Predicting Non-performing Loans Through Financial Ratios of Small and Medium Entities in Lebanon
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The purpose of the study is to examine the ability of financial ratios in the prediction of the financial state of the Small and Medium Entity in Lebanon. Specifically, the financial state of the company is determined by the classification of the loans of the SME, which can be performing or non-performing.
An empirical study was performed using a data analysis conducted on the financial statements of 222 SMEs in Lebanon for the years 2011 and 2012, of which 187 are subject to performing loan and 35 are subject to non- performing loans.
The Altman Z-scores were calculated, the independent sample t-test was performed, and models were developed using the logistic regression.
Empirical evidence from this study showed first, that the Altman Z-scores were able to predict the solvent state of SMEs having performing loans, but were unable to predict the bankrupt state of the SMEs having non-performing loans. Second, the independent sample t-test revealed five financial ratios that are significantly different between SMEs having performing loans and non-performing loans during the years under study, which are the following: liquid assets/current assets, total liabilities/total assets, total equity/total assets (disregarded subsequently, for being complimentary to the previous ratio), sales/total liabilities, and working capital/total assets. Third, a logistic regression model was developed for each year under study and accuracy results were deducted and displayed in a table showing the percentage of accurately classified companies (solvent and bankrupt), in addition to the type I and type II errors.
This study recommended not applying the Altman Z-score models on SMEs in Lebanon due to the low accuracy results registered. Moreover, some financial ratios with predictive ability are worth focusing on during the analysis of the financial health of a company, which are liabilities/assets, liquid assets/current assets, sales/liabilities and working capital/assets.
Finally, since the logistic regression models developed in this study using only quantitative variables and a sample of 222 SMEs did not result in high accuracy levels, further research conducted on a larger sample using qualitative variables such as years of experience of the SME in the market, geographical location, history of repayment in the bank, overall macro-economic indicators… could add to the predictive ability of the logistic regression model in Lebanon.
Description
Keywords
Citation
Nasr, M. R. (2015). Predicting Non-performing Loans Through Financial Ratios of Small and Medium Entities in Lebanon (MBA thesis, Haigazian University)