Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/24987
Title: Impact of bank capital on non-performing loans: New evidence of concave capital from dynamic panel-data and time series analysis in Malaysia
Authors: Yaman Hajja
UniKL BiS
Keywords: bank capital
dynamic panel-data
macroeconomic determinants
Malaysia banking system
non-performing loans
vector autoregression
Issue Date: 2020
Publisher: John Wiley and Sons Ltd
Citation: Hajja, Yaman. (2020). Impact of bank capital on non-performing loans: New evidence of concave capital from dynamic panel-data and time series analysis in Malaysia. International Journal of Finance & Economics. 1. https://doi.org/10.1002/ijfe.2305
Abstract: Amid the steep expansion in Malaysia household debt, we investigate the role of bank capital in disciplining non-performing loans (NPLs) after controlling for the macroeconomic environment. Utilizing generalized method of moments (GMM) on a dynamic panel-data of 19 commercial banks and stress testing of NPLs using vector autoregression (VAR) on aggregated-monthly time series of the banking system, we provide new evidence that the capital (in the past and in the future) is a concave function of NPLs implying that increasing the capital will initially increase the NPLs until NPLs reach a maximum threshold (under the moral hazard effect), after which more capital build ups will succeed in decreasing NPLs (under the disciplinary or regulatory effect). We also find that higher levels in GDP growth and lending interest rate and are associated with more NPLs, while higher inflation is associated with less NPLs. Monetary expansion, that is, higher money supply growth raises NPLs in banks, while competition between banks and higher liquidity of the stock market are NPLs reducing.
Description: This article is index by Scopus
URI: http://hdl.handle.net/123456789/24987
ISSN: 10769307
Appears in Collections:Journal Articles



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.