Myths and Facts: A review of Otting testimony on proposed changes to the Community Reinvestment Act

(Download)

The Comptroller of the Currency, Joseph M. Otting, submitted testimony in advance of his appearance before the House Financial Services Committee on Jan. 29, 2020. His testimony cited and challenged NCRC’s analysis of a proposal to overhaul rules that enforce the Community Reinvestment Act. This is NCRC’s response to Otting’s testimony.

MYTH:  “Nothing in modernizing CRA regulations encourages or legalizes redlining.” (Page 7)

FACT:  §25.11 of the proposed rule would not examine bank mortgage lending in low- and moderate-income (LMI) geographies/neighborhoods, just bank lending to LMI borrowers. How will the regulator know if redlining is or is not occurring? This is the agency’s response to prevent gentrifying loans going to middle- and upper-income (MUI) borrowers in low-income areas, but there are better regulatory responses to displacement than taking the regulator’s eye off potential redlining of LMI neighborhoods.

The proposal also suggests banks could fail in nearly 50% of their assessment areas (AAs) and still pass or get an outstanding rating, allowing them to neglect retail lending in half of the AAs with low- and moderate-income (LMI) borrowers and communities without any penalty.

MYTH: Critics allege that proposal would rely on a single metric (Page 7)

FACT: The proposal associated the single metric with a presumptive rating, which would suggest it is the dominant measure. Also, the single metric has proposed ratings that correspond to various ratios (11% is Outstanding, 6% is Satisfactory, etc.). The other component of the new test would be the retail test which is only pass/fail. Currently, retail lending counts for 50% on the large bank exam. The new and weaker retail test would only be pass/fail and thus a minor measure compared to the single metric. 

MYTH: Critics allege that proposal strays from original statutory focus on LMI (page 7 & 8)

FACT: Eliminating home lending to LMI tracts would make it harder to provide homeowner opportunities to LMI people in LMI communities that are not gentrifying by reducing incentives for banks to make these loans. In addition, there are other ways to deal with the possibility of displacement associated with gentrification as raised by NCRC in our ANPR letter. For example, examiners could count only loans to LMI people in LMI tracts that are gentrifying (NCRC has a methodology for identifying tracts that are gentrifying. 

MYTH:  “The proposal would require examiners to consider a retail lending test, virtually identical to that recently described by Federal Reserve Governor Brainard.” (Page 8)

FACT:  No, there are key differences in the conjunction and the count based on what we know of both proposals so far. §25.11 would require banks to meet “either the associated [demographic] comparator or the associated [peer] comparator” – either demographics OR peers, WHICHEVER IS LESS.  In contrast, Gov. Brainard was clear, their proposal would look at BOTH – demographics and in-market competitors/peers. Why does it matter?  Well, under the OCC’s proposed rule, there is really no incentive for banks to do better than their peers. Under the proposed rule, they only have to do 65% of what peer banks do to pass.

Gov. Brainard herself noted the data, modelling and tailoring differences in a speech this month saying: “We were hopeful our proposed approach could be incorporated into the proposed rulemaking that was released last month in order to seek public comment on a range of options.”

MYTH:  The proposal would eliminate churning, which is the practice of banks purchasing loans made to LMI people shortly before CRA exams (Page 8)

FACT: There are instances of the same loans to LMI people being sold and purchased over and over again. However, the OCC’s proposal would penalize banks that sell their loans to the secondary market within 90 days of making them. This penalizes selling loans to get more capital to make more loans, which could significantly reduce overall levels of LMI lending. Instead, the way to deal with churning is to simply say banks cannot purchase loans in a certain time period before their CRA exams. The Dodd Frank improvements to HMDA data would enable regulators to do this because they have access to a loan ID for each loan. The public does not have this data but the regulators do. 

MYTH:  “…the list of approved activities are more true to the statute’s original intent and refocuses on LMI individuals…”  (Page 8)

FACT:  The proposed rule replaces the “primary purpose” focus of the current CRA Interagency Q & A with §25.04, which awards far more CRA credit for bank activities that only “partially benefit” LMI people and communities. FDIC Board Member Martin J. Gruenberg added, the proposal “…expands eligible and qualifying CRA activities to include some of what banks already do in the ordinary course of business, thereby diluting the effectiveness of CRA.” (e.g. public roads, bridges, tunnels, sewer treatment and collection, parks and public safety facilities that partially benefit or serve LMI people).

MYTH:  “…the proposal maintains the branch’s central role…” (Page 9)

FACT:  Today, bank branches are examined under a CRA service test – 25% of the exam. The OCC’s approach drops the CRA service test in favor of a branch credit in the OCC’s dollar volume metric – multiplying the percentage of LMI branches by .01. A bank with 30% of their bank branches in LMI census tracts would receive .003% credit – a mere 5% of a presumptive satisfactory rating or 2.7% of a presumptive outstanding CRA rating.  

Today, failure to achieve sufficient branching distribution in most of a bank’s local communities would jeopardize the bank’s overall rating to a far greater degree than a proposal that could allow banks to fail in nearly half their local communities. The .01 branch credit added to the dollar volume metric proposed in §25.10 gets lost – swallowed up in the OCC’s “CRA Evaluation Measure.”

Also, while it is true that the proposal eliminated the distinction between full and limited scope AAs, it would still delegate about half the AAs to no impact since banks can fail in half their AAs.

MYTH: “…the proposal states that a bank must have at least a satisfactory in a significant portion of assessment areas to get a satisfactory at the bank level and specifically asks for input on whether that threshold should be as high as 80 percent.” (Page 9)

FACT: The proposed rule also states that: “At the bank level, a bank’s presumptive rating would be based on the comparison of its average bank-level CRA evaluation measure to the established empirical benchmark, except that a bank could not receive a satisfactory or an outstanding unless it also received that rating in a significant portion, such as more than 50 percent, of its assessment areas and in those assessment areas where it holds a significant amount of deposits, such as more than 50 percent.”

MYTH: “In actuality, banks have received credit for financing athletic facilities, including professional sport stadiums, under the current CRA framework for decades.” (Page 9)

FACT: Perhaps there are anecdotes, but the examples that the Comptroller cites under today’s stricter qualifying criteria – repairs to local high schools and municipal facilities, multiuse and colleges facilities that serve minority and LMI areas – are a far cry from those contemplated by bank investments in Opportunity Zone (OZ) funds in §25.04(c)((11). OZ funds can finance any type of project in an OZ.  NCRC examined 37 recently constructed and proposed stadiums across the country and 17 were in LMI OZs and another 12 in either an LMI census tract or an OZ.  Because of the scope and expense of stadium construction, this could become an easy way for banks to meet almost all their CRA obligations to a community, by financing low-risk and profitable stadium construction, expansion or improvements. The question for the Comptroller is whether he is contemplating the financing of jumbotrons or community park facilities.

MYTH:  “The reality is that there are many examples where agencies act independently based on the needs of the institutions they oversee and the communities served by those institutions.” (Page 10)

FACT:  Comments from CRA stakeholders across the spectrum have been consistent and aligned in their desire to have one set of final rules, to ensure clarity and consistency across agencies and across banks.

MYTH: “…the proposed rule would result in less CRA activity. Not true at all…misperception created by a handful of groups circulating faulty research…” (Page 10)

FACT:  The Philadelphia Federal Reserve study and NCRC’s nationwide CRA loss projection based on its methodology was developed in advance of the OCC’s proposed rule. It remains illustrative of the impact of CRA today. The forecast is also relevant to the OCC’s proposed retail lending test, which would eliminate CRA review of mortgage lending in LMI geographies/neighborhoods. 

Also, if CRA exams diminish the importance of AAs (especially if a bank can fail in a large number of AAs), then declines of this magnitude are possible. 

To be clear, concern about the proposed rule resulting in less CRA activity has been joined by a diverse coalition as well as other sectors impacted by the proposal.   

MYTH:  “…bankers will be incentivized to do more because there will be greater regulatory certainty about what counts and how much an activity counts for CRA credit.” (Page 11)

FACT: The general performance standards and presumptive ratings proposed in §25.12 rely on an overly determinative dollar volume metric (§25.10), inflated by an expanded list of qualifying activities and definitions (§25.04 and 25.03) with more partial credit and more credit anywhere in the country. It will incentivize banks to hit the metric by doing big, easy and the most profitable financing.

Getting small dollar credit or bank financing in underserved communities or to more complex deals will get more difficult, because the proposed rule expands opportunities for CRA credit, but undermines the CRA obligation. The CRA obligation drives as much or more of bank behavior than where they can get credit. The proposed rule undermines CRA’s statutory and regulatory design to date, which has been to encourage banks to enter LMI markets that are thinly-traded, to overcome market failures, negative and informational externalities.

MYTH: “…the potential for advocates to “stuff the ballot box” when it comes to public comments… publishing letters purportedly signed by hundreds of its member organization.” (Page 11)

FACT: The National Community Reinvestment Coalition (NCRC) was named for CRA and founded over 25 years ago to safeguard the law against weakening in Washington, but also to demonstrate the power of the law across the country. NCRC represents hundreds of housing counselors, CDCs, EDCs, nonprofit developers, CDFIs, small business centers and other CRA community stakeholders across the country. We have negotiated $150 billion in community benefit agreements with banks around bank mergers and acquisitions since 2016.

NCRC members and other national organizations have relied on NCRC’s CRA expertise and on the organization to provide leadership on CRA advocacy. Throughout this regulatory reform process, hundreds have cited NCRC’s research in letters, articles and other materials; joined NCRC on letters, in dozens of meetings, on dozens of calls all towards providing feedback to the OCC, the FDIC, the Federal Reserve and the U.S. Congress on CRA regulatory reform that bests executes the law’s purposes, makes it even more effective and that meets local community needs.

The proposed rule announced by the OCC and the FDIC in December 2019 dilutes the law’s effectiveness and does not meet local credit needs. NCRC will continue to provide leadership on CRA, speak up about it and organize around the flaws in the substance of the OCC and FDIC’s proposed rule and the rulemaking process the agencies are undertaking, leaving out key data and analyses and expediting a rule that needs more public review.

Print Friendly, PDF & Email