Corporate Bankruptcy Prediction: Bridging the Gap Between SME and Large Firm Models

Abstract

Research on corporate bankruptcy prediction has garnered renewed interest due to economic crises and regulatory changes. Most studies focus on large enterprises, leaving a gap in understanding bankruptcy prediction in small and medium-sized enterprises (SMEs). This study carries out a systematic literature review to examine the evolution of this topic, focusing on SMEs. Using a structured methodology based on PRISMA, we analysed 541 academic papers, categorising them into two groups: (i) SMEs and (ii) non-SMEs. Our findings reveal key distinctions between the two groups, particularly regarding the definition of bankruptcy, financial and non-financial predictive factors, and the types of models applied. While statistical models, such as logistic regression and discriminant analysis, remain dominant in SME-focused research, artificial intelligence-based techniques are gaining traction. The study also identifies a lack of comparative studies assessing model effectiveness for SMEs across different economic contexts. Based on these insights, we propose a framework to enhance future research in corporate bankruptcy prediction, emphasising the need for models that integrate macroeconomic variables, governance factors, and alternative risk assessment techniques tailored to SMEs. Our findings contribute to bridging the gap between theory and empirical research, offering practical implications for financial institutions, auditors, policymakers, and SME managers in mitigating bankruptcy risks.