I’ve written at least a dozen articles about the dramatic impact the new CRA is going to have on banks. Until now, only about 1.2% of bank CRA examinations end with a failing grade. But, by the estimates of the regulators, if the new rule had been in effect during 2018-2020 that failure rate could have been ten times higher! Think about that. Historically, only about 1 in 100 CRA exams ended in a failing grade. But under the new rule the regulators estimate that 1 in 10 banks will fail their Retail Lending Test - and failing the Retail Lending Test means you fail the entire CRA exam.

It’s understandable that anxiety is widespread among bank CRA officers given those scary numbers. I’ve been contacted by innumerable bankers asking how they can figure out how their bank will fare under the new rule when it becomes completely “applicable” at the beginning of 2027. This article will outline the steps a bank can take to estimate what their performance may look like under the new rule.

It’s impossible to precisely determine what your numbers and the benchmark numbers you will be compared to will be in 2027, but that doesn’t mean you can’t approximate those numbers and get a ballpark estimate of whether you will attain a satisfactory performance rating (and gain some insight into where you look strong and where you may have some problems).

The process begins by understanding how the new CRA will measure performance. Although there will be 2 tests for Intermediate Banks and 4 tests for large banks, the most important test of all is the Retail Lending Test. The new CRA (and the current CRA) demands that a bank must attain a satisfactory performance rating (or a low satisfactory performance “conclusion”) on the Retail Lending Test, or it cannot attain a “composite” satisfactory performance rating at the institution level. I will write another article about the Community Development Financing Test, which is important too, but getting a satisfactory rating on the Retail Lending Test is the sine qua non for a successful CRA performance rating.

The first step to estimate how your bank may look under the new rule is to determine your Facility-based assessment areas and your new Retail Lending Assessment Areas because that’s where most of your Retail Lending Test performance activity will be concentrated.

The Facility-based assessment areas are where you maintain deposit-taking facilities, just like you do now. Except that the regulators have made the radical and destructive decision to remove the flexibility in the current regulation that allows banks to delineate assessment areas that a bank “can reasonably be expected to serve”. Now (beginning April 1, 2024) a “large” bank must include whole counties in their assessment areas, even if it’s not reasonable to expect a bank to serve the entire county. This will create a very big problem for banks that have branches very near the borders of counties where the bank doesn’t have a branch (It will also create problems for banks in very large counties where they may have a single branch or two). Those banks will need to decide if they annex those counties to their FBAA. This will require those banks to self-assess their performance with or without those nearby counties (as will be explained later in this article).

Large banks will also need to determine if they will have any Retail Lending Assessment Areas, and if so, where they are. Since the new rule defines a RLAA as any MSA or statewide non-MSA, or multistate MSA in which a bank has originated or purchased 150 or more closed-end mortgages or 400 or more small business loans, making those determinations should not be too difficult. Keep in mind that a RLAA triggered by the closed-end mortgage count does not mean small business loans are evaluated unless that activity equals or exceeds the threshold for small business loans. The same is true in reverse. So, it’s possible to have RLAA where only closed-end mortgages or small business loans are evaluated.

Once you have determined your FBAAs and RLAAs, the next step is to estimate how you will perform under the “distribution tests” in comparison to the “calibrated” benchmarks. The two distribution tests are the “geographic” and “borrower” distribution tests. The geographic distribution test measures your “penetration rates” in the low-income and the moderate-income census tracts and the borrower distribution test measures your lending to low- and moderate-income mortgage borrowers and to very small businesses or small farms. Under the new rule, your activity is based on the count (originations and purchases) of your “major product line” loans activity within each Assessment Area.

The calibrated benchmarks under the Retail Lending Test are determined by how you compare to other lenders in your Assessment Area (the “market” benchmark) and how you compare to certain demographics (the “community” benchmark). Specifically, you need to calculate the market and community benchmarks as follows:

• Geographic benchmarks lending in the low- and moderate-income tracts:
  • Market Benchmark: The penetration rate of all other closed-end mortgage lenders (mistakenly called “peers” by many bankers) in the low- and moderate-income tracts and
  • the percentage of small business loans (and small farm loans if you are a farm lender) in low- and moderate-income census tracts
  • Community Benchmark: The relative percent of owner-occupied houses in the low- and moderate-income tracts and the relative percent of businesses
  •  in the low- and moderate-income tracts
• Borrower benchmarks – lending to low-income and moderate-income mortgage borrowers and to very small business and small farms
  • Market Benchmark: the penetration rate of mortgage lending (closed-end mortgages) to low- and moderate-income borrowers and
  • the penetration rate of all reporting lenders to very small businesses and small farms (based on GAR)
  • Community Benchmark: the percentage of families that are low- and moderate-income in the assessment area and
  • The percentage of businesses in the assessment area that are very small (based on GAR)

Once you have the information above you can do the computations to see how you compare to these benchmarks. First calculate the market and community benchmarks and then apply an 80% “multiplier” to the market benchmark and a 60% “multiplier” to the community benchmark. Why am I suggesting those multipliers? Because they determine the minimum performance level necessary to attain at least a low satisfactory performance rating under the new CRA.

Some of the demographic benchmarks may be difficult to get. For example, the business demographics are difficult to get because they are controlled by a single source (yes, the regulators effectively gave a monopoly to a vendor). However, I've always thought the market benchmarks are much more important because they measure the real need for credit whereas the community benchmarks, although useful, are simply demographics that don't necessarily coincide with credit needs.

So, it’s really not that difficult to develop an estimate of your CRA performance and to determine if you will qualify for a satisfactory performance “supporting conclusion” under the new CRA.