Skip to navigation

Skip to main content

Rural Migration News

contact us

The H-2A Program and AEWRs

 Click here to download this blog post as a PDF file

December 18, 2020

The H-2A program expanded in 2020 despite high US unemployment. The US Department of Labor (DOL) reduced the wages that must be paid to H-2A workers in 2021 and 2022, which should encourage faster H-2A expansion.

DOL certified the applications of 13,552 farm employers to fill 275,430 jobs with H-2A workers in FY20. The leading H-2A states were FL with 14 percent of H-2A jobs certified, GA and WA with 10 percent each, CA with nine percent, and NC with eight percent. The top five states accounted for over half of all job certifications.

Florida-based employers were certified to fill 60,100 jobs with H-2A workers, but only 36,600 of these jobs were in Florida. Many FL-based FLCs move H-2A workers to other states, explaining why almost 40 percent of the jobs certified to FL-based employers were in other states.

Employers specify the occupation or title of the farm jobs they seek to fill with H-2A workers. In FY20, over 88 percent of jobs certified were for crop farm workers, followed by six percent for equipment operators and four percent for animal farm workers.

The largest H-2A employer was the NC Growers Association with 10,650 jobs certified, followed by Fresh Harvest with 5,300, WALFA with 4,400, and Foothill Packing with 3,200. The NCGA accounted for almost half of H-2A jobs certified in NC, Fresh Harvest accounted for a fifth of jobs certified in CA, and WAFLA accounted for a sixth of jobs certified in WA.


Farm employers must pay H-2A workers and US workers in corresponding employment the higher of the Adverse Effect Wage Rate (AEWR), the prevailing wage in the area, the wage in an employer’s collective bargaining agreement, or the federal or state minimum wage. The AEWR is usually the highest of these wages.

Since June 1, 1987, the AEWR for the current year is the average hourly earnings of non-supervisory field and livestock workers for the state or region during the previous year, as determined by the USDA NASS Farm Labor Survey (FLS), which asks farm employers to report the earnings and hours worked of their hired workers for the week that includes the 12th of the month for January, April, July, and October.

Analysts divide earnings by hours worked to calculate average hourly earnings, which were published for multistate regions and FL, CA, and HI by job title or occupation in 2019. Over the past three decades, the year-to-year change in FLS earnings and thus AEWRs has been highly variable, up 15 percent or more one year and down five percent or more in other years. Nevada-based Peri & Sons sued DOL in 2019 when the AEWR for the Mountain II states rose 20 percent between 2018 and 2019, but the suit was dismissed.

AEWRs by Occupation

DOL’s July 26, 2019 Notice of Proposed Rule Making proposed to change from one AEWR per state to “separate AEWRs by agricultural occupation to better protect against adverse effect on the wages of similarly employed workers in the US” (p36171). Employers specify the job title or occupation of the job they want to fill with H-2A workers in Section E of From 9142A, such as agricultural equipment operators (45-2091), crop farm workers (45-2092), and animal farm workers (45-2092). These three job titles accounted for 96 percent of job certifications in FY20.

In 2020, there was one AEWR per state or multistate region, so the AEWR for equipment operators was the same as the AEWR for crop workers, for example, $11.71 in Florida.

How will DOL set AEWRs by occupation? DOL proposed to look first to USDA’s FLS to obtain the average hourly earnings for the employer-specified occupation. If the FLS reports wage data for the occupation, the FLS average earnings for one year would become the AEWR for the next year, continuing the current practice but of using the previous year’s average earnings, but specifying a different AEWR for each job title.

If the FLS does not generate an hourly earnings estimate for a particular occupation, DOL would turn to Occupational Employment Statistics ( to determine an AEWR. The OES does not collect data from farm employers, only from nonfarm businesses that provide services to farms such as farm labor contractors. The OES collects data in wage ranges. Employers report to OES, for example, the number of employees who earn $12.50 to $15 an hour. DOL proposed to use the mean or average OES wage to set the AEWR for job titles specified by employers and not covered by the FLS.

DOL’s justification for setting AEWRs by occupation, and to use the OES rather than the FLS to determine AEWRs when necessary, is to protect skilled US workers from being adversely affected by H-2A workers. However, in a November 5, 2020 final rule (p35), DOL said that the AEWRs for 99 percent of farm jobs would be set using 2019 FLS data, including job titles 45-2041 - Graders and Sorters; 45-2091 - Agricultural Equipment Operators; 45-2092 - Farmworkers and Laborers, Crop, Nursery, and Greenhouse; 45-2093 - Farmworkers, Farm, Ranch, and Aquacultural Animals; 53-7064 - Packers and Packagers, Hand; and 45-2099 - Agricultural Workers.

DOL’s New Methodology

On September 30, 2020, USDA announced that it was ending the FLS because the employment and earnings data that were being collected for workers who are hired directly by farmers were available from other sources. The UFW sued, and a federal judge ordered USDA to collect the 2020 data in order to set AEWRs for 2021.

On November 5, 2020, DOL issued a final rule changing the methodology for adjusting AEWRs. For the occupations of 99 percent of the H-2A jobs that were certified in FY20, DOL froze 2020 AEWRs until January 1, 2023, after which they will increase with the annual increase in the Employment Cost Index (ECI) for wages and salaries. The ECI rose about half as fast as the FLS over the past decade, about 2-3 percent a year versus 4-5 percent.

The occupations affected by the 2021-22 wage freeze and the adjustment thereafter with the ECI include (p35) 45-2041 - Graders and Sorters; 45-2091 - Agricultural Equipment Operators; 45-2092 - Farmworkers and Laborers, Crop, Nursery, and Greenhouse; 45-2093 - Farmworkers, Farm, Ranch, and Aquacultural Animals; 53-7064 - Packers and Packagers, Hand; and 45-2099 - Agricultural Workers.

For other job titles such as supervisor and construction laborer, DOL will use the mean wage for the occupation in each state’s OES data and, if there is no statewide OES data for the occupation, the national average OES wage for that occupation.


DOL’s November 5, 2020 final rule estimates that employers will save an average $171 million a year (p7 and p80) in response to its efforts to better protect construction laborers, drivers, supervisors, and a few other workers “against adverse effects on an occupation basis.” If the DOL final rule had been in effect for 2020, AEWRs would have risen in about 1,000 certified jobs and fallen in 274,000 certified jobs.

DOL on p86 notes that an average of 8,000 unique US employers (EINs) file about 13,000 H-2A certification applications a year (some employers file several applications). This means that less than two percent of the 500,000+ US farm employers who reported expenses for hired labor in the 2017 COA, and less than eight percent of the 105,000 US agricultural employers who were registered with state UI authorities across the US in 2020, apply for H-2A workers.

DOL estimates (p96) that freezing AEWRs at 2020 levels and adjusting them by the ECI will “transfer” $13 million a year from 45-2091 equipment operators, $120 million a year from 45-2092 crop workers, and $8 million a year from 45-2093 animal workers to H-2A employers, for a total of $141 million a year. This worker-to-employer transfer is offset by a gain of $18 million a year in higher wages for other occupations, primarily supervisors, drivers, and construction laborers, for a net transfer from H-2A workers to employers of $124 million a year.

DOL developed examples of wage bill changes that assume more than full-time work. In calculating average QCEW employment, DOL assumes that a full-work year is 52-40 hour weeks or 2,080 hours. DOL’s AEWR example assumed a work-year for 45-2091 equipment operators of 306-7 hour days or 2,142 hours and for 53-7062 laborers a work year of 280-8 hour days or 2,240 hours. This is likely the opposite of what would occur in practice, where equipment operators normally have longer hours than laborers.

The 50,400 US workers in corresponding employment, meaning they work on farms with H-2A workers, will also see their wages frozen, transferring another $29 million to employers (p97). Total wage savings from the AEWR freeze are thus $141 million plus $29 million or $170 million year in 2021 and 2022. DOL did not estimate any additional wage savings that may arise if frozen AEWRs could speed the expansion of the H-2A program nor did it consider the effects of the AEWR freeze on the wages of US farm workers who are not employed on farms with H-2A workers.

DOL considered and rejected several other options to set and adjust AEWRs that would have reduced employee-to-employer wage transfers. DOL rejected continued reliance on the FLS to set wages by occupation, which it estimated would result in an annual transfer of $18 million a year from workers to employers as wages for most workers fell after higher wage equipment operators are withdrawn from the FLS sample. DOL also rejected reliance on OES data to set all AEWRs by occupation, which it estimated would result in an annual employee-to-employer wage transfer of $66 million a year (pp99-100).

DOL rejected continued reliance on the FLS, arguing that the 2021-22 wage freeze followed by ECI adjustments “allows specific OES wages for workers in higher-paid agricultural occupations, such as supervisors of farmworkers and construction laborers on farms, while simplifying the AEWR for SOC codes set by the FLS AEWR and tying it to the ECI index.” (p100). DOL rejected the switch to OES data to set all AEWR wages by occupation because the FLS “is a superior wage source to the OES for those [almost all] occupations.” (p100).


The AFBF welcomed DOL’s November 5, 2020 final rule, saying it would provide “stability and predictability in labor costs.”. The AFBF emphasized that the ECI (solid blue line) has increased at a lower and more stable rate than the FLS (dashed black line) over the past decade.

The FLS (black dashed line) rose faster and with more variability over the past decade than the ECI (blue line)

Worker advocates condemned the cancellation of the FLS survey and the DOL November 5, 2020 final rule. They sued successfully to require USDA to conduct the FLS for 2020, and sued DOL to block the AEWR wage freeze, the switch to ECI to adjust the AEWR beginning in 2023, and using the OES for some job titles.

The suits against DOL are likely to raise several points, including the 2021-22 wage freeze, alternatives to the ECI for AEWR adjustments, using prevailing wage studies to set AEWRs, and potential knock-on effects of lower AEWRs on US workers.

First, the 2021-22 wage freeze is a clear transfer of wages from workers to employers that DOL justified to give farmers time to adjust to the new methodology. However, since AERWs based on the FLS have been changing each year, do employers need two years to adjust to the slower-rising ECI?

Second, why did DOL select the ECI to adjust farm worker wages beginning in 2023? The ECI does not cover agriculture, and the few farm workers it includes are those brought to farms by nonfarm support firms; farm workers are grouped with construction and mining workers in SOC 45-47. The federal government uses the ECI to adjust military and federal employee pay and Medicare reimbursements, while state and local governments use the ECI to adjust professional consulting rates and personnel costs. The ECI does not appear to be used to adjust wages in any temporary worker program.

If DOL wanted to switch away from the FLS to adjust AEWRs, there were better possibilities, including the annual change in the weekly wage of agricultural workers from the QCEW. The QCEW covers about 80 percent of US agricultural employment and 100 percent in six states including CA and WA, excludes H-2 workers in many states (not CA and WA), and the annual changes in QCEW weekly wages closely tracked the annual changes in FLS average hourly earnings over the past decade. QCEW data are not available by occupation, but are available by NAICS industry codes at the state level, so that crop worker AEWRs could be adjusted separately from animal worker AEWRs and support worker AEWRs.

Third, DOL’s overall goal in changing the AEWR methodology was to better protect the relatively few H-2A workers in farm-related occupations, including 3,256 construction laborers in FY20, 577 drivers, 379 supervisors, and 100 helpers who collectively accounted for 1.5 percent of the 275,000 jobs certified in FY20. DOL will use OES data to set AEWRs for these job titles.

An alternative to freezing AEWRs, adjusting with the ECI, and using the OES for missing job titles would be to conduct more prevailing wage surveys. If SWAs conducted prevailing wage surveys, they could be used to set AEWRs for the job titles not covered by the FLS and collect information on wage systems and working conditions. SWA prevailing surveys would have the advantage of increasing SWA knowledge of farm labor markets and providing more information to DOL to evaluate employer job orders.

Fourth, DOL did not consider the effects of freezing AEWRs for two years and then adjusting them with the ECI on US farm workers who are not employed alongside H-2A workers. Average annual employment in US agriculture is about 1.5 million, so that 200,000 H-2A workers who are in the US an average six months fill 100,000 year-round equivalent jobs, less than seven percent of average agricultural employment.

A $15 AEWR may prompt US workers to also expect $15 an hour, while a frozen AEWR may also freeze US worker wages. Average annual farm wages of $40 billion have been increasing by four percent a year or $1.6 billion. If the US farm worker wage bill does not increase because of the AEWR freeze, farm employers would save $1.6 billion a year and then an additional $800 million a year if the AEWR rises by two percent a year with the ECI rather than four percent a year with the FLS.

The AEWR freeze and switch to ECI adjustment clearly saves employers money. The question is whether the savings are confined to H-2A workers and US workers employed alongside them or extend to all US farm workers.


DOL incorporated “adverse effect” into its regulations in 1951. In 1956 and 1958, DOL found an adverse effect of Braceros on US workers because some growers who employed Braceros paid lower wages than growers who employed US workers. DOL established a national minimum wage for Braceros of $0.50 an hour in 1958 at a time when US farm workers were not covered by the FLSA.

In 1959, DOL asked four consultants to evaluate the effects of almost two decades of Bracero employment on US farm workers. They concluded that “wage rates in activities which Mexicans [Braceros] are employed have lagged behind the rising wage level for farm work generally” (1959, pp4-5) and urged DOL to determine a wage “rate necessary to avoid adverse effect on domestic wage rates” in crops and areas dominated by Braceros (1959, p283).

DOL introduced statewide AEWRs in May 1962 of $0.95 an hour in AZ, $1 an hour in CA, $0.90 in CO, $1 in KS, $0.75 in NM, and $0.70 in TX (DeLone, 1992, p108).. These AEWRs were usually higher than prevailing wages, and had to be offered to any US workers employed by farmers seeking certification to hire Braceros (Martin, 2008, p13). In 1964, DOL set AEWRs for 1965 by increasing each state’s 1950 average agricultural wage by the increase in manufacturing wages between 1950 and 1962. In 1967 DOL adjusted AEWRs using the increase in the “average national agricultural wage” between 1963 and 1965 (DeLone, 1992, p110).

DOL in 1968 began to change AEWRs each year based on the change in the annual wage of field and livestock workers in each state or region over the previous two years, although DOL did not describe this methodology in regulation until 1976 (DeLone, 1992, p111).

There were revisions to H-2 regulations in NPRM issued January 25, 1977, when DOL asserted that “normal adjustments in wages and working conditions should bring sufficient US workers to an occupation,” sending a signal that DOL expected the number of H-2 workers to shrink over time. DOL issued a final rule in 1978 that included requiring employers to raise piece rate wages when AEWRs rose so that workers would not have to work harder to earn the higher AEWR.

DOL on January 16, 1981 adopted a new methodology to calculate AEWRs after concluding that “the current AEWR methodology ha[d] not successfully achieved the purpose of preventing wage deflation of similarly employed US workers” because AEWRs were pegged to initial statewide wages that had been depressed by the presence of guest workers.

DOL considered five AEWR methodologies before issuing the final 1981 rule, including a national AEWR based on the average hourly earnings of piece rate workers, national AEWRs by crop activity, statewide AEWRs based on the federal minimum wage and adjusted by the percentage change in field worker wages, a national AEWR that was 25 percent above the federal minimum wage, and continuing to adjust AEWRs to reflect the increases in the average hourly earnings of field and livestock workers in the FLS.

DOL selected the piece rate worker option, which would have generated the highest AEWR, because DOL found that most guest workers earned piece rate wages. However, the new AEWR methodology was withdrawn by the incoming Reagan Administration in June 1981, a year in which USDA did not conduct the FLS, so the 1981 AEWRs were left in place in 1982. DOL adjusted statewide AEWRs based on the change in average weekly wages from the QCEW in 1983, 1984, and 1985, but in 1986 DOL reverted to the FLS to adjust the AEWRs because USDA had restarted the FLS survey.

DOL in 1987 made AEWRs the average hourly earnings of field and livestock workers as measured by the FLS in the previous year, which reduced AEWRs by 16 to 24 percent. Worker advocates sued. The DC Circuit Court of Appeals allowed DOL to use the new methodology that reduced AEWRs, but required DOL to explain why it reduced AEWRs.

DOL’s first justification for using the previous year’s average hourly earnings of field and livestock workers was rejected in April 1988, prompting DOL to re-issue the rule unchanged but with the argument that DOL could not measure any wage-depressing effects of legal guest or unauthorized workers in farm labor markets. Furthermore, with the enactment of IRCA in 1986, DOL argued that there would be fewer unauthorized workers to depress farm wages. The DC Circuit Court of Appeals upheld the methodology of using the previous year’s FLS field and livestock hourly earnings to set AEWRs for the current year, saying it represented DOL’s best effort to balance “providing an adequate supply of labor and protecting the jobs of domestic workers.”

After almost six decades, controversy continues over the AEWR methodology. On several occasions, as in 1980-81, 2008-09, and perhaps again in 2020-21, outgoing administrations made major changes to the AEWR that were reversed by incoming administrations.

DeLone, Sarah. 1992. Farmers, Growers and the Department of Labor: The Inequality of Balance in the Temporary Agricultural Worker Program. Yale Journal of Law and Liberation. Vol 3. No 1.

DOL. 2019. Temporary Agricultural Employment of H-2A Nonimmigrants in the United States. July 26.

DOL. 2020. Adverse Effect Wage Rate Methodology for the Temporary Employment of H-2A Nonimmigrants in Non-Range Occupations in the United States . November 5.

DOL certified 13,552 applications to employ 275,430 jobs with H-2A workers in FY20

Applications Received1
FY Q1 (Oct-Dec) Q2 (Jan-Mar) Q3 (Apr-Jun) Q4 (Jul-Sept) % Change FY 2019
14,131 3,545 6,129 2,681 1,776 8.0%
1 Includes all applications submitted for processing during the reporting period.
Data as of September 30, 2020. All figures are rounded and calculated based on data submitted by employers for processing.


Applications Processed
Determination FY Q1 Q2 Q3 Q4
Total Processed 14,063 2,111 6,820 3,095 2,037
  - Certified 13,552 1,991 6,654 2,964 1,943
  - Denied 247 62 73 67 45
  - Withdrawn 264 58 93 64 49
Positions Requested 286,900 43,178 100,248 88,136 55,338
Positions Certified 275,430 40,844 97,375 85,044 52,167
Processed Timely 1 296.8% 97.8% 96.8% 95.9% 96.0%
1 Percent of complete applications resolved no later than 30 days before the start date of need. A complete H-2A application is defined as one containing all the documentation (e.g., housing inspection report, workers’ compensation, recruitment report) necessary for OFLC to issue a final determination no later than 30 days before the start date of need.
Data as of September 30, 2020. All figures are rounded and calculated based on data submitted by employers for processing.


Review of Positions Certified FY 2020 EOY
(% of total certified FY 2020 EOY)
Top 10 States of Employment
Florida 39,064 14.2%
Georgia 27,614 10.0%
Washington 26,832 9.7%
California 25,453 9.2%
North Carolina 22,052 8.0%
Louisiana 11,332 4.1%
Michigan 9,912 3.6%
Arizona 8,602 3.1%
New York 8,482 3.1%
Kentucky 6,952 2.5%
Top 10 Occupations
(based on SOC Codes)
Farmworkers and Laborers, Crop 242,585 88.1%
Agri. Equip. Operators 15,396 5.6%
Farmworkers, Farm, Ranch, & Aqua. Animal 10,898 4.0%
Construction Laborers 3,256 1.2%
Agri. Workers, All Other 1,155 0.4%
Heavy and Tractor-Trailer Truck Drivers 577 0.2%
Graders and Sorters, Agri. Products 467 0.2%
First-Line Supervisors of Agri. Crop and Horticultural Workers 379 0.1%
Farm Labor Contractors 143 0.1%
Helpers – Production Workers 100 0.04%
Top 10 Employers 1
(based on unique FEINs)
North Carolina Grower’s Assoc., Inc. 10,650 3.9%
Fresh Harvest, Inc. 5,326 1.9%
WAFLA 4,358 1.6%
Foothill Packing, Inc. 3,199 1.2%
Farm Op Kuzzens H2A, LLC 2,998 1.1%
Rancho Nuevo Harvesting, Inc. 2,864 1.0%
Overlook Harvesting Company, LLC 2,589 0.9%
Templabor, LLC 2,252 0.8%
Zirkle Fruit Company 2,191 0.8%
R & R Harvesting, Inc. 2,056 0.8%
1 All calculations are aggregated based on the unique FEIN submitted by the employer. Multiple appearances of the same employer name may occur when associated with multiple FEINs.
Data as of September 30, 2020. All figures are rounded and calculated based on data submitted by employers for processing.


Martin, Philip. 2008. Evaluation of the H-2A Alien Certification Process and the US Farm Labor Market. KRA. DOL. ETA.

Subscribe via Email

Click here to subscribe to Rural Migration News via email.