Around 47% of the U.S. housing stock was built in the 1980s and earlier. The median age of owner-occupied homes climbed to 42 years old in 2024, up from 31 in 2005 according to the latest data from the American Community Survey[1]. The U.S. owner-occupied housing stock has aged measurably, particularly since the Great Recession, as residential construction has consistently fallen short of demand, contributing to a persistent housing shortage.
Currently, new home construction faces headwinds such as rising material costs, a persistent labor shortage, and elevated interest rates. These challenges have contributed to an insufficient supply of new construction, making the nation’s owner-occupied housing stock significantly older over time. As a result, the aging housing stock signals a future growing remodeling market. Older structures require reinvestment to add new amenities or need repairs or replacements of aging components.
From 2020 to 2024, new construction added nearly 3.6 million owner-occupied homes, accounting for only 4% of total owner-occupied housing stock as of 2024. Relatively newer homes built between 2010 and 2019 made up around 9% of the stock, while those constructed between 2000 and 2009 constituted 15%. In contrast, around 47% of the owner-occupied homes were built before 1980, including around 34% built before 1970.
Due to a modest supply of housing construction, the share of relatively newer owner-occupied homes (those built within the past 14 years) has declined greatly, from 18% in 2014 to only 13% in 2024. Meanwhile, the share of older homes that are at least 45 years old has increased significantly, rising from 39% in 2014 to 47% in 2024. This shift further reflects the ongoing aging of the U.S. housing stock, highlighting the growing importance of the remodeling sector to address the growing needs of homeowners nationwide.
[1] : Census Bureau did not release the standard 2020 1-year American Community Survey (ACS) due to the data collection disruptions experienced during the COVID-19 pandemic. The data quality issues for some topics remain in the experimental estimates of the 2020 data. To be cautious, the 2020 experimental data is not included in the analysis.