Work

Text-enhanced empirical analysis of digital inclusive finance on rural poverty reduction in China

Year: 2026

Type: article

Abstract: Based on information asymmetry and transaction cost theories, this study developed a text-enhanced empirical framework to evaluate the poverty reduction effects of Digital Inclusive Finance (DIF). Taking Lankao County as a representative case, township-level structured data were combined with text-derived indicators. The results show that the per capita disposable income of rural residents increases by 56.78 yuan and the incidence of poverty decreases by 0.043% for each unit increase in the DIF index. Textual signals provided significant incremental explanatory power: for each unit increase in the frequency of policy attention, income increased by 32.45 yuan. Mechanism analysis showed that DIF reduced poverty through two key pathways: “accurate identification” (mediated by credit availability) and “dynamic monitoring” (mediated by risk resilience), accounting for 34.9% and 27.9% of the total effect, respectively. Heterogeneity analysis further showed that the effects of DIF were more pronounced in remote areas (with a coefficient 0.021 higher) and among low-income groups (with a coefficient 0.036 higher), highlighting its inherent inclusiveness. These findings provide methodological insights for the application of text enhancement analysis in the field of DIF, and provide practical significance for improving the accuracy and effectiveness of poverty reduction policies.

Source: Discover Artificial Intelligence

DOI: 10.1007/s44163-026-02557-6

Authors: Shijie Wu, Dejing Lin

Institutions: Lingnan University, Yango University

Language: English

Topic: Microfinance and Financial Inclusion

Subfield: Economics and Econometrics

Field: Economics, Econometrics and Finance

Domain: Social Sciences

SDGs: No poverty, Reduced inequalities, Decent work and economic growth

Open Access status: gold

Estimated APC paid: $1,590