Turning ‘No Hits’ Into Home Runs

By all appearances, financial institutions have never had it better. Not only has credit card usage tripled since 1990, but home mortgage loans topped $6 trillion in 2002¹. But appearances can be deceiving. As household loans have grown, so has market saturation, especially in the credit card, new purchase mortgage and general loan sectors.

To continue enjoying steady growth, financial institutions must find new revenue streams. Many are finding those sources in traditionally underserved markets the growing Hispanic community, the exploding immigrant population, and other traditionally underserved minority groups.

These growing populations are logical places to look for corporate growth. According to the U.S. Census Bureau, minorities will account for nearly 80 percent of the total population increase of the United States through 2010. The nation’s Hispanic population alone increased 58 percent between the 1990 and 2000 censuses, accounting for 40 percent of the increase of the nation’s total population during that period. Further sweetening the pot, the Census Bureau also estimates that 33.1 million immigrants live in the United States, an increase of 2 million since the 2000 Census.

The potential is staggering. In addition to steady growth in these underserved populations, immigrants are three times as likely as all adults to rank home buying as their top priority².

As lucrative as this scenario is, penetrating this largely untapped market is rife with challenges. Culturally, minority groups often distrust commercial financial institutions, preferring instead to save their money in nontraditional ways either by keeping it in the home or investing in a cultural savings club, which often replaces traditional savings accounts and lending methods.

But the cultural challenges associated with attracting these underserved groups pale in comparison to the practical and logistical challenges. For the general marketplace, financial institutions have long relied on information from credit bureaus to validate loan applicants. These well-established bureaus collect credit data from a variety of credit grantors. Combining that data with information about length of home ownership and employment, the credit bureau uses existing standards to interpret the data and translate it into credit risk scores.

That tried-and-true method of credit evaluation, however, doesn’t work with underserved populations. Because many people from underserved populations have been in the United States a short time or operate within their own communities, they often have little or no experience with traditional financial institutions-and little or no credit ratings for credit bureaus to use as a source of credit validation. Often, the result of submitting these applicants to the traditional credit process is a “thin file” or “no hit”, meaning that little or no data is available. Based on this lack of information, financial institutions often reject the applicant even one that may have made an excellent customer if approved.

Realizing the significant potential of this growing community, forward-thinking lenders are using creative thinking to rework their credit evaluation processes. By incorporating nontraditional sources of credit information, they believe they may be able to better identify and process credit-worthy customers automatically, much like traditional credit evaluation

Sources of nontraditional credit information may include:

• Public records, such as court records, drivers licenses, evictions databases, tax liens and vehicle registration
• Private consortium data, such as debit bureau databases, insurance companies, Non-U.S. credit bureaus and sub-prime credit bureaus
• Rental/utilities, such as cable companies, household utilities, landlord databases and telephone companies
• Retail/marketing data, such as loyalty databases, rent-to-own center databases, retailer lists and surveys/product registration

Although accessing these nontraditional sources of data clearly is a great way to augment credit evaluation for potential customers in underserved markets, creating the infrastructure to do so is daunting. Not only does it involve choosing the best sources from the hundreds available, but it involves convincing the organizations to share data, building the pipes to receive the data, and reworking existing credit evaluation models to incorporate new types of data and criteria.

The largest financial institutions may be able to undertake such large and complex projects, but for most, the process is expensive and time-consuming. For these institutions, it often makes sense to contract with an outside consulting firm, who can build gateways from the financial institution to a central hub, which processes vital information about the applicant and uses a pre-determined set of intelligent business rules to determine which nontraditional data sources to query in order to make the credit determination. The data resulting from this complex search can be returned to the lender either as raw data, an actual credit score, or a further action the institution should take, such as requesting specific additional information from the applicant. This structure generally works on a per-transaction or monthly fee basis, making it more cost-effective for financial institutions.

To make this process successful, CIOs should engage in the following steps:

• Choose the best and most appropriate sources of nontraditional data from the hundreds available. The institution’s management team can rank the sources based on its own internal criteria or can use a consulting firm’s core group of pre-approved sources, adding others specific to their business model.
• Negotiate and build relationships with each chosen data provider. This can be time-consuming and frustrating, not only because of the vast number of potential data providers, but because of many providers’ reluctance to participate in this process. Utilities-one of the best sources of nontraditional data-often hesitate to participate in these projects because of privacy laws, potential conflict with the Fair Credit Reporting Act, and lack of incentive. Large financial institutions may have better luck convincing these utilities, as would consulting firms with numerous financial customers and infrastructure.
• Develop intelligent business rules for data evaluation. Either with the help of a consulting firm or with internal staff, financial executives develop intelligent business rules that will govern how a given applicant might be handled and what data sources might be tapped. Here is a simple example: If an applicant applying for a mortgage lives in Minnesota, receives a “No Hit” or “Thin File” response from a credit bureau and has lived in Minnesota for less than six months, the business rules might dictate searching the following databases:a. Minnesota driver’s license bureau

  b. Debit bureau, such as eFunds

  c. Midwest Utilities Data Consortium

  d. National marketing residence database, such as Acxiom Residence file

  Once the system has searched the chosen databases, a business rule might direct it to return the following information:

  a. Number of years the applicant has held a Minnesota driver’s license

  b. Applicant’s debit score

  c. Number of times the applicant has been delinquent on a utility account during the past six months

  d. Applicant’s last three addresses

• Build pipes to receive the data. This can be expensive and time-consuming, but with a competent IT staff, it’s possible to build these pipes in-house. Alternatively, you can contract with an Application Services Provider (ASP) to provide those pipes. By using a hosted environment infrastructure integrated with your existing lending processes, you will help protect your process from the stress of constant change.
• Rework existing credit valuation models to incorporate new types of data and criteria. Virtually all financial institutions today use automated predictive models to evaluate information received from credit bureaus and other sources. These work well for traditional segments of the marketplace, but may not handle nontraditional information properly. Financial institutions must rethink what constitutes an acceptable credit risk within the underserved market, revising credit scoring mechanisms. Although some institutions may be able to rework existing models to incorporate these features and functions, others may be best served by developing new models to handle underserved markets and nontraditional data sources. In general, this requires a combination of IT staff and financial institution leadership.

These methods, although daunting, are beginning to work well in financial institutions that have decided to move forward. Here is some advice to ease the transition:

• Before contracting with a data provider, create a test with the provider using your own data to see what type of hit rate you achieve.
• Think beyond the traditional ways of assessing credit risk that have been commonplace for the past 20 years and start thinking how to creatively use new and different information in “risk appropriate” ways.
• Make sure your staff knows how to use and interpret this new data.

Although moving to this new model may result in some initial challenges, the benefits of new revenue streams, cost reductions and improved customer satisfaction show promise to outweigh the short-term pain. If done properly, leveraging new, nontraditional data sources can greatly increase revenue, chiefly from more loans and more customers-including underserved markets, expanded relationships with existing customers, increased fees, and larger loans. And, there is a “win-win” situation. Not only will lenders reap the benefits of new revenue streams, but many consumers, primarily those from underserved markets, will be able to realize their “American Dream” of owning their own home.

By tapping these traditionally underserved markets, financial institutions will continue to enjoy steady growth into the next century.

¹U.S. Federal Reserve Board of Governors, U.S. Department of Commerce

²“Reaching, Emerging, and Underserved Home Ownership Markets” by Andrew I Schoenholtz and Kristin Stanton Jones of The Georgetown University, Institute for the Study of International Migration, February, 2003.

Walter Howell, Executive Vice President for AMS’s Commercial Sector, directs the company’s global IT business with financial institutions as well as communications, media and entertainment organizations. Mr. Howell has more than 25 years experience in strategizing, managing, and directing global IT initiatives. For more information on AMS’s solutions in the financial services space, visit http://www.ams.com/credit/.