Figure 2. BancorpSouth Lending Penetration (2011-13) and Branch Distribution (12/31/13), Memphis MSA
The map highlights two critical issues within the Memphis MSA for BancorpSouth.First, relative to other areas, the bank was significantly less penetrated in majority-minority neighborhoods. The bank had an application penetration rate of 3.7 percent in census tracts that were not majority-minority. This figure compares unfavorably to a 1.3 percent application penetration rate in census tracts that were majority-minority.
Second, the geographic distribution of BancorpSouth branches were not representative of the distribution of majority-minority census tracts. While 43 percent of the census tracts within the Memphis MSA were majority-minority, only 22 percent of BancorpSouth’s branches, by the end of 2013, were located in majority-minority census tracts.
Both of these issues – poor penetration and lack of branches in majority-minority census tracts – were identified within the CFPB’s consent order. As the map shows, these issues were particularly acute within the inner city of Memphis, TN, where:
- BancorpSouth was less penetrated than in outlying majority-majority areas; and
- Branches tended to be located just outside the boundaries of clusters of majority-minority neighborhoods.
Using Models to Identify and Prioritize Risk
So far, we have analyzed public data to identify a potentially high risk market for redlining (i.e., the Memphis MSA) and have used mapping to show where that risk may be concentrated. Yet there may be additional factors that should be considered that can explain the observed patterns. We may find that after accounting for these factors – such as competitive intensity, real estate activity, product mix, homeownership levels and building types (e.g., single family homes, mobile homes, etc.) – the overall risk level may be lowered or even eliminated.
Regression models may be developed to control for these potentially explanatory variables. The outcome of this effort may provide a more precise assessment of risk and help prioritize the neighborhoods that offer the greatest realizable opportunity to lower redlining risk. Rather than having to consider all 159 majority-minority census tracts in the Memphis MSA, a well-designed modeling plan may pinpoint for BancorpSouth a small subset of specific areas where it can most effectively and efficiently:
- improve market penetration;
- open new branches; and,
- increase its marketing presence.
A Learning Opportunity for Banks and Mortgage Lenders
The BancorpSouth consent order is just the latest in a series of cases in recent months that show the increased focus regulatory agencies and the DOJ have on redlining issues. While other Fair Lending issues related to underwriting and pricing were identified in the consent order, the exclusion of branches and marketing activities from high-minority concentration areas and the exclusion of these areas from its assessment area under the Community Reinvestment Act were primary arguments in the CFPB’s complaint against BancorpSouth.
For other banks and mortgage lenders, the BancorpSouth case provides a great learning opportunity for incorporating redlining into their compliance management system. The complaint offers a new data point in understanding how the CFPB and other agencies screen lenders for redlining violations. The problematic lending patterns and branch distributions identified for BancorpSouth, Hudson City Savings, Associated Bank, etc. can be leveraged as benchmarks or other elements in an effective compliance monitoring system.
At ADI, we analyze these cases to refine our tools and techniques for identifying and managing Fair Lending risk, and we are reviewing this latest case for new insights that can be implemented in our client solutions. We encourage you to evaluate your compliance management system for any opportunities to identify and address the issues that BancorpSouth and others have encountered, before your examiners do.
About the Author