Search Results (31)

Robust credit risk prediction in emerging economies increasingly demands the integration of external factors (EFs) beyond borrowers’ control. This study introduces a scenario-based methodology to incorporate EF—namely COVID-19 severity (mortality and confirmed cases), climate anomalies (temperature deviations, weather-induced road blockages), and social unrest—into machine learning (ML) models for credit delinquency prediction. The approach is grounded in a CRISP-DM framework, combining stationarity testing (Dickey–Fuller), causality analysis (Granger), and post hoc explainability (SHAP, LIME), along with performance evaluation via AUC, ACC, KS, and F1 metrics. The empirical analysis uses nearly 8.2 million records compiled from multiple sources, including 367,000 credit operations granted to individuals and microbusiness owners by a regulated Peruvian financial institution (FMOD) between January 2020 and September 2023. These data also include time series of delinquency by economic activity, external factor indicators (e.g., mortality, climate disruptions, and protest events), and their dynamic interactions assessed through Granger causality to evaluate both the intensity and propagation of external shocks. The results confirm that EF inclusion significantly enhances model performance and robustness. Time-lagged mortality (COVID MOV) emerges as the most powerful single predictor of delinquency, while compound crises (climate and unrest) further intensify default risk—particularly in portfolios without public support. Among the evaluated models, CNN and XGB consistently demonstrate superior adaptability, defined as their ability to maintain strong predictive performance across diverse stress scenarios—including pandemic, climate, and unrest contexts—and to dynamically adjust to varying input distributions and portfolio conditions. Post hoc analyses reveal that EF effects dynamically interact with borrower income, indebtedness, and behavioral traits. This study provides a scalable, explainable framework for integrating systemic shocks into credit risk modeling. The findings contribute to more informed, adaptive, and transparent lending decisions in volatile economic contexts, relevant to financial institutions, regulators, and risk practitioners in emerging markets. Full article

This study, which investigates the impact of foreign bank entry on the efficiency and sustainability of domestic banks in developing countries using a meta-frontier analysis to estimate efficiency scores, presents findings of significant importance to banking and finance. By incorporating financial, social, and environmental sustainability proxies—such as efficiency, loan portfolio composition, and macroeconomic conditions—this study assesses whether foreign competition enhances or undermines the long-term stability of domestic banking sectors. The results show that while foreign banks can improve financial efficiency, they may destabilize domestic banks, notably smaller or less capitalized institutions. Additionally, the findings suggest that banks with higher investments in SME lending and green projects demonstrate better social and environmental sustainability. Policymakers and financial institutions must consider these dual effects when promoting foreign bank entry. Full article

We analyze the loan portfolios of United States banks from 2013 to 2023, showing that high environmental, social, and governance (ESG) banks have larger shares of consumer loans and commercial loans and smaller shares of construction loans and real estate loans. We also find that the governance pillar (G) is more tightly related to the bank loan composition compared to the environmental (E) and social (S) pillars. Furthermore, we show that construction loans and real estate loans decrease more considerably with bank ESG scores inside countries with high gas emissions, i.e., where ESG issues would arguably be more serious. Our interpretation is that sustainable banks are reluctant in lending money for real estate projects, exposing them to potentially high ESG risk. These findings contribute to developing a deeper insight about the relationship between ESG and bank lending, which, in the previous literature, has been treated more frequently in aggregate terms instead of separating loan types. Our outcomes suggest that sustainability is crucial for the availability of funds in the real estate sector, delivering important insights to bank and real estate managers, besides policy makers. Full article

The purpose of this paper is to investigate the intertemporal relationship between the non-performing loan ratio and bank lending and to analyze factors affecting loan growth using data from Mongolian commercial banks. There has been a lack of research on Mongolian banks’ lending behavior due to their short history. Thus, this paper investigates the effect of the non-performing loan ratio on total loan growth using an ordinary least squares (OLS) regression model with panel data. We used bank-related variables such as the loan-to-deposit ratio, provision-to-gross loan portfolio ratio, equity-to-asset ratio, and liquidity ratio, and economic variables such as the real gross domestic product (GDP) growth rate, interest rate, and inflation rate. The results of this paper show that non-performing loans have a significant negative impact on total loan growth. The implication of this result is that non-performing loans affect banking efficiency, which, in turn, affects financial stability and the real economy. Moreover, high non-performing loans reduce banks’ profits. Also, this paper found that loss reserve and the liquidity ratio have a positive effect on total loan growth, while the effects of the loan-to-deposit ratio and the equity capital ratio were not found to be significant. Additionally, from a macro perspective, the inflation rate has a positive effect on the total loan growth rate, while the interest rate has a positive effect on total loan growth rather than a negative effect. And real gross domestic product (GDP) growth does not affect the total loan growth rate. Full article

This comparative study analyzed the impact of loan portfolio composition on the efficiency of different types of banks in India—public sector, old private, and new private banks—in the period between 2013 and 2022. Efficiency was evaluated using data envelopment analysis (DEA). The study considered four loan variables—term lending, working capital, priority sector lending, and secured lending in proportion to the overall loans—as independent factors against the efficiency score as the dependent variable, using a random-effects generalized least squares (GLS) regression framework. The results indicate that there were no significant effects on the efficiency of old private banks, except for working capital, which had a marginally negative impact on bank efficiency. Working capital, priority sector lending, and term lending have been found to significantly impact the efficiency of new private banks. Only term and working capital loans significantly affected the efficiency of public sector banks. Full article

Credit risk scorecard models are utilized by lending institutions to optimize decisions on credit approvals. In recent years, ensemble learning has often been deployed to reduce misclassification costs in credit risk scorecards. In this paper, we compared the risk estimation of 26 widely used machine learning algorithms based on commonly used statistical metrics. The best-performing algorithms were then used for model selection in ensemble learning. For the first time, we proposed financial criteria that assess the impact of losses associated with both false positive and false negative predictions to identify optimal ensemble learning. The German Credit Dataset (GCD) is augmented with simulated financial information according to a hypothetical mortgage portfolio observed in UK, European and Australian banks to enable the assessment of losses arising from misclassification costs. The experimental results using the simulated GCD show that the best predictive individual algorithm with the accuracy of 0.87, Gini of 0.88 and Area Under the Receiver Operating Curve of 0.94 was the Generalized Additive Model (GAM). The ensemble learning method with the lowest misclassification cost was the combination of Random Forest (RF) and K-Nearest Neighbors (KNN), totaling USD 417 million in costs (USD 230 for default costs and USD 187 for opportunity costs) compared to the costs of the GAM (USD 487, USD 287 and USD 200). Implementing the proposed financial criteria has led to a significant USD 70 million reduction in misclassification costs derived from a small sample. Thus, the lending institutions’ profit would considerably rise as the number of submitted credit applications for approval increases. Full article

The lead arrangers of syndicated loans often have lending relationships with the borrowers, while other lenders participating in the syndicate largely engage in an arm’s length transaction. Relatively little is known about how these relationships affect the shares of syndicated loans that the lead arrangers retain in their portfolio. Using a random sample of 10,328 syndicated loans made to 7316 nonfinancial U.S. firms over the period 1987 to 2013, this paper investigates the impact of lending relationships on the shares of loans retained. The results show that lending relationships are associated with a significant reduction in retained shares. These results are robust to alternative estimation techniques, such as propensity score matching and binary endogenous treatment models, which are employed to address endogeneity concerns. Furthermore, the results provide additional evidence that the existence and strength of lending relationships lead to decreased retained shares, particularly for non-top-tier lead arrangers. Moreover, the findings also demonstrate that when lead arrangers have lending relationships with borrowers, they retain significantly smaller shares whether the loans are made to informationally opaque, small, or speculative-grade-rating firms. Overall, the findings of this paper have important implications for lenders seeking to reduce their risk exposure in syndicated loans. Full article

Recent studies indicate that lending portfoliocomposition in Islamic banks is concentrated towardsdebt-based lending portfolio; however, the ideal lending portfoliocomposition in Islamic banks should be an equity-based lending portfolio. This article explores the effects of the internal governance factors on lending portfolio compositionofIslamic banks in the GCC Region. The internal governance factors investigated are board of directors’ characteristics (size and independence), Shariah supervisory board attributes (size and cross-membership), and ownership structure (family and government). The generalized least squares (GLS) method is used to examine the relationship between the study variables. The results indicate that two characteristics of the board of directors, size and independence, and two attributes of the Shariah supervisory board, Shariah board size and Shariah board cross-membership, have significant effects on lending portfolio composition of Islamic banks in the GCC Region. However, the rest of the internal governance factors have no effects on lending portfolio composition of Islamic banks in the GCC Region. These significant results add new contributions to the literature in the area of internal corporate governance of Islamic banks. The article concludes with suggestions for regulators and policy makers in the GCC Region with regard to the ideal characteristics of the board of directors and the optimal attributes of the Shariah supervisory board in Islamic banks as well as directions for future studies in this area of research. Full article

The severe effects of climate change and its anticipated negative influence on the future of the globe has prompted more research into the attainment of carbon neutrality. While carbon neutrality is a paramount issue, human socio-economic well-being which is mostly influenced by economic activities cannot be overlooked. This study investigates the effect of financial sector activities on CO₂ emission in five economic sectors and three economic bodies. The financial sector variables utilized are derived from the undertakings of institutions such as banks, stock exchanges, and insurance companies. Using a sample of 39 countries between 1989 and 2018, this paper provides a global perspective of the profound impact financial sector activities have in different economies on CO₂ emission reduction. The feasible generalized least squares (FGLS) regression model, as well as the random and fixed effects model with regards to Durbin–Wu–Hausman, are used to analyze the data. The generalized method of moments (GMM) is also adopted as the robustness method. Our findings show that for emerging economies, all major activities of the financial sector aggravated CO₂ emission levels in all major CO₂ emitting economic sectors. The developing and developed economies also show a similar trend. In the emerging economies, virtually all activities carried out by the financial sector have a significant negative impact on CO₂ emissions at the 1% or 5% significance level, thereby hampering CO₂ emission mitigation efforts. However, increased long-term bank lending to non-major economic sectors is found to alleviate CO₂ emissions in developing economies. This is also the situation with increased numbers of import insurance. Meanwhile, CO₂ emissions are found to decrease with increased net portfolio investments and numbers of insurance on exports. These findings not only imply that financial sector activities play a fundamental role in CO₂ emission mitigation but also serve as a reminder for financial policymakers that the decisions they make have an inevitable impact on the attainment of carbon neutrality in their economies. Full article

The systematicness of banks is an important driver of financial crisis. Overlapping portfolios and assets correlation of banks’ investment are important reasons for systemic risk contagion. The existing systemic risk models are all analyzed from one aspect and cannot reflect the real situation of the banking system. In the present paper, considering the overlapping portfolios and assets correlation, a contagion network model with multi-channel risk is proposed, which is with interbank lending (direct contagion channel), overlapping portfolios (indirect contagion channel), and assets correlation (indirect contagion channel). In addition, the model takes investment risk as an impact factor and learns the operation rules of the banking system to help banks compensate for liquidity through asset depreciation. Based on the proposed model, the effects of assets correlation, assets diversity, assets investment strategy, interbank network structure, and the impact of market density on risk contagion are studied and analyzed quantitatively. The method in this paper can more truly reflect the banking system risk than the existing model. This paper provides a solution for quantitative analysis of systemic risk, which provides powerful tools for macroprudential stress testing and a reference for regulatory authorities to prevent systemic risk. Full article

Decentralized finance (DeFi) protocols use blockchain-based tools to mimic banking, investment and trading solutions and provide a viable framework that creates incentives and conditions for the development of an alternative financial services market. In this respect, they can be seen as alternative financial vehicles that mitigate portfolio risk, which is particularly important at a time of increasing uncertainty in financial markets. In particular, some DeFi protocols offer an automated, low-risk way to generate returns through a “delta-neutral” trading strategy that reduces volatility. The main financial operations of DeFi protocols are implemented using appropriate algorithms, but unlike traditional finance, where issues of value and valuation are commonplace, DeFis lack a similar value-based analysis. The aim of this study is to evaluate relevant DeFi performance metrics related to the valuations of these protocols through a thorough analysis based on various scientific methods and to show what influences the valuations of these protocols. More specifically, the study identifies how DeFi protocol valuations depend on the total value locked and other performance variables, such as protocol revenue, total revenue, gross merchandise volume and inflation factor, and assesses these relationships. The study analyzes the valuations of 30 selected protocols representing three different classes of DeFi (i.e., decentralized exchanges, lending protocols and asset management) in relation to their respective performance measures. The analysis presented in the article is quantitative in nature and relies on Granger causality tests as well as the results of a fixed effects panel regression model. The results show that the valuations of DeFi protocols depend to some extent on the performance measures of these protocols under study, although the magnitude of the relationships and their directions differ for the different variables. The Granger causality test could not confirm that future DeFi protocol valuations can be effectively predicted by the TVLs of these protocols, while other directions of causality (one-way and two-way) were confirmed, e.g., a two-way causal relationship between DeFi protocol valuations and gross merchandise volume, which turned out to be the only variable that Granger-causes future DeFi protocol valuations. Full article

The purpose of this paper is to investigate the association between FinTech payments and consumer financial satisfaction with cashless payments using data from the 2017 China Household Finance Survey. This study defines computer payment and mobile terminal payment using a cell phone or pad as payments with FinTech. The results indicate that payments with FinTech are positively associated with financial satisfaction with cashless payments. Furthermore, this result holds in the eastern and central groups of China, but not in the western group, where payments with FinTech are not associated with financial satisfaction with cashless payments. Similarly, the positive association does not hold for consumers with low financial literacy. Moreover, analyses on the mediating effects imply that payments with FinTech play roles through three perceived mediators. Specifically, payments with FinTech help increase consumers’ perceived convenience and perceived popularity as well as reduce perceived risk, which eventually improves financial satisfaction with cashless payments. These findings have implications for consumer policymakers, such as improving the development of FinTech, noticing the heterogeneity in terms of location, and guiding consumers to correctly understand the risks associated with FinTech. Surrounding this issue, future studies may also explore other mediators related to psychology and expand the connotation of Fintech from payments with FinTech to lending and portfolio investments with FinTech. Full article

Global financial systems are increasingly interconnected, and risks can spread more easily, potentially causing systemic risks. Research on systemic risk based on multi-layer financial networks is relatively scarce, and studies usually focus on only one type of risk. This paper develops a model of the multi-layer financial network system based on three types of links: firm-bank credit, asset-bank portfolio, and interbank lending, which simulates systemic risk under three risk sources: firm credit default, asset depreciation, and bank bankruptcy. The impact of the multi-layer financial network structure, default risk threshold, and bank asset allocation strategy is further explored. It has been shown that the larger the risk shock, the greater the systemic risk under different risk sources, and the risk propagation cycle tends to rise and then decline. As centralized nodes in the multi-layer financial network system, bank nodes may play both blocking and propagation roles under different risk sources. Furthermore, the multi-layer financial network system is most susceptible to bank bankruptcy risk, followed by firm credit default risk. Further research indicates that increasing the average degree of firms in the bank–firm credit network, the density of the bank-asset portfolio network, and the bank capital adequacy ratio all contribute to reducing systemic risk under the three risk sources. Additionally, the more assets a bank holds in a single market, the more vulnerable it is to the risks associated with that market. Full article

Changing weather patterns may impose increased risk to the creditworthiness of financial institutions in the agriculture sector. To reduce the credit risk caused by climate change, financial institutions need to update their agricultural lending portfolios to consider climate change scenarios. In this paper we introduce a framework to compute the optimal agricultural lending portfolio under different increased temperature scenarios. In this way we quantify the impact of increased temperature, taken as a measure of climate change, on credit risk. We provide a detailed case study of how our approach applies to the problem of optimizing a portfolio of agricultural loans made to corn farmers across different corn producing regions of Ontario, Canada, under various climate change scenarios. We conclude that the lending portfolio obtained by taking into account the climate change is less risky than the lending portfolio neglecting climate change. Full article

Academic research on food consumer behavior related to food safety has developed extremely rapidly in the last decades, and a sizable amount of knowledge has been accumulated in this interdisciplinary field. This information set, as big data, lends itself to bibliometric analysis. Based on the Web of Science database and on a statistical analysis of more than 26.6 thousand articles containing more than 3.4 million bibliometric pieces of information, the current article offers a systematic analysis of these statistical data. The dynamics of relevant publications show an exponential character. The field is dominated by researchers from welfare states; however, food safety is a more important problem in developing states. There are dynamic changes in the portfolio of journals, but Bradford’s law cannot be proven. The explanatory power of Lotka’s law has been decreasing, proving the de-concentration of relevant authors. Besides traditional disciplines like consumer science, food chemistry, microbiology, and technology, new disciplines, e.g., sociology, cultural anthropology, postmodern techniques, and the real-life study of consumer behavior, going beyond the application of traditional techniques, are gaining importance. There are three key challenges for further research: (1) contribution to a deeper understanding of inherent laws governing the food-consumer-environment system; (2) quantification of results for decision-makers to enhance the efficiency of policy preparation; (3) widening the scope of research in geographical terms, better involving the developing world, and in sociological terms, focusing on the specific needs of vulnerable groups. Full article