Final Look at Local Housing Markets in July and a Comment on July Sales from Tom Lawler
First, from housing economist Tom Lawler:
Comments on the NAR’s Existing Home Sales Report for July
In its existing home sales report released last week, the National Association of Realtors estimated that existing home sales in July ran at a seasonally adjusted annual rate of 4.01 million, up 2.0% from June’s estimated pace and up 0.8% from last July’s seasonally adjusted pace. The NAR’s estimate for July was significantly higher than my “early read” estimate, which also happened to be the same as the consensus forecast. CR Note: Tom’s early estimate was 3.92 million SAAR.
Even more surprising, the NAR estimated that the median existing home sales price in July was up just 0.2% from the previous July, and the median existing single-family home sales price was up just 0.3% from a year earlier. I was expecting a YOY gain in the median existing home sales price of 2.1%.
While I normally don’t comment about my “misses,” after looking at a larger sample of realtor/MLS reports than available in my “early read” I decided I should make a few observations.
On median sales prices, the NAR estimated that the median existing home sales price (both total and SF) in the Northeast region last month was up only 0.8% from last July. I have state realtor reports from every state in the Northeast, and the NAR’s estimate is inconsistent with these realtor reports. Below is a table showing YOY increases in median sales prices reported by state realtors in the Northeast.
As this table shows, the median sales price for most Northeast states increased by a lot more than the NAR’s estimate for the region, with one exception being Vermont, where the median SF sales price was down 4.7% YOY. However, Vermont only accounts for about 1.6% of total Northeast sales, so the Vermont results would have only a very modest effect on the region’s overall statistics. The sales-weighted average YOY increase in the Northeast’s median sales price for July was just over 5%.
While I haven’t done the same analysis for other regions, it seems crystal clear that the NAR’s estimate for the YOY increase in median home sales prices for July is too low.
Flipping now to sales: In my “early read” estimate I had assumed that this July’s seasonal factor would be about the same as last July’s. In fact, the NAR’s implied seasonal factor for this July was 1.26% lower than last July. Given business day counts and timing of holidays this surprised me, though I don’t have access to the NAR’s seasonal adjustment program. My seasonal factor “miss” accounted for about half of my total sales estimate miss.
In terms of my unadjusted sales forecast, virtually all of my miss came from the South region, I had predicted a YOY sales decline in the South of about 1.7%, while the NAR reported a YOY increase of 1.17%.
I now have more state realtor reports from the South that I did on August 15th, including state realtor/MLS reports for Alabama, Delaware (actually reports from all three counties), DC, Florida, Georgia, Kentucky, Louisiana, Maryland, North Carolina, South Carolina, Tennessee, Texas, and Virginia. Using these reports, and using Redfin sales reports for Arkansas, Mississippi, Oklahoma, and West Virginia, the implied YOY % change in home sales in the South last month was -1.14%, compared to the NAR’s estimate of a 1.17% increase. If the NAR’s estimate of sales in the South were consistent with state realtor/MLS reports, and if its seasonal factor were accurate, then it would have report existing home sales at a SAAR of 3.97 million instead of 4.01 million.
I am sure most of you will find this report boring. I just thought I’d share because when I miss by a significant amount I like to know why.
CR Note: Actually, Tom’s early estimate didn’t miss by that much!
Final Look at Housing Markets in July and a Look Ahead to August Sales
After the National Association of Realtors® (NAR) releases the monthly existing home sales report, I pick up additional local market data that is reported after the NAR. This is the final look at local markets in July.
There were several key stories for July:
Sales NSA are down YoY through July, and sales last year were the lowest since 1995!
Sales SAAR (seasonally adjusted annual rate) have bounced around 4 million for the last 2 1/2 years.
Months-of-supply is above pre-pandemic levels (this is the highest level for July since 2016).
The median price is barely up YoY, and with the increases in inventory, some regional areas will see more price declines - and we might see national price declines later this year (or in 2026).
Sales at 4.01 million on a Seasonally Adjusted Annual Rate (SAAR) basis were slightly above the consensus estimate.
Sales averaged close to 5.40 million SAAR for the month of July in the 2017-2019 period. So, sales are about 26% below pre-pandemic levels.
Closed Sales in July
In July, sales in these markets were down 0.6% YoY NSA. Last month, in June, these same markets were also up 4.9% YoY Not Seasonally Adjusted (NSA). The NAR reported sales in July were down 0.5% YoY NSA, so this sample is very close.
Important: There were the same number of working days in July 2025 (22) as in July 2024 (22). So, the year-over-year change in the headline SA data was similar to the NSA data.
Notes for all tables:
New additions to table in BOLD.
Northwest (Seattle), Jacksonville Source: Northeast Florida Association of REALTORS®
Totals do not include Atlanta, Denver and Minneapolis (included in state totals)
Comparison to 2019 ONLY includes local markets with available 2019 data!
For next month (August 2025 sales): There were one fewer working days in August 2025 (21) as in August 2024 (22). So, the year-over-year change in the headline SA data will be more than the NSA data (there are other seasonal factors).
August sales will be mostly for contracts signed in June and July, and mortgage rates averaged 6.82% in June and 6.72% in July (somewhat lower than for closed sales in July).
My early expectation is that we will see existing home sales (SA) up slightly in August compared to August 2024 (3.93 million SAAR).
Sales were up 0.8% year-over-year compared to July 2024. This followed 5 consecutive months with sales either down or unchanged year-over-year. The next two months will also have easy year-over-year comparisons.
Total sales last year (2024) were the lowest since 1995 (lower than any year during the housing bust), so it says something that sales are tracking slightly lower in 2025!
Months of Supply
Here is a look at months-of-supply using NSA sales. There are 14.1 months of supply of condos in Miami-Dade!
New Listings in July
For these areas, new listings were up 7.0% year-over-year.
Last month, new listings in these markets were up 5.8% year-over-year.
New listings are now up year-over-year, but down 14% compared to July 2019 activity for the cities that I have 2019 data.
Active Inventory in July
Inventory was up 20.1% year-over-year. Last month inventory in these markets was up 22.6% YoY. Inventory picked up more than usual this year.
Also note the significant regional differences.
More local data coming in September for activity in August!