Public Comments | Immigration

EPI comments on DOL’s proposed changes to the Adverse Effect Wage Rate methodology for H-2A visas for temporary migrant farmworkers

Submitted via

Brian Pasternak
Administrator, Office of Foreign Labor Certification (ETA)
U.S. Department of Labor
200 Constitution Avenue NW, Room N-5311
Washington, DC 20210

Re: Adverse Effect Wage Rate Methodology for the Temporary Employment of H-2A Nonimmigrants in Non-Range Occupations in the United States
RIN: 1205-AC05, DOL Docket No. ETA-ETA-2021-0006

Dear Mr. Pasternak:

The Economic Policy Institute submits this comment in response to the Department of Labor (DOL)’s Notice of Proposed Rulemaking on the Adverse Effect Wage Rate Methodology for the Temporary Employment of H-2A Nonimmigrants in Non-Range Occupations in the United States (the “Proposed Rule”).1 Despite its shortcomings, which are discussed below, the Proposed Rule would make meaningful improvements to wage protections in the H-2A temporary foreign agricultural worker program. These changes to the Adverse Effect Wage Rate (AEWR) methodology will broadly benefit both U.S. farmworkers and farmworkers coming into the country on H-2A visas.


The H-2A temporary work visa program has grown in popularity among farm employers, and as a result, over recent years, the size of the H-2A program has increased rapidly. In fiscal year 2021, the DOL’s Office of Foreign Labor Certification (OFLC) certified more than 317,000 H-2A positions, nearly double the number certified just five years prior. The range of duties performed by these workers has also broadened, with nearly 15% of positions allocated to roles outside of crop labor.

The H-2A workforce now represents 10 percent of the farm labor force nationwide working in crops, and over half of the jobs certified in 2021 were concentrated in the five states of Florida, Georgia, California, Washington, and North Carolina. The rapid expansion of H-2A has meant that new entrants to the farm workforce are now primarily workers from Mexico with H-2A visas.

Despite the rapid growth of H-2A, little has been done to improve the legal and regulatory framework in H-2A in a way that would better protect both migrant workers and U.S. workers seeking jobs in agriculture, improve wages and working conditions, or to curb the countless cases of abuse and exploitation of H-2A workers.

Although they are legally authorized to work, H-2A “guestworkers” or “temporary migrant workers” employed with these temporary visas are among the most exploited laborers in the U.S. workforce because the employment relationship created by the visa programs leaves workers powerless to defend and uphold their rights; this is certainly the case in H-2A, as reports from the media, advocacy groups, and government audit reports have pointed out time and time again.2

The abuses often start before H-2A workers even arrive in the United States—many are required to pay exorbitant fees to labor recruiters to secure U.S. employment opportunities, even though such fees are usually illegal. Those fees leave them indebted to recruiters or third-party lenders, which can result in a form of debt bondage. After arriving in the United States, H-2A workers may find out the job they were promised doesn’t exist. 3

U.S. workers seeking jobs in agriculture are also vulnerable to being discriminated against and kept out of farm jobs because employers sometimes prefer exploitable H-2A workers, as news reports and legal settlements have detailed.4 And government oversight of the H-2A program by DOL has been woefully inadequate: companies that are frequent and extreme violators of these rules are often allowed to continue hiring through H-2A and other temporary work visa programs with impunity.5

As a result, considering the many flaws in the H-2A program and its rapid growth—the need for effective worker protections in the H-2A program is greater than ever—and protections and labor standards for H-2A migrant workers and U.S. workers in agriculture must be improved. Therefore we commend the Department of Labor (DOL) for seeking to improve the H-2A’s Adverse Effect Wage Rate (AEWR). However, we suggest that DOL consider our comments and adjust the proposed rule to reflect the suggestions offered herein.


As EPI research has also demonstrated, farmworkers are among the lowest-paid workers in the entire U.S. workforce, even lower than other comparable low-wage workers. As Figure A below shows, farmworkers earned just $14.62 per hour on average in 2020, which is just 60%—or three fifths—of what production and nonsupervisory nonfarm workers earned ($24.67), who are the most appropriate cohort of workers outside of agriculture to compare with farmworkers.6

Figure A

The farmworker wage gap in 2020: Farmworkers earn very low wages compared with other workers: Average hourly wage rate for farmworkers and H-2A workers compared with average hourly wages of other workers, 2020

Type Amount
H-2A farmworkers $13.68
Nonsupervisory farmworkers $14.62
Workers with less than HS $15.06
Workers with HS diploma only $20.09
Nonsupervisory nonfarm $24.67
All workers $28.78
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: All values are for 2020 and in 2020 dollars. HS = high school. H-2A wage is the national average Adverse Effect Wage Rate for 2020, as reported by the U.S. Department of Labor, and does not reflect the average wage paid to the H-2A workers who were ultimately employed in 2020. Nonsupervisory nonfarm workers’ wage represents the average hourly earnings of production and nonsupervisory employees, total for the private sector, not seasonally adjusted. Nonsupervisory farmworkers’ wage is the gross average hourly wage of field and livestock workers. Data for all workers, and for workers with a high school diploma and less than high school, can be found at the Economic Policy Institute State of Working America Data Library. 

Notes: All values are for 2020 and in 2020 dollars. HS = high school. H-2A wage is the national average Adverse Effect Wage Rate for 2020, as reported by the U.S. Department of Labor, and does not reflect the average wage paid to the H-2A workers who were ultimately employed in 2020. Nonsupervisory nonfarm workers’ wage represents the average hourly earnings of production and nonsupervisory employees, total for the private sector, not seasonally adjusted. Nonsupervisory farmworkers’ wage is the gross average hourly wage of field and livestock workers. Data for all workers, and for workers with a high school diploma and less than high school, can be found at the Economic Policy Institute State of Working America Data Library. The H-2A visa program is described by the U.S. Department of Labor as a "temporary agricultural program [that] allows agricultural employers who anticipate a shortage of domestic workers to bring nonimmigrant foreign workers to the U.S. to perform agricultural labor or services of a temporary or seasonal nature."

Sources: Author’s analysis of USDA Farm Labor Survey data and nonfarm wage data from the BLS Current Employment Statistics survey; EPI analysis of CPI-ORG microdata; Office of Foreign Labor Certification historical state AEWRs

Author’s analysis of USDA, Economic Research Service using data from National Agricultural Statistics Service, Farm Labor Survey, and nonfarm wage data are from Bureau of Labor Statistics, Current Employment Statistics survey; EPI analysis of Current Population Survey Outgoing Rotation Group microdata (Economic Policy Institute, State of Working America Data Library); Office of Foreign Labor Certification, “Historical State AEWRs: Adverse Effect Wage Rate by State from 2014 to Present,” U.S. Department of Labor.

Copy the code below to embed this chart on your website.

Farmworkers have very low levels of educational attainment, and compared to the two groups of workers with the lowest levels of education in the United States, farmworkers earn even less on average: Figure A also shows that nonsupervisory farmworkers at $14.62 per hour earned 44 cents per hour less than the average wage earned by all workers without a high school diploma ($15.06), and farmworkers earned roughly $5 less per hour than the average wage earned by all workers with only a high school diploma ($20.09).

Many farmworkers employed through the H-2A visas program fared even worse in 2020. Figure A shows that the wage paid to most farmworkers with H-2A visas—known as the Adverse Effect Wage Rate (AEWR)—was even lower, with a national average of $13.68 per hour, according to DOL. But many H-2A farmworkers earned far less in some of the biggest H-2A states. In Florida and Georgia—where a quarter of all H-2A jobs were located in 2020—H-2A workers were paid the lowest state AEWR, at $11.71 per hour.

A previous analysis I coauthored also helps illustrate just how little farmworkers earn. Using data on consumer expenditures from the Bureau of Labor Statistics (BLS) about consumer spending on fresh fruits and vegetables, in conjunction with other data, we explored the question: How much would it cost to give farmworkers a significant raise in pay, even if it was paid for entirely by consumers? The answer is, not that much. A 40% increase in pay for farmworkers would cost just $25 per household in increased spending on fruits and vegetables per year.7 About the price of a couple of 12-packs of beer, a large pizza, or a nice bottle of wine.


As noted above and also reported by the U.S. Department of Agriculture’s (USDA) Economic Research Service (ERS), farmworkers earned just 60% of what production and nonsupervisory nonfarm workers earned, who are the most appropriate cohort of workers outside of agriculture to compare with farmworkers. In the proposed rule, DOL also cites another key data point that contextualizes farmworkers wages within the broader trends in the agricultural industry:

The ERS data also indicates that labor costs as a share of total gross farm income has not risen significantly over the past two decades, with the ERS concluding that “[a]lthough farm wages are rising in nominal and real terms, the impact of these rising costs on farmers” incomes has been offset by rising productivity and/or output prices,” and adding that “labor costs as a share of gross cash income do not show an upward trend for the industry as a whole over the past 20 years.”8 [emphasis added]

As the ERS data DOL has cited show, farms have become more productive and increased income at the same time that labor costs have risen, and thus labor costs for farmers have not risen as a share of income for the past 20 years. Data on farmworker wages and the share of labor costs disprove that the claim which is often made and repeated by farm employers and agribusiness lobbyists and representatives—i.e., that wages are rising too quickly for farmworkers and that the AEWR for H-2A workers is too high and rising too quickly, and thus not consistent with labor market trends. In fact, such claims are not credible and not based on any data or evidence.

Farmers also simultaneously claim that a labor shortage exists and that it is difficult to find agricultural workers, while expecting wages to remain the same and not rise in response to said shortage. As DOL rightly points out in the proposed rule, it is a rule of economics that wages rise in response to a labor shortage, and wages should “increase by an amount sufficient to attract more workers until supply and demand [are] met in equilibrium.”9 It is irrational to claim that there is a labor shortage in the farm labor market, but not expect wages to rise—and therefore unreasonable to ask DOL to use the AEWR to protect farmers from the natural operation of the free market. The AEWR is simply a tool that allows the wages of H-2A workers to reflect ongoing labor market trends in the United States, and is in place because DOL has a statutory mandate to prevent adverse effects to U.S. workers in its administration of the H-2A program.10

Finally, some employer groups are making unreasonable comparisons to support their argument that the AEWR wage is too high. For example the National Council of Agricultural Employers (NCAE), in their comment on the proposed rule, notes that H-2A worker wages are similar to starting salaries for teachers in Nebraska who hold a college degree:

A starting teacher in rural Hemingford, Nebraska, with a BS in education can expect an annual salary of $39,919. An H-2A worker or a domestic worker in corresponding employment with a 6th grade education working at the farm adjacent to the school, would receive an hourly AEWR rate of $16.47 or $34,258 on an annualized basis. The AEWR is making the case that maybe a high school or college education is not all that valuable, after all.11

The NCAE’s comparison is inappropriate on two fronts. First, it takes the Nebraska hourly wage rate and annualizes it for someone working 8 hours per day and 40 hours per week for an entire year (52 weeks) at the AEWR for Nebraska, $16.47 per hour—and suggests that if H-2A workers are paid so handsomely at that hourly rate, it diminishes the value of a college degree. Presumably, NCAE member must be aware that most farmworkers do not work 40 hours per week for an entire 52 weeks. We know this is true because, for example, there is a discrepancy between data in the USDA’s Census of Agriculture, which shows there are 2.4 million hired agricultural workers on farms, and the Quarterly Census of Employment and Wages (QCEW), which shows there are 1.6 million year-round full-time-equivalent (FTE) jobs in agriculture. Other research also shows that in California, the ratio of farmworkers to FTE jobs is two-to-one.12 Research I coauthored explains the large gap between FTE earnings and the actual earnings of farmworkers in more detail, showing that in 2015, workers in California who received their primary earnings from agricultural employers earned an average of $17,500—less than 60 percent of the average annual wage of a full-time equivalent (FTE) worker—and explains how employers and news reports often repeat this false narrative of farmworkers who earn well over $30,000 per year.13 When thinking about and analyzing the wages of farmworkers, it’s of the utmost importance to consider what they’re actually paid—not what they would earn if they worked full-time and year-round, since very few of them do.

The second flaw in the NCAE’s argument is related: the vast majority of H-2A workers, like U.S. farmworkers, do not work 40 hours per week for 52 weeks. In fact, DOL disclosure data show that the average duration of H-2A job certifications are 6 months.14 That means that the average H-2A farmworker in Nebraska is only likely to earn $17,129 during their time in the United States.

Finally, the teacher example is misleading because it purports to use teaching jobs as an example of a good-paying jobs that offers a decent middle-class life. Unfortunately, there is reliable evidence showing that teachers in the United States are woefully underpaid, and numerous examples in news reports of teachers who, for example, work three jobs and donate plasma to make ends meet.15 The underpayment of teachers around the country has led to many walkouts and strikes by teachers demanding better pay and working conditions in recent years. EPI data show that in Nebraska, teachers there see a weekly pay penalty of 17.7%.16 Thus, using a profession like teaching where workers have been undervalued and underpaid for years, and comparing them to farmworker wages to argue farmworkers are overpaid, is dishonest at best. What the NCAE’s example does instead is support the arguments of those advocating for better pay for teachers: If anything, their pay has eroded so far that it is now being compared to the pay of farmworkers, who earn some of the lowest wages in the entire U.S. workforce according to just about any metric.


The H-2A statute requires employers who would like to hire temporary migrant workers to obtain a labor certification from DOL stating that no U.S. workers are available for the position and that the employer is offering wages and working conditions that will not “not adversely affect the wages and working conditions of workers in the United States similarly employed.”17 These statutory protections are essential for U.S. farmworkers—both documented and undocumented—and their families because farmworker wages are much lower than similarly-situated workers in other sectors.18 The H-2A program’s adverse effect standard helps to ensure that agricultural employers respond to fundamental market incentives to improve wages and working conditions before turning to a foreign labor source.

The AEWR and other H-2A wage requirements function to protect all workers in agricultural workplaces. This includes H-2A workers themselves, who lack bargaining power because of the structure of the H-2A visa program. H-2A workers can work only for those employers who sponsored their visa applications, meaning that many workers perceive a binary choice between accepting whatever wages their employers offer and losing the opportunity to work in the United States. And before they even get to the United States, migrants are often charged illegal fees just to access temporary H-2A jobs, leaving them in debt, and they’ve also been the victims of fraud and even human trafficking as a result of the H-2A the recruitment process.19

H-2A workers who might try and ask for higher wages are often met with employer retorts about the low cost of living and high poverty rates in their home countries—or worse, with threats to have the workers deported or blacklisted. And because workers understand that retaliation is commonplace, even if it is illegal, many simply accept underpayment as the cost of working in the United States. In fact, surveys of H-2A workers show that legal violations are commonplace, revealing how little power H-2A workers have in the workplace.20 Indeed, the total control that employers can exercise over H-2A workers due to their precarious legal status leads many hiring employers to illegally discriminate against U.S. workers by preferring to hire H-2A workers.

For years, the AEWR has played a central role in fulfilling DOL’s mandate to prevent adverse effect to U.S. workers, and we support the Department’s decision to continue this role. The AEWR requirement appropriately “reflects a longstanding concern that there is a potential for the entry of foreign workers to depress the wages and working conditions of domestic agricultural workers.”21 As the Department explains in the Proposed Rule, “[t]he AEWR acts as ‘a prevailing wage concept defined over a broader geographic or occupational field,’” thereby “protect[ing] against localized wage depression that might occur in prevailing wage rates.”22 The potential for wage depression in the H-2A context is heightened because the program is uncapped, giving employers “access to a potentially unlimited number of foreign workers.”23

The H-2A program, however, contradicts DOL’s description of the AEWR as a “floor below which wages cannot be negotiated, thereby strengthening the ability of this particularly vulnerable labor force to negotiate over wages.”24 We agree that, in theory, the AEWR should serve as a wage floor. Unfortunately, the AEWR in practice acts as a wage ceiling. DOL has long accepted the interpretation that U.S. workers who demand more than the AEWR are not “available” within the meaning of 8 U.S.C. § 1188(a)(1)(A).25 It must be remembered that in practice, the AEWR is in fact a floor, and that growers must reasonably negotiate with both U.S. and H-2A workers who request wages above the AEWR.

In addition, we note that the AEWR is merely one piece of a larger set of wage protections in the H-2A program. Employers seeking to use the H-2A program must pay their workers the highest of four different wage sources: (1) the state or federal minimum wage; (2) the agreed-upon collective bargaining rate; (3) the AEWR; or (4) the applicable prevailing wage.26 This four-pronged wage structure is carefully calibrated, with each piece designed to protect U.S. workers in a different way. Unfortunately, the current prevailing wage system is broken: most states have not had any prevailing wage findings in years.

As a result, it is difficult to assess the impact of this AEWR rule independent of broader changes to the H-2A wage protections. Any impending changes to the prevailing wage system are particularly relevant to this rulemaking in light of the Proposed Rule’s effort to ensure market rate wages for specific occupations that normally receive higher compensation than crop farming. The prevailing wage system has historically provided protection for more specific job duties, but the lack of prevailing wage findings in most states prevents it from doing so today. Given the need for reforms in the prevailing wage system, we are disappointed that the Department has chosen to limit the scope of this rulemaking, but we nonetheless welcome the significant improvements in the Proposed Rule


The Proposed Rule makes two key changes to the AEWR methodology that will benefit workers. First, it revives the Department’s longstanding reliance on the wage findings of the USDA’s Farm Labor Survey (FLS) to set the AEWR for field and livestock positions. Second, it recognizes the growth of non-field and non-livestock positions in the H-2A program and seeks to ensure that U.S. and H-2A workers in those positions are paid market-rate wages by relying on the Bureau of Labor Statistics Occupational Employment and Wage Statistics (OEWS) data.

We strongly support DOL’s decision to continue setting the AEWR for field and livestock workers at the wage rate determined by the FLS. For nearly 35 years, DOL has relied on the FLS wage findings to set the AEWR. The FLS is an annual survey of farm employers that reports a combined average wage for the six primary field and livestock occupations at the state, regional, and national level. As the Department has repeatedly recognized, the FLS is the best source of accurate data on most farmworkers’ wages.27 However, the previous Administration departed from this longstanding approach in November 2020 when it issued its Final Rule on the AEWR methodology. The 2020 AEWR Final Rule would have abandoned the FLS, frozen worker wages for two years, and then adjusted upward annually based on changes in the Employment Cost Index. Such a drastic change of approach would have been disastrous for workers, and the Department was rightly enjoined from enforcing the 2020 AEWR Rule.

We also support the use of OEWS wage findings to set the AEWR for any H-2A job opportunity that does not fall within the six occupations covered by the FLS’s combined field and livestock category—but offer caveats and suggestion about the OEWS data set in the following section.


While we support the proposed continued general rule of requiring H-2A employers to pay the highest of the Adverse Effect Wage Rate (AEWR), the prevailing wage, wage established in a collective bargaining agreement, state minimum wage, or federal minimum wage, the terms of the proposed rule should be modified slightly with respect to the occupations that will be set by DOL’s Occupational Employment Statistics (OEWS) data set.

When wage data for a specific, narrower occupation is unavailable in the FLS to set the AEWR, DOL proposes to use the statewide average wage for the occupation that is reported the OEWS survey for the occupation according to the corresponding Standard Occupational Classification (SOC) code. However, wage rates for the same occupation can vary widely in a state, especially in a large one like California, which has counties with very high living costs (e.g., Monterey) and some with less expensive living costs (e.g., many of the counties in the San Joaquin Valley). The OWES also reports wage survey data by local area, which are reported as Metropolitan Statistical Areas (MSAs), counties, or nonmetropolitan areas. Therefore, using the statewide OES wage to set the AEWR and ignoring local wage rates available from the OEWS makes little sense.

The local wage rates are easily accessible on Foreign Labor Certification Data Center’s website, which is public and free.28 The FLC Data Center’s online wage library hosts wage rates surveyed by the OEWS and makes them available by local area and SOC code, and lists the average for the area and occupation. Searching for the local OEWS wage for the occupation will not cause any additional burden on employers than searching for the statewide OEWS wage rate for the occupation would cause. In fact, most prevailing wage rates in two other separate temporary work visa programs managed by DOL—the H-2B and H-1B visas—are set by occupation and local area through use of the FLC Data Center website.

While DOL has proposed to make the AEWR methodology more reflective of occupations with higher wage rates that are not reflected in the FLS, it should go one step further and require payment of the highest wage available to set the AEWR from among all of the data sources available which include the state or regional FLS wage, the statewide OEWS wage for the occupation, or the local OEWS wage for the occupation (rather than picking the statewide OES in every case and ignoring the local OES wage). This will help ensure that farmworkers in counties with higher wages will not be at a disadvantage, and will help prevent downward pressure on local wages that would be caused by farmworkers in higher-wage areas being paid lower wage rates that reflect the statewide average.

In sum, in order to prevent downward pressure on farm wages, the updated prevailing wage methodology should be the highest among: (1) the local prevailing wage rate established by a SWA; (2) the wage established in a collective bargaining agreement; (3) the state minimum wage; (4) the federal minimum wage; (5) the state or regional wage reported by FLS; but when an OEWS wage applies, it should be (6) the local wage reported in the OEWS for the occupation; or (7) the statewide wage reported in the OES for the occupation. 


While the FLS is currently by far the best available data set for understanding wage trends in agriculture and therefor for setting the AEWR, we realize that it is imperfect for the reasons DOL has outlined. However, while we support DOL’s decision to supplement the FLS with data from the OEWS, we nevertheless do not believe that the OEWS is an adequate substitute for the FLS because of its flaws with respect to data collection in the agricultural industry. DOL expresses concern that a future administration could decide to terminate the FLS—a realistic concern we share considering the previous administration’s attempt to do just that. Given such a possibility, we suggest that DOL begin to significantly invest in and improve the OEWS, but warn that without such improvements, using OEWS data for the majority of H-2A jobs instead of the FLS would artificially distort the earnings of farmworkers and lead to much lower AEWRs.

In a previous comment to DOL which I coauthored about H-1B wage levels, I discussed some of the shortcomings of the OEWS data set and the need for improvement.29 Some of those same concerns apply here, but there are added concerns with respect to agricultural wages.

The strength of the OEWS survey is that it has data on nearly 800 SOC occupations, and for all regions in the United States. It’s the only wage data source we know of that does so, and another benefit is that each year DOL calculates the mean and median wage for each occupation, as well as the 10th, 25th, 50th (i.e. median), 75th, and 90th percentile wages for each occupation at the national level—and the mean and median at the state level—publishing all of it on the BLS website in an easy to read and understand format.

However, the OEWS is far from perfect. The OEWS covers so many occupations and regions, that often the sample sizes are quite small, diminishing the validity of the result. And when employers fill out the OEWS survey form, rather than writing in the actual salaries of their employees, employers choose from a range of salaries (hourly and/or annually), which diminishes the accuracy of the results.30 And finally, the BLS conducts the OEWS survey every year but pools three years of data for its results, which as BLS notes, means there are “limitations associated with this estimation procedure in that it requires ‘updating’ for the earlier years of data and limits the usefulness of OEWS data for time series analysis.”31

We would urge the DOL conduct an audit of the entire OEWS data set to ensure geographic and occupational sample sizes are sufficient to return valid wage data. There are many examples of OEWS-based wage data for high-skilled H-1B occupations on the website utilized by OFLC to set H-1B wages—the Foreign Labor Certification (FLC) Data Center website—that return implausibly low wages. For example, Level 1 wages for a Software Developers, Applications job (SOC 15-1132, retrieved from the 07/2020 – 06/2021 All Industries Database) in Northeast Minnesota are $9.87 per hour, 32 which is such a low wage that it is even less than even the Minnesota state minimum wage of $10.33 per hour.33 Similarly the Level 1 wage for Electrical Engineers (SOC 17-2071) is $10.43 for College Station Texas and the Level 4 wage is a meager $22.76.34 The Level 4 wage for Electrical Engineers reported by OFLC for College Station, Texas is less than half the national average hourly wage of $50.96 for SOC code 17-2071.35 Clearly there are significant methodological problems with the data being reported. We suggest that DOL identify reported FLC wages that are far below the national average and investigate such instances further to validate the data.

When it comes to agricultural wages, the OEWS is especially deficient in terms of reflecting the wages of farmworkers because it is limited to non-farm employers, which in the agricultural sector generally refers to farm labor contractors (FLCs). Farms increasingly rely on FLCs for certain kinds of work, such as construction or equipment operation,36 but the majority of field and livestock workers nationally are still directly hired by farms rather than FLCs, as both the National Agricultural Workers Survey and QCEW data show.37

OEWS data relying on only on wages paid by FLCs is a problem because of the FLC business model and the foreseeable results. The FLC model of employment may increase the incidence of employment law violations by separating the main beneficiary of the labor provided by farmworkers—the farm operator or “lead” employer—from the farmworkers who perform the work. Farms that rely on FLCs are a textbook example of what David Weil has called a “fissured” workplace,38 where the relationship between the worker and the lead employer is fissured, or broken, via the use of a temp agency or subcontractor (in this case the FLC). Unsurprisingly, fissuring often results in lower wages for workers,39 in part because the subcontractor (the FLC) keeps a percentage of the wages earned by the workers, and farm operators do not provide the farmworkers who work on their farms with fringe benefits because they are employees of the FLC.

The results of this flawed business model are clear: FLC wages for many field and livestock workers are lower than those paid by farms that directly hire. DOL’s own studies have indicated that farm labor contractors and other employers in the support services field, on average, paid wages that were 14% lower than those paid by farm operators,40 and QCEW data also show lower wages paid by FLCs compared to direct-hire workers.41 FLCs often operate on incredibly thin margins and are notorious for labor and employment violations.42 In fact, a 2020 study I coauthored found that FLCs accounted for 24% of all wage violations investigated by the Wage and Hour Division in the agricultural sector from 2005 to 2019, despite making up only 14% of overall agricultural employment.43

Thus, until and unless DOL makes significant investments in, and improvements to the OEWS data, they will continue to be inadequate as a substitute for the FLS if it is ever discontinued. The OEWS’s reliance on wage data collected exclusively by FLCs with a fissured business model and lower wages will artificially depress the AEWR. Therefore, DOL should take steps to improve the collection of farmworker wage and earnings data under the OEWS; for example, by expanding the population surveyed by the OEWS to include direct-hire farm employers. It should also generally improve the OEWS to improve its annual results as outlined above.


We commend the Department for its efforts to improve the AEWR methodology and strengthen its protections for U.S. and H-2A farmworkers. The Proposed Rule still needs improvement as discussed in this comment, but it reverses harmful changes sought by the previous Administration and effectively responds to the dynamics of the growing H-2A program.

Nevertheless, it is worth mentioning that the only rational and durable solution for stabilizing the farm labor force is for Congress to craft solutions that do not rely on temporary immigration statuses and indentured temporary migrant workers like H-2A workers, and that pay farmworkers a living wage, as well as keep them protected in the fields and allow them to become permanent members of American society. Congress should do this by, first and foremost, passing a law that provides a path to citizenship for the current members of the farm labor force that lack an immigration status. Next, Congress should limit the time that any future H-2A workers employed in agriculture may remain in a temporary status by allowing those workers to self-petition for permanent residence after a provisional period working in H-2A or another temporary status. In addition, Congress should appropriate more funding to DOL to audit a significant share of H-2A employers every year to ensure compliance with regulations and job contract terms, and pass a law that permanently bans any employer from hiring through H-2A or any other temporary work visa programs if that employer has violated any labor or employment laws.


Daniel Costa

Director of Immigration Law and Policy Research
Economic Policy Institute



1. 86 Fed. Reg. 68174 (Dec. 1, 2021).

2. See for example, Centro de los Derechos del Migrante, Ripe for Reform: Abuses of Agricultural Workers in the H-2A Visa Program, April 2020; Jessica Garrison, Ken Bensinger, and Jeremy Singer-Vine, “The New American Slavery,” BuzzFeed News, July 24, 2015; Southern Poverty Law Center, Close to Slavery: Guestworker Programs in the United States, February 2013; U.S. Government Accountability Office, H-2A and H-2B Visa Programs: Increased Protections Needed for Foreign Workers, GAO-15-154, reissued May 2017.

3. See for example, Centro de los Derechos del Migrante, Recruitment Revealed: Fundamental Flaws in the H-2 Temporary Worker Program and Recommendations for Change, n.d.; United Nations Office of the High Commissioner for Human Rights, “End of Visit Statement, United States of America (6–16 December 2016) by Maria Grazia Giammarinaro, UN Special Rapporteur in Trafficking in Persons, especially Women and Children,” Washington, D.C., December 19, 2016. See also United Nations Office of the High Commisioner for Human Rights, “Debt Bondage Remains the Most Prevalent Form of Forced Labour Worldwide—New UN Report” (press release), September 15, 2016; United Nations Human Rights Council, Report of the Special Rapporteur on Contemporary Forms of Slavery, Including Its Causes and Consequences, Thirty-third session, Agenda item 3, Promotion and protection of all human rights, civil, political, economic, social and cultural rights, including the right to development, July 4, 2016; United Nations Office on Drugs and Crime, The Role of Recruitment Fees and Abusive and Fraudulent Practices of Recruitment Agencies in Trafficking in Persons, 2015.

4. See for example, Jessica Garrison, Ken Bensinger, and Jeremy Singer-Vine, “‘All You Americans Are Fired,’” BuzzFeed News, December 1, 2015; U.S. Department of Justice, “Justice Department Settles Claim Against Florida Strawberry Farm for Discriminating Against U.S. Workers” (press release), June 11, 2019.

5. See for example, Ken Bensinger, Jessica Garrison, and Jeremy Singer-Vine, “Employers Abuse Foreign Workers. U.S. Says, by All Means, Hire More,” BuzzFeed News, May 12, 2016.

6. Daniel Costa, “The farmworker wage gap continued in 2020: Farmworkers and H-2A workers earned very low wages during the pandemic, even compared with other low-wage workers,” Working Economics blog (Economic Policy Institute), July 20, 2021. For previous years comparing farmworkers with production and nonsupervisory nonfarm workers, see Economic Research Service, “Wages of Hired Farmworkers” in “Farm Labor,” U.S. Department of Agriculture, last updated August 18, 2021.

7. Daniel Costa and Philip Martin, “How much would it cost consumers to give farmworkers a significant raise? A 40% increase in pay would cost just $25 per household,” Working Economics blog (Economic Policy Institute), October 15, 2020.

8. 86 Fed. Reg. 68185, citing Economic Research Service, Farm Labor, U.S. Department of Agriculture, last modified Aug. 18, 2021, at footnote 70.

9. 86 Fed. Reg. 68185, citing 80 Fed. Reg 24146, 24158-24159 at footnote 73.

10. 8 U.S.C §1188.

11. Comment submitted to the proposed rule by Robert Roy, President and General Counsel for the National Council of Agricultural Employers, January 25, 2022.

12. Philip Martin, Brandon Hooker, Muhammad Akhtar, Marc Stockton, How many workers are employed in California agriculture? California Agriculture, Volume 71, number 1, pages 30-34, August 23, 2016.

13. Philip Martin and Daniel Costa, “Farmworker wages in California: Large gap between full-time equivalent and actual earnings,” Working Economics blog (Economic Policy Institute), March 21, 2017.

14. See for example Philip Martin, “The H-2A farm guestworker program is expanding rapidly: Here are the numbers you need to know,” Working Economics blog (Economic Policy Institute), April 13, 2017.

15. Katie Reilly, “‘I work 3 jobs and donate blood plasma to pay the bills.’ This is what it’s like to be a teacher in America,” Time, September 13, 2018.

16. Sylvia A. Allegretto and Lawrence Mishel, Teacher pay penalty dips but persists in 2019: Public school teachers earn about 20% less in weekly wages than nonteacher college graduates, Economic Policy Institute, September 17, 2020.

17. 8 U.S.C. § 1188(a)(1).

18. Daniel Costa, “The farmworker wage gap continued in 2020: Farmworkers and H-2A workers earned very low wages during the pandemic, even compared with other low-wage workers,” Working Economics blog (Economic Policy Institute), July 20, 2021.

19. See for example, Farmworker Justice, No Way to Treat a Guest, 2011; Daniel Costa, Temporary work visa programs and the need for reform: A briefing on program frameworks, policy issues and fixes, and the impact of COVID-19, Economic Policy Institute, February 3, 2021; Centro de los Derechos del Migrante, Recruitment Revealed: Fundamental Flaws in the H-2 Temporary Worker Program and Recommendations for Change, n.d., accessed December 10, 2020; Mary Bauer and Meredith Stewart, Close to Slavery: Guestworker Programs in the United States, Southern Poverty Law Center, Feb. 19, 2013; Centro de los Derechos del Migrante, Ripe for Reform: Abuses of Agricultural Workers in the H-2A Visa Program, April 2020.

20. Centro de los Derechos del Migrante, Ripe for Reform: Abuses of Agricultural Workers in the H-2A Visa Program, April 2020.

21. 2010 Final Rule, 75 Fed. Reg. 6883, 6891 (Feb. 12, 2010).

22. Proposed Rule, 86 Fed. Reg. at 68176 (quoting 75 Fed. Reg. at 6893).

23. Id.

24. Id.

25. See Hernandez Flecha v. Quiros, 567 F.2d 1154, 1156 (1st Cir. 1977).

26. 20 C.F.R. § 655.122(l).

27. Proposed Rule, 86 Fed. Reg. at 68178; 2010 Rule, 75 Fed. Reg. at 6898.


29. Daniel Costa and Ron Hira, “EPI comments on DOL Request for Information on determining prevailing wage levels for H-1B visas and permanent labor certifications for green cards,” Economic Policy Institute, public comment submitted for Request for Information on Data Sources and Methods for Determining Prevailing Wage Levels for the Temporary and Permanent Employment of Certain Immigrants and Non-Immigrants in the United States, Request for Information, DOL Docket No. ETA-2021-0003, RIN: 1205-AC00, June 1, 2021.

30. Bureau of Labor Statistics, OES questionnaire form, U.S. Department of Labor, Form Approved O.M.B. No. 1220-0042 Rev. March 2020.

31. See Section A, Question 4 in Bureau of Labor Statistics, U.S. Department of Labor, Occupational Employment and Wage Statistics, Frequently Asked Questions.

32. Retrieved May 30, 2021:

33. Minnesota Department of Labor and Industry, “Minimum Wage Increases Jan. 1, 2022,” August 19, 2021.

34. Retrieved May 30, 2021:

35. Occupational Employment and Wages, May 2020, 17-2071 Electrical Engineers, retrieved May 30, 2021.

36. Proposed Rule, 86 FR at 68181

37. The latest National Agricultural Workers Survey (which does not include H-2A workers) showed that in 2017-18, 89% of farmworkers worked directly for a farm employer. QCEW data also shows that direct hire crop employment is higher than FLC employment; see Rural Migration News, “Farm Labor: Employment Down, Wages Up,” University of California, Davis, June 21, 2021.

38. David Weil, The Fissured Workplace: Why Work Became So Bad for So Many and What Can Be Done to Improve It. Cambridge, Mass.: Harvard University Press, 2014.

39. A number of studies show a wage penalty for subcontracted/outsourced workers. For example, see Arindrajit Dube and Ethan Kaplan, “Does Outsourcing Reduce Wages in the Low-Wage Service Occupations? Evidence from Janitors and Guards.” Cornell University ILR Review, Cornell University, January 1, 2010; Deborah Goldschmidt and Johannes Schmieder, “The Rise of Domestic Outsourcing and the Evolution of the German Wage Structure.” The Quarterly Journal of Economics, Oxford University Press, vol. 132(3), 1165–1217, 2017; and Andres Drenik, Simon Jäger, Pascuel Plotkin, and Benjamin Schoefer, Paying Outsourced Labor: Direct Evidence from Linked Temp Agency-Worker-Client Data.Econometrics Laboratory, University of California, Berkeley, September 2020.

40. 75 Fed. Reg. 6883, 6898 (Feb. 12, 2010).

41. Rural Migration News, “Farm Labor: Employment Down, Wages Up,” University of California, Davis, June 21, 2021.

42. Maria Perez, “What led to a migrant worker’s death from heatstroke?” USA Today, December 17, 2021.

43. Daniel Costa, Philip Martin, and Zachariah Rutledge, Federal labor standards enforcement in agriculture: Data reveal the biggest violators and raise new questions about how to improve and target efforts to protect farmworkers, Economic Policy Institute, December 15, 2020.

See related work on Immigration

See more work by Daniel Costa