This study examines the effect of credit risk management (CRM) on the financial performance of Saudi Arabian banks and investigates whether this relationship differs between Islamic and conventional banking models. Using panel data from 40 banks covering 2020–2024, the study incorporates key credit-risk indicators including NPLA/PLAL, PLAL/TLA, NPLA/TLA, TLA/TAS, and LDR and applies multiple regression and group-comparison tests. The results reveal that CRM significantly influences profitability, with higher non-performing loan ratios reducing ROE, while stronger lending intensity (LDR) and higher loan concentration (TLA/TAS) enhance performance. Comparative tests indicate substantial differences in credit-risk profiles across bank types but no significant difference in financial performance levels. However, interaction-term analysis demonstrates that the impact of credit-risk indicators on ROE varies meaningfully between Islamic and commercial banks. Overall, the findings underscore CRM’s essential role in sustaining profitability and highlight the moderating effect of banking model structures within Saudi Arabia’s Basel-aligned regulatory environment.

ABSTRACT This study investigates the effect of corporate environmental, social, and governance (ESG) performance on credit risk using a sample of manufacturing firms listed on China's Shanghai and Shenzhen A‐share markets from 2009 to 2021. Employing fixed effects, the generalised method of moments, and instrumental variable models, we find that stronger ESG performance is significantly associated with lower credit risk, as measured by the distance to default. Mediation analysis reveals that this relationship operates primarily through enhanced profitability and improved external governance. In contrast, Tobin's Q acts as a negative channel, potentially reflecting market overvaluation and inefficiencies. ESG's impact also varies across firm types: the risk‐reducing effect is most pronounced among non‐state‐owned enterprises (NSOEs), firms based in eastern provinces, and those in the growth or decline stage of the corporate lifecycle. Further analysis shows that environmental (E) and social (S) pillars drive credit improvements, whereas the governance (G) score has an insignificant effect. Our findings provide theoretical and empirical insights into the ESG–credit risk nexus, highlighting the importance of sector‐specific, regionally sensitive ESG strategies in emerging markets.

As a core part of the financial industry, credit operations are accompanied by significant risks. Therefore, accurate credit risk control is crucial for financial institutions' lending decisions and overall risk management. In this paper, we propose a hybrid machine learning framework (Neural network-Ensemble learning-Reinforcement learning Hybrid Framework, NERHF) for efficient credit risk control. The framework utilizes neural network algorithms to extract features from credit data, enhancing the accuracy and robustness of credit risk prediction. Further, based on the extracted features, ensemble learning algorithms are employed for credit risk prediction. Finally, the improved deep reinforcement learning algorithm Pre-DDQN is applied to generate optimal credit risk control strategies for different combinations of key credit indicators, aiming to mitigate default risks. Experimental results show that NERHF demonstrates significant advantages in credit risk prediction, especially when using recurrent neural networks for feature extraction in conjunction with lightweight gradient boosting machine algorithms. Additionally, the Pre-DDQN algorithm outperforms comparative algorithms in credit risk control, highlighting its potential for practical applications.

Integrating Environmental, Social, and Governance (ESG) metrics into supply chain finance is critical for promoting sustainable development. However, the dynamic mechanisms through which real-time ESG performance influences credit allocation and, consequently, shapes credit risk and environmental risk exposures for financial institutions, remain poorly understood, especially when compared to traditional static and retrospective ESG evaluations. To address this, we developed an agent-based model that simulates interactions among green enterprises, a financial institution, and a regulator, featuring a dynamic credit algorithm that adjusts credit lines based on real-time ESG scores. Our simulations demonstrate that ESG-driven credit policies significantly boost green technology adoption among SMEs, raising adoption rates from 20% to over 85% under strong incentives, which in turn drives a substantial reduction of the supply chain’s carbon footprint by more than 50%. Notably, this environmental benefit is achieved without a commensurate surge in credit risk, as the non-performing loan ratio only experienced a moderate increase. Additionally, sensitivity analysis reveals a non-linear relationship between the ESG weighting in credit decisions and environmental outcomes, identifying a critical threshold for policy effectiveness. Our findings offer risk managers and policymakers evidence-backed strategies for designing dynamic incentives that effectively promote supply chain decarbonization while managing associated financial risks.

Review on Credit Risk Assessment Using Machine Learning Algorithms

Data Asset Collateralization, Credit Risk Pricing, and Corporate Digital Transformation

ABSTRACT Earnings-based debt covenants in Japan typically reference GAAP earnings, particularly “ordinary income,” an earnings measure unique to Japan defined as earnings before taxes and special items. We posit that ordinary income serves as an effective summary indicator of firms’ operating and financing performance and that its use in debt covenants helps lenders reduce their contracting costs. Consistent with this prediction, we find that the use of ordinary income in earnings-based loan covenant is negatively associated with the number of covenant types, suggesting that it reduces the need for additional covenants. Interestingly, we also find that EBITDA is more useful than other earnings measures, including ordinary income, in explaining credit risk and predicting future cash flows. Overall, our evidence suggests that ordinary income is employed primarily to economize on contracting costs, rather than to serve as an information signal. Data Availability: Data are available from the public sources cited in the text. JEL Classifications: G21; G32; M41.

This study examined how certain Ethiopian microfinance institutions manage credit risk. Microfinance institutions are a crucial means of raising substantial funds in developing nations. Secondary data sources were collected from specific microfinance institutions for this investigation. The years 2010 through 2018 were consecutively included in the collected data. The data were analysed using a balanced regression model. Purposive sampling was used to select the sample, and a quantitative research methodology was employed to achieve the studys objectives. The studys target population is the whole Ethiopian microfinance industry. An explanatory research design is used to achieve the studys aim. For analytical reasons, credit risk management is measured as the dependent variable. Moreover, explanatory (independent) factors include the loan-to deposit ratio, average loan balance per borrower, company development, managerial performance, and total assets. Credit risk is positively correlated with total assets, the loan-to-deposit ratio, management performance, and average loan balance per borrower, as indicated by a regression analysis in EViews 8. The study also recommended that companies enhance their overall asset value and management performance by implementing training and other strategic measures.

Credit risk classification plays a crucial role in supporting financial institutions, especially microfinance institutions, in assessing the ability of customers to repay loans. This study aims to develop a credit risk classification model using the Random Forest method, which is known for its accuracy and robustness in handling classification problems. The research uses a dataset obtained from a microfinance institution consisting of various customer attributes such as income, age, loan amount, repayment history, and employment status. The dataset is preprocessed and divided into training and testing sets to evaluate model performance. The Random Forest algorithm is then applied to build a classification model that categorizes customers into three credit risk levels: low, medium, and high. The results show that the Random Forest model achieves a high level of accuracy, with a classification precision of 89%, recall of 87%, and F1-score of 88%. These findings indicate that Random Forest is an effective technique for credit risk classification and can be implemented by microfinance institutions to support better decision-making in credit approval processes. This research also highlights the potential of machine learning techniques in enhancing credit risk management and minimizing non-performing loans.

Liquidity, Credit and Leverage Risk as Determinants of Profitability: An Empirical Study of Pakistan s Commercial Banks

The relationship between Islamic Commercial Bank health indicators and non-performing financing requires careful examination amid recent economic volatility. Researchers analyzed quarterly financial data from eleven Indonesian Islamic Commercial Banks spanning 2019-2024 using panel data regression through E-views software, employing the RGEC framework with Financing to Deposit Ratio (FDR), Return on Equity (ROE), Operating Expense to Operating Income Ratio (BOPO), and Capital Adequacy Ratio (CAR) as independent variables, while Non-Performing Financing (NPF) served as the dependent variable. Statistical analysis reveals that FDR, ROE, BOPO, and CAR jointly lack predictive power for NPF variations, with the model explaining merely 3.079% of variance (Adjusted R-Square). Individual variable testing identifies FDR as the sole significant predictor, demonstrating an inverse relationship with NPF—higher financing distribution correlates with lower default rates when lending remains selective, whereas ROE, BOPO, and CAR show no meaningful association with non-performing financing levels. These findings challenge conventional assumptions about financial ratio utility in predicting credit quality deterioration, as the model's weak explanatory power suggests internal financial metrics alone offer insufficient understanding of NPF dynamics. Future research should integrate external determinants including macroeconomic indicators, regulatory policy shifts, and institutional risk management practices to develop more robust predictive frameworks for Islamic banking credit risk.

Purpose: This study aims to examine the effect of Non-Performing Loans (NPL) and Capital Adequacy Ratio (CAR) on the Return on Assets (ROA) of primary dealer banks in Indonesia during the period 2020–2024. The research is motivated by the vital role of the banking industry in maintaining financial system stability and supporting the national economy, with primary dealer banks serving as the government's strategic partners in issuing and trading Government Securities (SBN). Methodology/approach: The study employs a quantitative approach using panel data regression analysis with EViews 13. The population consists of 17 primary dealer banks, and 16 of them are used as the research sample for the 2020–2024 period. Findings: The results reveal that NPL has a significant negative effect on ROA, indicating that higher credit risk adversely impacts bank profitability. Conversely, CAR shows no significant effect on ROA, implying that capital adequacy does not directly influence profitability levels. Practical and Theoretical contribution/Originality: This research provides new insights by employing primary dealer banks in Indonesia as the research subject, which have been rarely used in previous studies examining similar topics. Theoretically, the study contributes to the understanding of the relationship between risk management, capital structure, and profitability within the context of Indonesia’s primary dealer banking system. Research Limitation: The study is limited to the variables NPL and CAR, and focuses only on primary dealer banks within the 2020–2024 period. Another limitation lies in the incomplete historical data of primary dealer banks, which may restrict the comprehensiveness of the analysis. Future research is recommended to include additional financial and non-financial variables as well as extend the observation period to provide a broader understanding of factors affecting banking financial performance in Indonesia.

This study analyzes banking credit performance and credit risk in Indonesia before, during, and after the Covid-19 pandemic. The research applies a descriptive analytical method using secondary data obtained from the Financial Services Authority (OJK), focusing on commercial banks. Before Covid-19, banking credit was in a stable condition, with normal lending and credit growth exceeding the growth of third-party funds (DPK). This indicates that banks were active in distributing credit supported by sufficient funding. When Covid-19 occurred, economic performance declined significantly, and credit distribution contracted. For three consecutive quarters, both economic growth and credit growth recorded negative figures as banks became more cautious due to shrinking business activities and rising uncertainty. In the post-pandemic period, spanning nine quarters up to Q4 2024, the banking sector showed strong recovery aligned with economic improvement. Credit growth increased sharply and was noticeably higher than DPK growth, demonstrating restored confidence and better financial intermediation capacity. Regarding credit risk, banks managed non-performing loans (NPL) effectively both before and after Covid-19, maintaining NPL ratios below the 5 percent regulatory threshold. During Covid-19, banks strengthened reserves through increased allowance for impairment losses (CKPN) to keep NPL stable, although this temporarily reduced profitability as reflected in lower return on assets (ROA). Overall, Indonesian banks showed resilience through prudent risk management.

Explainable AI (XAI) is critical in domains like credit scoring where model decisions must be transparent and accountable. This survey paper compares three local explanation techniques—SHAP, LIME, and ensemble Hybrid approach that integrates both. We evaluate these methods on consistency, variability, and suitability for regulatory environments. Emphasis is placed on use in credit risk modeling, with insights drawn from both literature and practical evaluation.

This study explores the link between user sentiment and credit risk on FinTech lending platforms using sentiment analysis techniques like Latent Dirichlet Allocation (LDA) and the Liu Hu method. Analyzing data from 2020 to 2023, findings reveal Kiva leads with 91.16% positive feedback and a 4.7-star rating but fewer reviews (617). LendingClub, with 1.58K reviews, has mixed sentiment (56.08% positive, 39.99% negative) and a lower rating (3.3 stars). Plenti achieves 58.33% positive sentiment but lower coherence, while Mintos balances sentiment (66.69% positive) with the largest review base (100K+). Results show platforms with higher positive sentiment and topic coherence mitigate credit risk more effectively, underscoring the value of user feedback in optimizing marketplace lending. The study offers actionable insights for FinTech stakeholders to improve app performance and user-centric financial solutions through effective sentiment analysis.

This study examines the effect of risk management implementation on the financial performance of Bank BJB (PT Bank Pembangunan Daerah Jawa Barat dan Banten, Tbk) during the period 2015–2024. The research aims to analyze how credit risk (Non-Performing Loan), liquidity risk (Loan to Deposit Ratio), and capital adequacy (Capital Adequacy Ratio) influence Return on Assets as a measure of financial performance. Employing a quantitative explanatory design, the study uses secondary data from Bank BJB’s annual reports, financial statements, and good corporate governance reports. Data analysis is performed through multiple linear regression with classical assumption tests to ensure model validity. The results indicate that Non-Performing Loan has a negative and significant effect on Return on Assets, while Loan to Deposit Ratio and Capital Adequacy Ratio have positive and significant effects. These findings confirm that effective risk management significantly contributes to profitability and financial stability. The novelty of this research lies in its integration of risk management variables within the context of regulatory evolution, digital transformation, and post-pandemic recovery in Regional Development Banks. The study contributes to both academic literature and practical policy by providing empirical evidence on how prudent risk management can enhance bank performance and resilience. Keywords: Risk Management; Financial Performance; Non-Performing Loan; Loan to Deposit Ratio; Capital Adequacy Ratio

Real estate loans account for a relatively high proportion of commercial bank loans, the current stage of the real estate market is depressed, the huge real estate loans on the bank has caused a huge burden, the impact of the bank's credit, uncertain real estate policy will affect both the quality of the stock of real estate loans, but also affect the commercial banks to the real estate industry's credit investment, all of which ultimately affects the risk taking of commercial banks. Based on this, this paper first combs through the real estate central level real estate general tone policy and financing policy since 2023, and then selects the micro panel data of 16 listed banks in China from 2010-2023 to analyze the impact of the stock of real estate loans on banks' credit risk. The results show that a rise in real estate loan stock significantly increases commercial bank credit risk. The research in this paper provides a basis for the government to strengthen real estate policy regulation with comprehensive measures and block the transmission of corporate risks to the financial sector.

This study evaluates the impact of three flagship MSME credit schemes—Pradhan Mantri MUDRA Yojana (PMMY), Credit Guarantee Fund Trust for Micro and Small Enterprises (CGTMSE), and Prime Minister’s Employment Generation Programme (PMEGP)—on inclusive economic growth in India. Using data from 2015 to 2024 and policy analysis, it assesses each scheme’s contribution to credit access, risk mitigation, and employment generation. PMMY has shown sustained growth and high disbursement efficiency; CGTMSE has expanded guarantee coverage, especially in the post-pandemic period; and PMEGP has delivered cost-effective job creation despite recent fluctuations. The findings underscore the collective role of these schemes in advancing financial inclusion and entrepreneurship, offering policy insights to enhance equity, resilience, and long-term sustainability in MSME support.

This paper analyzes solvency dynamics in Ecuador’s mutualist segment by modeling the joint behavior of the productive-assets-to-total-assets ratio (PATR) and portfolio-specific delinquency rates. Using monthly supervisory data from the Superintendencia de Economía Popular y Solidaria (SEPS) for the full universe of four mutualist institutions (2022–2025), we estimate a multivariate Gaussian Hidden Markov Model on system-level aggregates. The model identifies latent regimes that summarize configurations of asset productivity and segmented credit risk, distinguishing relatively sound conditions from episodes of heightened vulnerability. Model selection is based on information criteria, complemented by convergence checks, distributional diagnostics, and alternative covariance specifications to assess robustness. The approach is explicitly framed as diagnostic rather than causal or prescriptive: it does not replace simple thresholds nor calibrate capital buffers, but organizes supervisory information into interpretable solvency states with associated frequencies and expected durations. The framework is transparent and reproducible and provides a baseline for future extensions with longer samples and richer covariates.

The COVID-19 pandemic has caused significant global economic disruptions, particularly increasing non-performing loans (NPLs) in the banking sector. This study examines the impact of bank specific factors and macroeconomics on NPLs in banks across ASEAN countries. Panel data from 39 publicly listed banks over the period 2010–2024 were examined using the Generalized Method of Moments (GMM) and system GMM approach. The results indicate that lagged NPLs, return on equity, total assets, credit growth, GDP growth, and lending interest rates significantly influence NPLs, while the Tier 1 capital ratio, non-interest income, and unemployment rate show no significant effect. This study provides important implications for bank management and policymakers in enhancing credit risk management and strengthening banking supervision within the ASEAN region.