Good Borrowers Research Articles

The Imitation Game: The Imitation Game: How Encouraging Renegotiation Makes Good Borrowers Bad

Abstract We show that commercial mortgage borrowers behave opportunistically to attempt to obtain principal reductions. We develop a model in which lenders cannot perfectly observe borrowers’ use values and renegotiation is costly. We then exploit a tax rule change that reduced the cost of renegotiation. Consistent with the model predictions, borrowers with high private use values of the property are more likely to transfer into special servicing when lenders have a higher capacity to negotiate principal reductions after the rule change. Our results suggest adverse consequences of principal forgiveness for lenders.

Character Counts: Psychometric-Based Credit Scoring for Underbanked Consumers

Psychometric-based credit scores measure important personality traits that are characteristic of good borrowers’ behaviors. While such data can potentially improve credit models for underbanked consumers, the utility of psychometric data in consumer lending is still largely understudied. The present study contributes to the literature in this respect, as it is one of the first studies to evaluate the efficacy of psychometric-based credit scores for predicting future loan defaults among underbanked consumers. The results from two culturally diverse samples of loan applicants (Sub-Saharan Africa, n = 1113; Western Europe, n = 1033) found that psychometric scores correlated significantly with future loan defaults (Gini = 0.28–0.31) and were incrementally valid above and beyond the banks’ own credit scorecards. These results highlight the theoretical basis for personality in financial behaviors, as well as the practical utility that psychometric scores can have for credit decisioning in general and the facilitation of financial inclusion for underbanked consumer groups in particular.

Credit scoring: Does XGboost outperform logistic regression?A test on Italian SMEs

The old-fashioned logistic regression is still the most used method for credit scoring. Recent developments have evolved new instruments coming from the machine learning approach, including random forests.In this paper, we tested the efficiency of logistic regression and XGBoost methods for default forecasting on a sample of 35,535 cases from 7 different business sectors of Italian SMEs, on a set of 28 banking variables and 55 balance sheet ratios for verifying which approach is better supporting the lending decisions.With this aim, we developed an efficiency index for measuring each model's capability to correctly select good borrowers, balancing the different effects of refusing the loan to a good customer and lending to a defaulter. Also, we computed the balancing spread to quantify the different models' efficiency in terms of credit costs for the borrower firms.Results show that different sectors report different results. However, generally speaking, the two methods report similar capabilities, while the cutoff setting can make a substantial difference in the actual use of those models for lending decisions.

Government Lending as a Tool to Mitigate the Effect of Asymmetric Information

This paper analyzes the effect of asymmetric information on investment efficiency and the ways in which government credit can mitigate the inefficiency caused by asymmetric information. The focus is on credit rationing. In the model, the project return and the project risk are assumed to be independent random variables. Under this assumption, asymmetric information produces two types of inefficiency: exclusion of good borrowers and inclusion of bad borrowers. Using a subsidy, the government can improve the composition of borrowers (average project return) in the sector where information is opaque (opaque sector) and increase the volume of loans in the opaque sector. The best way to improve investment efficiency is that the government sets the lending rate above the profit-maximizing level (lower lending rate adjusted for the default rate), bids up the funding rate to raise more loan funds, and makes up the difference between the high funding rate and the low adjusted lending rate with a subsidy. A subsidized loan guarantee covering a moderate portion of loan losses can produce similar outcomes. This result is a dramatic deviation from the conventional notion of a subsidized loan, which almost automatically means a lower lending rate.

THE NEW LAW ON MICRO BANKS AND WHAT IT WILL CHANGE IN THE FINANCIAL MARKET OF GEORGIA

Microfinance is the provision of financial services to low-income groups. By classical definition it contributes to the improvement of the socio-economic condition as well as financial self-sufficiency of the low-income costumer and improves their living standards. Today, microfinance institutions around the world serve up to 156 million borrowers and manage a loan portfolio of USD 187 billion (Impact Finance Barometer, 2022: 3). Formation of Georgian microfinance market began in the 1990s and was developed mainly with the same scenario as in Balkans, Eastern Europe and Central Asia. During the first 10 years after adoption of the new Law on Microfinance Organizations and commercialization of the industry, a rapid growth and fast development of the sector was recorded (2006-2016), which has been then changed by a negative trend and market decline. The main reasons of the crisis are: tighter regulation, market saturation, growing competition from commercial banks, over-indebtedness, weak corporate governance, etc. In July 2022, the draft law on the activities of micro banks was registered in the Parliament of Georgia, according to which a new financial institution focused on lending to entrepreneurial and agricultural activities - a micro bank - will enter the market from 2023. The draft law was prepared based on the recommendation of the International Monetary Fund, which advised the National Bank to introduce a micro bank as a new institutional form (International Monetary Fund, 2020: 19). As a result of the law, 4 types of financial institutions will be presented in the financial market of Georgia since 2023: commercial banks, micro banks, microfinance organizations and loan issuing entities. The new regulatory framework should give microbanks the following advantages over the current status of MFIs: The maximum loan amount per borrower will be increased from GEL 100,000 to GEL 1 million, which allows microbanks to actively enter the SME lending segment, retain good borrowers, grow faster faster in the market and compete with banks. Having the right to mobilize the deposits, microbanks will be able to better diversify their funding sources, reduce the cost of local currency borrowing (based on savings in hedging) and, accordingly, reduce the interest rate on loans for thier clients, which will also increase their competitiveness with banks. By offering bank account services, microbanks will be able to provide customers with an almost full range of banking services. After legally differenciating from “consumption” and “pawnshop” type of financial institutions, microbanks will be able to attract more investments from responsible and reputable international financial institutions. At the same time, the new draft law contains certain potential risks that need to be properly assessed and analyzed: The increase of loan size limit from 100,000 to 1 million GEL is a good for portfolio growth. However, most companies lack the experience and qualification to properly manage risks of such large loans. Past experience also confirms that the worst portfolio quality and problem loans MFIs have in large loans (over GEL 10,000). Administration of the article, according to which at least 70% of the loans of a microbank must be disbursed for business purposes, seems very difficult. As of today practically no MFIs can meet this criteria. Adding the new products (deposits, account management) requires a high-quality core banking systems and automation of many processes. Consequently, microbanks will have to mobilize additional financial and human resources to build this infrastructure, which may take several years and entail significant costs. In conclusion, it can be said that the draft law is likely to have a positive impact on the financial sector. Competition in the financial sector will increase, which will lead to the provision of better quality and diverse financial services to customers.

Financial supply chain analysis with borrower identification in smart lending platform

The popularity of the online peer-to-peer (P2P) lending platform in the financial supply chain (FSC) has grown tremendously in the past few years. However, it is pretty challenging to develop an efficient financial supply chain for Micro, Small, and Medium-Sized Enterprises (MSMEs) by providing credits and services. In this research, an FSC model analyzes by emphasizing the retailer and supplier relationship. The retailer predicts the market demand to maximize profit, and accordingly, orders to the suppliers and suppliers utilize the production capabilities in limited monetary value. In such a case, the supplier also needs to apply for loans to meet the production commitment. With an intelligent lending platform, applicants can get loans quickly and more conveniently from the lenders. Still, identifying defaulter borrowers is a difficult task on the peer-to-peer lending platform. The main focus of the lenders is to maximize profit and minimize risk by giving the loan to non-defaulter suppliers. The present study describes the importance of the financial parameters (λ) of the retailers and suppliers (MSMEs) and their impact on the related supply chain. We provide the proof of dependency of the financial parameter on the cost of debt and the ratio of the debt-to-equity by developing two propositions. We propose an innovative k-Random Boosting Classifiers (k-RBC) algorithm for identifying potential good and bad borrowers to capture this case. The time complexity of the k-RBC algorithm is O(n2log(n)). The results obtained from the study show a significant improvement in comparison to the outputs from existing approaches on the same datasets. Our algorithm gives 90% accuracy to identify potential good borrowers, whereas existing algorithms achieve up to 87% accuracy. Furthermore, the importance of the borrowers’ features and its impact are analyzed of the lending platform in FSC.

The Imitation Game: How Encouraging Renegotiation Makes Good Borrowers Bad

The Imitation Game: How Encouraging Renegotiation Makes Good Borrowers Bad

Factors Affecting Default in the Loan Portfolio of Microfinance Institutions in Cameroon

The main objective of microfinance is to provide funds to those who are excluded from the banking system. But in order to attain this objective they have to deal with the issue of default in their loan portfolios. The aim of this paper is to analysis the factors that affect the default of borrowers in microfinance institutions. The discriminant analysis reveals that 96.6% of the bad borrowers are correctly classified and 92.1% of the good borrowers are correctly classified and that other debts, age of borrower, borrowers income and number of dependents significantly affect the likelihood of default. We suggest that credit officers in microfinance institutions should be keen on these factors when granting loans.

Loan Officers Impede Graduation from Microfinance: Strategic Disclosure in a Large Microfinance Institution

One of the most important puzzles in microfinance is the low rate of borrower graduation to larger, more flexible loans. Utilizing observational and experimental data from a large Chilean microfinance institution, we demonstrate that loan officers impede borrower graduation due to common features of their compensation contracts. Our partner lender offers both microloans and larger, more flexible graduation loans, and relies on loan officer endorsements to determine borrower graduation. Loan officers are rewarded for the size of their portfolio and repayment, and so are implicitly penalized when good borrowers graduate. In an experiment designed to isolate strategic disclosure, we modify compensation to reduce this implicit penalty and document that loan officers withheld endorsements of their most qualified borrowers prior to the shift. Graduated borrowers endorsed after the shift are 34% more profitable for our partner lender than those endorsed beforehand. A back-of-the- envelope calculation suggests that strategic behavior of loan officers accounts for $4.8-29.2 billion in lost social value from forgone borrower graduations in microfinance worldwide. Our experimental design may prove useful for other experiments within firms.

Are you a good borrower? Mining interpretable pattern structures in credit scoring

Purpose The purpose of this study is to show that closure-based classification and regression models provide both high accuracy and interpretability. Design/methodology/approach Pattern structures allow one to approach the knowledge extraction problem in case of partially ordered descriptions. They provide a way to apply techniques based on closed descriptions to non-binary data. To provide scalability of the approach, the author introduced a lazy (query-based) classification algorithm. Findings The experiments support the hypothesis that closure-based classification and regression allow one to both achieve higher accuracy in scoring models as compared to results obtained with classical banking models and retain interpretability of model results, whereas black-box methods grant better accuracy for the cost of losing interpretability. Originality/value This is an original research showing the advantage of closure-based classification and regression models in the banking sphere.

Government Lending As a Tool to Mitigate the Effect of Asymmetric Information

This paper analyzes the effect of asymmetric information on investment efficiency and the ways in which government credit can mitigate the inefficiency caused by asymmetric information. The inefficiency caused by asymmetric information critically depends on the way in which the project return is related with the project risk. To produce more general results, I assume that the project return and the project risk are independently distributed random variables. Under this assumption, asymmetric information produces two type of inefficiency: exclusion of good borrowers and inclusion of bad borrowers. Asymmetric information results in investment inefficiency by making the borrower composition in the sector where information is opaque (opaque sector) suboptimal and under-allocating credit to the opaque sector. In the case of credit rationing, the government can improve the borrower composition and the sector allocation using a subsidy, thereby improving investment efficiency. The most interesting result is a positive effect of a higher lending rate on investment efficiency. Government credit is much more likely to improve investment efficiency when the lending rate is set at a level above the private lenders' profit-maximization level than when the lending rate is set at a level below it. When lenders raise the lending rate, borrowers with a higher repayment probability drop out because the borrowing rate adjusted for the repayment probability (adjusted borrowing rate) is higher, and borrowers with a lower project return drop out because they cannot afford to pay a higher borrowing rate. Thus, a higher lending rate is likely to improve the borrower composition in terms of the average project return. The best way to deal with credit rationing, therefore, is to set the lending rate at a high level (low adjusted lending rate) to weed out low-return projects, bid up the funding rate to increase the loan volume in the opaque sector, and make up the difference between the high funding rate and the low adjusted lending rate with a subsidy. In this way, the government loan can accommodate many high-risk, high-return projects. This result is a dramatic deviation from the conventional notion of a subsidized loan, which almost automatically means a lower lending rate. A subsidized loan guarantee covering a moderate portion of loan losses can deliver similar outcomes.

Combining multiple probability predictions in the presence of class imbalance to discriminate between potential bad and good borrowers in the peer-to-peer lending market

Credit risk scoring predictions represent an effective guide for lenders to discriminate between potential good (who will repay the loan) and bad (who will default) borrowers in the online social lending market. A common characteristic of such a market is a lower percentage of defaulted borrowers than non-defaulted borrowers; thus, the sample is class imbalanced. Class imbalance may affect the accuracy of default predictions, as classifiers tend to be biased towards the majority class (good borrowers). We analyse the default prediction performance when combining class rebalancing methods with different regression and machine learning techniques. We also propose to combine multiple probability predictions to improve the predictive performance. The analysis is based on a book of loans (with a three-year term) funded in the 2010–2015 period though the online platform of Lending Club. The results show that some measures of predictive accuracy tend to improve when the scoring models are trained using a rebalanced, rather than an imbalanced sample, except when the extreme gradient boosting approach is applied. Finally, we find that combining multiple probability predictions via regularised logistic regression may help to improve the predictive accuracy.

Leverage dynamics and credit quality

We study a dynamic model of collateralized credit markets with asymmetric information, which allows for a rich set of signaling strategies based on the path of debt and repayment. Whether credit history reveals private information about credit quality depends on the degree of uncertainty in collateral value. When uncertainty is low, good borrowers fully and costlessly separate by deleveraging, that is borrowing a sufficiently high amount such that subsequent repayment reveals the presence of unobservable income. When uncertainty is higher, good borrowers pay an adverse selection cost through a higher interest rate because bad borrowers could default, and asymmetric information is not always resolved.

On the duration of sovereign ratings cycle phases

• We analyze the drivers of sovereign rating phases via discrete-time Weibull models. • For speculative-rated sovereigns, reputation build-up as good borrowers is gradual. • The reputation of investment-grade sovereigns remains unchanged as time passes. • The length of both phases significantly depends on the country's economic conditions. • Better governance quality helps to reduce the duration of speculative-grade phases. Using long-term sovereign ratings data for a panel of 130 countries over the last three decades, we investigate the duration and determinants of sovereign rating phases through the lens of discrete-time Weibull models. We find that the likelihood of the end of the ‘speculative-grade’ phase increases as time goes by (i.e. there is positive duration dependence), but the ‘investment-grade’ phase is not duration dependent. Thus, for sovereigns rated as speculative, the build-up of reputation as good borrowers is a gradual process, whereas the reputation of investment-grade sovereigns solidifies and remains unchanged as time passes. However, the length of both phases significantly depends on the country's economic conditions. In particular, lower inflation, stronger growth and sounder fiscal policies shorten (prolong) the speculative- (investment-) grade phase. In addition, better governance quality helps to reduce the duration of speculative-grade phases.

The Benevolence of Time, Sound Macroeconomic Environment and Governance Quality on the Duration of Sovereign Ratings Phases

Using long-term sovereign ratings data for a panel of 130 countries over the last three decades, we rely on discrete-time Weibull models to investigate the duration and determinants of sovereign ratings phases. We find that the likelihood of the end of the 'speculative-grade' phase increases as time goes by (i.e. positive duration dependence), but the 'investment-grade' phase is not duration dependent. Thus, for sovereigns rated as speculative, the build-up of reputation as good borrowers is a gradual process, whereas the reputation of investment-grade sovereigns solidifies and remains unchanged as time passes. However, the length of both phases has proven to be significantly dependent on the country's economic conditions and the quality of governance. In particular, lower inflation, better economic performance and sounder fiscal policies shorten (prolong) the speculative- (investment-) grade phase. In addition, a better perception of the quality of public services and the ability of the government to formulate and implement sound policies and regulations help to reduce the duration of speculative-grade phases.

Exploring the Performance of Synthetic Minority Over-sampling Technique (SMOTE) to Predict Good Borrowers in P2P Lending

Exploring the Performance of Synthetic Minority Over-sampling Technique (SMOTE) to Predict Good Borrowers in P2P Lending

Credit risk assessment for unbalanced datasets based on data mining, artificial neural network and support vector machines

PurposeCredit scoring datasets are generally unbalanced. The number of repaid loans is higher than that of defaulted ones. Therefore, the classification of these data is biased toward the majority class, which practically means that it tends to attribute a mistaken “good borrower” status even to “very risky borrowers”. In addition to the use of statistics and machine learning classifiers, this paper aims to explore the relevance and performance of sampling models combined with statistical prediction and artificial intelligence techniques to predict and quantify the default probability based on real-world credit data.Design/methodology/approachA real database from a Tunisian commercial bank was used and unbalanced data issues were addressed by the random over-sampling (ROS) and synthetic minority over-sampling technique (SMOTE). Performance was evaluated in terms of the confusion matrix and the receiver operating characteristic curve.FindingsThe results indicated that the combination of intelligent and statistical techniques and re-sampling approaches are promising for the default rate management and provide accurate credit risk estimates.Originality/valueThis paper empirically investigates the effectiveness of ROS and SMOTE in combination with logistic regression, artificial neural networks and support vector machines. The authors address the role of sampling strategies in the Tunisian credit market and its impact on credit risk. These sampling strategies may help financial institutions to reduce the erroneous classification costs in comparison with the unbalanced original data and may serve as a means for improving the bank’s performance and competitiveness.

Do bad borrowers hurt good borrowers? A model of biased banking competition

This paper explores a two-bank model in which, first, one bank correctly estimates the probability of low-quality loan repayment while the other overestimates it, and second, both banks have identical convex costs when granting loans. In this context of optimistically biased banking competition, we show how the unbiased bank follows the biased competitor as long as the bias of the latter is not too large. This would favour bad borrowers, who get better credit conditions at the expense of good borrowers. As a consequence, the presence of a biased bank increases welfare as long as the expected default rate is sufficiently high. Contrariwise, in subprime markets, biased banking competition would be socially harmful.

The Legacy and the Tyranny of Time: Exit and Re‐Entry of Sovereigns to International Capital Markets

AbstractWe use a continuous‐time Weibull model (without and) with a change‐point in duration dependence to investigate the duration of the exit and re‐entry of sovereigns to international markets. We find that, as the reputation of debtor countries as good (bad) borrowers solidifies over time, those episodes are more likely to end—the “legacy of time.” Debtor countries take advantage of the “benefit of doubt” of creditors during short exits. When exits are long and the reputation as a bad borrower emerges, no more “complacency” makes it more difficult to borrow again in international markets—the “tyranny of time.”

Strategic group lending for banks

Credit institutions often refuse to lend money to small firms. Usually, this happens because small firms are not able to provide collateral to lenders. Moreover, given the small amount of required loans, the relative cost of full monitoring is too high for lenders. Group lending contracts have been viewed as an effective solution to credit rationing of small firms in both developing and industrialized countries. The aim of this paper is to highlight the potential of group lending contracts in terms of credit risk management. In particular, this paper provides a theoretical explanation of the potential of group lending programs in screening good borrowers from bad ones to reduce the incidence of non-performing-loans (NPL). This paper shows that the success of firms involved in selected group lending programs is due to the fact that co-signature is an effective screening device: more precisely, if lenders make a proper use of co-signature to screen good firms from bad ones, then only firms that are good ex-ante enter group lending contracts. So, the main argument of this paper is that well designed group lending programs induce good firms to become jointly liable, at least partially, with other good firms and discourage other – bad-firms to do the same. Specifically, co-signature is proven to be a screening device only in the case of a perfectly competitive bank sector.