Likely Impact of the FRB HOEPA Proposals

In December 2000 the Federal Reserve Board invited public comments on its proposed revisions to Regulation Z that would alter the criteria for a mortgage loan to be covered by the Home Ownership and Equity Protection Act (HOEPA). The proposals were intended to curb a variety of mortgage market abuses that have been termed "predatory lending." However, on numerous occasions since then, the Board has reaffirmed its intent not to impede the general growth of the legitimate subprime mortgage market. The problem is that loans that are deemed HOEPA loans raise the lender's costs, because such status triggers extra disclosures, prohibits certain contractual features and exposes the lender to additional legal liability and regulatory penalties. Consequently, the FRB proposal to expand the number of loans qualifying as HOEPA loans will inevitably raise the costs to subprime lenders.

Setting aside the issue of whether broader HOEPA coverage will reduce predatory tactics, the big question is how many more loans will be impacted by the FRB proposal. Will it affect just a small portion of the subprime market not currently covered by HOEPA, or will lenders' costs be raised for a significant portion of the market? The lack of comprehensive data about either the size of the subprime market or the characteristics of subprime loans has limited the Board's attempts at estimating the extent of current HOEPA coverage (as a proportion of all subprime mortgage loans) or quantifying the proposed expansion of coverage.

A new study sponsored by the American Financial Services Association (AFSA) provides evidence of a much larger impact on the subprime mortgage market than either the FRB or conventional wisdom expected. The report utilizes information from a large population of mortgage loans originated by the subprime divisions of nine major national lenders. The database was compiled by PriceWaterhouseCoopers in late 2000. All of the loans in the data set are closed-end loans secured by residential real estate (either first or second lien). More specifically the data set includes all such loans originated by the subprime divisions of the participating companies between July 1, 1995 and June 30, 2000, a total of 1.4 million loans. How representative are these loans of the entire subprime mortgage market? There are no reliable statistics for the entire market, but we do know that for 1998, the origination volume in the data set was equivalent to about 39 percent of the volume originated that same year by all subprime lenders required to report under the Home Mortgage Disclosure Act.

Using the AFSA database, economists Gregory Elliehausen and Michael Staten (Credit Research Center, McDonough School of Business, Georgetown University) undertook an analysis of the impact of the FRB proposals. Under current rules a loan is covered by HOEPA if it satisfies either part of a two-part test. The first test ("APR test") involves the annual percentage rate (APR) on the loan. If the APR exceeds the rate on a U.S. Treasury security of comparable maturity by 10 percentage points or more, then the loan is subject to HOEPA protections. In the second test ("fees test") loans with non-interest fees that exceed the lesser of 1) 8 percent of the loan amount or 2) a fixed dollar amount that adjusts each year with the consumer price level ($465 in 2001) are covered by HOEPA.

The Board has proposed lowering the APR trigger for HOEPA coverage to 8 percentage points above the Treasury security of comparable maturity. In addition, it has proposed to add premiums on single-premium credit insurance to the fees test. Both the current and proposed HOEPA triggers apply equally to first and second lien mortgages.

Elliehausen and Staten applied both the APR and fees tests to the 1.3 million originated loans in the analysis sample to calculate the percent of loans that would fall under HOEPA coverage under both the current and proposed rules. The chart below displays current HOEPA coverage by type of lien and also illustrates how much of the coverage is attributable to the APR test alone.

During the 1995-2000 sampling period, 12.4 percent of originated first mortgages in the data set were HOEPA loans. Nearly three quarters of these qualified as HOEPA loans based on the APR test alone. A much higher incidence of HOEPA coverage occurs among second mortgages in the data set. Nearly half (49.6 percent) of all second mortgages originated during the 1995-2000 sampling period were HOEPA loans. Higher contract interest rates account for the large majority of the increased coverage. These estimates of current coverage are substantially higher than estimates that the Federal Reserve Board has cited as benchmarks from other surveys.

Printer-Friendly Chart

The next chart ("HOEPA Coverage Under FRB Proposal") displays a dramatic increase in the percent of loans subject to HOEPA protections under the Board's proposal to lower the APR trigger and incorporate premiums on financed credit insurance purchases into the "fees" test. The new proposal would bring 37.6 percent of all first mortgages and 81.1 percent of second mortgages under HOEPA coverage. The chart's separation of the impact of the APR test alone vs. the APR-plus-fees test is particularly instructive under the new proposal. For first mortgages, the proposal to lower the APR trigger to 8 percentage points above the corresponding Treasury security by itself boosts the percent of covered loans from 8.9 percent to 25.8 percent. The inclusion of insurance premiums into the "fees" test affects an additional 12 percent of all first mortgages that apparently have single premium credit insurance but do not have an APR above the trigger point. The chart also shows that under the Board's proposal, a large majority of all second mortgages made during the sample period would have become HOEPA loans. Lowering the APR trigger would boost the percent of second mortgages that were HOEPA loans from 46.8 percent to 74.5 percent. The inclusion of insurance premiums into the "fees" test affects an additional 6.6 percent of loans of all second mortgages that have single premium credit insurance but do not have an APR above the trigger point.

Printer-Friendly Chart

Given the concern that broader HOEPA coverage may reduce the supply of subprime credit, an important policy question is whether the proposed HOEPA regulations impact some groups of borrowers (and loans) differently from others. Generally speaking, analysis revealed that for both first and second lien mortgages, the percent of loans covered by HOEPA (current and proposed) is higher for loans to borrowers who have lower incomes or whose credit history reveals them to be higher risk. HOEPA coverage is also higher for smaller loans and those with shorter terms to maturity. The chart below ("Percent of Loans Covered Under FRB HOEPA Proposal, By Borrower Income") illustrates the disproportionate impact of proposed HOEPA regulations on loans to low-income consumers. Under the Board's proposal 51.8 percent of loans to borrowers with incomes less than $25,000 would be covered by HOEPA vs. 14.3 percent of loans to borrowers with incomes of $75,000 or more. A similar pattern is evident on second mortgages. Under the proposed revisions HOEPA coverage would be extended to 86.2 percent of loans to borrowers with incomes less than $25,000 and 67.2 percent of loans to borrowers with incomes greater than $75,000.

Printer-Friendly Chart