Maximizing the Benefits From Credit Reporting

[Pages:19]TRANSUNION WHITE PAPER

Maximizing the Benefits From Credit Reporting

Michael E. Staten, Take Charge America Endowed Professor Norton School of Family and Consumer Sciences University of Arizona

1 Introduction

4 The Special Function Served by Credit Reporting

5 How does Comprehensive Credit Reporting Improve Credit Decision-Making?

8 Benefits That Result From Comprehensive Full-File Reporting Systems

13 Structuring the Credit Reporting System to Maximize Benefits

15 Conclusion

Credit bureaus have evolved worldwide in recent decades to reduce the costs of assessing borrower risk. Credit reporting benefits both lenders and consumers, at the same time it increases economic activity. However, some types of reporting systems generate larger benefits than others. The value of a credit reporting system to its host economy depends on the nature and extent of the information collected and shared, as well as on the organizational and ownership structure of the reporting system. This paper provides evidence that a full-file, comprehensive credit reporting system maximizes the economic benefits from credit reporting. It also explains why independent, privately held credit bureaus have been found to provide more complete credit history data and more innovative information products and services for lending decisions.

The credit report helps lenders pierce the "fog of uncertainty" surrounding each new loan applicant. More information about a borrower's past credit history and the handling of existing accounts improves the lender's risk assessment process. The result is a better match of borrowers to loans. Full-file reporting of negative information on borrower delinquencies and positive information on accounts paid as agreed produces the maximum benefits. Simulations illustrate how full-file reporting (especially with widespread lender participation) increases the ability of the shared data to predict borrower risk. Full-file reporting facilitates more accurate prediction of repayment probabilities and encourages repayment by giving consumers a stake in establishing and protecting a good credit reputation. Because full-file reporting increases a lender's ability to recognize borrowers for successful use of credit, it also increases both the availability of credit as well as the likelihood that loans are repaid.

When comprehensive data on consumer credit histories is readily available, it significantly reduces the costs to new lenders of entering loan markets, enhances competition and lowers credit prices. This levels the playing field so that new entrants can overcome the advantage of established lenders in reaching creditworthy customers.

Broader access to credit made possible by full-file credit reporting systems generates macroeconomic growth advantages for the host countries. Cross-country studies have found that consumer spending is more sensitive to changes in income in countries with less-developed consumer credit markets. Credit markets that make loans accessible to large segments of the population provide a cushion that dampens the impact of household income interruptions, and provides consumers a credit "bridge" while they look for new employment. Portable credit reputations also increase the mobility of the workforce by reducing the burden on consumers of accepting new jobs in distant locations in response to structural shifts in the economy. In this way, structural economic shifts can cause temporary disruptions without crippling long-term effects. Full-file reporting also can make capital more mobile by supporting the securitization of consumer loan receivables. Comprehensive credit reports and credit scores give investors objective information on the quality of the underlying loan accounts. Armed with this granular information to quantify risk, lenders can tap into the global capital markets to obtain funds for making loans.

Securitization has supported enormous growth in credit card, automobile and mortgage lending in the U.S. In addition, comprehensive personal credit reports on small business owners have triggered an expansion in small business lending, especially to relatively risky marginal borrowers who would not otherwise receive credit.

But, not all reporting systems produce equivalent benefits. The structure of the reporting system plays a significant role in bringing the benefits to the market. Credit bureaus have evolved as either market-based, privately owned organizations or government controlled public credit registries. A recent study of 129 countries found that credit reporting agencies that distributed a broader range of data and provided legal incentives to ensure the quality of the data were also associated with significantly more private credit. More specifically, the characteristics of credit reporting agencies that are associated with significantly more private credit include 1) distribution (to end users) of both positive and negative data, 2) data from a cross-section of lenders, including banks, retailers, trade creditors, and utilities, 3) five or more years of historical data for borrowers, and 4) data collected for all loans of value above 1percent of per capita income. All of these reporting are hallmarks of full-file reporting.

Independent, privately owned credit bureaus produce full-file reporting more often than do public credit registries. The for-profit structure of the typical private credit bureau gives it greater incentive to ensure data accuracy, as well as to collect data from a variety of sources to increase the value of the information it sells. Consequently, the core database assembled and maintained by private credit bureaus is potentially more valuable to both private sector customers and the financial regulators than would be the data assembled by public credit registries. In addition, private credit bureaus also have an incentive to make data products available to a wider variety of end-users than do public credit registries, helping to account for the higher volume of lending in countries with private credit bureaus. These data products include the development of innovative decision tools (e.g., fraud prevention; identity verification and theft prevention; risk scoring models) that help to manage and reduce risk. Neither public credit registries nor industry-sector credit reporting consortiums have comparable incentives to develop a wide array of information products.

Maximizing the Benefits From Credit Reporting

The sharing of information among lenders regarding borrowers' past credit experience is the foundation of modern risk management.

Credit bureaus, either publicly run or privately owned, have evolved worldwide in recent decades to formalize the exchange of credit histories and reduce the costs of assessing borrower risk. Credit reporting systems generate benefits to lenders and consumers and increase economic activity. However, some types of systems generate larger benefits than others, depending on the nature and extent of information shared and also on the organizational and ownership structure of the reporting system. This paper will examine the significant economic benefits that are achievable through a well-developed credit reporting system.

The paper is organized as follows. Section 1 briefly discusses the economic function of credit bureaus, and their evolution as either market-based, privately owned organizations or government controlled public credit registries. Section 2 explains why information sharing through credit reporting improves credit decisionmaking. The results of several simulations illustrate that broader information sharing (more credit information reported for each borrower; more lenders participating in the sharing) increases the power of the data in predicting borrower risk. Section 3 documents the variety of benefits to consumers, lenders and the economy that can derive from a well-developed credit reporting system. The more comprehensive the credit reporting is, the greater the benefits. Section 4 considers how the organizational structure of a credit reporting system influences the magnitude of the benefits. Recent evidence from cross-country studies suggests that benefits are maximized when privately owned credit bureaus compete in the market to offer information products and decision tools.

3

? 2008 TransUnion LLC All Rights Reserved

Maximizing the Benefits From Credit Reporting

1. The Special Function Served by

Credit Reporting

The twin threats of adverse selection and moral hazard confront every loan officer examining a new loan application. Adverse selection poses a significant barrier to the entry of new lenders into credit markets. New entrants have no prior experience with local borrowers to draw upon, greatly increasing the difficulty of assessing an applicant's risk. As a result they are likely to attract applications from higher-risk borrowers who have been rejected by established lenders. Information about borrowers' past credit experience that is shared across lenders through a credit bureau intermediary can reduce this problem.

Moral hazard presents lenders with a different problem. Once a loan is obtained, borrowers have a greater incentive to default when the expected future consequences are low. But, a reputation for past default that is readily communicated to potential lenders can raise those costs, thereby reducing the moral hazard by boosting the borrower's incentive to repay.1 Credit bureaus facilitate that information sharing. Economist Daniel Klein has observed that the credit bureau has the distinction of being "the most standardized and most extensive reputational system humankind has ever known."2 To the extent that an interest in preserving reputations encourages consumers to alter their behavior, then through the market's evolution of the credit bureau, "commerce promotes morality as though it were guided by an invisible hand."3

The emergence of the credit bureau as an important third-party participant in credit markets institutionalizes the sharing of information and reduces the costs of assessing borrower risk. Economic research has shown that lenders can benefit as a group if they commit to exchanging information about borrowers and create an enforcement mechanism that ensures accuracy of the information shared (Padilla and Pagano, 1997). The third-party credit bureau serves as both the clearinghouse and enforcer. As a result, loan interest rates and loan default rates are lower than in the absence of information sharing.4

Around the globe, the pooling of borrower credit histories has become commonplace, although much of the reporting infrastructure has been established in just the past three decades.5 Credit information sharing may take place on a voluntary basis through private credit bureaus that are set up either through lender consortiums or by independent firms. In many countries the information sharing and pooling may be mandatory through Public Credit Registries (PCRs) set up and run by the country's central bank.6 In some countries, both types of credit bureaus serve the market.

Japelli and Pagano (2006) synthesize the results of two detailed cross-country surveys that were conducted to determine the extent of credit reporting and when it originated. One survey covered 49 countries (Japelli and Pagano 2002); the other survey was commissioned by the World Bank and surveyed 77 countries (Miller 2003). Together, the surveys revealed that before 1950 less

than 20 percent of surveyed countries had a private credit bureau, and less than 5 percent had a PCR. By 2000, 60 percent of countries surveyed had a private bureau and 50percent of countries had a PCR.

Private credit bureaus are usually structured around reciprocal agreements, where furnishers of data (creditors) voluntarily agree to contribute accurate data (usually in prescribed formats) in exchange for access to consolidated reports on potential customers. The level of detail in the report varies widely across countries. The threat that a data furnisher will be denied future access to reports if the furnisher fails to report (or knowingly contributes inaccurate data) is often used to reduce the "free-rider" problem inherent in sharing. Reported data ranges from a simple statement of current or past delinquencies (negative information) to more detailed statements that itemize account balances, credit limits, and account age, by type of account (positive information). In some countries, credit reports also include information on borrower assets and employment.

Because sharing with a PCR is usually compulsory, all lenders are covered, but PCR reporting is typically required only for loans that meet or exceed a certain loan-size threshold.7 Credit accounts with balances below the threshold are not reported, or may only be reported if seriously delinquent. Frequently, only the borrower's aggregate balances are reported to lenders making inquiries, not the itemized listing of accounts and creditors. Because there is typically credit activity below the PCR loan-size

? 2008 TransUnion LLC All Rights Reserved

4

Maximizing the Benefits From Credit Reporting

threshold, private credit bureaus still operate in some countries with PCRs by focusing on the market segment below the PCR threshold. Generally speaking, private credit bureaus offer more detailed information on individual accounts than do PCRs and may merge credit info with other types of data to create valuable information products for lenders.8

Jappelli and Pagano (2006) summarize the major implications of economic studies of the development of credit reporting and its influence on the functioning of credit markets. The sharing of information among lenders via credit reporting:

1. Facilitates more accurate prediction of repayment probabilities (reducing adverse selection problems)

2. Encourages repayment (reducing moral hazard)

3. Enhances competition (by reducing the costs to new lenders of entering loan markets)

4. Prevents overextension by borrowers as lenders gain a more comprehensive view of borrower credit obligations

2. How does Comprehensive Credit

Reporting Improve Credit Decision-Making?

Piercing the "Fog of Uncertainty" The credit report helps lenders pierce the "fog of uncertainty" surrounding each new loan applicant from the lender's viewpoint. Generally speaking, more information about a borrower's past credit history and the handling of existing accounts improves the lender's risk assessment process. The result is a better match of borrowers to loans.

The credit reporting environment varies widely around the world. The difficulty and cost of risk evaluation rises and falls accordingly. Some countries (e.g., United States, Canada, United Kingdom) have credit reporting systems characterized by comprehensive, full-file reporting that yields a credit report for each consumer containing both positive and negative information about a borrower's credit experience across all types of loan products. At the other end of the reporting spectrum are credit reporting systems that produce consumer credit files containing only negative information (delinquencies, chargeoffs, bankruptcies, etc.). Essentially, a consumer in a negativeonly reporting country would have either derogatory information in their credit report or no information at all.

The problems for a lender trying to assess the applicant's risk in a negative-only reporting environment are readily apparent. Such credit reports give a lender little or no information for lowerrisk borrowers who use credit responsibly. The lender can not discern the length and breadth of the consumer's past credit experience nor can the lender determine the consumer's current credit obligations. The consumer gets no benefit from

handling credit responsibly in the past, and the lender can't tell the extent to which the consumer is burdened with other credit obligations at the time of the application. Only when the file contains some negative information does it help the lender at all. Otherwise, the applicant remains shrouded in a fog of uncertainty.

In between the full-file comprehensive reporting systems and the negativeonly systems are a host of intermediate reporting environments that contain some positive information, but not a complete history of a consumer's credit experience. Common examples are reporting systems that evolved from lender consortiums within a particular segment of the industry. For example, banks may have historically participated in the exchange of information within the banking community about their consumer loan experience, but did not share that information with non-bank creditors. Retailer creditors or finance companies may have developed their own sharing arrangements within their segment of the loan market. Within the consortium, lenders would agree to share negative or positive data, or both. However, even with positive data present, a credit report produced by any one consortium would be incomplete, because it would contain the information from lenders in only one segment of the industry.

Across all reporting environments, comprehensive, full-file reporting provides the greatest benefit to risk evaluation. The empirical case for the advantages of full-file reporting is relatively recent, but the results are now well established, having been replicated

5

? 2008 TransUnion LLC All Rights Reserved

Maximizing the Benefits From Credit Reporting

by researchers several times with different datasets. Barron and Staten (2003) provided a comparative assessment of benefits from reporting environments as part of a World Bank project to explore the role of credit reporting infrastructure in developing economies. Their report offers a set of simulations that demonstrate the benefits of increasingly comprehensive information about a borrower's credit profile. In doing so, they simulate the effects that country-by-country differences in credit reporting have on credit availability and loan performance. One simulation is described below, comparing a reporting environment in which full-file information is available for risk assessment vs. an environment in which only negative information is available.

Because credit reports in the U.S. provide the most comprehensive profile of consumer credit histories (full range of negative and positive credit account data, including account balances and percent of credit line utilized for revolving credit accounts) of any credit reporting in the world, they provide a very useful analytical tool for the simulations. By using a large set of credit report data elements from U.S. credit reports to build a predictive credit scoring model, and then removing particular data fields that in other countries are either banned by regulation or unavailable due to limitations in local credit reporting systems, the simulation can identify the reduction in predictive power attributable to the missing information. This technique quantifies in two ways the cost imposed on lenders and consumers by the missing data. First, it reveals the increase in predicted delinquency rates for a group of accepted

loans, relative to what lenders can achieve when more information is available about a borrower's credit experience. It also identifies the decrease in the number of loans that can be approved for a given pool of consumer applicants, while maintaining a target delinquency rate.

Table 1 illustrates the change in predictive power associated with expansion in the information available to the credit scoring models. Under each of the scenarios depicted in the table, the models were used to calculate individual credit scores for each borrower in a sample. Individual borrowers were ranked according to their credit score. The authors then picked var-

ious "loan approval rates" (e.g., 60 percent) and reported the corresponding percent of borrowers who would default (defined as reaching 90 days past due) on their newly opened accounts within two years. To illustrate, at a targeted approval rate of 60 percent (i.e., accept the top 60 percent of all borrowers in the sample), the model built on a negative-only reporting system produced a 3.36 percent default rate among accepted applicants, compared to a 1.95 percent default rate for the full-file model. In other words, the default rate under the negative-only reporting rules is 72 percent higher than if full-file information was available to creditors.

TABLE 1

Effect on Default Rate for Various Approval Rates

Target Approval Rate

40% 60% 75% 100%

Negative-Only Model 2.91% 3.36% 4.10% 9.38%

Source: Barron and Staten, 2003

Comprehensive, Full-File Model

1.15% 1.95% 3.09% 9.38%

Default Reduction 153.0%

72.3% 32.7% 0.0%

TABLE 2

Effect on Credit Availability for Various Default Rates

Target Approval Rate

3% 4% 5% 6% 7% MEAN

% of Customers Who Obtain a Loan

Negative-Only Model

Comprehensive, Full-file Model

39.0%

74.8%

73.7%

82.9%

84.2%

88.9%

90.6%

92.8%

94.6%

95.6%

100%

100%

Source: Barron and Staten, 2003

Percent Increase with Full-file Model

47.5% 11.1% 5.3% 2.4% 1.0%

? 2008 TransUnion LLC All Rights Reserved

6

Maximizing the Benefits From Credit Reporting

Next, consider the implications of the two reporting systems for extending credit to deserving borrowers. Suppose the economics of the lender's operation dictate an optimal default rate of 3 percent. Table 2 shows that the negative-only reporting model could approve only 39.8 percent of applicants without exceeding the target default rate. However, under the full-file system, 74.8 percent of applicants could be approved. In other words, for every 10,000 applicants, the full-file system would approve 3,500 deserving borrowers that the negative-only system would have rejected. These results clearly indicate that a credit reporting environment that does not provide comprehensive credit experience on consumers produces either a significant increase in the likelihood that originated loans will default (and consequently higher loan prices) or a significant reduction in the availability of credit.

How can this be? The reason for the improved performance of the full-file model is intuitive: when risk assessment tools have less information available to them, creditors have greater difficulty piercing the "fog of uncertainty" that surrounds new borrowers. Consequently, creditor efforts are less effective at matching loans to borrowers who will repay as agreed. For any pool of approved loans, more of the loans go to borrowers who will default, and more borrowers are rejected who would have repaid. The negative impact on worthy borrowers is greatest for those who are young, have short time on the job or at their residence, have lower incomes, and are generally more financially vulnerable. These are precisely the borrowers for

whom the ability to see successful handling of credit on the credit report is most important, to offset attributes that otherwise make them appear to be higher risk.

Barron and Staten (2003) also report simulations of credit reporting that is fragmented by the type of lender originating the loan. This occurs frequently around the world in countries where the evolution of credit data repositories was driven by industry affiliation. In several Latin American countries (Brazil, Mexico, Argentina) as well as Japan, banks historically shared information about their customer loan experience and created databases that captured only bank loan experience. Non-bank creditors often were not allowed to draw information from the repository, so pooled their own loan experience to form a separate database. Barron and Staten simulate a segmented reporting system by comparing the fullfile scenario to a reporting system in which only credit experience compiled by retailers was available to a lender. The simulations demonstrated that fragmented reporting also results in significant deterioration of a scoring model's predictive power.

A corollary to these findings, and one that can be demonstrated statistically through simulations like the ones described above, is that in any reporting system, the more lenders that report, the better. The intuition follows directly from the results above. Lenders who choose not to report their customers' experience, or report only negative information and omit accounts in good standing, impose a predictive drag on the rest of the data. Credit reporting

is a good example of an action that produces what economists call "externalities." A non-reporting creditor, especially a large one, imposes an "external" cost on all other creditors, because omission of its data impairs the value of the remaining credit bureau information for everyone. The predictive power of models built with reported data is diminished, and the fog of uncertainty surrounding a given borrower is a little bit greater when lenders know that the consumer may have an account (of unknown size and payment status) with a non-reporting creditor.

The simulation approach developed by Barron and Staten has been applied by other researchers to different databases and other reporting environments. A study by Turner (2003) used commercially available scoring models and U.S. data to show the impact of a negative-only reporting environment in which only serious delinquencies (90 days past due or worse) are reported. Powell, Mylenko, Miller and Majnoni (2004) used credit report data from Brazil and Argentina and found dramatic increases in loan acceptance rates resulting from full-file information. Still another study by Turner (2007) used Colombian credit files and a commercial scoring model to show the predictive boost from the availability of full-file information. Although the magnitude of the gain from full-file reporting varies across these studies due to differences in the depth and quality of the information in the respective credit reports across countries, all the studies showed a significant gain to risk assessment when full-file information is used rather than negative-only information.9

7

? 2008 TransUnion LLC All Rights Reserved

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download