daily brief
Residential Construction Spending Edged Down in June; the One-Month Estimate Has a Wide Margin of Error
The Census Bureau's August 3 release estimated private residential construction spending at a $877.1 billion seasonally adjusted annual rate in June, 0.3 percent below the revised May estimate. Census reports a margin of error of plus or minus 1.3 percent for that monthly change, so the release does not establish a durable national housing trend or a change in any household's affordability.
Published Aug 3, 2026, 8:00 PM · Approved by Darnell Dickerson, publication owner (scheduled-publisher instruction)
Confirmed facts
- The U.S. Census Bureau estimated total construction spending during June 2026 at a $2,166.5 billion seasonally adjusted annual rate, 0.1 percent below the revised May estimate of $2,168.5 billion and 3.2 percent below the June 2025 estimate of $2,237.7 billion.
- Census estimated private construction spending at a $1,622.5 billion seasonally adjusted annual rate in June 2026, 0.1 percent below the revised May estimate of $1,624.5 billion.
- Private residential construction spending was estimated at a $877.1 billion seasonally adjusted annual rate in June, 0.3 percent below the revised May estimate of $879.9 billion. Census reports a plus-or-minus 1.3 percent margin of error for that monthly change.
- Census reported that construction spending in the first six months of 2026 totaled $1,046.9 billion, 3.5 percent below the $1,084.5 billion reported for the same period in 2025; the reported margin of error for that comparison is plus or minus 1.0 percent.
- The Census release separately reported private nonresidential construction and public construction, so its total-construction figure is broader than residential construction.
Editorial analysis
- The release supports describing a slightly lower estimated residential construction-spending rate for June, but the stated margin of error is larger than the reported 0.3 percent monthly movement. That makes the one-month change insufficient evidence of a durable direction on its own.
- Construction spending is an estimate of the value of work put in place, not a measure of home-sale prices, rents, mortgage payments, or a household balance sheet. It can provide context about housing-related activity without resolving those separate questions.
What remains unknown
- The release does not determine whether homes will become more or less affordable, or what will happen to mortgage rates, home prices, rents, or residential construction in future months.
- It does not show a particular household's housing payment, financing terms, income, or ability to purchase or rent a home.
- The national estimates do not establish conditions in a specific city, neighborhood, building type, or household-income group.
Evidence
- U.S. Census Bureau, Value of Construction Put in Place, June 2026 (CB26-126)
primary · accessed Aug 3, 2026, 7:58 PM · supports: The release date, June 2026 total, private, private-residential, private-nonresidential, and public construction-spending estimates and their comparisons with revised May estimates.; The stated margins of error for the total, private, private-residential, and first-six-month comparisons.; The first-six-month total and comparison with the same period of 2025.
- U.S. Census Bureau, Value of Private Construction Put in Place, June 2026 data table
primary · accessed Aug 3, 2026, 7:58 PM · supports: The table identifies private residential construction as including improvements and states that residential improvements exclude expenditures to rental, vacant, or seasonal properties.
Update log
- — Initial publication from the Census Bureau's August 3, 2026 Construction Spending release and its linked private-construction table. Confirmed facts, analysis, and unknowns are separately labelled; the one-month estimate is qualified by Census's stated margin of error. Published by the scheduled publisher under the publication owner's current instruction.
Numbers Uncovered provides source-backed editorial context, not investment advice. Reported facts, analysis, and unknowns are intentionally separated.