In her Wednesday speech blasting the OCC/FDIC CRA rewrite, FRB Gov. Brainard said that, “The CRA plays a vital role in the ecosystem supporting economic opportunity in LMI communities in both rural and urban areas.” It is indeed easy to infer that banks would not make significant community-service commitments without a statutory mandate – after all, banks are first and foremost for-profit companies with no more of a duty to serve LMI households than a gas station has to pump for free. However, once past the intuitively-obvious cattle-prod conclusion, CRA’s impact on LMI communities is far from straightforward. Gov. Brainard wants a CRA rulebook akin to the current one with changes only to make it still tougher yet somewhat more clear. If CRA as is hasn’t done much, then CRA as she wants it will do little more. The OCC/FDIC proposal rips up the rulebook, but its construct is based largely on what it calls “stakeholder” input, not substantive research. Looking to objective data as a guide through this political minefield, it’s clear that CRA’s impact has neither democratized credit as advocates argue nor made it riskier as opponents allege nor even for that matter necessarily made much of a difference. To revise CRA to achieve the urgent economic-equality objectives on which all of the banking agencies agree, each needs to step back and look at the data.
In a 2018 EconomicEquality blog post, we anticipated progressive plaints about Administration-sponsored CRA reform, looking at a comprehensive literature survey for guidance to a meaningful reform agenda. Evaluating CRA’s impact on mortgage finance, researchers found that, “CRA has expanded access to credit in LMI communities, but the magnitude of the increase and the mechanisms of the impact of CRA are far from conclusive.” As a HUD study we cited then concluded, affordable-housing finance is at an “all-time low.” Maybe without CRA it would have been worse, but CRA is still far from the LMI household panacea on which the Fed premises its proposal.
Indeed, CRA as is may well have perverse results. A 2019 study from the libertarian Cato Institute updates prior research, concluding that CRA failed because credit may well be flowing to areas with LMI households, but not to the actual LMI households in these areas. Instead, this research finds that credit flows most freely to “gentrifiers,” citing Washington, D.C. as what I would suspect is an all-too apposite example. According to this study, about two-thirds of CRA-eligible credit goes to gentrifiers, not LMI households hoping to stay put.
A 2018 study from the National Community Reinvestment Coalition (NCRC) challenges Cato’s conclusions, using its own research survey to conclude that a “significantly” weaker CRA would deprive LMI communities of up to $105 billion of financing over the five years it examines. However, it reaches this conclusion because it thinks that as much as twenty percent of lending to LMI areas would evaporate with CRA liberalization, not based on looking to see who gets CRA credit now and if LMI-area loans are actually for LMI households or small businesses. What a “significant” weakening of CRA might entail and how certain data in the study were “extrapolated” to NCRC’s conclusion is even less clear.
Although the OCC and FDIC support much of their proposal based on “stakeholder input” – translation, lobbying by banks, community groups, and perhaps even some communities – the Fed’s alternative is grounded on “metrics based on data” and “guided by stakeholder input.” However, following the Fed’s data as described by Gov. Brainard, the bulk of it either tells us how unequal America is – which we knew – or how banks have to date complied with CRA – a round-trip back to keeping CRA as is. In fact, Gov. Brainard’s speech states that “consistent data on CRA-eligible activities were not readily available,” with the Fed instead building a massive database of 6,000 CRA evaluations. If the Fed could link its CRA-performance metrics with changes in LMI communities that then are correlated or even better caused by CRA lending, then the predicate assumption of Gov. Brainard’s speech would be validated. Without it, we’re back to relying on stakeholder input.
The agencies will get a whole lot of it at House Financial Services Committee hearings later this month, but I’m not sure we’ll be much the wiser for it. Nor, I suspect, will constructive CRA reform advance any more quickly following this going-over. Too bad – even if CRA hasn’t ever made all that much of a real difference, it sure could.