The Boston Fed study
Ed Morrissey nods approvingly at Mayor Bloomberg’s comments on the mortgage crisis. He quotes from an Investor’s Business Daily piece criticizing a 1992 study by the Boston Federal Reserve on housing discrimination:
The unusual full-court press was predicated on a Boston Fed study showing mortgage lenders rejecting blacks and Hispanics in greater proportion than whites. The author of the 1992 study, hired by the Clinton White House, claimed it was racial “discrimination.” But it was simply good underwriting.
It took private analysts, as well as at least one FDIC economist, little time to determine the Boston Fed study was terminally flawed. In addition to finding embarrassing mistakes in the data, they concluded that more relevant measures of a borrower’s credit history — such as past delinquencies and whether the borrower met lenders credit standards — explained the gap in lending between whites and blacks, who on average had poorer credit and higher defaults.
The study did not take into account a host of other relevant data factoring into denials, including applicants’ net worth, debt burden and employment record. Other variables, such as the size of down payments and the amount of the loans sought to the value of the property being bought, also were left out of the analysis. It also failed to consider whether the borrower submitted information that could not be verified, the price of a cosigner and even the loan amount.
When these missing data were factored in, it became clear that the rejection rates were based on legitimate business decisions, not racism.
This topic was addressed by Jeff Madrick and Frank Partnoy in a recent review of Gretchen Morgenson and Joshua Rauser’s Reckless Endangerment: How Outsized Ambition, Greed, and Corruption Led to Economic Armageddon in the New York Review of Books. They write:
Their scathing criticism of a Federal Reserve Bank of Boston study published in 1992, which demonstrated prejudice against minorities in the distribution of mortgages in the Boston area, is an especially disturbing example of their one-sided reporting. They assert that this study, which they say influenced Congress’s adoption of affordable lending goals, was deeply flawed. They mock the primary author of the study, the economist Alicia Munnell, and state that the Boston Fed made “a fool of itself.”
But they don’t point out that the Boston Fed’s study was later subject to a stringent peer-review process and was published in 1996 in the respected American Economic Review. Indeed, few research papers have been as closely scrutinized. The published results, which corrected some methodological errors, showed persuasively that there was in fact a bias in the mortgage markets. In 1998, another peer-reviewed paper—also ignored by the authors—analyzed the criticisms of the Boston Fed study, as well as other research on mortgage bias, and concluded that the study was fundamentally correct. “The bottom line is that the results related to race are extremely robust,” wrote the author.