Abstract
This paper uses data from a panel of more than 400 Italian banks for the period 2001–2015 to examine the main determinants of loan loss provision (LLP), which are classified as either discretionary (income smoothing, capital management, signalling) or non-discretionary (related to the business cycle). The possible effects of the double-dip recession of 2008–9 and 2011–15 are also examined. The results suggest that LLP in Italian banks is countercyclical, with non-discretionary components and macroeconomic shocks playing a significant role. Moreover, LLP is less cyclical in the case of local banks, since their loans are more collateralised and their behaviour is more strongly affected by supervisory activity.
Highlights
During the last decade, the European economy has experienced one of the deepest recessions of the post–war period
The banking sector was significantly affected by the crisis: bad loans piled up, both reducing revenues and increasing loan loss provisions (LLP), which led to further revenue losses
The Italian economy was severely affected by a double-dip recession that was deeper and longer than those experienced by other Eurozone countries and had a bigger impact on non-performing loans and on LLPs of Italian banks
Summary
The European economy has experienced one of the deepest recessions of the post–war period. The income smoothing hypothesis implies that banks should decrease (increase) LLP when earnings are expected to be low (high) This hypothesis is tested using the ratio of earnings before interest, taxes and LLP to total assets (ISi,t), as in Anandarajan et al (2006), Bouvatier and Lepetit (2008), Soerdamono et al (2012) and Bouvatier et al (2014). Banks ca use LLP to signal their financial strength (Ahmed et al, 1999; Kanagaretnam et al, 2005) To test this hypothesis, we use the one-year-ahead change of earnings before taxes (SIGNi,t = ISi,t+1 − ISi,t/ISi,t), an adjusted version of the weighted one considered by Anandarajan et al (2006), Bouvatier and Lepetit (2008), Soedarmono et al (2012) and Bouvatier et al (2014). The sign of the coefficient on earnings could be either positive or negative: if banks use provisions to smooth earnings, the expected sign is positive; a negative sign is possible owing to pro-cyclical effects
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
