Resumen de Identifying the Determinants of Indebtedness in Rural Telangana: A Quantile Regression Approach
The high incidence of indebtedness and the increasing number of farmer suicides in India is a cause of concern. The 77th round of the National Sample Survey (NSS) 2019 by the Government of India shows the severity of the indebtedness in rural India, particularly in some of the southern states, including Telangana and Andhra Pradesh. The purpose of this paper is to understand the nature, sources, and purposes of rural credit in Telangana and also identify the main determinants of indebtedness in the state. Design/methodology/approach: The study uses household-level data provided in the 77th round of the NSS Survey 2019, to get information on the socio-economic profile, nature, sources, and purposes of loans of sample households. As the OLS method may not be sufficient to get a comprehensive relationship between indebtedness and its determinants, we use quantile regression to identify the determinants of indebtedness using the same data set. Findings: The results indicate that 85% of households are indebted, with an average outstanding loan of Rs. 1,10,500. Approximately 71% of the households are engaged in agriculture, and informal sources account for 54% of the total borrowing, with 53% of the loans being utilized for agricultural purposes. Quantile regression models demonstrate that total earnings, consumption expenditure, and education strongly impact debt at lower quantiles. Bank loans and debt have a negative relationship at higher quantiles, while the loans from moneylenders. are significant at the 10th quantile. The quantile graphs reveal a non-linear relationship between the outstanding loan and its determinants. Originality: The study uses the quantile regression method, which allows for analyzing determinants of debt across different quantiles rather than only at the mean distribution (OLS). It shows that households at the lower end of the debt distribution are more likely to be affected and benefit more from policy interventions.
