The New York Times suggests that the problem is the mortgage brokers; immigrants
and African Americans don't trust banks for a variety of reasons. And they did
trust mortgage brokers who were members of their communities, and steered them
to expensive loans that earned fat commissions.
I don't say that this
isn't the problem--I'm sure it's a least some part of it. But there's a problem
with this sort of analysis. There are a number of different metrics that go into
loan quality, and therefore what a buyer should pay for their loan:
Expenses, especially outstanding debt
Most of the studies I've seen
indicating that minorities are steered into pricier loans look at just one, or
at most two, of these factors. But they all matter, particularly the size of the
This isn't the first time McArdle has brought up this subject; it's at least the third. "It is not plausible to argue that the banks knew the loans would go bad . . . and nonetheless jammed billions of them into their portfolios, McArdle sniffed in December 2007, in response to "Nobel-prize-winning economist Gary Becker"'s assertion. "So far I've seen little evidence that, taking these things into account, banks are discriminating against minority borrowers," she insisted after a New York Times article about loan discrimination in January 2008. And in July 2008 she still didn't see a problem after another blogger discussed a similar credit bias situation.
If the companies were statistically discriminating against African Americans,
giving them worse loan terms than they really qualify for, they should be paying
off those loans at higher rates than whites.
They're not. Most of the aggregate research I've seen fails to reject the null
hypothesis that there is no discrimination in loan markets, which means that if
there is discrimination, it is not catching huge numbers of people who are more
likely than their loan terms would suggest to pay their bills on time. Just to
be clear, we're not talking about research that says that blacks who get a
higher interest rate don't pay off at the same rate as whites who get a lower
one--you can't blame the default rate on the higher interest rate. We're talking
about the fact that minorities do not outperform their own loan class. If loan
companies really were discriminating, issuing sub-prime mortgages and car loans
to credit-worthy minorities should be a license to print money.
What a mess. "Most of the aggregate research." "Reject the null hypothesis." "Sub-prime mortgages to minorities should be a license to print money." "There's a problem with this sort of analysis." "Most of the studies I've seen indicating that minorities are steered into pricier loans look at just one, or at most two, of these factors. But they all matter, particularly the size of the loan." Weasel words, free of fact and analysis. McArdle doesn't refute facts, she hen-pecks at the methods used to gather information. That way she doesn't actually have to prove anything, she just casts enough aspersions on the data to confuse the issue. When source after source after source after source brings up a problem, dismissing it out of hand begins to look like bigotry and callous indifference instead of honest disagreement.
She was corrected then and she might be corrected again, but facts don't matter. In the Free Market Fantasyland in McArdle's head, a company will never do anything short-sighted or underhanded because the Free Market Fairy will punish it. Reality is unimportant--the fantasy must live on.