ABA Letter to CFPB Updating an Earlier Comment on the Proposal to Raise Coverage Thresholds

On June 10, 2019, ABA filed comments on the Consumer Financial Protection Bureau’s (Bureau) proposal to raise the coverage thresholds for collecting and reporting data about closed-end mortgage loans and open-end lines of credit under the Home Mortgage Disclosure Act (HMDA) rules (the NPR). This updated communication is sent to update the data submitted in that letter, to assure the Bureau has the most accurate information on this important rulemaking.

Our June 10th letter indicated our support for an increase in the threshold for reporting data on closed-end mortgages, stating that a higher closed-end coverage threshold would more appropriately balance the burden on lower-volume depository institutions while at the same time maintaining sufficient reporting to achieve HMDA’s purposes. Although thresholds should not be so high as to impair HMDA’s ability to achieve its statutory purposes, we indicated that our research reflected that such thresholds for reporting data about closed-end mortgage loans could be increased considerably. Our comments noted that our independent analysis of the 2017 HMDA data in partnership with Asurity Technologies found that closed-end thresholds could be raised to 200 without causing meaningful loss of data. Asurity Technologies’ comparisons show that raising the thresholds to 100 would lead to a loss of 105,522 transactions (1.43%); raising it to 150 would yield a loss of 184,541 (2.51%); and raising it to 200 would mean a loss of approximately 268,651 transactions (3.65%). Under this count, doubling the threshold size would exclude an additional 163,129 transactions, reducing reported originations by only 3.65%. Given the benefits and additional cost savings, this step was highly worthwhile.

Updated analysis by Asurity Technologies of the 2018 HMDA data demonstrates that the Bureau could actually raise thresholds to higher levels without significantly affecting data quality. In analyzing the 2018 data, the updated comparisons demonstrate that raising the thresholds to 100 would lead to a loss of 105,862 transactions (1.37%); raising it to 200 would yield a loss of 253,927 (3.28%); and raising it to 250 would mean a loss of approximately 328,400 transactions (4.25%). Accordingly, more than doubling the threshold size would exclude an additional 222,538 transactions, reducing reported originations by only 4.25%. In short, closed-end thresholds could be raised to 250 without causing meaningful loss of data. Considering the costs of HMDA compliance, fully discussed in our initial comment to the ANPR on HMDA Data Points and Coverage, we urge the Bureau to revisit these calculations and consider this higher threshold.

Further analysis by Asurity Technologies shows that if the threshold had been set at 250 in HMDA Submission year 2018, then the loss of the data would not adversely impact the data on lending in low-to-moderate Income tracts or majority-minority tracts in any county or MSA. The percentage of loans in such tracts does not change significantly across all counties and MSAs when the data is removed from analysis.

Once again, we appreciate the Bureau’s initiatives to alleviate regulatory burdens on community banks, and we urge the Bureau to consider the higher HMDA thresholds for collecting and reporting data on closed-end mortgage loans. This would be a significant step towards burdens reductions for small institutions.