ETHNIC SHIFT IN EASTERN CROP AGRICULTURE:
Replacement or Displacement?
by Rick Mines
I am going to describe a model for how agricultural employers shift ethnic groups. I'll support this model by macro and micro data, including a case study I did myself. And, I'll try to explain why employers follow the model. In my view, a model for understanding employer behavior, facilitates policy discussion.
There has been a marked tendency for crop employers on the East Coast to favor the progressive substitution of Mexican and Guatemalan-born labor for other groups. We do not know whether this process is due to replacement or displacement of U.S. workers. However, there is considerable evidence that the process is occurring.
Do the employers intentionally or without knowing displace US workers who want to remain in farm jobs or do US workers voluntarily depart leaving job opportunities for arriving foreigners? Below, I present both macro-level and case study data which may allow you to come to your own conclusion.
Table 1: Ethnic Shifts in California Agriculture (1966-1995)
California Assembly UC EDD Survey California NAWS
Year 1965 1983 1994-1995
White US Born 43.90% 4.50% 1.10%
Hispanic US Born 16.50% 4%
Hispanic Mexican Born 71.30% 93.50%
Asians & Native Americans US Born 6.80% 6.80% 0%
Blacks 3.30% 0.90% 0.10%
This process began in the West and Texas years ago but got a major push from the Bracero program and a huge illegal influx of the 1942-1964 period. It is difficult to demonstrate this ethnic shift over time from data sources. However, for California surveys done in 1966, 1983 and the current National Agricultural Workers Survey (NAWS) allow for some comparisons. Although the ethnic identification questions were asked somewhat differently in the three surveys, we can see a definite shift toward Hispanic immigrants in California over the years. In 1966, Hispanics were still a minority. By 1983 the foreign born Hispanics were already
over 70 percent and by the middle 1990s well over 90 percent of crop workers were foreign born Hispanics. (See Table 1)
This ethnic shift later spread through networks of legal and undocumented workers to the rest of the country in the 1960s through the 1990s. As we will see below the shift to certain parts of the East Coast is quite recent and continues today.
Table 2: The Sequencing of the Substitution of Mexican and Guatemalan Labor for U.S. Labor
Order of Incorporation of Foreign Labor
pick, pull, cut
hoe, thin, tie, prop, transplant
prune, spray, irrigate, grove maintenance
pack, haul, ship
recruit, monitor, check, transport
How does this process work? First, employers find it convenient or preferable to shift to foreigners in the tedious harvest tasks like picking fruit, vegetables and pulling tobacco. Then, some time later employers make the same ethnic shift in the pre-harvest tasks like thinning, hoeing, and transplanting. Also, during this same period growers shift their semi-skilled longer-term tasks from domestics to foreigners. These include pruning, spraying, irrigating and soil and grove maintenance. A further shift occurs in the post-harvest tasks of packing and shipping. (This tends to have the most domestic female, often family, labor). Finally, some employers turn to the foreign born for their supervisory jobs. The shifts, of course, occur differently for each crop as a result of many local
conditions including availability of local labor, difficulty of the tasks, and access of the employer to foreign workers.
We have corroboration of this process from the National Agricultural Workers Survey (the NAWS). The NAWS is a random sample survey done since 1988 on a national basis. Each year it interviews about 2,500 crop farmworkers in three cycles during the year. I use 1989-1995 data for this paper.
I divided the country into two regions. (1) The eastern US which includes everything east of the Mississippi, except Florida. (2) And the rest of the country. These two regions very roughly encompass the areas of traditional foreign incorporation into the farm labor market (the West) and the area of more recent incorporation (the East).
Table 3: All Years Together-Percent Distribution of Ethnic Groups by Eastern and Western US (1989-1995)
White Other US Born Hispanic Born Mexican Other Latin Americans Other Foreign Born Total
Western 6.55 2.9 10.17 73.43 3.55 3.41 53.49
Eastern US 38.37 3.36 12.74 42.0 2.5 1.04 46.51
Total 21.35 3.11 11.37 58.81 3.06 2.31 100
* West includes Florida
The first point is that over the whole period most of the native born farmworkers are in the eastern region. We can see that although most farmworkers are foreign born in both regions--most of the whites and other US born are in the East. Eighty four percent of the whites and over half of the blacks and Hispanic Americans are in this eastern region. By contrast, the majority of the foreign born (2/3rds) are in the West or Florida.
Table 4: Nationwide: Place of Birth of Farmworkers (1989-1995)
Place of Birth for All Farmworkers
Foreign Born US Born
1989 61.02 38.79
1990 59.19 40.78
1991 58.51 41.35
1992 62.45 37.02
1993 68.76 31.22
1994 68.55 31.4
1995 70.24 29.76
The second point is that nationally the proportion of foreign born among the farmworker population is increasing over time from about 60% in 1989 to 70% today. (See table 4) However, this increase in the proportion of foreign born is entirely due to a shift in the relative share of foreign born in the eastern section. (Table 5) Meanwhile the proportion in the western United States and Florida did not change in recent years because it was already highly foreign born by 1989. (Table 6)
Table 5: East: Place of Birth of Farmworkers (1989-1995)
Place of Birth for East Farmworkers
Foreign Born US BORN
1989 39.5 60.5
1990 38.07 61.93
1991 25.93 74.06
1992 39.68 60.32
1993 56.94 43.06
1994 57.17 42.83
1995 58.59 41.41
Total 45.54 54.47
Table 6: West: Place of Birth of Farmworkers (1989-1995)
Place of Birth of Farmworkers for Western US and Florida
Foreign Born All US Born
1989 83.34 16.66
1990 82.03 17.97
1991 78.38 21.61
1992 82.09 17.91
1993 81.12 18.88
1994 77.13 22.87
1995 79.35 20.64
Total 80.38 19.62
In the eastern US, the proportion of foreign born increased from about 40% in the early years to 60% in the later years. In the West, it stayed at about 80% throughout. Furthermore, the vast majority of this increase in the foreign born is due to an increase in Mexicans. The proportion of this group increased from about 35% to 50% during this period in the Eastern U.S.
The way the increase occurred supports the hypothesis that employers progressively shifted by task to new ethnic groups. In fact, the data show that the increase in the proportion of foreign born Mexicans and Guatemalans occurred principally in certain tasks. The data show that employers during the 1990s in the East shifted to the foreign born for non-Harvest workers. Since the tasks of all types in the West were already dominated by the foreign born at the beginning of the period we do not observe a shift there. Virtually all (70 percent or more) were foreign born. (See Table 8) In addition, in the East, where the overall proportion
of the foreign born increased, the increase of the foreign born was not particularly into the harvest tasks during this period since they already had the large majority (over 70%) even at the beginning. (Table 7)
In the East, the foreign born did become more common in first the preharvest and the semiskilled jobs and eventually in the post-harvest jobs in the 1990s. (Table 7) These shifts correspond closely to the model suggested above. (See Table 2) Namely, the employers shift first the heavy harvest tasks, next the preharvest and semiskilled and lastly the postharvest jobs are given to the foreign born.
Table 7: East:Percent of Foreign Born Farmworkers in EAST by Task
Foreign Born in the East
Harvest PreHarvest Post Harvest Semi-Skilled
FY 1990-1991 72.36 10.94 3.95 4.2
FY 1992-1993 68.31 53.28 11.89 45.17
FY1 1994-1995 73.23 46.86 46.32 52.87
Table 8: West:Percent of Foreign Born Farmworkers in WEST by Task
Foreign Born Farmworkers by Task West US
Harvest PreHarvest Post Harvest Semi-Skilled
FY 1990-1991 83.53 70.18 75.51 83.26
FY 1992-1993 89.22 69.64 68.39 79.13
FY1 1994-1995 88.67 67.75 67.23 79.61
The case studies done by anthropologists and sociologists over the years in the Eastern United States support the model of ethnic shift by task. Social scientists have chronicled the shift to Mexicans and Guatemalans from other ethnic groups. (Friedland, Nelkin, Heppel, Kissam, Griffith, Amendola)
The original groups in the postwar period East of the Mississippi in general were local whites and blacks, black migrants from Florida, Puerto Rican migrants from the island, and more recently Haitians and Southeast Asians. The recent trend according to this case study work seems to be inexorably toward the Latin Americans.
The sociologists have shown that the networks tend to behave similarly across ethnic groups.
1. During times of labor surplus when employers could be particularly choosey, networks of solo males tended to be more important. During times of relative scarcity, more families could find work. This was true of African American, Haitian, Puerto Rican and other Latin American groups.
2. Also, these case study researchers have shown that the labor markets have been segmented with different ethnic groups predominating in different segments at different periods.
Below, I will give 4 examples from the recent case study literature on the Eastern U.S. farm labor market.
1. Delmarva vegetables, potatoes, tomatoes
In this area, small farms have for some years hired locals for long term work and themselves organized pre-harvest and post-harvest activities. At the same time, many of them hired farm labor contractors, or multi-state firms to organize the harvest using migrant labor. This migrant labor originally was African American (mostly from Florida) but has shifted to mostly Latin American immigrant labor. (See Kissam and Griffith)
2. Southern Michigan asparagus, strawberries, blueberries, pickling cucumbers and apples
Again a two-tiered system was in place at least by 1930s. Small farmers have put together several jobs for long term workers. Some of these workers enter into sharecropping arrangements with employers using their whole families to fulfill their contracts. Again, farm labor contractors have brought in peak season migrants. In the early years, the migrants were southern blacks and whites, later they shifted to Tex- Mex families and Legal Resident Mexicans, and finally a shift has begun to solo male Mexicans based in Florida and Mexico. (Kissam and Griffith)
3. New Jersey Nurseries
The Glassboro labor exchange in the heart of the vegetable and nursery region of Southern New Jersey used to hire 12,000 (in the 1950s and 1960s) Puerto Ricans from the island each year, but now the agency is inactive. The nursery jobs were being rationed in the early 1990s with the better core jobs going to the remaining Puerto Ricans and the less desirable jobs to the Mexicans. (Pfeffer).
4. South Carolina and Georgia Peaches
Local blacks in Georgia, and Florida migrant blacks in South Carolina did the harvest tasks in the postwar period. However, with the mid-1980s, the shift to foreign born Hispanics in the harvest tasks took off; a coincidence of higher yields and the availability of workers due to the 1983-4 freeze in Florida made the shift permanent. As of the early 1990s, the African Americans had the semi-skilled and pre and post- harvest jobs the Mexicans the harvest tasks but the markets for the nonharvest jobs were shifting. (Amendola)
In all four cases, the segment where the locals continued to hold sway were precisely the pre, post and semi-skilled tasks while the newcomer immigrants had come to dominate in the more arduous tasks. However, when these case studies were done at the beginning of the 1990s, there was already a noted tendency for some of the immigrants to settle and locate work in the sectors heretofore dominated by U.S. labor. Our macro data cited above tend to confirm that the shift of the foreign born into the US worker dominated tasks continued unabated in the 1990s.
NORTH CAROLINA AND VIRGINIA PIEDMONT TOBACCO
Now that you have a model for what has occurred along the East Coast, I'd like to give you one in-depth example. This is the case of Tobacco workers in the piedmont area of North Carolina and Virginia. I believe this example will help form an opinion about whether this process is displacement or replacement.
The data I will share with you was gathered over 10 years ago when I worked for the GAO. I spoke to dozens of participants along the border of Virginia and North Carolina. Including growers, association presidents, supervisors, recruiters, H2 workers and illegal Mexican workers. According to the participants this is what happened.
According to the growers in 1986, they were having a hard time making a profit. Inexpensive labor was crucial to their success since about half of production costs went to labor. The revenue from a pound of tobacco was $1.50/lb, while costs were $1.20 not counting the 30 cents rent for the quota. For example if a grower raised the wages he paid to workers form $4 to $6, he would increase his gross costs by approximately 16%. According to the growers, this meant the difference between profit and loss.
The growers were convinced they needed the Mexicans. I asked one grower who owned $70,000 of tobacco quota what would happen if the Mexican stop coming. He said, "I would lose $70,000 overnight."
The area has mostly small growers and many are not highly educated. Tobacco growing goes back hundreds of years in the area. First, the growers used slaves, some of the poor whites were sharecroppers. Then, the sharecrop system became the norm with blacks doing most of the work. There was a period in the postwar period when U.S. migrants, locals, and students did the heavy work. And, since 1972, Mexicans have done most of the heavy work. (See Table 9)
Table 9: Periodization of the Different Labor Regimes in VA
1700-1865 small growers with slaves and some white sharecropping
1865-1965 sharecropping about 60% blacks
1965-1972 transition use of high school kids, migrants, sharecroppers
1972-present Mexicans increasingly predominate in the heavy field tasks
How did the transformation occur?
In the late 1960s as a combination of the urban migration, industrial development in the cities, the civil rights movement, and other causes, labor became relatively scarce. The children of the sharecroppers were not staying on the farms. The students did some of the work but they were not sufficient.
In 1972, July came and the crop was going unharvested. The stalks were turning black in the fields. The local Virginia Employment Commission (VEC) placement officer went up to Hendersonville, North Carolina to the apple country and picked up a truck load of Mexican apple thinners. They worked out so well that other growers went and found more Mexicans in Hendersonville. According to the growers, they saved the harvest that year. And ever since the Mexicans have done the majority of the field tasks.
The switch happened fast. As the head of the local growers' association put it: "before 1972 I had never seen a Mexican." Piedmont tobacco reached a turning point.
The Recruitment System:
The network recruitment system swung quickly into high gear. One Mexican "Julio" from Cuernavaca became the right hand man of the leading grower in the area. His brother and father who lived in Cuernavaca began recruiting his countrymen from the same region. By the next year, a permanent relationship had been set up between Southside Virginia and Julio's network.
Julio's relatives screened and transported the workers to the border where Julio went to pick them up and brought them to southern Virginia. In 1978, there were extensive INS raids in Southside tobacco. As a result, the Virginia growers shifted to the H2 guestworker program.
Why did the growers shift so easily from local to labor from Cuernavaca and stay with that labor indefinitely? Why didn't they use local or legal US migrant labor?
Two Groups Compared:
A comparison of the two groups explains why growers preferred Mexican labor. The Mexicans were all men, young, homogeneous, uncomplaining, already screened for the best workers, and willing to commit themselves to whatever task without being distracted by family obligations. Meanwhile, the US workers were a mixed group. Many were women and children. They did not have to stick with the job if conditions became intolerable
since they had other options. Plus, there was the general impression among all observers that the local workers did not want the work as much as the foreigners.
As one black sharecropper said about African American workers, "you go to pick them up (the African Americas), blow and blow your horn and only 10 in 100 come out."
Meanwhile, the recruitment of the Mexican labor force is absolutely free to the growers. Julio's father charged the workers a $245 fee in 1986 to get on the H2 list. He used this money to pay all his expenses of recruiting and screening workers and made sure they got to the border. In effect, the workers themselves with about 10% of their earnings were subsidizing the recruitment system.
To recruit legal US workers, the Virginia growers would have had to spend considerable time and effort. We saw successful recruiters of U.S. labor in Roanoke, Virginia apples and on the Eastern Shore of the Delmarva Peninsula among potato packers. But, these growers maintained long term ties to these U.S. workers and to their crew leaders. They often telephoned and sent cards off season. And, they paid for travel to the job and guaranteed employment through the season.
The growers in these cases had out of pocket expenses to maintain these U.S. worker networks. Much of these recruitment costs under the system organized by Julio were taken out of the workers' earnings.
Complements or Substitutes?
Like described for the other case study industries, Piedmont tobacco has ethnic segments. The barnwork (hanging the tobacco leaves), and the transplanting at that time (in 1986) was still done mostly by locals while the heavy field tasks of hoeing, suckering, pulling was done by the Mexicans. Did the presence of the Mexicans actually maintain the existence of the industry and keep the jobs held by locals? Or did the presence of the Mexicans drive the low wage workers out into the rest of the economy and lower wages there?
It is unclear. Most African Americans resented the presence of the Mexicans. They accused them of having driven down the wages; as one put it: "these foreigners have come in here and pushed us out of our own country." On the other hand, another African American tobacco worker told me: "the youngins don't work baca no more."
Probably both processes were going on at the same time. Some jobs were saved by the presence of the Mexicans and some downward pressure on wages occurred for all low wage workers in the area.
However, I must emphasize that up to now I have referred only to the local workers.
No Real Out-of-State Recruitment
One point seems irrefutable. The growers who use the low-cost, homogeneous Mexican networks from Cuernavaca never gave out-of-state U.S. worker competitors a real chance. Everyone involved is aware of the preference growers have for foreign-based networks of workers without long-term legal standing in the U.S.
As the VEC certification officer in Richmond at the time said, "You know that anyone who asks for an alien wants one. The job order is set up so they can get out of using domestics". In the first years of the program in the late 1970s, the employment service produced large numbers of referrals. But the growers didn't want their smooth running recruitment system disrupted. Employment Service agents from Florida, Texas and North Carolina all agreed that the workers who were sent up to Virginia were discouraged from staying.
As the agent from North Carolina stated, "We sent many people but they didn't want them. They gave them poor housing, poor jobs, and chased them away. I finally stopped sending them until the legal aid people made us do it again."
The agent from Texas thought he could place workers if the Mexican option didn't exist. "If the growers didn't have these other workers to fall back on, the Texans going up there would be treated different. There would be housing, transportation and so on. Right now, it's not worth it."
We did our own investigation that year. Thirty-three workers had been referred to the jobs. The growers filed reports that all had not showed up, quit or refused to work.
We reached and interviewed 12 of these "rejected" workers in our study. Eight of them disputed the grower claims. Some were told to dig ditches and fill them up. One other left after being physically threatened.
I described a system (progressive task substitution) through which US workers are being replaced or displaced by newcomers among the foreign born. And, I could not be certain which one is occurring. The process began on the West Coast years ago but now has spread to the East and is nearing completion.
One issue is clear. That the foreign-based networks are and will continue to provide cheaper labor. As shown by the North Carolina/Virginia case study, the system lowers costs to growers by a subsidized recruitment system. Also, the newcomer labor is homogeneous, uncomplaining, and willing to devote itself around the clock to the needs of the grower.
It is possible that this kind of replacement can continue indefinitely. Namely, new immigrants may replace older ones as soon as they settle down. This may not be advisable since it serves as an entry gate into our economy for thousands of new unskilled workers every year.
This process can keep wages and other working conditions very low in the farm labor sector. Since most farmworkers under such a system cannot earn enough to stay doing farm work, as new cheaper ones come they take the places of the veteran workers.
Whether this is displacement or replacement may be less important than that it may not be very good policy to encourage this constant influx.