Step towards a general bankruptcy prediction model for listed non-financial firms: The use of accounting, market, and macroeconomic variables

The purpose of this research is to move a step forward toward developing a bankruptcy model that provides globally relevant predictors for firms listed on stock exchanges across different economies. This study has developed a general bankruptcy model including accounting, market and macroeconomic variables that are universal and consistent for non-financial firms across the globe. Data for the study was obtained from higher-income upper middle income and lower-middle-income (a total of 18) countries for the period 2001–2017. In the first stage, the predictor variables that affect firms’ bankruptcy are identified. In the second stage, the effect of predictors on the dependent variable is estimated at a time one year (t-1) and two years (t-2) prior to bankruptcy. In this step, logistic regression is used to check the impact of accounting, market and macroeconomic variables on corporate bankruptcy. The study identified that accounting variables: liquidity ratio, profitability ratio, asset turnover ratio and financial leverage ratio are consistent across all the models. In addition, market variables: change in stock return, past excess return and market to book value impact the bankruptcy of firms. The study also found that macroeconomic variables including change in Gross domestic product, change in retail price index, change in real effective exchange rate, change in real interest rate and change in stock market capitalization are significant predictors of bankruptcy. Countrywide bankruptcy prediction models are plentiful and have been used widely in different countries. This study provides predictors of bankruptcy that are consistent during different times period and across different countries.