H-1B Labor Condition Application Positions

Limitations

Federal fiscal years (October 1 - September 30), not calendar years: `year` is the fiscal year in which DOL's Office of Foreign Labor Certification issued its determination on the labor condition application (LCA) - its decision date - the basis DOL itself reports on. DOL publishes each year as quarterly disclosure files, some of them cumulative year-to-date extracts that repeat earlier quarters and some carrying a case re-determined within the year twice, so each year here is the union of its quarter files with one row per case number, the latest determination kept (FY2021 826,305 rows -> 528,902 cases; FY2023 644,607 rows -> 543,580 cases; FY2025 596,552 rows -> 594,821 cases). Counting follows DOL's rule: positions certified are the positions on LCAs certified in the year, including LCAs certified and then withdrawn by the employer within the same year; positions requested are the positions on every LCA decided in the year except those certified in an EARLIER year and withdrawn in this one, whose positions DOL counted in the year they were certified. With that rule Georgia's statewide positions certified equal DOL's published LCA Selected Statistics ('positions certified' by state of employment) exactly in every year DOL lists Georgia (FY2020 24,849, FY2021 26,770, FY2022 31,395, FY2023 25,259, FY2024 25,706, FY2025 26,301); the FY2026 statistics list only the top 10 states and Georgia is not among them. Nationally the same rows and rule reproduce DOL's applications processed (certified, denied, withdrawn) and positions requested and certified exactly in every year except two - FY2023 has 7 more certified applications and 7 more positions certified and requested than DOL's 508,096 / 925,187; FY2025 has 2 more certified applications and 2 more positions certified and requested than DOL's 561,505 / 910,968 - a few re-determined cases that the quarter files carry twice and that DOL's own tabulation resolves differently from the one-row-per-case rule. The difference is national only: Georgia's figures match DOL exactly. FY2026 is PARTIAL: it holds decisions from October 1, 2025 through June 30, 2026 only (DOL's cumulative year-to-date Q3 file) and will be replaced when DOL posts the full year; do not compare it with complete years. October 2025 holds only 1,459 determinations nationally (55 in Georgia) against 34,444-80,448 in each later month - the October 1 - November 12, 2025 federal government shutdown. An LCA is not a visa. Certified positions are employer attestations about wages and working conditions that DOL approves BEFORE the employer petitions USCIS (where cap-subject H-1B petitions go through the annual lottery) and before any worker is admitted, so they overstate H-1B employment: employers certify more positions than they fill, a petition can lose the lottery or be denied, and one worker can generate several LCAs (a new job, a change of employer, an amendment or an extension each needs one). DOL counts a 'Certified - Withdrawn' LCA as certified in the fiscal year of its original certification, and a case re-determined in a later fiscal year - typically certified in one year and withdrawn in the next - counts in both years: as a certification in the first and as an application with no positions in the second (3,608 Georgia cases appear in two years). Geography is the LCA's FIRST worksite (worksite state 'GA', county and ZIP), DOL's own state-of-employment basis. An LCA may list up to ten worksites; the others (in DOL's separate worksite files, not used here) may lie in other counties or states, so each LCA is counted once, at its first worksite, and an LCA whose first worksite is outside Georgia is not counted even if it lists a Georgia site (DOL's worksite files show about 3,400 such LCAs a year). Employer, agent, attorney and preparer addresses are never used. Worksite counties are checked against the worksite street address. The Census Bureau's geocoder located the Georgia worksite addresses in 2020 Census geography and matched 102,462 LCAs (72%) to the same street address (house number and street, or intersection) in the same ZIP code. For 91,979 of them the address lies in DOL's county. For 7,322 LCAs (10,831 positions certified; FY2020 958, FY2021 853, FY2022 1,296, FY2023 1,226, FY2024 1,381, FY2025 946, FY2026 662) it lies in another county and a second sign agrees DOL's county is wrong - DOL's county holds under 5% of the worksite ZIP's residents, or the address lies at least 500 m from DOL's county, too far for a small placement error to explain, or property records confirm the address's county (861 LCAs) - so the address's county is used (for 12 LCAs whose street quadrant differs from DOL's, only on the ZIP test). It also gives a county to 2,597 LCAs DOL left blank (3,021 positions certified). For 576 LCAs (595 positions certified) the address points to another county but lies within 500 m of DOL's county, where a geocoder can err, and DOL's county holds at least 5% of the ZIP's residents, so DOL's county is kept; some are DOL errors and some geocoder errors, and the geocoder alone cannot tell which. 52 LCAs (49 positions certified) at addresses where county tax or property records contradict the geocoder are set aside, so DOL's county stands. Where the address could not settle it, DOL's own worksite county is sometimes impossible for the worksite ZIP code - none of the ZIP's 2020 population lives in it (such as Fulton for a Cumming worksite in Forsyth County, or Decatur County for the city of Decatur in DeKalb). Such a county is replaced only when the right one is certain: at least 99% of the ZIP's population lives in one other county, and the worksite city is the ZIP's postal name or a Census place in that county, and not also a place or postal name in the stated county. 112 Georgia LCAs (116 positions certified) were moved to the ZIP's county this way (FY2020 16, FY2021 10, FY2022 15, FY2023 20, FY2024 16, FY2025 13, FY2026 22); they sit in 31 ZIP codes, 26 wholly inside the new county and 5 at least 99% inside it. Another 533 LCAs (568 positions certified) keep a stated county that is impossible for their ZIP, because the evidence falls short of certain: in 147 the ZIP straddles a county line, in 138 the city does not match the ZIP (the ZIP itself may be mistyped) and in 248 the city lies on both sides of the county line. Their county may be wrong. DOL left the worksite county blank on most LCAs decided July 1 - September 30, 2025 (the FY2025 Q4 file): 3,673 of that quarter's 4,341 Georgia LCAs (FY2025 has 22,251 in all), holding 4,282 positions certified; the quarter's other Georgia LCAs keep DOL's county. Of the 3,676 Georgia LCAs with no usable county across all years, 2,597 are placed by their street address (above) and 1,060 of the rest in the county holding most of the population of their worksite ZIP code (2020 Census ZIP-county crosswalk); 491 of those sit in ZIPs that straddle a county line. Where the street address locates an LCA with certainty, the ZIP's dominant county is the address's county on 92.1% of the FY2025 LCAs DOL left blank (92.2% of FY2025 Q3 LCAs, 92.8% over all years), so roughly 8% of the ZIP-recovered FY2025 counties may be a neighboring county. 19 FY2025 LCAs could not be placed (worksite ZIP 08648, 1716, 28217, 30006, 30099, 30158, 30320, 31412, 31999, 33613: not a Georgia ZIP in the crosswalk) and count in the statewide rows only, so FY2025's county rows sum to 19 applications, 82 positions requested and 82 positions certified fewer than the state row. The worksite county spelled 'FORSYTHE' (43 FY2020 LCAs) is Forsyth, and 'GEORGIA' given as a county is treated as blank. `case_status` is DOL's determination with DOL's rule made explicit: 'certified' and 'certified_withdrawn' (certified this fiscal year, then withdrawn by the employer in the same year) together are DOL's certified applications; 'withdrawn' and 'certified_withdrawn_earlier_year' (certified in an earlier fiscal year, withdrawn in this one; no positions) together are DOL's withdrawn applications; 'denied' is DOL's denied. `occupation_group` groups the SOC code: DOL's files switch from the 2010 to the 2018 SOC during FY2022, so 'software_developers' joins 2010 15-1132 and 15-1133 to 2018 15-1252 (BLS crosswalk), and the other groups are SOC major groups, stable across the two vintages. `visa_class` is the program: H-1B, H-1B1 (Chile, Singapore) or E-3 (Australia) - all file the same LCA; filter it to 'h1b' for H-1B alone. County rows are totals only (every categorical 'all'); the breakdowns are statewide. No suppression: rows are applications (job slots), not people, so the domain's county small-cell rule does not apply, small county counts are published, and every zero is a real zero. Wages, employer identity, industry and the H-1B-dependent flag are not published here.