Minority Subprime Borrowers - Minorities Pay More for Home Ownership

Minority Subprime Borrowers
October 2002
Consumers Union Southwest Regional Office

This study reinforces (for Texas) the findings of several national studies: race matters. The race/ethnicity of borrowers is a powerful factor in the penetration of subprime lending in Texas communities. Our study shows that subprime loans are concentrated in geographical areas with a higher concentration of minority residents. Even after accounting for other factors, the likelihood of getting a subprime loan increases for minority borrowers, especially Black borrowers. Among higher income borrowers, the distinction between subprime lending to Whites and subprime lending to minorities is stark.

The Consumers Union study

This study analyzes Texas refinance loans from 1997 to 2000. We identify subprime lending (loans from HUD identified subprime companies) at the state and city (MSA) level. With 2000 census information about tracts, we identify patterns of subprime lending using standard regression techniques and cross tabulations.(1)

To compare subprime and prime loans by the race, income and ratio of loan size to income at the MSA level, we combine four years of Home Mortgage Disclosure Act (HMDA) data for some of the analysis. This ensures a large enough loan pool for comparison even among smaller subgroups of borrowers. This means that the absolute ratio we calculate for supprime penetration is significantly lower than the 2000 subprime penetration rates calculated by other researchers.(2)

To gain an understanding of point-of-sale lending practices that might lead large numbers of people into the high priced subprime market, we also reviewed consumer complaints filed with the Office of the Consumer Credit Commissioner and interviewed individual consumers. This information is essentially anecdotal and incorporated here only to suggest possible reasons why so many people take high priced home loans.

Most vulnerable neighborhoods

In our first report in this series, Consumers Union calculated that residents who refinance their homes using subprime loans pay an estimated average $1,944 per year in additional interest alone.(3) Here we find that the people who most often pay the "subprime premium" live in lower income, high minority neighborhoods with a higher than average share of elderly residents.

While subprime companies currently control about a third of the refinance loan market statewide, they take a much larger share of refinance loans in some communities. Consumers Union examined 643 Texas census tracts with the highest subprime penetration (over 54 percent subprime), and compared these tracts to tracts with median subprime penetration or less. We found that residents in high subprime tracts were older (more people over 65), had lower incomes and were nearly 80 percent minority. People in these areas can least afford to pay so much more every year for a home secured loan.(4)

Minority borrowers hit hardest

This study of HMDA data in Texas confirms the findings of numerous studies released across the country: race of the borrower is probably the strongest factor distinguishing loans with subprime companies.

The American Association of Retired Persons (AARP) recently reported that Texas and New York enjoy the highest subprime penetration rates in the nation.(5) The Center for Community Change (CCC) reports that Houston, San Antonio and Austin rank in the top ten cities in the nation for subprime penetration among Black borrowers. The top cities in the nation for subprime penetration among Hispanic borrowers are all in Texas: El Paso, San Antonio and Corpus Christi.(6)

Confirming the findings of CCC, our analysis demonstrates that minority borrowers disproportionately took refinance loans from subprime companies over the past four years-and race of the borrower is a powerful factor predicting whether a person will get a prime or subprime loan.

Even among higher income people, minority borrowers took significantly higher share of subprime loans

We find a large disparity between Black and White borrowers in the subprime arena-and an equally significant disparity between predominantly White and minority neighborhoods-even among higher income people. By 2000, 58.1 percent of Black borrowers took refinance loans from subprime lenders, compared to 39.8 percent of Hispanic borrowers and 23.5 percent of White borrowers.(7)

Mr. Smith of Dallas owned his home outright when he decided to get a refinance loan to fix it up. He borrowed $42,000 from Full Spectrum Lending, the subprime affiliate of Countrywide, and used it to pay off a car and credit cards.

"We was kind of hurried when we got that loan," he told Consumers Union. "When we finally looked at it, we realized it was high." Based on an ad from his credit card company, Mr. Smith borrowed at 15 percent APR interest. In November 2000, when he took out this loan, conventional mortgage interest rates averaged 7.75 percent, according to the Federal Reserve.

"I was looking through the papers, but they didn't explain the percentage rate," he said. "I had thousands of papers, God Almighty!...I don't think I asked the right questions. It ought to be explained to you. I had to go over them and over them, and over the phone [with the loan company]."

According to Mr. Smith, he has a credit score of around 600. "I have one little item from when I was in the hospital and I just didn't agree with the charge. I had a dispute and I'm not going to pay it, so they sent it to the credit bureau."

He recently decided to get rid of this high cost loan by refinancing with a new company at only 8.11 percent APR. Unfortunately, his payoff amount at the time he closed the new loan was actually higher than the $42,000 he originally borrowed and he's still trying to figure out why.

Black and Hispanic families in Texas statistically earn less and enjoy fewer assets, which can affect credit scores and loan underwriting. Therefore we analyzed all home purchase and refinance loans to borrowers earning more than $60,000 annually (1.5 times the state's median income) and borrowing less than 2.5 times their reported income. We found that among these higher income people taking a reasonable home refinance loan, minority borrowers still took refinance loans from subprime companies in almost the same proportions. While 16.7 percent of upper income Whites refinanced using a subprime company, 30.9 percent of Hispanics and 46.4 percent of Black borrowers refinanced through a subprime company statewide.

Clearly the minority concentration of a neighborhood is a factor in the rate of subprime refinance. Using four years of data to ensure a large number of minority refinance loans at every minority concentration, Consumers Union plotted the subprime refinance penetration rate for all upper income borrowers (earning more than $60,000), and for upper income minority borrowers.(8) We find that, even for upper income borrowers, use of subprime companies rises significantly as the minority concentration of the neighborhood rises. Upper income minority borrowers used subprime companies slightly more often in all areas.

In a separate analysis, we also looked at upper income Black borrowers living in predominantly Black areas, and upper income Hispanic borrowers living in Hispanic areas. These Black borrowers refinancing their homes in predominantly Black neighborhoods took loans from subprime companies at very high rates, over 40 percent in the most concentrated Black areas. Upper income Hispanic borrowers also took more refinance loans from subprime companies if they live in predominantly Hispanic neighborhoods, but the difference is not as sharp.

Race a powerful factor

Clearly the ethnic composition of the neighborhood is a factor in subprime penetration. To determine the significance of the race of the borrower independent of the neighborhood, Consumers Union predicted the probability that a consumer would end up with a subprime refinance loan holding constant other factors: the gender of the borrower, the minority concentration, elderly concentration and income of the neighborhood, borrower income, loan amount, population density, and the loan to income ratio. Public data contains no information about the credit history of borrowers.(9)

We find Black borrowers are 3.9 times as likely to end up with a refinance loan from a subprime company than non-Blacks (combining all other borrowers), after accounting for all these other factors. We find that Hispanics are 1.6 times as likely to end up with a refinance loan from a subprime company as non-Hispanics (combining all other borrowers).

The power of borrower race to predict that the borrower will get a loan from a subprime company shores up the findings of our recently released studies on women borrowers and elderly borrowers. We previously reported that women get loans from subprime companies more than men, but that subprime companies enjoy their highest penetration rates among Black women borrowers and in high elderly/high minority neighborhoods.

Race at the MSA level

In order to look fairly at all Texas towns and cities, including smaller cities with fewer loans to selected minority borrowers, Consumers Union reviewed four years of lending data together. This analysis will present a low actual penetration rate for subprime lending in these cities, because subprime lending grew throughout the four year period. Over the four year period, subprime penetration rates average 17 percent, but in 2000 subprime companies made more than a third of refinance loans.

On the other hand, we can reasonably rank larger and smaller cities together, with some surprising results. In general, subprime penetration ranks highest in Texas' medium size cities, and in many of these cities minority borrowers are far more likely to get subprime loans than White borrowers-with even stronger disparity ratios than large cities like Houston and Dallas.

Gulf Coast and Rio Grande Valley cities have the highest subprime penetration in the state over the four year period. Borrowers in Beaumont-Port Arthur, followed closely by El Paso and McAllen/Edinburg/Mission, took the highest share of subprime refinance loans.

Compared to White borrowers, Black borrowers in Tyler took far more subprime refinance loans (45.3 percent compared to 6.5 percent, or a disparity ratio of 6.99). Black borrowers in Bryan/College Station, San Angelo, and Longview/Marshall also took a far higher share of loans from subprime companies. Among upper income Black and White borrowers, the discrepancies were greatest in the state's largest cities, although this may be partially because there were too few upper income minority borrowers in most smaller cities for appropriate comparison. The Kileen-Temple area, a military town with a significant minority population, has the second largest disparity (after Houston) between subprime lending to Black and White borrowers.

Hispanic borrowers generally took the highest share of subprime loans (compared to white borrowers) in Lubbock, followed by Bryan/College Station and Victoria. Higher income Hispanic borrowers took a disproportionate share of subprime loans in Austin, Houston, and San Antonio compared to upper income White borrowers.

Maps of specific MSAs are perhaps the easiest way to understand the distribution of subprime lending in Texas. This report includes a digital map for review of any Texas MSA. The maps confirm that subprime lenders found a market in the minority, older, and frequently poorer neighborhoods in most Texas cities.

Recommendations

Subprime lending can have disastrous consequences for low income and minority communities. People who take out home refinance loans they cannot afford face foreclosure if they don't keep up the payments. People who started out with good credit can lose their credit if unaffordable payments slip into past due status-even if they keep paying and don't go all the way to foreclosure.

For many minority communities, home equity is the most important way families develop wealth over the long term. High cost refinance turns family wealth into cash, cash that is frequently turned back over to the lender in high loan fees.

To prevent the stripping of equity from the most vulnerable minority families, the Texas Legislature should reduce the high fees and costs associated with home refinance. The AARP, the National Consumer Law Center (NCLC) and others have defined loans as "high cost" if they have an interest rate that equals or exceeds six percentage points over the weekly average yield on five year treasury bills (currently about 3.5 percent but more typically ranging from 4 to 6.5 percent over the period of this study). These groups also define "high cost" as loans that contain fees in excess of three percent of the loan amount.(10) The Texas Legislature should set standards for "high cost" loans:

prohibit the financing of fees, closing costs, or other lender charges (including "prepaid" points) if the fees rise above three percent of the loan amount. Current Texas law limits fees to three percent of the loan amount, but lender origination fees and points are not counted in that cap. Therefore, fees quickly rise to eight or ten percent of the loan and are usually financed by increasing the amount borrowed to cover the higher costs.
require loan counseling for any borrower getting a high cost loan during the existing 12 day waiting period before the loan closes; and
prohibit lending without due regard to repayment ability.

Even borrowers getting subprime loans where the rate does not exceed six percent above the weekly 5-Year treasury bill rate deserve additional protections to preserve their home as an asset. Along with AARP and NCLC, we recommend:
limiting "discount points" to legitimate charges that actually provide a substantial benefit to consumers. The AARP, the Self Help Credit Union and NCLC have created standards for "Bona Fide Discount Points" that would eliminate many of the problem fees consumers face at closing.

________

Notes

1 Owner occupied, single family, refinance home loan records for tracted areas of the state where 2000 census information is available (402,639 loans). We excluded refinance loans made by HUD identified manufactured home specialty lenders. In order to factor in more borrower information and census data, additional records were eliminated for the regression analysis.

2 Bradford, Calvin, Risk or Race? Racial Disparities and the Subprime Refinance Market, Center for Community Change, May 2002. This study finds subprime penetration rates of more than 40 percent in several Texas cities by 2000, and rates of more than 60 percent among Black borrowers in selected Texas MSAs.
The statistics calculated by Bradford differ from those of Consumers Union because Bradford calculated the subprime penetration as a share of conventional loans. Consumers Union calculated loans from subprime companies as a share of all mortgage lending, including FHA, VA, and FmHA. Since minority borrowers also take a disproportionate share of FHA and VA loans, the inclusion of these loan types in the base "market" results in a lower overall estimated penetration of subprime conventional lending for minority borrowers.

Over all, Consumers Union finds that refinance loans by subprime companies grew from 6.3 percent of the statewide refinance market in 1997 to 34.8 percent in 2000. The total number of refinance loans made by subprime companies increased from 2630 loans in 1997 to 23,480 loans in 2000.

3 Women Subprime Borrowers, Consumers Union SWRO, October 2000, p. 2.

4 We split the census tracts that had 2000 census information about age of residents and minority concentration into quartiles ranked by refinance subprime penetration. The top quartile of tracts (643 total) had a subprime penetration of 54.54 percent up to 100 percent of all refinance loans. We calculated the mean minority concentration and population over 65 for these high subprime penetration tracts, and for the tracts with subprime penetration at or below the median (the bottom two quartiles, or 1682 tracts with subprime penetration at or below 40 percent).

5 PCI Services, Inc., "Subprime Refinance Loans as a Ratio of All Refinance Loans," AARP Public Policy Institute, no date.

6 Bradford, Calvin, "Risk or Race? Racial Disparities and the Subprime Refinance Market," The Center for Community Change, May 2002, p. 37 and 42.

7 Subprime lenders also now make a considerable number of home purchase loans. By 2000, 5.1 percent of White borrowers took home purchase loans from subprime companies, compared to 9.5 percent of Hispanic borrowers and 17.2 percent of Black borrowers. None of the top twenty purchase mortgage lenders statewide are subprime companies, but three of the top twenty home purchase lenders to Black borrowers are subprime.

8 Minority borrowers include Black, Hispanic, Asian, Native American, and those who identifed as "other." Where fewer than 30 upper income Black or Hispanic borrowers refinanced their homes over the four year period, we eliminated the area from the "upper income" analysis.

9 Consumers Union's regression model is based on a logistic regression of 262,329 refinance loans originated in Texas over 1997-2000 as reported under HMDA. The base of data is smaller because we eliminated all loans where the lender did not report the race or gender of the borrower, and where income or loan amount were not reported or reported as zero.

The regression controls for the sex, race and income of the borrower, the size of the loan, the size of the loan in relation to the income of the borrower, the presence of a co-applicant, the racial composition and population density of the home's census tract, as well as the relative wealth of that tract to the rest of the MSA. The logistic model successfully predicts the outcome 74 percent of the time.

10 American Association of Retired Persons Public Policy Institute, "Home Loan Protection Act," November 2001.

View Files Sorted By Office: Consumers Union OPI, New York - Washington DC Office
West Coast Regional Office - Southwest Regional Office - Consumer Policy Institute

[ Health ] [ Finance ] [ Food ] [ Product ] [ Other ]
[ About CU ] [ News ] [ Tips ] [ Resources ]
[ New Files ] [ Home ]