Application of the Z-Score and extended Springate H-Score models in predicting bankruptcy risk of listed real estate firms in Viet Nam

NGUYEN HOANG CHUNG

Ho Chi Minh University of Banking; Email: [email protected]

TRAN THI MY TRINH

Thu Dau Mot University; Email: [email protected]

Abstract

This study examines the predictive behaviour and underlying determinants of corporate distress scores for 55 real estate firms listed on the Ho Chi Minh and Hanoi stock exchanges over the period 2018-2025, an interval that captures both the COVID-19 demand shock and the 2022-2023 corporate bond market freeze. Drawing on an unbalanced panel of 440 firm-year observations, the analysis follows a two-layer framework: the four-variable Altman Z″-Score and an eight-variable extended Springate H-Score are first computed and used to classify firms into safe, grey, and distress zones, after which panel regressions identify the financial fundamentals that drive each score. Estimator selection is governed by formal diagnostics-the Breusch-Pagan Lagrange multiplier, Hausman, modified Wald, and Wooldridge testsleading to feasible generalized least squares for the Z-Score model and a cluster-robust fixed-effects estimator for the extended H-Score. Leverage and operating profitability are the dominant determinants in both models, whereas the conventional current ratio is statistically irrelevant, signalling a sector-specific liquidity trap rooted in illiquid inventory. The extended H-Score explains substantially more variation (within-R² ≈ 0.96) than the Z″-Score. Distress incidence peaked sharply during 2022-2023.

Keywords: Bankruptcy prediction, Z-Score, Springate H-Score, financial distress, panel data, Viet Nam real estate

(The article was published in the Vietnam Economic & Financial Review, Vol. 2, Issue 4, August 2026).

For more details, please click here