Mandate on Department of Housing and Urban Development's Alternative Credit Pilot Program

Gao ID: GAO-10-876R July 30, 2010

When potential borrowers apply for a mortgage loan, lenders typically use borrowers' credit scores--which are based on their credit payment histories, debt, length of credit history, new credit accounts or requests, and types of credit used--to help determine their creditworthiness. However, some borrowers have limited or no credit histories, and lenders are unable to determine their creditworthiness using this traditional method. For these nontraditional borrowers, lenders may assess creditworthiness through alternative means, including the compilation of performance on rental payments; utility, phone, and cable television bills; and insurance or tuition payments. While some lenders have developed statistical scoring methods to determine borrowers' creditworthiness based on these data, others rely on the judgment of their staff to make determinations on a case-by-case basis, according to Department of Housing and Urban Development (HUD) officials. HUD permits lenders that originate mortgages insured by HUD's Federal Housing Administration (FHA) to establish a borrower's credit history through alternative means and has provided guidance to FHA lenders for evaluating these nontraditional credit histories. According to HUD officials, 7,319 nontraditional borrowers were approved for FHA-insured mortgages in fiscal year 2009. Section 2124 of the Housing and Economic Recovery Act of 2008 (HERA) requires HUD to develop a pilot program establishing an automated process for providing FHA lenders with alternative credit rating information for borrowers who have insufficient credit histories to determine their creditworthiness using traditional methods. Section 2124 of HERA also directs GAO to report by July 30, 2010, on (1) the number of additional borrowers served using the automated process and (2) the impact of the process on the safety and soundness of FHA's insurance funds.

In July 2009, HUD published a notice in the Federal Register to solicit comments on the design of the pilot program. In September 2009, HUD awarded a contract to a consulting firm to study the feasibility of developing an automated process and to propose a design for the pilot program. The contractor reported to HUD in June 2010 on its findings and concluded that there are sufficient sources of alternative credit data--and scoring methods using such data--to move forward with the pilot program. Specifically, the contractor noted that several vendors aggregate data that are often used for making credit decisions through alternative means and that cover a large portion of the nontraditional borrower population. In addition, other vendors have developed methodologies and tools to analyze these data to produce nontraditional credit scores, which could be used to support the pilot program if they can be independently validated as being effective for managing credit risk. Based on these findings, in July 2010 the contractor recommended that HUD implement the pilot program in two phases. In the first phase, called the proof-of-concept phase, the contractor proposed to examine the performance of existing FHA-insured mortgages that were approved using alternative credit information. Specifically, the contractor would analyze statistical relationships between mortgage performance and alternative credit data to determine whether there are indicators, criteria, and parameters that could be used in an automated decision tool, according to HUD officials. The contractor recommended allowing 300 days for the completion of the proof-of-concept phase. If the proof-of-concept phase is successful, they will move to a production phase, during which the contractor would create an automated decision tool based on their findings and HUD would recruit lenders to participate. This second phase would continue for several years in order to evaluate the performance of the mortgages approved using the decision tool. HUD officials said they have reviewed the contractor's reports and agree with the findings and recommendations. They plan to exercise the option year on their contract with the consulting firm in order to begin the proof-of-concept phase in fiscal year 2011, provided that funding is available.



The Justia Government Accountability Office site republishes public reports retrieved from the U.S. GAO These reports should not be considered official, and do not necessarily reflect the views of Justia.