25
A credit scoring model for Vietnam's retail banking market Thi Huyen Thanh Dinh a,1 , Stefanie Kleimeier a, a Limburg Institute of Financial Economics (LIFE), Faculty of Economics and Business Administration (FdEWB), Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands Available online 22 June 2007 Abstract As banking markets in developing countries are maturing, banks face competition not only from other domestic banks but also from sophisticated foreign banks. Given the substantial growth of consumer credit and increased regulatory attention to risk management, the development of a well-functioning credit assessment framework is essential. As part of such a framework, we propose a credit scoring model for Vietnamese retail loans. First, we show how to identify those borrower characteristics that should be part of a credit scoring model. Second, we illustrate how such a model can be calibrated to achieve the strategic objectives of the bank. Finally, we assess the use of credit scoring models in the context of transactional versus relationship lending. © 2007 Elsevier Inc. All rights reserved. JEL classification: G21; G28; O16 Keywords: Retail banking; Emerging markets; Credit scoring; Default risk 1. Introduction Are retail loans special? Advocates of relationship lending tend to answer this question with yes. Since retail borrowers are typically individuals or entrepreneurs, information about them is not readily available but needs to be obtained through close contact over the life of a bankclient relationship. Relationship lending is thus based on soft, proprietary information. Lending decisions are not only determined by exact characteristics such as the borrower's income or collateral but to a large extent also by qualitative information such as the borrower's character, reputation, or standing in the community. As relationships are expensive to maintain and as Available online at www.sciencedirect.com International Review of Financial Analysis 16 (2007) 471 495 Corresponding author. Tel.: +31 43 3883733; fax: +31 43 3884875. E-mail addresses: [email protected] (T.H.T. Dinh), [email protected] (S. Kleimeier). 1 Thi Huyen Thanh Dinh worked on this study while being a master student at the Maastricht University. Since then she has moved to De Lage Landen International B.V., Vestdijk 51, 5611 CA, Eindhoven, The Netherlands. Tel.: +31 40 2339174; fax: +31 40 2338626. 1057-5219/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.irfa.2007.06.001

Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

  • Upload
    tonyag

  • View
    28

  • Download
    2

Embed Size (px)

DESCRIPTION

Credit scoring model for VN retail banking

Citation preview

Page 1: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Available online at www.sciencedirect.com

International Review of Financial Analysis 16 (2007) 471–495

Acredit scoringmodel for Vietnam's retail bankingmarket

Thi Huyen Thanh Dinh a,1, Stefanie Kleimeier a,⁎

a Limburg Institute of Financial Economics (LIFE), Faculty of Economics and Business Administration (FdEWB),Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands

Available online 22 June 2007

Abstract

As banking markets in developing countries are maturing, banks face competition not only from otherdomestic banks but also from sophisticated foreign banks. Given the substantial growth of consumer creditand increased regulatory attention to risk management, the development of a well-functioning creditassessment framework is essential. As part of such a framework, we propose a credit scoring model forVietnamese retail loans. First, we show how to identify those borrower characteristics that should be part ofa credit scoring model. Second, we illustrate how such a model can be calibrated to achieve the strategicobjectives of the bank. Finally, we assess the use of credit scoring models in the context of transactionalversus relationship lending.© 2007 Elsevier Inc. All rights reserved.

JEL classification: G21; G28; O16Keywords: Retail banking; Emerging markets; Credit scoring; Default risk

1. Introduction

Are retail loans special? Advocates of relationship lending tend to answer this question with‘yes’. Since retail borrowers are typically individuals or entrepreneurs, information about them isnot readily available but needs to be obtained through close contact over the life of a bank–clientrelationship. Relationship lending is thus based on soft, proprietary information. Lendingdecisions are not only determined by exact characteristics such as the borrower's income orcollateral but to a large extent also by qualitative information such as the borrower's character,reputation, or standing in the community. As relationships are expensive to maintain and as

⁎ Corresponding author. Tel.: +31 43 3883733; fax: +31 43 3884875.E-mail addresses: [email protected] (T.H.T. Dinh), [email protected] (S. Kleimeier).

1 Thi Huyen Thanh Dinh worked on this study while being a master student at the Maastricht University. Since then shehas moved to De Lage Landen International B.V., Vestdijk 51, 5611 CA, Eindhoven, The Netherlands. Tel.: +31 402339174; fax: +31 40 2338626.

1057-5219/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.irfa.2007.06.001

Page 2: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

472 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

relationship lending might not always be profit maximizing, banks have introduced transactionallending tools where lending decisions are based on quantifiable borrower characteristicssupported for example by a credit scoring model (CSM). In 1996, already 97% of all US banksused credit scores for credit card loan applications and 70% for small business loans. CSMs havealso spread globally to banking markets in other developed countries. By now, the most widely-used CSMs are those of Fair Isaac and Co. Inc. who developed models for small business tradecredit in 1998, small business credit in 1999, and finally personal credit in 2001.2 In contrast, inthe retail banking markets of developing countries lending practices tend to be poor and banks,i.e. small banks or micro-financiers lack the necessary data on borrower characteristics and theircredit histories to design reliable CSMs. It is therefore not surprising that only a few of the largestbanks in developing countries use CSMs. However, as banking markets in developing countriesare maturing, banks face competition not only from other domestic banks but also fromsophisticated foreign banks. Given the substantial growth of retail credit and increased regulatoryattention to risk management, the development of a well-functioning credit assessment frame-work is essential.

In essence, a CSM provides an estimate of a borrower's credit risk – i.e. the likelihood that theborrower will repay the loan as promised – based on a number of quantifiable borrowercharacteristics. Given its widespread use in developed countries, the benefits of credit scoring aswell as the methodological issues surrounding credit scoring are well established in the academicliterature.3 Evidence regarding CSMs for retail banking markets in developing countries,however, is extremely limited as only two studies exist: Viganó (1993) for Burkina Faso andSchreiner (2004) for Bolivia. Both studies analyze small business loans. In contrast to businessloans in developed countries, reliable up-to-date financial data is often lacking and the boundarybetween the private and business property of an entrepreneur is often vague. Thus, CSMs forthese loans need to rely – just like CSMs for consumer loans – on the personal information aboutthe entrepreneur.4 Schreiner's (2004) study is practitioner-oriented and discusses the benefits ofcredit scoring as well as its implementation in general. Viganó (1993) specifically addresses theproblem of selecting the relevant borrower characteristics for the CSM. Despite being considered“the best scorecard for microfinance in the literature” (Schreiner, 2004), Viganó's (1993) analysisuses very limited samples containing 31 to 100 loans. In contrast, our study is based on thecomplete retail loan portfolio of a Vietnamese commercial bank. For our CSM, we can utilize dataon more than 56,000 loans including in addition to small business loans also consumer loans,mortgages and credit card loans. Our study is thus the first to provide a CSM that is applicable tothe overall retail lending activities of a bank in a developing country.

In particular, we identify which borrower characteristics a bank needs to collect and how thesecan be combined into a scoring model. Though we start with a similar set of borrowercharacteristics as many developed-country retail CSMs, two of our four most importantcharacteristics are unique to Vietnam – gender and loan duration – while commonly includedcharacteristics related to income and employment are not part of our CSM. In this sense we cannotfully agree with Allen, DeLong, and Saunders (2004) who conclude for small business loans that

2 See Mester (1997) and Allen et al. (2004).3 See for example Mester (1997) and Blöchlinger and Leippold (2006) for the former and Altman (1968), Hand and

Henley (1997), Altman and Saunders (1998) and Thomas (2000) for the latter literature.4 In this sense, the small business loans considered in this study are different from those discussed by Altman and

Narayanan (1997) and as summarized in Table 2 of Allen et al. (2004). Even though these studies consider business loansto developed as well as developing countries, they focus on the prediction of bankruptcy based on financial data of thebusiness alone.

Page 3: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

473T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

“what is striking is not so much the models' differences across countries of diverse size and invarious stages of development but rather their similarities.” Due to the detailed nature and largesize of our sample, we can furthermore illustrate how CSMs can be calibrated to achieve thebank's strategic objectives. Though this issue has been investigated for banks in developedcountries, our study is the first to analyze this for a developing-country bank. Firstly, we find thatour proposed CSM leads to an increase in profits, indicating that our bank's non-performing loantarget is adequately selected. Secondly, we provide a simple rule for risk-based loan pricing whichleads to the same profit margin as the currently applied single interest rate for all borrowers.Thirdly and finally, we present credit scoring as part of a lending-decision-making process that isbuilt on the principles of transactional lending but leaves room for relationship lending. Here wepropose a more specific CSM with two cut-off points. Such calibration allows banks to definethose loans that are accepted or rejected on the CSM alone as well as those which are furtherexamined by the credit officer. Overall, the bank benefits from reduced cost of loan assessmentwhile optimally using its relationship lending tools, i.e. its loan officer's knowledge about theborrower.

The remainder of the paper is structured as follows: Section 2 presents a general CSM ascurrently applied for retail credit and illustrates the modeling steps and decisions that have to betaken. In Section 3 a model is estimated using the retail loan population of a Vietnamesecommercial bank. Specific applications of the model regarding bank profitability, risk-based loanpricing, and relationship lending are explored in Section 4. Section 5 concludes.

2. A credit scoring methodology for retail loans

CSMs are commonly structured along the lines of Altman's (1968) Z-score model.5 First, theCSM must be developed and estimated. Typically, a CSM uses historical loan and borrower datato identify which borrower characteristics are best able to distinguish between defaulted and non-defaulted loans. During this developmental stage, decisions about the model's estimation methodhave to be made and all potentially relevant borrower characteristics have to be identified andcoded. Finally, the relevant borrower characteristics have to be selected and their impact ondefaults established — in other words, the model has to be estimated. Now the CSM can beapplied to new loan applications for which the probability of default (PD) is not known. Based onthe estimated CSM, a credit score can be calculated for each new loan applicant where a higherscore indicates better expected performance of the borrower and thus a lower PD. This score mustbe compared to the CSM's cut-off rate to determine whether the loan application is accepted,rejected or requires further assessment. Thus, as part of the CSM's developmental stage,calibration is also necessary such that an optimal cut-off rate, which is in line with the lender'sobjectives is determined.

2.1. The estimation method

The development of a CSM starts with the decision about the basic form of the model, i.e. itsestimation method via decision trees, linear probability models, logit or probit regression models, or

5 For an overview of the bankruptcy prediction literature since Altman's original study see Byström, Worasinchai, andChongsithipol (2005) or Philosophov and Philosophov (2002). The former paper is also the only one to investigate theissue for a developing country. See Laitinen (1999) for a discussion of the related literature on the prediction of credit riskratings.

Page 4: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Table 1Variables commonly used in retail credit scoring models

Commonly used variables inindustrialized countries (Crook et al., 1992)

Variables included in the credit scoring model

Crook et al. (1992) — UK Schreiner (2004) — Bolivia Viganó (1993) — Burkina Faso

Postcode Postcode (1) Date of disbursement Customer's personal characteristics(age, sex, religion, martial status, education,employment sector and place, etc)

Age Employment status (2) Amount disbursedNumber of children Years at bank (3) Type of guaranteeNumber of other dependants Current account (4) Branch Data on the enterprise (type, professional skills,

number of employees, productivity, profitability, etc)Whether an applicants has a home phone Spouse's income (5) Loan officerSpouse's income Residential status (6) Gender of the borrowerEmployment status Phone (7) Sector of the firm Profitability (main and secondary revenue, revenue stability)Employment category Years at present employment (8) Number of spells or arrearsYears at present employment Deposit account (9) Length of the longest

spell of arrearsAmount and composition of assets(total assets including money and deposits)

Income Value of home (10)Residential status Outgoings (11) Financial situation

(initial and current amount of loans received,defaults, loans granted)

Years at present address Number of children (12)Estimated value of home Applicant’s income (NI)Mortgage balance outstanding Mortgage balance outstanding (NI) Investment plans

(presence of investment plan, other sources of finance)Years at bank Charge card (NI)Whether a current account is heldWhether a deposit account is held Customer's relationship with bank

(past loans with bank, savings account with bank, etc)Whether a loan account is heldWhether a check guarantee card is heldWhether a major credit card is held Bank's control of credit risk

(loan destination, disbursement form, method of repayment,loan amount and maturity, collateral,contractual conditions on interest rate, etc)

Whether a charge card is heldWhether a store card is heldWhether a building society card is heldValue of outgoings

When available, a rank indicating the importance of the variable in the credit scoring model is given in brackets. (NI) indicates that a variable was considered but finally not included.

474T.H

.T.Dinh,

S.Kleim

eier/International

Review

ofFinancial

Analysis

16(2007)

471–495

Page 5: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

475T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

multiple discriminant analyses. Based on findings by Boyle, Crook, Hamilton, and Thomas (1992),Desai, Crook, and Overstreet (1996, 1997), Henley (1995), Srinivasan and Kim (1987), and Yobas,Crook, and Ross (2000) as reported by Thomas (2000), we opt for the logistic regression method.Thus, we first collect a sample of existing loans and observe their borrower characteristics and theex-post default status. A borrower's ex-ante PD is unobservable. The logistic regression techniqueovercomes this problem by directly estimating this probability and has therefore been themethodology of choice for retail credits. This technique assumes the existence of a continuousvariable Zj which is defined as the probability that loan-applicant j defaults and can be modeled as alinear function of a set of variables x which describe the applicant:

Zj ¼ W Vx ¼ w1xj1 þ w2xj2 þ N þ wkxjk ð1Þwhere wk is the coefficient of the kth variable and xjk is the value of variable k for applicant j. Zj isknown as Z-Score of the jth applicant. Zj is ex-ante unobservable and default can only be defined ex-post as a 0–1 dummy. Eq. (2) thus derives the default probability π using an iterative maximumlikelihood estimation method. Here, larger values of π reflect a higher PD.

pj ¼ 1

1þ e�ðW VxÞ ð2Þ

2.2. Variable identification

Next, a choice has to be made among the variables x that should initially be considered for Eqs.(1) and (2). There is no overall consensus regarding the number or type of initial variables. SomeCSMs such as First Data Resources' model for credit cards lenders (Mester, 1997) or CaisseNationale de Crédit Agricole du Burkina Faso's model for rural microfinance (Viganó, 1993) startwith as many as 50 variables. Table 1 presents a list of the more commonly-used variables forretail CSMs. In CSMs for corporate loans, the variables are relatively similar across countries andlimited to financial statement data (Allen et al., 2004). For retail loans, direct measures of thefinancial strength of the borrower such as income or value of the home are included. Given thelimited availability of such proxies for individuals, proxies that measure the financial strength of aborrower indirectly (education, household size, years at employer, or postal code) also need to beincluded. There seem to be only a few variables that are typical for developing countries such asgender, religion, branch, and loan officer. This can be explained by the fact that banks in mostdeveloped countries are prohibited by law to use variables such as gender, whereas banks indeveloping countries are not. Branch and loan officer variables can reflect branch policy, bonuses,experience, or training levels. Their inclusion in the CSM for Bolivia might indicate thatdifferences across branches and loan officers are more pronounced in less-developed bankingmarkets. Except for these few variables, however, it seems that a retail CSM for a developingcountry can generally start with the same set of variables as a retail CSM for a developed country.

Note that already at this early stage, a certain selection of variables has to be made. Even banksthat do not use a CSM are collecting information from loan applicants during the lending-decisionmaking process. This information can serve as a starting point. Practical experience will howevershow that not all variables are fully available. Values may be structurally missing (e.g. questionwhich are asked conditionally on the responses to previous questions). Alternatively, missingvalues might be due to a deficient information system in some branches or due to customers whothemselves were not willing to completely fill in the application forms. Excluding all variables

Page 6: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

476 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

with missing values might substantially reduce sample size and lead to a loss of valuableinformation whereas including variables with only a few valid observations might lead tounreliable results. In order to balance these two problems, we first identify those variables whichshow a substantial number missing values. Second, we rely on the loan officers' expertknowledge and feeling for the data and characteristics to identify which of these variables can andwhich cannot be excluded from the model (see Henley & Hand, 1997).

2.3. Variable coding

After the choice of the initial set of independent variables has been made, qualitative as well asquantitative variables need to be coded. The coding of quantitative variables is required when therelationship between the variable and default is not linear. As Thomas (2000) argues, instead oftrying to map such a relationship as a straight line, borrowers can be grouped into a number ofcategories such that a categorical variable is created. The later approach is more commonly usedin credit scoring mainly because it can also be applied to qualitative variables and we thus adoptthis approach for all our variables. To complete the coding of the variables, a value has to beattached to each category. Boyle et al. (1992) show that variables can be coded based on thedistribution of defaulted and non-defaulted loans in the sample. If a variable has m categories, letgi be the number of good (non-defaulted) loans who belong to the ith category and bi the numberof bad (defaulted) loans that belong to the ith category. G and B are the total number of good andbad loans in the whole sample, respectively, such that

G ¼Xm

i¼1

gi and B ¼Xm

i¼1

bi ð3Þ

Instead of using a simple coding rule6, estimates of probability odds of the good and badloans in the ith category are commonly used. We follow Crook et al. (1992) and code ourvariables as7

lnðgi=biÞ þ lnðB=GÞ ð4ÞFor qualitative variables for which the number of possible categories is very large, coding all

categories becomes infeasible. For such cases, we follow Thomas' (2000) and Boyle et al.'s(1992) recommendation and aggregate values of similar PD measured as bi / (gi+bi).

2.4. Variable selection and estimation of the CSM

Once the variables have been coded, Eq. (2) can be estimated. As is evident from retail CSMsin developed countries, the model can initially contain a large number of variables. Whereas thismight be statistically feasible for large samples, there are practical considerations. Too manyquestions in an application form deter loan applicants, who will not answer all questions or applyfor a loan elsewhere. Efficiency and applicability thus stipulate that the number of variables isreduced. It is furthermore critical to determine not only how many but also which variables to

6 A simple rule assigns, for example, the value of 1 to the category with the lowest PD, the value of 2 to the categorywith the second highest PD, etc.7 In our sample ln(B/G) is constant. In practice, however, when banks are updating their CSM at regular intervals, this

value will change with every update.

Page 7: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

477T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

incorporate in the model. We therefore apply a forward as well as a backward stepwise methodwhich sequentially adds or withdraws variables to maximize the model's predictive accuracy(see Henley & Hand, 1997). Now, the final version of the CSM can be determined and thecoefficients wj of Eq. (2) can be estimated.

2.5. Calibration

Finally, the CSM should be tested for its predictive accuracy. This is best done out-of-sampleas the model predicts over-accurately in-sample. First, the PD for each loan in the calibration-sample is estimated. This PD is then compared to a cut-off value to establish whether the loanapplicant will be a good (non-defaulting) or bad (defaulting) borrower. Initially, a cut-off value of50% can be chosen such that an applicant who's estimated PD is greater (smaller) than 50% willbe classified as a bad (good) loan. This classification is then compared to the observed event ofdefault to establish the accuracy of the model. A classification table as shown in Table 2 iscommonly employed at this stage. Gg represents the number of correctly classified good loanswhereas Gb represents the number of good loans that are incorrectly classified as bad loans.Similarly, Bb represents the number of correctly classified bad loans whereas Bg represents thenumber of bad loans that are incorrectly classified as good loans. The percentage of correctlyclassified (PCC) loans serves as an accuracy measure. The percentage of correctly classified goodloans (PCCgood) is defined as the proportion of correctly classified good loans to the total numberof observed good loans. Correspondingly, the percentage of correctly classified bad loans(PCCbad) is defined as proportion of correctly classified bad loans to the total number of observedbad loans. Finally, the percentage of correctly classified total loans (PCCtotal) is defined as thenumber of correctly classified loans relative to the total number of loans. Though simple to use,PCC may not always be an appropriate accuracy measure. It implicitly assumes that the costs ofmisclassification of bad and good loans are equal. For banks, one classification error may,however, be much more expensive than another. Another, often violated assumption of PCC isthat the class distribution is constant over time and relatively balanced (Provost, Fawcett, &Kohavi, 1998). Baesens et al. (2003) use two additional accuracy measures called sensitivity(SENS) and specificity (SPEC). In contrast to PCCbad, which relates the predicted bad loans to thetotal number of observed bad loans, SPEC is defined as proportion of correctly classified badloans to the total number of predicted bad loans. Correspondingly, SENS is defined as numberclassified good loans relative to the total number of predicted good loans. Banks might want to

Table 2Predictive accuracy of credit scoring models

Observation Prediction PCC

Non-default Default

Non-default Gg Gb PCCgood=Gg / (Gg+Gb)Default Bg Bb PCCbad=Bb / (Bb+Bg)

PCCtotal= (Gg+Bb) / (Gg+Gb+Bb+Bg)SENS Gg / (Gg+Bg)SPEC Bb / (Bb+Gb)

For the abbreviations Gg, Gb, Bb and Bg the capital letter indicates the actual nature of the loan whereas the subscriptindicates the predicted nature. Based on these four measures, the accuracy of a credit scoring model can be assessed by thepercentage of correctly classified loans (PCC), sensitivity (SENS) and specificity (SPEC).

Page 8: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

478 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

minimize both Bg and Gb at the same time. However, reducing Bg comes at the expense ofincreasing Gb, and vice versa. Thus banks should take into account its differences in cost arisingfrom misclassified good versus bad loans. For most banks the cost of Bg will be higher than thecost Gb and the CSM can be calibrated based on SENS. In Section 4.1. we introduce a moreadvanced method that takes these two costs explicitly into account.

3. An application to the Vietnamese retail lending market

In 1987, Vietnam started its transformation to a market economy. Part of this process is thereplacement of the monopoly of state-owned banks by a two-level banking system consisting of anational central bank on one level and state-owned as well as commercial banks on another level.8

Projects to modernize the inter-bank market, to create an international accounting system, and toallow outside audits of major Vietnamese banks are ongoing. However, the banking systemcontinues to suffer from lack of capital, inadequate provisions for possible loan losses, lowprofitability, inexperience in capital markets, and the slow pace of institutional reform. Withrespect to risk assessment and management, there are numerous difficulties including a lack oftransparency in non-performing loan disclosure. There is for example no uniform definition of anon-performing loan. It is thus not surprising that non-performing loan rates range from 3% basedon banks' official financial reports to 35–70% based on figures by the World Bank or IMF.9 Inorder to improve risk management in light of Basel II, Vietnam's central bank has been reviewingits risk management regulations. As part of a broader strategy, which also addresses the banks'business strategy, assets and liability management, and internal audit, all state-owned commercialbanks and joint-stock commercial banks have been asked to develop a comprehensive creditmanual which takes international practices in risk management into account. The development ofa proper CSM is an important building block within this credit manual.

To illustrate the current state of credit assessment in Vietnam, Table 3 summarizes the mostadvanced credit scoring system as it is currently used by one of Vietnam's commercial banks. Thisbank scores applicants based on 14 variables in a two step procedure. In a first assessment round, aloan applicant is evaluated based on the nine criteria listed in Panel A. If her score is higher than acertain threshold, she will be evaluated again based on the criteria listed in Panel B. Otherwise, she isrejected directly without any further consideration. Based on the total score from both rounds, theloan applicant will be rated as shown in Panel C.Note that the variables used to score applicants in thefirst round overlap with those listed in Table 1. This finding enforces our earlier conclusion thatdespite the difference in economic development between Vietnam and industrialized countries, thesame factors are relevant for the lending decision. The variables used in the second round measureexplicitly the customer's relationshipwith the bank and thus highlight the importance of relationship-lending aspects in the decision process. Furthermore, the credit assessment system presented in Table3 is a qualitative method where scores and weights of each variable are not the result of a statisticalapproach but based on experience and judgment of the credit officer. More importantly, with thissystem, the bank cannot estimate a PD and therefore cannot exploit this estimate in improving its riskmanagement. Given the high levels of non-performing loans, the development of a more

8 In 2005 when the data for this study was collected, this second level of the Vietnamese banking system contained fivestate-owned commercial banks, one social policy bank, 31 foreign bank branches, 40 foreign credit institutionrepresentative offices, five joint-venture commercial banks, 36 domestic joint-stock commercial banks, seven financecompanies, and the Central People's Credit Fund System with 23 branches and 888 local credit funds.9 See Loi (2006).

Page 9: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Table 3An example of credit assessment in Vietnam

Panel A: Variables considered in the first round of credit assessment

Variable Categories

Age 18–25, 26–40, 41–60, N 60 (years)Education post graduate, graduate, high school, less than high schoolOccupation professional, secretary, businessman, pensionerTotal time in employment b 0.5, 0.5–1, 1–5, N 5 (years)Time in current job b 0.5, 0.5–1, 1–5, N 5 (years)Residential status owns home, rents, lives with parents, otherNumber of dependents 0, 1–3, 3–5, N 5 (people)Applicant's annual income b 12, 12–36, 36–120, N 120 (million VND)Family's annual income b 24, 24–72, 72–240, N 240 (million VND)

Panel B: Variables considered in the second round of credit assessment

Variable Categories

Performance history with bank(short-term)

new customer, never delayed, payment delay less than 30 days, payment delay morethan 30 days

Performance history with bank(long-term)

new customer, never delayed, delay during 2 recent years, delay earlier than 2 recentyears

Total outstanding loan value b 100, 100–500, 500–1000,N 1000 (million VND)Other services used savings account, credit card, savings account and credit card, noneAverage balance in savings account

during previous yearb 20, 20–100, 100–500,N 500 (million VND)

Panel C: Loan decision

Applicant's rating Score Loan decision

Aaa ≥ 400 Lend as much as requested by borrowerAa 351–400 Lend as much as requested by borrowera 301–350 Lend as much as requested by borrowerBbb 251–300 Loan amount depends on the type of collateralBb 201–250 Loan amount depend on the type of collateral with assessmentb 151–200 Loan application requires further assessmentCcc 101–150 Reject loan applicationCc 51–100 Reject loan applicationc 0–50 Reject loan applicationd 0 Reject loan application

479T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

sophisticated CSM seems indeed necessary. In the remainder of this section, we develop such amodel based on the retail loan portfolio of another of Vietnam's commercial banks.

3.1. Sample selection and variable identification

To develop a CSM, all retail loans signed between 1992 and 2005 were extracted fromthe database of one of Vietnam's commercial banks. This loan population contains stilloutstanding as well as repaid mortgages, consumer loans, credit card loans or business loans toborrowers from all over Vietnam. The bank classifies loans with more than 90 days of paymentdelay or at least three consecutive payment delays as defaulted. Defaulted loans account for4.9% in the total population. Currently, the bank records more than 30 loan characteristics.

Page 10: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Table 4Descriptive statistics of the Vietnamese retail loan sample

Panel A: Variables initially considered for the Vietnamese retail credit scoring model

Variable Categories and their default frequencies

Income(in million VNDper month)

0.5 to 1.5 1.6 to 3.5 3.6 to 8.0 morethan 8.0

other

7.93% 1.51% 0.29% 0.13% 4.07%Education post

graduateuniversitygraduate

collegegraduate

high schoolgraduate

non-high schoolgraduate

other

0.30% 1.23% 9.02% 2.40% 3.96% 1.48%Occupation farmer,

housewife,student

entrepreneur,farm owner,consultant / broker

doctor,professional,researcher,lawyer,teacher

police,military,journalist

other

3.37% 0.23% 1.90% 6.07% 0.85%Employer type self-employed,

entrepreneurjoint-venture,foreign

joint-stock,state-owned

other

0.22% 0.22% 1.06% 3.98%Time with employer

(in years)0 to 2 3 to 5 6 to 10 11 to 20 21 to 35

2.31% 2.09% 2.72% 4.44% 3.37%Age (in years) 18 to 24 25 to 35 36 to 45 46 to 64 more than 65

1.98% 2.94% 3.27% 2.91% 4.66%Gender male female

3.80% 2.49%Region north centre south except

Ho Chi MinhCity

Ho ChiMinh City

0.32% 26.16% 2.14% 0.20%Time at present address

(in years)0 to 2 3 to 5 6 to 10 11 to 20 21 to 60 other

0.71% 0.68% 1.35% 2.95% 4.86% 34.89%Residential status home owner,

home not usedas collateral

home owner,home usedas collateral

living withparents

rental other

3.36% 0.64% 2.94% 2.36% 0.71%Marital status married single widowed,

divorcedother

3.40% 2.19% 2.27% 5.75%Number of

dependants0 1 2 3 more than 3

1.06% 3.06% 2.46% 4.33% 6.69%Home phone yes no

0.92% 8.53%Mobile phone yes no

0.40% 4.41%Loan purpose business house collateralized general

creditcredit card

5.05% 0.95% 11.96% 1.29% 0.40%

480 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

Page 11: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Table 4 (continued)

Panel A: Variables initially considered for the Vietnamese retail credit scoring model

Variable Categories and their default frequencies

Collateral type real estate mobile asset fixed asset,machine

no collateral

0.41% 7.13% 2.86% 7.30%Collateral value

(in million VND)0 to 50 51 to 100 101 to 500 501 to

1000more than 1000

5.57% 4.62% 0.68% 0.24% 0.20%Loan duration

(in months)lessthan 13

13 to 24 25 to 36 37 to 48 more than 48

2.37% 1.59% 3.12% 7.35% 2.16%Time with bank

(in years)lessthan 13

13 to 24 25 to 36 more than 36

2.38% 5.62% 4.30% 0.95%Number of loans 1 2 3 more than 3

7.94% 1.49% 0.74% 0.15%Current account yes no

1.34% 4.22%Savings account yes no

0.11% 3.54%

Panel B: Loan types

Loan purpose Number of loans Loan size (in millions of VND)

Average Standard deviation

Business loan 19,956 127.3 385.0Mortgage 6759 243.7 98.9Collateralized loan 3673 54.2 522.1General credit 21,798 26.6 34.3Credit card loan 3851 158.8 208.9

Panel C: Differences in quantitative variables between defaulted and non-defaulted loans

Average characteristic Defaulted loans Non-defaulted loans

Income (in million VND per month) 1.85 8.36Time with employer (in years) 13.38 11.93Age (in years) 46.52 43.25Time at present address (in years) 28.68 18.95Number of dependants 3.51 2.88Collateral value (in million VND) 38.24 330.92Loan duration (in months) 61.91 28.22Time with bank (in years) 15.15 10.60Number of loans 1.26 3.66

In Panel A, for each variable the categories are shown in the top rows followed by the default probabilities in the bottomrow in italics. In Panel B, the number of loans describes our full sample of 56,037 loans which can be either repaid or canstill be outstanding. The loan size indicators are based on outstanding loans only.

481T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

There are, however, many missing values. Using all 30 variables is not feasible as less than10% of the applicants have full information for all characteristics. In order to obtain asufficiently large sample of loans with complete information we rely on the expert knowledgeof the loan officer and identify 22 relevant variables as listed in Panel A of Table 4 with

Page 12: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

482 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

relatively few missing values.10 This leads to a sample of 56,037 loans. This sample contains3.3% defaulted loans which is only slightly lower than 4.9% in the total loan population.11

This could imply that a borrower who sufficiently fills in the loan application form has a lowerPD than someone who does not properly complete the form. This is also consistent with one ofstrategies used in practice for coping with missing values as stated by Henley (1995): “a refusalto answer a particular question may be indicative of greater risk”. Our sample of 56,037 loansis thus based on 22 variables where the criteria for selecting these variables are based on theexpert knowledge. These criteria can overrule the requirement of no missing values and oursample thus contains the category ‘other’ which covers missing values.

As can be seen in Panel A of Table 4, the bank distinguishes five borrower groups based on theirloans' purpose: Customers borrowing money to finance their business (business); customersborrowing money for purchase or maintenance of their house (house); customers borrowingmoney to finance mobile assets such as the purchase of a car or motorbike where the asset serves ascollateral (collateralized); customers borrowing money for living expenses or consumptionwithout collateral (general credit); and customers borrowing money via the bank's credit card(credit card). Thus, our sample includes consumer as well as business loans. In Vietnam in generaland in our bank in particular, these latter loans are used to finance relatively small and privatebusinesses. Due to the lack of reliable up-to-date financial data on these borrowers and due to thefact that the boundary between the private and business property of an entrepreneur is often vague,these loans can only be assessed based on the personal information of the entrepreneur. In terms ofloan numbers, Panel B of Table 4 shows that business loans and general credits are the mostfrequently occurring loan types which account for 36% and 39% of all loans, respectively. Theother three loan types account for only 5% to 10% of all loans, each. In terms of value, however,housing loans constitute a larger share of the bank's loan portfolio due to their large average size of243.7 million Vietnamese Dong (VND). Furthermore, the variations in loan size within one loantype category can be substantial. Whereas the relatively small standard deviation for mortgagesand general credits indicates that these are more homogeneous loans with respect to size,collateralized loans and business loans are more heterogeneous. Furthermore, default varies byloan purpose ranging from as little as 0.40% for credit card loans to 11.96% for collateralized loans(see Panel A). The low default rate for credit card loans is driven by credit rationing as the bankreserves these loans for their best customers. The high default rate of collateralized loans reflectsthat fact that collateral is a signal for high risk and thus correlated with the risk characteristics of thecustomer such as age or income. Comparing defaulted and non-defaulted loans based on Panel C ofTable 4 reveals further differences. Most strikingly, defaulted borrowers have lower income,longer-term loans, and less collateral. They have been the bank's customer for a longer time butreceived fewer loans during this period. The negative correlation between income and loanduration is common in retail lending as a loan of a given size can only be repaid over a longerperiod of time when the borrower's income is small. The difference in collateral value can be

10 We thus exclude the following characteristics: Credit line, living expense, frequency of principal payment, frequencyof interest payment, outstanding amount in savings account, outstanding amount in debit account, sub-income andnumber of days in arrear.11 The population PD of 4.9% is in line with the PD reported on banks’ financial statements but rather far below thealternative figures of 35% to 70% reported by the World Bank. For the purpose of this study we assume that the dataprovided to us by the bank accurately records all defaults. As long as any potential mis-reporting is not structurallyrelated to borrower characteristics (i.e., the variables included in our CSM), this assumption is acceptable as it does notaffect the estimation of the model but would only lead to an over-statement of the predictive accuracy of the model.

Page 13: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

483T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

explained by the fact that defaults for mortgages, which have very valuable collateral, is rarewhereas defaults for collateralized loans, where less valuable assets such as machines are posted ascollateral, are much more frequent. When discussing the coding of the variables in Section 3.2. wewill investigate in more detail the relationship between borrower characteristics and default.

Finally, note that because the sample contains only information of loan applicants who wereaccepted, it is not representative of the population of future applicants. We also cannot perceivehow those applicants who were rejected would have performed if they had been accepted. Asolution is the out-of-sample calibration of the model described in the previous section. Wetherefore divide the 56,037 loans into two sub-samples. The initial sample contains 30,994 loansof which 1026 (3.3%) are in default and the hold-out sample contains 25,043 applicants of which798 (3.2%) are in default.

3.2. Variable coding, selection and model estimation

To estimate the CSM, the forward stepwise method is used to select among the 22 initialvariables. This method starts with a model without any independent variables and sequentiallyadds variables. At each step, the variable that leads to the greatest improvement in predictiveaccuracy – in terms of the highest score statistic conditional upon a significance level of less than5% – is added. The process continues until no variable with a significance level of less than 5%can be found. Based on the forward stepwise method, 16 of the initial 22 variables are included inthe CSM. To ensure that the selected variables are the most powerful predictors, the backwardstepwise method is applied as well. Here the starting point is a model that contains all 22 initialvariables. At each step, the method eliminates the weakest variable so that only the strongestpredictors are kept for the final CSM. As expected, the backward stepwise selects the same 16variables as the forward stepwise method. The six remaining variables are excluded due to one oftwo reasons. First, they have insignificant coefficients and do not contribute to the explanation ofthe dependent variable's variance. Second, they are eliminated because of their correlation withincluded variables. Together with adding the most predictive variables to the model, at every stepof the backward stepwise regression, the tolerance (1−Ri

2) of every excluded variable iscalculated. Those variables with tolerances of less than 80% (i.e. 20% or more of the variance inthe variable is explained by variation in the other variables) are considered for deletion. In the caseof multicollinearity, the inclusion of all variables leads to inferior results in the hold-out sample.

The 22 variables initially selected for inclusion in the CSM overlap with the list of commonlyused variables of Table 1. In line with our general discussion in Section 2.2., we find that ourvariables are not unique for Vietnam in particular or even for developing countries in general.Exceptions are gender and region (similar to Schreiner's (2004) branch variable) which seem to beexclusively used in developing countries. When categorizing and coding the variables, however, wetake the specific circumstances in Vietnam into account. The categories as well as the associated PDof all 22 variables can be found in Panel A of Table 4. Outlined below are the details regarding thesecategories, the specific circumstances in Vietnam, and the reasons why we expect these borrowercharacteristics, i.e. those captured by the finally selected 16 variables, to be relevant.

Regarding education we expect that better educated people have more stable, higher-incomeemployment and thus a lower PD. We therefore distinguish borrowers by their educational degreeranging from post-graduate to non-high school graduate. In 2002, however, less than 0.5% of theVietnamese working population held a graduate or post-graduate degree and only 13% held anundergraduate degree (GSO, 2002). The largest share of the working population thus falls into thelower two educational categories. Our loan sample broadly reflects these demographics but

Page 14: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

484 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

contains more highly educated borrowers: 30% of borrowers hold a post-graduate, graduate, orundergraduate degree, 39% hold a high-school degree, and 31% have less or other education.Default frequencies do however not consistently decline with increasing education. Collegegraduates, for example, have the highest default frequency of 9.02%.

Gender can no longer be included in CSMs for many industrialized countries as it is deemeddiscriminatory. In contrast, Schreiner (2003) argues that fair discrimination – for example basedon the statistical default rates of men versus women – is acceptable as it is based on quantifiabledata. The alternative – subjective scoring – discriminates equally if not more. Overall, there isample evidence that women default less frequently on loans (Schreiner, 2004) but most of thisgender effect disappears when other risk factors that are correlated with gender are accountedfor.12 We thus initially include gender and rely on the variable selection procedure to determinewhether gender remains in the final CSM. In Vietnam there is furthermore a specific relationshipbetween gender and income which in turn might be predictive of default. The GSO (2004) reportsa higher income for women than for men. For wage earners, the average monthly income percapita per month is 228 thousand VND if the head of the household is a woman compared to 139thousand VND if the head of the household is a man. The same holds on aggregate for all incomesources but here the difference between women and men is less pronounced with 589 versus 489thousand VND, respectively.13

Region represents the area of the country where the borrower lives. As people of similar wealthtend to live in the same location (a suburb might attract richer residents and the resulting increasein housing and property prices make this suburb prohibitively expensive for poorer households),this geographic criterion can indicate a borrower's level of financial wealth. A typical proxy is anarea's postal code but in Vietnam a postal code is not part of the address. Instead, we approximatethe region with the branch where the loan is issued. According to the bank's policy, borrowers canonly obtain loans from their local branches and branch location therefore coincides with theborrower's residential area. The numerous branches were categorized into four regions of thenorth, center, south (excluding Ho Chi Minh City) of the country, and Ho Chi Minh City. Notethat given this coding, region might not only be a proxy for the borrower's wealth but also reflectthe current credit assessment ability of the different branches. In our sample, it appears thatborrowers from Ho Chi Minh City are the best borrowers and those from the center are the worst.This can be understood in terms of the average income per person which is highest in Ho ChiMinh City, followed by the north and south, while the center is the poorest area in Vietnam (WorldBank, 2004).

Time at present address represents the number of years that the borrower has been living at hiscurrent address. Crook, Hamilton, and Thomas (1992) find that default risk drops with an increasein this variable. In this sense, time at present address might be a proxy for the borrower's maturity,stability, or risk aversion. However, this relationship does not hold in Vietnam where default ratesincrease with the time at the present address. In Vietnam, people who acquire financial wealthtend to seek better living conditions and thus often move to a new home in a better area. Thus,changing address might be a signal that a borrower's financial wealth is high and/or improvingrapidly. Under these conditions, he is better able to repay his loan.

12 For Bolivian microfinance, Schreiner (2004) shows that women in comparison with men are less risky borrowers asthey are more likely to be traders than manufacturers and to have smaller businesses that require smaller, shorter loans.13 Note that for this as well as all other binary variables, coding the variable based on Eq. (4) leads to a more precisemeasure than simply assigning values of 0 and 1.

Page 15: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

485T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

Residential status indicates whether borrowers own their home, rent, or live with their parents.Ex ante, the relationship between residential status and default is unclear. On the one hand,residential status can indicate financial wealth in particular in the case of home ownership. On theother hand, residential status can indicate increased financial pressure on the borrower's incomethrough insurance fees, taxes, or electricity cost. Crook et al. (1992) find that borrowers who aremost likely to default belong to the “other” category whereas borrowers living with their parentsare least likely to default. We expect that this ranking will be different for our sample since thereasons of having a certain residential status in Vietnam are dissimilar from those in industrializedcountries. Almost 90% of all borrowers in our sample own their homes. This properly reflects theVietnamese population as a whole where 95% of all households own their house. Houseownership is with 98% somewhat higher in rural areas compared to 86% in urban areas (GSO,1999). If this house is used as collateral, default rates are among the lowest whereas default ratesare highest when this house is not used as collateral. This can be explained by the importanceof owning a house in Vietnamese society, which leads to a strong aversion of many borrowersto loose their house.

Marital status can matter if it has an effect on the responsibility, reliability, or maturity ofborrowers. In our sample, however, default rates are higher for married than for single borrowers.This result can be due to the fact that marital status is typically related to number of dependantswhich in turn reflects financial pressure on the borrower and her ability to repay a loan. Anotherlinkage exists between marital status and age. The Committee for Population, Family andChildren (2003) finds that in 2002, Vietnamese women marry mainly in their 20s. Among 20- to24-year old women only 46% are married compared to 80% to 90% of older women. Thus, theage group which shows the lowest default rates (18 to 24 years) also has the most unmarriedpeople. However, as 87% of borrowers are married, it is unclear ex ante how informative maritalstatus will be in contrast to other variables such as age or number of dependents.

Number of dependants represents the number of people that the borrower has to support. Crooket al. (1992) separate this variable into two categories, the number of children and the number ofother dependants. In this study, we measure the total number of dependants though this variablemostly reflects the number children. As the number of dependants increases, so does the pressureon the borrower's income due to higher expenses such as school fees. In Vietnam as inindustrialized countries, the default rate increases steadily with the number of dependents. Forexample, moving from zero to three dependants quadruples the default risk from 1.1% to 4.3%.Note that our categories from zero to more than three dependents are reflective of the typicalhousehold size in Vietnam. In 2002, 57% of all households had four or less members (thus threedependants in addition to the borrower). Larger households have commonly five or six memberswhile very large households are with less than 10% rare in Vietnam (Committee for Population,Family and Children, 2003).

Home phone measures whether a customer has a home phone or not and can indicate howeasily the bank can keep contact with that borrower. Thus Crook et al. (1992) find that nothaving a home phone is associated with a higher default risk. We expect this relationship to bepresent in Vietnam – possibly stronger than in developed countries – as the percentage ofhouseholds with a home phone was only 27% in 2004. There are however big differencesacross areas: 74% of urban households have a telephone whereas only 11.5% of ruralhouseholds do. In Vietnam, home phone and default are not only linked due to the ability tocontact the borrower but also due to the fact that home phone and income are correlated. Whileonly 16% of households in the lowest 20% of the income distribution have a telephone, 74% ofhouseholds in the highest 20% of the income distribution have a telephone (GSO, 2004). In our

Page 16: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

486 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

sample the percentage of borrowers with a home phone lies with 70% above the populationaverage.

As the first two variables that measure the borrower's banking relationship, loan purposedescribes how the funds are used and collateral typemeasures what type of collateral supports theloan. Collateral type is incorporated in by Schreiner (2004) who classifies guarantees as none,personal, multiple, or other guarantee and Viganó (1993) who includes dummies for pledges,sales or personal guarantees. Our categorization is different due to specific features of theVietnamese retail banking market. Next to non-collateralized loans, we distinguish between realestate assets, mobile assets and fixed assets. Given the state of Vietnam's legal system, banks donot accept personal guarantees as these are difficult to enforce. In our sample there is clear relationbetween loan type and collateral type, partly due to the overlapping variable definitions.Generally speaking, requiring collateral is a signal of risk. Most defaults occur in thecollateralized or business loan class. For business loans, 97.52% are collateralized with realestate, 1.4% with fixed assets such as machines, 0.49% with mobile assets and 0.06% with othercollateral. Only 0.53% of business loans are not collateralized. At the other end of the distribution,credit cards do not require collateral and default rates are low. This can be due to the fact that acredit card is still very new product, which is offered only to those borrowers who have a veryclose and good relationship with the bank. The exemptions to the rule are housing loans. Though75% are collateralized, less than 1% is in default. This might indicate borrower's risk aversionregarding the loss of his house and is also consistent with Vietnamese culture in which peopleconsider their houses as a very important feature of their life. Finally, there is an overlap betweenthe collateral type and residential status as only borrowers who live in their own home can use thisas collateral. Note however that many borrowers, who own more than one property, use a propertythat they do not live in as collateral for a (business) loan. This could again be a strong signal forthe cultural importance of owning your home. As a related proxy, collateral value reflects thevalue of collateral in millions of VND. The higher its value, the higher the incentive for theborrower, who does not want to loose her collateral, to repay.

Loan duration measures the maturity of a loan in months. Usually, this is a feature of the loanand as such the outcome of the negotiation between borrower and bank and is thus excludedfrom CSMs. However, the Vietnamese situation is different. What we measure here is the loanduration as proposed by the borrower and not as negotiated between the bank and borrower. Thus,this variable reflects the borrower's intention, risk aversion, or self-assessment of repaymentability.

Time with bank measures the length of the banking relationship in years. In the context ofrelationship lending, it can be assumed that the longer a customer stays with a bank, the morethe bank knows about him and the lower the default risk becomes. Whereas this is confirmedfor industrialized countries (Crook et al., 1992), it does not quite hold in Vietnam. The highestdefault rate can be found for those borrowers who have been with bank for 13 to 36 yearswhereas borrowers who have been with bank both shorter and longer are less likely to default.This could reflect the fact that the Vietnamese banking market is being reformed and that creditofficers still have room for preferential credit allocation. As reform progresses and CSMs areput in place, the effect of this variable is expected to change and should thus be updatedregularly.

Number of loans counts the number of loans a customer has received during her wholerelationship with the bank. Many borrowers have not only a sequence of historical loans but alsohave more than one loan at a time. As a defaulted borrower has difficulties in receiving a newloan, this proxy can be informative about default risk. As expected, default is least frequent for

Page 17: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Table 5The estimated credit scoring model

Included variables Estimated coefficient Standard error Significance level

Time with bank − 1.774 0.121 0.0%Gender − 1.557 0.222 1.0%Number of loans − 0.938 0.051 1.4%Loan duration − 0.845 0.080 3.7%Savings account − 0.750 0.104 3.1%Region − 0.652 0.030 13.6%Residential status − 0.551 0.278 44.6%Current account − 0.492 0.208 10.4%Collateral value − 0.402 0.096 9.8%Number of dependants − 0.356 0.096 9.9%Time at present address − 0.285 0.054 2.5%Marital status − 0.233 0.101 68.1%Collateral type − 0.190 0.057 53.0%Home phone − 0.181 0.047 3.4%Education − 0.156 0.067 60.3%Loan purpose − 0.125 0.054 3.3%Constant − 3.176 0.058 4.6%

487T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

repeat borrowers: Whereas 7.94% of first-time borrowers default, only 0.15% of borrowerswith three or more loans do so. This variable thus clearly reflects how important the bank–borrower relationship, i.e. a good credit record, is.

Current account is a binary variable that indicates whether a borrower holds a current account.In Vietnam, the concept of a personal bank account is still unfamiliar to most of the population. In2006, only about 6% of the population has a bank account. With 35%, this percentage is howeverhigher for the urban middle class and has been increasing substantially from only 12% in 2001.14

Having a bank account is mostly an indicator of modern lifestyle among young people and thisvariable can thus be indicative of education or financial wealth (see also Crook et al., 1992). In oursample, 34% of borrowers have a current account at the time of the loan request and these haveindeed a lower default risk. The high growth in current accounts might substantially change therelationship between default and current account ownership in the future. Consequently, thisvariable needs to be updated regularly if included in a CSM.15

Table 5 shows the estimated CSM and reveals that among our 16 variables time with bank isthe most important predictor, followed by gender, number of loans, and loan duration. Com-pared to Crook et al.'s (1992) CSM for the UK, six out of their 12 predictors are incorporated inour CSM: time with bank (ranked number 1st here versus 3rd by Crook et al.), savings account(5th vs. 9th), region (6th vs. 1st), residential status (7th vs. 6th), current account (8th vs. 4th),and home phone (14th vs. 7th). Compared to Crook's longer list of 24 most typical variablesthere is an overlap with nine of our variables. As such, our model reflects – to some extent – the

14 Figures are obtained from two surveys conducted in 2006: (1) A survey conducted by Visa International as reported inthe Thanh Nien News on January 8, 2007. (2) TNS's VietCycle survey as reported in the Vietnam Investment Review No795 on January 8, 2007.15 We also consider income, occupation, employer type, time with employer, age, and the ownership of a mobile phone.As these variables are not part of our final CSM, they are not motivated here in detail. Their categories and defaultfrequencies are, however, listed in Panel A of Table 5 for illustrative purposes.

Page 18: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Table 6Predicted versus observed probabilities of default in the hold-out sample

Panel A: Default probabilities by loan group

Group Number of loans Probability of default (in %)

Predicted

Observed Minimum Average Maximum

1 5091 0.01 0.00 0.01 0.022 4956 0.08 0.03 0.06 0.123 5007 0.27 0.13 0.24 0.414 4929 1.11 0.42 0.77 1.795 5060 49.31 1.80 15.41 96.81total 25,043 3.19 0.00 3.33 96.81

Panel B: Predictive accuracy with a cut-off of 0.50

Observation Prediction

Non-default Default PCC

Non-default 24,136 109 99.55% = PCCgood

Default 397 401 50.25% = PCCbad

97.98% = PCCtotal

SENS 98.38%SPEC 78.63%

Panel C: Predictive accuracy with an optimal cut-off of 0.1995

Observation Prediction

Non-default Default PCC

Non-default 23,698 547 97.74% = PCCgood

Default 199 599 75.06% = PCCbad

97.02% = PCCtotal

SENS 99.17%SPEC 52.27%

Panel D: A credit scoring model with two cut-offs of 0.01 and 0.21

Observation Prediction

Non-default Marginal Default

Non-default 19,023 4,681 541Default 40 155 603Bg/B 5.01%Bb/B 75.56%

See Table 2 for definitions of the accuracy measures.

488 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

international norm of credit scoring. The inclusion of variables such as collateral type, loanduration, or gender is however unique to developing countries. Particularly, gender and loanduration are very effective predictors ranking 2nd and 4th in the CSM, respectively. Firstregarding gender, recall the above discussion whether gender is truly indicative of default orsimply reflects underlying risks. For Vietnam, we have to conclude that gender helps in theproper assessment of credit risk even when other risk factors such as loan purpose or collateralvalue are taken into account. As women are coded with a higher value than men in our CSM,the negative coefficient of gender implies that Vietnamese women default less frequently than

Page 19: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

489T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

men. Furthermore, we argued above that gender is indicative of income, a fact which canexplain the absence of more traditional income proxies (income, occupation, employer type ortype with employer) from our model. Second, loan duration, as a measure of the borrower'sintention, risk aversion, or self-assessment of repayment ability, is unique to the Vietnamesesituation. It indicates that Vietnamese banks cannot only rely on their own assessment of theborrower but also on the borrower herself, who appears to adequately and honestly state herown loan capacity. In this context, time with bank and number of loans as 1st and 3rd mostimportant predictors reflect the borrower's relationship with the bank and thus the role that suchrelationships play in Vietnam's loan market. To this extent, our CSM is consistent with the lessformalized credit assessment approach of another Vietnamese bank presented in Table 3, whichalso relies heavily on relationship proxies, i.e. in its second round of credit assessment.Regarding the accuracy of our CSM, its explanatory power increases from step to step leadingto a final adjusted R-square of 0.578. However, to properly assess the accuracy of the model, itshould be applied out-of-sample.

3.3. Model calibration and out-of-sample testing

When applied out-of-sample, we find as expected a significant difference in the predictedPD between observed defaulted and non-defaulted borrowers. The average (median) of thepredicted PD is 1.73% (0.2%) for non-defaulted loans compared to 49.05% (51.66%) for defaultedloans. However, the ranges of predicted probabilities are rather large and overlap for the two loangroups. For defaulted loans PD ranges from 0.01% to 96.81% compared to 0.00% to 73.54% fornon-defaulted loans. A closer look at Panel A of Table 6 reveals the observed versus predicted PDfor different sub-samples which are generated by sorting all loans of similar predicted PD into fivegroups of approximately equal size. By looking at the average estimated PD of 3.33% compared tothe observed PD of 3.19% we can see that our model slightly overestimates default. There is,however, a pattern: The PD of the low-risk groups 1 to 3 are underestimated whereas the maximumestimated probability is far beyond the observed PD. The average predicted PDs indicate thatdefault risk is rising steadily but only jumps above the 1% level for group 5. A more detailedaccuracy analysis is thus warranted using the PCC, SENS, and SPEC indicators.

Panel B of Table 6 reports these accuracy measures for an ad-hoc cut-off value of 50%.Regarding PCC, the overall accuracy is surprisingly high with almost 99%. Looking at defaultedversus non-defaulted loans separately reveals that this high accuracy is mainly driven by the non-defaulted loans as PCCgood (99.55%) is much higher than PCCbad (50.25%). For industrializedcountries, Desai et al. (1996) report accuracy values of up to 86.90% for PCCtotal and 41.38% forPCCbad. For Bolivia, Schreiner's (2004) different calibrations are able to achieve a maximumaccuracy for PCCtotal of 91%, PCCgood of 99%, or PCCbad of 71%. Overall, our results are in linewith these findings. The sensitivity measure implies a PD for the bank's loan portfolio of (100%—SENS)=1.62% because all applicants with a predicted PD of less than 50% will be accepted and98.38% of these are expected to repay the loan. Our SENS accuracy is slightly higher than thatreported in other studies. For example, Baesens et al.'s (2003) survey reports sensitivity measuresof up to 96.36%. Finally, SPEC indicates that among the rejected loans (with a predictedPDN50%), 78.63% would have actually defaulted whereas 21.37% would have been repaid. Notethat SPEC is an abstract measure since in practice a bank cannot check whether its rejectedloan applicants would have defaulted or not. The 21.37% incorrectly rejected loans should beinterpreted as opportunity cost for the bank. In contrast, the 1.62% incorrectly accepted loanscreate actual cost for the bank.

Page 20: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

490 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

The CSM's calibration, i.e. the determination of the optimal cut-off, depends on the preferencesof the bank. Our bank – though it does not have a CSM in place but relies on subjective creditassessment – has set a target for non-performing loans. Assuming that this target is 0.83%, we haveto calibrate the CSM to a sensitivity of 99.17% (SENS=100% – 0.83%=99.17%). At the firstsight, one might think that in order to achieve this target, bank should only accept customers whohave predicted PD of less than 0.83%. Such a cut-off would however lead to a SENS of 99.33% andthus an expected PD of 0.77%. Instead, a cut-off of 19.95% as shown in Panel C of Table 6achieves the desired result. Comparing the accuracy of this cut-off with that of the ad-hoc cut-off of50% reveals the following dynamics: By sacrificing a small amount of good loans (PCCgood dropsfrom 99.55% to 97.74%), the number of incorrectly accepted bad loans increases substantially(PCCbad increases from 50.25% to 75.06%). In terms of SENS and SPEC, the improvement by0.79% in sensitivity is achieved at a cost of a 26.36% reduction in specificity.

4. Credit scoring and bank strategy

In the previous section, the cut-off value of 19.95% is chosen for the bank to achieve itsassumed default target of 0.83%. Is this optimal if the bank follows other strategies such asminimizing the time spent on credit assessment or maximizing return? In order to answer thisquestion, we look at alternative ways to calibrate the CSM.

4.1. Profitability

A CSM's benefit can lie in the effect that the new credit assessment policy has on the bank'sprofitability. Schreiner (2003) suggests measuring the change in profits in terms of opportunitycost. In particular he focuses on the ratio of good loans lost (Gb) per bad loans avoided (Bb), whichis to some extent related to SPEC. Panel A of Fig. 1 shows how this ratio changes for different cut-off values and Panel B calculates the effect on profit as

Dprofit ¼ cost per bad loan ⁎ Bb � benef it per good loan ⁎ Gb ð5Þ

For the cut-off of 19.95% identified as optimal in the previous section, the tradeoff is foregoing0.91 good loans for each bad loan that is avoided. In terms of profit changes, we assume differentrelationships between the cost per bad loan and benefit per good loan. Data for developing countriesis not available and we therefore take the evidence from the credit card market in industrializedcountries as a benchmark. Here, practitioners argue that it takes at least 10 good loans to cover thecost of one bad loan (Schreiner, 2003). Under this assumption, a cut-off of 19.95% leads to anincrease in profits, which would indicate that our bank's target of 0.83% non-performing loans iswell set. Overall, however, the change in profits is relatively stable for a cut-offs of 7% to 20%before falling more substantially. A further look at Panel B of Fig. 1 reveals that as the cost-benefit ratio becomes smaller, so does the effect on profits. With a cost-benefit ratio of 10-to-5,an increase in profits can only be obtained with cut-off value of 7% or higher.

4.2. Risk-based pricing

Economic theory suggests that in a competitive market, prices will equal marginal cost. If themarket price is higher, high profit margins will induce competitors to enter the market and biddown the price. This principle implies that each loan should be priced in accordance with the cost

Page 21: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Fig. 1. Credit scoring and bank profitability.

491T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

of funds, origination and servicing costs, and the implicit costs associated with potential default.While the fixed costs do not generally vary within the same product category, expected defaultcost can vary across borrowers and should be calculated as PD times loss given default (LGD).The risk-based interest rate R could thus be derived in a simple model as

R ¼ cost of fundsþ overhead costþ expected default costþ profit margin ð6Þ

As Vietnam moves towards a more and more competitive retail banking market, borrowers willshop around and any bank offering too high interest rates for good borrowerswill loosemarket share.At the same time, banks offering too low interest rates for bad borrowers will experience erodingprofits. Risk-based pricing therefore seems to be necessary if a bank wants to survive in Vietnam'semerging banking market. Currently, however, our bank charges the same annual interest rate for allmedium- and long-term loans. Based on our CSM, we can suggest a simple pricing strategy.

Page 22: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

Table 7Risk-based loan pricing

Borrower class

Low risk Moderate risk High risk All borrowers

Range of predicted PD 0.00%–0.04% 0.05%–0.30% 0.31%–19.95% 0.00%–19.95%Number of loans 7424 7557 7842 22,823Overhead cost 1.00% 1.50% 2.00% 1.50%Cost of funds 9.00% 9.00% 9.00% 9.00%Profit margin 2.00% 2.00% 2.00% 2.00%Cost of default 0.10% 0.20% 2.00%Loan interest rate 12.10% 12.70% 15.00% 13.30%LD = average observed PD 0.13% 0.33% 1.98% 0.83%LGD 100.00% 100.00% 100.00% 100.00%Cost of default = LD ⁎ LGD 0.13% 0.33% 1.98% 0.83%Loan interest rate 12.13% 12.83% 14.98% 13.33%

The abbreviations are probability of default (PD) and loss given default (LGD).

492 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

As Table 7 shows, we split the bank's borrowers into three risk classes containing about 7500borrowers each. The predicted PD for the low risk class ranges from 0.00% to 0.04%, for themoderate risk class from 0.05% to 0.30%, and for the high risk class from 0.31% to 19.95%. Giventhe targeted maximum of 0.83% non-performing loans, any applicant with an estimated PD ofmore than 19.95% will be rejected. Based on our discussions with Vietnamese bankers, a 2.00%profit margin and an average 1.50% overhead cost appear to be reasonable and realistic as-sumptions. However, customers with higher risk usually create higher costs because they takemore of the credit officer's time. To be consistent with customer-specific pricing, overhead cost of2.00%, 1.50%, and 1.00% will be allocated to the high-, moderate-, and low-risk borrowers,respectively. The cost of funds can be approximated by the bank's deposit interest rate. In Vietnamin 2006, the annual deposit interest rate on local currency deposits is about 9%. Finally, theexpected cost of default PD⁎LGD has to be considered. Our CSM provides estimates for PD but itis beyond the scope of this study to estimate LGD. LGD will be at least loan-type specific sinceeach type of loan is associated with a different type of collateral. Default cost can be quitesignificant. As a benchmark note that in 1997 risk-premia for US consumer loans ranged from aslow as 0.10% to more than 2.00% for the highest risk borrowers (Sangha, 1998). In line with thisstudy, we assign 2.00% default cost to the high-risk borrowers, 0.20% to the moderate-risk and0.10% to the low-risk borrowers. As Table 7 shows, the bank could charge 12.10%, 12.70%, and15.00% to the different risk classes. If, on the other hand, the bank decides to charge the same priceto all borrowers, it could charge an average interest rate of approximately 13.30% to earn the sameprofit margin. As the second set of calculations shows, these assumptions would be consistent witha LGD of 100% and a PD based on the realized PD of the risk class. In practice, PD could be basedon the historic PD obtained from the loan portfolio which is used when estimating the CSM.16

4.3. Transactional versus relationship lending

Should banks rely exclusively on their CSM, i.e. transactional lending, or should there be roomfor relationship lending? The benefits of credit scoring are clear. Mester (1997), for example,

16 For an in-depth analysis of the interaction between risk-based pricing, profit-maximizing cut-offs, and adverseselection problems see Blöchlinger and Leippold (2006).

Page 23: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

493T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

reports that credit scoring has substantially reduced the length of the loan approval process fromtwo weeks to 12.5 h for small-business loans in the US, or from nine to three days for consumerloans in Canada. Once a CSM is in place, these benefits can be achieved at a cost of $1.50 to $10per loan. In order to reduce the cost, time, and effort that credit officers spend on loan assessment orto allocate officers' time more productively, banks might thus be inclined to implement a strictcredit policy based on their CSM alone. In industrialized countries, internet banks rely onautomated loan systems which include CSMs. Mester (1997) reports that Banc One exclusivelyuses credit scores for loans up to $35,000 and approves a further 30% of all loans below $1 milliononly based on scores. Furthermore, a regional bank in Pennsylvania assesses all small businessloans under $35,000 based on the CSM alone. The downside, however, can be substantial.Schreiner (2003) recalls a Bolivian example of a finance company that relied exclusively on itsCSM and went bankrupt.

Credit scoring is based on quantifiable criteria and ignores qualitative risk. Its underlyingassumption is that qualitative risk is either irrelevant or not measurable. Is this assumptionreasonable? If qualitative risks matter but are not captured indirectly via the quantitative variablesin the model, the estimated PD will be too low.17 To the extent that credit officers can discover thisqualitative risk through their relationship with the borrower and to the extent that this risk is notalready reflected in variables measuring the bank–customer relationship such as ‘time with bank'or ‘number of loans', a combination between transactional and relationship lending is valuable.Credit scoring can thus either precede or follow traditional loan assessment. Schreiner (2003)argues that banks should apply their CSM only to loans that are already conditionally approved by thecredit officer. In contrast, Mester (1997) reports that banks in the US have credit officers reviewing theCSM's decision and that many banks focus on those loan applicants with scores close to the cut-offpoint. For the First National Bank of Chicago this led to a substantial change in the credit decisions forsmall-business loans. About 25% of the applications rejected by the CSM were later approved andanother 25% of the applications accepted by the CSMwere later rejected by the credit officer. Based onour hold-out sample, we can only illustrate this second approach.18 In particular, we will illustrate howour CSM can be calibrated for two cut-offs C1and C2 with 0bC1bC2b1. Such a calibration allowsbanks to define loans that need some further examinations instead of rejecting or accepting themdirectly. All loan applicantswith an estimated PDbC1 are accepted outright whereas all applicants withPDNC2 are rejected outright. The remaining loans with (C1bPDbC2) are classified as marginal loanswhich are consequently reviewed by the credit officer. The bank's credit assessment can thus be acombination of the transactional lending via CSM and relationship lending.

Panel D of Table 6 reports the results of such a CSM. C1 and C2 are determined according to asomewhat ad hoc decision rule which can be interpreted in the context of PCC. This rule specifiesthat (1) the proportion of incorrectly classified bad loans should be less than 5% of all bad loans,i.e. Bg/Bb5%, and (2) that the proportion of correctly classified bad loans should be at least 75%of all bad loans, i.e. Bb/B≥75%. The emphasis on bad loans is justified by the relatively highpenalty costs associated with over looking a potentially bad loan. The trade off is the gain fromreducing the number of good loans subjected to evaluation. If evaluation cost were available, C1

and C2 can be determined so that total evaluation cost is minimized. For an even more advancedapproach, evaluation cost can be included in the model presented in Section 4.1. and C1 and C2

can be chosen based on a profit maximization rule. However, since these costs have not yet been

17 See Schreiner (2003), Box 4 for a more in-depth discussion.18 Note however that Schreiner's (2003) approach is to some extent implicit in our sample as it is restricted to approvedloans and does not cover all loan applications.

Page 24: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

494 T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

precisely estimated by our bank, the above decision rule has been applied resulting in the cut-offpoints C1=1% and C2=21%. The model predicts 19,063 good, 4836 marginal and 1141 badloans. Regarding the actual bad loans, 40 were classified as good, 155 as marginal and 603 as bad.The implication for credit assessment is that the 40 loans will not be evaluated by the credit officerbut incorrectly accepted outright. The 155 loans will be examined together with the 4681 othermarginal loans. If the credit officer can correctly spot these 155 loans, the bank would reject 758of the 798 actual bad loans. In contrast, the CSM with a single cut-off of 19.95% (50%) leads to arejection of only 599 (401) of the observed bad loans. Regarding the observed good loans, ourtwo cut-off points do not lead to any substantial change compared to the models with one cut-offpoint. This is due to the criteria imposed on the cut-off points. Assuming again that the creditofficer can correctly spot the 4681 observed good loans among the marginal loans, the bankwould accept 23,704 of the 24,245 observed good loans. In comparison, the CSM with one cut-off at 19.95% (50%) achieves a correct prediction of 23,698 (24,136) loans.

5. Conclusions

As banking markets in developing countries mature, banks not only benefit from retail creditgrowth but also face increasing competition and regulatory attention to risk management. Cancredit scoring help in these circumstances? Based on our evidence from the Vietnamese retailloan market, the answer is ‘yes’. First and foremost, a CSM can reduce loan default. Be replacingits informal credit assessment method with a CSM, our bank can expect a decrease in its defaultratio from 3.3% to 2%. In addition, banks can use a CSM to implement risk-based pricing tomanage its loan portfolio composition. By setting a lower interest rate for less risky borrowers, abank will be more competitive in this loan market segment and be able to attract more low-riskborrowers. We have also shown how banks can calibrate the CSM to manage profits. Yet, in thecase of our Vietnamese bank, these effects were small. Finally, a CSM reduces the time and thuscost spent by the loan officer on loan assessment. In contrast to corporate or government loans,retail loans are small and their credit assessment is thus more costly per $ or VND of loanvolume. A CSM can thus be calibrated with this consideration in mind. Overall, banks canbenefit from reduced loan default, increased competitiveness, and reduced cost of loan assess-ment while optimally using its relationship lending tools, i.e. its loan officer's knowledge aboutthe borrower. In order to achieve these benefits, it is, however, important to align a CSM to thelocal market as the predictive borrower-characteristics and their default probabilities arecountry-specific.

References

Allen, L., DeLong, G., & Saunders, A. (2004). Issues in credit risk modeling of retail markets. Journal of Banking andFinance, 28, 727−752.

Altman, E. I. (1968). Financial ratios, discriminant analysis, and the prediction of corporate bankruptcy. Journal ofFinance, 23, 589−609.

Altman, E. I., & Narayanan, P. (1997). An international survey of business failure classification models. FinancialMarkets, Institutions and Instruments, 6, 1−57.

Altman, E. I., & Saunders, A. (1998). Credit risk measurement: Developments over the last 20 years. Journal of Bankingand Finance, 21, 1721−1742.

Baesens, B., Gestel, T. V., Viaene, S., Stepanova, M., Suykens, J., & Vanthienen, J. (2003). Benchmarking state-of-the-artclassification algorithms for credit scoring. Journal of the Operational Research Society, 54, 627−635.

Blöchlinger, M., & Leippold, M. (2006). Economic benefit of powerful credit scoring. Journal of Banking and Finance,30, 851−873.

Page 25: Credit Scoring Model foCredit scoring model for VN retail bankingr VN Retail Banking Mkt_Huyen 2007

495T.H.T. Dinh, S. Kleimeier / International Review of Financial Analysis 16 (2007) 471–495

Boyle, M., Crook, J. N., Hamilton, R., & Thomas, L. C. (1992). Method s for credit scoring applied to slow payers. In L. C.Thomas, J. N. Crook, & D. B. Edelman (Eds.), Credit scoring and credit control (pp. 75−90). Oxford: OxfordUniversity Press.

Byström, H., Worasinchai, L., & Chongsithipol, S. (2005). Default risk, systematic risk and Thai firms before, during andafter the Asian crisis. Research in International Business and Finance, 19, 95−110.

Committee for Population, Family and Children, ORC Macro, 2003. Vietnam Demographic and Health Survey 2002.Committee for Population, Family and Children, Hanoi, http://www.measuredhs.com/aboutsurveys/search/

Crook, J. N., Hamilton, R., & Thomas, L. C. (1992). A comparison of discriminations under alternative definitions ofcredit default. In L. C. Thomas, J. N. Crook, & D. B. Edelman (Eds.), Credit scoring and credit control (pp. 217−245).Oxford: Oxford University Press.

Desai, V. S., Crook, J. N., & Overstreet, G. A. (1996). A comparison of neutral networks and linear scoring models in thecredit union environment. European Journal of Operational Research, 95, 24−37.

Desai, V. S., Crook, J. N., & Overstreet, G. A. (1997). Credit scoring models in the credit union environment using neutralnetworks and genetic algorithms. IMA Journal of Mathematics Applied in Business and Industry, 8, 323−346.

GSO. (1999). Population and Housing Census Vietnam 1999. General Statistics Office of Vietnam (GSO), Hanoi,http://www.gso.gov.vn

GSO. (2002). Establishment census of Vietnam 2002. General Statistics Office of Vietnam (GSO), Hanoi, http://www.gso.gov.vnGSO. (2004). Living standard survey 2004. General Statistics Office of Vietnam (GSO), Hanoi, http://www.gso.gov.vnHenley, W. E. (1995). Statistical aspects of credit scoring. Open University, Dissertation.Henley, W. E., & Hand, D. J. (1997). Statistical classification methods in consumer credit scoring: A review. Journal of the

Royal Statistical Society — Series A (Statistics in Society), 160, 523−541.Laitinen, E. K. (1999). Predicting a corporate credit analyst's risk estimate by logistic and linear models. International

Review of Financial Analysis, 8, 97−121.Loi, H. T. (2006). Credit risk and credit access in Asia: Trends and developments in insolvency system and risk

management: The experience of Vietnam. Source OECD Emerging Economies, 2006(8), 281−289.Mester, L. (1997 September/October). What's the point of credit scoring? Federal Reserve Bank of Philadelphia Business

Review, 3−16.Philosophov, L. V., & Philosophov, V. L. (2002). Corporate bankruptcy prognosis: An attempt at a combined prediction of

the bankruptcy event and time interval of its occurrence. International Review of Financial Analysis, 11, 375−406.Provost, F., Fawcett, T., & Kohavi, R. (1998). The case against accuracy estimation for comparing classifiers. Proceedings

of the Fifteenth International Conference on Machine Learning, 445−553.Sangha, B. S. (1998). A systematic approach for managing credit score overrides. In E. Mays (Ed.), Credit risk modeling

design and application (pp. 223−244). Chicago: Glenlake Publishing.Schreiner, M., 2003. Scoring: The next breakthrough in microcredit? CGAP Occasional Paper No. 7, http://www.cgap.orgSchreiner, M. (2004). Scoring arrears at a microlender in Bolivia. Journal of Microfinance, 6, 65−88.Srinivasan, V., & Kim, Y. H. (1987). The Bierman–Hausman credit granting model: A note. Management Science, 33,

1361−1362.Thomas, L. C. (2000). A survey of credit and behavioral scoring: Forecasting financial risk of lending to consumers.

International Journal of Forecasting, 16, 149−172.Viganó, L. (1993). A credit scoring model for development banks: An African case study. Savings and Development, 4,

441−482.Yobas, M. B., Crook, J. N., & Ross, P. (2000). Credit scoring using neural and evolutionary techniques. IMA Journal of

Mathematics Applied in Business and Industry, 11, 111−125.