As I write this article the Department of Justice is pursuing a record-breaking number of “redlining” cases against banks and mortgage lenders. The “Anti-redlining Initiative” was announced October 22, 2021, by Attorney General, Merrick Garland. In his press release AG Garland stated, “We know full well that redlining is not a problem from a bygone era but a practice that remains pervasive (emphasis added) in the industry today.” This language hints at an adversarial attitude that may be the reason for a big change in one of the “tools” used for fair lending enforcement, the idea of “REMAs”. A critical concept used by the DOJ and the prudential bank regulators in almost all redlining cases is called a “Reasonably Expected Market Area” or “REMA”. And recent changes to that concept by the regulators have set the stage for potential abuse with misleading redlining analyses predicated on unrealistic markets. 

What is a REMA and how is it critical to the prosecution of most redlining cases?

The term reasonably expected market area cannot be found in the so-called “fair lending” laws nor in any fair lending regulations. It appears to be a concept that originated with bank examiners sometime between 2007 and 2009. The earliest official example I could find was in the 2009 Interagency Fair Lending Examination Procedures where it is identified 6 times and as “Step 1” in the “comparative analysis for redlining”. 

The CRA assessment area can be a convenient unit for redlining analysis because information about it typically already is in hand. However, the CRA assessment area may be too limited. The redlining analysis focuses on the institution’s decisions about how much access to credit to provide to different geographical areas. The areas for which those decisions can best be compared are areas where the institution actually marketed and provided credit and where it could reasonably be expected to have marketed and provided credit. Some of those areas might be beyond or otherwise different from the CRA assessment area.

The 2009 procedures document did not instruct examiners how to identify a REMA and how to distinguish it from a bank’s CRA Assessment Area. However, in more recent years the agencies have published documents that provide guidance to examiners on how to configure a REMA. Those documents indicate that examiners, when configuring a REMA, should consider:

• Where a bank has generated loan applications
• Where a bank has marketed its products
• The bank’s method of attracting business
• Where a bank has originated loans
• Where a bank’s branches and loan production offices are located
• Mortgage subsidiaries
• The bank’s CRA Assessment Area
• Channels by which mortgage applications are generated (mortgage brokers, mortgage realtors, etc.)
• Marketing campaigns

Through 2020 the Agencies broadcast webinars about REMAs and the factors considered by examiners when determining what a REMA is for a bank under examination. 

The question of how to define a REMA is not just another technical question. It is a question of great importance because how a market is defined not only affects a bank’s performance, it dictates the “performance context” which creates the race and ethnicity demographic profile of the community that will be the basis for potential redlining accusations. 

Shortly after the DOJ launched its “Anti-redlining Initiative” the factors that had been the basis for defining a bank’s REMA for fair lending analysis apparently were significantly changed. Representatives from the OCC, FDIC, and FRB during 2022 began to announce that the reasonable factors that had been cited in examiner manuals for the delineation of a REMA were now replaced with areas no smaller than Metropolitan Statistical Areas (“MSAs”) or Metropolitan Divisions (“MDs”). This is a radically different basis for determining a REMA that will have profound consequences for many community banks.

MSAs and MDs can be extremely large areas with thousands of census tracts and millions of people. The Los Angeles MSA consists of more than 3,100 census tracts and the Los Angeles MD contains 2,494 census tracts. A community bank with only 3 or 4 or 5 branches cannot reasonably be expected to serve such large REMAs. But that appears to be the new basis for the approach to REMAs and it will have very profound and very adverse consequences for community banks because it follows that when a market is too big to be adequately served the appearance of redlining is much more likely. And once the inquiry into potential redlining is initiated it can be time-consuming and expensive, even when a bank is exonerated of any wrongdoing.

Bankers and bank associations should express concern to regulators and political leaders about this new aggressive and misleading approach to redlining analysis. Fair lending is a legitimate public policy, but so too should be fair regulating.