Credit is a method of selling goods or services without the buyer having cash in hand. A credit card is only an automatic way of offering credit to a consumer. Today, every credit card carries an identifying number that speeds shopping transactions. Credit cards are an important fact of life.

A credit card is part of a system of payments named after the small plastic card issued to users of the system. The issuer of the card grants a line of credit to the consumer from which the user can borrow money for payment to a merchant or as a cash advance to the user.

Credit cards are approached two ways in the theoretical literature. A credit card may be classified as a medium of borrowing or as an alternate transaction medium. Although in reality the card-holder probably does use the card both as a means of taking short term loans and as a way to delay payment for reasons other than the lack of available funds (Amanada Swift King, vol. 43, pg 59)

Increasing money supply and availability of huge volumes of credit report data are giving rise to a new branch of analytic. Credit risk analytics is concerned with attaching a credit worthiness score to a borrower by performing statistical procedures. A credit score rates how risky a borrower the person is the higher the score, the less risk the person poses to creditors. Credit scoring models are developed by analyzing statistics and picking out characteristics that are believed to relate to creditworthiness. Different scoring models are used for different purposes. For example, Auto financing could employ a different model than installment loans.

Credit risk is the risk of loss due to a debtor's non-payment of a loan or other line of credit. Financial firms and banks had developed rating systems to determine an individual or company's credit worthiness. Creditworthiness has to do with the ability of a borrower to pay current debt in a timely manner. Within the context of the ability, several basic factors come into play. An evaluation of the creditworthiness of a borrower involves identifying the presence of resources that may be used to repay debts, the willingness of the debtor to use those resources for repaying debt, and a history of choosing to repay debt obligations in a timely manner.

The process of credit scoring is very important for banks as they need to separate `good borrowers' from `bad borrowers' in terms of their creditworthiness. Credit scoring is a structure in which creditworthiness of client is evaluated. The methods generally used for credit scoring are based on statistical pattern as well as judgmental decisions.

Credit scoring is a system creditors use to help determine whether to give you credit. Information about you and your credit experiences, such as bill-paying history, the number and type of accounts you have, late payments, collection actions, outstanding debt, and age of your accounts is collected from credit applications and your credit report.

Using a statistical program or judgmental decision, creditors compare this information to the credit performance of consumers with similar profiles. A credit scoring system awards points for each factor that helps predict who is most likely to repay a debt. Total number of points (credit score) helps predict how creditworthy you are; how likely it is that you will repay a loan and make payments when due (Avery, Bostic, Calem, and Canner, 1996; Freddie Mac, 1996; and Pierzchalski, 1996).

The growing importance of credit scoring in the allocation of mortgage credit led to the current debate about scoring's impact on the flow of credit to certain segments of the applicant population. Proponents of credit scoring and of the “automated underwriting” process that benefits from its use, argue that it lowers the overall cost of making credit available to consumers, while simultaneously increasing the speed and objectivity of the underwriting decisions (M. Cary Collins). For example, Calem and Wachter (1999) argue that scoring benefits lenders and borrowers alike by increasing the efficiency of the credit review process and reducing the likelihood of delinquency. Detractors of credit scoring models argue, however, that the underwriting variables employed and the weights assigned to each variable are based on the payment performance of traditional consumers. As such, scores generated by these models may not accurately portray the creditworthiness of underrepresented groups in the applicant pool, such as low-income and minority applicants groups that constitute a larger fraction of first-time home buyers. In particular, scoring models typically omit certain nontraditional indicators of credit performance, such as rent and utility payment histories, which are important components of credit performance for many low-income applicants (M. Cary Collins).

Credit scoring models calculate the score on the basis of the information contained in individual's credit report. Credit applications along with individual's occupation and length of employment are also taken into consideration while computing the credit score of an individual.

Credit scoring models, or score cards, can contribute to bank safety and soundness as long as bank management actively participates in the development, implementation, monitoring, and validation of the model. Credit scoring is particularly applicable to accounts where it is desired to keep the cost of the credit granting decision as low as possible, consistent with some predefined level of risk.


Credit reporting was born during the 19th century, when small retail merchants traded financial information about their customers. These "merchant associations" eventually organized into small credit bureaus. In the early days of credit reporting and scoring, as the 19th century came to an end and the prosperity heralded in by the second wave on the Industrial Revolution settled into the middle class, it was an informal system of information exchange between local retailers.

Credit scoring first materialized in the late 1950s to support lending decisions by the credit departments of large retail stores and finance companies. Statistically based credit decision-making systems were pioneered during the late 1950s but only saw mainstream use during the 1990s as the depth and breadth of electronic credit information increased (Edward M. Lewis 1994). These statistically based techniques are commonly referred to as “credit scoring” models. Initially, scoring models were employed in the consumer credit portfolios of most major banks and credit card issuers to increase the speed of the credit decision, enhance the uniformity of the decision process, and reduce the overall costs of decision making. The relative homogeneity of this type of credit and the wide availability of performance data on applicants made the initial implementation of scoring models by these lenders successful and profitable (M. Cary Collins)

The advents of credit cards in the 1960s, lenders have used credit scoring, both application and behavioral scoring to monitor and control default risk. Banks use credit scoring models when approving and pricing loans (Noel Capon spring 1982, p. 83).

By the end of the 1970s, most of the nation's largest commercial banks, finance companies, and credit card issuers used credit-scoring systems. Over these two decades, the primary use of credit scoring was in evaluating new applications for credit, and creditors used their own experience and data, sometimes with the aid of consultants, to develop the models. Although often available at the time from local credit bureaus (today more commonly referred to as credit-reporting agencies), credit history records were limited in scope and relatively expensive to access. Thus, lenders essentially had no practical way of obtaining the complete credit histories of non customers and so could not effectively target them for solicitations on the basis of credit history.

By the late 1980s much had changed. Creditors were no longer restricted to the credit histories of their own customers and credit applicants. Rather, lenders could purchase the generic credit history scores of individuals who were not their account holders and, with that data, market consumer credit products tailored to various credit scores to appropriate potential borrowers.

As the 20th century began to take hold, credit scoring was shaped by modern era inventions in transportation and communications, such as the automobile and the telephone. In a sense, as these became more common, the world became a smaller place. People could travel farther to do business and creditors could easily share information throughout a broader region.

The use of credit scoring then spread to additional loan products including home mortgage and small-business lending. Scoring technologies also were applied in new ways, such as in assessments by institutions of whether to purchase individual loans or pools of loans backing securities. Finally, credit-scoring technologies were developed to focus on outcomes beyond credit-risk assessment to include, for example, account profitability and various aspects of account management


Despite the explosion of banking services, lending is the core business of the commercial banks. From the mechanical viewpoint, the lending means to provide money temporarily on condition that the amount borrowed will be returned. Making loans are in the riskiest category of consumer loans and are typically sold in a market to individuals. In general it's a relatively series of actions involving two parties. These activities range from the initial loan application to the successful or unsuccessful repayment of the loan. Increases in the sum of loans also increase the probability of defaults. The primary problem of any lender is to differentiate between "good" and "bad" customers, means that customer who can pay debt from those who cannot pay debts. Such differentiation is possible by using a credit-scoring method.

The important of consumer credit in the U.S, economy has grown markedly through the 20th century. A combination of growth in supply and form of credit and increased consumer demand has led to an average annually compounded rate of growth in consumer credit outstanding of 7.5% from 1991. (Board of Governors of the Federal Reserve System 1976a) Despite the growth in credit availability, many consumers are unable to gain access to credit that they need and believe that they deserve.

Canner and Lucket (1992) reviewed the credit card market in the earl's 1990. They indicated that the behavior of credit card rates can be analysis from two sides. On the supply sides, card companies face relatively high operation cost and high risk of default. On the demand side they maintained that cardholders are relatively unresponsive to rate change.

The ever increasing ability to offer credit has important sales and profit implications for marketers, just as the ability to obtain credit has important quality of life implications for consumers. However, despite the growth in credit availability, many consumers are unable to gain access to the credit that they need and believe they deserve. The importance of this issue was recognized by Congress, which in 1974 passed the Equal Credit Opportunity Act prohibiting discrimination in the granting of credit on the basis of sex and marital status (ECOA 1975). In 1976 the Act was amended to include race, color, religion, national origin, receipt of income from a public assistance program, and age as proscribed characteristics. Further, in 1977, the Federal Trade Commission decided to devote a significant percentage of its then increased resources to the handling of all forms of credit abuse problems (Advertising Age 1977).

Credit scoring is particularly applicable to accounts where it is desired to keep the cost of the credit granting decision as low as possible, consistent with some predefined level of risk. It is very applicable, therefore, to high-volume small business accounts where the dollar amount is small and the risk is high. Using a credit scoring system for these types of accounts will tend to significantly reduce the cost of the credit granting decision. By design, credit scoring provides consistent, objective decisions, can identify potential high-volume customers and can enable increased credit limits to low-risk accounts

One of the aims of the credit scoring process is to predict whether an applicant will, if accepted, repay on time or not. This prediction is needed to decide which ones, of a large number of applicants, to grant credit and which ones to decline


2.1 Different Method of Scoring System:

2.1a: Judgmental Scoring System:

The ability to quantify credit risk, the risk that a borrower will not pay back a loan as agreed is central to the core aspects of lending, soliciting accounts, extending credit, pricing (that is, setting the interest rate or fees or other terms), and managing existing credit accounts. As noted earlier, systems in which the credit decision is made manually by a loan officer or other person are referred to here as judgmentalsystems.

A Judgmental System is basically a method of evaluating the credit worthiness of customers by using a formula or a set of rules based upon a combination of internal and external credit experiences. Realistically, companies that are not using a formal credit scoring system are using some version of a judgmental based system with the exception that the value placed on a given response is weighted subjectively by the evaluator, who after gathering the answers to a series of questions, applies judgment, based upon experience to come up with a ship/don't ship decision.

A judgmental scoring model is primarily based on traditional standards of credit analysis. It considers factors like bank and trade references, payment history, credit bureau ratings and financial statement ratios to produce the total credit score of the individual. Judgmental scoring model is comparatively easier to interpret and is almost free from ambiguity.

The conceptual framework for judgmental credit decisions has endured for many decades. This framework consists of the three "c's" of credit, character, capacity and capital, often joined by collateral and conditions, and indicated primarily by credit history and such other characteristics as income, occupation and residential stability (Noel Capon spring 1982, p. 83).Judgmental Scoring System is based on traditional standard of analyzing credit. This system is based fully on judgment that whether to grant loan to customer or not.

In a typical system a number of predictor characteristics were chosen for their ability to discriminate between those who repaid their credit (goods) and those who did not (bads), and points were awarded to different levels of each characteristic. An individual applicant was judged on the relationship between his/her summated score across characteristics and independently set accept/reject cut-off values. Early systems employed such characteristics as occupation, length of employment, credit bureau clearance, personal references, marital status, bank account, neighborhood, collateral, length of residence, income, rent, life insurance ownership, sex and race (Noel Capon spring 1982, p. 83).

This method is of approving or denying credit based on the lender's judgment rather than on a particular credit scoring model. Judgmental credit analysis entails evaluating the borrower's application and using prior experience dealing with similar applicants to determine credit approval. This process avoids using any algorithms or empirical process to determine approvals.

Traditionally, judgmental systems have operated under the assumption that the experiences loan officer or credit manager can use his professional instincts to screen credit risks. This flexible policy allows the managers to occasionally take a change on a poor risk applicant if the manager has a “hunch” the applicant will repay. Similarly, if all indices of creditworthiness are positive but the manager has an intuitive feeling that the loan should be denied, the loan is denied. It is this subjective elasticity which has made judgment systems attractive to many creditors. Several creditors who use judgmental systems consider informality and flexibility to be the system's major assets.

Judgmental-based scoring models only let you know what the ”quality of risk” is with a customer, they are ranking systems where the company with the highest score is considered the lowest risk and the company with the lowest score is considered the highest risk. Judgmental scoring does not tell you what the probability or odds are that a given company will pay its bill within any particular time period. Also, when the two methods are compared to each other by analyzing the same set of customers over the same time period, statistical-based scoring always outperforms judgmental scoring in predicting those customers that will become serious credit risks.

The potentially inconsistent treatment of information is another reason that a judgmental system may reach an outcome that differs from a statistically based decision. Judgmental systems rely on the experiences of individual loan officers to discern the factors that will be good predictors of loan repayment and to identify the tradeoffs among those factors. Differences in loan officers' experiences may lead them to consider different factors and make different tradeoffs among factors.

This system makes decision based on what they value the most by weighting them while having the ability to leverage multiple data source.

2.1b: Numerical Scoring Systems

Assessments of credit risk have been conducted as long as credit has been offered: Lenders collect information that they believe is relevant to the question of whether a loan will be repaid, and the summary of that information determines whether to make the loan.

Credit decision is made mechanically on the basis of a statistical model are commonly termed credit-scoring systems. Numerical scoring systems, first developed in the mail order industry in the 1930s and later used by large personal finance companies, were an attempt to address these concerns (Smalley and Sturdivant 1973, Wonderlic 1952)

Statistically based credit decision-making systems were pioneered during the late 1950s but only saw mainstream use during the 1990s as the depth and breadth of electronic credit information increased. These statistically based techniques are commonly referred to as “credit scoring” models. Initially, scoring models were employed in the consumer credit portfolios of most major banks and credit card issuers to increase the speed of the credit decision, enhance the uniformity of the decision process, and reduce the overall costs of decision making. The relative homogeneity of this type of credit and the wide availability of performance data on applicants made the initial implementation of scoring models by these lenders successful and profitable (M. Carry Collins , p. 103).

Although the ability to make credit decisions on a quantitative rather than a judgmental basis represented an important advance, the widespread diffusion of quantitative methods did not occur until development of the necessary computer technology in the early 1960s. In computer-based systems, hereafter termed credit scoring systems, the computational power of the computer is employed to identify, from a creditor's own historic files, those characteristics that best discriminate between the goods and the bads and to determine the point values for the various levels of each selected characteristic (Noel Capon spring 1982, p. 83). Statistically based procedures draw on types of information that will be similar for all borrowers and evaluate the data through a mathematical process that yields a numerical score.

Numerical Scoring is a method based upon statistical analysis, usually in the form of a multivariate regression model that can estimate the probability that a new customer or existing customer will default or become delinquent. These systems are very similar in operation to judgmental systems except that the factors used and their assigned weights are based on a statistical analysis of a company's past operations (optimally), i.e., under the best circumstances, historical data in an electronic format is available so that a company's own history can be used as the basis for the Behavior Scoring system.

Statistical model works in almost the same way as judgmental model. Statistical model considers many factors at the same time. This model analyzes multivariate (any procedure which involves two or more variables) correlation (A reciprocal relation between two or more things) to assign statistically derived weights used in the model. The factors are normally obtained from individual's credit files and also from the credit bureau reports. Statistical Scoring Model can also be described in terms of a scorecard, a pooled scorecard, and a custom scorecard. A scorecard applies data from one firm, whereas a pooled scorecard applies data from more than one firms and a custom scorecard mixes the data acquired from both the statistical model and judgmental model.

In the hearings on the amendments creditors argued that adherence to the law would be improved if credit scoring systems were used. They contended that whereas credit decisions in judgmental systems were subject to arbitrary and capricious behavior by credit evaluators, decisions made with a credit scoring system were objective and free from such problems. Regulation B thus envisioned two categories of credit decision systems, statistically sound and empirically derived credit scoring systems, and all others not satisfying the criteria of statistical soundness and empirical derivation, which are termed judgmental systems. This distinction has practical importance. For example, although age is a proscribed characteristic under the Act, if the system is statistically sound and empirically derived, it can be used as a predictive characteristic, provided that the elderly receive the maximum points awarded to any age category. The appropriate manner in which both types of systems should be used was spelled out in the Regulations (Noel Capon spring 1982, p. 83)

Presently credit scoring systems are used extensively, especially among major credit granters. It is claimed that their use reduces bad debt losses, that more consumers are granted credit, and that organizational consistency in decision making is achieved. Further, the costs of granting credit are reduced, since less skilled personnel are required and fewer credit reports need be purchased (Credit Card Redlining 1979, p. 234-240; Fair, Isaac and Company 1977).

The critical distinction between extant credit scoring systems and other methods of credit evaluation is the absence, in credit scoring, of an explanatory model. While judgmental systems are based, however imperfectly, upon a credit evaluator's explanatory model of credit performance, credit scoring systems are concerned solely with statistical predictability. Since prediction is the sole criterion for acceptability, any individual characteristic that can be scored, other than obviously illegal characteristics, has potential for inclusion in a credit scoring system (Noel Capon spring 1982, p. 85)

In evaluating information, statistical systems rely on automated statistical procedures, not on the experience and judgment of loan officers. The statistical procedures consider many credit-related factors simultaneously, statistically identify the relative ability of these factors to measure risk, and assign corresponding weights to each factor.

2.2: Probability of Default

The modern concept of risk has emerged as a central issue in finance over the last three decades. Gallati (2003) defines risk as a condition with potential exposure to adversity. Risk here is defined as a condition in which there is a possibility that the actual outcome will deviate from the expected. Credit risk or default risk is the oldest form of risk associated with financial markets. Risk is considered as a fundamental factor that influences financial behavior. Therefore a vast effort to allocate and manage resources in a way to decrease risk is required. Credit risk (default risk) is defined by Lopez and Saidenberg (2000) as “the degree of value fluctuations in debt instruments and derivatives due to changes in the underlying credit quality of borrowers and counterparties”

Default means to fail to comply with the terms of a contract, especially to fail to pay back a debt. A default event happens if the value of a borrower's assets falls short of the value of debt. When a consumer has been late on a payment, it is possible that other creditors, even creditors the consumer was not late in paying, may increase the interest rates the consumer is paying. This practice is called universal default.

Probability of default (PD) is a parameter used in the calculation of economic capital or regulatory capital for a banking institution. This is an attribute of a bank's client.

The probability of default is also like chance that a loan will not be repaid and will fall in the default. It is calculated for each client. The credit history of counterparty and nature of investment are taken into account to calculate probability of default.

Under the proposed new Basel capital accord (Basel II) 1, banks can also use their default probability estimates for calculating regulatory capital. Therefore, accurate default prediction is in the interest of banks as well as regulators.

There are many alternatives for estimating probability of default. It may be estimated from historical database of actual defaults using modern techniques like logistic regression, swap bounds and option on common stock.

Default probability can also be measured as the number of account that defaults over a 12-month window divided by the total number of open accounts at the observation point (Credit Card redlining 1979).

After a quiet period in credit card defaults in the early 1990' default and personal bankruptcy began to increase sharply after 1995; and this phenomenon has become a serious issue for banks and policy makers (Domowitz and Eovaldi, 1993).

The previous literature on consumer credit can be categorized in two parts the studies on consumer credit applicants examining the lenders' decision to grant the loan and the studies on consumer credit clients examining the borrowers' ability to pay the loan.

There are many studies on examine and improving the rejection and acceptance criteria of credit lenders' decisions. Jappelli (1990), for example, investigates lenders' and borrowers' behaviors in consumer rationing activities for the United States' credit market in 1983. He found that most of the applicants are rejected because of their credit history, their age or their income, amount of collateral, offered by borrowers to secure a loan in case of delinquency, is another important factor affecting credit-granting decision. Time spent at current job, time spent at current address, type of work, family size, sex, and race is found to be less effective on credit decision. They also found that debt is strongly correlated with economic factors. In a different study, Crook (2001) reported that income, homeownership and family size positively affect the United State household debt level. The refusal and acceptance model manage both good and bad borrowers similarly. This parameter can be reliable on basis of that good borrower should be accepted on above mention factors.

2.3 Effect of Variables on Probability of Default

2.3a: Loans/ Debts

Debt can be defined an amount owned to a person or organization for funds borrowed. It can be represented by a loan note, bond, mortgage or other form starting repayment terms. These different forms all imply intent to pay back an amount owned by a specific date. A debt is created when a creditor agrees to lend a sum of assets to a debtor. In modern society, debt is usually granted with expected repayment; in many cases, plus interest.

Consumer debt is consumer credit which is outstanding. It is debt which is used to fund consumption rather than investment. It is the debt which is held by an individual as credit card. Credit card debt is an example of unsecured consumer debt. The use of credit cards has increased substantially worldwide in recent years and is replacing currency as a method of payment.

Consumer Debt results when a client of a credit card company purchases an item or service through the card system. Debt accumulates and increases via interest and penalties when the consumer does not pay the company for the money he or she has spent.

Debt issued by consumers is an understudied asset class. There has been particularly little academic study of recent trends in default on this debt. Between 1994 and 1997 the number of personal bankruptcy filings in the United States rose by about 75%. The 1.35 million filings in 1997 represented well over 1% of U.S. households. Delinquency rates on credit cards rose almost as sharply (Federal Reserve Bank of Cleveland 1998).

The results of not paying this debt on time are that the company will charge a late payment penalty and report the late payment to credit rating agencies. Being late on a payment is sometimes referred to as being in "default". The late payment penalty itself increases the amount of debt the consumer has.

The creditworthy borrowers obtained additional credit and it is these borrowers who accounted for most of the increase in default. Many analyses cite growth in the number of credit cards offers and in the sizes of credit cards limits, among other changes in the supply of consume credit is the most important factor behind the increase in default (David B.). The higher income people use more credit cards this result increase in probability of default. The card holders with higher balance have a higher probability of default.

Credit risk poses the most significant risk to banks involved in credit card lending. Since credit card debt is an unsecured line of credit, repayment depends primarily upon a borrower's capacity to repay. The highly competitive environment for credit card lending has provided consumers with ample opportunity to hold several credit cards from different issuers and to pay only minimum monthly payments on outstanding balances. As a result, borrowers may become overextended and unable to repay, particularly in times of an economic downturn or a personal catastrophic event.

A first variable which is found to influence default behavior is the percentage of the total credit line which the consumer has used. A high debt balance to credit line ratio should increase the probability of default for a card user. A strategic factor exists for consumers in this variable. A consumer facing default may try to obtain more credit line in order to avoid this situation. It is found to have significant predictive power in explaining the tendency for default (Domowitz and Eovaldi, 1993). Number of credit card will increase debt burden on individual customer and then customer will fall into category of default. Default probability will increase with increase in amount of debt.

The provision of consumer financing has become a pervasive tool for many marketers worldwide (Glass- man 1996). Simultaneously, consumer debt has soared to record levels and an alarming number of households are finding themselves in financial difficulties (Andelman 1998, Monthly Review 2000). In lending credit card debt burden is one of the most important factor if customer having debt burden it is possible that he or she will not be granted cerdit.

One of the main sources of data for lenders is other banks, good and standardized data requires a strong banking system. Banks are social accountants: they keep track of how much money their clients keep on their accounts, and that signals to others about customers creditworthiness (Stiglitz and Weiss 1988). Banks as lenders are also responsible for monitoring and keeping track of people's credit behavior.

According to Norvilitis (2003), there are conflicting findings in the literature about which psychological factors causes creating debt. Some studies conclude that people with large amount of personal debt are not demographically different than others.

Livingston and Lunt (1992) found that debt is common among individuals with high incomes and less children. High income people will find the way to avail more than one credit card, which resulted in high debt burden and then they fall into category of default. As debt burden increases the probability of default will also increase. The probability of default and debt are positively related.

Calem and Mester (1995) investigated the situation of the people's inability about paying their credit card debts and found that card holders with high unpaid debts have higher probability of default. Card debt also increase with the increase with the ratio of total minimum require payment from all credit cards to income and from the percentage of total credit line which has been used by the consumer.

Mester (1995) find that card holder with the higher balance have higher probability of default. Dey, Mumy (2005) find that higher credit worthiness of the borrower, the lower is their likelihood of default.

Work by Ausubel (1997) and Domowitz and Sartain (1999) both find that there is positive correlation between credit card debt and personal bankruptcy filings. The potentially serious impact of credit card default on the general state of the economy has prompted a number of researchers to explore the default issue.

Calem and Mester (1995) test the argument of Ausubel' 1991 paper that irrational consumer behavior and problems account for the failure of competition in the credit card market. They also examine default in this market and find that cardholders with higher balances have a higher probability of default. The number of credit card having to an individual is direct related to probability of default. The bigger the credit stocks the higher difficulty in paying its dues and payments.

Dunn and Kim (1999) found that the variables, the total minimum required payment to income ratio, the percentage of total credit line which the consumer has used and the number of credit cards on which the consumer has charged to the credit limit have statistically significant positive effects on the probability of credit card default. They set up that the total credit card debt to income ratio has no dramatic power on the default probability. In short it has been founded that raised in the number of loans and debts will falls his ability of paying bills and outstanding. In other words the higher the bundle of credit, the higher the risk of default.

Ausubal (1991) pointed out that there is a contribution from irrational card holder's attitudes for the rise of credit card debts in USA. Calem and Mester (1995) examined the situation of the people's inability about paying their credit card debts and found that card holders with high unpaid debts have higher probability of default. In their study, Black and Morgan (1998) stated that there are important effects of the social and demographic factors about the rise of not paying the credit card debts

Ausubal (1997) and Domowitz and Sartain (1999) have found a positive relationship between credit card debts and personal bankruptcy. He believes that if client is having more than one credit card he may default in paying one of the payments. In this sense there wee can expect positive correlation between surplus debt and default probability. The relationship between debt and default is an indicator of personal bankruptcy.


Income refers to the consumption and savings opportunity gained by an entity within a specified time frame, which is generally expressed in monetary terms.

It is well known that the maximization of lifetime utility for a consumer facing perfect capital markets leads to the independence of current consumption from current income. But most empirical estimates indicate that consumption is more sensitive to current income than predicted by the life-cycle hypothesis with perfect markets.

Income is an important variable. However, the points relationship does not increase monotonically; rather, the points fluctuate wildly as income increases (Noel Capon spring 1982). Consumer having lesser source of income will be granted lower amount of credit line and wise versa.

Zeldes (1989), using the Panel Survey of Income Dynamics, performed an excess sensitivity test by splitting the sample into high- and low-liquid-asset households and found that consumption is too sensitive to current income for low wealth families but not for the rest of the population. On the other hand, Runkle (1991), using similar sample splits does not reject the implications of the PIH/LCH. Zeldes focuses on consumer behavior relating to appropriate use of income and he founds that lower class consumer use more credit card to finance their consumptions.

The debt-to-income ratio used in many previous studies acts as a rough proxy for the more detailed behavior that is embodied in these two variables. Since credit cards have expanded the set of decisions that a consumer must make in the use of a debt instrument, they have therefore expanded the possibilities for the employment of strategy by consumers. These important aspects of credit card use are not adequately reflected in the debt-to-income ratio.

The minimum paymentvariable was chosen as an independent variable because according to Dunn and Kim (1999), many card users using revolving credit are maximizing their utility subject to the minimum required monthly payment constraint rather than an income constraint. Therefore, it is possible for the minimum required payment to income ratio variable to have ability of predicting default behavior. It is expected that a high ratio should increase the probability of default.

While the total minimum required payment to income ratio is more relevant to a consumer's ability to avoid default in the short run (month-to-month), the debt to income ratio balancingis important in terms of showing a consumer's overall debt condition in the long-run (Dunn and Kim, 1999). Purchaser endeavor to uphold their lifestyle and expenditure over their life time even through their income and wealth may fluctuate, so they use credit card to maintain their life style and take benefit of minimum payments. If consumer having more credit cards than his capacity to pay than he will be incapable to disburse his outstanding and he will fall into category of default.

There are studies in the literature related to credit attitudes of the individuals. In a study, Davies and Lea (1995) analyzed the students' attitudes toward debt in UK and found out that students with higher incomes tend to have higher debts (from Hayhoe et al, 2000). They found that income does not affect credit-granting decision and being a male significantly decreases the chance of being granted a loan. Students who be expecting future income higher than their present income will borrow from their prospect income.

The lack of significance of education, income, or home-ownership in directly influencing default is noteworthy since banks traditionally have relied heavily on these three characteristics in assessing the credit-worthiness of loan applicants. It also suggests that the behavior of consumers in using credit cards which is captured in our financial variables is not affected by these socioeconomic characteristics, although they may influence the individual components of our financial variables (e.g., total credit line, etc.).

The incomevariable was added to model as an independent variable to know whether an increase on income increases the probability of default or not. It is expected that a person who has higher income is more likely to use credit card for their expenses and may have higher probability of default. Xiao et al. (1995) indicate that in terms of other financial behaviors, individuals and families with high income tend to be more willing to use credit cards. According to Livingstone and Lunt (1992), the increase of debt and ability to repay debt is also affected by income. There is moderate positive correlation between income and loan size because of policy of the institution, thus, the amount of credit is determined according to borrower's income level.

The life-cycle hypothesis (Modigliani and Brumberg 1954) posits that consumers attempt to maintain their lifestyle and consumption baskets over their lifetime even though their income and wealth may fluctuate over time. Specifically, older consumers can borrow from their past savings and consume at levels beyond their current incomes. Conversely, young consumers who expect future incomes to be higher than their present income can "borrow from their future income" to support their present lifestyle. These processes have been referred to as consumption smoothing (Shefrin and Thaler 1988). Income is used to predict repayment behavior of customer if customer earns enough amount of money and if he is able to pay debts he should be granted loan

Livingston and Lunt (1992) found that debt is common among individuals with high income. High income people have more burden of debt. Past studies indicate that there is a close relationship between spread of credit card usage in a country and its stage of socio-economic development. With increased level of socio-economic and technological development, credit card usage particularly increases in developing countries. Credit card usages will increase in order to fulfill socio economic wants. Payday loan approval requires a steady income source.

Studies credit card usage behavior in an advanced developing country an empirical research study conducted in urban Turkey indicates that there are certain relationships between socioeconomic and demographic characteristics of Turkish consumers and their credit cardholding and usage behaviors. In light of the survey findings, offers a number of marketing strategy guidelines for credit card issuers in developing countries to remain competitive as well as creating more market growth opportunities in this growing service industry. If consumer is willing to take debt beyond the bank's assessment of their ability to handle to that level of debt he will fall in area of default. If consumer is using only one or limited credit cards he will be able to manage payments and there will be no or less chance of default.

Crook (1996) replicates Jappelli's (1990) study with 1989 data and examines whether the client characteristics, which predict the probability of households being credit constrained, have changed or remained the same between years 1983 and 1989 in the United States. According to his study, more years of schooling of a household head would also be expected to increase future income, with consequent increases in the household's demand for credit and the supply.

Roszbach K. and Jacobson T. (1998) built a statistical model in order to measure the risk of sample loan portfolio and show how the model helps to evaluate alternative lending policies. They found that income does not affect credit-granting decision and being a male significantly decreases the chance of being granted a loan. In addition, homeowners have more chance of being granted a loan.

Cox and Jappelli (1993) found that the demand for credit to be positively related to permanent earnings and net worth and negatively related to income. If consumer have easy access to large amount of credit, they are likely to infer that their lifetime income is high and their willing to use credit will also be high.

A study, Davies and Lea (1995) analyzed the students' attitudes toward debt in UK and found out that students with higher incomes tend to have higher debts (from Hayhoe et al, 2000). Duca and Rosenthal (1993) found that the credit demand of young households is positively related to wealth, income and household size. This can be summarized as that consumer with low income will default probability very high.

Slocum and Mathews (1970) testes whether social class and income can be considered as indicators of consumer credit behavior by using data obtained from 2,032 commercial bank credit card holders in the USA. They found out that while members of different income segments exhibits different credit card use patterns, social class is not the most useful market segmentation variable for the credit card behavior of consumers and concluded that income level is better indicator of consumer credit card behavior than social class.

In their studies Mathews and Slocum (1969) and Slocum and Mathews (1970) found out that cardholders with low income and socioeconomic status use cards to generate revolving credit more frequently than do rich and high status card holders. There is positive relation between income and loan size because of the policy of lending determine credit limit according to the customer income level.

A consumer who is confident that income will increase substantially will rationally take on more credit than will an otherwise similar consumer who does not think that a substantial income increase is likely. the higher the expected income growth rate, the more the consumer will borrow for a given probability; and the higher the probability of income increase, the more the consumer will borrow for a given real income growth rate. (Yu-Chun Regina Ghang2 and Sherman Hanna)

The level of current income affects the relative importance of the disutility caused by the social embarrassment of default; On the other hand, the steepness of labor income profile affects the desirability of access to credit. For households with a hump-shaped labor income profile (high school graduates), default is more likely to occur early and late in life. For consumers with very steep labor income profiles (college graduates) the probability of default is very low, at all ages.

In order to understand the effects of introducing default, we have to keep in mind that the option of being able to discard liabilities at any time in the future in effect provides the consumer with an additional insurance policy. Consumption will be higher for lower levels of wealth. However, for wealthier households the benefits of filing for bankruptcy are smaller; the value of the option of default decreases and consumption functions converge as cash on hand increases.

In this model the default rule is always such that there is an asset level below which default is triggered. In other words, if in a given period the consumer has a low income realization, outstanding debt is at its limit and his assets are below this equilibrium trigger level, he chooses to default.

Dunkelberg and Smiley (1975) and Shay and Dunkelberg (1975, p.188) examine what they claim to be “the popular notion that the low-income credit user subsidizes the high income users.

2.3c: AGE

Age refers to the length of time a client has lived. Age has a large effect on default probability. Bank examines each customer while granting loan. Age effect is one of the factor consider that cause the probability of default. There is tenuous evidence that the probability of default decreases with age. Young customer has more probability of default.

Age of a credit card holder and demand for credit can be related to each other with the lifecycle theory with financial needs generally increasing with age (e.g., buying a house / car, providing college education for children), therefore older individuals are more likely to have greater debts than younger individuals (Cameron and Golby, 1990). The customer age is therefore used to predict default probability. It is argued that young customer have the higher default probability than older ones, the variable is expected to have negative impact. The points awarded for age have a curved relationship.

In a study, Kaynak and Harcar (2001) investigated consumer attitudes and intentions towards credit card ownership in Turkey and found that the age group between 36 and 45 is more likely to own credit cards than any other group. Barker and Sekerkaya (1992) reported that the middle age group is the most likely to hold and use credit cards. Young consumer use more credit card in order to fulfill their needs and to increase their standard of living.

The numerical results indicate that married couples and homeowners tend to pay their debt on time. On the other hand, credit default risk decreases when the income and age increase. The correlation between age, marital status and interest rate credit category is found positive.

Although older American spends increasingly large sums of money on goods and services, it is widely believed by financial service providers and retailers that these individuals make relatively little use of credit cards. Using a large national sample of respondents from different age groups, finds that older adults use credit cards as frequently as younger adults when circumstances and opportunities for consumption in both groups are similar. Age-related declines in use of credit cards may reflect changes in lifestyles and other circumstances associated with age. Discusses implications of the results for retailers and consumer credit lenders.

Adcock etal. (1977) indicated that there is a negative relationship between age and use of credit card (from Lee and Kwon, 2002). Young investors, with long time horizons, have ample time to recover from large market losses and are making large contributions (relative to the size of their account balance). In addition to the many “earning years” in front of the young investor, the capacity to take investment risk is very high. In the years eligible for retirement (age 50 to 65) the ability to absorb risk gradually declines as preservation of retirement wealth becomes more important. Eligible retirees have fewer “earning years” in front of them. Upon retirement, many glide paths continue to roll down equity exposure as the individual's time horizon (lifespan) gets shorter.

The lifecycle path, focal point on asset growth for many years proceeding to the age of 50 and then begins to accentuate principal continuation as retirement come closer. In a cross-section of prime borrowers, middle-aged adults borrow at lower interest rates and pay fewer fees relative to younger and older adults. Averaging across ten credit markets, fee and interest payments are minimized around age 53. Here it is observed was that experience rises with age, but analytical abilities decline with age.

The 20-45 year old segment of the population is positively associated with card use as people in that age group are more likely to acquire debt than individuals in other stages of their life cycle. , the insignificance of the population variable indicates that the 18-45 age group is a less important determinant of check-credit than credit-card use.

Tabulations of bank card use by economic and demographic characteristics have been provided by Johnston (1975), Mandell (1973), and Shay and Dunkelberg (1972). Card use is found to increase with income, education, and social class, and to be lower among retired people.

2.3d: SAVING

Personal saving is defined as personal disposable income minus personal consumption expenditure. In other words income that is not consumed by immediately goods and services and money is saved. Saving can occurs, as with cooperate retain earning it can be calculated as profit minus dividend and tax payment.

Saving is different from savings. The former refers to an increase in one's assets and increase in net worth, where as the latter refers to one part of one's assets, usually deposits in saving account. Saving refers to an activity occurs over time, a flow variable, where as savings refers to some thing that existed at any one time.

Asset correlations are a major driver of regulatory capital requirements in the new Capital. The Basel Committee on Banking Supervision in April 2003 issued its third proposal for a revision of the standards for banks' capital requirements. A significant part is devoted to the evaluation of the credit risk of retail loan portfolios. Two important input parameters are default probabilities and correlations.

HARALD (2009) assumes that probability of default happens if the value of a customer's assets falls short of the value of debt. He further describes that, the correlation of the asset values of two borrowers is an indicator for the degree of co-movements. Default correlations can be analytically derived from the asset correlations, it means that consumer are becoming more credit worthy if they reduce consumption and increase saving,

Hayashi (1985), using the 1963 Survey of Financial Characteristics of Consumers, estimates consumption for high-saving households. He interprets the estimates as a description of desired consumption without credit constraint.

According to the life-cycle theory, the consumption and saving behavior of people changes with income, wealth, age and different socioeconomic conditions during different periods of their life (Modigliani and Brumberg, 1954) and for age, the model indicates that saving should be positive for individuals in their working period and negative for the older ones, therefore wealth should be hump-shaped (Modigliani, 1986). It can be summed as saving through different consumption with respect to age. Saving is more when customer is in his earning stage. Saving shows decline to a particular age.

Saving is one of the variables that help customer to avoid default. Individual will not select for bankruptcy unless he has strong believes that his future cash flow will cover obligation. Many points are awarded for maintenance of either a checking or savings account, irrespective of the balances (Noel Capon spring 1982). Saving helps customer to pay debt. Saving takes very high value in assessing credit card. If customer lacks of money he can pay back from his saving.

Economic investment theory models developed by Fisher (1930) and Hirshleifer (1970) suggest consumers may increase their utility through judicious selection of debts and assets (Herendeen 1975). A consumer expecting a continued growth in real income might borrow to smooth consumption over the life cycle.

The real-time on the exemption level is positive suggesting that consumers hold higher savings in regions where they are allowed to keep more of them in case they file for bankruptcy. This evidence is also conformed by the estimation done by Lehnert and Maki (2002). In short household are likely to grasp low return assets and have high cost debt with higher income level.

Saving is essential for consumers who never choose to default when assets are very high. In case the when client go unemployed the probability of default is higher and saving help them to overcome default. This is due to the fact that in that case default generates asset level is positive and constant over the life cycle. This is logical if consumer keeps in mind that in case of unemployment, crash in income impel consumption level down and marginal utility is high, accordingly the stigma effect becomes comparatively less, for all ages, including middle aged consumers. The assets level is not very high because saving is very valuable for consumer.

Households with both high income and high assets can avoid repaying their debts in bankruptcy as long as their assets are below the applicable exemption level. Debtors can often increase their financial benefit from bankruptcy by shifting assets from nonexempt to exempt categories before filing . Households with large amounts of assets have the most to gain from generous bankruptcy exemptions in a manner similar to income, assets affect households access to credit and may also affect their demand. It is easy for consumers to use past income in the future because it can be stored in the form of savings and investments to be used later

2.3e: Occupation/ Employment:

Occupation represents the work place where the customer is working. For the occupation characteristic of a credit scoring system employed by a multinational bank, the occupation of farm foreman and laborers, enlisted personnel, clergymen, entertainers, farmers and ranchers and government and public officials received few points (Credit Card Redlining 1979)

An applicant's occupation is an important characteristic. However, to be gainfully employed in the categories of driver, labor, or outside gains no more points than being unemployed. Occupation is an important characteristic, and the unemployed category achieves the highest possible point score (Noel Capon spring 1982, p. 86). In assessing customer credit worthiness unemployed client are not awarded credit.

It is observed that most of the post war upward trend in personal bankruptcy and default probability has taken place in periods of economic expansion, with unemployment rates falling. It is in periods of low uncertainty and low unemployment rates that buyer-stock consumers wish to consume and borrow more (Carroll 1992). Obtaining loan or credit card depends mainly on client circumstances. Bank considers that do client have assets, employment at the time of applying credit card. If the customer is unemployed so the chances of obtaining approval from a financial lender will be lesser.

When cardholder's don't have any income, personal saving and any valuable assets and being unemployed then that customer will come into category of default. Unemployment variable reflects the cash flow to individuals from the job market.

Variables that were significantly negatively associated with bank credit card outstanding per capita, as hypothesized, included the farm income share of statewide personal income and unemployment variance. They indicate that credit loan holdings were significantly positively associated with per capita personal income.

The empirical work in this study supports a number of theory about the determinants of credit-card use many of which have not been previously confirmed. They suggest that credit-card outstanding per capita are high in states where consumer real income and bank wages are high and unemployment variance is low.

Occupation has a significant relationship with probabilities of default. The model makes the following important contributions. First, it delivers a natural explanation for the observed fact that most of the post war upward trend in personal bankruptcy has taken place in periods of economic expansion, with unemployment rates falling. It is in periods of low uncertainty and low unemployment rates that buffer-stock consumers wish to consume and borrow more (see Carroll 1992). The financial lender will look at the amount of loan compared to the total amount of client assets should they need to be sold to repay the loan. This will only happen if client default on the loan repayments. Many unemployed people who need small sums of money often apply for a no credit history check low interest credit card and these types of clients are not rewarded credit line.

The study considers by Peterson (1976) in which he performs an un-structured regression of bank card debt using interstate data. The debt variable is written as a function of customer income, unemployment rate, and dummy variables. He founds that debt is higher in higher income states and low in unemployment variance states because a rich, job secure person demands more.

23.f: Number of Dependent:

House hold size of cardholder has a positive relation with default probabilities which implies that those customers with large household size tend to default more than whose household size is smaller. This may not be unconnected to the problem of both high children and perhaps adult dependence. This means that operator of large family size has more chance of default than less family member.

Özlem Özdemir (2004) build a conceptual model in order to explain the relationship between consumer credit clients' payment performance and credit category, he found that the clients' payment performance decreases, when clients live with their families.

Livingston and Lunt (1992) found that debt is common among individuals with high incomes and less children


The preceding chapter discussed theories. These theories help to develop theoretical frame work that guides in calculation Credit Scores.

3.1 Relationship Factors Affecting Credit-Rationing

The length of the relationship between a borrower and a potential lender should be an obvious determinant of whether the lender extends credit to the borrower. The longer the relationship, the greater the opportunity for the lender to monitor the borrower and mitigate the adverse selection problem between borrower and lender. Accordingly, LENGTH, defined as the duration (in years) of the family's oldest loan account with the potential lending institution, is introduced as a relationship variable.

Relationships can also be built through interaction over the multiple services provided by a financial institution, which reflects the breadth of a family's relationship with the potential lender. Accordingly, we define ACTIVITY as the total number of asset accounts and loans with a family's potential lending financial institution.

The essence of the Stiglitz and Weiss (1981) argument is that informational irregularity between borrower and lender leads to moral hazard and adverse selection effects, leading to refusal of loans to some among an observationally identical population of potential borrowers. It is reasonable to assume that through the gathering of information pertaining to certain observable characteristics of borrower creditworthiness, some of which are observable over time, a lender is better able to make a rational judgment about a potential borrower's request for credit. If the borrower is able to provide such information to the bank about his creditworthiness, he is able to improve his chances of obtaining a loan and to do so at a relatively lower cost. Consistent with Peterson and Rajan (1994), we label these interaction terms "relationship.

The degree of a borrower's dependence on a potential lender as a source of financial services should influence whether the borrower receives credit. If an individual, through his dealings with multiple lenders, incurs benefits from all but shares the costs with the lenders, then the value of private information pertaining to the borrower is less valu- able to any individual lender and may lead to reduction in the incentive for that lender to extend additional credit to the borrower (Bulow and Shoven 1978).

Also, as Cole (1998) argues, the number of sources of financial services/loans may also capture borrower quality in that a lower-quality borrower may be forced to shop around for a loan. Hence, potential lenders may be less willing to extend credit to borrowers with multiple sources of financial services and/or loans-after controlling for other relevant borrower related factors.

Cole (1998) also discusses the notion of preexisting relationships between a borrower and a potential lender and argues that such relationships generate useful information in ascertaining a firm's creditworthiness. Consistent with Cole, we introduce variables that capture whether an individual (or family) maintains different asset accounts with the potential lender.

Some author considerably beyond Cole in identifying the various asset accounts that an individual could maintain with a potential lender. These finer distinctions are important because the type of information that a lender obtains from monitoring a checking/savings account is likely to be different from the information obtained by monitoring an individual's individual retirement or Keogh account.

According to Diamond (1991), the age of a client should influence whether it receives credit simply because a client is in business for a longer period of time has generated reputation effects through the display of its ability to survive the critical start-up period. Consistent with this intuition, in our case of individual borrowers we include AGE, defined as the age of the head of the household, as a public information proxy

3.2 Evaluation of Credit Scoring

Credit scoring process is used to envisage an applicant will or will not repay credit on time. This prediction is to decide which ones from large number of applicant, to grant credit which ones to declines. Thus this process will predict default probability of an applicant. This can be comparing with cut off score whether to reject applicant or to accept applicant.

In early days numbers of characters were chosen in order to predict goods and bad customers. Good customers are defining those who paid their credit and those who did not paid there debts are consider as bad customers. An applicant for credit is assessed in credit scoring system by adding the points received on the variety of application characteristics to appear to a total score.

The score is treated as cut off scores. If customer score exceed the cut off, credit is awarded automatically, while if it falls below the lower cut-off, credit is automatically denied. If the score is between the two cut-offs, credit history information is obtained, scored, and the points added to the total score obtained from the application blank. If this new score is above a new higher cut-off, credit is awarded; if not, credit is denied (Neol Capon, 1982)

The creditor sets his/her cut-off values on the basis of the probabilities of repayment and nonpayment associated with the various point scores and the trade offs between type I and type II errors. The higher an acceptance cut-off is set, the lower the type I error (accepting applicants who fail to repay), while the lower a rejection cut-off value, the lower the type II error (failing to accept applicants who would have repaid) (Neol Capon, 1982)

Major National Retailer's Final Scoring Table for Application Characteristics







Zip Code


Time with Employer

Zip Code A




Less than 6 months


Zip Code B




6 months-5 years


Zip Code C




5 year-8 years


Zip Code D




8 years-15 years


Not Answered




15 year or longer


Bank Reference





Checking only






Saving Only






Checking and Saving


Military enlisted


Not answered


Bank Name


Military officer


Finance Company Reference

No Reference


Office staff




Not answered




Other Reference Only


Type of Housing









Not answered






All other




Not answered



Time at Present Address



Less than 6 months




6 months-1 year




1 year-3 years




3 years-7 years


All other


7 years-12 years


Not answered


12 years 6 or longer


Not answered



4.1: Hypothesis:

Hypothesis: Local banks have less probabilities of default than foreign banks.

The customer loan market function by measuring customer is replicated by the lack of variation in time series and cross functional data. Foreign banks used judgmental decision method which is basically a method of evaluating credit risk through internal and external credit experience. It is a traditional method of calculation risk through history, occupation, and residential address of customer. In the recent years local banks had moved from the judgmental review of loans to computerized credit scoring method. In computerized scoring method lender collect information which he believes is relevant for calculation credit worthiness of customer.

It is observed that judgmental scoring model leads to higher probabilities of default when compare with the both judgmental and computerized scoring method. Some studies identify scorecards variables that produce the disproportion in applicant rejection rates. Research suggest that foreign banks should focus more on computerizes scoring model.

4.2: Variables:

Financial literature classifies number of variables as indicator of credit card default. The hypothesis development and variable choice is motivated by both experimental and academic consideration. The selected variables should be related to the default probabilities and credit card rationing theories

The more responsive quantify is to default risk, the more it can rapidly reflect the changes and therefore the more effectual factor will be early warning factor. The following factors are taking on as analysis.

4.2a Default probabilities:

Percentage of account charged off during the 12 months outcome Default Probabilities is measured as that number of accounts that default over 12 month window divided by the total number of open accounts at the observation point ( Credit Card Redialing 1979)

4.2b: Income:

Annual house hold income from all sources before tax.

4.2c: Occupation:

The occupation is representing with the work place where the customer is working. For the occupation characteristic of a credit scoring system employed by a multinational bank, the occupation of farm foreman and laborers, enlisted personnel, clergymen, entertainers, farmers and ranchers, government and public officials received few points ( Credit Card Redialing 1979)

4.2d: Loans/ Debts

Debt can be defined an amount owned to a person or organization for funds borrowed. It can be represented by a loan note, bond, mortgage or other form starting repayment terms. Consumer Debt results when a client of a credit card company purchases an item or service through the card system.

4.2e: Number of Dependent:

Number of dependent is the number of people to whom client is supported by giving his support through his income.

4.2f: Saving:

Personal saving is defined as personal disposable income minus personal consumption expenditure. In other words income that is not consumed by immediately goods and services and money is saved. Saving can occurs, as with cooperate retain earning it can be calculated as profit minus dividend and tax payment.


This research study credit scoring methods and its relationship with probability of default. This chapter aim to describe the methodology used to provide data to investigate them.

4.1a: Data Collection:

The data used in this study is derived from State Bank of Pakistan, over the period May 2006 to May 2009. The number of banks registered at State Bank of Pakistan is 24 banks. The available data does not exceed to may 2009 so data collection is limited t this time. For data analysis, a clear definition of default is required. Default is define when client is not able to meet his/ her obligation on due date.

4.1b: Data Sample:

This study aim is to compare credit scoring method applied on Local and Foreign Banks of Pakistan. The population of our study regarding banks offering credit card facilities to its customer determines the following numbers. There are eleven banks in Pakistan offering Credit Card Facilities, out of which eight are local banks and three are foreign banks (source: State Bank of Pakistan). A random sample of six banks (three local and all of three foreign banks) are selected (Roseo, 1975)

4.1c: Dependent and Independent Variable:

The critical destination between computerized credit scoring system and judgmental scoring system is the absence of an explanatory model. While judgmental systems are based, however imperfectly, upon a credit evaluator's explanatory model of credit performance, credit scoring systems are concerned solely with statistical predictability. Since prediction is the sole criterion for acceptability, any individual characteristic that can be scored. (Credit Card Redlining 1979)

Default probability is taken as dependent variable and independent variables are age, income, number of dependent, occupation, saving and loan references.

4.1d: Method:

The process of credit scoring is important for banks as they need to segregate bad customers form good customers in terms of credit worthiness. Banks has to reveal hidden data about their customers. Both methods computerized scoring method and judgmental decision methods can be measure through different methods. Here we use independent T test in order to find which decision method has less probability of default.


This chapter focusing on reporting the findings of the study

The t-test has been used for testing difference between means of Foreign and local Banks. This study was viewed that there is significant difference in default probability of both banks.





Std. Deviation

Std Error Mean


























































































The study was conducted to compare the default probability of two types of banks local banks and other is foreign banks. The first table labeled Group Statistics gives descriptive statistic for both groups. The “Mean Difference” statistics indicates the scale of the difference of the means. The above table shows that local banks default (M. 631.89, S.D. 12.191) are less than foreign banks (M. 66.241, S.D. 92.2427) so we reject the null hypothesis. There is the large difference between the means of variables that include age (Local Bank M. 97.87, Foreign Bank M. 109.361), debt (Local Bank M. 135.898, Foreign Bank M. 144.803), and occupation (Local Bank M. 92.93, Foreign Bank M. 115.398). This result succeeds the null hypothesis and shows that there is a significant difference local and foreign banks probabilities of default.

5.2: Independent Sample Test:

In this table we get result of two tests Levenes's Test of Equality of variance and T-Test for equality of variance. The table contains two sets of analysis, the first one assuming equal variance into two groups and second one assuming unequal variance. The Levenes's Test tells us which statistic to consider analyzing the equality of means. It tests the null hypothesis that both banks have same probabilities of default. A small value of significant is associated with this test indicates that two groups have unequal variance and null hypothesis is falls.

The T Test result (with equal variance not assumed) shows t statistics of -4.76 with 199.273 degree of freedom. The correspondence two tailed p valued is 0.000, which is less than 0.05 therefore, we reject the null hypothesis at 5% significant level, which means that the average default probabilities of two banks are significantly different from each other i.e. that default probabilities of two banks are not same.

In the second case (age) the significance value is also 0.001 which is less than 0.05 we reject H0. In the other cases, income (p value 0.269), dependent (p value 0.937), saving (p value 0.061) hypothesis is not supported because p value is greater that 0.05 so we accept the null hypothesis. It means in cases of income, saving and number of dependent probability of default are same in local and foreign banks.


This final chapter discusses the empirical finding in chapter 5. It aims to advise potential implications of the result in the light of the theoretical models. It also addresses direction of the future research.


As a result of the endorsing progress toward the use of credit scoring in credit card lending, scoring is at the front position of the policy discuss surrounding fair lending and possible dissimilar impact. Using 2006-2009 loan application data on credit cards from a State bank of Pakistan, we develop a framework for investigating whether judgmental method or computerized credit scoring leads to more considerable approach. We hypothesize that judgmental credit scoring systems result in larger probability of default. Our findings confirm this hypothesis.

The findings are important for the outcome of computerized credit scoring on credit card applicants. Proponent of credit scoring resulted that scoring improves the impartiality of loans judgment and lower the cost and time requirement. It shows that use of computerized scoring model as the criteria in underwriting credit cards had an improvement on loan application, results in less probability of default.

Finally the results have important implication for the bank supervisor. Banks and credit card companies, use credit scores to evaluate the potential risk posed by lending money to consumers and to mitigate losses due to bad debts.

Source: Essay UK - http://turkiyegoz.com/free-essays/finance/selling-goods-or-services.php

Not what you're looking for?


About this resource

This Finance essay was submitted to us by a student in order to help you with your studies.


No ratings yet!

  • Order a custom essay
  • Download this page
  • Print this page
  • Search again

Word count:

This page has approximately words.



If you use part of this page in your own work, you need to provide a citation, as follows:

Essay UK, Selling goods or services. Available from: <http://turkiyegoz.com/free-essays/finance/selling-goods-or-services.php> [17-12-18].

More information:

If you are the original author of this content and no longer wish to have it published on our website then please click on the link below to request removal:

Essay and dissertation help


SWOT Analysis Nivea Men | Schmirgelpapier Rolle 115mm x 5m Schleifmittel Korn 40 - 400 Schleifpapier | Download