Розробка індексу кредитних стандартів в Україні за допомогою впорядкованих логістичних моделей

Abstract

This study expands the understanding of the dynamics of credit standards in Ukraine. This publication makes an important contribution to the analysis of the supply of bank lending. A panel ordered logit model was used to build the Credit Standards Index (CRI) for different market segments. The study uses survey data on bank lending conditions at the bank level, which makes it unique, as this data is usually used at the aggregated level. The accuracy of the model is 65% for corporate loans, 84.4% for mortgages, and 77.6% for consumer loans. It is found that at the system level, banks in Ukraine tightened corporate lending standards during the crises in 2014–2015 and 2022, while standards for mortgages and consumer loans were tightened only in 2022. Based on the results of the simulation, it is determined that banks weakened lending standards only for consumer loans in 2021. It is found that the model weakly classifies banks that have softened or strengthened corporate lending standards, but for mortgages and consumer loans, the results are noticeably better. The increase in interest rates in the interbank market affects the tightening of credit standards for corporate loans. According to the results of the study, real GDP growth contributes to the weakening of standards for corporate and mortgage loans. Competition between banks may weaken standards for all market segments, but strengthen them for mortgages. Expectations of the prospects for the real estate market affect both the strengthening and weakening of mortgage lending standards. In the field of consumer lending, borrowers’ solvency expectations determine the change in credit standards in both directions. The results are important from the standpoint of using the index to study the determinants of lending in Ukraine, as the index approximates the supply of banks, which is an unobservable variable.