The way it works comes down to ratings. The federal regulators who oversee the CRA — the Federal Deposit Insurance Corporation, the Office of the Comptroller of the Currency and the Federal Reserve Board — look at how much banks are lending to low‐income borrowers (those who earn less than 80 percent of median income in their metropolitan area) and to those in low‐income census tracts (below 80 percent of the area’s median income). The regulators assign a rating to each bank and take this rating into consideration when deciding on banks’ applications to expand or merge. Bad CRA performance can cost a bank billions of dollars in forgone business opportunities.
While the goals of the CRA are worthy, policymakers should care about its consequences, since there is no guarantee that the additional credit will reach historically marginalized communities who otherwise would not get it.
The nation’s capital has by some measures experienced the most intense gentrification process of any major American city. A full 22 percent of the capital’s census tracts have seen the arrival of significant numbers of new and more affluent residents since 2000 — a greater proportion than any other U.S. metropolitan area.
And in Washington, D.C., most CRA lending appears to be going not to the underserved but to the people gentrifying their neighborhoods. Of the more than $1.3 billion of CRA‐eligible mortgage loans made in Washington in 2017, 65.5 percent went to borrowers whose incomes exceed 80 percent of the local median. Yet these mortgages earn banks CRA points because the borrowers reside in one of D.C.’s 88 eligible census tracts.
The five blocks of 11th Street described earlier are among them, as they belong to a tract where the median family income ($48,621 in 2017) is only around 50 percent of the D.C. median ($95,995). Yet few of the low‐income families living in those blocks could afford to borrow $800,000 for a home. Instead, it is more likely that home will go to a new, more affluent resident. Not only that, but a bank will get CRA points regardless of whether the borrower is truly underserved.
It is tempting to blame banks for the preponderance of higher‐income borrowers among CRA‐eligible loans. But the fault lies with the incentives created by CRA regulations themselves.
Banks are supposed to lend in low‐income areas without incurring additional risk. Absent this safeguard, public policy would cause banks to make ill‐advised loans, leaving taxpayers to pick up the eventual tab. Even with the statutory language, some evidence has accumulated over the years showing that the CRA does sometimes encourage risky lending. But by directing credit to higher‐income “gentrifiers” in CRA‐eligible areas, banks can meet the letter of the CRA without increasing the amount of likely losses on their balance sheets.
This practice may, however unwittingly, end up accelerating the displacement of poorer residents. Consider, for example, that in five tracts in D.C.’s Park View and Petworth neighborhoods, more than 80 percent of mortgage loan volume in 2017 went to borrowers earning more than the CRA threshold, even though all of these tracts have median incomes well below that threshold.
Can the CRA change to more effectively bring sound credit to low‐income borrowers? Not easily. Politicians and bureaucrats are not well‐placed to weigh credit risk against profitability and social value, as the most recent financial crisis showed. In the run‐up to 2008, successive administrations and congressional acts gradually ratcheted up the share of mortgages guaranteed by Fannie Mae and Freddie Mac that went to low‐income borrowers. The consequence was a $317 billion bailout and an explosion in foreclosures.
To be sure, there are ways in which CRA regulations can be made more transparent and less onerous. Allowing banks to pay third‐party providers such as online lenders and community organizations to fulfill their CRA obligations would simplify compliance, encourage these firms to specialize in underserved areas, and put an explicit price on the costs that the CRA imposes on the financial sector. This price would reflect, among other things, any additional risk involved in CRA lending. Furthermore, a system of tradable obligations would modestly improve upon the present structure of bureaucratic CRA assessments, which account for 7.2 percent of bank compliance costs.
Yet even such a reform would not address the CRA’s bias in favor of higher‐income residents.
For that reason, the usefulness of the CRA should be reconsidered altogether. The spread of large bank branch networks and the rise of nonbank lending since the 1990s have changed the landscape that the CRA’s drafters inhabited. Today, online lenders and credit unions, neither of which are subject to the CRA, make a greater share of their loans to low‐income borrowers than banks do. Furthermore, credit discrimination against protected classes can be better addressed by other laws, such as the Fair Housing and Equal Credit Opportunity acts. Thus, repealing the CRA would not foreclose access to affordable credit by vulnerable communities.
It was never the goal of the CRA to promote lending to high‐income borrowers who can get it without the help of Uncle Sam. Yet that is where much CRA lending now goes. Today, innovation and competition are doing a better job of promoting financial inclusion than a four‐decade‐old law.