Credit Decisions Research Articles

How Can Machine Learning Contribute to the Credit Scoring Process in SME Lending?

This research investigates the potential of machine learning to enhance the credit scoring process in small and medium-sized enterprise (SME) lending. SMEs are vital to global economies, accounting for over 90\% of businesses and more than half of total employment worldwide, according to The World Bank (2023). In emerging economies, SME lending is particularly crucial, providing access to capital that supports economic growth and development. However, assessing the credit risk of SMEs poses unique challenges due to limited financial histories and variable cash flows, making accurate credit scoring essential for lenders. Machine learning offers promising solutions for the risk assessment process in SME lending, leveraging algorithms and large datasets to identify patterns and improve decision-making. By applying techniques such as artificial neural networks, decision tree learning, gradient boosting, logistic regression, random forest, and support vector machines, machine learning can enhance the accuracy and efficiency of credit scoring models. This research explores how these machine learning methods can reduce biases, improve risk prediction, and ultimately support sustainable lending practices by enabling lenders to make more informed credit decisions. The findings aim to illustrate the transformative potential of machine learning in SME credit scoring, offering a framework for financial institutions seeking to integrate advanced data-driven technologies in the credit assessment process.

Outlier Detection Using Gaussian Mixture Model Clustering to Optimize XGBoost for Credit Approval Prediction

Credit approval prediction is one of the critical challenges in the financial industry, where the accuracy and efficiency of credit decision-making can significantly affect business risk. This study proposes an outlier detection method using the Gaussian Mixture Model (GMM) combined with Extreme Gradient Boosting (XGBoost) to improve prediction accuracy. GMM is used to detect outliers with a probabilistic approach, allowing for finer-grained anomaly identification compared to distance- or density-based methods. Furthermore, the data cleaned through GMM is processed using XGBoost, a decision tree-based boosting algorithm that efficiently handles complex datasets. This study compares the performance of XGBoost with various outlier detection methods, such as LOF, CBLOF, DBSCAN, IF, and K-Means, as well as various other classification algorithms based on machine learning and deep learning. Experimental results show that the combination of GMM and XGBoost provides the best performance with an accuracy of 95.493%, a recall of 91.650%, and an AUC of 95.145%, outperforming other models in the context of credit approval prediction on an imbalanced dataset. The proposed method has been proven to reduce prediction errors and improve the model's reliability in detecting eligible credit applications.

Creditworthiness Criteria According to the 5Cs Model and Credit Decision: The Moderating Role of Intellectual Capital

The current study aims to examine how creditworthiness criteria based on the (5Cs) model with its dimensions (character, capacity, capital, conditions, and collateral) on the credit decision in the presence of intellectual capital as a moderating variable applied to the branches of Jordanian commercial banks in the southern governorates of Jordan. The study included 449 populations from these branches, with a sample size of 205 employees. The authors employed a quantitative approach and utilized Macro PROCESS v4.2 by Andrew F. Hayes for analysis within SPSS v.28 software to achieve the study objective and test its hypotheses. According to the findings of the study, there is a positive effect of creditworthiness criteria based on the (5Cs) model on the credit decision. Furthermore, the study discovered that there is a presence of a moderating role of intellectual capital on the relationship between creditworthiness criteria based on the (5Cs) model and the credit decision. The study recommends that banks should prioritize investing in training programs for employees to improve their ability to assess credit risk accurately.

Enhancing transparency and fairness in automated credit decisions: an explainable novel hybrid machine learning approach

This paper uses a generalised stacking method to introduce a novel hybrid model that combines a one-dimensional convolutional neural network 1DCNN with extreme gradient boosting XGBoost. We compared the predictive accuracies of the proposed hybrid architecture with three conventional algorithms-1DCNN, XGBoost and logistic regression (LR) using a dataset of over twenty thousand peer-to-peer (P2P) consumer credit observations. By leveraging the SHAP algorithm, the research provides a detailed analysis of feature importance, contributing to the model’s predictions and offering insights into the overall and individual significance of different features. The findings demonstrate that the hybrid model outperforms the LR, XGBoost and 1DCNN models in terms of classification accuracy. Furthermore, the research addresses concern regarding fairness and bias by showing that removing potentially discriminatory features, such as age and gender, does not significantly impact the hybrid model’s classification capabilities. This suggests that fair and unbiased credit scoring models can achieve high effectiveness levels without compromising accuracy. This paper makes significant contributions to academic research and practical applications in credit risk management by presenting a hybrid model that offers superior classification accuracy and promotes interpretability using the model agnostic SHAP framework.

Textual analysis and credit scoring: a new matrix factorization approach

Credit scoring models are important for financial institutions’ credit decisions. This study examined how variables are extracted from loan statements and whether textual variables can improve the accuracy of the default model. We used a combination of forward selection and non-negative matrix factorization to extract variables from loan statements. We also built a credit scoring model using both loan statement and numerical data. The results show that in the comparative analysis, the credit scoring model built using the optimal cut-off logistic regression model and the two types of data had the highest accuracy. Moreover, compared with the credit scoring model constructed using the deep learning method based on word vectors, the credit scoring model in this study had better interpretation. The regression analysis revealed that the variables from the loan statement have a significant effect on the default status.

Risk Analysis of Mortgage Loan Default for Bank Customers and AI Machine Learning

Abstract Risk Analysis of Mortgage Loan Default for Banks is a crucial issue. On one hand, it concerns the quality of the bank's credit decisions, and on the other hand, it affects the rights of homebuyers to obtain financial support. In 2008, the U.S. subprime mortgage crisis sparked a global financial meltdown. What began with the collapse of the housing market quickly spread throughout the global financial system, resulting in the failure of numerous banks and a widespread economic recession. Recently, the banking sector has increasingly leveraged Artificial Intelligence (AI) and Machine Learning (ML) to enhance decision-making processes, particularly in assessing the risk of mortgage loan defaults. This paper aims to explore the application of ML techniques to predict and analyze the risk of default among bank customers, thereby enabling financial institutions to make more accurate and informed lending decisions. Keywords: Mortgage Loan Default Risk Analysis AI ML K-MANS LTV.

Cracking the Code: Hidden Choices and Visible Impacts Pattern Recognition in Corporate Finance

AbstractResearch in corporate finance suffers from bounded rationality due to static modeling. Adopting factor analysis, an unsupervised machine learning approach, and balance sheet information (accounts) over time, we find underlying dynamic latent corporate finance decisions. Our study identifies three latent corporate finance decisions adopted by executives in Pakistan, (1) long-term capital investment, (2) short-term debt credit, and (3) financial flexibility. The order of the decisions and the empirical tests highlight agency problems rooted in familial ownership concentration. We find that long-term capital investment and short-term debt credit decisions significantly reduce the firms’ present and future performance. Conversely, managers do not embrace financial flexibility, despite its ability to improve performance. The study highlights a contradiction, firms bounded by ownership concentration prefer control over performance and do not make decisions to optimize and protect minority shareholders’ wealth, depicting a moral hazard problem.

Loan application due diligence and financing small firms in Africa

ABSTRACT Banks gauge and verify business potential when processing small business loan applications by conducting due diligence on the firm. Although research has examined the factors that influence small business financing, little attention has been given to understanding the due diligence process that informs credit decision making. We address this gap using signal theory to analyze and develop insights from 24 qualitative interviews. Our study identified 11 key hard and soft signals loan officers use to access small business credit risk during due diligence. The key signals include key person risk, change in leadership risk, articulation of company value, adaptation to change, separating business and personal finances, personal reputation, transparency and willingness to share information, data accuracy and completeness, regular account operation, faulty budget and turnover assumptions, and legitimacy of collateral. The implications for small and medium-sized enterprises finance theory and practice are discussed.

Climate risk and split ratings: evidence from Korea

ABSTRACT This study examines the effects of climate risk on split ratings. Numerous studies have examined the correlation between financial reporting quality and split ratings. However, few studies have explored the effect of business ethics on split ratings. Thus, this study focuses particularly on environmental factors within the broader context of Environmental, Social, and Governance (ESG), that have recently attracted significant attention. We conducted a multiple regression analysis of the impact of climate risk on split ratings using a dataset of companies listed on the Korea Exchange. The analysis indicates that climate risk, which encompasses a firm’s greenhouse gas emissions and energy consumption, contributes to the split ratings of corporate bonds. This observation implies that heightened exposure to climate risk amplifies the likelihood of split ratings owing to diminished accounting transparency resulting from practices such as earnings management and a decline in the quality of disclosures. This study not only expands the understanding of the determinants of split ratings theoretically, but also offers investors empirically supported insights into the relevance of considering climate risk in their investment and credit decisions. Furthermore, this suggests the necessity for government policymakers to establish a monitoring system aimed at alleviating potential information asymmetry issues for firms facing elevated climate risk.

The Influence of Financial Literacy, Credit Requirements, and Financial Report Quality on Access to Financing for Micro, Small and Medium Enterprises

This study examines the influence of financial literacy, credit requirements, and the quality of financial reports on access to financing for MSMEs in the micro-trade sector in Yogyakarta City. Using a quantitative approach, data were collected from 80 MSME respondents selected through purposive sampling. Data were gathered via questionnaires with a Likert scale and analyzed using Partial Least Square (PLS) version 4.0 to evaluate the measurement and structural models. The findings reveal that financial literacy significantly influences access to financing, indicated by a T-statistic of 2.234 and a P-value of 0.026. This suggests that higher financial literacy among MSME owners facilitates easier access to financing. Credit requirements also significantly affect access, with a T-statistic of 3.259 and a P-value of 0.001, highlighting the impact of complex credit conditions on MSME financing accessibility. However, the quality of financial reports does not significantly affect access, as shown by a T-statistic of 1.057 and a P-value of 0.290, implying that financial reports are not the primary factor in credit decisions for MSMEs. This study emphasizes the importance of improving financial literacy and addressing credit requirements to enhance financing access for MSMEs.

Digital transformation, information and lending distance -Evidence from Chinese bank branch data

This research aims to empirically investigate the relationship between bank digital transformation and lending distance. In traditional financial geography theory, banks have a natural geographical proximity preference in their credit decision-making, which might lead to suboptimal credit allocation. Can the rapid development of digitalization improve the situation that banks are only capable of undertaking ex ante due diligence and ex post monitoring of proximate borrowers, given that fintech has greatly improved the overall efficiency of the financial industry? Using bank loan data of listed enterprises in China from 2013 to 2021, this paper finds that the digital transformation of banks can significantly extend the lending distance between banks and enterprises. The moderating effect analysis suggests that distance extending effect of bank digitalization is more pronounced for firms with poor information quality and in competitive banking sector. Further, digital transformation extends the lending distance without increasing risk preferences, as evidenced by the empirical finding that the non-performing loan ratio remains unaffected by the increased distance. Also, geographical expansion in lending distance reduces loan concentration. The research in this paper supports the conclusion that banks' digital transformation positively impacts corporate credit allocation and is beneficial to achieving financial stability by improving loan performances.

The Effect of Financial Literacy and Credit Procedures on Credit Taking Decisions

Purpose: This study aims to determine the effect of Financial Literacy on Credit Decision Making. To assess the effect of Credit Procedures on Credit Taking Decisions. Moreover, to determine the impact of Financial Literacy and Credit Procedures on Credit Decision Making. Research Design and Methodology: The population in this study were debtors of Bank BRI Jayapura City Unit Office who applied for credit in 2021-2022, totaling 1,714 debtors. The sampling technique used in this study is proportional random sampling. Data sources use primary data and secondary data. The statistical analysis method used in this research is Multiple Linear Regression Analysis based on the T Statistical Test technique, F Statistical Test, and the Coefficient of Determination with the help of SPSS version 29 software. Findings and Discussion: The results showed that Financial Literacy has a positive and significant effect on Credit decision-making, and credit procedures have a positive and significant effect on credit decisions. Financial Literacy and Credit Procedures simultaneously affect the decision to take credit at BRI Bank Jayapura City Unit. Implications: This research is expected to significantly influence future studies in economics, particularly financial management. The results are also likely to serve as a reference and comparison material for further research related to credit decision-making.

FinTech Credit and Entrepreneurial Growth

ABSTRACTBased on automated credit lines to vendors trading on Alibaba's online retail platform and a discontinuity in the credit decision algorithm, we document that a vendor's access to FinTech credit boosts its sales growth, transaction growth, and the level of customer satisfaction gauged by product, service, and consignment ratings. These effects are more pronounced for vendors characterized by greater information asymmetry about their credit risk and less collateral, which reveals the information advantage of FinTech credit over traditional credit technology.

A Review of Research on the Impact of Equity Investment on the Debt Financing of Invested Enterprises

To promote the theoretical and empirical research on the linkage of “stock loan debt insurance”, this paper summarizes the current situation of English and Chinese literatures research on the impact of VC/PE on the debt financing of invested enterprises. At present, the relevant research on the impact of VC/PE on the debt financing of invested enterprises in English literatures mainly focuses on the role and mechanism of third-party VC/PE, the role and mechanism of bank-affiliated VC/PE. The relevant research on the impact of VC/PE on the debt financing of invested enterprises in Chinese literatures mainly focuses on the direct role of third-party VC/PE and the mechanism of third-party VC/PE. From the relevant research results of English literatures and Chinese literatures, scholars generally believe that VC/PE has a positive impact on the debt financing of invested enterprises. The certification effect, corporate governance effect and ownership effect of VC/PE are the main mechanisms. In general, scholars’ research on ESG equity investment decisions of equity investment institutions for technology-based SMEs is relatively weak; in addition, scholars have rarely studied the impact of ESG equity investment intervention on the credit behavior of banks and technology-based SMEs. Under the concept of ESG, the academic research on the theoretical quantification and empirical estimation of the influence mechanism of ESG equity investment on the bank credit decision-making of technology-based SMEs needs to be carried out urgently.

Impact of Financial Literacy, Transaction Interests and Financial Inclusion on Credit Decisions on Peer to Peer (P2P) Lending in the West Medan Sub-District

The study aims to find out the impact of financial literacy, transaction interests and financial inclusion on peer to peer lending decisions in the West Medan sub-district. This research data is primary data using a questionnaire as a tool to collect data. Data collection technique with snowball sampling, which is the technique of determining samples that are initially small quantities, then these samples are asked to choose their friends to be used as samples. The population in this study was either online borrowers or those who used peer to peer lending of 202 users in the West Medan sub-district. The statistical analysis used in this study is a multiple linear regression analysis using SPSS (Statistical Package For the Social Science). The results showed that financial literacy has a positive and significant influence on the Peer to Peer Lending Credit Decision. Transaction interests have a negative but significant impact on the peer to peer lending credit decision. Financial inclusion has a significant and positive impact on peer-to-peer lending. The results of the F test have shown that financial Literacy, Transaction Interest and Financial Inclusion have a significant impact upon the Peer to Peer Lending credit Decision simultaneously.

What Makes a Smart Bank: The Case of Brde, a Last-Mile Subnational Development Bank in Brazil

Development banks are considered mission-oriented financial institutions that must act to fill gaps and market failures, supporting innovation with long-term financing in order to play a fundamental role in smoothing the risks associated with boom and bust cycles. . The existence and role of public development banks have been discussed for more than a century. Thus, this work proposes a case study with the analysis of the Subnational Development Bank - BRDE, in which there is a reference based on authors who discuss regional development and, subsequently, the case of BRDE, with the aim of describing the operation of a small bank in an emerging economy, its strengths and challenges. Therefore, it was possible to conclude that banks that have, at the same time, clarity of vision and mission and alignment of programs, flexibility to adjust to economic and social reality, solid financial results, robust governance attributes and autonomy - professional and collegial governance multilevel that increases the costs of both undue political pressure and bad credit decisions. From a financing point of view, the bank stands out for its technical capacity, which are critical but are important institutions to reinforce that its capillarity and knowledge of the context and local markets are useful for development. Thus, these banks provide financial instruments to expand financing at local and micro-regional levels, directly or through commercial banks and cooperatives. KEYWORDS: Regional development, banks, financial institutions, BRDE.

The effect of liquidity and company size on company value with profitability as a moderation variable

The purpose of this study was to determine the effect of liquidity and company size on firm value with profitability as a moderation variable. This study used 29 samples with 5 years of observation with the results of 145 observation samples in F&B companies listed on the Indonesia Stock Exchange for the period 2018-2022. The analysis technique used in this study is Moderated Regression Analysis (MRA), using Eviews 11 software. The results show that liquidity has a positive effect on the value of the company, this finding shows that companies with high profitability can have higher company values, even with low liquidity. These findings are important for investors and creditors in making investment and credit decisions. For company management, it is recommended to optimize inventory such as ensuring that inventory levels remain ideal for smooth operations and low storage costs. Strict credit policies and incentives for faster payments should be implemented to accelerate the recovery of trade receivables.

Does KAMs Disclosure Affect Trade Credit Decision? Evidence from Egypt

Does KAMs Disclosure Affect Trade Credit Decision? Evidence from Egypt

Environmental Penalties and Financing Punishment: Evidence from Incremental Bank Loans

ABSTRACT This study examines the impact of environmental penalties on corporate incremental bank loans. The results show that both the frequency and degree of environmental penalties significantly reduce corporate incremental bank loans, indicating that corporate environmental penalties lead to “financing punishment” by damaging corporate reputation and increasing risks. This financing punishment can be alleviated by good CSR performance and CSR assurance, indicating that exemplary CSR performance and assurance can provide an “insurance” effect to influence bank loans when a firm has a negative event. We also find that this financing punishment has intra-industry peer effects, suggesting that the environmental penalties incurred by one firm can influence the broader credit decisions of commercial banks within the same industry.

Combining E-Scores with Scenario Analysis to Evaluate the Impact of Transition Risk on Corporate Client Performance

Scenario analysis is a comprehensive approach to assess the impact of climate-related transition risk on businesses. Environmental, social, and governance (ESG) scores are popular tools with financial institutions (FI’s) for green-scoring practices and since they characterise a company’s performance from an ESG perspective, they have been criticised for enabling “greenwashing” when used within the context of climate risk. Commercially available ESG scores are also available for listed entities, while FI counterparties are often unlisted. This study develops a methodology for creating in-house environmental scores (E-scores), which are then used to effectively choose appropriate transition pathways to be used in company-specific forward-looking scenario analysis. Such scenario analysis can be used to forecast the company’s financial position, including the cost of its greenhouse gas (GHG) emissions, and quantify the impact of transition climate risk on specified metrics. The choice of metrics depends on what the results of the analysis are used for. Two metrics are identified for being useful for risk management and credit decisions: future profitability and weighted average carbon intensity. Finally, the study demonstrates how this process can be implemented with a practical worked example, using only publicly available data.