Chapter 7

Keeping the Promise

America has the best housing finance system in the world, and automated underwriting is poised to make it better still. By making mortgage lending simpler, fairer and more affordable, automated underwriting promises to help put hundreds of thousands more American families into homes they can afford and keep.

To keep pace with the changing mortgage marketplace, automated underwriting will undergo continuous updating and refinement. Building on the benefits these systems already have brought, Freddie Mac is introducing automated underwriting to new markets and experimenting with new applications of the technology. We also are using the knowledge gained in the development of Loan Prospector to improve today’s manual underwriting and offer new products to mortgage borrowers.

Expanding Access to Automated Underwriting

Freddie Mac, in cooperation with industry participants, is exploring a wide range of applications for Loan Prospector. For example, our pilot programs will explore using automated underwriting to identify borrowers who would benefit most from homeownership counseling as well as from intensive mortgage servicing.

In July 1996, Freddie Mac and HUD launched a pilot program to evaluate benefits that automated underwriting could bring to FHA borrowers. In announcing the pilot, HUD Secretary Henry Cisneros noted:

This pilot represents the National Partners in Homeownership at its best . . . . This pilot brings together Freddie Mac, mortgage lenders and FHA in a test of how new technologies can make FHA work better for more American families.

Working with HUD, Freddie Mac built a version of Loan Prospector that provides risk classifications based on actual FHA default insurance claims. Ten pilot participants—representing state housing finance agencies, nonprofit housing groups and large and small mortgage lenders—will join with HUD and Freddie Mac to assess the benefits and possibilities of the new automated underwriting service.

Continual Refinement

Developing Loan Prospector expanded our understanding of which loans perform well—and why. As a result, Freddie Mac is able to offer new mortgage products, such as mortgages with down payments as low as 3 percent. Knowledge gained about the layering of risks and the factors that offset those risks was key in making this traditionally high-risk mortgage product available to borrowers.

Over time, improvements in Loan Prospector will generate new insights and lending opportunities. For example, Freddie Mac participates in the NeighborWorks Campaign for Homeownership®, which provides extensive pre- and post-purchase homeownership counseling. Initiatives like these ultimately will generate data that can be used to assess the impact of counseling on loan performance. Potentially, future versions of Loan Prospector could be modified to recognize counseling as a compensating factor.1

Opening Doors to Homeownership

Sprinkled throughout these pages are real-life examples of how automated underwriting helps families for whom traditional loan-approval methods were not working. Stories such as theirs will proliferate in the coming years:

• Louise Beyler, a self-employed borrower whose debt ratios seemed too high to qualify her for a mortgage, succeeded in buying her home
  
• Edward and Margaret Hartman, whose previous loan did not close until the moving van was arriving, got loan approval in two days
  
• Alejandro Hernandez, whose loan application was a study in complex layered risks, avoided a brush with loan rejection
  
• Dawn Clark, whose difficult loan- origination experience led her to question whether she, as an African-American, had been treated fairly, will find the future more reassuring
  
• Ray Evans, who initially was told he did not qualify for a conventional loan because his down payment was too low, got one after he switched to a lender using automated underwriting
  

By accurately assessing risk and accounting for compensating factors, automated underwriting will provide lenders with added confidence to serve more and more borrowers.

Keeping the Promise

All told, highly predictive statistically based systems such as Loan Prospector promise to open doors to homeownership for nearly three-quarters of a million new borrowers, 80 percent of whom will be minority and low-income families:

• 70,000 additional renters will be able to afford their first homes because of the reduction in closing costs
  
• 400,000 additional minority borrowers will succeed in obtaining mortgages because of a level playing field
  
• 250,000 additional borrowers who did not fit the standard mold will be able to obtain a mortgage
  

The cost savings resulting from automated underwriting will be substantial:

• Had every mortgage cost $400 less to originate in 1995 as a result of automated underwriting, borrowers could have saved $2 billion
  
• Had every subprime mortgage undergone automated underwriting review in 1995, the borrowers qualifying for conventional financing could have saved up to $100 million a year in reduced interest costs
  

With Loan Prospector, Freddie Mac is building on a 25-year history of improving the mortgage market by reducing costs and expanding the market to more borrowers.

The people of Freddie Mac are committed to seeing that automated underwriting lives up to its promise. We are proud to play a pioneering role in making the world’s best housing finance system even better.

1. Beth Prentice, “Automated Underwriting: Friend or Foe?” Stone Soup, The NeighborWorks® Partnership Report, Summer 1996. As a Neighborhood Reinvestment Corporation director recently concluded: “Opinions . . . are not good enough. We need cold, hard facts to demonstrate clearly the risk mitigation provided by comprehensive counseling systems such as NeighborWorks® Full Cycle LendingSM. . . . [M]y opinion is that a lending system that pays more attention to a borrower’s proven credit experience and attitude can be a positive trend for lower-income families.” Also see Roberto G. Quercia and Susan M. Wachter, “Homeownership Counseling Performance: How Can It Be Measured?” Housing Policy Debate, Volume 7, Issue 1, 1996.