Macroeconomic and Structural Determinants of Non-Performing Loans in ASEAN+4

Abstract

Background: Non-performing loans (NPLs) are a critical indicator of banking sector fragility, reflecting deteriorating credit quality and posing systemic risks to financial stability. Despite extensive research in European and advanced economy contexts, empirical evidence on NPL determinants across the ASEAN+4 region remains limited, notwithstanding the region's deep economic interdependencies and recurring exposure to major global shocks.
Purpose: This study identifies and empirically analyses the macroeconomic and structural-institutional factors influencing NPL dynamics across ASEAN+4 economies over the period 2013 to 2023.
Design/Methodology/Approach: Static panel data analysis is employed across 10 economies, namely Indonesia, Malaysia, Thailand, Vietnam, the Philippines, Cambodia, China, Japan, South Korea, and India. Seven explanatory variables are examined: GDP growth, inflation, lending interest rate, unemployment rate, exchange rate, the Civil Justice Index from the World Justice Project, and a COVID-19 dummy variable. Model selection among Pooled Least Squares, Fixed Effect Model, and Random Effect Model was conducted via the Chow, Hausman, and Lagrange Multiplier tests, with the Random Effect Model selected as optimal. Data were sourced from CEIC Data, the World Bank World Development Indicators, and the World Justice Project, and processed using EViews 9.
Findings/Result: Unemployment rate and lending interest rate exert positive and statistically significant effects on NPL ratios at the 1% significance level, with coefficients of 0.524 and 0.288 respectively, confirming that deteriorating labour market conditions and higher borrowing costs erode borrowers' debt-servicing capacity. Inflation exhibits a negative and significant effect at the 5% level (coefficient: 0.159), consistent with the debt-deflation channel whereby moderate inflation reduces the real burden of outstanding obligations. GDP growth, exchange rate, the Civil Justice Index, and the COVID-19 dummy do not demonstrate statistically significant effects within this specification. The model is statistically significant overall (F-statistic p-value: 0.0005; R-squared: 0.220).
Conclusion: Labour market conditions and lending interest rates are the primary macroeconomic drivers of NPL accumulation across ASEAN+4 economies. Policymakers should prioritise employment-supportive measures and prudent interest rate management as pre-emptive credit risk containment tools. The non-significance of civil justice and pandemic variables suggests their effects may be mediated by country-level heterogeneity or require longer horizons to materialise.
Originality/Value: This study contributes to the literature in three ways: it provides rare panel econometric evidence on NPL determinants for the underrepresented ASEAN+4 region; it jointly models macroeconomic and structural-institutional variables within a unified framework; and it explicitly incorporates the COVID-19 shock, advancing understanding of pandemic-era credit risk dynamics in emerging economies.

Keywords:
non-performing loans, macroeconomic determinants, civil justice, panel data, ASEAN+4