housing_supply_county_year
County-year grain. Total housing units, units per 1k population, units by age of structure. Stub — pending acs_county_supply_2024 raw load (ACS B25001 + B25034).
tier —admin_record
overview
overview
All 3,222 counties (planned). STUB - pending acs_county_supply_2024 load.
current vintage — 2020-2024 ACS 5-Year (planned)
history — ACS B25001: 2009-present
provenance
source & licensing
authority
U.S. Census Bureau
dataset
Building Permits Survey; ACS 5-Year Estimates
source url
license
—
citation
U.S. Census Bureau. ACS 5-Year Estimates, Tables B25001 (total housing units), B25034 (year built), 2020-2024.
schema
fields
| name | type | definition |
|---|---|---|
| _ingested_at | timestamp | Timestamp when this record was written to BigQuery.Pipeline metadata field. |
| _ingestor_version | string | Version of the ingestor script that produced this record.Pipeline metadata field. |
| _reliability_tier | string | Data quality tier (A/B/C) assigned at ingestion.Pipeline metadata field. |
| _source_dataset_id | string | Internal dataset identifier linking to meta.datasets.Pipeline metadata field. |
| country_id | string | ISO alpha-2 country code (always 'US' for domestic tables). |
| county_idkey | string | 5-character FIPS code identifying the county.Joins dim.counties on county_id. |
| county_name | string | Human-readable county name corresponding to county_id. |
| housing_units_per_1k_pop | float64 · rate per 1,000 population | Number of housing units per 1,000 residents in the county. Higher values indicate greater housing availability relative to population size. |
| pct_units_pre_1940 | float64 · percent (0–100) | Share of total housing units constructed before 1940, expressed as a percentage. Higher values indicate an older housing stock, which is associated with greater maintenance burden and potential lead paint or structural concerns. |
| state_id | string | 2-character FIPS code identifying the state.Joins dim.states on state_id. |
| total_housing_units | int64 · count | Total number of housing units in the county, including occupied and vacant units. Higher values indicate a larger overall housing stock. |
| units_built_1940_1969 | int64 · count | Number of housing units constructed between 1940 and 1969 inclusive. Represents mid-century housing stock that may require significant maintenance or modernization. |
| units_built_1970_1999 | int64 · count | Number of housing units constructed between 1970 and 1999 inclusive. Captures mid-to-late 20th century housing stock, which may reflect suburban expansion and post-war development patterns. |
| units_built_2000_plus | int64 · count | Number of housing units with a year of construction of 2000 or later. Higher values indicate a larger share of newer, more recently built housing stock. |
| units_built_before_1940 | int64 · count | Number of housing units constructed prior to 1940. Higher values indicate a larger share of older housing stock, which may signal deferred maintenance needs or historic preservation contexts. |
| yearkey | int64 | Reference year of the observation. |
relationships
joins
primary key
county_id, year
common joins
dim.geographies on county_id
housing_vacancy_county_year on (county_id, year)
housing_tenure_county_year on (county_id, year)
usage
how to use this table
method
ACS B25001 total housing units; B25034 by structure age. Per-1k-pop = total / population x 1000. Currently STUB.
do not use for
New construction (use Census Building Permits Survey); commercial properties.
known issues
STUB - not yet loaded.
last updated · May 5, 2026