The Impact of Fintech on the Nexus Between Household Debt and Financial Crises: A Global Perspective

Four distinct datasets have been employed in our research. To capture our dependent variable (i.e., the occurrence of a financial crisis), we primarily rely on the Systemic Banking Crises Database. This comprehensive dataset has been sourced from the work of Nguyen et al. (2022), who extended the previously established Systemic Banking Crises Database developed by Laeven and Valencia (2020) to encompass data spanning from 1970 to 2019. The second data set was gathered from the Global Debt Database, containing information regarding household debt, non-financial firm debt, and government debt. This dataset provides a comprehensive record of the total gross debt within the private and public non-financial sectors across an extensive panel of 190 countries, with records dating from 1950 to 2021. To enhance data completeness, certain missing values were augmented based on the research by Mian et al. (2017). We acquired the fintech development index from the CrunchBase database, recognized as one of the largest and most reputable repositories of startup company information. This database has been increasingly utilized in recent scholarly publications(Bernstein and Korteweg, 2017). The final category of data about our control variables was drawn from the World Bank.

本研究共采用四组独立数据集。为获取因变量（即金融危机发生情况），本研究主要采用系统性银行危机数据库（Systemic Banking Crises Database）。该综合数据集源自Nguyen等人（2022）的研究成果，他们对Laeven与Valencia（2020）构建的经典系统性银行危机数据库进行了扩展，数据覆盖时段延伸至1970年至2019年。第二组数据集取自全球债务数据库（Global Debt Database），涵盖住户债务、非金融企业债务与政府债务相关信息。该数据集完整记录了1950年至2021年间，覆盖190个国家的大型跨国面板中私人与公共非金融部门的总毛负债规模。为提升数据完整性，本研究参照Mian等人（2017）的研究对部分缺失值进行了补全。本研究从CrunchBase数据库获取金融科技发展指数，该平台作为全球规模最大、声誉最卓著的初创企业信息仓储库之一，近年来已被诸多学术研究（Bernstein与Korteweg，2017）广泛采用。最后一组控制变量数据取自世界银行（World Bank）。