Authors: Paraskevi Boufounou, Evdoxia Papadopoulou

Title: Comparative Predictability of Corporate Failure Models: Evidence from the Greek Market during Crisis

Abstract

The accurate prediction of corporate bankruptcy is of a great concern to investors, creditors, financial institutions and governments. Corporate failure prediction, otherwise known as financial distress prediction, reserves a key role in economic decision-making. Especially, it becomes of prior importance to countries facing prolonged economic crisis, like Greece facing a severe financial crisis since 2009.

This study uses two classical statistical methods, the logistic regression (logit) and the probability regression (probit), to examine the usefulness of financial ratios in predicting the financial distress of companies. This research conducted analyses both listed and non-listed companies in the Greek Stock Exchange Market for the time period of 2005 to 2018, taking under consideration for each company, the last five years of operations.

The findings suggest that the popular ratios of profitability and liquidity tend to give warning signs for an upcoming failure most of the time, but also maybe somewhat deceiving. This tends to happen because high ratios by themselves do not necessarily imply that the company has sufficient sources to pay its liabilities. It is evidenced that the bigger the sample and the more specific the time period, the better the warning signal and the foresight of financial deterioration and the detection of financially distressed companies.