The determinants of the unlikely-to-pay and the flows towards performing and bad loans

Using a novel panel data-set, covering the period 2010-2016, the study detects: a) determinants of the unlikely-to-pay loan; b) their out-flows to both other categories of non-performing- and performing loans. Results show that the determinants tested in the literature on non-performing loans are in part confirmed for unlikely-to-pay and in part rejected. In particular, results differ depending on the variable chosen as a measure of the credit portfolio quality. They vary depending on whether we adopt the stock- or the flow-measure, which reveals the importance of the proxy used in literature, regulations, and controls specifically related to unlikely-to-pay. Considering the out-flows from unlikely-to-pay, findings suggest having a specific office/unit to manage non-performing loans increases flows to performing loans and reduces flows to other categories of non-performing loans. Referring to the coverage of unlikely-to-pay, banks with a higher ratio show higher flows to other categories of non-performing loans. The first part of our study, about the determinants of UTP ratio and flows of UTP from performing loans, is useful for banks in order to prevent new UTP, the second part of the analysis, about the flows that reduce UTP, is relevant for encouraging the transition from UTP to performing loans and for preventing their worsening to bad loans.