Every number on Beacon CRE's growth pages is a published U.S. federal statistic (Census Bureau, Bureau of Labor Statistics or IRS), shown with its own period and source; changes compare the latest period with 5 years earlier, and ranks compare only areas with a figure for the same period.
Figures last refreshed
Summary sentences on each page are written by fixed rules from these numbers, not by AI. Figures are informational only, not investment advice.
Headline measures
9 measures appear on the county, metro and rankings pages. Each tile on a page shows its own period: one page can mix years, because each agency publishes on its own schedule.
Latest: 2025 (3,144 counties). Change compared with 2020. Series from 2020.
How it is computed
The Census Bureau's resident population estimate for July 1 of each year (so "since 2020" means July 1, 2020, not the April 1 census day). Change is the percent change from 5 years before the latest period, or from the earliest period loaded when that year is missing.
Agency releases
Annual
Net domestic migration
People moving in from other U.S. counties minus people moving out
Net domestic migration for the year to June 30: people who moved in from other U.S. counties minus people who moved out. Moves to and from other countries are counted separately, as international migration. Per 1,000 residents divides by the same year's population estimate; counties with fewer than 1,000 residents get no per-resident rate. The multi-year total adds every year loaded, and the components table adds natural change and the Census adjustment.
Agency releases
Annual
Jobs
Average annual employment covered by unemployment insurance
Latest: 2025 (146 counties). Change compared with 2020. Series from 2020.
How it is computed
The annual average of monthly employment covered by unemployment insurance, all industries and ownerships (private and government). Change is the percent change from 5 years before the latest period, or from the earliest period loaded when that year is missing. QCEW: NC/SC only. Outside North and South Carolina, jobs are private-sector employment from County Business Patterns.
Latest: 2025 (146 counties). Change compared with 2020. Series from 2020.
How it is computed
Total covered wages divided by covered jobs, per week, all industries and ownerships, in dollars of each year (not adjusted for inflation). Change is the percent change from 5 years before the latest period, or from the earliest period loaded when that year is missing. QCEW: NC/SC only: pages outside North and South Carolina show no wage figure.
Latest: 2025 (3,143 counties). Change compared with 2019. Series from 2010.
How it is computed
The annual average unemployment rate (BLS series period M13). Change is in percentage points, measured from 2019 (the last year before the pandemic spike) when that year is loaded, otherwise from 5 years earlier.
Agency releases
Monthly
New homes permitted
Housing units authorized by building permits in the year
Latest: 2025 (3,024 counties). Change compared with 2021. Series from 2021.
How it is computed
Housing units authorized by building permits in the calendar year, from the Census Bureau's annual county files: units in 1-unit, 2-unit, 3-4-unit and 5+-unit buildings, including the Bureau's estimates for places that did not report. Permits count homes authorized, not homes finished. Per 1,000 residents divides by the same year's population estimate; counties with fewer than 1,000 residents get no per-resident rate. Change is the percent change from 5 years before the latest period, or from the earliest period loaded when that year is missing.
Latest: 2020–24 (3,144 counties). Change compared with 2016–20. Series from 2016–20.
How it is computed
The median household income from the American Community Survey 5-year estimates, in dollars of the survey's last year. The change compares the latest 5-year window with the earliest window loaded; the two windows can share a year and are not adjusted for inflation, so read the change as a rough guide.
Agency releases
Annual
Private-sector employment
Paid employees of private employers in mid-March (County Business Patterns)
Latest: 2023 (3,142 counties). Change compared with 2018. Series from 2018.
How it is computed
Paid employees of private employers in the pay period including March 12 (County Business Patterns). Government jobs and the self-employed are not counted. Change is the percent change from 5 years before the latest period, or from the earliest period loaded when that year is missing.
Agency releases
Annual
Net migration (IRS tax returns)
People on tax returns moving in minus moving out, U.S. and abroad; the sum of the member counties (moves between them cancel out)
People on tax returns (exemptions) who moved in minus those who moved out between one filing year and the next, U.S. and abroad; people who do not file are not counted. Metro pages add up the member counties, so moves between them cancel out.
Agency releases
Annual
Population components (births minus deaths, international migration and the Census adjustment) come from the same Census population estimates, from 2021 on (the 2020 figures cover April to July only).
Changes and small numbers
Percent changes compare the latest period with 5 years earlier, or with the earliest period loaded when that year is missing. The comparison period is printed next to every change.
Unemployment changes are in percentage points from 2019, so the 2020 pandemic spike is not used as the starting point.
Small base: when the latest or the starting figure is below a floor (population 1,000, jobs 100, private-sector employment 100, new homes permitted 20), a percent change swings on a handful of people, jobs or homes, so the page shows "small base" and makes no change, per-resident or rank claim.
Per-1,000-resident rates need at least 1,000 residents.
Survey drop-outs: a county whose latest building-permit figure is older than the survey's current year has left the survey; its page says "not in the survey after <year>" instead of showing an old figure as current, and it is not ranked.
Revisions: each data refresh replaces a source's whole series, so when an agency revises earlier years (a new population vintage, for example), the revised figures replace the old ones.
How ranks are built
Each county is ranked within its state and against every U.S. county with a comparable figure. "County" includes county equivalents: independent cities, parishes, and Alaska boroughs and census areas.
Population growth: percent change over the period shown; 1st is the highest; 3,107 of 3,144 counties ranked nationally.
Net domestic migration per 1,000 residents: per 1,000 residents in the latest year; 1st is the highest; 3,108 of 3,144 counties ranked nationally.
Private-sector job growth: percent change over the period shown; 1st is the highest; 3,092 of 3,144 counties ranked nationally.
Median household income: the latest value; 1st is the highest; 3,107 of 3,144 counties ranked nationally.
Average weekly wage: the latest value; 1st is the highest; no U.S. rank (loaded for North and South Carolina only).
Lowest unemployment rate: the latest value; 1st is the lowest; 3,108 of 3,144 counties ranked nationally.
New homes permitted per 1,000 residents: per 1,000 residents in the latest year; 1st is the highest; 3,005 of 3,144 counties ranked nationally.
Who is left out of a rank:
Counties with fewer than 1,000 residents (on every measure).
Counties whose figure is for a different period from most others (for example a county missing from the latest release), so an old figure is never ranked against new ones. That is why the number ranked differs by measure and is below the number of counties.
Counties with a small-base change (see above) or without a comparable figure.
National ranking pages list counties with at least 10,000 residents (the top 100), numbered among themselves; each county's own page ranks it against every U.S. county with the same figure.
On North and South Carolina county pages, the opening sentence ranks population and job growth among the 146 counties of both Carolinas; the rank table ranks within the state and the U.S. Metro and micro areas are ranked separately from each other.
States and metro areas
State figures add up the state's counties (counts only: population, migration, jobs, permits and private employment).
Carolinas metro pages add up their member counties, only for years every member has; medians, wages and rates stay on each county's page. A metro with less than 98% of its land in North and South Carolina is marked as covering its Carolinas counties only.
Other U.S. metro and micro area pages use the figures each agency publishes for the area itself (OMB Bulletin 23-01 delineation; ACS 2016-20 and CBP 2018 use the delineation of their year). IRS migration is the sum of the member counties.
Which pages are indexed
A county page is indexed by search engines only when it has at least 4 of these 6 core figures: population, net domestic migration, private-sector employment, unemployment rate, new homes permitted, median household income. Today 3,143 of 3,144 county pages qualify.
A ranking page is indexed only when at least 10 counties are ranked on it.
National metro and micro area pages are held to a stricter bar: at least 5 of the 6 core figures. A one-county area is never indexed, because its figures are its county's and the county page already has them (580 of 900 areas). Today 320 metro and micro area pages qualify.
Pages below the bar still answer and carry their links (noindex, follow), but they are left out of the sitemaps and of IndexNow submissions.
How often data refreshes
Beacon CRE's data pipeline checks the federal sources every day. A new release reaches this site after it passes our checks, and every page shows the date its figures were last refreshed. The figures on this site were last refreshed on . The agencies publish on these cycles:
Every figure on the growth pages is a work of the U.S. Government and in the public domain (17 U.S.C. 105). You may reuse it; please cite the agency. Beacon CRE's page text and layout are not public domain.
Using federal data does not mean any agency endorses Beacon CRE. More on every layer: data sources and licences.
What is not included
Forecasts or projections on these free pages: every figure here is a published past value. Only the paid Carolinas quarterly report has an outlook (see below).
Margins of error: ACS figures are survey estimates with sampling error, which is wide for small counties, so nearby ranks may not really differ.
Inflation adjustment: income and wage changes are in each year's dollars.
Flags on imputed permits: Census estimates for places that did not report are included in the permit counts and are not marked.
QCEW: NC/SC only. Outside North and South Carolina, jobs are private-sector employment from County Business Patterns.
Government jobs and the self-employed in private-sector employment (CBP).
Labour-force counts, commuting flows, industry mix and wages by industry.
A rental vacancy rate: the report's vacant-homes figure counts every vacant home, seasonal and second homes included.
Map, parcel and listing layers cover North and South Carolina only for now.
The $99 county report
The report for a county outside North and South Carolina is built from the same federal sources as these pages, with a source next to every number. It adds these figures, which are not on the free pages:
Median gross rent
Vacant homes, incl. seasonal and second homes
New homes permitted, last 12 months
Rent growth, 5 years
Permits in 5+ unit buildings
IRS income moving in (net AGI)
The last-12-months permit count comes from the Census Bureau's monthly county files, which are preliminary and cover only places that report monthly. The report's own methods page says:
This edition uses federal county statistics only. Parcel, listing, sales, tract, map and site-readiness layers cover North and South Carolina only for now, so they are not in it.
Ranks put 1st as highest, except vacancy and unemployment (1st is lowest). A rank compares only counties whose figure is for the same period; counties without one are left out.
State values are the state's own published figure where the source has one. Otherwise rates are population-weighted means of the state's counties (marked *). IRS migration has no state value: adding counties together would count moves inside the state.
U.S. values are the median county (half of U.S. counties are above it), not a national total, so a county can be compared with a typical one of any size.
Neighbouring counties share a border (Census county adjacency file), across state lines.
Private-sector employment is County Business Patterns (paid employees in mid-March); it excludes government and the self-employed. Five-year changes compare the latest year with five years earlier, or with the earliest year loaded when that year is missing.
IRS migration counts tax returns (households) and exemptions (people) that moved between counties from one filing year to the next; non-filers are not counted.
ACS figures are 5-year estimates with sampling error, which is wide for small counties.
North and South Carolina county reports add Carolinas layers; each report lists its own sources.
The free growth pages show only published past values. The paid quarterly report for a North or South Carolina county adds a 5-year and 10-year outlook for some measures (population, jobs (county business patterns), homes permitted per year, median gross rent), each with an 80% range, counted from the latest year of data.
Each method is backtested: it forecasts past years from the data available at the time, and its errors are compared with simple baselines ('same as the latest year' and straight-line trends).
A model is used only if it beat the best baseline in those tests, by a statistical test at the 10% level, with at least 3 past years tested.
The report prints a cell only when such a model exists for that county, measure and horizon, with its tested average error and the number of past years it was tested on. Trend-only, untested and 'no reliable forecast' cells are left out, and a measure with no tested cell is not shown.
The 80% ranges come from the backtest errors. At 5 years and beyond, outcomes fell inside the range less often than 80% of the time for several measures, so read the range as a guide, not a guarantee.
An outlook is an estimate, not investment advice.
Corrections
Found a number that doesn't match its source? Email hello@beaconcarolinas.com with the page and the figure. We check it against the agency's file and fix or explain it.