The complex relationship between credit and liquidity risks: a linear and non-linear analysis for the banking sector

This article explores the reciprocal link between credit risk and liquidity risk in Tunisia. To the best of our knowledge, no study has examined the linear and non-linear relationships between credit risk (CR) and liquidity risk (LR) taken in both directions. We utilized a sample of Tunisian banks from 2000 to 2018 to investigate this link in both causative directions and within a linear and non-linear framework. Unlike previous investigations, we used two empirical approaches. The linear link was assessed using the Seemingly Unrelated Regression (SUR) model, whilst the non-linear correlation was investigated using the Panel Smooth Transition Regression (PSTR) model.The results of the linear analysis show that credit and liquidity risks are positively related in both directions. The non-linear analysis proves that there is a threshold impact in both connections. More specifically, we discovered that the NPLs ratio, which measures credit risk, is 9.87%, while the LTD ratio measures liquidity risk, which is 102%. Below this threshold, there is a negative and significant relationship; beyond these thresholds, the effect is positive but only significant for the influence of credit risk on liquidity risk.