NCRC’s Guide To The 2023 Community Reinvestment Act Final Rule

December 2023

Photo: Ernie Journeys on Unsplash

Positive Changes

  • The Retail Lending Test is much more objective and transparent thanks to these changes. The expansion of the test’s reach and reformulation of its scoring process for large banks should make this CRA component’s incentives much stronger.
  • The expansion of eligible activities in the Community Development Financing Test sharpens the test’s message to banks and rewards creative, proactive investment. Newly eligible community development activities in Native Land Areas, in weather resiliency and financing that contributes to the health and well-being of LMI households are all significant positive steps.
  • Special Purpose Credit Programs (SPCP) are now explicitly encouraged under CRA. This provides confidence to continue expanding an innovative approach to lending that has taken off in the 25-plus years since the last significant update to CRA rules.

Unhelpful Changes

  • Raised thresholds will leave CRA under-enforced at many institutions. Fewer banks will be covered by positive changes in the final rule than would have been covered under the proposed rule due to ill-advised changes to asset and loan  thresholds.
  • New auto lending provisions will apply to few lenders. Though incorporating auto lending is a positive grand-scheme change, the final rule greatly restricts the change’s reach: Only institutions where car loans are more than half of all retail lending will be evaluated on their auto lending.
  • Two key areas of evaluation have ‘only contribute positively’ caveats that mean poor performance doesn’t hurt bank scores. Affordable and sustainable bank retail products can increase a bank’s score, but unaffordable or unsustainable ones cannot lower a score. This weakens CRA’s incentives and signals to banks that these are lower priorities.

Missed Opportunities

  • The rules perpetuate CRA’s troublesome racial blind spot. Though CRA was designed to address racist policies and business practices, institutions will still not be evaluated on the demographic mix of their borrowers.
  • Nothing has changed to strengthen the link between CRA performance and merger or branch-siting review processes. Advocates encouraged the agencies to more strongly invite community input in merger reviews and to update the if-then consequences of poor CRA performance on branch closure and merger review processes, but changed nothing there.
  • Weather-resiliency incentives fall short on climate change challenges. In addition to incentivizing weather-resiliency projects on the ground, regulators could have required an analysis of the climate impacts of banks’ financing.


Kevin Hill, Senior Policy Advisor, NCRC

The Community Reinvestment Act (CRA) requires banks to meet the credit needs of the entire community, particularly low- and moderate-income (LMI) households. The CRA accomplishes this through an ongoing evaluation of a bank’s loans, investments and services. The CRA is implemented by the Office of the Comptroller of the Currency (OCC), the Federal Deposit Insurance Corporation (FDIC) and the Federal Reserve. The three agencies recently finalized the most significant update to the CRA rules since 1995, a monumental task considering how much banking and the economy has changed since then.

The updated rules offer several improvements to the metrics and benchmarks for evaluating CRA performance in lending, as well as helpful updates to the definition of community development. CRA exams will also evaluate large banks on how well their online banking services serve LMI households, a major part of banking today that was nonexistent during the last major change to CRA in the mid-90’s.

The agencies also missed several opportunities, most significantly by not incorporating a racial analysis into CRA lending examinations. There is also no change to how CRA performance will affect bank mergers, which is the main penalty for inadequate CRA performance.

The initial guide that follows summarizes the requirements and definitions in the new rules. It does not and cannot definitively assess the agencies’ work: Ultimately, final judgment of these new rules will depend upon real-world outcomes once the changes have come fully online in 2026. If these rules lead to increases in LMI homeownership and other improvements in bank performance inside and outside of their branch networks, including securing much-needed community development financing on Native Lands, then that will be cause to celebrate. However, if these rules have little impact on our most pressing national issues, such as the persistent racial home ownership gap and the threat climate change poses to economic security, then we will have confirmation that the agencies – or Congress – need to take additional steps.

This outline of the new CRA rules will be complemented in the coming months with further analysis zeroing in on specific areas of note in the package.

Asset Thresholds: A Significant Redefinition Of “Large,” “Intermediate” And “Small” Banks

Bank asset thresholds in the final rule are unchanged from the proposed rule. As a result, 778 fewer banks will have a community development test. Another 216 banks will no longer have a Service test that evaluates their branch network, nor be subject to the new Retail Services and Products test of their products.[1]

Here is how the CRA definitions of large, intermediate and small banks have changed with the finalization of the rule:

  • Large Banks – $2 billion in assets and above (previously $1.5 billion)
  • Intermediate Banks – between $600 million and $2 billion in assets (previously $376-$1.5 billion)
  • Small Banks – $600 million in assets and below[2] (previously $376 million)[3]

The agencies note that these new thresholds only lead to a 2% reduction of total deposits covered by the community development and service tests. But it decreases the total number of banks with a service test by 15% and exempts 31% of previously-covered banks from a community development test.[4]

The agencies also justify the higher thresholds by raising concerns with the capacity of intermediate and small banks, but many capacity concerns are addressed in other portions of the rule. For instance, the new community development test is optional for intermediate banks as they can choose to keep the current test. They also enjoy a modified version of the new Retail Lending Test without retail-based assessment areas and a further qualified exemption from the new outside retail lending assessment areas if more than half of their loans are inside their branch network. Small banks also have the option to retain their current Lending Test and have no additional requirements for community development or branches.[5]

Performance Tests & Test Weights: Which Banks Are Being Evaluated on What

Here are the types of tests facing each category of bank under the new rules, including breakdowns of how each mix of evaluations will be weighted in calculating an overall grade. A separate detailed breakdown of the tests themselves – both what’s changed and how newly-introduced assessments will function – follows in a subsequent section.

Large Banks – $2 billion in assets and above[6]

  • Retail Lending Test – 40% weight of state, MSA and institution/total rating
  • Community Development Financing Test – 40% weight of state, MSA and institution/total rating
  • Retail Services and Products Test –  10% weight of state, MSA and institution/total rating
    • Qualitative evaluation of products that can only increase a rating; A failure to provide affordable credit or deposit products cannot cause CRA downgrading.
  • Community Development Services Test – 10% weight of state, MSA and institution/total rating

Large banks naturally tend to have a greater geographic reach corresponding to their larger market share. The new rules include a specific subset of provisions for those large banks with the most extensive geographic reach, defined as having 10 or more separate facility-based or retail lending-based CRA assessment areas. This provision aims to reduce the likelihood of rewarding a bank that is excelling in a handful of markets while performing poorly in the rest. Such banks cannot receive passing CRA grades unless they earn “Low Satisfactory” or higher in at least 60% of their facility- and retail-based assessment areas at the state, multi-state MSA or institution/overall level.[7]

In a bank’s first exam after the new rules have kicked in, this 60% pass rule applies only to large banks with at least 10 facility-based assessment areas. From the second exam cycle onward, large banks with at least 10 facility-based or retail-based assessment areas will also face the 60% pass rule.[8]

Intermediate Banks – $600 million to $2 billion in assets

  • Retail Lending Test (with modifications) – 50% weight of state, MSA and institution/total rating
  • Previous Community Development test with opt-in option for new Community Development Financing test – 50% weight of state, MSA and institution/total rating
  • Intermediate banks are exempt from the 60% pass rule for large banks described above.[9]

Small Banks – $600 million in assets and below

  • Previous lending test, with opt-in option for new modified Retail Lending Test – 100% of state, MSA and institution/total rating
  • Small banks are exempt from the 60% pass rule for large banks described above.[10]

Limited Purpose Banks – defined as banks that do not offer retail loans and services on a regular basis. These banks will have a community development finance test.[11]

The final rule also maintains the current practice of allowing a strategic plan option. For banks to get a strategic plan they must justify how the bank’s activities are outside the scope of the performance tests and why being evaluated pursuant to a plan would more meaningfully reflect its CRA performance.[12] In addition, draft strategic plans will be placed on regulator websites for a 60 day comment period, as well as the banks website and notices must be placed in newspapers of general circulation in each of the banks facility-based assessment areas. This is a considerable improvement over the previous process for strategic plans, that only required notices in newspapers[13]

Changes to Assessment Areas

Assessment areas are the markets where a bank’s performance is analyzed. There are three different types of assessment area which may apply in a given bank’s CRA evaluation.

Facility-based Assessment Areas

These areas are what has shaped CRA evaluations for years, applying to banks at all asset levels. They continue to play a role in the modernized CRA rules, now alongside new assessment areas discussed below. Facility-based assessment areas are determined by where the bank has its main office, physical branches or unstaffed deposit-taking facilities like ATMs. While largely retained from the previous rules, the shape of facility-based assessment areas has changed subtly for the largest banks, which must now include the entire county surrounding a qualifying facility. Intermediate and small banks can still use partial counties as the rules had previously allowed for all sizes of bank.[14]

Retail Lending Assessment Areas

These are one of the major new features of CRA regulations going forward. They will apply only to a subset of the industry. Large banks that do less than 80% of their retail lending inside their branch network (or facility-based assessment areas) will now have retail lending assessment areas in markets where they’ve originated either 150 closed-end mortgages or 400 small business loans in each of the last two years.[15]

Banks covered by this new assessment area category will not be evaluated for community development work outside their facility-based assessment area. Retail lending assessment areas are subject to the Retail Lending Test only. That test, discussed in detail below, will analyze closed-end mortgages and small business lending, depending on which of these loan types trigger the loan threshold.[16]

The specific geographies included in the retail lending assessment area differ depending on the nature of the market in question.

In urban/suburban marketsthe retail lending assessment area is the entire Metropolitan Statistical Area (MSA), excluding any counties where they have branches that are included in the facility-based assessment areas

In rural markets, all of the counties in a non-metropolitan area of a state, excluding any counties where the bank has branches or where it did not originate closed-end mortgage or small business loans.[17]

These final definitions and conditions around the new retail lending assessment areas and corresponding CRA exam test are significantly different from the versions of each first suggested in the draft rule. The proposed rule would have covered substantially more banks. The regulators initially proposed lower thresholds of lending to trigger the assessment areas – 100 closed-end mortgages or 250 small business loans – and planned for them to apply to all large banks. Instead, banks that do as much as 20% of their retail lending outside their facilities-based assessment area will be exempt entirely and the rest will benefit from the softened loan thresholds compared to the proposed rule. [18] Based on an analysis of data from 2018-2020, the agencies estimate that these changes decreased the number of banks that would have to create retail lending assessment areas by half, from 125 to 63, as well as decreasing the number of new retail lending assessment areas by nearly half, from 1,591 to 863.[19]

Using 2018-2020 data, regulators estimate that 310 large banks (83% of large banks – excluding wholesale, limited purpose and strategic plan banks) will not be required to create retail lending assessment areas. They estimate that 50 large banks (13%) will be required to create between one and 10 such areas. Five banks will be required to create more than 50 of them.[20]

The regulators also estimate that large banks would have received failing ratings in 22.4% of retail lending assessment areas and that this would have resulted in 7% of large banks receiving a “Needs To Improve” on their overall Retail Lending Test.[21]

Outside Retail Lending Areas

The new rules also introduce new nationwide assessment areas that exclude the bank’s facility-based and retail lending assessment areas, as well as any non-metropolitan counties where the bank did not originate or purchase any loans.[22]

All large banks will be evaluated in this new fashion. Intermediate banks will have Outside Retail Lending Areas if 50% or more of their loans and purchases – calculated using a combination of loan counts and dollars – are outside of their facility-based assessment areas. Other banks may opt in to this new system but are not required to have Outside Retail Lending Areas.[23]

Revisions To The Retail Lending Test

The Retail Lending Test which has been a core mechanism of CRA evaluations for decades has been overhauled in a litany of ways in the new rules. This test evaluates large and intermediate banks on the distribution of their loans to LMI borrowers  and within LMI census tracts, as well as loans to small businesses and farms.

The Retail Lending Test includes three sets of metrics – first a Retail Lending Volume Screen, then metrics to determine the major product lines of the bank and to analyze those major product lines against relevant market and community benchmarks for the distribution of loans. The components of the Retail Lending Test are weighted differently, following a complex and flexible set of rules discussed below.

These changes add more transparency to how a bank’s lending is evaluated. But the rule also applies various compensating factors in the Retail Lending Test that create an opening for poor performance to be excused, which could reduce consistency in how the updated lending test is implemented.

Retail Lending Volume Screen

This metric is used to determine if a bank is making a reasonable number of loans in a facility-based assessment area compared to its deposits held in that assessment area. Regulators will create two loan-to-deposit ratios. The first is a bank’s Bank Volume Metric, calculated by dividing a bank’s total dollars of lending in specific categories by the total dollar amount of the deposits the bank holds in that specific facility-based assessment area. Categories of lending assessed in this metric include closed-end home mortgage loans, open-end home mortgage loans, multifamily loans, small business loans, small farm loans and auto loans in some cases.[24] Automobile lending is only automatically included in the Bank Volume Metric if auto lending accounts for more than 50% of a bank’s total retail lending in each of the two years before the first year of the evaluation period. In such cases, auto lending in the Outside Retail Lending Area will also automatically be evaluated.[25] Other banks that don’t trigger that 50% threshold could also choose to have their auto lending included.[26]

The regulators will compare this Bank Volume Metric to a Market Volume Benchmark, which is created by dividing aggregate retail lending by deposits of all banks that maintain at least one branch in the assessment area. The Market Volume Benchmark excludes auto lending, but otherwise uses the same definition of retail lending as the Bank Volume Metric.

Small and intermediate banks were already evaluated using loan-to-deposit ratios. Large banks were already evaluated on the volume of their lending in their assessment areas. But the new Retail Lending Volume Screen adds additional clarity to how regulators are determining whether a bank makes a sufficient number of loans in the communities where it holds branches.[27]

For large banks, the Bank Volume Metric must be at 30% of the Market Volume Benchmark, or else the bank will not pass in that facility-based assessment area unless the bank has an “acceptable basis” for not meeting the threshold. If the regulators determine the bank has an “acceptable basis” then its score will be determined by the distribution tests of its major product lines.[28]

In determining an “acceptable basis” for missing the 30% benchmark, regulators will consider the financial condition of the bank, the presence of other lenders in the area and the business strategy of the bank. Due to the wide variance in bank business strategies, the final rule does not list specific exemptions for particular strategies. It instead leaves this to the CRA examiners on a case by case basis, which could reduce consistency in implementation.[29]

Regulators reviewed HMDA and CRA data from 2018-2020 and found that 3.3% of facility-based assessment areas of banks above $50 billion in assets would not meet the Retail Lending Volume Screen, a very low percentage.[30]

The final rule also notes that “a bank would fail to meet the credit needs of its entire community if it makes too few loans relative to its community presence, capacity and local opportunities, even if those loans happened to be concentrated among, for example, LMI borrowers and LMI census tracts.”[31]

The retail lending volume screen matters less for intermediate banks and small banks that opt into the new lending test, as they can still pass based on the results of the distribution analysis.[32]

Credit card lending and other types of consumer loans such as unsecured small dollar loans will no longer be evaluated on the lending test, with the exception of small business credit card lending.[33] Previously, consumer lending was considered if it constituted a substantial majority of a bank’s business, or if the bank requested it to be considered.[34] Regulators point out that large banks can receive positive consideration on the Retail Services and Products test for credit products that are responsive to the needs of LMI individuals. But it is important to note that this test can only contribute positively to a bank’s CRA rating – and it will not include an analysis of which borrowers are actually getting consumer loans.[35]

Purchased loans continue to count the same as originations and will be used on the Bank Volume Metric as well as in the distribution analysis. However, purchased loans are excluded from market benchmarks on the distribution analysis.[36] Regulators will exclude purchases done to “inappropriately enhance” lending performance by specifically looking for loans that are rapidly resold after evaluation periods, as well as purchased loans that have been considered on the CRA exams of more than one bank.[37] Regulators should also look for purchases made in facility-based assessments in order to avoid requiring retail lending assessment areas or outside retail lending assessment areas, as both the thresholds for these assessment areas include purchases.[38]

Major Product Line

The second stage of the Retail Lending Test determines which loan types the regulators will consider when evaluating the number of loans made or purchased to LMI borrowers, in LMI census tracts, or to small businesses or farms in the case of business loans. This step of the process works differently in different types of assessment areas.

In a Facility-based Assessment Area or Outside Retail Lending Area, a bank’s originated and purchased closed-end home mortgage loans, small business loans, small farm loans and automobile loans (for lenders whose auto loans account for 50% or more of total retail lending) would qualify as a major product line if the loans in the product line comprise 15% or more of the bank’s lending across all product lines in the area. The 15% threshold will be determined by averaging the percentages of both loan count and dollar volume for all product lines, for example, if a bank’s auto lending accounts for 10% of its total retail lending dollars and 22% of its total retail loan count in its facility-based or outside assessment area, then the combined average would be 16% and auto lending would qualify as a major product line.[39] 

In Retail Lending Assessment Areas, the product lines that trigger the creation of the new assessment area are automatically deemed “major” for the Retail Lending Test: Closed-end mortgages (when 150 or more are made in back-to-back years) and small business loans (when 400 or more are made in back-to-back years). Small farm lending in retail lending assessment areas is excluded from the test.[40]

Here the final rule makes a notable change from the initially proposed rule: Open-end mortgage lending (such as home equity lines of credit) and multifamily lending have been removed from consideration as a major product line and banks will not be evaluated on their distribution of these loans on the Retail Lending Test. Instead these loans will be used for the Retail Lending Volume Screen.[41]

If a bank has no major product line in a facility-based assessment area, meaning no product line met the 15% threshold, then the Retail Lending Volume Screen is used to evaluate the bank’s performance in that facility-based assessment area.[42]

Distribution Analysis

Large and intermediate banks that pass the Retail Lending Volume Screen, or fail with an acceptable basis, will be analyzed on the distribution of each of their major product lines made to LMI borrowers, within LMI census tracts and to small businesses and farms with annual revenues between $250,000 and $1 million. In a helpful change, the LMI borrower and LMI census tract metrics will be evaluated separately rather than combined as in the past. In other words, lending to low-income and moderate-income borrowers will be evaluated separately, just like lending to low-income and moderate-income tracts.[43]

A bank’s lending distributions will be compared to both a market benchmark and a community benchmark. Market benchmarks are determined by an aggregate of all lending to each category (LMI borrower, LMI census tract, lower-revenue business or farm) and the community benchmark reflects local demographic data.[44] Taken together, then, the benchmarks allow examiners to both see which banks are doing more or less than their peers, and which banks are being more or less responsive to the population and neighborhoods in their assessment areas.

Examiners will use loan counts and not dollar amounts in these assessments, in order to not discourage smaller dollar loans. For example, when evaluating a bank on closed-end mortgage lending to borrowers with low incomes, the regulators will divide the total number of LMI-borrower mortgages a bank originated and purchased by the total number of closed-end mortgages the bank made to all borrowers in the assessment area.[45]

The agencies have tied specific ranges to each performance rating for both the market and community benchmarks and will use whichever benchmark is lower for each major product line. For example, if 115% of a bank’s market benchmark is 30% and 100% of its community benchmark is 40%, the bank’s lending would be compared against the 30% market benchmark figure.[46] NCRC recommended a weighted average approach instead of choosing the lower of the community or market benchmarks. Selecting the lower benchmark could result in lower thresholds that inflate ratings. For example, in assessment areas in which the market benchmark is considerably lower than the community benchmark, all lenders could be under-performing in making loans to borrowers with LMI or LMI communities. In these situations, using the lower market benchmark could excuse poor performance.

These ranges have been reduced from the original proposal. [47]

Regulators will publish data for determining each benchmark annually. But the community benchmark data will be released in advance[48] – that is, the 2025 benchmark would be announced in 2024 – while the market benchmark will be retrospective[49], with the official 2024 benchmark data being released in 2025.

Both community and market benchmarks will be set based on data within a bank’s evaluation period, not based on data from years prior to the evaluation. The agencies will annually release data to help with tracking benchmarks.[50]

Retail Lending Test Weights & Calculations

Once examiners determine which are a bank’s major product lines, they then must decide how to appropriately weight different major products in scoring performance – and how to weight different assessment areas when combining them into a single rating for the bank across its entire geographic footprint. The new rules provide a detailed and complex process by which those weighting decisions will be made. To understand best how this process works, it helps to unpack it into two stages: First, how the bank’s multiple ratings in its different specific assessment areas are calculated; then, how the ratings of those assessment areas are combined together to yield a Retail Lending Test rating for the bank as a whole.

Determining Ratings In A Specific Assessment Area: A Weighted Average of Product Line Scores

Each major product line will receive a product line score based on the following four categories with two based on geography and two based on the borrower: The geographic categories are the same for each product line, but the borrower categories depend on the loan type.[51]

Geographic Categories – All major product lines are evaluated based on Census designations of low-income tracts and moderate-income tracts, with each of these tract categories receiving its own product line score.

Borrower Categories

Closed End Mortgage and Auto Loans

  • loans to borrowers with low incomes,
  • loans to borrowers with moderate incomes,

Business Lending

  • loans to businesses below $250,000 in gross annual revenue,
  • loans to businesses with $250,000 to $1 million in gross annual revenue,

Farm Lending

  • loans to farms below $250,000 in gross annual revenue,
  • loans to farms with $250,000 to $1 million in gross annual revenue

Examiners will use the Distribution Analysis method described above of comparing to community and market benchmarks to determine the specific product line score, or points, the bank receives for each of these four categories:[52]

  • Outstanding (10 points);
  • High Satisfactory (7 points);
  • Low Satisfactory (6 points);
  • Needs to Improve (3 points);
  • Substantial Noncompliance (0 points)

Each geographic category and borrower category will get a product line score. Then, to generate an overall score for geographies and borrowers in a specific assessment area, the regulators will create a weighted average based on demographics. For example, if the breakdown of LMI families in an assessment area is 70% low-income and 30% moderate-income, then the low-income component of the borrower category assessment would receive a weight of 70% and the moderate income component would receive a weight of 30%.[53]

Then the agencies will calculate the average for both borrower and geographic categories for each major product line in an assessment area to come up with the final product line score.[54] Borrower income average and geographic income scores will be weighed equally for the final product line score.[55]

Combining Weighted Average of Different Product Line Scores

To combine the different product line scores in a given assessment area into a Retail Lending Test Area Score for that area, the agencies will weight each product line score based on a combination of loan dollars and loan count associated with the product line in that particular assessment area.[56]

For example, if a major product line contained 50% of a bank’s loans in a Retail Lending Test Area in dollar amount and 30% of a bank’s loans in that area in loan count then the weight assigned to the product line score would be 40%.[57]

After combining the different product line scores to get the Retail Lending Test Area Score, they will apply this to the following range of performance ratings to get the final rating for that assessment area.[58]

  • Outstanding – 8.5 or more
  • High Satisfactory – 8.4 to 6.5
  • Low Satisfactory – 6.4 to 4.5
  • Needs To Improve – 4.4 to 1.5
  • Substantial Noncompliance – 1.4 or below
Determining Retail Lending Test Ratings

After determining Retail Lending Test Area Scores for each assessment area, the regulators will combine these scores using weighted averages to come up with a rating for the Retail Lending Test for each state and each Metropolitan Statistical Area (MSA), as well as the institution’s overall rating.

For states and multi-state MSA’s, the weights of Retail Lending Test Area Scores will be determined by an average of two factors;   (1) the bank’s share of total  deposits held in the assessment area from within the specific state or multi-state MSA, and (2) combined average of the loan count and dollar amount of loans from the  assessment area within the specific state or multistate MSAs. After applying these weights to the Retail Lending Area Test Scores, they will then be averaged to come up with a final result, or Retail Lending Test conclusion. That conclusion will then be compared to the same performance ranges described above used to determine the rating for specific assessment areas.[59]

Institutions’ overall lending test ratings follow the same method. Each assessment area’s Retail Lending Test Area Score will be weighted based on the average of (1) the banks share of deposits in the assessment area, and (2) the banks share of loans using loan dollars and loan counts. These scores will then be combined and averaged, and that final result will be compared to the same performance ranges used to determine state ratings.[60]

The agencies used 2018-2020 data to estimate how large and intermediate banks would fare on the new Retail Lending Test. They determined that 9.2% of the banks in their sample would have received failing ratings of “Needs To Improve” or “Substantial Noncompliance.”[61]

The final rule does not include a statistical model to identify markets where all lenders may be underperforming. The agencies note that they intend to develop statistical models that would predict the level of market benchmarks expected in facility-based assessment areas and retail lending assessment areas if they had been adequately served by lenders. The agencies are considering how to develop an appropriate statistical model and the preamble to the final rule notes they intend to solicit additional feedback from the public in developing such a model.[62]

Community Development Financing Test

All large banks face a new Community Development Financing Test created in the final rule. Large banks were previously evaluated separately on their community development loans, investments and services. Now those loans and investments will be combined under the Community Development Financing Test.[63] Intermediate banks can choose whether to adopt that new combined Community Development Test or keep their old one, which has been renamed the Intermediate Bank Community Development Test.[64]

All banks may now get credit for community development activities done anywhere in the country.[65]

Unlike the Retail Lending Test, the Community Development Financing Test is largely a qualitative evaluation: It does not include thresholds for determining ratings, although the test does include metrics and benchmarks to guide examiners in evaluating community development performance.[66]

The test also includes consideration of the impact and responsiveness of a banks’ community development loans and investments.

How the Community Development Financing Test Works

Regulators will determine the annual value of a bank’s community development loans and investments by dollar amount. In a change from the previous CRA, banks will now get credit for the remaining value of loans on the bank’s balance sheet. Previously banks would only get credit for remaining values of investments.[67]

The Community Development Financing Test includes components tied to facility-based assessment areas, to state borders, to MSAs and to nationwide performance. Additionally, bank activities relevant to the Community Development Financing Test will now also be evaluated for impact and for how responsive they are to community needs.

Facility-based Assessment Areas

Large banks will receive a rating on their Community Development Financing Test for each facility-based assessment area.[68] Specific ratings will be assigned to a point scale, using the same scale used for the Retail Lending Test. For example, Outstanding performance would merit 10 points.[69]

For each facility-based assessment area, examiners will create a Bank Assessment Area Community Development Financing Metric based on a bank’s community development loans, investments and deposits. The regulators will also create a local Assessment Area Community Development Financing Benchmark and two national MSA and Nonmetropolitan Nationwide Community Development Financing Benchmarks to compare a bank’s performance against other banks, but without assigned ratings based on performance ranges.[70]

  • Bank Assessment Area Community Development Financing Metric – This is calculated by dividing the dollar amount of a bank’s loans and investments that benefit or serve the specific facility-based assessment area by the dollar amount of deposits the bank has in the assessment area.[71]
  • Assessment Area Community Development Financing Benchmark – This is a local benchmark specific to the assessment area, calculated by summing all large-bank community development loans and investments that benefit or serve the assessment area and dividing that by the sum of large-bank deposits in the assessment area[72]
  • MSA & Nonmetropolitan Nationwide Community Development Financing Benchmarks – These two national benchmarks are calculated similarly to the local benchmark. Both national benchmarks are calculated by dividing the total dollar amount of community development loans and investments from large banks that serve either urban MSAs or rural nonmetropolitan areas by the dollar amount of deposits large banks hold in either MSAs or nonmetropolitan areas. The regulators will compare the Bank Assessment Area Community Development Financing Metric to either the MSA or Nonmetropolitan benchmark depending on whether the assessment area is located in an MSA or nonmetropolitan area.[73]
State Community Development Financing Evaluation

Banks will get a state rating for each state they have branches in, as long as the branch is not located in a multistate MSA, in which case it will be evaluated as part of the bank’s separate Multistate MSA Community Development Financing Evaluation as described below.[74]

To determine the state rating, regulators will use two different components. One uses statewide benchmarks and metrics. The other is facility-based and factors into the rating in similar fashion to the Retail Lending Test’s method for combining assessment areas.[75]

Statewide Component for Determining Rating

  • Bank State Community Development Financing Metric –  This is calculated similarly to the assessment area metric. It will divide the dollar amount of a bank’s community development loans and investments in the state, regardless of whether they are in a facility-based assessment area or not, by the bank’s deposits in the state.[76]
  • State Community Development Financing Benchmark – Similar to the assessment area benchmark, this is calculated by dividing total community development loans and investments from all large banks in the state by deposits of all large banks in the state. Deposits will use annual averages of deposits based on the years in the evaluation period.[77]
  • State Weighted Assessment Area Community Development Financing Benchmark – This benchmark is determined by weighing each of the Assessment Area Community Development Financing Benchmarks for each facility-based assessment area the bank has in a particular state. These benchmarks are weighted based on a combined average of the percentage of retail loans in the facility-based assessment area (using both loan counts and dollars) and deposits.[78]
Facility-Based Component for Determining Rating

The weight of each Facility-based Assessment Area’s Community Development Test score will be determined by an average of two factors; (1) the bank’s share of total customer deposits held in the assessment area from within the specific state or multi-state MSA, and (2) combined average of the loan count and dollar amount of loans from the specific assessment area within the specific state or multistate MSAs. After applying these weights to all of the facility-based Community Development Test scores, they will then be averaged to come up with the facility-based component of a state’s Community Development Financing Test rating.

The weights applied to the facility-based component or statewide component will be determined by where the bank lends and has deposits in the state. Specifically, the regulators will average the percentage of a bank’s loans (using both loan count and dollar amount) and deposits in its facility-based assessment areas. Banks with more of their loans and deposits in facility-based assessment areas will have the facility-based metrics and benchmark matter more in determining the final rating and banks with less of their loans and deposits will have their statewide metrics and benchmarks weighed more[79]

Multistate MSA Community Development Financing Test Evaluation

Large banks will be rated on their community development loans and investments in their multistate MSAs if they maintain branches in two or more states in the multistate MSA. Washington, DC, is an example of a multistate MSA as the metropolitan area includes the District of Columbia and counties in Maryland and Virginia. Other examples include Cincinnati and Kansas City.[80]

The Multistate MSA Community Development Financing Test Evaluation will be done the same way as the state test, with the same MSA specific metrics and benchmarks.[81]

Nationwide Area Community Development Financing Test Evaluation

The nationwide overall rating for the Community Development Financing Test will be determined similarly to the state test, with nationally specific metrics and benchmarks and two components reflecting aggregate nationwide performance and a weighted average of performance in all the bank’s facility-based assessment areas based on loans and deposits. The weight of these components will be determined by the percentage of loans and deposits in facility-based assessment areas in the same fashion used for generating state ratings.[82]

Bank Nationwide Community Development Investment Metric and Benchmark

The one difference between national and state-level community development evaluations is the creation of the Bank Nationwide Community Development Investment Metric and Benchmark.

These provisions reflect comments submitted by NCRC and NCRC members requesting a separate metric for investments in order to ensure that combining loans and investments on the Community Development Financing Test does not lead to banks decreasing investments.

In response to those comments, the agencies added an additional metric and benchmark for banks with more than $10 billion in assets, dubbed the Bank Nationwide Community Development Investment Metric and the Nationwide Community Development Investment Benchmark respectively. The bank metric will analyze a bank’s investments as a share of deposits and compare that to the national benchmark of all other banks above $10 billion in assets. Investments in mortgage-backed securities will not be counted in this analysis.[83]

However, bank performance under this metric and benchmark can only contribute positively to a bank’s performance.

This is a change from the status quo. Previously, all large banks would have a separate evaluation of investments factored into their overall rating whether its effect was negative or positive.

The preamble to the final rule notes that regulators will look for an “appreciable decline” in investments or loans caused by combining them and may establish separate tests. They have not defined an “appreciable decline,” however – neither by establishing what magnitude of decline would trigger a response, nor by establishing if declines will be scrutinized at national, regional, state or MSA level.[84]

Since intermediate banks already have a community development test that combines loans and investments, it would have been helpful for the agencies to do a comparison of the amount of loans compared to investments of banks at different asset sizes. If intermediate banks pursue more loans compared to investments, particularly more than large banks near the intermediate threshold that have similar capacity, this would support the intuitive theory that combining these two activities could lead to a decrease in investments.

Impact and Responsiveness Review

All facility-based, state, multistate MSA and national community development evaluations also include a review of the impact and responsiveness of the banks community development loans and investments.

The regulators have identified twelve factors to consider in evaluating impact and responsiveness, while noting that they would consider other factors as well[85]

The twelve Factors of Impact and Responsiveness assess whether a given community development activity…:

…Serves persistentpoverty counties, defined as those with poverty rates of 20% or more for 30 years;[86]

  • Many persistent-poverty counties are rural. Examples of persistent poverty counties include many parts of the Mississippi Delta, Appalachia, “colonias” in the Rio Grande River Valley, Native Land Areas and Fresno and Tulare counties in California’s San Joaquin Valley.[87]

2.  …Serves census tracts with a poverty rate of 40% or higher;[88]

  • Unlike persistent poverty counties, these census tracts are mainly urban[89]

3.  …Serves areas with low levels of community development financing;[90]

4.  …Supports a Minority-Owned Depository Institution (MDI), Women-Owned Depository Institution (WDI), Low Income Credit Union (LICU), or Community Development Financial Institution (CDFI), excluding certificates of deposit with a term of less than one year;[91]

5.  …Benefits or serves individuals, families, or households with low income;[92]

6.  …Supports small businesses or small farms with gross annual revenues of $250,000 or less;[93]

7.  …Directly facilitates the acquisition, construction, development, preservation or improvement of affordable housing in High Opportunity areas;[94]

  • High Opportunity Areas are defined by the Fair Housing Finance Agency (FHFA) as (1) areas designated by HUD as a “Difficult Development Area”; or (2) areas designated by a state or local Qualified Allocation Plan as High Opportunity Areas and where the poverty rate falls below 10% (for metropolitan areas) or 15% (for nonmetropolitan areas).

8.  …Benefits or serves residents of Native Land Areas;[95]

9.  …Is a grant or donation;[96]

  • These must be tied to CRA-eligible community development work.

10.  …Is an investment in projects financed with Low Income Housing Tax Credits or New Markets Tax Credits;[97]

11.  …Reflects bank leadership through multi-faceted or instrumental support; [98]

  • This covers activities that entail multiple forms of support provided by the bank for a particular program or initiative. For example, a loan to a community-based organization serving LMI individuals coupled with a service supporting that organization in the form of technical assistance that leverages the bank’s financial expertise would activate this provision of the assessment. Instrumental support would cover activities that involve a level of support or engagement on the part of the bank such that a program or project would not have come to fruition, or the intended outcomes would not have occurred, without the bank’s involvement.[99]

12.  …Is a new community development financing product or service that addresses community development needs for LMI individuals, families, or households.[100]

The agencies will provide a summary of a bank’s impact and responsiveness review data, such as the volume of activities by impact and responsiveness review category.[101]

Large banks will be required to collect and maintain data related to impact and responsiveness factors. This data could help the regulators create additional metrics and benchmarks related to impact, since the regulators cited a lack of information as a challenge for establishing performance measures in the final rule.[102]

The agencies are considering issuing additional guidance to examiners “in the near term” to provide additional clarity on how to apply the impact and responsiveness review. This guidance may include examples of criteria that examiners could consider in evaluating the impact and responsiveness of a bank’s community development loans and investments, including:

  • the percentage of a bank’s community development loans and investments that meet one or more impact and responsiveness factors;
  • the dollar amount and number of community development loans that meet one or more impact and responsiveness factors; and
  • reasons for providing more or less weight to the impact and responsiveness component of the Community Development Financing Test.[103]
Definitions of Community Development

The CRA requires large and intermediate banks to pursue community development loans, investments and services on an ongoing basis.

Defining what is included in CRA eligible community development is a way of setting priorities. The specifics of those definitions are thus a vital element of CRA implementation. The updating and expansion of CRA-eligible community development activities detailed below is only one way in which the final rule attempts to shore up this key line of communication between regulators and banks.

The agencies have also committed to producing a non-exhaustive illustrative list of community development activities that would receive CRA credit, and to creating a formal pre-approval confirmation process to give institutions confidence. The agencies will jointly issue a publicly available illustrative, non-exhaustive list of examples of loans, investments and services that qualify for community development consideration.[104] The rule creates a formal process for banks to request confirmation that activities are CRA-eligible community development. Regulators may impose conditions in confirming that activities are eligible,[105] but regulators declined to open this formal process to anyone other than banks.[106]

Those two novel efforts will help banks orient themselves effectively. But the most important things for community development practitioners to understand on this front are in the definitions of community development activity themselves.

The  final rule establishes that banks will receive full credit for community development loans, investments and services if they meet any of the following criteria.[107] Several of the eligible categories have changed little from the previous rules. But there are also significant overhauls to affordable housing, and important new eligible categories to encourage community development on tribal lands and investments that will help LMI communities survive the changing climate.

Majority Standards

This criterion has two components. First, the activity must support one or more of the community development categories enumerated below. Second, it must meet one or more of the specific majority standard criteria established for each community development category. The category specific criteria vary, but often specify that the majority of people benefiting are LMI.[108] These standards are intended to ensure that most of the activity benefits LMI households, small businesses and farms, or residents of tribal lands, as applicable.[109]

Bona Fide Intent Standard

The agencies seek to encourage beneficial activity by extending CRA credit to certain projects where it is evident the bank had a genuine intent to fulfill the law’s goals but may not be able to demonstrate such under the Majority Standards. An activity meets the Bona Fide Intent Standard if:

  • The housing units, beneficiaries, or proportion of dollars necessary to meet the majority standard are not reasonably quantifiable;
  • The loan, investment or service has the express, bona fide intent of one or more of the community development purposes; or,
  • The loan, investment or service is specifically structured to achieve one or more of the community development purposes.[110]

In addition to the majority and bona fide intent standards, banks will also receive full credit for loans, investments, or services to MDIs, WDIs, LICUs and CDFIs, and community development related to LIHTC-financed projects.[111][112]

Community Development Categories

Community development must fit one or more of these 11 categories in the final rule in order to be CRA-eligible:[113]

  1. Affordable housing, with new provisions discussed below
  2. Economic development
  3. Community supportive services (activities that assist, benefit, or contribute to the health, stability, or well-being of LMI individuals[114])
  4. Revitalization or stabilization
  5. Essential community facilities
  6. Essential community infrastructure
  7. Recovery of designated disaster areas
  8. Disaster preparedness and weather resiliency, a new category discussed in detail below
  9. Any of items 4 through 8 effected within Native Land Areas specifically, a significant expansion of CRA’s incentives discussed further below
  10. Activities with Minority-owned depository institutions (MDIs), Women-owned depository institutions (WDIs), low- income credit unions (LICUs), or Community Development Financial institutions  (CDFIs)
  11. Financial literacy

Three of these are new or significantly updated in ways that merit further discussion:

Affordable Housing

Four types of affordable housing activities are eligible for CRA credit:[115]

  • Affordable rental housing developed in conjunction with Federal, State, local and tribal government programs
  • Multifamily rental housing with affordable rents, also referred to as naturally occurring affordable housing (NOAH)
  • Activities supporting affordable LMI homeownership
  • Purchases of mortgage-backed securities that finance affordable housing

The definition of naturally occurring affordable housing, or NOAH, requires further unpacking.

For a development to qualify for affordable housing community development credit under the NOAH provision, the final rule requires that monthly rent for the majority of units will not exceed 24% of area median income (AMI).[116] This is a softening of the stance taken in the proposed rule,[117] which conditioned affordable housing CRA credit upon a lower income level (18% of AMI) that more effectively addressed affordability.

It is not enough for naturally occurring affordable housing to be priced according to the rule’s standards. It must also meet one or more of the following criteria:[118]

  • The housing is located in a LMI census tract.
  • The housing is located in census tract in which the median renter is LMI and the median rent does not exceed 24% of AMI.
  • The housing is purchased, developed, financed, rehabilitated, improved or preserved by any nonprofit organization with a stated mission of, or that otherwise directly supports, providing affordable housing.
  • The bank can provide documentation that a majority of the housing units are occupied by LMI individuals or families.

The originally proposed rule also offered another possible criteria for defining NOAH: That property owners would make a written pledge to keep rents affordable for at least five years or the length of financing. The agencies took this criteria out because they determined that neither the regulators nor the banks would be able to enforce these pledges.[119]

The agencies declined to condition NOAH qualification, or affordable housing CRA eligibility in general,  in two important ways NCRC and many NCRC members had advised. Banks do not need to show that their affordable housing projects will not displace LMI residents, nor that the projects follow responsible lending practices. The preamble to the final rule notes that CRA examiners will “retain the discretion to consider whether an activity reduces the number of housing units affordable to low- or moderate-income individuals,” a nod at the general problem of displacement. But displacement can occur even if rents are affordable. And this does not cover the quality of the housing, which was meant to be addressed through responsible lending practices such as screening for developers and property owners that frequently harass tenants or don’t keep housing up to code.[120] Although the agencies declined to adopt standards for responsible practices, the direction to examiners at least gives advocates standing to raise these issues on CRA exams in the future.

The final rule also maintains the practice of allowing partial CRA credit for housing that does not meet the majority affordability standard, but requires this to be done in conjunction with a government affordable housing plan, program, initiative, tax credit or subsidy. The amount of partial credit is based on the percentage of housing units affordable to LMI households. This is the only instance where the regulators will allow partial credit. All other community development activities must meet majority standards.[121]

Disaster preparedness and Weather Resiliency

Major new provisions referred to in the proposed rule as “climate resiliency” are re-labeled as “weather resiliency” activities in the final rule. The agencies made this change to add clarity to the types of activities that qualify.[122] The rule defines this category around activities that “assist individuals and communities to prepare for, adapt to and withstand natural disasters or weather-related risks or disasters.”[123] These activities must:

  • benefit or serve targeted census tracts such as LMI census tracts;
  • be done in conjunction with either (1) a plan, program or initiative of a federal, state, local or tribal government, or (2) a mission-driven nonprofit organization that is focused on benefiting or serving targeted census tracts; or,
  • benefit or serve residents of targeted census tracts, and not directly result in forced or involuntary relocation of LMI individuals in the targeted census tracts.[124]

The final rule’s inclusion of mission-driven nonprofit partners as a means of fulfilling this criterion is responsive to comments on the proposed rule from NCRC and others, who noted that many state and local governments do not have plans or programs in place for addressing the effects of the climate crisis, or “disaster preparation or weather resiliency” in the language of the rule.[125]

The rule provides several examples of activities that would qualify under the disaster preparedness and weather resiliency category, including the construction of flood control systems in a flood-prone targeted census tract, promoting green space in targeted census tracts in order to mitigate the effects of extreme heat, community solar projects and microgrid and battery projects that could help ensure access to power to an affordable housing project in the event of severe storms.[126]

The agencies note that LMI communities are more affected by “weather-related risks” but largely avoid a discussion of how climate change is increasing the frequency and severity of these risks. They specifically declined to consider activities related to decarbonization and transition to clean energy as eligible, citing difficulties with determining how it would benefit residents of LMI census tracts and other targeted census tracts. However, financing that increases energy efficiency should qualify if it meets the criteria.[127]

This approach is more focused on addressing the weather-related symptoms of climate change than the fossil fuels that are causing it.

The CRA would be more effective at protecting and preserving the livelihoods of LMI communities and communities of color if it evaluated banks on the environmental impact of their loans, investments and underwriting of bonds and equities. Many banks have already made climate commitments and some even include them in merger applications as examples of how they serve the convenience and needs of the public. One way to partially address this would be to create an impact and responsiveness criteria for the environmental effects of a bank’s financing.

Native Land Areas: CRA Adds A Vital New Tool

The expansion of CRA’s community development test incentives to projects in tribal areas is a big deal. The change is simple on a mechanical level: Similar activities that were always rewarded under the test are now explicitly eligible if they serve Indian Country. What makes this change noteworthy is the severity of the deprivation facing the millions of people living in such places. Read NCRC’s new report in conjunction with Native Community Capital for a much more detailed dive into those challenges.

Other Significant CDF Test Changes

In a change from the previous CRA rules, certain direct loans to small businesses or small farms may be considered under both the Community Development Financing Test and the Retail Lending Test. In order for a retail loan to a small business or farm to qualify under the economic development category of community development, it needs to meet a size and purpose test, as well as be done in conjunction or in syndication with a government plan, program or initiative, such as an SBA loan.[128]

To meet the size test, the loans must be to businesses or farms with less than $1 million in revenue, or that meet size eligibility standards of Small Business Development Centers (SBDCs) or a Small Business Investment Company (SBIC).[129]

To meet the purpose test, the loans must have the purpose of promoting permanent job creation or retention for LMI individuals or in LMI census tracts.[130]

NCRC urged the agencies to also reward activities that support media outlets whose leadership or audience reflect traditionally underserved groups, such as Black, Indigenous and people of color (BIPOC) communities. The agencies did not do so explicitly – but it would seem that loans to such businesses would still be rewarded in the CDF test provided they meet the size and purpose criteria. Allowing loans that meet certain requirements to count under both the Retail Lending Test and the Community Development Financing Test could encourage banks to seek out loan opportunities with businesses that also serve a community development purpose.

Retail Services and Products Test

This is the updated Service test that applies to large banks. In general, this test evaluates a bank’s branch network, the availability and accessibility of retail banking services and products and the responsiveness of services and products to community needs, particularly for LMI households as well as small businesses and farms. Similar to the Community Development Financing Test, this test uses some quantitative metrics and benchmarks, but it does not tie performance ranges to specific ratings like the Retail Lending Test. Examiners will instead determine ratings while considering metrics and benchmarks, as well as community needs and other qualitative factors.[131]

Retail Banking Services

Banks with assets above $10 billion will be evaluated on

  • Branch availability and services
  • Remote service facility availability (ATMs for example)
  • Digital delivery systems and other delivery systems

Large banks with assets between $2 billion and $10 billion will not be evaluated on digital and other delivery systems, unless the bank requests to be evaluated on it and collects the required data, or if the bank does not maintain any branches.[132]

All large banks’ branch networks will be analyzed in their facility-based assessment areas by comparing the percentage of their branches in census tracts at each income level (Low, Moderate, Middle and Upper) to three separate benchmarks determined by the percentage of (1) census tracts, (2) total households, businesses and farms and (3) full service bank branches, at each census tract income level in the specific assessment area.[133]

Positive consideration will also be given for branches in middle- and upper-income census tracts if the bank can document the extent to which LMI households use services offered at the branch.[134]

Positive consideration will also be given for branches on Native Lands and in distressed or underserved nonmetropolitan middle-income census tracts.[135]

Regulators will continue reviewing the impact of opening and closing branches on the accessibility of banking services.[136] Regulators will also consider the “reasonableness” of bank branch hours in LMI census tracts compared to middle- and upper-income census tracts, including a comparison of whether branches offer extended or weekend hours.[137]

Regulators will also evaluate the range of services provided at branch locations that improve access to financial services or decrease costs for LMI customers. The following services are provided as examples, but this is not considered to be an exhaustive list:[138]

  • Bilingual and translation services
  • Free or low-cost check cashing services, including but not limited to check cashing services for government-issued and payroll checks
  • Reasonably priced international remittance services
  • Electronic benefit transfers

The availability and responsiveness of a bank’s digital and other delivery systems – including mobile banking, one of the many key shifts in banking that CRA modernization here addresses for the first time – will be evaluated on:[139]

  • Digital activity by individuals in census tracts at each income level, such as the number of checking and savings accounts opened digitally and numbers of such accounts that remain active at the end of each year
  • Range of digital and other delivery systems;
  • Bank strategies and initiatives to serve LMI individuals with digital and other delivery systems (including the cost, features and marketing thereof)
Retail Banking Products

The evaluation of credit and deposit products can only contribute positively to a bank’s rating on the Retail Services and Products test, meaning that banks cannot be downgraded for failing to offer responsive credit and deposit products. The evaluation of retail banking products will be done on a nationwide level and not at the assessment area level.[140]

All large banks will be evaluated on their credit products. Large banks with more than $10 billion in assets will also have their deposit products analyzed. Only credit and deposit products are included in this evaluation, with no assessment of insurance or investment products.[141]

The following categories of credit products will be evaluated:[142]

Credit products and programs that facilitate home mortgage and consumer lending targeted to LMI borrowers

  • Small-dollar mortgages and consumer lending programs that utilize alternative credit histories are provided as examples of responsive products/programs.[143]

Credit products and programs that meet the credit needs of small businesses and farms, including those with revenues of $250,000 or less

  • Examples include micro loans of $50,000 or less, or patient capital loans with longer terms.[144]

Credit products and programs conducted in cooperation with MDIs, WDIs, LICUs, or CDFIs

  • This category includes loan purchases from these institutions.[145]

Low cost education loans

Special purpose credit programs (SPCPs)

  • SPCPs were not included here in the initially proposed rule, but have been added into the final rule after NCRC and many others submitted comments arguing for the change. SPCPs do not have to come with income restrictions to qualify for CRA credit.[146]

Regulators will determine how responsive credit products in these categories are by reviewing their quantity, features, accessibility and affordability.[147]

Large banks above $10 billion in assets will be evaluated on the cost, availability, usage and responsiveness of their deposit products.[148]

Examples of responsive deposit products include savings accounts targeted toward LMI customers such as Family Self-Sufficiency Accounts,[149] “second-chance accounts” and accounts certified under the Cities for Financial Empowerment Fund’s “Bank On National Account” standard, which precludes overdraft and insufficient funds fees.

To evaluate usage of deposit products, regulators will look at:

  • the number of responsive accounts opened and closed during each year of the evaluation period in census tracts at every income level;
  • the percentage of total responsive deposit accounts compared to total deposit accounts for each year of the evaluation period; and,
  • marketing, partnerships and other activities that the bank has undertaken to promote awareness and use of responsive deposit accounts by LMI customers.[150]

To determine state and multistate MSA ratings, regulators will apply a weighted average to the banks ratings in each facility-based assessment area based on a combined average of the share of loans and deposits in the assessment area. Only branches, remote service facilities and retail banking services will be used to make conclusions at the facility-based assessment area, state and multistate level.[151]

To determine the overall rating, regulators will consider how the bank fared on the evaluation of Retail Services at the state and multistate level, as well as determine if the bank will receive positive consideration for credit and deposit products. The exact weighing of these two components is left to examiners.[152]

Community Development Service Test

This test evaluates large bank performance in providing community development services. Similar to the Community Development Financing Test and the Retail Services and Products Test, this is a largely qualitative test that uses some quantitative metrics, but it does not establish specific performance ranges tied to ratings.[153]

Community development services are evaluated based on:[154]

  • The number of services attributable to each community development category;
  • The capacities in which board members or employees of a bank or its affiliates serve (for example, do bank staff serve on boards of nonprofits);
  • Total hours of community development services performed by the bank;
  • Any other evidence demonstrating that the bank’s community development services are responsive to community development needs, such as the number of LMI individuals that are participants or the number of organizations served; and,
  • The impact and responsiveness of the bank’s community development services that benefit or serve the facility-based assessment area.

Scores will be given for each state and each multistate MSA, and a separate overall score generated by applying weighted averages of loans and deposits from each facility-based assessment area. Scores can be adjusted upward by community development services performed nationwide outside of facility-based assessment areas.[155]


Given CRA’s legislative purpose of addressing redlining by evaluating banks on how well they meet the financial service needs of the entire community, it’s disappointing that the CRA continues to leave race out of the process. NCRC offered several recommendations for doing so, such as creating a disparity study that would identify racial groups and communities with significant gaps in loans and investments  and factoring a bank’s performance at serving these identified groups into the exam process.

Instead, the agencies will start publishing data on the lending performance of large banks by race and ethnicity in all of their assessment areas using Home Mortgage Disclosure Act (HMDA) data.[156] This information will be put on the agencies’ websites, but will not be republished in CRA exams or in any way affect a bank’s CRA rating.

Banks will also have to update their public notices to note that their HMDA data is available on the CRA regulators’ websites, which is not much of a change since these notices already had to include that HMDA information was available on the Consumer Financial Protection Bureau’s own website.[157] Once Section 1071 data is available, banks required to report this data will update their public notice to note that their small business and farm lending data is available on the CFPB’s website. The agencies should work to ensure that data is presented on websites in the most user-friendly way possible and should provide links to tables on CRA exams.

Effect of CRA Performance on Applications

CRA performance is considered when banks apply for mergers, as well as in applications to open or relocate branches. The possibility of getting these applications denied is the main consequence of failing CRA performance.[158]

The regulators proposed no changes to how CRA performance will be used to evaluate such applications.[159]

The regulators did note that the majority of commenters supported the idea of requiring, or at least encouraging, the use of community benefits plans in merger applications. This is important to note in upcoming rulemakings related to bank merger review, which like the CRA is also in dire need of an update.[160]


Although the effective date of the final rule is April 1, 2024, the applicability date for the majority of the provisions is January 1, 2026.[161] Banks will have until January 1, 2027, to comply with the reporting requirements related to assessment area data, operating subsidiaries and other affiliates and community development loans and investments with a third party. From that point forward, such data must be reported by April 1 each year beginning in 2027.[162]

[1] CRA Final Rule. Page 49

[2] CRA Final Rule. Page 42

[3] “Interagency Overview of the Community Reinvestment Act Final Rule” Page 2.

[4] These percentages came from comparing the estimated number of banks that would be reclassified as either intermediate or small banks discussed by the agencies on page 44 of the final rule, to the number of banks currently evaluated as intermediate (1,416 intermediate banks) or small (2,512 small banks) based on asset data from the FDIC’s BankFind Suite accessed on 11.6.2023.

[5] CRA Final Rule. Page 49

[6] CRA Final Rule. Pages 879-882

[7] CRA Final Rule. Page 886-889

[8] CRA Final Rule. Page 889

[9] CRA Final Rule. Page 891

[10] CRA Final Rule. Page 891

[11] CRA Final Rule, p. 421

[12] CRA Final Rule. Page 840

[13] CRA Final Rule. Pages 841-843

[14]CRA Final Rule. Pages 330 and 334-335

[15] CRA Final Rule. Pages 345, 348

[16] CRA Final Rule. Pages 349

[17] CRA Final Rule. Page 361

[18] CRA Final Rule. Page 349

[19] CRA Final Rule. Page 350

[20] CRA Final Rule. Pages 349-351

[21] CRA Final Rule. Pages 351-352

[22] CRA Final Rule. Page 388

[23] CRA Final Rule. Page 384

[24] CRA Final Rule. Page 444

[25] CRA Final Rule. Page 454

[26] CRA Final Rule. Page 455

[27] CRA Final Rule. Pages 438-439

[28] CRA Final Rule. Pages 444 and 495

[29] CRA Final Rule. Pages 472 and 475

[30] CRA Final Rule. Page 490

[31] CRA Final Rule. Page 472

[32] CRA Final Rule. Page 445

[33] CRA Final Rule. Page 514

[34] CRA Final Rule. Page 438

[35] CRA Final Rule. Page 456-461

[36] CRA Final Rule. Page 466

[37] CRA Final Rule. Pages 638 – 640

[38] CRA Final Rule. Pages 356 and 384

[39] CRA Final Rule. Pages 445, 499, 533 and 539

[40] CRA Final Rule. Pages 445, 513 and 543

[41] CRA Final Rule. Page 444

[42] CRA Final Rule. Page 470

[43] CRA Final Rule. Page 562

[44] CRA Final Rule. Page 446

[45] CRA Final Rule. Pages 552 – 553

[46] CRA Final Rule. Page 446

[47] CRA Final Rule. Pages 609, 616 – 620

[48] CRA Final Rule. Page 592

[49] CRA Final Rule. Page 592

[50] CRA Final Rule. Pages 592 and 595

[51] CRA Final Rule. Page 627-628

[52] CRA Final Rule. Page 626

[53] CRA Final Rule. Page 624

[54] CRA Final Rule. Page 624

[55] CRA Final Rule. Pages 624 and 628

[56] CRA Final Rule. Page 631

[57] CRA Final Rule. Page 631

[58] CRA Final Rule. Page 631

[59] CRA Final Rule. Page 652-654

[60] CRA Final Rule. Page 654

[61] CRA Final Rule. Page 659

[62] CRA Final Rule. Page 646

[63] CRA Final Rule. Page 401

[64] CRA Final Rule. Page 405

[65] CRA Final Rule. Page 393

[66] CRA Final Rule. Page 731

[67] CRA Final Rule. Page 747

[68] CRA Final Rule. Page 781

[69] CRA Final Rule. Page 782

[70] CRA Final Rule. Page 759

[71] CRA Final Rule. Page 765

[72] CRA Final Rule. Page 772

[73] CRA Final Rule. Page 773

[74] CRA Final Rule. Pages 782 and 786

[75] CRA Final Rule. Page 782, 787 and 790

[76] CRA Final Rule. Page 783

[77] CRA Final Rule. Page 784

[78] CRA Final Rule. Page 784 and 789

[79] CRA Final Rule. Page 790 and 795

[80] CRA Final Rule. Page 796

[81] CRA Final Rule. Page 796-797

[82] CRA Final Rule. Page 804

[83] CRA Final Rule. Pages 737 – 738 and 801

[84] CRA Final Rule. Pages 401 – 402

[85] CRA Final Rule. Pages 294-296

[86] CRA Final Rule. Page 301

[87] CRA Final Rule. Page 302 and US Census Bureau – Persistent Poverty

[88] CRA Final Rule. Page 301

[89] CRA Final Rule. Page 303

[90] CRA Final Rule. Page 301

[91] CRA Final Rule. Pages 306-307

[92] CRA Final Rule. Page 309

[93] CRA Final Rule. Page 311

[94] CRA Final Rule. Page 313

[95] CRA Final Rule. Page 315

[96] CRA Final Rule. Page 318

[97] CRA Final Rule. Page 319

[98] CRA Final Rule. Page 321

[99] CRA Final Rule. Page 320

[100] CRA Final Rule. Page 322

[101] CRA Final Rule. Page 297

[102] CRA Final Rule. Page 779

[103] CRA Final Rule. Page 779

[104] CRA Final Rule. Page 286

[105] CRA Final Rule. Page 290

[106] CRA Final Rule. Page 291

[107] CRA Final Rule. Pages 129-130

[108] CRA Final Rule. Page 130

[109] CRA Final Rule. Page 130

[110] CRA Final Rule. Page 131

[111] CRA Final Rule. Page 132

[112] CRA Final Rule. Page 132

[113] CRA Final Rule. Page 119

[114] CRA Final Rule. Page 10

[115] CRA Final Rule. Pages 139-140

[116] The rule describes its definition of affordably-priced housing as “30% of 80%” of AMI, which we simplify here to simply 24% AMI. The agencies’ language reflects the consensus definition of housing affordability as rent that is no more than 30% of a household’s income. Pegging that 30% figure to an income level below the area median represents one approach to targeting new housing development to those most in need. The agencies decided to use 80% as that sub-median peg, rather than the 18% of AMI (or 30% of 60%) figure preferred by many advocates.

[117] CRA Final Rule. Page 151

[118] CRA Final Rule. Pages 151-152

[119] CRA Final Rule. Pages 156-157

[120] CRA Final Rule. Page 159

[121] CRA Final Rule. Page 134

[122] CRA Final Rule. Page 251

[123] CRA Final Rule. Page 252

[124] CRA Final Rule. Page 257

[125] CRA Final Rule. Page 257

[126] CRA Final Rule. Page 254

[127] CRA Final Rule. Pages 252-254

[128] CRA Final Rule. Pages 189-193

[129] CRA Final Rule. Page 189

[130] CRA Final Rule. Page 189

[131] CRA Final Rule. Pages 670 and 673

[132] CRA Final Rule. Pages 674 and 676

[133] CRA Final Rule. Pages 679-680

[134] CRA Final Rule. Page 685

[135] CRA Final Rule. Page 685-686

[136] CRA Final Rule. Page 688

[137] CRA Final Rule. Pages 688-689

[138] CRA Final Rule. Pages 688 and 1083

[139] CRA Final Rule. Pages 692 and 694-695

[140] CRA Final Rule. Page 726

[141] CRA Final Rule. Page 699

[142] CRA Final Rule. Pages 706 and 1086

[143] CRA Final Rule. Page 707

[144] CRA Final Rule. Page 707

[145] CRA Final Rule. Page 707

[146] CRA Final Rule. Page 709

[147] CRA Final Rule. Pages 699 and 705

[148] CRA Final Rule. Pages 711-713

[149] CRA Final Rule. Page 713

[150] CRA Final Rule. Pages 717-719

[151] CRA Final Rule. Page 722

[152] CRA Final Rule. Page 722

[153] CRA Final Rule. Pages 813-814

[154] CRA Final Rule. Page 1094

[155] CRA Final Rule. Page 1094-1095

[156] CRA Final Rule. Page 11

[157] CRA Final Rule. Pages 994-996.

[158] CRA Final Rule. Page 923

[159] CRA Final Rule. Page 923

[160] CRA Final Rule. Page 924

[161] CRA Final Rule. Pages 11-12

[162] CRA Final Rule. Page 12

Print Friendly, PDF & Email
Scroll to Top