Predicting Corporate Financial Distress in Four Southeast Asian Countries

Abstract

Quantifying the risk of financial distress, which can lead to risk of default and the emergence of zombie firms, is of considerable interest to academics, investors, and regulators. This study examines and predicts financial distress likelihood among non-bank publicly listed firms in four Southeast Asian countries—Indonesia, Malaysia, the Philippines, and Thailand—over the period 2005–19 using panel logit models. We evaluate predictive performance through out-of-sample forecasts. Estimates from the 2005–19 sample are used to forecast financial distress in 2020, after which the estimation window is updated annually through 2022 to generate one-year-ahead forecasts. This recursive forecasting framework provides a built-in test of the robustness and stability of model estimates and predictions. To improve forecasting accuracy, we use Least Absolute Shrinkage and Selection Operator (LASSO) techniques. The results show that the panel logit models perform reasonably well in identifying distressed firms, while LASSO delivers the expected improvement in predictive accuracy. These findings provide useful insights for investors, financial institutions, and policymakers seeking to identify vulnerable firms and mitigate financial risks.