Does the threshold matter? The impact of the monitoring activity on non-performing loans: Evidence from the Italian banking system. Issue 2 (11th February 2019)

• Record Type: Journal Article
• Authors: Piatti, Domenico; Cincinelli, Peter
• Is Part Of: Managerial finance. Volume 45:Issue 2(2019)
• Page Start: 190
• Page End: 221
• Publication Date: 2019-02-11
• Subjects: Banking industry -- Data Envelopment Analysis -- Italian banks -- Non-performing loans -- Credit monitoring -- Threshold panel; G21 -- G30; Corporations -- Finance -- Periodicals; Business enterprises -- Finance -- Periodicals; 658.1505
• DOI: 10.1108/MF-02-2018-0077
• Languages: English
• ISSNs: 0307-4358

Abstract

Abstract : Purpose: The purpose of this paper is to investigate whether the quality of the credit process is sensitive to reaching a particular threshold level of non-performing loans (NPLs) and, more importantly, whether higher NPLs ratios could make the monitoring activity ineffective. Design/methodology/approach: The empirical design is composed of two steps: in the first step, the authors introduce a monitoring performance indicator (MPI) of the credit process by combining the non-parametric technique Data Envelopment Analysis with some financial ratios adopted as input and output variables. As second step, the authors apply a threshold panel regression model to a sample of 298 Italian banks, over the time period 2006–2014, and the authors investigate whether the quality of the credit process is sensitive to reaching a particular threshold level of NPLs. Findings: This paper finds that, first, when the NPLs ratio remains below the threshold value estimated endogenously, an increase in the quality of monitoring has a positive impact on the NPLs ratio. Second, if the NPLs ratio exceeds the estimated threshold, the relationship between the NPLs ratio and quality of monitoring assumes a positive value and is statistically significant. Research limitations/implications: Due to the lack of data, the investigation of NPLs in the Italian industry across loan types combined with the monitoring effort by banks management was not possible. The authors plan to investigate this topic in future studies. Practical implications: The identification of the threshold has a double operational valence. The first regards the Supervisory Authority, the threshold approach could be used as an early warning in order to introduce active control strategies based on the additional information requested or by on-site inspections. The second implication is highlighted in relation to the individual banks, the monitoring of credit control quality, if objective and comparable, could facilitate the emergence of best practices among banks. Social implications: A high NPLs ratio requires greater loan provisions, which reduces capital resources available for lending, and dents bank profitability. Moreover, structural weaknesses on banks' balance sheets still persist particularly in relation to the inadequate internal governance structures. This means that bank management must able to recognise in advance early warning signals by providing prudent measurement together with an in-depth valuation of loans portfolio. Originality/value: The originality of the paper is twofold: the authors introduce a new proxy of credit monitoring, called MPI; the authors provide an empirical proof of the Diamond's (1991) economic intuition: for riskier borrowers, the monitoring activity is an inappropriate instrument depending on the bad reputational quality of borrowers.