# June carries 10.4% of a year's purchase mortgages and January carries 5.8% > TC Bulletin computed the monthly distribution of purchase originations across eight years of FHFA's National Mortgage Database. The peak month runs about 1.8 times the trough month, and the curve is stable enough across very different rate environments to plan staffing against. **Section:** The Desk **Published:** August 19, 2026 **Byline:** TC Bulletin Staff **Canonical URL:** https://tcbulletin.com/desk/purchase-origination-seasonality-june-january **Publisher:** TC Bulletin (tcbulletin.com) ## Key points - Averaged across 2015 to 2019 and 2022 to 2024, June accounts for 10.4% of a calendar year's home purchase mortgage originations and January accounts for 5.8%. - The four months from May through August carry roughly 39.5% of the year's purchase originations between them. The four months from November through February carry roughly 26.8%. - The ratio between the peak month and the trough month is about 1.8 to 1, which is the swing a desk staffing to average monthly volume will feel twice a year. - The pattern held through both the low-rate years before the pandemic and the high-rate years after it, which is what makes it usable for planning rather than a description of one cycle. - These are financed purchases only. NAR put cash at 26% of existing-home transactions in July 2026, and cash sales do not appear in this data at all. --- Every coordinator knows the summer is busier than the winter. What is harder to answer, and what actually decides whether to take on another agent or another contractor, is by how much and how reliably. FHFA's National Mortgage Database can answer that, because it publishes monthly counts of purchase mortgage originations going back to 1998. TC Bulletin downloaded the monthly national file and computed each calendar month's share of its own year's purchase originations, averaged across eight years: 2015 through 2019, and 2022 through 2024. The pandemic years 2020 and 2021 were excluded because the refinance boom and the disrupted spring of 2020 distort the shape. ### Share of the year's purchase originations, by month | Measure | Value | Note | | --- | --- | --- | | January | 5.8% | | | February | 6.1% | | | March | 8.3% | | | April | 8.7% | | | May | 9.8% | | | June | 10.4% | Peak month | | July | 9.6% | | | August | 9.7% | | | September | 8.6% | | | October | 8.3% | | | November | 7.3% | | | December | 7.6% | | ## What the shape means for a book of files If a desk closes 240 purchase files a year and the work arrived evenly, that is 20 a month. It does not arrive evenly. On this distribution, June brings about 25 and January brings about 14. A coordinator staffed for the average is overloaded for four months and underused for four months, and the two transitions are where quality problems and missed deadlines tend to appear. The grouping is more useful than any single month. May through August together carry roughly 39.5% of the year, close to two fifths of annual volume inside one third of the calendar. November through February together carry roughly 26.8%. The spring build from February to June is the steepest stretch on the curve, rising from 6.1% to 10.4% in four months. > **Originations are not closings, and the gap is the point.** A mortgage originates at closing, so this curve describes when files completed, not when they opened. A June closing was typically a contract signed in April or May. A coordinator reading this curve for intake planning should shift it back by roughly four to eight weeks, which is where the pending-sales series covered separately by TC Bulletin becomes the better timing tool. ## Why the pattern survives different rate environments The two sets of years averaged here are not alike. The 2015 to 2019 stretch was a low-rate, rising-volume market. The 2022 to 2024 stretch covers the sharpest rate increase in four decades and a steep fall in transaction counts. Total volume moved enormously between them. The month-to-month shape did not move nearly as much, because it is driven by the school calendar, the weather and the moving season rather than by the cost of money. The curve describes when in the year work arrives, roughly independent of how much work there is in total. A coordinator can take the annual figure from wherever they like, their own book, a state association forecast, a national projection, and distribute it across the year with this shape. ## Three limits worth stating 1. This is financed purchases only. NMDB is built from mortgages, so all-cash transactions are invisible in it. NAR put cash at 26% of existing-home sales in July 2026, and cash concentration varies enormously by market and price band. 2. It is national. A coordinator in a market with a short building season or a heavy seasonal-rental economy will have a sharper curve than this, and one in a year-round market will have a flatter one. NMDB publishes state and census-division files that can be run the same way. 3. It excludes refinances entirely. Refinance volume follows interest rates rather than the calendar, which is why it is left out of a seasonal curve and why a desk taking refinance work should treat that stream as separately and less predictably timed. > **How this was computed.** TC Bulletin downloaded FHFA's file nmdb-new-mortgage-statistics-national-census-areas-monthly.csv on 19 August 2026, filtered to GEOLEVEL National, MARKET 'All Mortgages (Home Purchase)' and SERIESID TOT_ORIG, and expressed each month as a percentage of its own calendar year's total before averaging across the eight years named. FHFA's data dictionary states TOT_ORIG is reported in thousands of originations and that each sampled mortgage is weighted to represent twenty actual mortgages. Shares are rounded to one decimal place and sum to 100.2% as a result of that rounding. ### Primary sources for this story - [FHFA, National Mortgage Database aggregate statistics](https://www.fhfa.gov/data/nmdb): Monthly national file, downloaded and computed for this story - [FHFA, NMDB data dictionary and technical notes](https://www.fhfa.gov/document/d/nmdb/nmdb-aggregate-statistics-data-dictionary-technical-notes-2026q1.pdf): Confirms TOT_ORIG units and the five percent sample weighting - [NAR, Existing-Home Sales report for July 2026](https://www.nar.realtor/newsroom/nar-existing-home-sales-report-shows-1-7-decrease-in-july): Cash share of transactions, released 11 August 2026 - [TC Bulletin on housing data lead times](https://tcbulletin.com/desk/housing-data-lead-times-for-coordinators) --- ## Sources cited - FHFA National Mortgage Database: https://www.fhfa.gov/data/nmdb - FHFA NMDB data dictionary: https://www.fhfa.gov/document/d/nmdb/nmdb-aggregate-statistics-data-dictionary-technical-notes-2026q1.pdf Topics: market-volume TC Bulletin publishes independent trade coverage of the transaction coordination profession. Reuse is permitted with attribution and a link to the canonical URL above. See https://tcbulletin.com/syndication.