The Home Mortgage Disclosure Act (HMDA) requires covered institutions to make loan information available to the public. Regulators, examiners, academics, industry leaders, and members of the general public rely on these data to analyze the performance of mortgage lenders in achieving the goals established under the Equal Credit Opportunity Act (ECOA), the Fair Housing Act (FHA) and the Community Reinvestment Act (CRA).

With new reporting requirements on the horizon and increased regulator use of HMDA data in supervisory and examination responsibilities, the risk of HMDA reporting errors will only grow.

Inaccurate HMDA data hampers efforts to identify discriminatory patterns that might exist in home mortgage lending, and to monitor other compliance violations. This makes compliance with HMDA a critical focal point for enforcement, as actions taken by the Consumer Finance Protection Bureau (CFPB) in 2013 demonstrated. This enforcement focus will only grow as the new HMDA reporting requirements under consideration by the CFPB are adopted. These changes will add even more complexity and compliance risk to HMDA submissions.

Poor HMDA data integrity may also signal problems that can have unfavorable implications in a lender’s business intelligence (BI) system. BI systems are relied upon by lenders to inform important business decisions about marketing, operations, finance and other strategic aspects of the business. In many cases, data included in BI systems are sourced from the same loan origination systems that are used for HMDA submissions. The implication is that errors in HMDA reportable data points undermine the data executives use in their decision-making framework. This can lead to the formulation and execution of strategies and tactics based on erroneous data points.

HMDA reporting can be a significant challenge for covered mortgage lenders. The complexity of business systems and the required coordination among different people and systems raises numerous opportunities for reporting errors. We discuss, below, four common errors that lenders encounter during the HMDA reporting process and how you can minimize these errors in your HMDA compliance management system. These four errors are the first set of many other HMDA reporting errors that we will discuss in a series of upcoming articles.

1. Failure to Capture All HMDA-Reportable Transactions

Accurate HMDA reporting requires the identification of all reportable transactions in a timely fashion. For some lenders, the complexity of their business models can lead to missing reportable applications. Properties with multiple purposes or multiple dwellings and applicants with serial investment properties present challenges. Loans related to manufactured home purchases and home improvements as well as loans sold in secondary markets also commonly present scenarios that complicate the assessment of if and when an application should be reported. In addition, the process of managing applications leads that become applications but were not eventually submitted to underwriting may result in the exclusion of reportable applications that should be coded as withdrawn or closed for incompleteness.

2. Incorrect Loan Amount

Loan amount is one of the HMDA variables that many lenders find difficult to report accurately in their HMDA LAR. Errors can be caused by simple data entry mistakes (e.g., typographical errors, transposed numbers, inaccurate rounding, etc.), recording the base loan amount instead of the note amount, and changes in the amount requested during the application process from appraisals, counteroffers, etc. Examiners have identified this as a common reporting error to scrutinize during an audit of HMDA data, and lenders should be particularly mindful of the potential for errors in this field.

This enforcement focus will only grow as the new HMDA reporting requirements under consideration by the CFPB are adopted.

3. Inaccurate Geocoding

The HMDA fields that are the product of geocoding (MSA, State, County and Census Tract) are crucial for any geographic-based analysis of lending patterns, such as CRA and redlining analyses. Based on our experience, assigning the correct Census Tract presents the greatest geocoding challenge to lenders. The geocoding fields are unique among HMDA reporting requirements for many lenders because they depend on third-party software or service that can result in occasional or systemic errors. These errors can raise doubts about an entire HMDA submission.

Whether you rely on a manual process or an automated process that integrates your system with a third party geocoding provider, several factors can lead to reporting errors. These include failing to audit externally sourced geocoding results, using the wrong base geocoding year, and using the wrong address, among others.

4. Incorrect Rate Spread Calculation

Calculating the rate spread for each HMDA-reported loan requires the incorporation of market interest rate data published by the FFIEC. While there are tools that are helpful in automating the calculation and recording of these data for HMDA, the underlying calculation relies on the accuracy of three variables: the rate lock date, the action date and the amortization type.

Errors in recording any of these three variables will yield incorrect calculations. In our experience, capturing the appropriate rate lock date is one field that presents challenges to some lenders due to the possibility of it changing during the application process from rate re-locking requests by the borrower. Loan origination systems, as they are implemented, may prevent lenders from changing the original rate lock date and/or from recording re-lock dates, resulting in miscalculations of the rate spread.

What You Can Do to Mitigate These Errors

Managing these errors requires a solid grasp of HMDA reporting requirements and how data points are identified, entered into your loan origination system, and preserved for eventual reporting. When assisting clients, ADI recommends several steps for assessing and correcting common HMDA errors, including:

  • Review automated systems to ensure they are accurately identifying HMDA-reportable applications and loans;
  • Analyze excluded transactions, such as leads, to ensure they do not meet the definition of an application under HMDA;
  • Establish controls to minimize data entry errors;
  • Review and analyze processes that transfer data between systems to ensure transferred data are accurate and correctly mapped for the HMDA submission;
  • Perform an audit on a sample of applications to determine the accuracy of the recorded data based on source documentation; and
  • Conduct a full HMDA data scrub if error rates are above thresholds established by the CFPB.

With new reporting requirements on the horizon and increased regulator use of HMDA data in supervisory and examination responsibilities, the risk of HMDA reporting errors will only grow. Moreover, HMDA data will increasingly reveal the broad characteristics of markets and customers each reporting lender serves. It is critical, then, that lenders implement strong controls and a culture of compliance with HMDA reporting requirements. This will help lenders overcome current and future compliance challenges and maximize the quality of internal data that can be leveraged to support their marketing and growth strategies.

About the Author: Jonathon Neil

Jonathon is a Senior Consultant for ADI with expertise in Fair Lending compliance, CRA compliance, data mining, and geographic information systems. You can contact Jonathon at or 703.665.3707.