August 2026 Multifamily Spending: What Drove the Residential Gain

August 2026 multifamily spending was estimated to rise 0.2% from revised July, compared with a 1.1% increase in private residential spending overall. Improvements accounted for about 91% of the estimated residential gain, according to our calculations from the October 1 Census tables. For contractors serving apartment projects, the residential headline gives a stronger impression than the multifamily estimate supports.

The overall increase is also uncertain. Census reported a 90% confidence margin of ±1.3 percentage points around the 1.1% estimate, encompassing zero. August figures are preliminary, and the release cautions that individual categories can require several months to establish a trend. Census release and explanatory notes

The August 2026 construction spending mix

The October 1, 2026 release measures construction value put in place during August. The comparison below covers private residential construction; public housing construction is outside this breakdown. All dollar amounts are billions at seasonally adjusted annual rates, with July revised and August preliminary.

Source: Census private construction table, October 1, 2026. Improvements equal private residential spending minus new single-family and new multifamily spending; their percentage changes are our calculations from the published rounded levels.

On that basis, improvements supplied $8.718 billion of the $9.563 billion estimated monthly increase in the annual rate. New single-family work added $626 million and multifamily added $219 million. These are changes in annualized spending levels, not additional dollars spent within August.

The annual comparison puts the monthly lift in context: improvements remained furthest below their year-earlier level, while multifamily had the smallest estimated annual decline.

The improvements gain does not measure apartment renovations

Census residential improvements spending covers work on owner-occupied homes. The private table explicitly excludes improvements to rental, vacant and seasonal properties. Consequently, this month's estimated improvements increase does not establish stronger apartment-renovation demand. October table footnote

The estimation process also limits its usefulness as a sales forecast. Census derives improvements estimates from the Consumer Expenditure Survey and uses forecasts until reported spending becomes available, leaving the estimates subject to substantial revisions. Census explanation of improvements estimates, March 12, 2014

July revisions make the estimated gain smaller

The October release raised July's private residential annual rate by $13.705 billion—more than the estimated July-to-August increase.

Sources: September 1, 2026 release, Table 1 and October 1, 2026 release, Table 1. Improvements and revision amounts are calculated.

Comparing August with the original July total would produce an apparent 2.7% increase instead of 1.1%. For multifamily, the same mistake would suggest approximately 0.6% growth instead of 0.2%. Those larger figures combine the estimated monthly change with revisions to the starting point.

For a monthly monitoring sheet, replace July's component estimates together before calculating August's change. Keep the original release separately to track revisions and retain a record of what was known at the time.

Annual rates measure a pace, not a forecast

A seasonally adjusted annual rate takes a monthly estimate adjusted for recurring seasonal patterns and multiplies it by 12. It expresses that month's pace over a full year; it is neither the amount spent during the month nor a forecast of the year's eventual total. This is the standard annual-rate convention described in the BEA's June 2010 statistical appendix.

Value put in place measures construction work installed or erected during a period. It includes materials, labor and other project costs, including contractor profit and architectural services. Work on projects begun earlier contributes to the current month's estimate. Census explanation of value put in place

The release specifies that spending is not adjusted for price changes. A spending increase cannot by itself establish material-cost inflation, more physical construction or additional future starts. Our earlier explanation of value-put-in-place spending develops that distinction.

Match the indicator to your actual exposure

For a construction economics assessment, separate work already supporting revenue from projects expected to replace it. A multifamily subcontractor should compare remaining contracted work with project schedules and prospective awards before using the residential headline to justify staffing or equipment commitments. A lender can make the same distinction between draws on existing projects and demand for new construction loans.

These national estimates also need local context. In a housing data review, use permits, starts, customer bid requests and signed backlog to assess different parts of the pipeline. Check supplier quotes and appropriate price indexes separately before changing cost assumptions.