The Census Bureau reported a 10.5% drop in new home sales
See more of our coverage in your search results.
New single-family home sales ran at an annual rate of 607,000 in July, down 10.5 percent from June, the Census Bureau and HUD reported.
Here is the important part that often gets left out: the 10.5 percent decline is only an estimate, and the margin of error is very large. The real change could have been anywhere from a 24.5 percent drop to a 3.5 percent increase.
Because that range includes zero, the Bureau cannot confidently say sales actually went down. In simple terms, the data is too uncertain to tell whether sales fell, stayed about the same, or even rose a little.
Which numbers actually hold up
We ran the confidence interval on every headline figure in the release. Out of fifteen, three clear the bar.
| Figure | Change | Margin | Verdict |
|---|---|---|---|
| Midwest sales, vs July 2025 | –50.6% | ±10.6 | Significant |
| Midwest sales, vs June | –42.7% | ±13.4 | Significant |
| Homes for sale, vs June | +1.9% | ±1.2 | Significant |
| US sales, vs June | –10.5% | ±14.0 | Not significant |
| US sales, vs July 2025 | –6.3% | ±19.6 | Not significant |
| South sales, vs June | –13.0% | ±17.8 | Not significant |
| Median price, vs June | –2.3% | ±7.4 | Not significant |
| Months' supply, vs June | +12.9% | ±21.3 | Not significant |
The Northeast is the clearest illustration of why this matters. The release shows sales there up 95.5 percent from a year ago, which sounds like a boom. The margin is plus or minus 147.3 percent. The true change could be anywhere from a 52 percent decline to a 243 percent surge. That number carries no information at all.
None of this means the housing market is fine. It means one month of this particular survey can't settle the question. The Bureau says so plainly: it takes four months to establish a trend in new home sales, and preliminary estimates get revised by about 5 percent on average.
The Midwest is the real move
One regional number does survive the test, and it's a big one. Midwest new home sales fell to an annual rate of 43,000 in July, down from 75,000 in June and 87,000 in July 2025. That's a 50.6 percent decline year over year, and the confidence interval doesn't come close to zero.
A drop that size in a single region is worth watching. The Midwest is the smallest of the four regions in this survey and carries the second-highest sampling error, so it swings hard month to month. What makes July different is that the swing is large enough to clear its own margin, which the national figure isn't.
Sales in the South, the largest region by far at 383,000, were down 13 percent from June, but that one falls inside the noise range.
Inventory is the solid signal
The other statistically significant figure is the least dramatic and probably the most useful. There were 488,000 new homes for sale at the end of July, up 1.9 percent from June, with a tight margin of 1.2 percent. Inventory is genuinely rising.
At the current sales pace that's 9.6 months of supply, up from 8.5 in June. Analysts generally treat about six months as a balanced market, so nine-plus months means builders are sitting on more finished product than buyers are absorbing. Of those 488,000 homes, roughly 117,000 are already completed and waiting, and the typical completed home has been on the market for a little over three months.
That backlog fits the broader picture from the Bureau's other housing indicators, where permits pointed to a cooling single-family market and construction spending has been falling.
On price, the median new home sold for $393,800 in July and the average sold for $508,800. Neither change from June is statistically meaningful, but the gap between them is worth noticing: the average sits 29 percent above the median, which happens when a thin band of expensive homes pulls the average upward while the typical sale stays well below it.
The August report comes out September 24. Four of these in a row moving the same direction would be a trend. One is a data point with a wide error bar.
Sources
All figures are from the U.S. Census Bureau and Department of Housing and Urban Development's Monthly New Residential Sales report for July 2026 (Release CB26-128, August 25, 2026), including the regional, inventory, and stage-of-construction tables. Confidence intervals are the Bureau's own 90 percent intervals as published; the significance determinations are our arithmetic applied to those intervals. Sales figures are seasonally adjusted annual rates and are preliminary, subject to revision averaging about 5 percent.
Add CensusEasy as a preferred source on Google to see our stories first.
Which race is decreasing in the United States?
The non-Hispanic white population is the only major group that has shrunk since 2010, down 4.6 million to 192.2 million. It fell in 34 states and 74% of counties. Here is where, and why.
US trade deficit widens to $88.6 billion in July as computer imports hit a record
The July trade deficit is the largest since the March 2025 tariff rush. Capital goods imports set a record at $140.3 billion, computer imports have doubled this year, and Vietnam and Taiwan have both passed China as the largest bilateral deficits.
The US cities where the most adults hold a graduate degree
About 14% of American adults hold a graduate or professional degree. In Bethesda, Maryland it is 58.1%, with Brookline, Palo Alto, Newton, and Cambridge close behind. Every city on the list has a single institutional explanation.
How many new homes sold in July 2026?
New single-family home sales ran at a seasonally adjusted annual rate of 607,000 in July 2026, according to the Census Bureau and HUD. That is 10.5 percent below June's 678,000 and 6.3 percent below July 2025's 648,000, though neither change is statistically significant given the survey's margins of error.
Why isn't the 10.5 percent drop statistically significant?
The margin of error on that figure is plus or minus 14.0 percent, so the true change could range from a 24.5 percent decline to a 3.5 percent increase. Because that range includes zero, the Census Bureau cannot confirm whether sales rose or fell. The Bureau marks such figures with an asterisk and notes it takes four months to establish a trend.
What does a 9.6 month supply of homes mean?
It means that at July's sales pace, it would take 9.6 months to sell every new home currently on the market, up from 8.5 months in June. Analysts generally consider about six months a balanced market, so a figure above nine suggests builders are holding more inventory than buyers are absorbing. Roughly 117,000 of the 488,000 homes for sale are already completed.
