Data Dig : New Single-Family Home Sales and Starts Through July 2026

We last wrote that the supposed shortage of homes in the United States is a false narrative that masks the more accurate description of the housing affordability challenge as a mismatch between buyer expectations and the financial realities of new home construction. We examined the profitability of the largest new home builders and found profit margins with little remove for decline. Given the fixed and consistently rising cost of new home creation inputs (e.g., land, development costs, governmental fees, construction materials, and labor), we concluded that homebuilders whether for sale or rental could not rationally (read profitably) increase production and, therefore, housing supply and sell at lower prices. We also suggested that housing market data would confirm that supply will dwindle, not rise, if builder profitability remains depressed.

Sales, Prices, And Starts Declined, And Inventory Rose

After a strong start in the first quarter of 2026, the new single-family home data has deteriorated for buyers and builders. Continuing the decline that began earnestly in May, July 2026 brought no good news to the new single-family home market.

• The Census Bureau reported a seasonally-adjusted annual rate of 607,000 new single-family home sales in July 2026, 10.5% below the June 2026 rate of 678,000 and 6.3% below the July 2025 rate of 648,000.

• Reflecting weak sales in the lower price range where affordability challenges are concentrated, the median sales price of new houses dropped in July 2026 to $393,800 from $403,100 in June 2026 and $397,300 in July 2025. In contrast, homes in the higher price ranges sold more strongly in July 2026 with an average sales price of $508,800, 4.1% above June 2026 ($488,900) and 5.4% above July 2025 ($482,800).

  • Declining home prices in the lower range of the market certainly aid affordability but not by much. At the prevailing 6.91% 30-year mortgage rate, the monthly payment on a $403,100 (June 2026) home would be $2,657.31. The payment on July’s median home price would be $61.30 less.
  • The effect of an interest rate change is far greater, of course. At a mortgage interest rate of 6.0%, which was available just before the Iran [choose your descriptive noun], those two payments would be $2,416.79 and $2,361.03, roughly $235 lower.
  • At the 6.91% interest rate, the same monthly payment savings would require a home price reduction of approximately $36,000, a 9% or so price reduction from the June 2026 median price. Given that the largest new home builder net margins now range from 6.0% to 15.9%, a 9% price reduction would be financially devastating and certainly not sustainable for new home builders.

You likely have guessed already the impact of the sales deceleration on inventory.

• The seasonally-adjusted new-houses-for-sale inventory at the end of July 2026 was 488,000, slightly above the June 2026 estimate of 479,000 and a touch below the July 2025 estimate of 496,000. Because the number of sold homes dropped so much in July, the inventory count of 488,000 equates to 9.6 months of supply, which is 12.9% above June 2026’s 8.5 months and 4.3% above July 2025’s 9.2 months.

You will not have made any money in the prediction markets by guessing correctly that builders responded intelligently to the continuing sales decline by slowing new home construction starts by a huge amount.

• Builders started construction on a seasonally adjusted 808,000 homes, compared to 897,000 in June (an 11% decrease) and 959,000 in July 2025 (a drop of 18.7%).

If the United States has a housing shortage, new single-family home construction is not on a path to solving the problem nor will the direction of the path change anytime soon. New home builders are economically and professionally rational and respond quickly to data that forecasts lower revenue and smaller profit margin. Fewer sales and more standing inventory compels fewer starts.

For so long as buyers are unable or unwilling to buy (i.e., for so long as demand is weak), builders will not build (i.e., supply will shrink). Weak demand, falling prices, and rising months-of-inventory tell builders that supply is unprofitably abundant, not profitably scarce. Given the general view that a six-month supply of new homes signifies a market in balance, we can expect builders to use sales incentives to clear inventory and apply the brakes to their new construction for at least several months but likely into 2027.

The weak demand so clearly evident in the new home sales data also tells us that a supply shortage is not the bane of the housing market. Housing market advocates argue that increasing supply will drive down home prices. Our last post explained that the largely fixed and consistently rising costs of construction leave home builders little room to reduce prices of the homes that the builders think buyers want. Our simple calculations above showed that addressing affordability with lower home prices would erase new home builder profit near or below zero.

Fine Shovel To Keys, Supply Is Not The Problem, But What Is The Problem?

In upcoming posts, we will explore possible alternative causes and possibly even solutions to the housing affordability problem. Among the candidates on the agenda, which likely will grow and morph:

• Structural changes in land development and home construction financing as a result of the Great Financial Crisis of 2008;
• Post-GFC rising land, development, and construction material and labor costs owing to scarcity, inflation, tariffs, immigration decline, and probably other factors;
• Home buyer and tenant expectations for the features, size, and amenities in housing;
• Homeowner reliance on home equity as a substantial source of retirement wealth;
• Homeowner memory of the “high” price at which a neighbor sold a comparable house;
• The powerful attraction of a very low interest rate mortgage;
• Higher infrastructure costs, governmental fees, and code compliance costs;
• The unusually long low-interest-rate environment from 2008 to 2022.

Comment if you have suggestions for possible causes not listed above or a preference for which possible causes we next address in a Shovel To Keys post.