The case for market-based CRA reform

The Community Reinvestment Act was enacted in 1977 to hold banks accountable for “meeting the credit needs” of their local communities in a safe and sound manner.

Yet if the CRA was not in place today, would stakeholders want to enact it?

Many people may be inclined to answer affirmatively simply based on the CRA’s statutory goal. Furthermore, when it became law, many low-income and minority communities found mortgage and small-business lending hard to come by.

At the time, the CRA sought to address persistent and institutionalized credit discrimination going back to the 1930s, when the government-sponsored Home Owners’ Loan Corporation discouraged banks from providing credit to “hazardous” areas, colored red on HOLC-issued maps. Such “redlining” prevailed for decades.

Other anti-competitive regulation made credit discrimination difficult to overcome. Most states did not allow bank branching in 1977. Interest rate caps on savings accounts made it less advantageous to put money in the bank, particularly in the high-inflation environment of the 1970s. Together, these restrictions created local banking monopolies with little incentive to serve marginalized borrowers.

Thankfully, the U.S. banking landscape has changed for the better since 1977.

Branching is now permitted nationwide and has ushered in a 77% jump in the number of bank offices, despite major bank consolidation. Interest-rate caps on deposits are a thing of the past. More significantly, nonbank lenders make up an increasing share of mortgage and small business lending.

In fact, loans to low-income borrowers today constitute a bigger proportion of the nonbank portfolio than they do for banks. There is also evidence that fintech lenders cater more to underserved communities.

These structural changes have made credit markets more competitive and redlining less prevalent. They have also made the case for keeping the CRA less persuasive.

But there are other reasons to be wary of the CRA. Some empirical evidence suggests loans made right before CRA evaluations are riskier than other loans. Another study found that high CRA ratings correlate negatively with banks’ safety and soundness.

Even when CRA lending appears less risky, there is no guarantee that the loan dollars are going directly to the underserved communities that the Act seeks to help. In fact, in the DC area, as much as two-thirds of mortgages eligible for CRA credit go to borrowers that are not low- or moderate-income.

The CRA also fails to address some key present-day barriers to financial inclusion. These include the large number of unbanked households, totaling 8.4 million as of 2017; the growing problem of “banking deserts” lacking any branches within a 10-mile radius; and the considerable fees that low-income households must pay to keep a bank account.

Indeed, because the CRA adds to operating expenses, accounting for 7.2% of community bank compliance costs, it could be making these problems worse.

Partly in recognition of the CRA’s compliance burden, the Office of the Comptroller of the Currency, which among the CRA regulators is leading its reform, has suggested replacing the current system of qualitative assessments with a metric-based approach.

The hope is to make CRA evaluations less arbitrary and bureaucratic. Rather, proponents of this reform say it would make performance indicators more transparent and comparable across institutions and products.

But a single metric might not reflect the varied circumstances in which banks operate in different parts of the country. It would also turn the CRA into a quota system, explicitly contradicting the intent of its drafters.

There is a better way to get the simplifying benefits of a metric-based approach while introducing market incentives into CRA compliance. That is to allow banks to pay other lenders a fee for fulfilling their CRA obligations.

Instead of evaluating banks according to an expansive list of qualitative criteria, regulators would determine the amount of CRA lending and other activities that banks as a group must perform in an area. Banks could then either discharge those obligations on their own or pay someone else to do so on their behalf.

The advantages of this approach are manifold. First, it would encourage some lenders — whether banks, fintech firms, credit unions or community development financial institutions — to specialize in serving low-income and marginalized communities.

Second, it would give the lenders who contracted with banks an incentive to lend prudently, as they would bear the cost of bad loans. Third, the fee would explicitly reflect the cost of complying with the CRA. Regulators could even use local CRA fees as a measure of credit market saturation and prudential risk.

There are legitimate concerns, particularly after the last financial crisis, about politicizing the volume and allocation of credit. If unsound lending proliferates, experience teaches that taxpayers will end up paying the tab.

But politics already plays an outsized role in the CRA, with activist groups able to use low CRA ratings to press regulators to block bank expansions and mergers. A system of tradable obligations would not entirely depoliticize the CRA. But it would simplify compliance and challenge regulators to actually lay out unmet community credit needs that banks should fulfill.

Effective public policy to promote financial inclusion in 2019 looks as different from the CRA as today’s credit markets are from the unit banking system of yore. Stakeholders today would probably not enact the CRA as we know it; perhaps we should not even keep it. But at the very least, CRA regulations should adapt to the changing banking landscape.