Oh, Give Me A Home

By Tim Wise, The Black Commentator. Posted June 7, 2005.

According to government estimates, between two and three million cases of housing discrimination take place each year, victimizing persons of color who seek to either purchase or rent a place to live. This discrimination takes several different forms, some of which are rather blatant, and others of which appear more subtle: claiming that the last apartment was just rented, when it fact it wasn't; charging higher interest rates for the same mortgage loan offered to a white person on better terms, or flatly rejecting persons of color for a loan at all, even while offering the same loans to whites with similar credit.

Yet, despite decades of studies confirming the existence of housing bias, conservatives regularly devise and offer excuses for disparate housing outcomes, all of which presume that there are logical reasons why folks of color get loans less often, or on less favorable terms, and none of which reasons have the least bit to do with racism. As with white denial of racism in labor markets, the refusal of many in the white community to acknowledge the existence of bias in housing signifies an overwhelming need to rationalize inequality in the U.S.

Although skeptics downplay discrimination against persons seeking to buy a home, by saying that higher mortgage rejection rates for blacks are merely the result of having worse credit histories, the facts say otherwise.

The most comprehensive study of mortgage bias was conducted by the Boston Federal Reserve Bank and considered 38 different factors that could result in disparate lending. Among the factors examined were several measures of income, credit history, loan type and collateral. Even with all factors considered, blacks were still nearly 60 percent more likely to be rejected for a mortgage compared to similarly credit-worthy whites. Despite criticisms of the Boston Fed study, the research has held up to extensive scrutiny. Indeed, the methodology of the study was considered sound enough by Boston banks so as to gain their participation to begin with, no doubt because they expected (incorrectly as it turned out) that the research would exonerate them from claims of bias. Furthermore, two follow-up studies, both of which added control variables, found equal or higher levels of bias than were found in the original study.

Another study in Louisville sent black and white "testers" to banks with equal credit ratings and financial characteristics, and had them request conventional mortgages for the very same housing. Repeatedly, blacks were given less information or encouragement to apply, and were subjected to differential and unequal treatment in terms of loan prequalification. For example, blacks were often told their income and credit was inadequate to qualify for the loans they sought, while whites with identical incomes and credit were told they would qualify for the same loans.

Conservatives criticize studies that find evidence of mortgage bias, based on different outcomes for persons at the same credit rating, by arguing that default analysis shows different outcomes are justified. Specifically, they argue that since black default rates are higher than the rates for whites, at every level of pre-loan creditworthiness, banks are merely engaging in rational decision making when they reject blacks for such loans, aware that the risk of default is higher. But there are multiple flaws with this line of reasoning.

First, this argument ignores that default rates and foreclosure rates are far from the same, and it is only when loans are foreclosed that their default status becomes visible in data. Secondly, lenders control whether or not a late loan (technically in default) is going to be called in or not, and the available evidence suggests lenders are more aggressive in foreclosing on loans paid late by blacks than whites. In part, this is due to the ability to turn the lower-cost homes (with higher than average loan-to-value ratios) more quickly for greater profit once the loan is called in.

Another problem with the default analysis approach is that it assumes that since blacks are higher average credit and default risks, therefore, there is no discrimination when a particular black applicant for a loan gets turned down. But this argument extrapolates from group averages to individuals in a way that is not only illegal (it is unlawful to discriminate against a person because of the average characteristics of that person's racial group), but also irrational. After all, just because blacks as a group have higher default rates, doesn't mean that any given black loan applicant will likely default, and to treat them as if they would is to treat them on the basis of a statistical average over which they have no control, and which is likely to be wrong far more often than right (since most blacks will not default on their loans).

Finally, when loan default rates are not massively different between whites and blacks (and they aren't), lenders should care more about the average loss on a default, rather than the average rate of default between one type of borrower and another. As such, it is important to note that the expected monetary loss to a lender from a defaulted black loan is actually less than the average for loans to whites (understandable, since the size of the loan in the latter case is likely higher), so market theory would predict lower rates of foreclosure on black loans if discrimination were not operating.

Tim Wise is the author of White Like Me: Reflections on Race from a Privileged Son (Soft Skull, 2005) and Affirmative Action: Racial Preference in Black and White (Routledge, 2005). This is part three of a four-part series. Footnotes for this article can be obtained from the author at timjwise@msn.com. His writings can be found at www.timwise.org.

Liked this story? Get top stories in your inbox each week from AlterNet! Sign up now »