The Community Reinvestment Act is a law originally passed in 1977 that directed federal regulatory agencies to ensure that the banks they supervised were not discriminating against particular communities in making credit available.The onset of the subprime mortgage crisis triggered a flood of sloppy, lazy attacks on the CRA claiming that since the crisis was created by excess lending to the poor, and the CRA was intended to increase lending to the poor, the CRA must have caused the crisis. These arguments suffered from a mistaken premise (subprime lending had a modest negative correlation with income, but many subprime loans were used by the middle class to buy expensive houses in the suburbs and exurbs of California and Nevada) and a failure to check their facts (“Only six percent of all the higher-priced loans were extended by CRA-covered lenders to lower-income borrowers or neighborhoods in their CRA assessment areas, the local geographies that are the primary focus for CRA evaluation purposes.” — Randall Kroszner, former Fed governor appointed by President George W. Bush, in a Federal Reserve study that also found that subprime loan performance was no worse in CRA-covered zip codes than in slightly more affluent zip codes not covered by the CRA.)
Yesterday at a Cato Institute conference, Edward Pinto, chief credit officer at Fannie from 1987 to 1989 and currently a real estate financial services industry consultant (according to recent Congressional testimony), rolled out the new line. The new argument is a curious mirror image of the old argument (which Pinto himself may not have made): now the subprime explosion did not cause the housing bubble, but was caused by the housing bubble and … wait for it … the CRA caused the housing bubble, along with the affordable housing goals of Fannie and Freddie.
Before going further, it’s time for my favorite lesson on correlation and causality.
The idea that the housing bubble caused the explosion in subprime lending is not crazy. The worst excesses in mortgage lending happened in 2003-06, after housing prices had already reached historical highs. The idea is that with prices so high, lenders had to offer exotic mortgages (and stop checking for documentation) in order to make the houses affordable for new borrowers. (Of course, there should have been stronger safeguards against those exotic mortgages — consumer protection enforcement, better credit rating agencies, etc. — but that’s another topic.)
But the weirder part of the argument is that the CRA caused the housing bubble. A policy could push housing prices up by increasing the availability of credit in a way that increases borrowers’ buying power. However, that can only contribute to a bubble if (a) it increases the number of loans that cannot be paid off, making price rises unsustainable and (b) there is some continually-increasing aspect to the policy, without which prices should simply reset at a higher level.
Pinto gets into an argument with the Federal Reserve study (cited above) on the performance of CRA-covered loans, claiming that those loans are doing worse than the Fed claims; I can’t judge that without seeing something in more detail. But even so, there are a few missing elements to the causal chain. One is that the CRA should only have an effect in low-income communities, and unless the people buying houses in the Nevada desert were all people who had been priced out of low-income communities by the CRA, it’s hard to blame the real housing price craziness on the CRA. Another is that the CRA itself has provisions that say that lenders do not have to make loans that are unprofitable. A third is that if the CRA was forcing banks to make unprofitable loans, then you would expect the nonbank lenders to stay out of those market segments; in fact, we saw just the opposite.
Back in 2000, Cato had a different line on the CRA. Jeffrey Gunther wrote an article in a Cato journal arguing that the CRA should stand for “Community Redundancy Act” because competitive forces in the market made it unnecessary — lenders seeking profits would not discriminate against particular communities. Gunther cited subprime lending as an example of the type of profit-seeking innovation that made the CRA unnecessary. He noted exactly what CRA defenders argue today:
“If CRA were the driving force behind the recent increases in home-purchase lending in low-income neighborhoods, we would see evidence of a treatment effect. Lenders subject to the ‘CRA treatment’ [regulated banks] would have refocused their activity toward CRA objectives to a greater extent than lenders in the untreated control group [nonbank lenders]. However, there is little evidence of such a treatment effect. To the contrary, it was lenders in the control group that refocused their efforts in line with the mid-1990s boom in lending in low-income neighborhoods. In fact, lending in low-income neighborhoods grew faster than other types of lending at institutions not covered by CRA, whereas low-income lending grew at the same rate as other types of lending activity for CRA-covered lenders.”
Gunther’s optimism about subprime lending seems naive in hindsight, although it was shared by many prominent economists and policymakers from Alan Greenspan on down.
For the CRA to be the problem, the causal factor would have to be availability of credit in low-income communities. But from what I’ve read, it seems like today’s problem is no longer redlining — plenty of lenders were willing to lend to the poor. It’s predatory lending — they found that for various reasons it was easier to steer poor people into unnecessarily high-cost loans. Now, I’m no fan of policies to encourage homeownership in general. I think we have too many of them. But the CRA is primarily a policy to discourage discrimination, and that is something we unfortunately still need.
By James Kwak