HMDA Reporting: The Complete Guide for Mortgage Lenders

Understanding HMDA Reporting and Its Role in Fair Lending

The Home Mortgage Disclosure Act (HMDA) was enacted in 1975 to promote transparency in mortgage lending. It requires financial institutions to collect and report data about their mortgage lending activity, which regulators and the public use to detect discriminatory lending patterns and ensure equal access to credit. For mortgage lenders, HMDA reporting is not just a regulatory checkbox—it's a critical component of fair lending compliance and a key input for internal risk assessments.

HMDA data is published by the Consumer Financial Protection Bureau (CFPB) and used by regulators, researchers, and advocacy groups to analyze lending trends. When your institution's data reveals disparities, it can prompt closer scrutiny. This makes accuracy in every stage—from data collection to submission—paramount. In this guide, we break down the core HMDA reporting requirements: who must report, what data to collect, how to scrub for errors, and when and how to submit your LAR.

For broader context on federal mortgage compliance, see our Complete Guide to Federal Mortgage Compliance Regulations.

Who Must Report: HMDA Coverage Thresholds and Institutional Exemptions

Not every lender is automatically required to report. HMDA coverage hinges on several specific criteria. Determining your institution's status is the first step in HMDA compliance.

Depository Institutions: Banks, Savings Associations, and Credit Unions

A depository institution must report HMDA data if it meets all of the following:

• Asset-size threshold: The institution has total assets exceeding a certain threshold (adjusted annually by the CFPB). For 2024, that threshold is $56 million.
• Loan volume: The institution originated at least 25 closed-end mortgage loans or at least 200 open-end lines of credit in each of the two preceding calendar years.
• Location: The institution has a home or branch office in a Metropolitan Statistical Area (MSA).
• Regulatory status: It is federally insured or regulated, or the mortgage loans are insured, guaranteed, or supplemented by a federal agency.

Non-Depository Institutions: Independent Mortgage Companies

Non-depository lenders (mortgage companies not affiliated with a bank or credit union) have a slightly different test. They must report if:

• Loan volume: They originated at least 25 closed-end mortgage loans or at least 200 open-end lines of credit in each of the two preceding calendar years.
• Location: The company has a home or branch office in an MSA.
• Exemption: If a non-depository institution originated fewer than 500 open-end lines of credit in both of the last two years, it is exempt from reporting open-end data (as of the 2020 rule).

Partial Exemptions and Special Cases

Small institutions that do not meet loan-volume thresholds may qualify for a partial exemption. For example, they may be exempt from reporting certain data fields. Always verify current thresholds with the CFPB each year, as they are adjusted.

Key point: Even if partially exempt, you may still need to report some data, and fair lending implications remain. If you're unsure about your coverage status, automated compliance tracking tools like Reglith can help monitor threshold changes and flag when you cross a reporting trigger.

The Loan/Application Register (LAR): Core Data Fields Explained

Once you determine your institution must report, you'll need to compile the LAR—a dataset covering every application, origination, and purchased loan. The HMDA LAR contains numerous data fields, but understanding the categories simplifies the process.

Applicant and Borrower Demographics

These fields capture the identity and background of applicants and co-applicants. They are critical for fair lending analysis.

• Ethnicity, race, and sex: Collected based on visual observation or applicant self-identification. Reporting must include subcategories for ethnicities (e.g., Mexican, Puerto Rican) and races (e.g., Asian Indian, Chinese).
• Age: The applicant’s date of birth, used to calculate age.
• Income: The gross annual income relied on in making the credit decision.

Important: Collect this data using the government-mandated categories and separate collection forms to avoid steering or influencing responses. If the applicant declines to provide information, you must still report that the data is “not provided” or “not applicable.”

Loan and Property Characteristics

• Loan type: Conventional, FHA, VA, USDA/RHS.
• Loan purpose: Home purchase, refinance, cash-out refinance, home improvement, or other.
• Lien status: First lien, subordinate lien, or not secured by a dwelling.
• Property type: One-to-four family, manufactured home, multifamily.
• Property location: Census tract, state, county, and MSA.

Underwriting and Pricing Data

• Credit score: The credit score relied on in the decision, and the scoring model used.
• Debt-to-income ratio (DTI): The applicant’s monthly debt divided by income.
• Combined loan-to-value ratio (CLTV): Total loan amounts divided by property value.
• Interest rate: The initial rate for the loan.
• Rate spread: The difference between the APR and a survey-based average prime offer rate (APOR), used to identify higher-priced loans.
• Discount points: Amount paid to reduce the rate.
• Review indicators: Whether the file was reviewed under fair lending or other badging programs.

Automated underwriting systems often populate these fields directly. Ensure that your AI and Automated Underwriting Compliance process includes validation that HMDA data field mappings are accurate and consistent across systems.

Action Taken and Denial Reasons

• Action taken: Originated, approved but not accepted, denied, withdrawn, closed for incompleteness, etc.
• Denial reasons: If denied, you must report up to four primary reasons (e.g., debt-to-income ratio, employment history, collateral, insufficient cash).

Disaggregated Data Fields (Census Tract, Minority Census Tract, etc.)

• Census tract: The census tract where the property is located, critical for geographical analysis of lending patterns.
• Whether the tract is a minority or low-income tract, based on demographic thresholds.

Pro tip: Map every field back to your loan origination system (LOS) to ensure data flows correctly. Missing or mismatched data is a common source of scrub errors.

The Scrub Process: How to Validate Your HMDA Data Before Submission

A rigorous scrub process is the most important step in HMDA compliance. Submitting inaccurate data not only risks regulatory penalties but also undermines the integrity of fair lending analysis.

Step 1: Perform Internal Data Integrity Checks

Start with basic validation:

• Completeness: Verify all required fields are populated for each record. Use logic checks: If action taken is “originated,” then loan amount and rate must be present; if denied, denial reasons must be present.
• Consistency: Cross-check fields. For instance, a loan purpose of “home purchase” should correspond with a property type of one-to-four family (usually). If the lien status is not “first lien,” the loan cannot be a purchase-money mortgage.
• Format validation: Ensure ethnicity, race, and sex codes match the allowable enumerations. Use the CFPB’s HMDA Filing Instructions Guide for exact field definitions and valid values.

Step 2: Audit with Statistical Analysis

Look for outliers and patterns that could indicate systemic errors:

• Run frequency distributions on key fields like loan amount, income, and rate spread. Flag extreme values.
• Compare denial rates across demographic groups to identify unexplained disparities that might signal data collection issues rather than true lending differences.
• Analyze geographic distribution: if loans are concentrated in certain tracts, does it match your footprint?

Step 3: Cross-Reference with Other Systems

HMDA data often originates from multiple systems—LOS, servicing platform, pricing engine, and credit reporting interfaces. Reconcile:

• Loan amount and terms against the final closing documents.
• Applicant demographics against what was collected on the application (do not alter based on other documents unless a data collection error occurred).
• Underwriting data (credit score, DTI, CLTV) against the automated underwriting system results.

Step 4: Perform Targeted Resubmission Adjustments

When you find errors, correct them at the source to prevent recurrence. Document each change with the reason, date, and approver. Never bulk override fields without validating the root cause—this can introduce new errors.

Step 5: Validate Completeness of the LAR Universe

Ensure every covered application or loan is in the LAR. Perform a count reconciliation: compare the number of records in your LAR to the total loans reported in your systems for the same period. Missing records are as problematic as erroneous ones.

Step 6: Subject the LAR to a Third-Party Review

If possible, engage an independent auditor or a compliance peer to review a sample. Fresh eyes catch subtle issues. Automated compliance platforms like Reglith can continuously monitor your data submissions and flag anomalies before the filing deadline.

Remember: HMDA data errors can indicate potential UDAAP violations if they suggest systemic inaccuracies that could harm consumers.

Submission Deadlines and Filing Procedures

Annual Filing Deadline

HMDA data must be submitted by March 1 of the year following the calendar year of the data. For example, data for activity in 2024 must be filed by March 1, 2025. No extensions are granted.

Electronic Submission via the HMDA Platform

All submissions are electronic through the CFPB’s HMDA Platform. You must:

• Register your institution in the platform well in advance.
• Use the platform’s test filing function to validate your file format before official submission.
• Upload a pipe-delimited text file (.txt) that conforms to the HMDA Filing Instructions Guide specifications.

What Happens After Submission?

The CFPB reviews the data and may issue warnings or errors:

• Edits: The platform automatically checks for syntactical, validity, and quality edits. You must address all validity edits before final submission; quality edits require explanation.
• Resubmissions: If errors are discovered after the deadline, you must resubmit corrected data promptly. Late resubmissions are still subject to potential penalties.

Late submissions are a red flag to examiners and can trigger broader compliance examinations. Use a reliable compliance calendar to track the deadline and interim milestones.

Common Pitfalls and Enforcement Consequences

Pitfall 1: Incorrect Coverage Determinations

Failing to report when required can result in regulatory action. Conversely, over-reporting can overwhelm your compliance team. Reassess coverage annually.

Pitfall 2: Inconsistent or Missing Demographic Data

Collecting demographics in a non-compliant manner (e.g., using prohibited categories or failing to provide “not applicable” options) may lead to findings of discrimination under the Equal Credit Opportunity Act (ECOA) or HMDA.

Pitfall 3: Data Entry Errors in Key Fields

Common errors include misreporting loan purpose, lien status, or rate spread. A single-digit mistake in the census tract can skew fair lending analysis.

Pitfall 4: Inadequate Scrub Process

Rushing the validation phase often lets errors slip through. The scrub process should be iterative and involve multiple reviewers.

Penalties and Enforcement

Regulators can impose civil money penalties for HMDA violations. Penalties vary based on the severity and frequency of errors. Beyond fines, HMDA issues can trigger:

• Heightened supervision and more frequent exams.
• Reputational damage from public disclosure of erroneous data.
• Legal liability under fair lending laws if data inaccuracies mask discriminatory patterns.

Proactive compliance is the best defense. Many lenders integrate HMDA data monitoring into their broader compliance management systems and conduct periodic mock audits.

Key Takeaways

• Know your coverage: Review HMDA thresholds annually; asset size, loan volume, and location determine reporting obligations.
• Master the LAR fields: Understand each data point, especially demographics, underwriting details, and action taken, as they directly impact fair lending analysis.
• Implement a robust scrub process: Use multiple validation steps—logic checks, statistical audits, and cross-system reconciliation—to catch errors before submission.
• Respect the deadline: File your LAR electronically by March 1; late filings invite regulatory scrutiny and potential fines.
• View HMDA as more than compliance: Accurate data is a strategic asset for fair lending self-assessment and risk management. Automated tools can help you stay ahead of requirements.
• Stay interconnected: HMDA compliance intersects with TRID disclosures, UDAAP, and automated underwriting. Refer to our related guides for holistic compliance.