H1B_Bench

FY2025 · federal data only

Is H-1B used as cheap labor?

This page tests the claim with federal data. Each section states one result, its source, and its limits. For unemployed Americans who could fill these jobs, see the labor pool.

01

Most H-1B jobs have a legal minimum below the local median wage.

DOL sets the legal minimum pay for an H-1B job at one of four wage levels. Each level is a fixed point in the local pay range for the occupation. Level I is the 17th percentile. Level II is the 34th percentile. Level III is the 50th percentile, which is the local median. Level IV is the 67th percentile.

20.4%
Level I · 17th percentile
42.0%
Level II · 34th percentile
21.0%
Level III · 50th percentile
16.6%
Level IV · 67th percentile

In FY2025, 62.4% of filings were at Level I or II. For these jobs, the legal minimum is below the local median for the same work.

Limits: The level sets the minimum only. An employer can pay more than the minimum.

Source: DOL OFLC LCA disclosure data, certified H-1B filings, FY2025. Level percentiles: DOL proposed rule, Federal Register, March 27, 2026.

02

Employers that depend on H-1B workers pay less.

DOL calls an employer “H-1B dependent” when H-1B workers are a large part of its staff. For an employer with 51 or more full-time workers, the limit is 15%.

Employer typeFilingsLevel I or IIPaid exactly the legal minimumMedian offered pay
“H-1B dependent” employers131,46173.6%52.0%$105,997
Other employers407,98358.6%27.9%$127,878

“H-1B dependent” employers paid exactly the legal minimum on 52.0% of filings. Other employers did this on 27.9% of filings.

Limits: The “H-1B dependent” label comes from the employer's own filing.

Source: DOL OFLC LCA disclosure data, certified H-1B filings, FY2025.

03

Most H-1B filings offer less than the local median pay.

We compare the pay on each filing with the local median pay for the same occupation in the same area. The local median comes from DOL's own wage library.

57.1% of matched filings offer less than the local median. When a filing is below the median, the median gap is −$23,108 a year.

Employer typeMatched filingsBelow local medianMedian gap
All employers404,91757.1%−$6,834
“H-1B dependent” employers96,32570.0%−$18,293
Other employers308,59253.0%−$960

These large sponsors offer less than the local median most often. Each has 500 or more matched filings.

EmployerMatched filingsBelow local medianMedian gap
WIPRO LIMITED1,65590.3%−$19,198
TATA CONSULTANCY SERVICES LIMITED6,70890.3%−$20,425
TECH MAHINDRA (AMERICAS), INC1,06585.4%−$19,531
COMPUNNEL SOFTWARE GROUP, INC1,57278.9%−$16,689
Qualcomm Technologies, Inc.1,68477.3%−$23,192
Hexaware Technologies, Inc.50575.8%−$18,470
Deloitte Tax LLP83371.3%−$17,389
Deloitte & Touche LLP1,22470.3%−$17,888
Intel Corporation2,49069.9%−$10,777
INFOSYS LIMITED3,46067.1%−$16,744
LTIMindtree Limited2,13166.9%−$14,852
Fidelity Technology Group, LLC d/b/a Fidelity Investments1,15666.7%−$18,587
Amazon Development Center U.S., Inc.2,31166.4%−$18,846
ServiceNow, Inc.79465.1%−$13,092
UST Global Inc73665.1%−$9,438

Limits: Offered pay is the bottom of the offered range. The local median covers all workers in the occupation and area, at all experience levels. 404,917 of 501,934 yearly, full-time filings matched an area and a wage. DOL left the worksite county blank on most filings in its FY2025 Q4 file, so most of those filings are not in the count.

Source: DOL OFLC LCA disclosure data, certified H-1B filings, FY2025. DOL OFLC Online Wage Library, Level III wage, wage years 2024-25 and 2025-26. Method, committed before the results.

04

Filings with an employer-chosen survey offer less.

Each filing states where its wage floor came from. Most filings use government data. Some employers choose a private salary survey instead.

Filings with a private survey offer less than the local median 75.8% of the time. Filings with government data do this 55.3% of the time.

Where the wage floor came fromFilingsShareBelow local median
Government data (OEWS)498,48092.4%55.3%
Private survey, chosen by the employer33,7596.3%75.8%
Union contract6,5311.2%92.7%
DOL determination5900.1%81.1%
Federal contract wage840.0%92.3%

The survey publishers on the most filings:

Survey publisherFilingsBelow local median
Willis Towers Watson11,24172.3%
Radford (Aon)9,52177.6%
Aon6,24277.6%
Mercer3,09272.1%
AAMC1,16597.3%
Pearl Meyer1,03076.8%
U.S. Bureau of Labor Statistics2098.5%
CUPA-HR19093.7%

Limits: The source is the employer's own entry on the filing. A private survey can be lawful and accurate. This result compares outcomes; it does not judge any single survey. The local-median comparison uses only filings that matched in section 03.

Source: DOL OFLC LCA disclosure data, certified H-1B filings, FY2025. Method, committed before the results.

05

After DOL investigations, employers agreed to pay $123.2M in H-1B back wages.

DOL's Wage and Hour Division found H-1B violations in 2,886 cases. In these cases, employers agreed to pay $123.2M in back wages to 13,595 workers. DOL also assessed $11.7M in penalties.

99
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

Back wages by fiscal year of the findings. Each bar shows the number of cases on hover.

The employers with the most H-1B back wages:

Employer (as named by WHD)StateCasesBack wagesWorkers owed
Prince George's County Public SchoolsMD1$4,222,1461,046
Advanced Prfessional Marketing, Inc.NY1$2,865,216156
Patni Computer Systems Inc.MA2$2,411,537609
Computech CorporationMI1$2,250,00036
Byton North America CorporationCA1$1,749,18476
GlobalCynex, Inc.VA2$1,683,583343
Populus Group, LLCMI3$1,159,056596
Metropolitan Center for Mental HealthNY2$1,113,67715
Smartsoft International, Inc.GA1$997,732135
Sirsai Inc.WA1$762,806115

Limits: DOL investigates a small share of employers, so these cases measure enforcement, not the full rate of violations. The data holds concluded cases only, so recent years are incomplete. Back wages are amounts the employer agreed to pay. The data does not show if the employer paid. The totals count only cases that have a findings date.

Source: DOL Wage and Hour Division, concluded compliance actions, H-1B cases with findings from FY1999. Method, committed before the results.

06

In the same year they filed for H-1B workers, employers gave notice to lay off 21,743 workers.

A WARN notice tells the state about a large layoff before it occurs. We matched notices in Texas and California to employers with certified H-1B filings in the same fiscal year. The company names must match exactly.

153 companies filed 367 WARN notices to lay off 21,743 workers. In the same year, these companies had 21,923 H-1B filings certified.

The matched companies with the most workers in WARN notices:

Limits: WARN notices do not list jobs, so a layoff can be in different roles than the H-1B filings. WARN applies only to large layoffs, generally 50 or more workers at one site. The data covers two states, and California covers 9 of the 12 months. A notice under a site or brand name does not match, so these counts are lower than the true totals.

Source: DOL OFLC LCA disclosure data, certified H-1B filings, FY2025. Texas Workforce Commission WARN notices, Oct 2024 to Sep 2025. California EDD WARN report, Oct 2024 to Jun 2025. Method, committed before the results.

07

In the FY2024 lottery, 53.9% of registrations were for people with more than one registration.

Each year, employers register the people they want to hire. USCIS selects registrations at random. Up to the FY2024 lottery, each registration was one chance of selection. A person with registrations from several employers had several chances.

In FY2024, 408,891 of 758,994 eligible registrations were for people with more than one registration. USCIS cites evidence from the FY2023 and FY2024 lotteries. It opened fraud investigations and changed the rule.

FY2021241,299 for people with one registration, 28,125 for people with multiple registrations.
FY2022211,304 for people with one registration, 90,143 for people with multiple registrations.
FY2023309,241 for people with one registration, 165,180 for people with multiple registrations.
FY2024350,103 for people with one registration, 408,891 for people with multiple registrations.
FY2025423,028 for people with one registration, 47,314 for people with multiple registrations.
FY2026336,153 for people with one registration, 7,828 for people with multiple registrations.

From FY2025, USCIS selects by person, so more registrations do not give more chances. In FY2026, the share fell to 2.3%.

Cap yearEligible registrationsFor people with multipleShareSelected
FY2021269,42428,12510.4%124,415
FY2022301,44790,14329.9%131,924
FY2023474,421165,18034.8%127,600
FY2024758,994408,89153.9%188,400
FY2025new rule470,34247,31410.1%135,137
FY2026343,9817,8282.3%120,141

Limits: A person can have offers from more than one employer for lawful reasons. The data does not show which employers worked together on registrations. The share is the upper limit of this abuse, not a count of fraud. USCIS does not publish the employers behind these registrations.

Source: USCIS, H-1B Electronic Registration Process, historical data, cap years FY2021 to FY2026. Final rule, Federal Register, February 2, 2024. Method, committed before the results.

08

91.5% of green card filings for professional jobs were for a worker the employer already employed.

To sponsor a worker for a green card, the employer must show that no qualified U.S. worker applied for the job. The employer places job ads and reviews the U.S. applicants.

In FY2025, the worker already worked for the employer on 91.5% of certified filings for professional jobs. For all jobs, the share was 67.1%. The employer tests the U.S. labor market while its sponsored worker already works for it.

6,408 certified filings came from 147 employers that reported a layoff in the area, in the occupation or a related occupation, in the 6 months before the filing.

The DOL approval is for one employer only. A worker who moves to a different employer must start a new case with that employer. The DOL step alone took a median of 483 days. For 90% of filings, it took 510 days or less.

Limits: The green card file does not show the worker's visa status, so it does not show how many are H-1B workers. Both answers are the employer's own entries on the form. A layoff is lawful if the employer told the laid-off workers about the job and considered them. The data does not show that step. The wait is the DOL step only. The full green card wait is longer.

Source: DOL OFLC PERM disclosure data, certified filings, FY2025. Method, committed before the results.