DETERMINANTS OF NON-PERFORMING LOANS: EVIDENCE FROM INDIA

Tools

Makhija, Rishima (2017) DETERMINANTS OF NON-PERFORMING LOANS: EVIDENCE FROM INDIA. [Dissertation (University of Nottingham only)]

[thumbnail of DISSERTATION.pdf] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (1MB)

Abstract

We investigated macro-economic and bank-specific determinants of credit risk. While numerous studies on credit risk determinants have been undertaken, this one tested for the particular rise in NPLs since 2009 in Indian banking system. We used a dynamic panel model approach over the period of 2005-2016 to examine the determinants of credit risks in all commercial banks of India, including banks from different ownership. In this study, 14 factors representing potential determinants of the credit risk were incorporated to test for which of these variables were major contributor to burgeoning credit risk in Indian markets. One these variables, efficiency, was calculated using single-step stochastic frontier approach following Battese and Coelli (1995) model.

Our study recapitulated that bank-specific variables tend to have more influence on the credit risk than macroeconomic determinants for the observed time period. We found strong evidence for procyclicity in our data followed by levels of efficiency and performance indicator of a bank. In contrast to most studies, we found a positive correlation between diversification and problem loans. Beside from bank-specific variables, we also found empirical evidence for impact of macroeconomic environment on increasing problem loans. Public debt and external debt seemed maximum influence on NPLs followed by money supply and problem loans. However, in contrast to the existing literature, we found a negative relation between unemployment rate and lending rate with NPLs. This could possibly pertain to the fact of our observed period of study.

Item Type: Dissertation (University of Nottingham only)
Keywords: Non-Performing loans, dynamic panel data models, stochastic frontier approach, procyclicity, bad management
Depositing User: Makhija, Rishima
Date Deposited: 10 Apr 2018 13:35
Last Modified: 17 Apr 2018 15:18
URI: https://eprints.nottingham.ac.uk/id/eprint/45647

Actions (Archive Staff Only)

Edit View Edit View