Housing units (PEP)
On this page
Overview#
Housing units is the number of homes an area has, counting houses, apartments and mobile homes whether occupied or vacant, and it also measures how fast that number is growing. Housing-unit growth is the supply side of the market. Set against population or household growth, it shows whether building is keeping pace with demand.
Definition#
The number of housing units in an area as of July 1 each year, from the Census Bureau's housing-unit estimates. A housing unit is "a house, an apartment, a group of rooms, or a single room intended for occupancy as separate living quarters" (Building Permits Survey definitions). Vacant units are included. The estimates start from the census count and add new construction from building permits, less estimated losses.
Which version should I use?#
- PEP 2025 answers: how many housing units are there now, and how has that changed in recent years? It supplies the latest figure and the end of every change, so it is the version you read.
- PEP 2020 answers: how many housing units were there before 2020? It supplies the starting count for the 10-yr change. It is not a separate metric to choose.
How to use it#
Housing-unit growth tells you how fast a place is adding homes. Compare it with nearby places, the metro and the national figure.
High and low values
- High growth: often a fast-growing suburb or a metro with land and room to build. New supply can ease prices and rents, or it can be a sign that builders expect strong demand.
- Low growth: often a built-out city or a place with little demand. Where people are still moving in, slow building can push prices and rents up.
Scenarios
- Fast population growth: if units grow more slowly than population, homes are likely getting scarcer and costlier.
- A building boom: units growing well ahead of population can point to more vacant homes and softer rents ahead.
- Job losses: places that kept building through a downturn can be left with more homes than buyers or renters.
Use cases
- Finding where a metro is expanding: the counties adding homes fastest often mark where new growth is going.
- Screening for oversupply: finding places where the stock is growing much faster than nearby places.
- Sizing a market: the count gives the scale of a county's or metro's housing stock.
PEP 2025 version#
Census Population Estimates, 2025 release, published by the U.S. Census Bureau.
Metrics
Housing units (PEP)
The latest annual estimate, as of July 1. It is shown for counties and metros, with state and U.S. figures; tracts and ZIP codes do not carry the count. Choosing this metric on its own switches the map to counties.
Housing units (PEP), change
1-yr, 5-yr and 10-yr change: the percent change in housing units from 1, 5 or 10 years earlier to the latest July 1 estimate. For example, a rise from 100,000 to 103,000 units is +3%. Years from 2020 on come from this source; the 10-yr change starts in an earlier year, which comes from the PEP 2020 version below. Tracts and ZIP codes show their county's figure on the place card, in Index factors and in exports, marked as an estimate. Choosing this metric on its own switches the map to counties.
Considerations
- It counts homes, not households. Vacant units are included, so growth in units can run ahead of growth in the people living in them.
- Metro, state and U.S. figures are the sum of their counties. A metro uses its current member counties in every year.
- Units built or torn down without a permit are missed. The estimates are built from permits, so places with much unpermitted building or demolition can be off.
For limitations that affect all PEP Vintage 2025 metrics, see PEP Vintage 2025 considerations.
For full details and limitations, see the U.S. Census Bureau’s Population Estimates Program.
PEP 2020 version#
Census Population Estimates, 2020 release, published by the U.S. Census Bureau.
Metrics
Housing units (PEP), change
trendThe same 10-yr change described above. This source supplies the starting count, because the 10-yr window starts before 2020.
Considerations
- The 10-yr change crosses a change of census base. Its starting count comes from estimates built on the 2010 census, and its end from estimates built on the 2020 census.
For limitations that affect all PEP Vintage 2020 metrics, see PEP Vintage 2020 considerations.
For full details and limitations, see the U.S. Census Bureau’s Population Estimates Program.
References#
- 1Population and Housing Unit Estimateshttps://www.census.gov/programs-surveys/popest.html · US Census Bureau
- 2Building Permits Survey Definitionshttps://www.census.gov/construction/bps/definitions.html · US Census Bureau
Birdseye summarizes data from public and commercial sources. Each source’s publisher is the authority on its data and methods. See our Terms of Service.