The rise of digital finance: Financial inclusion or debt trap

1. Introduction

• Digital finance can reach people excluded from traditional finance by reducing information asymmetry and transaction costs, but it can expose less financially literate users to complex products with risks not yet known to the market.The introduction also identifies rising household debt as a growing concern.Source: §1 · p. 1
• The study adds empirical evidence from China on digital finance’s negative consequences, an area less studied than financial inclusion and previously addressed mainly through qualitative work.It aims to provide a fuller account of both positive and negative household outcomes and the channel through which distress may arise.Source: §1 · pp. 1–2
• The hypotheses trace a proposed pathway from widespread digital-finance use to easier credit access, higher household consumption, and greater debt-trap likelihood.The paper links easier credit access to consumption through changes in marginal propensity to consume.Source: §1 · p. 2
• Broadened access to digital finance increases credit-market participation and consumption through changes in marginal propensity to consume, while also raising financial-distress risk.The introduction frames this as a pathway from easier credit access to consumption and then debt-trap risk.Source: §1 · p. 2
• Using household-level data, the study examines China, where digital finance fills gaps in financial service but financial literacy is lower and legal gaps and inconsistencies remain.The paper presents documenting both positive and negative consequences, and explaining the distress channel, as contributions.Source: §1 · p. 2

2.1.1. Data

• The analysis uses CHFS household data from 2013, 2015, 2017, and 2019, paired with city-level DFI indices for 2013, 2015, and 2017.Source: §2.1.1 · p. 2

2.1.2. Variables

• Digital finance is measured by the total DFI index and subindices, while unpaid loans proxy credit-market access and self-reported distress proxies debt trap.The models also use household demographics and economic characteristics; continuous variables are winsorized at the 1% and 99% levels.Source: §2.1.2 · pp. 2–3, Table 1 · p. 3

2.2. Models

• The models test whether DFI relates to outstanding-loan status and consumption, examine debt and consumption, and test its relationship with financial distress.The first model specifies the natural logarithm of the DFI index or a subindex alongside control variables and confounding variables; debt trap is defined as a financial-distress indicator.Source: §2.2 · pp. 3–4

3.1. Digital finance and credit market participation

• A 1% increase in the DFI index is related to a 2.93% increase in the likelihood of households getting a loan and a 10.21% increase in the mean value of the debt dummy variable.Unreported tests find a 17,959 RMB increase in debt level for a 1% increase in the DFI index.Source: §3.1 · p. 4

3.2. Credit market participation and household consumption behavior

• The interaction-term coefficient is insignificant, indicating that debt does not significantly affect the marginal propensity to consume.Source: §3.2 · p. 4
• A 1% increase in the DFI index is related to a 27.29% increase in household consumption and a 4.30% increase in the marginal propensity to consume.For low, medium, and high Total DFI index values, a 1% income increase is associated with consumption increases of 3.35%, 4.78%, and 5.69%, respectively.Source: §3.2 · p. 4, Table 3 · p. 5

3.3. Digital finance and household financial distress

• Table 4 reports that a 1% increase in the DFI index is related to a 2.90% increase in households’ likelihood of falling into a debt trap and a 69% increase in the debt-trap variable’s mean.Source: §3.3 · p. 5

3.4. Endogeneity

• Using city-level smartphone ownership as an instrument, the positive significant coefficients of the DFI indices are consistent with the Section 3.3 results.The authors interpret this as evidence that their findings are not driven by endogeneity bias.Source: §3.4 · p. 5

4. Conclusion

• Wider digital finance use increases credit-market participation, raises households’ marginal propensity to consume out of liquidity and consumption, but also heightens financial-distress risk through borrowing.Table 4 reports tests analyzing the risk of falling into a debt trap; the supplied caption does not provide the table’s numerical results.Source: §4 · p. 5, Table 4 · p. 6
• Because Chinese households’ financial literacy is low, policymakers should provide appropriate digital financial literacy.Source: §4 · p. 5
• Because self-control cannot be educated and relates to credit non-payment and excessive debt burdens, policymakers should restrict credit by controlling loan purposes.Source: §4 · p. 5
• Policymakers should improve customer protections and ensure market transparency, competition, and fair pricing to reduce digital-finance risks (Ozili, 2020).Small lending platforms may inadequately manage risks, which could spill over from digital finance to traditional finance.Source: §4 · p. 6

• Pengpeng Yue handled conceptualization, methodology, software, formal analysis, data curation, and original-draft writing; Aslihan Gizem Korkmaz handled conceptualization, methodology, original-draft writing, and review and editing.Source: CRediT authorship contribution statement · p. 6
• Zhichao Yin provided supervision and resources, while Haigang Zhou provided supervision and methodology.Source: CRediT authorship contribution statement · p. 6

Funding

• The work was financially supported by the National Natural Science Foundation of China (72103010).Source: Funding · p. 6
• The robustness test uses smartphone ownership as an instrumental variable; *, **, and *** denote significance at 10%, 5%, and 1%, respectively.Source: Table 5 · p. 7, Funding · p. 7