Macroeconomic and bank-specific determinants of non-performing loans: A panel analysis on Malaysian commercial banks

Abstract

This study utilized dynamic and static panel data analysis to investigate the macroeconomic and bank-specific determinants of non-performing loans (NPLs) of commercial banks in Malaysia over the period of 2002-2013. The first objective of this study was to investigate the relationship between macroeconomics factor and NPLs. The study finds higher levels in GDP growth, lending interest rate, unemployment, sovereign debt, money supply M1 growth, and stock market index growth, increase NPLs in banks. Yet a rise in inflation, foreign direct investment, banking concentration and stock market turnover, reduce NPLs. The second objective was to investigate the relationship between bank-specific factors and NPLs. With regard to the impact of bank capital, the finding of the study support the concave capital in that NPLs continue to increases in capital until NPLs reach a threshold (Moral Hazard Effect), after which more capital build-up decrease the NPLs (Disciplinary or regulatory effect). With regards to the impact of charter value, the findings of the study support the disciplinary charter value that higher charter value reduces NPLs. With regard to the impact of managerial performance, the finding of the study support that better performance (Higher return on Equity-ROE) reduce NPLs and in efficiency increases NPLs. Yet, higher operating leverage increases NPLs. With regards to the impact of diversification, banks respond to higher NPLs by increasing non-interest income. The size of the bank, and off balance sheet activities. There is evidence supporting Too Big Too Fail “TBTF” that larger banks responds to Higher NPLs by increasing their financial leverage, i.e leverage conditional on size leads to more NPLs. However NPLs constrained up to a certain size level inferred by assessment of leverage marginal effects. For robustness, the study investigated the determinants of LLP (Loans Loss Provisions) as an alternative measure to NPLs in addition to utilizing VAR model on monthly data aggregates to test and confirm the concave capital. Moreover, the study provides a comparison between foreign and local banks in term of their NPLs and LLP determinants. The finding of this study bear several policy implications for regulator and policy makers in term of the macro-prudential and micro-prudential regulations in order to attain financial stability as well as for banks management.