Choose a table or county below. View data opens the actual table on a place profile, with its dates, units and source notes.
Open a county or state example to see the values. Table coverage counts supported profiles, not complete rows or every area. Original publisher links provide background documentation; they are separate from the data displayed here.
Fair Market Rents (HUD)
U.S. Department of Housing and Urban Development · Fair Market Rents, fiscal year 2026
- Data period
- Fiscal year 2026 (October 1, 2025 to September 30, 2026)
- Geography vintage
- HUD Fair Market Rent areas for fiscal year 2026
- Units and scope
- US dollars per month, gross rent including utilities.
- Pages with this table
- 3,154
Coverage and interpretation
- A Fair Market Rent is a HUD administrative benchmark for housing assistance (the 40th percentile of gross rents for standard units), not the median or observed rent in the county.
- HUD sets rents for its own rent areas, which can be a metro area or a part of one; the area shown is named in the column heading.
- A rent the source did not publish is shown as Not reported, never as zero.
- A county without a table is one for which HUD publishes no Fair Market Rent area at the county code.
Source periods, coverage and other tables · Original publisher: documentation and methods
HUD Fair Market Rents and Income Limits (FY2026, county)
U.S. Department of Housing and Urban Development · Fair Market Rents and Income Limits, fiscal year 2026, whole-county records
- Data period
- HUD fiscal year 2026 (October 1, 2025 to September 30, 2026)
- Geography vintage
- Counties and county equivalents as coded in HUD entity codes (five-digit county FIPS plus 99999), matched on that code; HUD publishes New England by town, so those counties have no record
- Units and scope
- Fair Market Rents: US dollars per month. Income limits and median family income: US dollars per year.
- Pages with this table
- 3,154
Coverage and interpretation
- HUD FY2026 administrative values, not market rents or measured incomes. A Fair Market Rent is HUD's 40th percentile gross rent estimate (rent plus utilities) for standard-quality rental units in the HUD rent area, used to set payment standards and rent ceilings for HUD programs; it is not the median or typical rent paid in the area.
- An income limit is the household income HUD uses to decide eligibility for assisted housing programs, set from the area median family income: extremely low income (HUD field il30), very low income (il50) and low income (il80), shown here for a four-person household. The median family income is HUD’s estimate for the income-limit area, not a survey of residents. HUD adjusts limits for household size, local housing costs and its extremely low income floors, so a limit is not an exact percentage of the median shown.
- HUD sets Fair Market Rents and income limits for its own areas, which can be a metro area, part of one or a nonmetropolitan county; the areas are named in each table’s notes and the same county can sit in different rent and income areas.
- A value HUD did not publish, or a county with no HUD record, is shown as Not reported, never as zero. Values are as HUD published them at retrieval; HUD can revise values during the fiscal year and some areas stay on earlier values after a reevaluation request.
- These are county figures: the county code is matched to the HUD record for the whole county. HUD publishes Fair Market Rents and income limits by town in Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island and Vermont, and HUD does not publish the Connecticut planning regions; those counties and regions carry no table here and are not estimated from towns.
Source periods, coverage and other tables · Original publisher: documentation and methods
Housing units authorized by building permits (Census Building Permits Survey)
U.S. Census Bureau · Building Permits Survey, county annual files
- Data period
- Calendar years 2025 and 2024
- Geography vintage
- County definitions in the permit files (Connecticut planning regions, as on the page); a county the permit files do not list has no entry
- Units and scope
- Housing units authorized, by structure size; valuation in dollars.
- Pages with this table
- 3,024
Coverage and interpretation
- Permits authorized are not units built or completed.
- Values are Census Bureau estimates that include imputation for permit offices that did not report every month. The reported-only row shows the part of the total that rests on permits actually reported.
- Valuation is the permit applicant’s stated construction cost in dollars, not market value. The state file reports thousands of dollars and is scaled by 1,000; the county file reports whole dollars.
Source periods, coverage and other tables · Original publisher: documentation and methods
Building permits authorized by structure size, 2021 to 2025: buildings, units and valuation (Census Building Permits Survey, county)
U.S. Census Bureau · Building Permits Survey, county annual files, 2021 to 2025
- Data period
- Calendar years 2021 to 2025
- Geography vintage
- Counties and county equivalents by the state and county FIPS codes in each annual file, matched by code to this site’s county list; Connecticut uses planning regions from 2023 and former counties before
- Units and scope
- Buildings and housing units authorized: counts. Valuation: dollars of stated construction cost.
- Pages with this table
- 3,036
Coverage and interpretation
- Building permits authorized are not buildings or units built or completed.
- Values are Census Bureau estimates that include imputation for permit offices that did not report; the source also publishes a reported-only set, and the “Of which units in permits actually reported (no imputation)” row follows that set. Imputed and reported flags follow the source and are not changed here.
- Valuation is the permit applicant’s stated construction cost in dollars, not market value. The county files report whole dollars.
- A year in which the area has no row in the source file is Not reported, never zero. Totals across structure sizes are sums computed by LocalFactBook from the four published sizes.
- Connecticut: the 2021 and 2022 county files use the former eight counties and the 2023 to 2025 files use the nine planning regions. Planning regions are Not reported for 2021 and 2022, and no former-county row is shown on a planning-region page.
Source periods, coverage and other tables · Original publisher: documentation and methods
Building permits authorized by structure size, 2021 to 2025: buildings, units and valuation (Census Building Permits Survey, state)
U.S. Census Bureau · Building Permits Survey, state annual files, 2021 to 2025
- Data period
- Calendar years 2021 to 2025
- Geography vintage
- States and the District of Columbia by the FIPS code in each annual file; Puerto Rico by code 72
- Units and scope
- Buildings and housing units authorized: counts. Valuation: dollars of stated construction cost.
- Pages with this table
- 52
Coverage and interpretation
- Building permits authorized are not buildings or units built or completed.
- Values are Census Bureau estimates that include imputation for permit offices that did not report; the source also publishes a reported-only set, and the “Of which units in permits actually reported (no imputation)” row follows that set. Imputed and reported flags follow the source and are not changed here.
- Valuation is the permit applicant’s stated construction cost in dollars, not market value. The state files report thousands of dollars; the values shown are the source figure multiplied by 1,000.
- A year in which the area has no row in the source file is Not reported, never zero. Totals across structure sizes are sums computed by LocalFactBook from the four published sizes.
- Puerto Rico is listed under code 43 in the 2021 state file and 72 afterwards; matched by code only, Puerto Rico is Not reported for 2021. The 2022 Puerto Rico row carries valuation in whole dollars where the rest of the file is in thousands, so Puerto Rico’s 2022 valuation is Not reported; its building and unit counts are shown as published.
Source periods, coverage and other tables · Original publisher: documentation and methods
FHFA conforming loan limits (2026, county)
Federal Housing Finance Agency · Conforming loan limits, calendar year 2026, all counties
- Data period
- Loans acquired in calendar year 2026
- Geography vintage
- Counties and county equivalents as coded in the FHFA list (state and county FIPS), matched on the five-digit code; Connecticut is listed by planning region and matched only where the project roster carries the same code
- Units and scope
- US dollars, maximum original principal balance of a mortgage.
- Pages with this table
- 3,222
Coverage and interpretation
- A conforming loan limit is the largest original loan balance Fannie Mae and Freddie Mac may purchase; it is a legal ceiling for loans acquired in 2026, not a home price, a typical loan or a measure of lending. Limits are shown for one- to four-unit properties as FHFA publishes them.
- The high-cost marker is derived from the published limits (one-unit limit above the $832,750 national baseline in FHFA's 2026 addendum); FHFA's file has no separate designation column. It includes areas with statutorily higher limits (Alaska, Hawaii, Guam, the U.S. Virgin Islands).
- A county with no FHFA row is shown no table, never zero. Limits are set per county or county equivalent, can differ across counties inside one metropolitan area only where FHFA lists them separately, and are not the FHFA House Price Index shown elsewhere.
Source periods, coverage and other tables · Original publisher: documentation and methods
Home-purchase mortgage originations, 2025 (HMDA, county)
Consumer Financial Protection Bureau and the Federal Financial Institutions Examination Council · Home Mortgage Disclosure Act data, activity year 2025, Data Browser, originated home-purchase loans
- Data period
- Activity year 2025 (loans originated in calendar year 2025), Data Browser snapshot files retrieved 2026-11-04
- Geography vintage
- Counties and county equivalents as coded in the HMDA file (Connecticut planning regions 09110 to 09190), matched on the five-digit FIPS code
- Units and scope
- Loans: number of loans. Total and median loan amount: current dollars, amounts rounded to the midpoint of a $10,000 interval.
- Pages with this table
- 3,222
Coverage and interpretation
- Counts and dollar amounts of originated home-purchase loans reported under HMDA, not all mortgage lending: only institutions covered by the Home Mortgage Disclosure Act report, so loans by institutions below the reporting thresholds and loans made outside the definition (refinancings, home-improvement loans, second liens, manufactured homes, second homes, investment properties, buildings of five or more units) are not counted.
- The definition is loan originated (action taken 1), home purchase (loan purpose 1), first lien (lien status 1), site-built (construction method 1), principal residence (occupancy type 1), one to four units (total units 1, 2, 3 or 4). The Data Browser accepts only two filters at once; the first two were applied by the source and the other four by this site from the same files.
- The county is the county of the property. Loans are counted in the activity year in which they were originated, as filed by lenders; lenders can resubmit, so later releases of the same year can differ.
- Loan amounts in the public file are rounded to the midpoint of a $10,000 interval. The total loan amount is the sum of those rounded amounts and the median is the median of them, in current dollars for the year; they are not home prices and not adjusted for inflation.
- A county with fewer than 5 loans in the definition is Not reported for all three figures. Not reported is not zero. A reader cannot tell none from one to four.
- Totals are not divided by population or by households, and places are not compared. Denial or approval figures, applicant characteristics and lender names in the source are not shown.
- County and county-equivalent areas use the five-digit codes in the HMDA file and are matched to this site’s county list by code. Connecticut’s file uses the nine planning regions (09110 to 09190), the same codes as this site’s roster. A few loans carry no county code in the public file; they are in no county figure. These are county figures, not figures for any city or place within the county.
Source periods, coverage and other tables · Original publisher: documentation and methods
Home-purchase mortgage originations, 2025 (HMDA, state)
Consumer Financial Protection Bureau and the Federal Financial Institutions Examination Council · Home Mortgage Disclosure Act data, activity year 2025, Data Browser, originated home-purchase loans
- Data period
- Activity year 2025 (loans originated in calendar year 2025), Data Browser snapshot files retrieved 2026-11-04
- Geography vintage
- States, the District of Columbia and Puerto Rico, matched on the two-digit FIPS code
- Units and scope
- Loans: number of loans. Total and median loan amount: current dollars, amounts rounded to the midpoint of a $10,000 interval.
- Pages with this table
- 36
Coverage and interpretation
- Counts and dollar amounts of originated home-purchase loans reported under HMDA, not all mortgage lending: only institutions covered by the Home Mortgage Disclosure Act report, so loans by institutions below the reporting thresholds and loans made outside the definition (refinancings, home-improvement loans, second liens, manufactured homes, second homes, investment properties, buildings of five or more units) are not counted.
- The definition is loan originated (action taken 1), home purchase (loan purpose 1), first lien (lien status 1), site-built (construction method 1), principal residence (occupancy type 1), one to four units (total units 1, 2, 3 or 4). The Data Browser accepts only two filters at once; the first two were applied by the source and the other four by this site from the same files.
- The county is the county of the property. Loans are counted in the activity year in which they were originated, as filed by lenders; lenders can resubmit, so later releases of the same year can differ.
- Loan amounts in the public file are rounded to the midpoint of a $10,000 interval. The total loan amount is the sum of those rounded amounts and the median is the median of them, in current dollars for the year; they are not home prices and not adjusted for inflation.
- A county with fewer than 5 loans in the definition is Not reported for all three figures. Not reported is not zero. A reader cannot tell none from one to four.
- Totals are not divided by population or by households, and places are not compared. Denial or approval figures, applicant characteristics and lender names in the source are not shown.
- A state figure is shown only when every county of the state has at least 5 loans in the definition, so that no county figure that is Not reported can be recovered by subtraction. It counts every loan in the state in the definition, including the few that carry no county code, and is tallied from the state file, not summed from the counties shown.
Source periods, coverage and other tables · Original publisher: documentation and methods
Homeless-assistance beds and units by project type (HUD Housing Inventory Count)
U.S. Department of Housing and Urban Development (Office of Community Planning and Development) · Housing Inventory Count (HIC) 2024, project-level data by Continuum of Care state
- Data period
- 2024 Housing Inventory Count, a one-night inventory each Continuum took in late January 2024 (a few in February)
- Geography vintage
- State of the reporting Continuum of Care (CocState) as published, joined to the site’s state list by USPS abbreviation
- Units and scope
- Number of year-round beds and of units for households with children, by project type.
- Pages with this table
- 52
Coverage and interpretation
- Beds are the file’s Year-Round Beds column, which equals the sum of its three household-type bed columns (households with children, without children, and with only children); the file’s seasonal, overflow and voucher beds are not included. Units are the file’s only unit column, Units HH w/ Children, so they count units for households with children only; no unit count is published for other households.
- Only current inventory (Inventory Type C) is counted. Beds under development (Inventory Type U) are excluded. A state is the state of the Continuum of Care that reported the project (CocState), which can differ from the project address; Continuums that cover parts of several states are reported under one state.
- Counts are a one-night January 2024 inventory of beds, not of people served, occupied beds or need. Totals include projects run by victim service providers, aggregated as the source does. A state with no project of a type shows 0 because the file is a roster of projects. Organization and project names and addresses are in the source and are not shown.
Source periods, coverage and other tables · Original publisher: documentation and methods
Homeownership and vacancy rates (Census Housing Vacancy Survey)
U.S. Department of Commerce, Census Bureau · Current Population Survey / Housing Vacancy Survey, annual statistics, state tables
- Data period
- Calendar years 2021 to 2025 (annual averages; 2025 omits October)
- Geography vintage
- States and the District of Columbia
- Units and scope
- Percent. Homeownership rate is the share of occupied housing units that are owner-occupied; rental vacancy rate is the share of the rental inventory that is vacant for rent; homeowner vacancy rate is the share of the homeowner inventory that is vacant for sale.
- Pages with this table
- 51
Coverage and interpretation
- These rates come from the Current Population Survey, a sample survey, and carry sampling error that is larger for small states; small differences between states or years may not be meaningful. The annual tables do not publish margins of error.
- Because of a lapse in federal funding the survey did not collect data for October 2025, so the 2025 annual averages are based on the remaining 11 months.
- These survey rates differ from American Community Survey tenure and vacancy figures, which use a different survey and universe. A value that is not a published number is shown as Not reported, never as zero.
Source periods, coverage and other tables · Original publisher: documentation and methods
House Price Index (FHFA)
Federal Housing Finance Agency · FHFA House Price Index master file, state series (traditional, all-transactions and purchase-only)
- Data period
- Latest quarter and earlier quarters of the quarterly series
- Geography vintage
- States and the District of Columbia
- Units and scope
- Index; the base period differs between the all-transactions and purchase-only series.
- Pages with this table
- 51
Coverage and interpretation
- The House Price Index measures price changes of repeat sales and refinancings; it is an index, not a price and not an average home value. The two indexes have different base periods, so compare changes, not index levels.
- The all-transactions index uses sales and appraisals; the purchase-only index uses only purchase prices. They differ and are shown separately.
- Changes are computed by LocalFactBook from the published index levels and are not published by FHFA.
Source periods, coverage and other tables · Original publisher: documentation and methods
House Price Index (FHFA)
Federal Housing Finance Agency · FHFA House Price Index master file, United States (traditional, all-transactions and purchase-only)
- Data period
- Latest quarter and earlier quarters of the quarterly series
- Geography vintage
- United States
- Units and scope
- Index; the base period differs between the all-transactions and purchase-only series.
- Pages with this table
- 1
Coverage and interpretation
- The House Price Index measures price changes of repeat sales and refinancings; it is an index, not a price and not an average home value. The two indexes have different base periods, so compare changes, not index levels.
- The all-transactions index uses sales and appraisals; the purchase-only index uses only purchase prices. They differ and are shown separately.
- Changes are computed by LocalFactBook from the published index levels and are not published by FHFA.
Source periods, coverage and other tables · Original publisher: documentation and methods
House Price Index by county, developmental (FHFA)
Federal Housing Finance Agency · FHFA House Price Index, annual county indexes (all-transactions, not seasonally adjusted; developmental)
- Data period
- Annual index for 1975 to 2025; newest year 2025 (Last updated: March 31, 2026)
- Geography vintage
- Counties and county equivalents as FHFA codes them (Connecticut planning regions)
- Units and scope
- Index levels (100 in the stated base year) and percent changes in the index to one decimal.
- Pages with this table
- 2,677
Coverage and interpretation
- FHFA describes these county indexes as developmental (experimental). They may be revised as new data arrive or technical adjustments are made.
- The index is calibrated with appraisal values and sales prices for mortgages bought or guaranteed by Fannie Mae and Freddie Mac. It does not cover all home sales: homes financed with other mortgages, and cash purchases are not in the underlying transactions.
- These are county figures. Each index measures cumulative nominal appreciation (not adjusted for inflation) and is not seasonally adjusted. Index levels are only comparable within one county and one base year. A level in one county says nothing about whether prices there are higher or lower than in another county.
- The base year is stated in each row. Where FHFA has too few transactions it leaves the value blank or marks it "." and the figure is shown as Not reported, never as zero; a county without a 2000 value has no index on a 2000 base.
- Changes over 1, 5 and 10 years are computed by LocalFactBook from the published index levels, as the percent change between two index values rounded to one decimal; FHFA does not publish these. They are changes in an index, not changes in the price of any home.
- Connecticut is shown by its nine planning regions, the geography FHFA uses.
Source periods, coverage and other tables · Original publisher: documentation and methods
Housing cost burden by tenure (HUD CHAS)
U.S. Department of Housing and Urban Development, Office of Policy Development and Research · Comprehensive Housing Affordability Strategy (CHAS) data, 2018-2022 ACS 5-year tabulation
- Data period
- CHAS 2018-2022 (American Community Survey 5-year estimates; HUD release of 2025-12-23)
- Geography vintage
- State, county and place summary levels as tabulated by HUD in the 2018-2022 CHAS; Connecticut counties as HUD codes them
- Units and scope
- Counts of households, rounded by HUD (0 stays 0, 1 to 7 is published as 4, 8 or more to the nearest 5); shares are one-decimal percentages computed from summed counts.
- Pages with this table
- 3,274
Coverage and interpretation
- These are HUD Comprehensive Housing Affordability Strategy (CHAS) tabulations of the American Community Survey 5-year estimates for 2018-2022 (released December 2025). Like all ACS data they come from a sample and carry sampling error; HUD does not publish margins of error for the CHAS counts, so none is shown, and small differences between places or between renters and owners should not be read as real differences.
- HUD rounds every household count before release: a count of 0 stays 0, a count of 1 to 7 is published as 4, and a count of 8 or more is rounded to the nearest multiple of 5. Counts and sums of counts here are therefore approximate, and a share built from them is approximate too.
- A household is cost burdened when its monthly housing costs, including utilities, exceed 30 percent of its monthly income, and severely cost burdened when they exceed 50 percent. The shares use as their base only the households whose burden HUD could compute; households with no or negative income, for which it cannot be computed, are left out of the base and of both counts. Counts are summed from the published cost-burden categories (30 percent or less, more than 30 percent up to 50 percent, and more than 50 percent, separately for renters and owners): the burdened count adds the last two and the base adds all three. No published percentage was averaged.
- A tenure whose category count HUD suppressed or left out is shown as Not reported, never as zero. HUD applies Census Bureau disclosure avoidance to these tabulations, which since 2018 has suppressed some previously available cells. Place figures are for the place as the ACS defines it, not the county containing it.
Source periods, coverage and other tables · Original publisher: documentation and methods
Housing units authorized by building permits (Census Building Permits Survey)
U.S. Census Bureau · Building Permits Survey, state annual files
- Data period
- Calendar years 2025 and 2024
- Geography vintage
- States and the District of Columbia
- Units and scope
- Housing units authorized, by structure size; valuation in dollars.
- Pages with this table
- 52
Coverage and interpretation
- Permits authorized are not units built or completed.
- Values are Census Bureau estimates that include imputation for permit offices that did not report every month. The reported-only row shows the part of the total that rests on permits actually reported.
- Valuation is the permit applicant’s stated construction cost in dollars, not market value. The state file reports thousands of dollars and is scaled by 1,000; the county file reports whole dollars.
Source periods, coverage and other tables · Original publisher: documentation and methods
Housing units authorized by building permits (Census Building Permits Survey)
U.S. Census Bureau · Building Permits Survey, United States national row of the state annual files
- Data period
- Calendar years 2025 and 2024
- Geography vintage
- United States
- Units and scope
- Housing units authorized, by structure size; valuation in dollars.
- Pages with this table
- 1
Coverage and interpretation
- Permits authorized are not units built or completed.
- Values are Census Bureau estimates that include imputation for permit offices that did not report every month. The reported-only row shows the part of the total that rests on permits actually reported.
- Valuation is the permit applicant’s stated construction cost in dollars, not market value. The state file reports thousands of dollars and is scaled by 1,000; the county file reports whole dollars.
Source periods, coverage and other tables · Original publisher: documentation and methods
HUD Income Limits (FY2026, state)
U.S. Department of Housing and Urban Development · State income limits and median family income, fiscal year 2026
- Data period
- HUD fiscal year 2026 (October 1, 2025 to September 30, 2026)
- Geography vintage
- States and the District of Columbia, matched on the state abbreviation of the HUD state file and its FIPS code
- Units and scope
- Fair Market Rents: US dollars per month. Income limits and median family income: US dollars per year.
- Pages with this table
- 50
Coverage and interpretation
- HUD FY2026 administrative values, not measured incomes. A state income limit is the household income HUD uses as an eligibility threshold for assisted housing programs, computed from the state median family income without HUD's local area adjustments: extremely low income (HUD field il30), very low income (il50) and low income (il80), shown for a four-person household. The median family income is HUD's estimate for the state, not a survey of residents.
- A value HUD did not publish is shown as Not reported, never as zero. Values are as HUD published them at retrieval and HUD can revise them during the fiscal year.
- HUD publishes no state-level Fair Market Rent, because Fair Market Rents are set for local areas, so none is shown. A territory for which HUD has no state record has no state table.
- The state figures are HUD’s statewide records and are not the average or total of the county or area records.
Source periods, coverage and other tables · Original publisher: documentation and methods
Low-Income Housing Tax Credit projects and units by period placed in service (HUD LIHTC, county)
U.S. Department of Housing and Urban Development · Low-Income Housing Tax Credit (LIHTC) project table, projects placed in service 1987 through 2022, from HUD’s ArcGIS FeatureServer, aggregated to the county
- Data period
- Projects placed in service 1987 through 2022; HUD service data last edited 2024-12-09 (UTC)
- Geography vintage
- Counties and county equivalents by the state code (STATE2KX) and county code (CURCNTY) in each record, matched by code to this site’s county list; Connecticut former counties are not matched
- Units and scope
- Projects: number. Units: housing units (a count, not people).
- Pages with this table
- 2,764
Coverage and interpretation
- HUD’s Low-Income Housing Tax Credit project table through 2022: counts of projects HUD’s file lists and the units they report, not the number of affordable homes that exist or are occupied in this county today. Projects placed in service in the last years are far fewer in the file (2020: 593, 2021: 48, 2022: 7 nationally), consistent with late reporting by allocating agencies, so the later periods are not a count of what was placed in service.
- The file covers projects placed in service since 1987 that allocating agencies reported to HUD. A place with no row has no project in the file; that is not a statement that it has none, because projects without a county assignment, or reported late, are not in a county table.
- Units are the file’s total-unit and low-income-unit fields as filed. An empty field or 0 is treated as no count (the file’s own substitute fields describe a total of 0 as missing or inconsistent), never as a zero total; a row where no project has a usable count shows Not reported. A few projects report more low-income units than total units; they are shown as filed. HUD’s substituted unit fields (N_UNITSR, LI_UNITR) are not used.
- Year placed in service is HUD’s field YR_PIS. The codes 8888 and 9999 are not years; HUD’s published data dictionary was not retrievable from the source host, so those projects are shown in their own row without interpretation.
- Projects HUD marks as no longer monitored (NONPROG: expired use or another reason) stay in the file and in every row above; the last row repeats them separately. That status does not say whether a project remains affordable.
- Project names, addresses, developers and contacts are not published here; only counts and unit totals for the area. Totals are not divided by population, and places are not compared or ordered.
- Public data of the U.S. Department of Housing and Urban Development (“HUD and the dataset and metadata authors assume no responsibility for the use or misuse of the dataset”); source line: U.S. Department of Housing and Urban Development, U.S. Census Bureau. This site is not endorsed by HUD.
- Counties are the county codes in HUD’s CURCNTY field (a mixture of 2010-era and later codes: Kusilvak, Alaska is 02158, while Valdez-Cordova, Alaska remains 02261). A code not on this site’s county list is left out of county tables and counted in the state table. Connecticut: the file uses the eight former counties (09001 to 09015), not the nine planning regions, so Connecticut counties are Not reported here and the projects are in the Connecticut state table.
Source periods, coverage and other tables · Original publisher: documentation and methods
Low-Income Housing Tax Credit projects and units by period placed in service (HUD LIHTC, state)
U.S. Department of Housing and Urban Development · Low-Income Housing Tax Credit (LIHTC) project table, projects placed in service 1987 through 2022, from HUD’s ArcGIS FeatureServer, aggregated to the state
- Data period
- Projects placed in service 1987 through 2022; HUD service data last edited 2024-12-09 (UTC)
- Geography vintage
- States, the District of Columbia and Puerto Rico by the state code (STATE2KX) in each record
- Units and scope
- Projects: number. Units: housing units (a count, not people).
- Pages with this table
- 52
Coverage and interpretation
- HUD’s Low-Income Housing Tax Credit project table through 2022: counts of projects HUD’s file lists and the units they report, not the number of affordable homes that exist or are occupied in this state today. Projects placed in service in the last years are far fewer in the file (2020: 593, 2021: 48, 2022: 7 nationally), consistent with late reporting by allocating agencies, so the later periods are not a count of what was placed in service.
- The file covers projects placed in service since 1987 that allocating agencies reported to HUD. A place with no row has no project in the file; that is not a statement that it has none, because projects without a county assignment, or reported late, are not in a county table.
- Units are the file’s total-unit and low-income-unit fields as filed. An empty field or 0 is treated as no count (the file’s own substitute fields describe a total of 0 as missing or inconsistent), never as a zero total; a row where no project has a usable count shows Not reported. A few projects report more low-income units than total units; they are shown as filed. HUD’s substituted unit fields (N_UNITSR, LI_UNITR) are not used.
- Year placed in service is HUD’s field YR_PIS. The codes 8888 and 9999 are not years; HUD’s published data dictionary was not retrievable from the source host, so those projects are shown in their own row without interpretation.
- Projects HUD marks as no longer monitored (NONPROG: expired use or another reason) stay in the file and in every row above; the last row repeats them separately. That status does not say whether a project remains affordable.
- Project names, addresses, developers and contacts are not published here; only counts and unit totals for the area. Totals are not divided by population, and places are not compared or ordered.
- Public data of the U.S. Department of Housing and Urban Development (“HUD and the dataset and metadata authors assume no responsibility for the use or misuse of the dataset”); source line: U.S. Department of Housing and Urban Development, U.S. Census Bureau. This site is not endorsed by HUD.
- State totals include every project in the file with that state code, including projects with no county assignment and Connecticut projects coded to former counties. Projects in American Samoa, Guam, the Northern Mariana Islands and the U.S. Virgin Islands have no page here.
Source periods, coverage and other tables · Original publisher: documentation and methods
Low-Income Housing Tax Credit Qualified Census Tracts and Difficult Development Areas (HUD, county)
U.S. Department of Housing and Urban Development · Low-Income Housing Tax Credit Qualified Census Tracts and non-metropolitan Difficult Development Areas, 2027 designations (year as in HUD’s item titles; published September and October 2026), from HUD’s ArcGIS FeatureServer, aggregated to the county
- Data period
- 2027 designations (year as in HUD’s item titles; published September and October 2026); HUD services last edited 2026-10-23 (UTC); DDA service endpoint https://services.arcgis.com/VTyQ9soqVukalItT/arcgis/rest/services/Difficult_Development_Areas_2026/FeatureServer/0
- Geography vintage
- Counties and county equivalents (2020 census tracts; Connecticut planning regions) by the county code in each HUD record, matched by code to this site’s county list
- Units and scope
- Tracts and counties: number. Flags are designation status, not people or dollars.
- Pages with this table
- 3,222
Coverage and interpretation
- Designations for the Low-Income Housing Tax Credit (Internal Revenue Code section 42(d)(5)), labelled as HUD’s item titles label them: 2027 designations (year as in HUD’s item titles; published September and October 2026). Neither HUD layer carries a year field, and the layers’ own descriptions contradict the item titles: the QCT layer description says “for 2024 … Fiscal Year 2024, Date Updated 10/2023”, the DDA layer description says “Date of Coverage: 2023-2024, Last Updated: 10-2023”, and the services are named “…_2026”. The year shown here is per the item titles (and the QCT item description, “Date of Coverage: 2027, Date Updated 09/2026”).
- Counts and designation flags, not a measure of need or of the housing built in the area. Tracts are 2020 census tracts; Connecticut tracts carry the planning-region county codes (09110 to 09190) in HUD’s file.
- The Qualified Census Tract service lists 14,539 designated tracts; it is taken as the complete designation list, so a county with none shows 0. The Difficult Development Area service lists 3,000 records: 286 non-metropolitan counties and 2,714 metropolitan ZIP Code Tabulation Areas.
- Metropolitan Difficult Development Areas are Not reported for every county: HUD designates them by ZIP Code Tabulation Area and its layer has no county field, so whether a county, or part of it, is in a metropolitan Difficult Development Area is not derived (it needs a ZCTA-to-county relationship, which is deferred). A county shown as not designated as a non-metropolitan area may still contain a metropolitan Difficult Development Area.
- Non-metropolitan counties are read from the county code in HUD’s DDA_CODE and checked against the record’s own ZCTA5 field. In one Puerto Rico record (Lares, 72081) the second number in HUD’s code reads 72141; the county is taken from the first number and the ZCTA5 field, and that second number is not used.
- Public data of the U.S. Department of Housing and Urban Development (“HUD and the dataset and metadata authors assume no responsibility for the use or misuse of the dataset”); source line: U.S. Department of Housing and Urban Development, U.S. Census Bureau. This site is not endorsed by HUD. Places are not compared or ordered.
Source periods, coverage and other tables · Original publisher: documentation and methods
Low-Income Housing Tax Credit Qualified Census Tracts and Difficult Development Areas (HUD, state)
U.S. Department of Housing and Urban Development · Low-Income Housing Tax Credit Qualified Census Tracts and non-metropolitan Difficult Development Areas, 2027 designations (year as in HUD’s item titles; published September and October 2026), from HUD’s ArcGIS FeatureServer, aggregated to the state
- Data period
- 2027 designations (year as in HUD’s item titles; published September and October 2026); HUD services last edited 2026-10-23 (UTC); DDA service endpoint https://services.arcgis.com/VTyQ9soqVukalItT/arcgis/rest/services/Difficult_Development_Areas_2026/FeatureServer/0
- Geography vintage
- States, the District of Columbia and Puerto Rico by the state code in each HUD record
- Units and scope
- Tracts and counties: number. Flags are designation status, not people or dollars.
- Pages with this table
- 52
Coverage and interpretation
- Designations for the Low-Income Housing Tax Credit (Internal Revenue Code section 42(d)(5)), labelled as HUD’s item titles label them: 2027 designations (year as in HUD’s item titles; published September and October 2026). Neither HUD layer carries a year field, and the layers’ own descriptions contradict the item titles: the QCT layer description says “for 2024 … Fiscal Year 2024, Date Updated 10/2023”, the DDA layer description says “Date of Coverage: 2023-2024, Last Updated: 10-2023”, and the services are named “…_2026”. The year shown here is per the item titles (and the QCT item description, “Date of Coverage: 2027, Date Updated 09/2026”).
- Counts and designation flags, not a measure of need or of the housing built in the area. Tracts are 2020 census tracts; Connecticut tracts carry the planning-region county codes (09110 to 09190) in HUD’s file.
- The Qualified Census Tract service lists 14,539 designated tracts; it is taken as the complete designation list, so a county with none shows 0. The Difficult Development Area service lists 3,000 records: 286 non-metropolitan counties and 2,714 metropolitan ZIP Code Tabulation Areas.
- Metropolitan Difficult Development Areas are Not reported for every state: HUD designates them by ZIP Code Tabulation Area and its layer has no county field, so whether a county, or part of it, is in a metropolitan Difficult Development Area is not derived (it needs a ZCTA-to-county relationship, which is deferred). A county shown as not designated as a non-metropolitan area may still contain a metropolitan Difficult Development Area.
- Non-metropolitan counties are read from the county code in HUD’s DDA_CODE and checked against the record’s own ZCTA5 field. In one Puerto Rico record (Lares, 72081) the second number in HUD’s code reads 72141; the county is taken from the first number and the ZCTA5 field, and that second number is not used.
- Public data of the U.S. Department of Housing and Urban Development (“HUD and the dataset and metadata authors assume no responsibility for the use or misuse of the dataset”); source line: U.S. Department of Housing and Urban Development, U.S. Census Bureau. This site is not endorsed by HUD. Places are not compared or ordered.
Source periods, coverage and other tables · Original publisher: documentation and methods
People experiencing homelessness on a January night (HUD point-in-time counts)
U.S. Department of Housing and Urban Development, Office of Community Planning and Development · 2024 Annual Homelessness Assessment Report (AHAR) to Congress, Part 1: Point-in-Time Estimates, 2007-2024 PIT Counts by State
- Data period
- Single night in the last ten days of January, 2007 to 2024
- Geography vintage
- States, the District of Columbia and Puerto Rico; Guam, the U.S. Virgin Islands, American Samoa and the Northern Mariana Islands are in the workbook but have no state page
- Units and scope
- People (all ages and household types) experiencing homelessness on one night; sheltered counts include emergency shelter, transitional housing and safe haven.
- Pages with this table
- 52
Coverage and interpretation
- Counts are point-in-time estimates of all people experiencing homelessness on a single night in the last ten days of January, as reported by each continuum of care; they are not annual totals and not a count of every person who was homeless during the year. Aggregate counts only.
- The sheltered figure is the source’s "Sheltered Total Homeless" (emergency shelter, transitional housing and, from 2010 on, safe haven combined; 2007 to 2009 have emergency shelter and transitional housing only). The unsheltered figure is the source’s "Unsheltered Homeless".
- The 2021 counts were affected by the COVID-19 pandemic: HUD reports that unsheltered counts in 2021 were disrupted and that sheltered counts should be viewed with caution, so the 2021 row is not comparable with other years. Some continuums of care did not conduct an unsheltered count in 2021, 2023 or 2024; the workbook notes that affected states for 2024 are AL, AR, CA, CO, FL, MD, NJ, OR, TX and VI, and for 2023 CA, GA, IL, MI, PR and WA, and that imputed values are used where a continuum did not count.
- Counts come from local continuums of care and can change with count method, weather and local coverage; a continuum that spans more than one state is reported by HUD under each state it covers. A value the source leaves blank is shown as Not reported, never as zero. Counts are HUD figures for the state; this site is not endorsed by HUD.
Source periods, coverage and other tables · Original publisher: documentation and methods
Subsidized households by program (HUD Picture of Subsidized Households)
U.S. Department of Housing and Urban Development · Picture of Subsidized Households, state summaries, 2025
- Data period
- Picture of Subsidized Households 2025: households reported as of December 2025
- Geography vintage
- States, the District of Columbia and territories as HUD codes them
- Units and scope
- Counts of units and households, whole percents, and whole US dollars (income per year; payment per month).
- Pages with this table
- 52
Coverage and interpretation
- These are HUD administrative figures for federally subsidized households (Picture of Subsidized Households, 2025 data, 2020-census geography). Households are reported to HUD on tenant certification forms as of December 2025 with effective dates spanning the prior 18 months; they are not a current count and are not a count of everyone in need of assistance.
- HUD suppresses a figure when fewer than 11 households were reported and marks it non-reporting when fewer than half of occupied units reported. A suppressed or non-reporting figure is shown as Not reported with HUD’s reason, never as zero. Subsidized units available are always shown when HUD publishes them. The household income and payment averages cover only households with a report; read them with the percent reported.
- Programs are HUD’s own categories. A row appears only for a program HUD lists for the area; a program with no row is one HUD shows no assisted units for. Program rows are not guaranteed to add to the all-programs row, and a Housing Choice Voucher row covers both tenant-based and project-based vouchers.
- The household payment is HUD’s average gross household contribution toward rent per month, which includes utilities; it is not the market rent and not the Fair Market Rent shown elsewhere on this page. Averages exclude zero and missing values as HUD defines them. In some areas HUD counts more households with a report than occupied units because the two counts come from different HUD systems; HUD then publishes the percent reported as 100.
- HUD assigns assisted units to a county by the address of the unit or voucher holder. Units HUD could not assign to a county are in the state figures and in no county table, so county tables need not add to the state table.
Source periods, coverage and other tables · Original publisher: documentation and methods
Subsidized households by program (HUD Picture of Subsidized Households)
U.S. Department of Housing and Urban Development · Picture of Subsidized Households, county summaries, 2025
- Data period
- Picture of Subsidized Households 2025: households reported as of December 2025
- Geography vintage
- 2020-census counties and county equivalents as HUD codes them (Connecticut planning regions)
- Units and scope
- Counts of units and households, whole percents, and whole US dollars (income per year; payment per month).
- Pages with this table
- 3,119
Coverage and interpretation
- These are HUD administrative figures for federally subsidized households (Picture of Subsidized Households, 2025 data, 2020-census geography). Households are reported to HUD on tenant certification forms as of December 2025 with effective dates spanning the prior 18 months; they are not a current count and are not a count of everyone in need of assistance.
- HUD suppresses a figure when fewer than 11 households were reported and marks it non-reporting when fewer than half of occupied units reported. A suppressed or non-reporting figure is shown as Not reported with HUD’s reason, never as zero. Subsidized units available are always shown when HUD publishes them. The household income and payment averages cover only households with a report; read them with the percent reported.
- Programs are HUD’s own categories. A row appears only for a program HUD lists for the area; a program with no row is one HUD shows no assisted units for. Program rows are not guaranteed to add to the all-programs row, and a Housing Choice Voucher row covers both tenant-based and project-based vouchers.
- The household payment is HUD’s average gross household contribution toward rent per month, which includes utilities; it is not the market rent and not the Fair Market Rent shown elsewhere on this page. Averages exclude zero and missing values as HUD defines them. In some areas HUD counts more households with a report than occupied units because the two counts come from different HUD systems; HUD then publishes the percent reported as 100.
- HUD assigns assisted units to a county by the address of the unit or voucher holder. Units HUD could not assign to a county are in the state figures and in no county table, so county tables need not add to the state table.
- Connecticut is shown by its nine planning regions, the geography HUD uses for 2025.
Source periods, coverage and other tables · Original publisher: documentation and methods
Occupied-unit rents are historical survey estimates. Fair Market Rents are administrative benchmarks. Neither is a list of homes available now, and permits are not completed homes.
Use the same definition, geography and period. Read uncertainty and missing-value notes beside the observation. A topic label alone does not establish that two figures are comparable.
These guides are under editorial review. Source display permission does not grant downloads, redistribution or commercial use. Public-source discovery is not approval for a new use.