Loan application volume
The number of home-purchase mortgage applications filed on properties in a tract, and its year-over-year change.
On this page
Not shown in the tool. The data behind this family is held and kept current, but it is not one of the measures the tool offers.
The number of home-purchase mortgage applications filed on properties in a tract.
An application enters the Home Mortgage Disclosure Act register once a lender acts on it, whatever the outcome. Originations, denials, withdrawals by the applicant, files closed for incompleteness, approvals the applicant did not accept, and preapproval decisions are all recorded actions (FFIEC LAR data fields). Applications counted here are those the register classifies as a home purchase, a category Regulation C keeps separate from home improvement, refinancing and cash-out refinancing (12 CFR 1003.4).
Applications measure attempted demand rather than completed sales. Someone who applies and is turned down still wanted to buy, and that intent is invisible in transaction records. Application counts pick it up.
Metrics in this family#
Purchase application volume
The annual number of home-purchase applications on properties in the tract, averaged across the three most recent published activity years.
This is a count, not a rate. A large or densely built tract will show more applications than a small one at the same intensity of demand, so it compares places of similar size more meaningfully than places of different size.
Purchase application growth
The change in that count from the previous year, expressed as a percentage.
A one-year change in a count is volatile at tract level, and a tract with few applications can post a large percentage swing on a handful of them.
Considerations#
The averaged count requires records in each of the three years of the window; tracts without a full three years are left without a value rather than given a partial one.
Loans a lender bought from another institution are not applications and are excluded, so the count is not inflated by loans changing hands on the secondary market. One property can generate more than one application if a borrower applies to several lenders, and the register carries no way to identify duplicates, so the count is of applications rather than of would-be buyers.
Cash purchases never generate an application and are absent entirely. In markets with heavy cash buying, application volume understates activity.
Coverage follows institutions above the Act's reporting thresholds rather than every lender, so areas served mainly by very small institutions show lower counts than their activity warrants. Because the reporting thresholds themselves changed after a September 2022 court order, the set of reporting institutions is not identical across all years. The published file is also revised as institutions resubmit. See the source page.
References#
- 1FFIEC LAR data fieldshttps://ffiec.cfpb.gov/documentation/publications/loan-level-datasets/lar-data-fields
- 212 CFR 1003.4https://www.consumerfinance.gov/rules-policy/regulations/1003/4/