox. In some cases, a sandbox has been Prior to establishing a sandbox, policy makers should set up only to find no applicants ready to test within its identify the most suitable approach by undertaking a boundaries; the regulators thus had to wind down the feasibility study and thoroughly assessing their own sandbox and pursue a different approach. fintech markets and the demand for, and appropriateness of, a sandbox as compared to other regulatory tools. For those markets where fintech activity is at an early stage, other fintech tools and innovation mechanisms might be more effective than a sandbox. These could Country Example: Estonia’s Fintech Market and include a test-and-learn approach with regulatory Fintech-Driven Initiatives forbearance provided on a case-by-case basis. Other Estonia, with a total population of 1.3 million people, innovation facilitators can also enable engagement has a burgeoning start-up scene: Tallinn, the capital, between regulators and innovators, including fintech is home to roughly 435 fintech start-ups.22 To respond committees and innovation hubs offering points of to the growing fintech sector while also managing the contact and guidance to firms. Still other methods resource and opportunity costs required to establish a include direct rule or regulatory change, proportional/ sandbox, the Estonian Financial Supervision Authority risk-based licensing regimes, or existing regulations (EFSA) opted to establish the in-house fintech Working extended to new technologies whose innovations Group in 2016. do not require a live-testing environment. The The Working Group operates much like an innovation Singaporean authorities have gone so far as to say that hub. It operates as a single point of contact and is a sandbox should be the tool of last resort and should made up of an informal group of regulators, including only be used in those cases where players don’t fully representation from the central bank, the anti-money- understand, are uncertain about, or are unable to meet laundering (AML) regulator, and the securities regulatory requirements from the onset. regulator. The members of the group provide fintechs Even when the regulator has the capacity to design with guidance on navigating the legal and regulatory and manage a regulatory sandbox and its risks, system and also help gather knowledge and build other complementary methods to advance fintech the capacity of policy makers within EFSA. Based innovations are potentially both easier and more on the group’s interaction and information exchange effective. In Kenya, for example, the CEO of BitPesa, with fintechs, it also works to develop proposals for recommends sponsored licensing18 programs as a way regulatory adjustment. In the past, consultations have for start-ups to build partnerships with incumbents in included data aggregation, payment services, and the industry and to scale beyond the caps imposed by crowdfunding requirements. sandboxes. Emerging market governments have also considered reciprocal licensing19 arrangements with other jurisdictions.20 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 11 Almost three years later, having successfully coordination will be needed, particularly given that implemented necessary reforms (including many fintech innovations cut across established fault strengthening the policy framework for research and lines of financial sector regulation. development as well as innovation policy) and gathered Multiple sandboxes are most prevalent where different the necessary market intelligence, the Working Group regulators control different areas of financial services. has unveiled plans for a sandbox to be established A country with separate and well-developed securities with the European Bank for Reconstruction and and banking sectors may have different needs to Development (EBRD).23 address. For instance, in Poland and Indonesia, authorities for banking and capital markets each opted Country Example: The Philippines — Scaling Up to launch their own sandboxes to develop an enabling Using a Test-and-Learn Approach financial ecosystem. In many countries, multiple The test-and-learn approach toward fintech sandboxes work together to enable innovation development, an alternative to the regulatory sandbox within the broader financial sector, reflecting the approach, has been used successfully in a few emerging different remits of different authorities. When economies, including the Philippines. In this approach, applied appropriately and with proper interagency regulators can use instruments such as letters-of-no- coordination, multiple sandboxes have shown objection or case-by-case waivers to allow innovators promise for holistically generating an enabling legal to operate in an environment free of specific regulation, and regulatory environment and enabling fintech. while allowing regulators to respond as results become more apparent through the testing process.24 In India, the RBI, the Securities and Exchange Board of India (SEBI), and the Insurance Regulatory and In 2004, taking an approach similar to one used in Development Authority of India (IRDAI) all launched Kenya, Bangko Sentral ng Pilipinas (BSP) allowed parallel sandboxes in 2019 with the basic goal of two large telecommunications firms to test new mobile supporting and encouraging responsible fintech. money products for consumers. At the time, it was a In Thailand, three different regulators (the Bank nascent market: there were no established regulations of Thailand (BOT), the Securities and Exchange or models for mobile money. BSP allowed the Commission (SEC), and the Office of Insurance telecommunications firms to proceed with testing new Commission (OIC)) launched regulatory sandboxes models of delivering financial services through non- focusing on innovation in different aspects of the bank entities, and BSP closely supervised the process. financial system. A similar approach was taken Five years later, in 2009, this experiment led to issuance in Hong Kong where different sandboxes cater to of “Guidelines on Use of Electronic Money.”25 different firm and institution types. The Hong Kong Monetary Authority (HKMA) sandbox allows Because sandboxes are very resource intensive and authorized banks to test new fintechs, the Securities have large opportunity costs, implementing a sandbox and Future Commission (SFC) sandbox allows both in a nascent fintech market may not be cost efficient licensed firms and start-ups to test,26 and the Insurance for regulators. In markets with few fintech firms, a Authority (IA) allows authorized insurers to test new sandbox may divert attention and resources away from fintech products. Firms that intend to conduct a pilot of other fintech initiatives or policy reforms. Even if a a cross-sectoral fintech product are requested to apply few firms enter the sandbox, policy makers may not to the most relevant sandbox. realize the full benefits in terms of identifying broader policy insights. Country Example: Complementarities of Thailand’s Multiple Regulatory Sandboxes 3.1.2 One or Many? In Thailand, three different regulators launched In some jurisdictions, different regulators run regulatory sandboxes: the Bank of Thailand (BOT), separate regulatory sandboxes, which may require the Securities and Exchange Commission (SEC), coordination. Some constructive tension between and the Office of Insurance Commission (OIC). Each sandboxes may be helpful in spurring innovation, but sandbox covers a different aspect of the financial 12 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Table 3.2. Countries with Multiple Fintech-Related Sandboxes and the Authorities Governing Them Fintech-related Country Authority Sandboxes (#) Thailand BoT, SEC, OIC 5 (3 by the BoT) USA CFPB, state gov’ts (Arizona, Kentucky, Nevada, Utah) 5 India RBI, IRDAI, SEBI 3 Brazil BCB, CVM 2 Hong Kong (SAR) HKMA & SFC, IA 2 Indonesia OJK, BI 2 Nigeria CBN & NIBSS, SEC 2 Singapore MAS 2 UAE (Abu Dhabi) ADGM, DFSA 2 Source: WBG analysis; see Appendix 3 for more detail. system: payments, remote identity verification, and significant growth in the number of venture capital insurance, respectively. firms, angel investors, and fintech accelerators (see Figure 3.1). Although direct linkage cannot be proven, The sandboxes, however, differ in approach, eligibility, the role of the BOT, OIC, SEC, and F13, combined and mandate. The BOT sandbox focuses on new, “never- with regulatory incentives (for instance, Thailand before-seen” innovations and thus far has focused on provides tax incentives to merchants who use card- quick-response (QR) codes and cross-border payments. accepting equipment) and stimulus through fintech The SEC sandbox allows fintechs to test new eKYC accelerators, cannot be ignored. (electronic know your customer) technologies, and the OIC sandbox has enabled insurers, agents, and InsurTech firms to test InsurTech innovations. The Figure 3.1. Investment growth in Thailand sandboxes also complement Thailand’s fintech hub, F13 (launched by the Thai fintech association), working together to develop a fintech ecosystem. The F13 hub 180 provides space for fintech start-ups to test and validate their services with customers. As a result of these 135 multiple initiatives, new regulations and initiatives were introduced for robo-advisory, peer-to-peer (P2P) 90 lending, eKYC, and QR payments.27 Although fintech growth in Thailand is not directly 45 attributable to the sandboxes only, since the launch of the various regulatory innovation facilitators, Thailand 0 has shown some competitive outcomes in line with 2012 2015 2017 2018 the “Thailand 4.0” national strategy to encourage Active VC & CVC in Thailand innovation. For instance, Thailand’s Global Talent Active Angel Investors in Thailand Accelerators Competitive Index for 2019 moved to a rank of 66 out of 125 countries surveyed, as compared to 70 in 2018 and 73 in 2017.28 In addition, Thailand has experienced Source: E&Y. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 13 Country Example: The Case of Hong Kong for Development and Research in Banking Technology The financial sector in Hong Kong is regulated by (IDRBT) as having the expertise to run a regulatory separate, sectoral regulators: the Hong Kong Monetary sandbox and innovation hub in collaboration with the Authority (HKMA),29 the Securities and Futures regulators. In response, the RBI, SEBI, and IRDAI all Commission, the Mandatory Provident Fund Schemes launched parallel sandboxes in 2019. While the basic Authority (MPFA), and the Insurance Authority. premise of supporting and encouraging responsible fintech is common to all three, their designs differ The Insurance Authority established its sandbox in somewhat, particularly regarding eligibility criteria December 2017, the same time that the Securities and testing environments. and Futures Commission and the HKMA announced their formation of new regulatory sandboxes or While multiple sandboxes can be useful, they may enhancements to existing sandboxes. All three also pose challenges for fintechs operating across sandboxes, at the time of the announcement, were sectoral boundaries and straddling the scope of to be linked together, providing a single point of authorities’ mandates. The parallel operation of entry for pilot trials of cross-sector fintech products. different sandboxes could introduce added layers of Coordination with the other regulators on fintech bureaucratic complexity for fintech firms that straddle issues is via their existing memorandums of more than one sector. understanding (MOUs). When a cross-sectoral issue Downsides can be minimized, and risks mitigated with has arisen, the lead authority tested the product in its effective interagency collaboration and coordination sandbox and coordinated and communicated results to align objectives and provide clear messaging to to its sister authorities. For example, the Insurance innovators. But differences in legal, regulatory, and Authority used its sandbox to test the distribution of supervisory practices and mandates may remain an insurance product via online banking channels that are difficult to align through coordination alone. by an insurance company that was part of a banking Effective cooperation may require alignment on group. Compliance with both insurance and banking new crosscutting regulations (on cloud computing or regulations (e-banking, agent banking rules, etc.) was artificial intelligence, for example) or some degree of checked before the product was allowed to graduate. regulatory arbitrage between different business models. Country Example: The Inception of the Sandbox 3.1.3 Cross-Border Sandboxes in India Cross-border sandbox coordination supports In July 2016, the Reserve Bank of India (RBI), India’s fintech firms to achieve scale, but the harmonization central banking institution, created an inter-regulatory required from different jurisdictions has proven to Working Group (WG)30 to study the scope and potential be a high threshold to cross. Regional and cross- of fintech and review the regulatory framework with border sandboxes can have benefits for firms, such which the industry must comply. The WG included as support for reciprocal licensing arrangements, but representatives from RBI, the Securities and Exchange their functioning in practice is still being tested. Board of India (SEBI), the Insurance Regulatory and Development Authority (IRDAI), and the Pension To remain sustainable, fintechs may require larger Fund Regulatory and Development Authority, as well customer bases than that of a single country. In Latin as from select financial entities regulated by these America and the Caribbean (LAC), approximately agencies, rating agencies and fintech companies. 20 percent of all fintechs operate in more than one jurisdiction, as many individual markets may be too On February 8, 2018, the government committee, small for business models to achieve scale, according representative of all the financial sector regulators and to a 2017 IDB study.31 The study led to the development select industry members, published its “Report of the of a regional sandbox in March 2019. Another example Working Group on Fintech and Digital Banking.” One is the Global Financial Innovation Network (GFIN), salient recommendation of the WG was to introduce a which was created as a platform for regulatory regulatory sandbox, and it recommended the Institute cooperation and collaboration on common challenges 14 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES or policy questions firms face across different support, collaboration, and harmonization between jurisdictions. One of the specific considerations for regulators on key inclusion issues, such as AML/CFT setting up GFIN was the potential to run a cross-border compliance and remittances. testing pilot. While the initial pilot did not see any successful fintechs, a number of invaluable lessons Country Example: Regional Sandbox Under the were learned that will be incorporated into the next Pacific Islands Regional Initiative (PIRI) phase. (See Box 1.) The Pacific Islands Regional Initiative (PIRI) launched If used effectively, cross-border sandboxes can a regional sandbox in March 2020. PIRI, supported allow fintech firms to benefit from reciprocal by AFI and UK aid, launched the Pacific Regional license arrangements, streamlined licensing, and Regulatory Sandbox Guidelines to support fintech reduced regulatory burden. For instance, the Bank development and regulation across seven central of Thailand, after its successful sandbox experiment banks, including: Banco Central de Timor-Leste, that enabled QR codes to come to market, directly Bank of Papua New Guinea, Central Bank of Samoa, partnered with other ASEAN central banks, including Central Bank of Solomon Islands, National Reserve in Cambodia, Japan, and Singapore, to develop a Bank of Tonga, Reserve Bank of Fiji, and Reserve regional interoperable cross-border payment via QR Bank of Vanuatu. code technologies.34 A cross-border sandbox covering The sandbox is designed to remove barriers to those countries may have accelerated roll-out of cross- innovation between the islands and to mitigate risks border QR payment solutions. by allowing members to act as a regional bloc rather Another potential advantage is the ability to help reduce than individual markets. This concept can potentially risks of regulatory arbitrage, so fintech firms denied help bolster expansion of interested fintechs to the licensing in one country are not granted access to global wider Pacific regional market which has a total GDP markets via another jurisdiction.35 Other potential of US$10.45 billion as of 2019.36 It is expected that, advantages of cross-border sandbox initiatives include in contrast to a national sandbox, this regional sandbox Box 1. Global Financial Innovation Network — The Global Sandbox In early 2018, the United Kingdom’s Financial Conduct Authority (FCA) proposed a global sandbox for firms to test innovative products, services, or business models across more than one jurisdiction. The Global Financial Innovation Network (GFIN) was formally launched in January 2019 by an international group of financial regulators and related organizations, including the World Bank Group, with the aim of creating a platform for shared knowledge and experiences. GFIN currently has a network of over 60 members and observers committed to supporting financial innovation. One of the considerations when setting up GFIN was the need for regulatory cooperation and collaboration on common challenges or policy questions facing firms across different jurisdictions. In response, a cross-border pilot workstream provided innovative firms with an efficient way to interact with regulators across different national jurisdictions. In January 2020, GFIN released a report on the lessons learned from the cross-border testing pilot. Prior to the launch of the pilot, market demand was unknown; however, it became clear with the volume of applications received that demand for the pilot was very strong. The GFIN members selected 8 firms from a set of 40 applications to develop a suitable testing plan.32 However, the lack of a streamlined application process for the various sandboxes proved too a high threshold for firms to cross. Based on the pilot testing phase, GFIN identified improvements for the testing experience, including launching a GFIN website as a single mechanism to better communicate with the market and to share information about cross- border testing more effectively; producing a publication with information on the types of activities that can receive support, to improve the application process for regulators and firms; and developing a single online application form with a set of common questions to collect information relevant to all jurisdictions/regulators.33 The members strongly believe that cross-border cooperation through the network will both help facilitate firm testing through the pilot project and enable the regulatory community to develop an increasingly collaborative approach going forward. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 15 will help firms tap into a larger, more diverse customer • Do we need a sandbox? What benefit does it provide base; reduce regulatory and legal bottlenecks; and us over other regulatory tools? ensure new business model sustainability and viability. • What is our specific objective? How will we While it is too early to draw any conclusions concerning measure success? the successes of the PIRI sandbox, in general, regional • Do we have the operational, technical, and financial approaches are aided by the geographic proximity capabilities to support a sandbox? of the participants, similarity in macroeconomic conditions, and presence of shared priorities — all • If a sandbox is the best tool, how (and where) should factors that can support the development of effective the sandbox be deployed? cross-border sandboxes. A feasibility assessment provides the opportunity to discuss, at the outset, metrics for measuring the 3.2 Design Considerations sandbox’s potential contribution to a particular industry or to the economy, rather than leaving this 3.2.1 Feasibility Assessments determination to after implementation. Sandbox Feasibility assessments are a critical first step that “success” depends to a large extent on results of such policy makers must undertake before setting up any initial assessments, and these findings should influence sandbox initiative and, where possible, at periodic the establishment, institutional arrangements, and intervals after. This is important for measuring the design of any sandbox. Success in regard to a sandbox legal ability and internal feasibility such as a regulators’ is subjective and depends directly on the initial resources — financial, technical, and physical — as benchmarks set for the sandbox. See Box 2 on the well as capacity to implement and operate a sandbox. World Bank Sandbox Simulation exercise and Section It is also important to check the viability based on 4 below for further details and guidance on measuring external criteria such as market demand gaps and the impacts of a sandbox. current fintech ecosystem. The United Nations Secretary-General’s Special Country Example: Rwanda’s Simulation Advocate for Inclusive Finance for Development Exercise (UNSGSA) reported that roughly a quarter of regulators The National Bank of Rwanda (BNR) set up its launched sandbox efforts without conducting an regulatory sandbox in 2018, as highlighted in Chapter initial feasibility assessment to determine whether IV of the official Gazette no. 14 of 02/04/2018, to it made sense to do so. Without a proper feasibility facilitate developing and adopting innovative financial study, sandboxes may launch prematurely or become technology, specifically within the payments space. unviable over time. In some instances, less resource- Since then, two further sandboxes have been created intensive regulatory tools may better address the needs within BNR alone, one for micro-insurance and another of regulators and industry alike. Operating a sandbox for deposit-taking institutions. In 2019, a World Bank has been overly cumbersome for some policy makers, team supported BNR with fine-tuning its regulatory resulting in low- and slow-functioning processes. In sandbox using a phase 2 Sandbox Simulation exercise some instances, a sandbox framework was put in place (see Box 2 for details). only to receive few or no applicants. These scenarios could have been avoided if proper assessments had The Sandbox Simulation exercise was attended by been conducted before and throughout the sandbox BNR key representatives from several departments, implementation process. including Bank Supervision, Policy, Payments Systems Supervision, Insurance Supervision, and Financial To conduct a feasibility assessment, government Inclusion; representatives of two other regulators, and regulatory officials should consider the concrete Rwanda Utilities Regulatory Authority (RURA) and demand for a sandbox and the unique policy goals a Kenya’s Capital Markets Authority (CMA), as well sandbox could address. They should ask themselves private sector incumbents and innovators, were also these fundamental questions at the outset:37 present. In total, 21 individuals participated. 16 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Box 2. Testing the Feasibility of a Sandbox — Simulation Exercises The World Bank Sandbox Simulation exercise38 is designed to provide financial regulators with insights into the design and implementation of regulatory sandboxes and to help “operationalize” the environment while considering country context. The simulation helps policy makers understand if a sandbox is viable and if the intended framework and its related operational plans are realistic in practice. The World Bank Group has helped many policy makers assess (i) the feasibility of implementing a sandbox as compared to other tools, (ii) the effectiveness of sandbox frameworks, and (iii) the alignment of the process to the goals and objectives of the sandbox. The simulation draws on our global expertise and entails cognitive, strategic, and practical exercises to support regulators in thinking through the structure of the sandbox and articulating its unique purpose and process in an environment that simulates the day-to-day processes of running a sandbox. Additional considerations for fine-tuning sandbox operations include identifying potential legal and operational frictions, including inter- and intra-regulator coordination; licensing and approval processes; staffing; safeguards; and testing; among others. Simulations are most often conducted with sandbox staff, members of the governance committees involved in key decisions, representatives from all relevant divisions and departments within the regulatory body (such as insurance, banking, payments, nonbank finance, etc.), and other regulators that might have their own sandboxes or whose remits might influence or affect the decisions or functioning of the sandbox; market innovators, too, are sometimes included. Outputs of Sandbox Simulation Phase 1 Conducted at Design & Pre-Launch Stage Phase 2 Conducted After Sandbox Is Operating Phase 1 Conducted at Design & Pre-Launch Stage Phase 2Conducted After Sandbox Is Operating • Identify sandbox objectives • Provide global knowledge and learnings • Tighten scope and remit • Familiarize attendees with established sandbox • Review existing best practice sandbox guidelines structuresIdentify clear eligibility and evaluation • Simulate each individual stage of sandbox process criteria (including licensing norms, etc.) with the use of carefully crafted case studies • Define testing and exit procedures particular to the country context: • Create sandbox prototype • Application review • Provide an understanding of relevant internal • Selection governance arrangements • Testing • Support creation of a sandbox manual • Reporting and risk mitigation • Shortlist communication options to publicize • Supervisory oversight sandbox • Post-test options (e.g., exit and licensing) • Identify areas of improvement in the design and process of sandbox structure • Distinguish areas of improvement in the governance procedures followed • Consider MOUs with other regulators or external bodies Before the detailed simulation exercise was conducted, application and evaluation through to test design and the WB team presented for four days on global sandbox implementation, culminating in an exit strategy. The models and approaches to sandbox governance and simulation forced participants to review and design processes. Using five case studies created uniquely for tests and supervision strategies for the realistic case the Rwanda context, the simulation took participants studies, with the aim of integrating lessons learned into through all aspects of the sandbox lifecycle, from the policy framework. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 17 The results and learnings helped the BNR refine listed international securities, and crowd-sourced the goals and objectives of the sandbox, define and funding activities. communicate clear and tangible objectives and • Imposed additional safeguards, such as instructions for innovators looking to apply to the disclosures, information about a provider’s sandbox, and highlight specific areas where the remuneration, associations and relationships with application form could be updated. The exercise also issuers of products, and the dispute resolution supported the BNR in identifying which waivers mechanisms available. they could consider without harming the regulator’s objectives and while the honing the governance The reform allowed ASIC to control how exemptions structure in place around the sandbox, including are granted and withdrawn and required fintech firms the cooperation mechanisms between different to notify clients that they are using the exemption. regulatory bodies. Moreover, certain baseline obligations continue to be applied during the course of the process, such as the Country Example: Australia’s Sandbox — A responsible lending obligations and obligations on Process of Iteration and Fine-Tuning handling client money and on preparing statements of In Australia, the Australian Securities and Investments advice where personal advice is provided. Breaching Commission (ASIC) revealed its first iteration of these obligations may lead ASIC to cancel a firm’s a sandbox in December 2016. Any eligible fintech exemption.39 company needed only to notify ASIC of its intention This example illustrates the need to fine-tune a sandbox to offer products and services within the sandbox rules. to respond to market demands and the regulator’s No further approvals from ASIC or other regulators changing needs. were required. The relatively restrictive parameters of the sandbox, Country Example: Morocco’s Bank Al Maghrib however, resulted in limited participation, with only — Strategic Choices for Fintech Support one start-up utilizing the sandbox in seven months. Faced with the digitalization of financial services ASIC therefore took further measures to improve the and the advent of new electronic payment and sandbox, and the government thereafter issued new fintech entrants, Bank Al Maghrib (BAM) began, draft legislation and regulations to create an enhanced as early as 2017, to think about how best to address regulatory sandbox. these developments, given its legal and regulatory framework. The new sandbox provides a “lighter touch” regulatory environment to allow fintech’s additional flexibility At first, BAM, with WB assistance, studied the while still testing their ideas. Safeguards remained sandbox option. However, after some deliberation, the same in the new legislation, but the following key this regulatory approach was abandoned for four main changes were made: reasons:40 • Extended exemption period from 12 months to 24 • BAM considered that its position, in principle, months. was to regulate the institutions under its control • Enabled ASIC to grant conditional exemptions fairly, not to regulate activities or institutions in an to financial regulations for the purpose of testing individualized manner. financial and credit services and products. • International experiences gathered were mixed and • Empowered ASIC to make decisions regarding how inconclusive, with few operational fintechs at the the exemption starts and ceases to apply. end of cohorts. • Broadened the categories of products and services • The main applications received from fintechs were that may be tested in the sandbox to include life within the current legal and regulatory framework. insurance products, superannuation products, Other applications required review of the banking law to carry out their activities. 18 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES • As the legislative processes are long and uncertain, mandates, as depicted by the CGAP-WBG innovation even those firms that successfully exit a sandbox facilitator survey as well as in desk-based research. might have been unable function unhindered in Figure 3.2 shows that civil law systems are the most the market, creating an unnecessary burden for the common type of legal system in which sandboxes fintech player. operate, but common law and hybrid legal systems follow closely, highlighting that no one system is more In light of these concerns, BAM looked to other suited than the others. approaches, such as accelerators or incubators, and to gradual implementation of more agile and Figure 3.2. Number of Fintech Sandboxes responsive regulation that could provide a framework by Legal Jurisdiction commensurate with the risks to consumers, financial integrity, and operational resilience. 40 To do this, BAM has decided, as part of its work 35 during 2019–2023 to formalize its digital strategy and 29 30 to develop and promote an environment conducive to fintech development along two axes: # of Sandboxes 25 23 20 • Creation of the one-stop-shop for fintech within 20 the SMP Supervisory Directorate, with a focus on 15 communication with the market, monitoring, and accreditation of fintechs. 10 • Integrating innovation into the central bank’s core 5 1 activities by creating an Innovation and Digital Lab 0 to foster innovation that supports the functioning Civil Common Hybrid Religious and operations of the Bank. Law Law System Law Legal System in Sandbox Jurisdiction 3.2.2 Interaction Between Legal Systems and Sandboxes Source: WBG research, Appendix 3. No definitive relationship exists between the legal The majority of sandbox initiatives in EMDEs did not system and the efficacy of a regulatory sandbox. require any amendments to laws or regulatory powers. Sandboxes have been implemented across a wide For instance, in May 2016 the FCA was the first variety of legal systems, and no one system has shown regulator to launch a regulatory sandbox initiative, more benefits than the others. However, in some allowing businesses to test out new, innovative financial countries, the regulators may have greater latitude services without incurring all the normal regulatory within their mandate to implement a sandbox, and consequences of engaging in those activities. This was the degree of autonomy regulators or supervisors are created under the existing powers available to the FCA given to make adjustments to regulations and their under the Financial Services and Markets Act under interpretations also varies. which it was created. Further, the sandbox was put in The Global Fintech Survey (GFS) conducted in 2019 place to directly support the secondary objective of by the IMF and the WBG, highlights that all regulators the regulator: to increase competition. This is similar report being keen to support firms; a sandbox is to India and South Africa, where regulators had the quicker to set up in some jurisdictions than others, power to set up a sandbox without needing an explicit however, and this depends largely on the powers law to provide approval. afforded the regulator. In other jurisdictions, however, laws and designated Common law countries, civil law countries, and those regulatory powers have required adjustment to initiate operating under a hybrid system have all established a sandbox and to designate an institutional mandate sandboxes, despite their differing supervisory roles and over fintech activities. Prior to establishing a sandbox, 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 19 for instance, the AFSA (Astana Financial Services or La Arenera; and a regtech mechanism to leverage Authority) in Kazakhstan, a civil law jurisdiction, innovations to help the regulator’s internal processes. required an amendment within its framework to include As part of the sandbox, the regulator issues a fintech objectives and give AFSA the power to waive temporary, two-year fintech license for sandbox or modify requirements in financial regulations.41 A graduates before deciding whether to permanently similar case occurred in Mexico, where a fintech law adopt any linked regulatory changes. If adopted, the mandated a regulatory sandbox initiative.42 changes are communicated through external circulars, which help to familiarize the financial sector with Country Example: Sandbox as Part of Mexico’s updated practices.45 Using this mechanism, Colombia Holistic Fintech Approach43 has thus far issued new regulations for cybersecurity, In Mexico — a civil law jurisdiction — a regulatory cloud computing, payments schemes, and QR codes sandbox is one of many complementary initiatives and is in the process of issuing fintech licensing and under the nation’s fintech law that work together to new anti-money laundering rules.46 build an enabling fintech ecosystem. Mexico enacted its renowned fintech law in March 2018 to encourage 3.2.3 Resources and Governance of Sandboxes innovation and extend regulatory perimeters to cover existing fintechs operating in the market. The law Sandboxes are highly resource intensive, and policy granted the various regulators the authority to supervise makers should be aware of this prior to initiating one. fintechs, set up a legal and regulatory framework for The WBG-CGAP innovation facilitators survey found fintech institutions, establish a fintech supervision that policy makers found the intensity of resources department within the Comisión Nacional Bancaria needed to implement a sandbox to be a major weakness y de Valores (CNBV) to oversee crowdfunding and of the approach. Moreover, many policy makers e-money, and launch regulatory sandbox(s). The significantly underestimated the resources required sandbox, supervised by a sandbox team, works to develop and operate a sandbox, with the estimated with both regulated and nonregulated entities to test cost varying considerably, from $25,000 to $1 million innovations. ($25,000 to $100,000 in EMDEs). In addition, the law launched an open banking Approaches to running a sandbox can differ substantially initiative and allowed transactions to be made using between countries. The two most common governance certain cryptocurrencies, among other initiatives. With models are (i) the dedicated unit, and (ii) the hub-and- the law passed, the phased secondary regulations can spoke model. The former requires countries to develop be adjusted and updated to hone the specifics of the and staff new departments specifically to implement operation of the law and the sandbox’s inter-agency the sandbox. An extreme example is the United collaboration arrangements. These efforts achieved Kingdom’s FCA, which has nearly 100 staff dedicated some early successes in building fintech expertise, to fintech; while not all are solely involved in running encouraging active engagement between policy the sandbox, they provide supporting policy and other makers and fintechs, and integrating policy makers and operational support to the framework. Bank Negara industry stakeholders within broader fintech forums Malaysia also established a new department, the domestically and abroad. Financial Development and Innovation Department, tasking it with operating the regulatory sandbox in coordination with the Financial Technology Enabler Country Example: Colombia — Introducing Group.48 In a hub-and-spoke model, only a skeletal Supportive Fintech Regulation via a Sandbox permanent staff count is maintained, and expertise is In 2018, Colombia’s regulator, Superintendencia drawn both from within the regulator and from outside, Financiera de Colombia (SFC), initiated a legislative as needed. For example, the Bank of Jamaica sandbox, change to launch its InnovaSFC program to encourage although housed in the Financial Market Infrastructure financial sector innovation with targeted regulatory (FMI) department, draws on various other regions of assistance.44 The program has three mechanisms: a the bank for support in determining eligibility and with hub to serve as a single contact point; the sandbox, testing individual firms. A third, less common model, 20 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES entirely outsources running the sandbox to an external to sufficiently adjust their legal or regulatory firm able to support capacity and technical resource framework to regulate or license appropriately the gaps, but this approach also requires substantial firms exiting the sandbox, unintentionally giving financial resources. In all three models, it is vitally those fintechs a competitive advantage by allowing important that supervisors are capable of (i) advising them to operate unhindered or with fewer rules than innovators effectively, (ii) designing relevant, robust incumbents. Lack of technical staff and capacity tests for firms admitted into the sandbox, and (iii) thus may lead to serious consumer protection risks understanding risks. as well as reputational risks for the regulator, who may be held responsible for undue consequences. A survey on innovation facilitators conducted by the Bank for International Settlements (BIS)49 revealed All the expertise necessary to review applications and that most regulators either have a dedicated team for support firms through testing is unlikely to be available their sandbox or are in the process of developing one. from a single set of resources. The skills needed can Only 28 percent of regulators reported that they did vary from supervision to technology to governance, not have a dedicated unit for their sandbox operations making it good practice to have subject matter experts (see Figure 3.3). available, either within the organization or externally, who are able to provide support when needed. In India, Regulators may find that regulatory sandboxes adjacent government bodies may be drawn on for demand more of their time and skill than they had insights as well. anticipated. Since regulators are required to assess complex innovation and innovator applicants, Country Example: Inside the Bank of Thailand’s define testing plans and performance metrics, and Regulatory Sandbox51 supervise participants during their time in the sandbox, ill-equipped and under-resourced sandbox Thailand’s central bank, the Bank of Thailand teams may pose risks to consumers. Despite this, (BOT), established a separate Financial Technology most EMDEs establishing a sandbox opt to hire Department adjacent to the Payment Systems Policy internally,50 even though internal staff may lack Department to oversee BOT’s regulatory sandbox. the technical skills or qualifications needed to The fintech department includes a mix of both IT and deal with complex new fintechs. In addition, some policy experts. The department works closely with regulators may not have the resources and capacity both industry players and other relevant departments in the BOT, such as the Payments System Policy Figure 3.3. Variation in Sandbox Resource Department, Bank Supervision Department, and the Structures Technology Risk Supervision Department. The BOT has also set up multiple cross-functional “squad” teams Currently comprised of representatives from various departments Developing Team who respond to new fintech technologies by sharing 8% different perspectives based on their expertise and learning from one another. No Dedicated Country Example: Leveraging Intra-Government Team 28% Expertise in India The regulatory sandbox set up by the Reserve Bank of India (RBI) has a dedicated staff of four to five personnel responsible for developing the enabling Have Dedicated Team framework and defining the operation and structure of 64% the sandbox. Source: BIS. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 21 For the first cohort, the team chose “retail payments” Sandboxes worldwide vary considerably in the testing as a theme, with the explicit intention of spurring durations used, from two weeks to two years. Testing innovation in the digital payments space and helping periods should be long enough to allow the regulators to offer payment services to the unserved and (and firms) to understand the market impacts of the underserved segments of the population. To support fintech product, but not so long as to mimic licensing the application and eligibility process, they sought without having met the full requirements. While input from the National Payments Corporation of India an optimum period should be decided based on the (NPCI), an initiative of the RBI and the Indian Banks’ regulator’s requirements, the testing period should be Association (IBA), with the primary aim of creating a a minimum of three months and a maximum of one robust infrastructure for India’s entire banking system year, with the option to extend. focused on innovations in retail payments and the For the sandboxes with available data on testing move toward a “less-cash” society. periods, the most common testing period is one year34 While the sandbox is still nascent and results are (33 percent of the sample of 45 sandboxes). About 16 still developing, this exchange of information across percent of sandboxes assess the testing period based on different areas of government provides a good example firm needs, and most provide the option of extending of the options for leveraging expertise across bodies to beyond the stated testing period if the firm and fulfill a common goal. authorities deem it necessary (see Figure 3.4). While the testing period varies with the type and objectives 3.2.4 Testing Durations of the sandbox, having a time-bound testing period is important to avoid using resources on underdeveloped All jurisdictions recognize the need for a defined or unviable innovations. and time-bound testing period. Because a sandbox is primarily useful for making evidence-based decisions on innovative products and services that can impact 3.2.5 Thematic Sandboxes Promoting Specific policy decisions, testing, the source of the evidence Technologies or Products collected, is one of its most important components and Evidence has shown that well-defined, thematic forms the crux of the framework. sandboxes can be effective in encouraging particular Figure 3.4. Variation in Testing Periods of Sandboxes 50 40 Percentage of Sandboxes 33 30 20 20 18 16 10 4 4 2 2 0 1 Year 6 Months 2 Years Varies by 8 Months 9 Months 3 Months 5 Months Applicant Testing Period Source: WBG Research, Appendix 3. 22 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES technologies or products to come to market. While services. The specialized sandbox streamlined the most (about 60 percent)35 fintech sandboxes are geared application processes for thematic innovations, while toward general fintech innovations, some specifically the broader sandbox continued to provide wider adopt themes such as enhancing blockchain technology, coverage for other innovative solutions. The first innovations in insurance technology, payment system specialized sandbox focused on eKYC and digital innovations, or digital authentication and verification onboarding in an attempt to evolve KYC regulation technologies (see Figure 3.5). historically performed in person. Under the specialized sandbox, two fintech companies and seven banks Thematic sandboxes can be effective in promoting tested new eKYC technologies.58 certain technologies and specific policy priorities. This can be done in several ways, including through One of the first participants of the regulatory sandbox, fintech challenges, as in the case of Sierra Leone MoneyMatch — an online cross-border remittance and Mozambique. Malta and Lithuania have created service provider — offered peer-to-peer remittance sandboxes geared toward promoting the use of services and tested digital onboarding by conducting blockchain technology, while Thailand’s thematic multiple video conferences to verify potential clients. sandbox encouraged development of standardized QR The firm created a platform to match individual buyers codes, resulting in QR codes becoming commonplace and sellers of currencies with a focus on SMEs who do for Thai consumers in Thailand and across borders. a lot of cross-border transfers over the course of one day. For verification, MoneyMatch used AI-powered Country Example: Bank Negara Malaysia’s third-party facial recognition. Using the sandbox for a Thematic eKYC Sandbox Track56 controlled roll-out and to test the effectiveness of the eKYC process, MoneyMatch successfully graduated Bank Negara Malaysia (BNM)57 introduced a in June 2019, exiting BNM’s eKYC sandbox and specialized thematic track for its sandbox. Termed receiving approval to operate and use its new KYC a specialized sandbox, it is intended to accelerate methods within the Malaysian market.59 innovations with clear potential to improve financial Figure 3.5. Sandbox Themes by Region 45 42 40 35 30 # of Sandboxes 25 20 15 10 7 6 6 4 4 5 2 0 DFS Blockchain InsureTech Payment DFS Related KYC/ RegTech Innovations Systems to Securities Digital ID Sandbox Theme South Asia North America Middle East & North Africa Latin America & Caribbean Europe & Central Asia East Asia & Pacific Africa Total Source: WBG research, Appendix 3. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 23 The sandbox results helped the BNM enable digital As a result, QR codes are now commonplace in verification and develop new eKYC policies. In Thailand. By end 2019, PromptPay registrations December 2019, the BNM issued an exposure draft reached 49.7 million65 and more than 3.7 million proposing requirements and guidance for eKYC merchants accept PromptPay QR payments (compared implementation.60 to 140,000 merchants accepting cards with 480,000 traditional POS devices).66 Daily transactions through Country Example: Blockchain Fintech Product PromptPay averaged 9.6 million, reaching a peak Testing in Lithuania of 13 million transactions. Moreover, e-payment transactions per user more than doubled from 63 in The Bank of Lithuania (BOL) launched its blockchain- 2019 to 135 transactions in 2020.67 based sandbox, LBChain, in March 2020, with the objective of accelerating the development and The sandbox results also helped build the rails for application of blockchain-based solutions in the cross-border QR payments, which are now available in financial sector and attracting more fintechs to the several ASEAN countries. Thai banks have partnered country. BOL is working with external service with foreign banks — for example, Krungsri with providers to build the LBChain platform, which MUFG in Japan — to enable Thai customers to use allows firms to test blockchain-based solutions while their Thai QR system in foreign shops.68 In addition, guiding them on applicable regulations and providing the BOT and the National Bank of Cambodia entered temporary relief on some supervisory requirements. into an MOU in 2019 to create an interoperable Cambodian-Thai QR system.69 While the regulatory sandbox is in its early phase of operation, it has become the testing environment for The successes described in the country examples are several different types of financial products. Of 21 context specific and must be evaluated in terms of the registrations, 6 fintech companies from 3 different unique circumstances and objectives of the jurisdictions countries were deemed eligible based on criteria implementing them. Often, the success of thematic including genuine innovation, consumer benefit, need sandboxes depends on the availability of supporting for testing in a live environment, readiness for testing, technology or financial sector infrastructure. For and a goal of providing financial services in Lithuania. instance, the success of Bank of Thailand’s thematic sandbox in promoting standardized QR codes in For its LBChain sandbox, BOL provides consultations on the market depended on the availability of BOT’s regulation as well as technical and technological support PromptPay system. Establishing a thematic sandbox to eligible fintech companies to support development of may not in itself guarantee success and will require products into market-ready solutions. These companies careful assessment by policy makers, but it shows have used LBChain to test their products, including promise in supporting wider market development. a KYC solution for AML compliance, a cross-border payment solution, smart contracts for factoring process management, payment tokens, a mobile POS and 3.2.6 Consumer Safeguards payment card solution, a crowdfunding platform, and an Regulators must ensure that participating consumers unlisted share trading platform.61 are not unduly exposed to risks from firms participating in the sandbox. To this end, as revealed in the WBG- Country Example: QR Codes Through the Bank CGAP Innovation Facilitator survey, 70 percent of of Thailand’s Thematic Sandbox62 regulators have safeguards in place to protect the The BOT’s thematic sandbox successfully encouraged consumers using their sandboxes, and 52 percent financial providers to test standardized QR Codes using granted temporary waivers from full licensing regimes the BOT’s PromptPay system.63 In 2017, the BOT but required full authorization at the end of testing (see accepted eight financial institutions into the sandbox Figure 3.6).70 Sandboxes are not meant as fertile ground to test QR codes,64 and five institutions successfully for fintechs looking to exploit regulatory loopholes, and exited the sandbox with approval to provide QR code they do not provide exemptions from extant laws and payment services to the general public. regulations. Instead, sandboxes prescribe proportional legal and regulatory requirements for specific 24 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Figure 3.6. Safeguards Commonly Used in a Sandbox Fit & Proper Assessment Limits on Funds Received from Clients Limits on # of Transactions Limits on # of Clients AML-CFT Rules Minimum Capital Requirements Compensation Scheme Complaints Handing Mechanism Disclosure Requirements 0 10 20 30 40 50 60 70 80 90 100 Percent Source: WBG-CGAP Innovation Facilitator survey. innovations, including (i) restricted authorization; consumer complaints against a sandbox firm have not (ii) rule waivers; (iii) individual guidance; (iv) no- been recorded in any jurisdiction, supervisors should enforcement-action letters; or (v) exemptions. The be attuned to this possibility; post-test consumer proportionality afforded to firms is time-bound and surveys might be a useful source of information. usually restricted to areas over which the jurisdiction has absolute control. 3.3 The Impact So Far A poorly designed sandbox can have severe Countries have established sandboxes as one among consequences for regulators and the consumers many mechanisms to stimulate innovation. Many they protect. Inadequately defined exit strategies sandboxes are established under the assumption that for firms with unfeasible business models can they can contribute to accelerating financial sector or cause serious consumer protection risks. Moreover, country-level outcomes. These targeted outcomes often when jurisdictions with relatively unsophisticated include increasing competition, promoting financial supervisory processes adopt sandboxes, monitoring inclusion, or developing specific technological or risks can potentially be more difficult. product-focused innovations. This section explores linkages between sandboxes’ policy objectives and Consumer protection measures within sandbox their impacts on the broader financial sector. frameworks include the need to provide transparent information to consumers on fees, data protection, and Although it is early to draw firm conclusions, the clarity on the firm’s status as part of a testing process growing number of cases provides useful lessons. and lacking a full license. Most jurisdictions include Emerging key trends include ensuring (i) the sandbox clear details in their application and eligibility criteria aligns with country needs; (ii) the sandbox has describing the need for sufficient funds to cover a focused, well-developed scope and objectives; liability for customer funds, should they be required. (iii) sufficient financial and technical resources are Other common protections include a thorough fit available; (iv) the fintech market is mature, ready, and and proper assessment and limits on the number has demand for a sandbox; (v) robust, well-thought- of transactions and number of consumers that can through exit plans exist; (vi) the sandbox engages in subscribe to the business model. Although cases of interagency and international coordination; and (v) 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 25 the sandbox is agile and flexible in its operation. This According to the WBG-CGAP survey, many authorities section explores some of these emerging lessons. reported instituting some type of regulatory change based on sandbox results. About 50 percent reported Most sandboxes are set up with the aim of achieving that live testing resulted in changed regulations, specific institutional and firm-level outcomes. These while 36 percent reported providing firms with full include improving the supervisory capacity of authorization to proceed (see Figure 3.7).72 regulators, testing or adjusting the appropriateness of current regulations, facilitating engagement between fintechs and regulators, and providing firms with Figure 3.7. Outcomes from Sandbox regulatory guidance. This section explores linkages Participation between a sandbox’s objectives and its impact on institutional arrangements, market development, and 60 individual firms. 50 50 % of Regulators Reporting Impact 3.3.1 Assisting Policy-Maker Decisions and Effecting Regulatory Change 40 36 Regulatory sandboxes are one of several mechanisms 30 authorities may use to facilitate innovation and drive regulatory change through evidence-based 20 policy decisions. Sandboxes are not necessarily uniquely positioned to test all innovations, but they 10 7 are useful in cases that require empirical evidence to support policy development. Like pharmaceutical 0 industry sandboxes, fintech sandboxes are uniquely Change in Issued Change in suited to providing the evidence base needed to Business Model Firm License Regulation support policy decisions and allow technological innovations or new business models into the market. Source: BIS. This is especially relevant for the financial sector when viewed with the regulatory mandate in mind. Several notable examples illustrate this effect. That said, a sandbox test environment is not necessary The Bank of Thailand’s thematic sandbox for QR for regulatory approval or licensing; their usefulness payments, already discussed, helped it engage with the depends largely on the business model being tested. industry and develop related common standards and Where regulatory requirements are unclear or missing business rules. Other sandboxes in Thailand helped or create barriers to entry disproportionate to the introduce regulations and initiatives for robo-advisory, risks, a regulatory sandbox can be beneficial. While P2P lending, and eKYC. Malaysia’s BNM sandbox these are not the only two circumstances under which helped authorities develop new eKYC policies and sandboxes are created, they show the clearest link issue an exposure draft proposing requirements and to direct benefits. Sandboxes can also help build the guidance for eKYC implementation. The Astana stakeholder consensus needed to endorse or support International Financial Center73 in Kazakhstan enabled broader regulatory understanding and change. For issuance of new frameworks on private e-currencies, instance, Kenya’s Capital Markets Authority (CMA) fintech, and crowdfunding.74 In another example, regulatory sandbox testing process led to updated Dubai’s the Financial Services Authority used its guidelines on debt-based crowdfunding, and in sandbox to develop regulations in tandem with testing Brazil, the Central Bank of Brazil reported that its and innovations undertaken by participating firms (see sandbox and the linked LIFT initiative has helped Box 3). the regulator assess risks and changes needed in the regulatory environment.71 26 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Country Example: Using the Sandbox Approach alternative data. The CFBP issued the firm a no-action to Adopt New Regulations in Kenya letter (NAL), referencing the application of the Equal In Kenya, the Capital Markets Authority (CMA) has Credit Opportunity Act (ECOA) and its implementing used the regulatory sandbox testing process to update regulation,77 for its use of artificial intelligence (AI) regulations by allowing firms to experiment outside and machine learning in its credit underwriting and the current regulatory framework. Upon exit from pricing models. the sandbox, participants are either granted a license A report by CFBP in August 2019 highlighted that the to operate in Kenya under existing regulations, or Upstart model increased approval of borrowers by 27 the CMA authorizes temporary operations until new percent relative to traditional models while offering regulations or guidelines are adopted according to 16 percent lower interest rates.78 Moreover, over the section 12 and 12A of the Capital Markets Act. past year it has kept defaults in check and reached For example, one participant, Pezesha, tested an profitability; however, the impact of the COVID-19 internet-based crowd-funding platform through which pandemic on this initiative is yet to be seen. As part of investors can provide loan facilities for small and the NAL, Upstart has continued providing CFBP with medium enterprises. The CMA uses these tests to create simulations on credit reporting, further proving that guidelines for debt-based crowdfunding in Kenya. AI can improve credit scores and credit approval. The Similarly, the Central Depository and Settlement sandbox tests have allayed the regulator’s concern over Corporation (CDSC), the fourth firm to enter the inherent bias in the alternative data and algorithmic sandbox, began testing its proposed screen-based decisioning and have shown that Upstart operates in securities lending and borrowing (SLB) platform for compliance with lending laws and regulations. a period of five months starting April 2020. If the test Sandboxes can provide a useful (virtual) space for succeeds, the CMA will update the current securities innovative start-ups to test new ideas and concepts lending and borrowing regulations to include the in accord with regulatory objectives. They play an screen-based model and will improve the uptake of the important role in providing the concrete empirical bilateral SLB product.75 Country Example: Stress Testing Underwriting Box 3. Dubai’s Progressive Approach Algorithms in the United States’ CFPB Sandbox to Developing Regulation Through a Upstart Network, a fintech that began to operate in Sandbox 2016, uses both traditional underwriting information Dubai’s Financial Services Authority (DFSA) and alternative sources of information — such as introduced an Innovation-Testing License (ITL) in 2017 employment history and educational background that allows eligible fintech firms to test innovations in the sandbox under a restricted license. DFSA used — to evaluate an individual’s creditworthiness. The this model to develop regulations in tandem with Upstart platform pools this alternative data while testing and innovation by firms in the sandbox.79 applying computing and machine learning to identify One example is Sarwa, a technology-based financial relationships and evaluate creditworthiness that might advisory firm that was the first fintech operator to not have been achieved using traditional assessments. receive an ITL. During the testing period, DFSA worked with Sarwa to understand the company’s Concerns about the fairness in algorithmic lending, the underlying operating model and then developed the use of nontraditional data, and the adherence to the fair appropriate regulation to allow its innovative products into the market while meeting the requirements lending laws set out in 1970 necessitated that Upstart for risk mitigation and consumer protection. Sarwa engage with the Consumer Financial Protection completed the regulatory test plan and graduated Bureau (CFBP), a U.S. consumer regulatory agency from the DFSA ITL to a full Dubai Financial Services that just launched its sandbox.76 Over a period of time, Authority license. Prior to receiving the full license, the platform was limited to use by UAE residents, but the CFPB reviewed the model using a series of tests; it is now available to users across the region. going for example, it processed the same loan application forward.80 using both traditional data and Upstart’s model with 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 27 evidence needed to make overarching decisions that While we should not underestimate sandboxes’ lead to regulatory change. When used with this purpose signaling benefits, it should be noted that other more in mind, sandboxes can provide unmatched benefits indirect objectives, including promoting competition, to the policy maker. However, while early evidence spurring innovation, and facilitating exchange with suggests that sandbox programs can result in regulatory market participants, can potentially be achieved change, interviews with some policy makers suggest through other market interventions. An example is that change is often attributed to the open engagement the Monetary Authority of Singapore (MAS), which between regulators and innovators and the specific provided regulatory guidance to approximately 140 guidance received by firms while in a sandbox. As firms in connection with its regulatory sandbox. Of the illustrated in the benefits section below, some of these applications received by MAS, 75 percent were later commonly cited benefits can be achieved using other, withdrawn or allowed to proceed without the need for less resource intensive, initiatives. It is thus difficult to a sandbox.82 This illustrates that if the objective is to quantify the direct impact of a sandbox on instituting reduce regulatory barriers or deepen understanding of regulatory change as compared to successes achieved particular technologies or business models, structures with other innovation facilitators. such as innovation hubs or guidance units set up by the regulator may be effective complements to a regulatory 3.3.2 Benefits for Regulatory Institutions sandbox. Moreover, these alternatives are cheaper and easier to implement, requiring less time or cost Sandboxes offer considerable value to policy investment, and do not lead to the distortionary makers seeking to increase their understanding effects of unlevel playing fields. and capacity to facilitate and regulate a range of fintech innovations. Sandboxes can provide a structured and arm’s-length process through which 3.3.3 Enhancing Financial Inclusion to strengthen dialogue and interaction with the Several sandboxes include a specific mandate industry. About 73 percent of regulators reported that to advance financial inclusion, including those implementing a sandbox contributed to building their in Bahrain, Malaysia, Sierra Leone,87 and India. capacity around fintech, and about 85 percent reported Others, like Jordan,88 have made the development that it helped them to assess the appropriateness of of a sandbox a core part of implementing a national their legal or regulatory frameworks.81 While testing financial strategy that bolsters inclusion by enabling innovative use cases can help build internal capacity an innovation environment (see Box 5). In Mexico, on different fintech innovations and encourage more the sandbox is linked to a financial inclusion mandate market-regulator dialogue, the added value unique under the fintech law.89 Figure 3.8 shows the regional to a sandbox is elevated when authorities focus on concentration and dominant themes of fintech-related applicants that can test existing policy frameworks sandboxes with a financial inclusion focus. against new technologies and business models. Country Examples: Jordan, Bahrain, and Sierra While most regulators we spoke with cited interactions Leone with the industry and regulator as an important benefit of sandboxes, the CGAP-WBG survey also Jordan linked its sandbox activities to its NFIS, which revealed that the vast majority of interactions in the includes a fintech pillar aimed at reducing bottlenecks surveyed jurisdictions took place outside the sandbox to financial inclusion and promoting financial access environment. Informal interactions with market through innovation. The sandbox contributes to participants themselves have revealed preference inclusion goals by testing innovations that support for guidance units within the regulator to provide inclusive and scalable products, services, and delivery targeted advice and direction. Used instead of, or as channels targeted toward Jordan’s priority populations: complements to, regulatory sandboxes, these structures women, youth, refugees, and low-income populations. promise to be more effective and better suited to most Some of the fintech initiatives to support inclusion business needs and able to build similar capacities include interoperable retail payment systems; digital through close engagement with the fintech sector. See payments, such as cash transfers and bill payments; the example in Box 4. 28 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Box 4. Leveraging Other Innovation Tools: Germany The Bundesbank and the Federal Financial Supervisory Authority (BaFin) have taken coordinated steps toward fintech innovation. But, for several reasons, nothing that can be defined as a regulatory sandbox has been set up in Germany. Germany takes an innovation hub approach, as do the majority of the European Union’s member states.83 An important institutional consideration, not only for Bundesbank but also for BaFin, was based on their legal mandate. Often the objective of using a “regulatory sandbox” is to proactively boost competition and, potentially, to relax regulatory or supervisory requirements to achieve this end. Yet, the promotion of competition is not a primary component of German financial supervisors’ mandate, and it was felt that setting up a sandbox could not be justified under strict interpretation of that mandate. They were also concerned that supporting or admitting firms to a regulatory sandbox could generate reputational issues or conflicts of interest that would undermine the supervisors’ fulfillment of their overall mandate. A second consideration is that much of the pertinent law and regulation around financial markets is developed and defined at the EU level due to the need to harmonize regulations across the area. When developing input for reform of EU legislation, authorities, including the Bundesbank and BaFin, coordinate through the European Supervisory Authorities, in particular the European Banking Authority in the case of banks and financial services providers. The EU has no single specific “regulatory sandbox.” Instead, the European Forum for Innovation Facilitators (EFIF), a network of over 35 innovation facilitators84 (including innovation hubs and some regulatory sandboxes) was established;85 there, supervisors could, for example, share information and technological experience. Bundesbank and BaFin are members of EFIF. At the national level, scope nevertheless remains to apply the principle of proportionality and a risk-based approach to supervision in a manner that eases constraints imposed on smaller fintech firms or firms that may pose less systemic risk. Accordingly, in several cases, the risks and size of new fintechs have defined supervisory practices. Interactions with fintechs have also led to adjustments in guidance and application of rules, as in the qualifications of directors, to adjust to the new fintech environment. Third, the Bundesbank and BaFin already had other measures through which to contact and support the start- up and fintech community, including regular “open door” consultation meetings where fintechs could interact with officials to learn to better understand how existing regulations or supervisory practices would apply to their business model. In addition, the Bundesbank, BaFin, and the Ministry of Finance often come together to work on topics of mutual interest. Another form of support to the fintech ecosystem in Germany is the Bundesbank’s Digital Office, which not only looks at trends in the market and interacts with incubators but also reviews how the central bank could use fintech to support its own internal functions. The Digital Office has a partnership with Frankfurt´s TechQuartier, a melting pot for entrepreneurs and innovators from the financial industry. Through this partnership, the Bundesbank is in close contact with the fintech scene. Besides innovation challenges, workshops, and other events, the supervisory garage is a key feature of the partnership. The supervisory garage gives fintechs the opportunity to enter into a simple and direct exchange with Bundesbank officials and provides assistance in navigating through regulations. The Bundesbank is also involved in considering the implications of new business models on market dynamics and stability.86 In 2016, BaFin launched a landing page for start-ups and fintechs. This landing page gives information on typical fintech business models and authorization requirements. In addition, BaFin launched a contact form for start-ups and fintechs on its homepage. The event, BaFin-Tech, hosted by BaFin, enables a broad dialogue with various stakeholders of the German fintech market. BaFin’s fintech unit, Technology-Enabled Financial Innovation, is engaged in identifying, understanding, and assessing fintech innovations and their relevance to and impact on the financial market. The unit aims to develop strategic positions, including on the need for regulatory or supervisory action, in relation to financial innovation. The fintech market in Germany is strong, and the market appears to be content with the role the regulators are playing. and innovative identity solutions to include vulnerable sandbox, and it continues to support both NFIS sections of society. These cannot be solely attributed implementation activities and sandbox activities. It is to the sandbox, however. The sandbox is coordinated important to note that an implementing body — such as and supported by the NFIS implementing entities. the NFIS Secretariat in this context — with sufficient In practice, Jordan’s Financial Inclusion Secretariat capacity or a separate dedicated unit to implement the played an integral role in establishing the regulatory sandbox is vital to its proper functioning. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 29 Figure 3.8. Fintech-related Sandboxes with Financial Inclusion Elements 5 4 4 13 # of Sandboxes 3 2 3 1 1 1 1 0 DFS Innovations KYC/Digital ID Payments API Solutions Blockchain InsureTech Africa Europe & Central Asia Middle East & North Africa East Asia & Pacific Latin America & Caribbean South Asia Source: WBG Research, Appendix 3. *A total of 23 sandboxes are related to financial inclusion or financial inclusion themes. Box 5. Fintech and National Financial Inclusion Strategies (NFIS) The first national financial inclusion strategies were launched around 2010, and by 2019 more than 45 countries had launched one and 39 others were in the process of doing so.90 Recently, these financial inclusion strategies have played a role in encouraging the use of fintech and digital financial services specifically to further financial inclusion goals.91 In a WBG-IMF study we noted that over 60 percent of jurisdictions, primarily in middle-income countries, reported incorporating fintech in an NFIS. These national strategies tend to focus on fostering adoption of fintech (41 percent of survey respondents), encouraging digitization of government processes (41 percent), and establishing a forum for public-private dialogue (33 percent).92 All low-income countries with financial inclusion strategies focus on encouraging fintech adoption and digitizing government services. In comparison, on average, about half the sample of middle-income countries contained strategies to encourage fintech adoption and encourage digitizing government services. About 75 percent of financial inclusion strategies in low-income countries focused on encouraging dialogue between fintech firms and financial sector incumbents, while on average less than 40 percent of strategies in middle-income countries had this as a focus. Fintech-related Actions Within National Financial Inclusion Strategies 100% 100% 75% 56% 60% 50% 40% 40% 32% 26% 28% 28% 28% 18% 12% 12% Encourage the adoption Direct public funding to Encourage digitization of Establish fora to promote the of fintech develop and foster fintech government services (with the dialogue with fintech firms and Government as payee or payer) financial sector incumbents Low income Lower middle income Upper middle income High income Source: Global Fintech Survey (GFS 2019). 30 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Other countries linking financial inclusion to sandboxes Currently little evidence shows that those sandboxes include Bahrain and Sierra Leone. The Central Bank set up with the key objective of financial inclusion of Bahrain (CBB) launched its sandbox initiative in have achieved more to that end than sandboxes set up June 2017 to foster market development that advances with, say, an innovation mandate. One reason for this financial inclusion. It is one of the few sandboxes that limited evidence may be that sandboxes have only been specifically identifies financial inclusion as a primary operating since 2016; however, it is equally probably objective within its framework, stating as its goals that sandboxes in all forms are able to support better “to promote effective competition, embrace new consumer-centric products and services and encourage technology, encourage financial inclusion and improve innovation in the wider economy. customer experience.” In addition, the integration of For instance, the majority of innovations that flow financial inclusion remits within the innovation office through a sandbox often operate within the payments, suggests that CBB is well positioned to monitor the clearing, and settlement spaces, which clearly link impact of financial inclusion throughout the initiative.93 to financial inclusion goals.99 Malaysia provides In Sierra Leone, the sandbox was initiated in April 2018 one of the most successful examples of a sandbox to facilitate new business models to promote greater contributing to financial inclusion when authorities financial inclusion, as well to achieve clear benefits opened up their sandbox to test remote customer for potential consumers. Evaluation criteria in the identification (eKYC). Four firms were allowed to test Sierra Leone sandbox framework require applicants to their products with the Bank Negara Malaysia (BNM) demonstrate how a proposed innovation can advance sandbox. Recognizing the usefulness of eKYC and its the country’s national financial inclusion strategy. The potential for financial inclusion, BNM subsequently framework also allows inclusion objectives to be bound drafted e-KYC guidelines promoting its use and to sandbox participants through requirements that the removing some of the friction in account opening. underserved be included in sandbox testing (collecting Evidence is thus insufficient to demonstrate a direct vital information and data about their needs) and/ causal relation between a sandbox created with an or be direct beneficiaries of the proposed innovation intention of financial inclusion and those created with after deployment. Incentives may also be offered to wider financial inclusion goals. However, it can be innovators who primarily address financial inclusion surmised that properly implemented sandboxes used objectives.94 While it is too early to assess the country- to encourage consumer-focused products and services, level impacts on financial inclusion, early evidence especially to fill a market gap, can potentially impact suggests that creating a focal point for fintech in Sierra broader financial inclusion goals. Leone, such as the sandbox team, has strengthened the overall fintech ecosystem by tapping into latent market demand and creating stronger links between the Bank 3.3.4 Assisting Private Sector Firms of Sierra Leone (BSL) and fintech companies.95 Most sandboxes aim to facilitate market entry of innovative firms that would otherwise struggle to Innovators can test new products and services that better establish themselves due to high regulatory thresholds. meet the needs of the underserved, but according to Evidence is mixed thus far on how successful regulators responding to the WBG-CGAP survey,96 less sandboxes are in this respect. While sandboxes are than a quarter of sandboxes tested focused on business often open to both regulated and unregulated firms, models or technologies that explicitly addressed the some fintechs attribute the ability to access markets to financial needs of the underserved.97 Similarly, the their sandbox participation. However, data collected UNSGSA found98 that regulatory sandboxes “are from policy makers show that many more fintechs have neither necessary nor sufficient for promoting financial been supported by innovation hubs than by sandboxes inclusion,” particularly if they are prioritized over more and accelerators combined.100 In fact, a recent study by impactful financial inclusion reforms. the CCAF and WBG showed that innovation offices Although some sandboxes do have an explicit mandate had assisted 12 times as many firms as sandboxes of financial inclusion, evidence is limited on the success covered in the same survey. of these initiatives in reducing barriers to inclusion. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 31 Other research101 has similarly concluded that certain products or services for up to 12 months without innovation hubs are potentially better suited to, and an Australian Financial Services license or credit more capable of, dealing with the wide variety of license.102 Firms can thus begin operation immediately companies that seek guidance on maneuvering the while keeping the ASIC notified of their plans, should regulatory landscape. the regulator wish to clarify their operations. Figure 3.9. Number of Firms Supported The regulator does not have a defined application by Innovation Offices and Regulatory process. Issues of an unlevel playing field do not arise, Sandboxes and firms are only allowed to operate within certain boundaries, that is, below 100 retail customers and within $A50,000 in individual customer exposure; 2,500 total maximum exposure for all clients is $A5 million 2,163 during the testing period, while complying with 2,000 responsible lending obligations and compensation arrangements. Once the thresholds are breached, they 1,500 must apply for a permanent license. 1,000 Another frequently cited benefit of sandboxes for firms is reduced compliance costs, which can lower 500 barriers to market entry. Although true, as with the 180 alternate avenues for guidance, the sandbox might not 0 be the best tool for reducing costs. Other regulatory Total Number of Firms Assisted by Innovation Office tools, including license exemptions, rule changes, or Total Number of Firms Assisted by Regulatory Sandbox proportionality, could potentially be better suited to Source: CCAF and WBG Regulating Alternative Finance survey. supporting firms in carrying out activities for which regulations already exist. Moreover, in some cases, if Country Example: Australian Licensing sandbox application processes are complex, unclear, Exemption Scheme or constantly evolving, firms may consider entry into In Australia, the Australia Securities and Investment sandboxes to be cumbersome. That said, the firms Commission (ASIC) put forth a fintech licensing with which we spoke that had been through a sandbox scheme. While still referred to as a sandbox, unlike process indicated that they had benefitted from the a supervised sandbox test, the ASIC fintech licensing process, the framework, and the relationship with the exemption allows eligible fintech companies to test regulator (See Box 6). Box 6. Reactions from Fintech Firms103 Indonesia: PrivyID Testing Digital Signatures within Bank Indonesia’s Sandbox Established in 2018, the Bank Indonesia (BI) FinTech Sandbox aims to provide a safe space to test Financial Technology Operators and their products, services, and business models. Fintechs entering the BI sandbox must be listed companies registered with the bank and meet other common sandbox criteria, such as providing innovative and relevant services targeted toward Indonesian customers, readiness to test, and others. PrivyID was an early entrant to BI’s sandbox. A member of the Indonesian FinTech Association (AFTECH), PrivyID was registered by the BI as Indonesia’s first digital identity and legally binding digital signature solution provider. Prior to entering the sandbox, PrivyID provided banks with digital signature technologies for some financial services. However, digital signature solutions for credit card applications were not as yet legal. This prompted PrivyID to apply to the sandbox to test digital solutions to replace hand-written signatures for credit card applications. In early 2019, PrivyID submitted its application to BI’s sandbox. The application process included a detailed interview with BI sandbox staff, adequacy requirement tests, and many follow-up information and documentation requests, all to a relatively tight timeline. The process as a whole took six months before PrivyID received the go- 32 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES ahead to launch its new digital signature solution for credit card applications. Once PrivyID was in the sandbox, many banks were keen to partner with it. The new technology expedited credit card approval processes from three to five days to 15 minutes. Over the course of one year, PrivyID and financial service providers processed e-signatures for more than 50,000 credit card applications. The sandbox provided other benefits to PrivyID. Inputs from Bank Indonesia helped PrivyID refine and adjust its solutions to better meet the needs of the consumer. For instance, PrivyID initially processed credit card applications by requiring users to send information to bank partners that then forwarded the application to PrivyID for verification. Once verified, PrivyID would message users (through SMS) with a unique user ID, a password, and a secure link for signature. However, some customers were unable to find their notifications or links for e-signature. This led PrivyID to include the added ability to use facial recognition against accredited identification documents with added one-time password controlled security methods that could also facilitate the digital signature. PrivyID exited BI’s sandbox in August 2020 after successfully completing all the tests, including no signature or application failures. Since then, providers must still obtain explicit approval from BI to utilize PrivyID’s digital signature solution. Moreover, service providers are required to seek permits from OJK (the Indonesia Financial Services Authority), as the regulator for e-KYC, to employ PrivyID’s solution. While the product still has a low take- up level, the firm indicated that it was very happy with the process and had benefitted immensely from the close relationship with the regulator. Rwanda: Riha Payment System Ltd. Riha, a subsidiary of the technology company Aurora Soft, is a mobile payments company based in Rwanda. Among its products is a payment aggregator platform and a service powered by artificial intelligence (AI) that offers end-to-end business solutions for merchants. Because it uses AI in its business model, Riha was not allowed to launch directly into the market. In 2018, when Riha first sought a way to test its digital payments product, it turned to the newly established sandbox of the Rwanda Utilities Regulatory Authority (RURA),104 because the Central Bank of Rwanda (BNR) did not yet have a sandbox facility. BNR, recognizing the implications of a sandbox for the financial market and the direct impact on consumers, set up a regulatory sandbox of its own, focusing on firms that do not clearly correspond to currently regulated products, for hybrid products, or products that use technology in a novel way. BNR’s process from that point showed agility and a willingness to learn from the market, but it did not come without its hurdles. The sandbox is applicable only to digital payment services (rather than to fintech in general) and was established by way of a secondary measure (Regulation No. 05/2018 of March 27, 2018, governing payment services providers), since it is framed within existing legislation. Entrants need to demonstrate that they (a) address a significant problem or issue or bring benefits to consumers or the industry; (b) improve accessibility, efficiency, security, and quality of payment services; (c) enhance efficiency and effectiveness in managing risks; or (d) address gaps in or open up new opportunities for financing or investments in the country. Since then a number of separate but connected sandboxes have been initiated in the BNR to specifically address gaps identified by regulators, including microinsurance and deposit-takers.105 One initial challenge for Riha was the relationship and mandate with other regulators operating within the jurisdiction, in this case RURA. Riha’s parent company is a technology firm, making it natural to first approach the utilities regulator about entering its existing sandbox. However, lack of clarity on the treatment of innovative business models and the overlying regulatory mandates created a convoluted application process. Moreover, Riha was required by the BNR to satisfy certain prerequisites before it could operate in the market. This included registering itself as a company, not solely as a subsidiary; hiring senior staff members, including a CEO (other than the founder) and a finance director; and appointing an auditor. Both the firm and BNR admit that learning occurred on both sides, demonstrating the value of learning while doing to bring agility and innovation to supervisory processes. Mr. Robert Ford, Managing Director of Riha, described it as a “win-win” situation for both sides and commended the BNR on its willingness to always confer and its readiness to admit when it was unfamiliar with the intricacies of the technology. Riha maintains that it would not have been able to operate and gain traction in the market without the support of BNR’s regulatory sandbox, and BNR attests to gaining an understanding of innovative concepts and the ability to evaluate whether its regulatory framework is fit for purpose. Riha has been in the BNR sandbox for close to two years, and currently no other firm operates there. The firm hopes to be fully authorized in the near future, but it is as yet unclear if this will require establishing a new license. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 33 3.3.5 Fostering Partnerships in the Market Network (AFIN),108 as part of its collaboration with the Monetary Authority of Singapore (MAS) and the Sandboxes can help develop partnerships in the International Finance Corporation (IFC), launched an marketplace, either directly or indirectly. Contrary API (application programming interface) Exchange to the commonly perceived risk that fintechs will (APIX) — effectively an API-themed testing disintermediate banks, evidence so far suggests that environment as a shared infrastructure. Through sandboxes encourage and facilitate partnerships partnerships, this cross-border, industry-led sandbox between them. Although, sandboxes are not the only has been effective in engaging fintech innovators means of fostering partnership models, they offer looking to develop solutions with participating specific beneficial features. For instance, some sandbox financial institutions (see Box 7 below). This designs require fintechs to partner with a licensed firm brought together developers, fintechs, incumbents, to apply, fostering cooperation and knowledge sharing. and the regulator to provide a platform for shared A 2016 study106 showed that more than half of the learning and mutual benefit. Shared platforms, like world’s largest 500 companies work with start-ups and blockchain protocols, open APIs, or shared security prefer partnering with them rather than developing their infrastructure may encourage third-party developers own technologies. This is a synergistic model, with to build next-generation products and reduce time to the incumbent getting new capabilities and the start- market. up gaining legitimacy and access to new distribution. Sandbox support for these partnerships has not been Country Example: Encouraging Fintech demonstrated, however. Partnerships in Brazil One of the most important roles for sandboxes has In May 2018, the Central Bank of Brazil (BCB) been creating continuous dialogue with regulators launched the Laboratory of Financial and Technological that allows participating firms107 to reassure investors, Innovations (LIFT), a fintech incubator and a sandbox thus helping to raise capital and create another kind to accelerate the development of start-ups and accept of partnership. However, this relationship can raise submissions from early-stage innovators.114 concerns about creating an uneven playing field Fintechs entering the BCB’s sandbox are supported specifically with those firms that have chosen not to go by participating organizations and LIFT partners, down the sandbox route. including the Federação Nacional de Associações Sandboxes can also help drive partnerships by dos Servidores do Banco Central (a Brazilian providing shared infrastructure. While not a typical nongovernmental organization) and industry partners, regulatory sandbox, the ASEAN Financial Innovation including researchers, developers, specialists, and Box 7. Early Successes from the ASEAN Financial Innovation Network The ASEAN Financial Innovation Network (AFIN), an initiative of MAS, IFC (International Finance Corporation, the private sector arm of the WBG), and the ASEAN Bankers Association (ABA), is a regional industry marketplace and technical sandbox intended to promote the growth and integration of innovative products and services that advance financial inclusion in the region.109 In September 2018, AFIN launched the API Exchange (APIX) platform, promoting an open architecture platform that enables fintechs and financial institutions to more readily discover, test, and co-design new end-to-end services.110 AFIN has been successful in raising fintechs’ visibility and in helping banks experiment with services based on an API. It aims to share insights and collaborate closely with regulators, to provide them with an opportunity to better understand the practical challenges fintechs face, and to inform efforts toward policy and operational harmonization, which, in turn, can further business and investment opportunities. The APIX ecosystem has thus far brought together more than 600111 fintechs and 150 financial institutions from 26 countries to partner and develop innovative cross-border solutions.112 The platform established numerous partnerships that have been accepted and are currently testing within the sandbox and has signed several MOUs with fintech associations in recent months to expand its regulatory coverage, including bringing Abu Dhabi,113 India, and Hong Kong within AFIN’s network and supporting the G20 TechSprint. 34 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES representatives from firms like Oracle, Amazon Web Figure 3.10. Benefits of Fintech According Services, IBM, and Microsoft. to Regulators Partners provide guidance, as well as products 100 and services to support sandbox applicants with 88 prototypes; in many cases, they partner with start- ups to launch in the sandbox and eventually in the 80 Brazilian market. This approach allows LIFT and 58 the BCB to help stimulate entrepreneurship, increase 60 55 Percentage competition, and introduce a new range of innovative solutions to enhance the Brazilian financial sector, 40 such as increasing financial education and inclusion, making credit cheaper, modernizing legislation, and 18 20 making the financial system more efficient. 0 3.3.6 Stimulating Market Competition Increase Increase Increase Improve Innovation Operational/ Competition Regulatory The role of sandboxes in fostering market Cost and Lower Compliance Efficiencies Barriers and competition depends largely on regulators’ of Fls to Entry Supervision objectives and mandates. Opening up room for competition is one potential result, but other Source: GFS survey. interactions have created unequal playing fields. Policy makers report mixed results when assessing A number of EMDE countries have cited attraction of whether a sandbox has led to increased competition players to their markets as a reason for developing a in their markets.115 For example, approximately, 88 sandbox. While this can be beneficial, it is not sufficient percent report that sandboxes have, in fact, attracted in and of itself as an objective for setting up a sandbox. innovators to their markets, particularly in the form Although jurisdictions like Singapore, Lithuania, and of larger, more established fintechs.116 In a survey of the Financial Services Authority in Seychelles have regulators in countries with sandboxes, approximately as part of their objectives making the market a more 55 percent suggested that one top benefit for the inviting environment for foreign firms, it is unusual to financial sector was increased competition and set up a sandbox with the sole intention of attracting lowered barriers to entry (see Figure 3.10).117 Some foreign players, and the success of such sandboxes has international fintechs have leveraged sandboxes to not yet been observed. enter new markets and credit early and sustained It is more common for competition benefits to arise engagement with policy makers as key factors in their not directly through the operation of a sandbox but success. Sandboxes such as the U.K. FCA sandbox indirectly through linked initiatives, as in Brazil and have been created specifically to increase market Jordan. A sandbox is likely to have a catalyzing effect, competition. This is in line with the FCA’s mandate as but it can have its greatest impact when operating a regulator, and firms’ time to market has been reduced, within a broader strategy or set of initiatives to enable with a 40 percent reduction in application processing fintech. In Singapore, for instance, the sandbox has time reported.118 However, not all jurisdictions have helped attract overseas start-ups to do business there, increased competition as a specific mandate, making contributing to making Singapore a smart financial facilitating competition a less imperative sandbox aim. hub.119 However, a number of other factors are at It is unclear, however, if a sandbox itself drives play that make Singapore an attractive ecosystem, competition and reduces time to market or if a sandbox including its very open economy and a financial system signals to firms that a regulator is open to innovation, with significant cross-border links (notably, dollar thereby attracting innovators to the market. Moreover, funding), particularly to China and ASEAN countries. once firms enter the market, no clear evidence suggests Also, foreign banks have a significant presence, and that firms continue to be active and profitable. the asset management industry channels funds to the 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 35 region from around the world.120 Singapore also has In response to the COVID-19 crisis, Jordan released one of the highest numbers of regulatory alliances and a specific cohort focused on solutions for consumers has cooperative arrangements with eight countries, battling the crisis and introducing competition to plug including the United Kingdom, Australia, and Japan. potential gaps in the market. On the flip side, a sandbox can also create imbalance Country Example: Increasing Demand for and cause level-playing-field concerns among firms InsureTech in Singapore that have chosen not to employ a sandbox process, To encourage fintech innovation in Singapore, the which can reduce sandbox effectiveness in achieving Monetary Authority of Singapore (MAS) launched its goals. Unlevel playing fields may arise in instances a fintech regulatory sandbox in 2016. PolicyPal, an where firms admitted into a sandbox have an “upper insurance technology-based company that uses AI to hand” in attracting investment due to reputational digitize insurance, registered in the MAS regulatory gains and exposure garnered within the sandbox. sandbox in March 2017;121 following six months of Firms accepted into a sandbox may reflect a regulator’s testing, it was the first to graduate.122 PolicyPal and unintentional “stamp of approval.” Some regulators the MAS worked together to optimize insurance have been accused of picking winners and losers, policy options for holders by assessing challenges and firms who have accessed a sandbox are viewed and identifying gaps in insurance policy. According to as having greater regulator interaction and therefore PolicyPal’s founder, the insurance sector grew rapidly a competitive advantage over firms not included in in 2018, when digital disruptions in the traditional the sandbox process. In the United Kingdom, 40 financial services sector created greater opportunities percent of start-ups that flowed through the FCA for InsurTech firms and insurers to introduce new sandbox received investment, either during or after the business models into the market.123 program.126 While in other markets, sandboxes linked to accelerator-like initiatives provide support such as Since PolicyPal entered the market, MAS has taken funding and fintech partnerships. This is mitigated on Inzsure, another InsureTech company aiming to somewhat, although not entirely, by having an open use its digital platform to provide end-to-end service and transparent application process with clearly and reduce transaction costs. A few companies in defined eligibility and entry criteria. the InsureTech space in Singapore now leverage AI, blockchain technology, and internet of things (IoT) Another risk is the possibility that detailed entry technologies, including companies in the region now requirements will stifle firms’ time to market. This is collaborating with the Singapore market.124 especially true for firms that might only need regulatory guidance. Applying to a sandbox takes significant time Country Example: Accelerating Fintech and resources without guaranteeing sandbox entry and Entrepreneurship in Jordan successful exit. At times, eligibility criteria are unclear, A core objective of the Central Bank of Jordan’s requiring even greater time and effort.127 The regulator regulatory sandbox is to contribute to the nation’s is also learning by doing, meaning larger firms could efforts to become a regional financial innovation and gain an unfair advantage because they have more time entrepreneurship hub by “encourage[ing] competition and other resources to invest. and increase[ing] effectiveness and security in money A separate but related risk is that once a firm exits transfers.”125 a sandbox, it may be able to operate with fewer The sandbox interacts with Jordan’s national fintech requirements than incumbents. This is especially hub to create a pipeline of incubated financial true in those jurisdictions with limited capacity technology innovations. The sandbox links with to adjust supervisory frameworks appropriately, other partner companies, like JoMoPay, Jordan’s leading to unlevel playing fields. This is a key risk in national e-payment and mobile payment platform, and jurisdictions where the sandbox merely functions to with a range of innovation facilitators to encourage waive an existing license requirement. While waivers competition and stimulate entrepreneurship within and reducing undue regulatory burdens is useful for Jordan. increasing market competition, other safeguards must 36 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES be introduced to ensure that AML/CFT risks and the achieve broader policy goals. For instance, resources risks of unlevel playing fields are well mitigated. spent on developing effective credit infrastructures or clearing and settlement institutions may be more 3.3.7 Enabling Fintech Market Development critical to enabling fintech than are sandboxes. Sandboxes by themselves are not a turnkey solution Many countries, such as France, Germany, and or a substitute for building effective, permanent Morocco, have eschewed regulatory sandboxes in regulatory frameworks to enable fintech. However, favor of other policy tools and strategies to enable in the right setting, sandboxes are a valuable innovation. Singapore, one of the original proponents tool for enabling fintech by providing empirical of the approach, uses sandboxes as the solution of last evidence and operating within a broader strategy resort. Sandboxes by themselves are not substitutes or set of initiatives. for building permanent regulatory frameworks that stimulate and support orderly adoption of innovation, Fintech and other forms of DFS can play a vital role including from fintech. They should be used as a in extending the reach and widening the access of targeted, temporary measure to achieve a limited set of financial services and, more broadly, the achievement clearly articulated objectives. of the Sustainable Development Goals (SDGs). The gains from fintech are particularly important for developing countries, as fintechs offer opportunities to Country Example: The United Kingdom and the address long-standing constraints. They have already Digital Sandbox helped to bring access to financial services to millions In May 2020, when the COVID-19 pandemic led most of consumers and MSMEs (micro, small, and medium of the world to go online, the FCA piloted a “digital enterprises) around the world. Regulatory sandboxes sandbox” to allow firms to test and develop proofs of can provide valuable insights for policy makers and concept in a digital testing environment while receiving promote innovation when operated within a strategic enhanced regulatory support to tackle the challenges of framework that enables fintech through a set of fintech- the COVID-19 pandemic.128 They are in the process of driven initiatives. However, policy makers should completing their first round of applications specifically be careful not to prioritize sandboxes over other, focused on preventing fraud and scams, improving potentially less resource-intensive, initiatives that can the financial resilience of vulnerable consumers, and Box 8. The Use of Landscape Assessments for Fintech Market Development A landscape assessment of the fintech sector can provide officials with a baseline understanding of the opportunities and challenges that the sector presents. It is important for government and regulators to actively monitor and evaluate the fintech space and look to develop flexible, risk-based regulatory approaches to fintech activities, especially in areas that may potentially promote the entry of smaller players and ensure access to affordable financial services. In April 2019, the Fintech Roadmap Committee of the Nigerian Capital Market unveiled its report, “The Future of Fintech in Nigeria.”129 The report identified fintech’s importance to the country’s capital market, and it both challenges and recommends suitable regulatory frameworks that could allow fintech to play a significant role in Nigeria’s capital market. Another instance is the first Fintech Landscaping Report published by the South African authorities in early 2020, which provides a holistic view of the fintech landscape in the country.130 The report described the landscape of fintech firms operating in the country and identified key enablers and inhibitors in the South African fintech ecosystem. The research helped estimate the market size of each segment and the potential growth per segment. In addition, a sample group of fintech entrepreneurs were interviewed to understand their journey, challenges, and perceptions of the regulatory environment. Feasibility studies and landscape assessments can provide regulators with a baseline with which to determine whether a sandbox is truly necessary to meet the challenges identified. Regulators can then develop metrics based on the information obtained from the initial landscape assessments to demonstrate whether the sandbox has been effective in meeting the needs of the regulator and/or industry. 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 37 improving access to finance for SMEs. While the regulatory coordination globally, and together the specifics are still being ironed out, the access to data, participants support fintech ecosystem development in a collaborative platform, and an API marketplace are several ways, including hosting an annual conference all being considered. This supports the development of on pertinent topics, collecting data on regulatory fintech products and services with a specific aim and bottlenecks and discussing solutions, and releasing end-goal. guidance to the market on key regulatory issues. They proactively assess emerging risks and opportunities Interestingly the digital sandbox also gives observers and recently set up a dedicated virtual space comprising the chance to participate in the sandbox, either to form a regulatory sandbox environment that allows live partnerships with other firms, provide mentorship, or testing of innovations and an innovation accelerator simply observe the process. to provide a collaborative, exploratory environment For fintech to thrive, a multidimensional approach is for financial sector regulators to learn from and work needed, including a gap analysis of existing laws and with each other. The intra-regulatory body also set up a regulations and an open dialogue between regulator Regulatory Guidance Unit to resolve specific questions and industry. A landscape assessment to consider regarding the policy landscape. the country context is a necessary first step for all These efforts are further complemented by the Fourth regulators. Conditions to be gauged include: (i) the Industrial Revolution South Africa partnership, an institutional mission and policy priorities, (ii) legal alliance between partners from the public and private and regulatory framework, (iii) maturity of the fintech sectors, academia, and civil society launched by the segment, (iv) capacity, (v) market conditions and president of South Africa in 2019 to bolster the drive feasibility, (vi) stakeholder ecosystem, and (vii) risks. toward a more digital economy. Regulatory sandboxes, while useful, cannot be used as the sole response to fintech in any jurisdiction and Country Example: Stimulating Demand for should be complemented by other mechanisms or Sierra Leone’s Sandbox133 initiatives to bolster the ecosystem. These can range Sierra Leone, a developing country in Sub-Saharan from key changes to infrastructure, such as payments Africa with high numbers of unbanked citizens, sought systems, or to development of new systems and to meet this challenge by promoting, attracting, and institutions, such as for digital ID. Other important catalyzing development of local financial services, issues may include developing funding options geared specifically to the underserved. To meet this for early-to-late stage companies; promotion of goal, the Bank of Sierra Leone (BSL) initiated a partnerships with incumbents and other players, as regulatory sandbox “pilot program” to encourage, in Indonesia; or MOUs with other countries, such as cultivate, and promote financial innovation. The BSL’s the United Kingdom’s fintech bridges with Australia, sandbox initiative is unique in the sense that the BSL China, Hong Kong, Singapore, and South Korea.131 stimulated demand for the sandbox in a market with limited financial sector depth, low levels of financial Country Example: Inter-Regulator Coordination inclusion, and a nascent fintech market. In partnership in South Africa to Support Growth of the Fintech with Financial Sector Deepening Africa (FSD Africa) Ecosystem and United Nations Capital Development Fund’s South Africa has a fast-growing fintech industry. (UNCDF) Mobile Money for the Poor, and with A collaborative effort by the regulators led to the support from USAID and the Last Mile Trust Fund, Intergovernmental Fintech Working Group (IFWG), the BSL launched the Sierra Leone Fintech Challenge in 2016, comprising representatives from the National in June 2017 to drive demand for the sandbox. Treasury (NT), South African Reserve Bank (SARB), Twenty innovators applied to Sierra Leone’s Fintech Financial Sector Conduct Authority (FSCA), National Challenge, and challenge winners received an Credit Regulator (NCR), Financial Intelligence Centre injection of capital and automatic entry into the first (FIC), and South African Revenue Service (SARS)132 cohort of the financial-inclusion-themed BSL sandbox The IFWG is one of the best examples of inter- pilot program. The first cohort of the BSL sandbox, 38 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES made up of four fintechs, launched in May 2018 and Country Example: Sandbox as Part of a included a mobile payment aggregator, a mobile Coordinated Strategy to Expand Fintech money cash transfer application for agriculturalists, a Ecosystem in South Korea135 financial literacy mobile application, and an electronic In early 2018, South Korea’s government designated money platform. fintech as a leading sector for innovation and has been implementing its Plan for Promotion of Fintech Country Example: Leveraging Multiple Innovation. In January 2019, the government Initiatives to Drive Fintech in Indonesia announced the following strategies to expand the The Indonesian fintech association, AFTECH,134 was fintech ecosystem: implementation of a regulatory established by the industry in 2016 as an umbrella sandbox; revamping outdated regulations; expanding organization for all companies or institutions in the investment in fintech; cultivating new industry sectors; financial sector that embrace technology to empower supporting global expansion; and enhancing digital their businesses. It was appointed by the Association financial security. of Digital Financial Innovation Organizers (IKD) of This strategy involves deregulation measures the Indonesia Financial Services Authority (OJK) and including 2019 amendments to the Supervisory focuses on policy advocacy and driving policy change Rules on Electronic Financial Transactions and the to enable fintech and support the Bank of Indonesia Special Act on Support of Innovation of Finance (BI) and OJK. To do so, AFTECH developed a strategic (Finance Innovation Act). Significant changes framework with three complementary policy tracks: under the Supervisory Rules include adaptations to • Financial inclusion: In line with Indonesia’s enable cloud computing and cloud-based services national financial inclusion strategy, AFTECH for processing of critical financial information. The activities focus on leveraging fintech, identifying Finance Innovation Act amendments include the digital incentives, and exploring public-private following deregulatory measures: partnerships to support financial inclusion. • Regulatory exemptions through the sandbox for • Support to the Bank of Indonesia: Activities focus innovative financial services for up to four years; on interoperability, collective risk management, and • A one-stop-shop model through which the Financial Open APIs. Services Commission provides rapid regulatory • Support to the OJK: AFTECH provides advisory services to firms; and coordination support to better link OJK’s regulatory • Core business outsourced to fintech companies sandbox to ten cross-industry working groups on without requiring separate regulatory approval each various themes. time under a “designated agent system.” AFTECH also works to communicate emerging trends A preliminary evaluation of the sandbox as a part of and policies and works closely with the media to the overall sector strategy suggests job growth in the ensure journalists are regularly updated and informed fintech space, with 225 jobs added by 23 new firms. on fintech topics. In addition, investment in the sector increased, with about 11 fintech firms attracting KRW 120 billion in investment as of 2019, and expansion of fintech firms to different global markets, specifically in Southeast Asia, the United Kingdom, Japan, and Hong Kong.136 3. GLOBAL EXPERIENCE AND LESSONS LEARNED 39 40 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 4. Evaluating Sandbox Impacts and Remaining Agile Sandboxes are relatively new regulatory instruments, and policy makers cannot always accurately predict market reactions and impacts from them. Translating objectives into measurable indicators and targets to ensure that progress is tracked and assessed can be challenging. Simple quantitative metrics often used by sandboxes, such as the number of firms admitted into the sandbox, are not wholly useful dimensions for quantifying achievements or testing policy implications. However, measuring the intangible benefits (such as catalytic change among policy makers) or the indirect impacts of a sandbox on national goals are harder to compute. A robust monitoring and evaluation system to measure the impact of a sandbox on country, regulatory and market level outcomes is a powerful and effective tool for identifying obstacles, demonstrating results, and ensuring the sandbox is moving toward its stated objectives. Having a clear hypothesis and logical framework can underpin sandbox success and regulators’ ability to craft relevant quantitative and qualitative metrics. Evaluation systems also help regulators pivot their sandboxes to better respond to industry needs (see Box 9). In some cases, a sandbox framework led to unsustainably large numbers of applicants or failed to attract more than a few (or any) applicants. These scenarios might have been avoided with proper assessment prior to sandbox implementation and monitoring with proper metrics throughout implementation, thus guiding regulators toward appropriate responses to evolving requirements. In Malaysia, evaluation results prompted the BNM to develop a specialized sandbox to standardize and streamline testing parameters for more efficient experimentation and data collection, and in South Korea and Hong Kong, regulatory sandbox evaluations helped sandbox teams document success in reaching policy objectives as well as identify opportunities to further improve the regulatory sandbox and promote the fintech industry. For policy makers looking to establish a sandbox and corresponding monitoring and evaluation system, we suggest a matrix structure with three measurement stages and four measurement levels that cut across them. (see Figure 4.1). 4. EVALUATING SANDBOX IMPACTS AND REMAINING AGILE 41 Box 9. Remaining Agile — Using Evaluation Results to Adjust a Sandbox Over Time The Monetary Authority of Singapore (MAS) launched a “Sandbox Express” in August 2019 to address the large number of applicants applying to the sandbox and the cumbersome approval and application processes that restricted potential firms from entering. Singapore’s Sandbox Express is intended to help encourage and speed up processes for experimentation and adoption of innovative technologies in the financial sector,137 specifically for firms with low and well-understood risks, allowing them to embark on their experiments more quickly within the predefined sandbox. Applications to the Sandbox Express are fast-tracked, and approval decisions are made within 21 days. The criteria the MAS uses to assess an application include (i) technological innovativeness of the financial service, and (ii) fitness and propriety of the applicant’s key stakeholders. The Sandbox Express operates in two predefined areas: insurance brokering and establishing market operators.138 The boundaries, expectations, and regulatory reliefs are predetermined, and applicants must declare they will fully comply with all expectations of the predefined sandbox, including providing clear disclosures and obtaining acknowledgements from users before onboarding them as customers. Figure 4.1. A Suggested Measurement Framework for a Regulatory Sandbox Measurement Framework Initial Measurement Ongoing Mid-Term and of Applicants and Cohorts Monitoring Final Evaluations Country-level, Financial Sector Outcomes Regulatory Outcomes Indicator Type Firm Level Outcomes Operational and Institutional Outcomes Source: WBG. • Initial Measurement: This first stage should focus test sandbox goals. The metrics can be monitored on defining indicators for sandbox applicants in by sandbox staff throughout the testing stage and line with the objectives of the sandbox framework. evaluated during the exit stage, with a focus on It should include not only business metrics but potential wider policy implications. Authorities regulatory and market outcomes, including those should avoid turning the assessment and monitoring that test assumptions and provide policy insights. of the applicant into a “check-the-box” exercise and These initial metrics defined by the policy makers should instead ensure the applicant has a clear link will help assess an applicant’s feasibility for the to enhancing sandbox policy goals. sandbox and how the applicant will contribute to and 42 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES • Ongoing Monitoring: Throughout the sandbox Each of the above measurement stages are intersected process, regulators should implement a series of by levels at which outcomes should be measured. ongoing assessments to measure the progress of These should include: (a) country-level financial the sandbox framework and the firms within it. sector outcomes, (b) regulatory outcomes, (c) firm- and Ongoing assessments often measure, for instance, market-level outcomes, and (d) operational outcomes. (i) continued suitability and relevance of each For each of these levels, especially useful indicators sandbox firm and cohort against sandbox metrics; and metrics — both qualitative and quantitative — that (ii) direct and indirect institutional changes and can provide insightful results have been highlighted in benefits that can be attributed to the sandbox; and the sections below. Like measurement stages, outcome (iii) operational efficiency of the sandbox process, areas might overlap. both for regulators and for firms that move through the sandbox process. Such assessments should 4.1 Country-Level Outcomes measure the progress and outcomes of the sandbox on an on-going basis and support policy makers in Country-level outcome indicators should focus on how remaining agile, understanding policy implications, well the sandbox contributes to broader financial sector and adjusting their sandbox and legal or regulatory outcomes. This may include, for instance, national framework as needed. financial inclusion goals, economic measures such as the ability to attract foreign talent and improve growth, • Periodic and Final Evaluations: Periodic and final or broader digital development. Understandably, these evaluations should be conducted at the end of a are difficult to measure or attribute to the workings sandbox process or after a defined duration. This is of a sandbox and hence should relate to goals clearly a point-in-time evaluation and should be positioned defined for the sandbox at its outset. to help determine the impact of a sandbox on broader financial sector and national goals, such Some examples of specific outcomes include nationwide as building institutional capacity, enabling firms rise in financial inclusion levels, increased ease of doing to come to market, growing the broader fintech business, or increased numbers of products or services ecosystem, or contributing to national financial targeting the unbanked. Outcomes can be measured inclusion progress. Such assessments often require through global data indicators, as in the World Bank broader data collection efforts and, in some cases, Findex, or in-country measurements, such as national econometric modeling. financial inclusion surveys assessing innovation uptake and/or feedback surveys from users of services offered While the stages have been listed as separate, they by firms participating in sandbox tests. can often have many overlapping elements. Aside from collecting data directly from the firm, sandbox Country Example: Sandbox Regulatory teams should consider complementary data sources, Outcomes in South Korea such as leveraging insights from stakeholders through The Korean government introduced a multisector consumer surveys or feedback forms, including regulatory sandbox in January 2019.139 As part of grievances and claims from customers through the overall sandbox, the financial sector sandbox complaint handling and other mechanisms. Market was launched by the Financial Services Commission research can also complement business metrics to (FSC)140 on April 1, 2019, under the Special Act on understand a firm’s market impact. Putting in place a Financial Innovation Support. plan to collect and leverage different data points and indicators is a useful exercise for sandbox teams. This The sandbox is not limited to financial sector will help policy makers conduct effective evaluation of innovations and prides itself on providing quick the sandbox’s impact and support the ability to adjust reviews (50 days on average, from application to sandbox operations and processes to the needs of the selection) that can result in temporary approvals. A policy maker, the consumer, and the market. “fast track” is available to further expedite review of cases similar to ones previously decided. 4. EVALUATING SANDBOX IMPACTS AND REMAINING AGILE 43 To evaluate the impact of the sandbox on overarching change in its eKYC regulation) is simple to measure, policy goals, the authorities implemented an evaluation knock-on effects and dependencies prove more against their own metrics of success. The 2019 difficult. evaluation results included the following outcomes:141 The metrics should test regulatory assumptions. For (i) fintech job growth: 23 fintech firms have added this, a clear understanding of the policy questions 225 more jobs; (ii) increased investment: 11 fintech that each applicant raises is critical; for instance, firms have been able to attract KRW 120 billion so far, should peer-to-peer lenders be considered as collective and KRW 10 billion worth of additional investment investment schemes and hence receive the same is expected within the next year; and (iii) global treatment? Do crypto exchanges pose undue risks to expansion: 7 fintech firms have either expanded their consumers? Clear questions will enable the policy business to overseas markets (Southeast Asia, United makers to design indicators that test the corresponding Kingdom, Japan, and Hong Kong) or are in discussions implications of a new technology or digitally enabled to do so. business model. Based on sandbox operations, the FSC recently Key considerations to developing policy-led indicators announced plans to further improve the regulatory include assessing risk, micro and macro shocks, sandbox using the following regulatory measures, behavioral reactions, sector-wide interactions, and among others, to promote fintech industry scale-up:142 contagion. One way to establish a universe of possible • Bring the total number of firms participating in the policy indicators is to use “stress scenarios” (see sandbox to over 100 by the end of 2020. Figure 4.2) to test market responses that in turn can inform regulatory policy or oversight. • Improve sandbox rules and practices: Figure 4.2: Sample stressors, regulatory implications, • Support protection of innovative ideas and and their corresponding measurement metrics technologies through intellectual property rights (e.g., provide legal advice or expedite patent Following are some examples that policy makers can dispute resolution). consider when assessing metrics to test particular regulations or policy frameworks: • Minimize the additional requirements imposed on the designated cases; and • Policy indicators: • In cases of mergers and acquisitions of the • Introducing new regulation or amendments designated services, grant continuation of to existing regulation to support digitization. designation status. This could include laws and acts, regulations, directives, circulars, guidelines, or explanatory • Provide budgetary support for testing, security notes. For instance, a new law to license payment inspection, office space, etc., as well as system providers (PSPs), regulate crowdfunding, one-on-one mentoring from designation to or support eKYC would constitute a change. commercialization of innovative financial solutions. • Contributing to policy projects through analytical insights. • Set up a supervisory framework tailored to supporting fintech firms. • Supervision indicators: • Adapting the supervisory process in response to 4.2 Regulatory Outcomes market developments. This level should include regulatory outcomes from • Implementing proportionate regulation to lower innovations operating within the sandbox or from the barrier to responsible innovation that might knowledge and intelligence gathered that impacts have been faced with unduly burdensome regulation, supervision, or policy. While direct regulatory requirements. regulatory change (as seen, for example, in Malaysia’s 44 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Figure 4.2. Sample Stressors, Regulatory Implications, and Their Corresponding Measurement Metrics Potential Risks and Stressors Deposit flight from incumbents Regulatory Implications to fintech banks Measurement Metrics Increased number of NPLs in Consumer protection fintech players AML/CFT compliance Safeguarding funds Pandemic situation Implications for oversight and Data privacy and cybersecurity supervision The failure of a bank where the AML / CFT fintech player held its funds Data breaches Stability risks Business continuity measures Loss of license or partnership with a correspondent bank Liquidity risks Competition implications Mystery shopping excercise Credit risks Operational resilience Insolvency • Including certain types of innovations within the • Case studies demonstrating the response of regulatory perimeter, as a direct consequence of incumbents to new entrants to the market. assessment of risk posed. • The nature of the support provided by the sandbox • Reviewing and addressing consumer protection that may have contributed to greater regulatory issues that may arise with new innovations. certainty for firms. • Interacting fairly and transparently with new • Evidence of how the sandbox has helped firms entrants. establish themselves in the market. • Improving inter- and intra-regulator coordination. • Number of new innovations, products, or services that have entered the market. • Assessing imbalances in level playing fields (i.e., monitoring of market distortion, anticompetitive • Increased volumes or values of particular services, behaviors, etc.). like e-lending, wealth management through robo- advisory services, online accounts, etc. • Adjusting or thoroughly evaluating current frameworks based on sandbox tests. • Increased number of financial service providers competing in the market and hence bringing in more 4.3 Firm and Cohort-Level Outcomes competition and consumer-centric products. As with other indicators, both qualitative and • Increased access to regulatory expertise to get quantitative indicators should be considered when innovative ideas to market. looking at the firm-level outcomes. Aside and beyond • Ratio of firms adapting to existing regulatory business metrics collected through firms, such as frameworks, compared to those requiring new or number of consumers, value of transactions, and so greatly modified regulation. on (see Box 9), the sandbox framework should also develop metrics to assess the sandbox’s impact on the • Firm reports around increased regulatory certainty, market and market players. guidance, and engagement from authorities. Metrics to test such hypotheses may be nuanced, but here are a few examples: 4. EVALUATING SANDBOX IMPACTS AND REMAINING AGILE 45 Box 10. Sample Quantitative Indicators to Measure the Operations and Functioning of a Sandbox The following list provides examples of indicators that policy makers can consider when assessing the firm benefits and operational efficiency of a sandbox. While these metrics may not provide a holistic measurement of a sandbox’s impact, they shed light on its efficiency and operational elements: • Number of applicants and number of applicants accepted into the sandbox. • Average length of time to accept applicants into sandbox, to test offerings, and to exit. • The time to come to market with and without a sandbox. • Number of sandbox tests (successful and unsuccessful). • Number of companies graduated from the sandbox or successfully completing testing. • Number of firms formally registered or with formal authorization to operate in the market following successful testing. • Number of firms successfully graduated from the sandbox that are currently still in operation. • Number of non-sandbox firms operating in the market under an adjusted legal or regulatory framework based on sandbox test. • Number of firms receiving investment as a result of the sandbox program. • Number of new consumers receiving financial services or products under the sandbox tests. Country Example: Evolving Sandboxes to Meet 4.4 Operational Outcomes Industry Needs — The Case of Hong Kong At the operational or institutional level, indicators The Hong Kong Monetary Authority (HKMA) should assess the ongoing appropriateness of the initially launched its fintech supervisory sandbox sandbox internally, analyze the resources and (FSS) in September 2016 as a program for incumbent capacities used during implementation, and evaluate if banks. During the first year of operation, however, the sandbox is contributing to overarching institutional HKMA received applications from technology firms goals. For instance, a regulator may hypothesize that requesting direct access to the FSS and soliciting a sandbox will strengthen an institution’s capacity to feedback on emerging fintech projects. Against this regulate fintech. To test this hypothesis, a regulator backdrop, HKMA upgraded to FSS 2.0 in 2017. This may want to consider indicators that herald particular version includes expanded access for both incumbents institutional changes such as the following: and nonbank technology firms; an FSS Chatroom to provide streamlined access, feedback, and support for • Widespread buy-in and catalytic change across the market participants; and increased formal coordination institution. between HKMA, the Insurance Authority, and the • Increased regulatory capacity in certain areas (e.g., Securities and Futures Commission on tests that may data analytics) in response to market trends. cut across multiple regulatory perimeters. By the end of August 2018, HKMA had received around 170 • Improved knowledge of the fintech sector and gaps requests to access the chatroom. Nearly 70 percent identified in business areas’ fintech knowledge and of these requests were made by nonbank technology know-how. firms from Hong Kong and overseas. • Signaling and market perception of regulator’s openness to enabling fintech. 46 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES • The establishment of new, related units or increased for the regulator (or for firms, with respect to departmental capacity of the sandbox. providing regulatory certainty). • New staff hired with backgrounds more aligned • Sufficient information provided by the sandbox to regulating or supervising particular elements of for departments to re-evaluate and, in some cases, fintech, per sandbox recommendations. adjust their regulatory frameworks. • An understanding of information flows from • New regulatory initiatives developed in part from sandbox trails, and how the information is used and the results of the sandbox. acted upon. • Adjustment of the sandbox itself, based on feedback • Cost-benefit analysis of capacity and resources used and lessons learned. by the sandbox compared to policy outcomes. • The infrastructural and operational cost of sandbox • Assessment of whether a sandbox was the most to regulators. appropriate tool or if other approaches (e.g., an innovation hub) could have yielded similar results Box 11. Evaluation — Review of the Regulatory Sandbox Framework In October 2017, the U.K. Financial Conduct Authority became the first regulatory agency to publicly release metrics concerning its sandbox framework and lessons learned.143 The analysis provides a thorough overview of the first year that the sandbox was in operation. The four main objectives of the FCA’s sandbox framework are: • Reducing the time and, potentially, the cost of getting innovative ideas to market; • Enabling greater access to finance for innovators by reducing regulatory uncertainty; • Enabling more products to be tested and, thus, potentially introduced to the market; and • Allowing the FCA to work with innovators to ensure that appropriate consumer protection safeguards are built into new products and services. The analysis established that the sandbox reduced the time and cost of getting innovative ideas to market, enabled firms and regulatory officials to understand how receptive consumers are to innovative products and services, and allowed the FCA to work with innovators to build appropriate consumer protection safeguards into new products and services. An additional benefit, but also a risk, is that the sandbox helped facilitate access to finance for innovators, hence potentially influencing investment outcomes that could be construed as potential conflicts of interest. It is understandably difficult to set wholly transparent and objective criteria for accession and graduation from the sandbox. The risk remains that the sandbox gives market power to the first candidates, allowing them to automatically gain a regulatory approval premium when seeking new investors. The FCA also noted numerous challenges to its sandbox framework, several of which sit outside the FCA’s purview to address. For instance, the FCA stated that certain firms’ access to banking services, including obtaining a bank account, was particularly problematic. Further, the smooth integration with APIs between start-ups and financial institutions is challenging. The FCA also noted that firms with certain business models had a greater difficulty meeting the initial regulatory requirements to become authorized, causing small-scale testing to be particularly difficult. The FCA has used the insights gained to inform future sandbox development and to feed into the FCA’s broader regulatory work. 4. EVALUATING SANDBOX IMPACTS AND REMAINING AGILE 47 48 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 5. Concluding Notes Regulatory sandboxes have become synonymous with fintech innovation and offer the unique benefit of providing the empirical evidence needed to substantiate decisions. However, they have more often been used as a signaling mechanism to show that the regulator is open to dialogue. In addition, as we have observed, although sandbox designs can differ widely, the degree of institutional or, rather, country-based isomorphism is hard to ignore. Sandboxes have had some vital direct benefits, such as introducing regulatory change and variations to the regulatory perimeter, and they have also influenced future supervisory methodology and, in rare cases, have supported the regulator’s competition mandate. For firms, sandboxes have been known to offer a faster route to market and a better understanding of the regulatory hurdles they need to cross. Moreover, from the lessons learned globally we can surmise that sandboxes have a number of indirect, often intangible, benefits, such as catalyzing the ecosystem, encouraging intra-regulator cooperation, and identifying or attracting firms, both local and further afield, to the market. But are these effects enough to validate the need for a sandbox? The answer lies in having clear, tightly defined objectives and conducting a cost- effectiveness threshold by comparing the costs and outcomes of alternative policy options. Despite successes, implementing a sandbox, particularly in EMDEs, is not always the right solution for unlocking financial innovation, and it can potentially pose unexpected burdens on regulators, as well as risks, such as creating unlevel playing fields in the market. Before embarking on creating a regulatory sandbox, authorities should step back and objectively review the environment in which they operate, specifically, (i) the existing legal and regulatory framework within that jurisdiction; (ii) the capacity and resources available to the regulator; (iii) the maturity and pervasiveness of the fintech market; and (iv) broader market conditions, including competition criteria. These factors require careful consideration and should be evaluated144 (along with other criteria) to understand whether a sandbox approach is the most appropriate for the given country context. 5. CONCLUDING NOTES 49 Once set up, the supervision of sandbox firms’ directly and indirectly to the wider policy and operations should be supplemented by frequent regulatory environment. A sandbox should contribute monitoring of the sandbox’s continued relevance to the design of, but cannot replace, an adequate legal and its ability to be agile and fail-fast and contribute and regulatory framework: Sandboxes don’t decrease — but rather, they increase — the need for skillful supervision. 50 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES APPENDIX Methodology, 1. Definitions, and Data Sources Employed This report in intended to complement “How Regulators Respond to Fintech: Evaluating Different Approaches — Sandboxes and Beyond” (2020)145 and was spurred by requests for further detail on the form, focus, and operation of the different sandbox approaches adopted. The analysis here is based on a number of inputs, including interviews with supervisory authorities, desk-based research and literature reviews, analysis of the survey by the World Bank Group (WBG) and the Consultative Group to Assist the Poor (CGAP) on innovation facilitators as well as the global fintech survey by the WBG and the International Monetary Fund (IMF). Defining sandboxes. This report defines sandboxes as a controlled, time-bound, live testing environment, which may feature regulatory waivers at regulators’ discretion.146 However, we have included all frameworks referred to as sandboxes by the jurisdictions in which they were created. The focus of this report is on sandboxes that support fintech or fintech-enabled innovations. Hence, the analysis in this report is limited to sandboxes operated by financial sector regulators with specific regard to enabling fintech innovations. Hence, sandboxes created by information and communication technology (ICT) or utilities authorities or any that are sector agnostic or not run by regulators have not been considered. In keeping with the WBG-IMF Bali Fintech Agenda (BFA), “fintech” refers to the “advances in technology that have the potential to transform the provision of financial services spurring the development of new business models, applications, processes, and products.”147 Defining the status of sandboxes. A comprehensive compilation of fintech sandboxes was undertaken as part of this report, with each sandbox categorized as either “announced” or “operational.” For this report, a regulatory sandbox is considered “announced” if the relevant authorities have not only stated their intention to open it but have also offered publicly available information on the sandbox process, relevant laws, and other details. To be categorized as “operational,” publicly available information must indicate that firms are already enrolled in the sandbox. APPENDIX 1. METHODOLOGY,, DEFINITIONS, AND DATA SOURCES EMPLOYED 51 Quantitative data sources employed. Two main • The IMF and World Bank Global Fintech Survey sources of quantitative data were used for the analyses (GFS 2019) collected responses from nearly 100 presented in this report. member countries on progress in relation to the 12 elements of the BFA. Collated data was drawn • The World Bank and CGAP Innovation Facilitator from the FSB-BCBS survey and the CCAF-WBG Survey (2019), conducted jointly between study on alternative finance. February and April 2019, gathered information on regulatory innovation facilitators, including Other reports and papers from standards-setting accelerators, sandboxes, and innovation hubs. bodies, individual jurisdictions, and international Approximately 31 responses were collected from organizations have also informed this paper and are regulatory agencies in 28 countries, including referenced in the endnotes. jurisdictions in Africa, the Americas, Asia, and Europe. 52 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES APPENDIX 2. Further detail on Innovation Hubs148 An innovation hub, or innovation office or lab as they are sometimes called, can provide a dedicated point of contact where firms can raise inquiries with competent authorities on fintech-related issues or seek nonbinding guidance on regulatory and supervisory expectations, including licensing requirements. Most commonly, hubs provide support, advice, guidance, and even, in some cases, physical office space, to regulated and unregulated firms. Single points of contact, a dedicated unit, an identified network of experts, or similar organizational arrangements can all be considered innovation hubs. In essence, an innovation hub can take any form that will be beneficial and suitable to the regulator while signaling to the market that the regulator is keen to interact with and enable the emerging field of fintech. Although providing guidance tends to be its most common function, a hub’s functions can range, for instance, from hosting and attending industry events to providing assistance in applying for authorization on new products. Hubs facilitate engagement between regulators and innovators, acting as forums for mutual learning as well as for policy and regulatory guidance. Supervisors may use innovation hubs to understand and monitor new business models and technologies as well as to identify regulatory and supervisory challenges associated with fintech. The Australian Securities and Investments Commission (ASIC) provides an example. It set up an innovation hub in 2015 to assist fintech start-ups in navigating the regulatory system and its laws, including by providing informal guidance from senior regulatory advisers about the overarching regulatory framework and questions relating to ASIC’s relief powers. For the regulator, this interaction provides information about emerging fintech issues that are potentially relevant to policy development. Other regulators, such as, Malaysia (Digital Finance Innovation Hub) and Thailand (OJK Infinity),149 have set up innovation hubs with players beyond the financial sector that not only provide regulatory clarity but also enable collaboration among service providers, including financial institutions, fintech start-ups, and academics. APPENDIX 2. FURTHER DETAIL ON INNOVATION HUBS 53 An innovation hub can be particularly useful for Recognizing the common challenges and the cross- jurisdictions considering a new approach to fintech, border nature of fintech, hubs have also been set since the hub can be less resource-intensive to up on a global level to support and encourage establish than a sandbox. Hubs can complement coordination among international regulators and to other approaches and are a good way for regulators pool resources. An example of this is the Bank of to gauge the interest and maturity of the market. In International Settlements (BIS), which established addition to requiring fewer resources, according to innovation hub(s) with the explicit intention of the WBG-CCAF survey on Regulating Alternative supporting central bank collaboration on research Finance,150 respondents report that innovation offices and innovation in financial technology and of supported a much larger number of firms than accelerating the banks’ digital efforts while pursuing regulatory sandboxes. their statutory objectives. 54 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES APPENDIX 3. Database of Global Sandboxes Data collected from the 73 sandboxes around the globe is tabulated on page 56. However for a more interactive experience, please go to Key Data from Regulatory Sandboxes across the Globe. APPENDIX 3. DATABASE OF GLOBAL SANDBOXES 55 Database of Sandboxes Currently Announced or in Operation (Up-to-date as of November 2020) General Information Nature of Regulator/Operator Legal System Eligibility Criteria Features of Sandbox WBG Type of Country Name of Type of Civil Law/ National Innovation Need Ready Genuinely Location of Description of Type of Current Financial Operational # of Status of Testing Date Region Economy Regulator/ Regulator Common Law/ Financial Provides Authorization to Innovative Firm Sandbox Sandbox Absence of Inclusion Status Firms in Firms in Period Established (AE/ Operator Hybrid System/ Inclusion Consumer to Participate Test Governing Focused Sandbox Sandbox EMDE) Religious Law Strategy Benefit in Sandbox Regulation? East Asia & AE Australia ASIC Securities Common Law - 3 3 3 3 - DFS related to Product - - Operational 7 "Fintechs can 12-24 Jun-16 Pacific (1) Regulator Securities (phase obtain ‘class months 1); financial waiver’ to advice, issuing test without a credit contracts license" and crowdsourced funding (phase 2) Middle East EMDE Bahrain CBB Central Bank Civil Law - 3 - 3 3 - General innovations Thematic, - 3 Operational 11 - 9-12 Jun-17 & North in DFS Cross- months Africa (2) border Latin EMDE Barbados CBB, FSC Central Bank Common Law - - 3 - 3 - General DFS (only Product 3 - Operational 1 - 8 Oct-18 America & one firm- Bitt Digital months Carribean Inc.- participated + (3) graduated. Tested digital wallet using blockchain tech. Latin EMDE Bermuda BMA Central Bank Common Law - 3 3 3 - - "InsurTech Product/ - - Operational 2 Modified - Sep-18 America & innovations" Policy license in Carribean Sandbox (4) Latin EMDE Brazil BCB Central Bank Civil Law 3 3 3 - 3 - "Laboratory of Product/ - 3 Operational 18 - 3 May-18 America & Financial and Policy months Carribean Technological (5) Innovations (LIFT)" Latin EMDE Brazil CVM Central Bank, Civil Law 3 3 3 - - - Specific for Product - - Announced - Will receive - Jan-20 America & Securities cryptocurrency and temporary Carribean Regulator, digital token authorization (6) Ministry of issuance Finance East Asia & EMDE Brunei AMBD Central Bank Common Law - 3 3 3 3 3 General innovations Product - - Operational 5 - 6-24 Mar-17 Pacific (7) in DFS months Europe EMDE Bulgaria MOF Ministry of Civil Law - - - - - - Sofia Regtech - - - Announced - - - Feb-20 & Central Finance Sandbox Asia (8) North AE Canada CSA Securities Common Law - 3 - 3 3 3 Securities focused Product/ - - Operational 10 - Varies Feb-17 America (9) Regulator sandbox Policy East Asia & EMDE China PBOC Central Bank Civil Law 3 - - - - - General innovations Product/ - - Operational 6 - - Dec-19 Pacific (10) in DFS Policy Latin EMDE Colombia SFC Financial Sector Civil Law - 3 3 - 3 - innovaSFC- Fintech Product/ 3 3 Operational 6 Granted - Oct-19 America & Regulator product/policy Policy license Carribean innovation (11) Reference 6. https://www.coindesk.com/brazil-financial-authorities-announce-regulatory-sandbox-for-blockchain 1. (a) https://www.bakermckenzie.com/en/-/media/files/insight/publications/2018/12/guide_intlguideregulatorysandboxes_dec2018.pdf; (b) https://techwireasia. 7. (a) https://www.ambd.gov.bn/development/fintech; (b) https://www.ambd.gov.bn/SiteAssets/fintech-office/FTSG%20v1_final.pdf com/2020/02/australian-government-makes-amendments-to-fintech-regulatory-sandbox/; (c) http://download.asic.gov.au/media/4112096/licensing-exemption- 8. https://emerging-europe.com/business/bulgaria-to-launch-first-regtech-sandbox-in-balkans/ for-fintech-testing-infographic.pdf 9. (a) https://www.securities-administrators.ca/industry_resources.aspx?id=1588; (b) https://www.securities-administrators.ca/industry_resources.aspx?id=1626; 2. https://www.tamimi.com/law-update-articles/central-bank-bahrain-cbb-launch-regulatory-sandbox-fintech-firms-2/ (c) https://www.osc.gov.on.ca/en/NewsEvents_nr_20170223_regulatory-sandbox.htm 3. (a) http://www.centralbank.org.bb/regulatory-sandbox/sandbox-participants; (b) https://www.ft-legal.com/resources-and-news/2019/07/16/barbados-regulatory- 10. (a) http://www.chinabankingnews.com/2020/04/28/china-to-expand-fintech-sandbox-trials-to-six-more-cities/; (b) http://www.chinabankingnews. sandbox-approves-first-participant/ com/2020/01/15/first-batch-of-trial-applications-for-beijings-fintech-sandbox-revealed/; (c) http://www.chinabankingnews.com/2019/12/06/beijing-launches- 4. https://cdn.bma.bm/documents/2019-03-28-05-10-19-BMA-Insurance-Regulatory-Sandbox-Innovation-Hub-Guidance-Note.pdf fintech-regulatory-sandbox-with-central-bank-backing/ 5. https://www.liftlab.com.br/docs/Regulamento_LIFT_en.pdf 11. https://www.colombiafintech.co/novedades/superfinanciera-lanza-sandbox-para-el-desarrollo-de-fintech 56 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES General Information Nature of Regulator/Operator Legal System Eligibility Criteria Features of Sandbox WBG Type of Country Name of Type of Civil Law/ National Innovation Need Ready Genuinely Location of Description of Type of Current Financial Operational # of Status of Testing Date Region Economy Regulator/ Regulator Common Law/ Financial Provides Authorization to Innovative Firm Sandbox Sandbox Absence of Inclusion Status Firms in Firms in Established (AE/ Operator Hybrid System/ Inclusion Consumer to Participate Test Governing Focused Sandbox Sandbox EMDE) Religious Law Strategy Benefit in Sandbox Regulation? Europe & AE Denmark DFSA Financial Civil Law - - - - - - General innovations - - - Operational - - 6-8 Dec-19 Central Asia Supervisory in DFS months (12) Authority Middle East EMDE Egypt CBE Central Bank Civil Law In 3 3 3 3 - General innovations Product/ - - Announced - - - Jul-19 & North development in DFS Policy Africa (13) Africa (14) EMDE Eswatini CBE Central Bank Hybrid System - 3 3 3 3 - General innovations Product/ - - Announced - - - Mar-19 in DFS Policy East Asia & EMDE Fiji RBF Central Bank Common Law 3 3 3 3 3 Specifies General innovations Product/ - 3 Operational - - Upto 12 Jan-20 Pacific (15) that foreign in DFS Policy months entities are included in 'intended participants' Europe & EMDE Georgia NBG Central Bank Common Law - 3 3 3 3 - General innovations Product/ In some - Announced - - Varies Feb-20 Central Asia in DFS + evaluation Policy cases (16) of regulation East Asia & EMDE Hong Kong HKMA, SFC Central Bank, Common Law - 3 - 3 3 - Fintech Supervisory Policy - - Operational 46 Sandboxes Varies Sep-16 Pacific (17) (China) Securities Sandbox only open Regulator to regulated activities using fintech. East Asia & EMDE Hong Kong Insurance Other Govt. Body Common Law - - - 3 3 - InsureTech Focus Product - - Operational - Varies Dec-17 Pacific (18) (China) Authority Europe & EMDE Hungary MNB Central Bank Civil Law - 3 3 3 3 - General innovations Product/ - - Operational - Receive - Dec-19 Central Asia in DFS Policy Financial (19) Innovation Testing Environment (FITE) license South Asia EMDE India RBI Central Bank Common Law 3 3 3 - 3 3 Retail Payments Product, 3 3 Operational - - 6 May-19 (20) Thematic months South Asia EMDE India IRDAI Other Govt. Body Common Law 3 3 3 3 - - Focus on insurance Product - - Operational 33 - 6 Oct-19 (21) products months South Asia EMDE India SEBI Securities Common Law 3 3 3 3 - - Focus on securities Product - - Operational - Firms will - Feb-20 (22) Regulator related products/ be granted services limited certificate of registration East Asia & EMDE Indonesia OJK Financial Civil Law 3 3 - - 3 - General innovations Product/ 3 3 Operational 1 - Upto 12 Sep-18 Pacific (23) Supervisor in DFS Policy months East Asia & EMDE Indonesia Bank Indonesia Central Bank Civil Law 3 3 3 - 3 - Payment System Thematic - - Operational 34 - 6 Dec-17 Pacific (24) Innovations months 12. https://www.computerweekly.com/news/252462611/Danish-financial-services-regulator-launches-sandbox-initiative 20. (a) https://www.gamechangerlaw.com/rbi-opens-applications-to-its-first-cohort-under-the-regulatory-sandbox/?utm_source=Mondaq&utm_medium= 13. (a) https://fintech.cbe.org.eg/home/sandboxCohorts?en; (b) http://www.shalakany.com/the-cbe-launches-its-first-fintech-regulatory-sandbox/14; (c) https:// syndication&utm_campaign=LinkedIn-integration; (b) https://rbidocs.rbi.org.in/rdocs/PublicationReport/Pdfs/ fintech.cbe.org.eg/home/sandboxCohorts?en, (d) http://www.shalakany.com/the-cbe-launches-its-first-fintech-regulatory-sandbox/ ENABLING79D8EBD31FED47A0BE21158C337123BF.PDF 14. https://www.centralbank.org.sz/fintech/sandbox/CBEFINANCIALTECHNOLOGYREGULATORYSANDBOXGUIDELINES.pdf 21. (a) https://www.business-standard.com/article/companies/irda-gets-close-to-170-applications-under-sandbox-approvals-by-end-of-fy20-119121600699_1.html; (b) https://teamleasecompliance.com/updates/article/6738/irdai-issues-guidelines-on-operational-issues-pertaining-to-the-regula/ (c) https://www. 15. https://www.rbf.gov.fj/getattachment/Left-Menu/Financial-Inclusion/Policy-Areas/FinTech-Regulatory-Sandbox-Guidelines-December-2019.pdf?lang=en-US asiainsurancepost.com/ecoinvestdemography/irda-approves-33-regulatory-sandbox-proposal 16. https://www.nbg.gov.ge/uploads/legalacts/project/2020/regulatory_sandbox_framework_200219_eng.pdf 22. (a) https://www.dataguidance.com/news/india-sebi-issues-framework-regulatory-sandbox; (b) https://www.sebi.gov.in/legal/circulars/jun-2020/framework-for- 17. https://www.hkma.gov.hk/eng/key-functions/international-financial-centre/fintech/fintech-supervisory-sandbox-fss/ regulatory-sandbox_46778.html; (c) https://www.asiainsurancepost.com/ecoinvestdemography/irda-approves-33-regulatory-sandbox-proposal 18. None 23. (a) https://www.credolab.com/news-press/ojk-officially-recognizes-credolab-as-first-credit-scoring-fintech-firm; (b) https://www.lexology.com/library/detail. 19. https://www.mnb.hu/en/innovation-hub/regulatory-sandbox aspx?g=ba27fa50-283e-4a2e-b969-386e412dfb64; (c) https://www.sebi.gov.in/legal/circulars/jun-2020/framework-for-regulatory-sandbox_46778.html 24. (a) https://id.rajahtannasia.com/media/2990/ahpclientupdate-2january2018.pdf; (b) https://id.rajahtannasia.com/media/2990/ahpclientupdate-2january2018.pdf APPENDIX 3. DATABASE OF GLOBAL SANDBOXES 57 General Information Nature of Regulator/Operator Legal System Eligibility Criteria Features of Sandbox WBG Type of Country Name of Type of Civil Law/ National Innovation Need Ready Genuinely Location of Description of Type of Current Financial Operational # of Status of Testing Date Region Economy Regulator/ Regulator Common Law/ Financial Provides Authorization to Innovative Firm Sandbox Sandbox Absence of Inclusion status Firms in Firms in Established (AE/ Operator Hybrid System/ Inclusion Consumer to Participate Test Governing Focused Sandbox Sandbox EMDE) Religious Law Strategy Benefit in Sandbox Regulation? Europe & AE Isle of Man Digital Isle of Other Gov't Body Common Law - - - - - - Blockchain Thematic - - Operational - - - Nov-19 Central Asia Man (Exec Agency Sandbox (25) within Dep't for Enterprise) Latin EMDE Jamaica BOJ Central Bank Common Law 3 3 3 3 3 - Delivery of payment Product - 3 Announced 4 (2 Issued Up to 24 Mar-20 America & services pending license months Carribean license) (26) East Asia & AE Japan FSA, Govt. of Financial Civil Law - 3 3 - 3 - FinTech Proof-of- Thematic 3 3 Operational 1st - 5 Nov-17 Pacific (27) Japan Supervisor, Other Concept (PoC) cohort- 5 months Govt. Body Hub: Customer firms, 2nd identity verification cohort- 2 + automating firms, 3rd customer suitability cohort: 5 determination firms Middle East EMDE Jordan CBJ Central Bank Hybrid System 3 3 3 - 3 - Specifies Thematic - 3 Operational - - Upto 12 Apr-18 & North crypto-currencies, months Africa (28) blockchain and DLT platforms, electronic payments, savings and financing, remittances, e-KYC and RegTech. Europe & EMDE Kazakhstan AFSA Other Govt. Body Civil Law - 3 3 3 3 - General innovations Cross- - - Operational 22 - - Jul-18 Central Asia in DFS border (29) East Asia & AE Korea, South FSC, FSS Securities Civil Law 3 3 - - 3 - General innovations Policy - - Operational 36 Partial 2 years Apr-19 Pacific (30) Regulator, in DFS + evaluation authorization Financial of regulation (upto 4 years Supervisor around blockchain (exemption technology from regulation after full authorization) Africa (31) EMDE Kenya KCMA Securities Common Law - 3 3 3 3 3 Innovation in Product, 3 - Operational 3 - Upto 12 Mar-19 Regulator Capital Markets Policy months Middle East EMDE Kuwait CBK Central Bank Hybrid System - 3 3 3 3 3 FinTech products Product, - - Operational - - 12 Nov-18 & North or services for Policy, months Africa (32) electronic payment Thematic of funds 25. https://www.digitalisleofman.com/sectors/blockchain-isle-of-man/ 30. (a) http://www.shinailbo.co.kr/news/articleView.html?idxno=1278649; (b) http://www.koreaherald.com/view.php?ud=20190110000327 26. (a) https://www.dataguidance.com/news/india-sebi-issues-framework-regulatory-sandbox; (b) https://www.sebi.gov.in/legal/circulars/jun-2020/framework-for- 31. (a) https://www.google.com/search?rlz=1C1GCEB_enUS886US887&sxsrf=ALeKk03B2jZOsV911tOb8ZaJsAI61t8R4Q%3A1586627118515&ei=LgKSXomAH-q regulatory-sandbox_46778.html CytMP56-JgAM&q=kenya+regulatory+sandbox&oq=kenya+regulatory+sandbox&gs_lcp=CgZwc3ktYWIQAzIECCMQJzoECAAQRzoHCCMQs 27.https://medium.com/tokyo-fintech/the-japan-fsa-regulatory-sandbox-b7e9f38e962e AIQJ0oJCBcSBTEyLTc0SgkIGBIFMTItMThQ3pIBWNvIAWCFywFoAXACeACAAWeIAZMPkgEEMjMuMpgBAKABAaoBB2d3cy13aXo&sclient=psy-ab&ved= 0ahUKEwiJz6jO9uDoAhVqgXIEHedXAjAQ4dUDCAw&uact=5; (b) https://disrupt-africa.com/2019/08/3-kenyan-fintech-startups-picked-for-cma-regulatory- 28. http://www.cbj.gov.jo/EchoBusV3.0/SystemAssets/9328fddf-3f3d-40d8-9ed3-d98bbc89db20.pdf sandbox/; (c) https://disrupt-africa.com/2019/08/3-kenyan-fintech-startups-picked-for-cma-regulatory-sandbox/ 29. (a) https://focus.world-exchanges.org/articles/kazakhstan-poised-be-forge-fintech-startups; (b) https://afsa.aifc.kz/files/pages/316/documents/39/list-of-fintech- 32. https://www.cbk.gov.kw/en/legislation-and-regulation/regulatory-sandbox/general-framework lab-participants.pdf 58 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES General Information Nature of Regulator/Operator Legal System Eligibility Criteria Features of Sandbox WBG Type of Country Name of Type of Civil Law/ National Innovation Need Ready Genuinely Location of Description of Type of Current Financial Operational # of Status of Testing Date Region Economy Regulator/ Regulator Common Law/ Financial Provides Authorization to Innovative Firm Sandbox Sandbox Absence of Inclusion status Firms in Firms in Established (AE/ Operator Hybrid System/ Inclusion Consumer to Participate Test Governing Focused Sandbox Sandbox EMDE) Religious Law Strategy Benefit in Sandbox Regulation? Europe & EMDE Kyrgyzstan BKR Central Bank Common Law - - - - - - General innovations Product - - Announced - - - Apr-20 Central Asia in DFS aimed (33) to lower barriers for introducing innovative technologies in the banking sector and the payment systems landscape. Europe & AE Lithuania BOL Central Bank Civil Law - 3 3 3 3 - Blockchain-based Product, - - Operational 6 - 6-12 Mar-18 Central Asia FinTech Solutions Thematic months (34) (LBChain Platform) East Asia & EMDE Malaysia BNM Central Bank Common Law 3 3 3 3 3 - Digital ID Solutions Thematic - 3 Operational - - - Oct-16 Pacific (35) Europe & AE Malta Malta Gaming Other Govt. Body Hybrid System - - - - 3 - General innovations - - - Announced - - - Jan-19 Central Asia Authority in DFS (36) Africa (37) EMDE Mauritius EDB Other Govt. Body Hybrid System 3 - 3 - 3 - General innovations Product, 3 - Operational 9 Regulatory - Sep-18 in DFS Policy Sandbox license Latin EMDE Mexico CNBV, MoF, Central Bank, Civil Law 3 3 3 3 3 - General innovations - - 3 Operational - Regulatory 2 years Oct-19 America & and Banxico Financial in DFS Sandbox Carribean Supervisor, license (38) Ministry of Finance Africa (39) EMDE Mozambique Bank of Central Bank, Civil Law 3 - - - - - General innovations Thematic - 3 Operational 5 - 6 May-18 Mozambique, Other Govt. in DFS months FSDMoc Body Europe & AE Netherlands AFM and DNB Central Bank, Civil Law - - 3 - 3 - General innovations Policy - - Operational - Partial Varies Jan-17 Central Asia Financial in DFS authorization (40) Supervisor Africa (41) EMDE Nigeria CBN and Central Bank, Hybrid System 3 - (only need - - - Fintech Industry Product - - Operational - - - Mar-18 NIBSS Other Govt. to register Innovation Sandbox Body with Financial Services Innovators to participate) Africa (42) EMDE Nigeria SEC Securities Hybrid System 3 3 3 - 3 - Innovation in DFS Product, - - Announced - - - May-18 Regulator related to Securities Policy Europe & AE Norway MoF Ministry of Civil Law - 3 3 - 3 - General innovation Product - - Operational - - - Nov-18 Central Asia Finance in DFS (43) South Asia EMDE Pakistan Securities and Securities Civil Law 3 3 3 3 3 - General innovation Product - 3 Announced - - - Feb-20 (44) Exchange Regulator in DFS Commission of Pakistan (SECP) 33. WB internal 39. http://fsdmoc.com/wp-content/uploads/2018/05/FSDMo%C3%A7-and-BdM_PRESSREALESE-ENG_Sandbox-Launch-17.05.18.pdf 34. https://www.lb.lt/en/lbchain 40. https://www.bakermckenzie.com/-/media/files/insight/publications/2020/05/a_guide_to_regulatory_fintech_sandboxes_internationally_8734.pdf?la=en 35. https://www.bnm.gov.my/index.php?ch=57&pg=137&ac=533&bb=file 41. (a) https://techpoint.africa/2019/12/10/nigeria-first-fintech/; (b) https://techcabal.com/2019/12/09/financial-service-innovators-launches-first-nigerian-industry- 36. (a) https://www.mga.org.mt/mga-launches-phase-2-of-its-sandbox-regulatory-framework/; (b) https://www.mfsa.mt/wp-content/uploads/2019/07/20190704_ innovation-sandbox-with-backing-from-flourish-efina-nibss-and-cbn/ MFSA-FinTech-Strategy-Pillar-1.pdf 42. http://sec.gov.ng/regulatory-sandbox-assessment/ 37. (a) https://www.edbmauritius.org/media/1659/guidelines-rsl.pdf; (b) https://www.edbmauritius.org/newsroom/posts/2019/january/edb-issues-regulatory- 43. https://iclg.com/practice-areas/fintech-laws-and-regulations/norway sandbox-licences-to-fintech-companies-for-their-innovative-projects/ 44. (a) https://www.secp.gov.pk/wp-content/uploads/2020/01/Press-Release-Jan-21-SECP-opens-first-cohort-under-the-Regulatory-Sandbox.pdf; (b) https://www. 38. (a) https://www.sandboxchallenge.com/en/home/; (b) https://www.mondaq.com/mexico/fin-tech/891724/mexico39s-regulatory-sandbox-in-2020; (c) https:// secp.gov.pk/document/secp-regulatory-sandbox-guidelines-2019/?wpdmdl=37476 iclg.com/practice-areas/fintech-laws-and-regulations/mexico#:~:text=The%20Mexican%20Fintech%20Act%2C%20issued,to%20carry%20out%20their%20 operations%3A&text=(ii)%20A%20regulatory%20sandbox%20for,licensed%20and%20non%2Dlicensed%20companies. APPENDIX 3. DATABASE OF GLOBAL SANDBOXES 59 General Information Nature of Regulator/Operator Legal System Eligibility Criteria Features of Sandbox WBG Type of Country Name of Type of Civil Law/ National Innovation Need Ready Genuinely Location of Description of Type of Current Financial Operational # of Status of Testing Date Region Economy Regulator/ Regulator Common Law/ Financial Provides Authorization to Innovative Firm Sandbox Sandbox Absence of Inclusion Status Firms in Firms in Established (AE/ Operator Hybrid System/ Inclusion Consumer to Participate Test Governing Focused Sandbox Sandbox EMDE) Religious Law Strategy Benefit in Sandbox Regulation? East Asia & EMDE Philippines BSP Central Bank Civil Law 3 3 3 - 3 - FinTech 'Test and Policy 3 - Operational - Operators Varies Jun-01 Pacific (45) Learn' Framework/ given letter of Sandbox 'no objection' Europe & EMDE Poland PFSA/ KNF Financial Civil Law - - - - 3 - General innovation Product - - Operational 8 - - Nov-18 Central Asia Supervisor in DFS (46) Europe & EMDE Russia CBRC Central Bank Civil Law 3 3 3 3 3 - DFS which require Policy - - Operational 15 - - Apr-18 Central Asia changes in legal (47) regulation + blockchain pilot Africa (48) EMDE Rwanda BNR Central Bank Civil Law In 3 3 3 3 - General innovation Product, - 3 Operational "Only 1 - 6 Oct-18 development in DFS Thematic fintech months enrolled to test mobile wallet solution" Middle East EMDE Saudi Arabia SAMA, SACMA Central Bank, Religious law 3 3 3 3 3 - General innovation Product, - - Operational 30 LOA- 6 Feb-19 & North Securities in DFS Policy temporary months Africa (49) Regulator restricted permission Europe & EMDE Serbia NBS Central Bank Civil Law - - - - - - Innovations in Policy - - Announced - - - Feb-20 Central Asia Payment Services (50) Africa (51) EMDE Sierra Leone Bank of Sierra Central Bank Common Law 3 3 3 3 3 - General fintech Product, - 3 Operational 4 (winners - 12 Apr-18 Leone innovations Thematic of a months fintech challenge) East Asia & AE Singapore MAS Central Bank Common Law - 3 3 3 3 - Fintech product/ Product/ - - Operational 3 - 6 Nov-16 Pacific (52) service innovation Policy months East Asia & AE Singapore MAS Central Bank Common Law - 3 3 3 3 - Insurance Policy Product - - Operational - - Upto 9 Aug-19 Pacific (53) Bots months Africa (54) EMDE South Africa IFWG Central Bank, Hybrid System - 3 3 3 3 - General innovation Policy, - - Operational - - 6 Apr-20 (collaboration) Financial in DFS Cross- months Supervisor, border Ministry of Finance, Other Gov't Bodies Europe & AE Spain Ministry of Ministry of Civil Law - - - 3 3 - General innovation Product - - Announced - - - Mar-20 Central Asia Economy & Finance in DFS (55) Business South Asia EMDE Sri Lanka CBSL Central Bank Hybrid System 3 3 3 3 3 - General innovation Product, 3 3 Operational - - 9 Feb-20 (56) in DFS Policy months Europe & AE Switzerland FINMA, Federal Financial Civil Law - - - - - - Innovations in Product, - - Operational - Issued Varies Dec-18 Central Asia Council Supervisor, Other Payment Services Thematic 'Innovator (57) Govt. Body & Blockchain license' 45. (a) https://www.gsma.com/mobilefordevelopment/wp-content/uploads/2012/06/Philippines-Case-Study-v-X21-21.pdf; (b) https://www.bis.org/review/r180716a. 52. (a) https://www.mas.gov.sg/-/media/MAS/Smart-Financial-Centre/Sandbox/FinTech-Regulatory-Sandbox-Guidelines-19Feb2018.pdf?la=en&hash=B1D36C055 htm AA641F580058339009448CC19A014F7; (b) https://asiatimes.com/2020/02/singapore-firm-graduates-from-regulatory-sandbox/ 46. (a) https://emerging-europe.com/business/poland-launches-first-regulatory-fintech-sandbox/; (b) https://www.knf.gov.pl/en/MARKET/Fintech/Regulatory_ 53. (a) https://www.mas.gov.sg/news/media-releases/2019/mas-launches-sandbox-express-for-faster-market-testing-of-innovative-financial-services; (b) https:// Sandbox www.marketsmedia.com/singapore-launches-sandbox-express/ 47. (a) http://www.cbr.ru/eng/press/event/?id=2407#highlight=sandbox; (b) http://www.cbr.ru/eng/fintech/regulatory_sandbox/#highlight=sandbox 54. (a) https://www.ifwg.co.za/regulatory-sandbox/; (b) https://international-adviser.com/fintech-programme-goes-live-in-south-africa/ 48. (a) https://www.uncdf.org/article/5216; (b) https://www.newtimes.co.rw/business/central-bank-grants-testing-approval-emerging-fintech-firm 55. (a) https://www.financierworldwide.com/digital-transformation-of-the-financial-system-in-spain#.XpRwJ8hKg2w; (b) https://www.mineco.gob.es/portal/site/ 49. (a) http://www.sama.gov.sa/en-US/Regulatory%20Sandbox/Documents/Regulatory_Sandbox_Framework_English_Nov4.pdf; (b) https://www.fintechfutures. mineco/menuitem. com/2020/04/saudi-arabias-fintech-sandbox-welcomes-nine-more-start-ups/ 56. (a) https://www.cbsl.gov.lk/sites/default/files/cbslweb_documents/press/notices/notice_20191230_fintech_regulatory_sandbox_in_2020_e.pdf; (b) https:// 50. https://www.ekapija.com/en/start-up/2054586/nbs-encourages-testing-of-innovative-ideas-in-payment-services-without-expenses-and economynext.com/sri-lanka-regulatory-sandbox-open-for-fintech-experiments-central-bank-57399/; (c) https://www.cbsl.gov.lk/sites/default/files/cbslweb_ documents/press/notices/notice_20200217-FinTech_Regulatory_Sandbox_of_CBSL_Framework_e.pdf 51. https://www.uncdf.org/article/3635/four-fintechs-approved-to-enter-the-sierra-leone-sandbox-programme 57. https://www.bakermckenzie.com/-/media/files/insight/publications/2020/05/a_guide_to_regulatory_fintech_sandboxes_internationally_8734.pdf?la=en 60 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES General Information Nature of Regulator/Operator Legal System Eligibility Criteria Features of Sandbox WBG Type of Country Name of Type of Civil Law/ National Innovation Need Ready Genuinely Location of Description of Type of Current Financial Operational # of Status of Testing Date Region Economy Regulator/ Regulator Common Law/ Financial Provides Authorization to Innovative Firm Sandbox Sandbox Absence of Inclusion Status Firms in Firms in Established (AE/ Operator Hybrid System/ Inclusion Consumer to Participate Test Governing Focused Sandbox Sandbox EMDE) Religious Law Strategy Benefit in Sandbox Regulation? East Asia & AE Taiwan FSC Financial Civil Law - 3 3 - 3 3 General innovation Product - - Operational 6 - Upto 12 Apr-18 Pacific (58) Supervisor in DFS months East Asia & EMDE Thailand BoT (launched Central Bank Hybrid System 3 3 3 3 3 - KYC and identity Product, - 3 Operational 6 Operate 6-12 Feb-20 Pacific (59) regulatory verification utility Thematic under limited months sandbox in scope: Dec 2016 for provide cross- FIs & NBFIs bank digital w/products ID verification related to to only new lending, savings payments and accounts fund transfers) East Asia & EMDE Thailand BoT Central Bank Hybrid System 3 3 3 3 3 - " Use Product, - - Operational 4 - Up to 12 Dec-17 Pacific (60) of blockchain for Thematic months letter of guarantee and cross-border fund transfer" East Asia & EMDE Thailand BoT Central Bank Hybrid System 3 3 3 3 3 - QR code standard Product, - 3 Operational 8 in - Up to 12 Jun-17 Pacific (61) for e-Payment Thematic sandbox, months 5 graduated East Asia & EMDE Thailand SEC Securities Hybrid System 3 3 - 3 3 - KYC for e-trading Product, - 3 Operational 10 - Up to 12 Jun-17 Pacific (62) Regulator platform Thematic (selected months through fintech challenge) East Asia & EMDE Thailand OIC Other Govt. Hybrid System 3 3 - - 3 - "InsurTech Product, - 3 Operational 5 - Up to 12 Jun-17 Pacific (63) Body (Office innovations" Policy months of Insurance Commission) Middle East EMDE Tunisia The Central Central Bank Hybrid System - - - - - - General innovations - - - Announced - - - Jan-20 & North Bank of Tunisia in DFS Africa (64) (BCT) Europe & EMDE Turkey Ministry of IT Other Govt. Body Civil Law 3 - - - - - Blockchain-based Product, - - Announced - - - Sep-19 Central Asia Solutions Thematic (65) Middle East EMDE UAE (Abu ADGM Other Govt. Body Common Law - 3 3 3 3 - Digital Sandbox for Product - 3 Operational - - - Nov-16 & North Dhabi) DFS (API Solutions) Africa (66) Middle East EMDE UAE (Abu DFSA Other Govt. Body Common Law - 3 3 3 3 3 Innovation testing Product 3 3 Operational 6 (first Sandbox 6-12 Nov-17 & North Dhabi) license: Fintech cohort), License: months Africa (67) product/ service 7 (second FinTech innovation cohort) Startups granted Innovation- Testing License 58. (a) https://www.inhouselawyer.co.uk/legal-briefing/major-changes-to-taiwan-financial-services-sandbox-experiments-token-offerings-and-internet-banks/; 63. (a) https://events.development.asia/system/files/materials/2019/04/201904-implementation-regulatory-sandboxes-global-overview.pdf; (b) https:// (b) https://www.businesswire.com/news/home/20190219006098/en/EMQ-Fintech-Startup-Accepted-Taiwan%E2%80%99s-Regulatory-Sandbox; (c) http://www. dfsobservatory.com/sites/default/files/DFSO%20-%20The%20State%20of%20Regulatory%20Sandboxes%20in%20Developing%20Countries%20-%20 winklerpartners.com/?p=9700 PUBLIC.pdf; (c) https://www.bangkokpost.com/business/1452594/insurtech-hub-in-the-works-to-help-industry-integrate 59. (a) http://silklegal.com/wp-content/uploads/2017/05/BOT-Guidance-on-Regulatory-Sandbox-full.pdf; (b) https://www.bot.or.th/Thai/PressandSpeeches/Press/ 64. https://allafrica.com/stories/202001210707.html News2563/n0663e.pdf 65. (a) https://www.ledgerinsights.com/turkey-blockchain-regulatory-sandbox/; (b) https://cointelegraph.com/news/turkey-announces-plans-for-a-national- 60. https://www.asiablockchainreview.com/thailand-blockchain-landscape-and-regulation/ blockchain-infrastructure 61. https://www.finextra.com/pressarticle/71573/five-banks-exit-thai-regulatory-sandbox-with-qr-payment-codes/retail 66. https://www.adgm.com/setting-up/fintech/overview 62. https://www.bakermckenzie.com/-/media/files/insight/publications/2020/05/a_guide_to_regulatory_fintech_sandboxes_internationally_8734pdf?la=en 67. (a) https://www.menabytes.com/whitepaper-fintech-sandboxes-mena/; (b) https://www.dfsa.ae/FinTech APPENDIX 3. DATABASE OF GLOBAL SANDBOXES 61 General Information Nature of Regulator/Operator Legal System Eligibility Criteria Features of Sandbox WBG Type of Country Name of Type of Civil Law/ National Innovation Need Ready Genuinely Location of Description of Type of Current Financial Operational # of Status of Testing Date Region Economy Regulator/ Regulator Common Law/ Financial Provides Authorization to Innovative Firm Sandbox Sandbox Absence of Inclusion Status Firms in Firms in Established (AE/ Operator Hybrid System/ Inclusion Consumer to Participate Test Governing Focused Sandbox Sandbox EMDE) Religious Law Strategy Benefit in Sandbox Regulation? Europe & AE UK FCA Financial Common Law 3 3 - 3 3 - General innovation Product, - 3 Operational 5 cohorts Restricted 3-6 Jun-16 Central Asia Supervisor in DFS Policy so far authorization months (68) (29 firms accepted into the last cohort) North AE USA Bureau of Other Govt. Body Hybrid System - 3 - - - - General innovation Policy - - Operational - - - Jul-18 America Consumer in DFS (69) Financial Protection North AE USA Arizona State Other Govt. Body Hybrid System - 3 3 3 3 - General innovation Product - - Operational 6 - 2 years Aug-18 America Regulators in DFS (70) North AE USA Kentucky State Other Govt. Body Hybrid System - 3 3 - 3 - InsureTech Focus Product - - Announced - - - Jun-19 America Govt. (71) North AE USA Nevada Other Govt. Body Hybrid System - 3 3 3 3 - General innovation Product - - Announced - - 2 years Jan-20 America Department of in DFS (72) Business and Industry North AE USA Utah Dept. of Other Govt. Body Hybrid System - 3 3 3 3 - General innovation Product - - Operational - - 2 years Jul-19 America Commerce in DFS (73) 68. https://www.bakermckenzie.com/-/media/files/insight/publications/2020/05/a_guide_to_regulatory_fintech_sandboxes_internationally_8734.pdf?la=en 72. http://business.nv.gov/Programs/Nevada_Sandbox_Program/ 69. https://debanked.com/2018/07/bcfp-launches-regulatory-sandbox-for-fintech-companies/ 73. (a) https://www.consumerfinancemonitor.com/2019/06/11/utahs-new-regulatory-sandbox/#:~:text=Regulatory%20sandboxes%20are%20laboratories%20 70. (a) https://www.consumerfinancemonitor.com/2019/06/11/utahs-new-regulatory-sandbox/#:~:text=Regulatory%20sandboxes%20are%20laboratories%20 for,sponsor%20of%20Utah’s%20legislation%2C%20Rep.; (b) https://news.bloomberglaw.com/banking-law/utah-becomes-second-state-to-launch-fintech- for,sponsor%20of%20Utah’s%20legislation%2C%20Rep.; https://www.forbes.com/sites/astanley/2018/03/23/arizona-becomes-first-u-s-state-to-launch- blockchain-sandbox regulatory-sandbox-for-fintech/#1db5e8621372 71. https://kentucky.gov/Pages/Activity-stream.aspx?n=PPC&prId=70 Note: The sandboxes included in this list have the following characteristics: (i) Sandbox is related to either only fintech innovations or fintech in combination with other sectors; and (ii) There is publicly available information/evidence that the Sandbox has been officially announced, launched or operational (defined as open for applications, or there is information on firms that already enrolled in sandbox). 62 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Glossary Anti-money laundering and countering the financing of terrorism (AML/ CFT): AML/CFT measures are defined by the Financial Action Task Force (FATF), the international standards setter in this area. The Basel Committee for Banking Supervision (BCBS) regularly issues guidance to facilitate banks’ compliance with their obligations in this area. Artificial intelligence (AI): AI is defined as an IT system that performs functions requiring human capabilities. AI can ask questions, discover and test hypotheses, and make decisions automatically, based on advanced analytics operating on extensive data sets. Machine learning (ML) is a subcategory of AI.151 Big data: Big data designates the large volume of data that can be generated, analyzed, and increasingly used by digital tools and information systems. This capability is driven by the increased availability of structured data, the ability to process unstructured data, increased data storage capabilities, and advances in computing power. Crowdfunding: Crowdfunding is the practice of funding a project or venture by raising monetary contributions from a large number of people. It is often performed today via internet-mediated registries that facilitate money collection for the borrower (lending) or issuer (equity).152 Distributed ledger technology (DLT): DLT, such as blockchain, is a means of recording information through a distributed ledger, i.e., a repeated digital copy of data at multiple locations. These technologies enable nodes in a network to securely propose, validate, and record state changes (or updates) to a synchronized ledger distributed across the network’s nodes.153 Fintech ecosystem: The fintech ecosystem is made up of consumers, financial institutions, fintech start-ups, investors, regulators, and educational institutions; its aims are to provide mutually beneficial cooperation among stakeholders to help deliver financial services at lower cost, higher speed, and better quality to more consumers.154 Fintech: Fintech offers advances in technology with the potential to transform financial services provision by spurring development of new business models, applications, processes, and products.155 Innovation facilitator: Innovation facilitators are public sector initiatives to engage with the fintech sector; they include regulatory sandboxes, innovation hubs, and innovation accelerators.156 GLOSSARY 63 Innovation hub/office: An innovation hub is an innovation facilitator set up by a supervisory agency to provide support, advice, or guidance to regulated or unregulated firms that are navigating the regulatory framework or seeking to identify supervisory policy or legal issues and concerns. Unregulated entities can engage with regulators to discuss fintech-related issues (share information and views, etc.) and seek clarification on conformity with regulatory frameworks and/or licensing requirements. Letter of no objection: No objection letters allow firms to operate in the open market, without a specific license, but with the implicit sanction of regulators. These letters can include restrictions and reporting requirements as deemed necessary by the regulator. Machine learning (ML): ML is a method of designing problem-solving rules that improve automatically through experience. Machine-learning algorithms give computers the ability to learn without specifying all the knowledge a computer would need to perform the desired task. ML also involves studying and building algorithms that can learn from and make predictions based on data and experience.157 New entrant: A new entrant is a prospective financial services provider that has not as yet been authorized by the regulator. No-enforcement-action letters: No-enforcement-action letters provide assurance to firms that the regulator will not take enforcement action against them, as long as they comply with the conditions specified in the letter. Peer-to-peer (P2P) lending: P2P is direct lending from savers to borrowers; traditionally the platform avoids intermediation by banks, but it also does not bear the risk of default.158 Reciprocal licensing: Reciprocal licensing allows firms with licenses in countries where the jurisdiction has a reciprocal license arrangement to easily passport into the country. Regtech: Regulatory technology, or regtech, involves new technologies to help regulated financial service providers streamline audit, compliance, risk management, and other back-office functions to enhance productivity and overcome regulatory challenges, such as the risks and costs related to regulatory reporting and compliance obligations. Regtech can also refer to firms that offer such applications. Regulatory accelerator or regtech lab: A regulatory accelerator is a partnership arrangement between fintech providers and central banks or supervisory agencies to accelerate growth or develop use cases, such as suptech or regtech, which may involve funding and/or authorities’ endorsement or approval for future use in central banking operations or in the conduct of supervisory tasks. Regulatory exemptions or waiver: Regulatory exemptions or waivers exempt a firm from requiring authorization to carry out a regulated activity or comply with a specific requirement. Regulatory sandbox: A regulatory sandbox is a controlled, time-bound, live testing environment, which may feature regulatory waivers at regulators’ discretion. The testing environment may involve limits or parameters within which firms must operate. Restricted or temporary license: Restricted or temporary licenses give firms licenses but set limits on the authorization, limiting, for example, the type of service that can be provided, the number of customers that can be served, or the time for which the license is valid. Sponsored licensing: Sponsored licensing allows firms to partner with existing license owners and to trade under that license. Suptech: Suptech is the use of innovative technology by supervisory agencies to support supervision. It is intended primarily to help supervisory agencies digitize reporting and regulatory processes, resulting in more efficient and proactive monitoring of risk and compliance at financial institutions.159 64 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES Endnotes 1. The IMF, WBG Bali Fintech Agenda (2018) defines fintech as “advances in technology that have the potential to transform financial services provision, spurring the development of new business models, applications, processes, and products.” 2. A regulatory sandbox is generally defined as “a controlled, time-bound, live testing environment, which may feature regulatory waivers at regulators’ discretion.” For the purposes of this paper, we have included all frameworks that the establishing authority refers to as a sandbox. 3. This signifies a small dedicated team (the hub) that can call upon satellite teams or personnel (the spokes) as needed. 4. The fear of contagion through contact has created incentives for increased use of digital financial services (DFS), which can be provided remotely and with no physical contact, often using mobile phones to access products and services. Fintech providers are a key part of this response. 5. WBG research; see Appendix 3 for full details. 6. For more on Innovation hubs please see Appendix 2; for a full list of all innovation facilitators by country, please see Annex 2 of Fintech Note No. 5, “How Regulators Respond to Fintech: Evaluating Different Approaches — Sandboxes and Beyond” (Appaya et al., World Bank, April 2020). 7. Ibid. This section draws on and summarizes elements from Fintech Note No. 5, which provides a more detailed analysis on types and definitions of sandboxes. 8. For a full list of all innovation facilitators by country, please see Annex 2 of Fintech Note No. 5, “How Regulators Respond to Fintech: Evaluating Different Approaches — Sandboxes and Beyond” (Appaya et al., World Bank, April 2020). 9. For the remainder of this document, unless stated otherwise, the term “sandbox” is used to refer to a regulatory sandbox. 10. https://www.hdruk.ac.uk/infrastructure/Sandbox/the-Sandbox-projects-2/. 11. See CFPB, Project Catalyst Report: Promoting Consumer Friendly Innovation (October 2016). 12. The sandbox has accepted 89 companies since its inception in 2016 and has just finished taking applications for its fifth cohort. ENDNOTES 65 13. Another way to differentiate sandboxes is according to whether the market or the regulator drives demand. However, for the purposes of this report, and to show the learnings from country case studies, we refer to the four primary sandbox types. 14. https://www.unsgsa.org/resources/news/regulatory-sandboxes-not-always-answer-regulating-inclusive/. 15. This paper only references the fintech-related sandboxes. Those created with wider purposes in mind, including those created by nonfinancial regulatory bodies, have not been included. 16. This paper is complemented by the CGAP paper, How to Build a Regulatory Sandbox: A Practical Guide for Policy Makers. 17. While similarities exist between a sandbox and a test-and-learn approach, the latter is a bespoke operating framework based on individual business models and thus can have issues of scalability. Moreover, test-and-learn frameworks are primarily innovator driven, while sandboxes are more likely to be regulator led. 18. “Sponsored licensing” here refers to an approach that allows firms to partner with existing license owners and to trade under that license. 19. “Reciprocal licensing” allows firms with licenses in countries with whom a jurisdiction has a reciprocal license arrangement to easily passport into the jurisdiction. 20. https://www.forbes.com/sites/elizabethrossiello/2019/02/14/forget-Sandboxes-why-governments-should- accept-licensing-from-other-jurisdictions/#2ed6fd0230e3. 21. https://fsdkenya.org/blog/groundbreaking-new-policy-and-regulatory-initiatives-may-spur-more-fintech- innovation-in-kenya/. 22. https://tracxn.com/explore/Fintech-Startups-in-Tallinn/. 23. https://www.ebrd.com/where-we-are/estonia/overview.html. 24. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3285938. 25. https://milkeninstitute.org/sites/default/files/reports-pdf/Fintech-in-the-Philippines-Update%20%281%29.pdf. 26. https://www.sfc.hk/web/EN/sfc-fintech-contact-point/sfc-regulatory-Sandbox.html. 27. https://www.ey.com/Publication/vwLUAssets/ey-thailand-fintech-landscape-report/$FILE/ey-thailand-fintech- landscape-report.pdf. 28. https://www.insead.edu/sites/default/files/assets/dept/globalindices/docs/GTCI-2017-report.pdf. 29. https://www.hkma.gov.hk/eng/, http://www.hongkong-fintech.hk/en/Sandboxes.html. 30. https://m.rbi.org.in/Scripts/PublicationReportDetails.aspx?UrlPage=&ID=892. 31. Inter-American Development Bank & Finnovista, Fintech Innovations that you may not know were from Latin America and the Caribbean (2017), available at: https://publications.iadb. org/handle/11319/8265. 32. Report: GFIN one year on. 33. Report: GFIN Cross-Border Testing Report. 34. https://www.bangkokpost.com/business/1882875/right-on-the-money. 35. See Cryptoslate, “Binance’s Bold Move to Usurp US Cryptocurrency Exchanges” (August 2019). 36. World Bank Open Data 37. Key considerations and guidance for policy makers to assess the feasibility of a sandbox can be found in our publication “How Regulators Respond to Fintech: Evaluating Different Approaches — Sandboxes and Beyond.” 38. The Sandbox Simulation exercise was conceptualized by CGAP and further developed by the World Bank Group. 39. Interview with ASIC. 66 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 40. Interview with Bank-al-Mahrib and WB project lead for Morocco. 41. https://aifc.kz/uploads/AFSA%20Annual%20Report%202018.pdf. 42. https://iclg.com/practice-areas/fintech-laws-and-regulations/mexico. 43. IADB Report: Regulatory Sandboxes in Latin America and the Caribbean. 44. SFC website. 45. https://www.superfinanciera.gov.co/jsp/Buscador/busqueda/BuscadorArchivos/idRecurso/1043144/f/0/c/0#. 46. WBG-CGAP Sandbox Survey. 47. Appaya, Jenik, “Running a Sandbox May Cost Over $1M, Survey Shows,” CGAP (August 2019), https://www. cgap.org/blog/running-Sandbox-may-cost-over-1m-survey-shows. 48. International Monetary Fund, Malaysia IMF Country Report (February 2020). 49. BIS. 50. GFS. 51. This box draws from the AFI report, “Creating Enabling Fintech Ecosystems: The Role of Regulators” (2020), https://www.afi-global.org/sites/default/files/publications/2020-01/AFI_Fintech_SR_AW_digital_0.pdf. 52. Interview with RBI. 53. https://www.npci.org.in/about-us-background. 54. WBG Research, see Appendix 3. 55. WBG Research, see Appendix 3. 56. https://fintechnews.my/17548/regtech-fintech-regulation-malaysia/ekyc-malaysia/. 57. https://www.bnm.gov.my/files/publication/fsps/en/2018/cp02.pdf. 58. https://fintechnews.my/17548/regtech-fintech-regulation-malaysia/ekyc-malaysia/. 59. https://fintechnews.my/20883/payments-remittance-malaysia/moneymwatch-graduate-bank-negara- malaysia-Sandbox/; https://dfsobservatory.com/sites/default/files/DFSO%20-%20The%20State%20of%20 Regulatory%20Sandboxes%20in%20Developing%20Countries%20-%20PUBLIC.pdf. 60. https://www.theedgemarkets.com/article/ekyc-banks-come-soon. 61. https://www.lb.lt/en/lbchain. 62. This box draws on https://www.nationthailand.com/Corporate/30331529; https://www.bot.or.th/English/ PressandSpeeches/Press/2020/Pages/n0963.aspx. 63. PromptPay is a system for money transfers that ties ID numbers or/and mobile phone numbers with bank accounts so that transferees can use any of their numbers instead of bank account numbers to transfer money. 64. https://www.nationthailand.com/Corporate/30331529. 65. https://www.bangkokpost.com/business/1882875/right-on-the-money. 66. Digital Financial Services, WBG (April 2020). 67. Digital Financial Services, WBG (April 2020). 68. Krungsri, “Krungsri-MUFG Team up to Drive Thai Financial Landscape Toward Global Arena; Launching Thai QR Code Payment in Japan for First Time.” News Release (October 16, 2018), https://www.krungsri.com/bank/ en/NewsandActivities/Krungsri-Banking-News/thai-qr-code-payment-injapan.html. 69. Pisei, Hin. “NBC Selects Three Banks for Pilot QR Code,” The Phenom Penh Post (June 18, 2019), https://www. phnompenhpost.com/business/nbc-selects-three-banks-pilot-qr-code-system. ENDNOTES 67 70. https://www.cgap.org/blog/do-regulatory-Sandboxes-impact-financial-inclusion-look-data. 71. https://www.bcb.gov.br/en/pressdetail/2308/nota. 72. WBG-CGAP Sandbox survey. 73. https://www.inform.kz/en/aifc-committee-launches-global-Sandbox-program_a3493933. 74. WBG – CGAP survey. 75. https://www.cma.or.ke/index.php?option=com_content&view=article&id=623:cma-press-release-fourth-firm- admitted-to-the-regulatory-Sandbox&catid=12:press-center&Itemid=207. 76. https://www.consumerfinance.gov/policy-compliance/notice-opportunities-comment/archive-closed/policy-no- action-letters-and-bcfp-product-Sandbox/. 77. https://files.consumerfinance.gov/f/documents/201709_cfpb_upstart-no-action-letter.pdf. 78. https://www.consumerfinance.gov/about-us/blog/update-credit-access-and-no-action-letter/. 79. https://www.dfsa.ae/MediaRelease/News/Robo-advisory-Firm-Sarwa,-First-to-Graduate-from-D. 80. https://www.menabytes.com/almakinah-coding-bootcamp-women-scholarships/. 81. GFS Survey. 82. https://www.finextra.com/finextra-downloads/newsdocs/unsgsa_report_2019_final-compressed.pdf. 83. https://esas-joint-committee.europa.eu/Publications/Reports/JC%202018%2074%20Joint%20Report%20 on%20Regulatory%20Sandboxes%20and%20Innovation%20Hubs.pdf. 84. https://esas-joint-committee.europa.eu/Publications/Reports/JC%202018%2074%20Joint%20Report%20 on%20Regulatory%20Sandboxes%20and%20Innovation%20Hubs.pdf. 85. https://esas-joint-committee.europa.eu/efif/efif-homepage. 86. Speech by Prof Dr Claudia Buch, Vice-President of the Deutsche Bundesbank, https://www.bundesbank.de/en/ press/speeches/digitalization-competition-and-financial-stability-799792#tar-5. 87. https://www.unsgsa.org/files/2915/5016/4448/Early_Lessons_on_Regulatory_Innovations_to_Enable_ Inclusive_Fintech.pdf. 88. http://www.cbj.gov.jo/EchoBusv3.0/SystemAssets/PDFs/2018/The%20National%20Financial%20 Inclusion%20Strategy%20A9.pdf. 89. https://www.unsgsa.org/files/2915/5016/4448/Early_Lessons_on_Regulatory_Innovations_to_Enable_ Inclusive_Fintech.pdf. 90. The World Bank actively supports authorities in developing and operationalizing national financial inclusion strategies, digital financial inclusion/fintech strategies, and fintech-driven initiatives like sandboxes. The WBG has assisted policy makers in establishing and linking fintech and financial inclusion initiatives in multiple countries, including China, Ethiopia, Indonesia, Jamaica, Jordan, Liberia, Mexico, Morocco, Nigeria, Pakistan, Peru, Saudi Arabia, and Sierra Leone, among many others. 91. In 2019, the WBG and the IMF conducted a global fintech survey (GFS) and collected responses from nearly 100 of our member countries on their progress in relation to the 12 elements of the BFA. 92. WBG-IMF, Fintech: The Experience so Far. 93. UNGSA (2019). 94. BSL, “Sierra Leone Fintech Challenge 2017” (2017), available at https://bit.ly/2PfiIFx; UNCDF, “The Sierra Leone Fintech Initiative in the Words of the Fintechs” (2018), available at https://youtu.be/hqSV-_qobOQ. 95. CBB, Regulatory Sandbox Framework (2017), available at https://bit.ly/2xvri8i. 68 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 96. Reported in the WBG-CGAP survey. 97. https://www.cgap.org/blog/do-regulatory-Sandboxes-impact-financial-inclusion-look-data. 98. https://www.unsgsa.org/resources/news/regulatory-Sandboxes-not-always-answer-regulating-inclusive/. 99. See “Payments Aspects of Financial Inclusion in the Fintech Era” (2020). 100. CCAF Survey; WBG-CGAP Survey. 101. Greg Chen, What Should We Realistically Expect from Regulatory Sandboxes?, CGAP (October 2017), https:// www.cgap.org/blog/what-should-we-realistically-expect-regulatory-Sandboxes; Buckley et al., Building Fintech Ecosystems. 102. ASIC website. 103. Interviews with fintech firms PrivyID and RiHa. 104. In 2017, RURA the utilities regulator (including telecommunications and information technology) set up a sandbox environment to facilitate testing of innovative business models. 105. Interview with BNR. 106. OpenAxel 2016 report. 107. FCA Regulatory Sandbox Lessons Learned report: https://www.fca.org.uk/publications/research/regulatory- Sandbox-lessons-learned-report. 108. An initiative between the Monetary Authority of Singapore, IFC, and the ASEAN Bankers Association for financial institutions and fintech firms to partner and collaborate via an integrated platform. 109. https://www.whub.io/startups/asean-financial-innovation-network. 110. https://www.mas.gov.sg/news/media-releases/2017/asean-financial-innovation-network-to-support-financial- services-innovation-and-inclusion. 111. As of July 2020. 112. https://www.finews.asia/finance/30413-asean-financial-innovation-network-manish-diwaan-afin-apix. 113. https://www.adgm.com/setting-up/reglab/overview. 114. https://www.centralbanking.com/fintech/4397616/Sandbox-initiative-central-bank-of-brazil. 115. 2018 GFS Survey. 116. GFS Survey. 117. GFS survey. 118. The Impact and Effectiveness of Innovate, April 2019. 119. https://www.nomurafoundation.or.jp/wordpress/wp-content/uploads/2017/09/NJACM2-1AU17-05_ SINGAPORE.pdf. 120. https://www.imf.org/en/Publications/CR/Issues/2019/07/15/Singapore-Technical-Note-Fintech-Implications- for-the-Regulation-and-Supervision-of-the-47113. 121. https://www.linkedin.com/company/policypal. 122. https://www.insurancebusinessmag.com/asia/news/breaking-news/insurtech-startup-policypal-is-first-graduate- of-mas-regulatory-Sandbox-77541.aspx. 123. https://www.asiainsurancereview.com/Magazine/ReadMagazineArticle?aid=42290. 124. https://www.asiainsurancereview.com/Magazine/ReadMagazineArticle?aid=42290 125. Central Bank of Jordan Sandbox Guidelines. ENDNOTES 69 126. FCA Lessons Learnt report. 127. https://www.fnlondon.com/articles/fcas-innovation-strategy-faces-criticism-20171107 128. https://www.fca.org.uk/firms/innovation/digital-Sandbox. 129. https://sec.gov.ng/report-of-the-fintech-roadmap-committee-of-the-nigerian-capital-market/. 130. Fintech Scoping in South Africa. 131. Department of Trade (DIT) and UK Treasury (HMT). 132. South Africa, https://www.ifwg.co.za/. 133. UNCDF https://www.uncdf.org/article/3635/four-fintechs-approved-to-enter-the-sierra-leone-Sandbox- programme. 134. Indonesia, https://fintech.id/id. 135. https://practiceguides.chambers.com/practice-guides/fintech-2020/south-korea. 136. FSC, Press Release, “Recent Progress on FSC’s Regulatory Sandbox” (November 5, 2019; in English). 137. Monetary Authority of Singapore, “MAS Launches Sandbox Express for Faster Market Testing of Innovative Financial Services” (August 2019), https://www.mas.gov.sg/news/media-releases/2019/mas-launches-sandbox- express-for-faster-market-testing-of-innovative-financial-services. 138. Remittances were also allowed until January 2020 due to the commencement of the Payment Services Act 2019. 139. The regulatory sandbox has four themes: ICT convergence, industry convergence, regional innovation, and financial innovation. The operation is coordinated by the Office for Government Policy Coordination and implemented by the Ministry of Science and ICT; the Ministry of Trade, Industry and Energy; the Ministry of SMEs and Start-ups; and the Financial Services Commission. 140. The Financial Services Commission is Korea’s government body responsible for financial policy and regulation across the financial sector. FSC has statutory authority over financial laws, regulations, and regulatory licenses. http://www.fsc.go.kr/eng/index.jsp. 141. FSC, Press Release, “Recent Progress on FSC’s Regulatory Sandbox” (November 5, 2019; in English). 142. FSC, Press Release, “Measures to Promote Fintech Scale-ups” (December 4, 2019; in English). 143. U.K. Financial Conduct Authority, “Regulatory Sandbox Lessons Learned Report” (October 2017), https:// www.fca.org.uk/publication/research-and-data/regulatory-sandbox-lessons-learned-report.pdf. 144. The WBG Report “How Regulators Respond to Fintech: Evaluating Different Approaches” provides guidance for policy makers on identifying and evaluating an appropriate regulatory approach to fintech. 145. http://documents1.worldbank.org/curated/en/579101587660589857/pdf/How-Regulators-Respond-To- FinTech-Evaluating-the-Different-Approaches-sandboxes-and-Beyond.pdf. 146. Ibid. 147. Definition from IMF-WBG Bali Fintech Agenda. See IMF & World Bank Group, The Bali FinTech Agenda (October 2018). 148. Excerpt from WBG, “Evaluating Regulatory Approaches to Fintech” (April 2020). 149. https://medium.com/@indonesiagodigital1/ojk-infinity-to-create-friendly-fintech-ecosystem-in-indonesia- 8f2afa7958b9. 150. World Bank and Cambridge Center for Alternative Finance, “Regulating Alternative Finance: Results from a Global Regulator Survey” (2019). 151. https://www.bis.org/bcbs/publ/d431.pdf. 70 GLOBAL EXPERIENCES FROM REGULATORY SANDBOXES 152. https://www.bis.org/bcbs/publ/d431.pdf. 153. https://www.bis.org/bcbs/publ/d431.pdf. 154. https://www.ey.com/Publication/vwLUAssets/EY-Fintech-ecosystem-playbook/$FILE/EY-Fintech-ecosystem- playbook.pdf. 155. Definition of from IMF-WBG Bali Fintech Agenda (available to policy makers). 156. Definition from FSB, https://www.fsb.org/wp-content/uploads/R270617.pdf. 157. https://www.bis.org/bcbs/publ/d431.pdf. 158. http://pubdocs.worldbank.org/en/148061509974150469/GFDR-2018-Chapter3.pdf. 159. https://www.bis.org/bcbs/publ/d431.pdf. ENDNOTES 71 CHAPTER TITLE 73