CARIBBEAN MIGRATION PROGRAM
The occasional papers of the Caribbean Migration Program are
published several times a year by the Center for Latin American Studies, University of Florida, Gainesville, The Caribbean Migration Program began in January, 1982, with support from the Tinker Foundation and the Ford Foundation. The Program's goals are to:
- train new scholars in the field of Caribbean migration
- review the existing state of knowledge concerning Caribbean
migration in order to identify priorities for research and
- generate and disseminate research in priority areas of
- establish collaborative relationships with other scholars
and institutions in the Caribbean and in the United States
working in the area of migration
- serve as an integrating mechanism and informational clearing
house for research and forums on all aspects of Caribbean
migration,- particularly to Florida and the Southeast.
Caribbean Migration Program activities include a Visiting Scholars program, predoctoral fellowships and graduate seminars on various aspects of the Caribbean and migration. We are completing an extensive bibliography on Caribbean migration and a roster of researchers in the area. The Center also publishes the Amazon Research Papers prepared under the Amazon Research and Training Program.
For further information write:
Center for Latin American Studies 319 Grinter Hall
University of Florida
Gainesville, Florida 32611
Terry L. McCoy
Charles H. Wood
Center for Latin American Studies
University of Florida
Gainesville, Florida 32611
Paper No. 2
Caribbean Workers in the Florida
Sugar Cane Industry
I- P-t- 2
CARIBBEAN WORKERS IN THE FLORIDA SUGAR CANE INDUSTRY (I)
Terry L. McCoy and Charles H. Wood
In this paper we report on a study of a labor program that has, for the last 39 years, brought workers from the Conmonwealth Caribbean
(known as the British West Indies prior to independence) to south
Florida to harvest sugar cane. The eight to nine thousand men in this
~work force currently constitute the largest legal foreign migrant labor
program in the United States, In the face of increasing pressure for
additional labor migration from Latin America and the Caribbean to the U.S., this labor force poses a number of theoretical and policy issues
associated with the international transfer of labor from less developed
to more developed economies. Congress is currently considering
legislation which would modify the current temporary worker (or 8-2)
In the context of some of the questions raised by such proposals, and in light of the general concern over the impact of transnational
labor migration, we present the findings of a survey of the Caribbean
cane cutters carried out during the 1980-81 harvest. The questionnaire
generated data regarding the socio-demographic characteristics of the
workers, their occupations back in the islands, the composition of their
households, the impact of seasonal migration on them and their
dependents, and the disposition of the money earned in Florida. Before
turning to an analysis of each of these variables, the introduction
reviews the history and organization of the off-shore program in the
Florida sugar industry in order to provide the context within which this
case of seasonal foreign labor operates.
*Its 1980-81 crop of 1,121,1400 tons of raw sugar made Florida the
; leading domestic producer of sugar in that year (see Table 1).C3) The
record crop exceeded the earlier peak by.60,000 tons and outdistanced the second largest producing state, Hawaii, by 100,000 tons (lrd
* .SiagrtNews,17, No. 2[April 1981: 1.) Florida accounts for nearly
20 percent of all (cane and beet) domestic production and 10 percent of
all sugar consumed in the U.S. (Ibi.,17, No. 4L [October 1981]:1).
The Florida industry also differs from the other U.S. producers on three i nter-relanted dimensionns its relatnivelyv recent or-igain; the fact ithat it-
flORIDA SUGAR IIZDUSrRY PRODUCrION
Acres Cane Average 960 Raw Value
CROP Total Harvested Ground Cane Basis Basis
YEAR Acres for Sugar Gross Tons Per Acre Short Tons Short Tons
(mans.) (mans.) (Thous.) (Gross Tons) (Thous.) (Thous.)
1960-61 50.7 48.9 1,618 33.1 157 160
1961-62 60.0 56.1 2,121 37.8 205 208
1962-63 116.8 114.3 4,219 36.9 375 380
1963-64 148.2 139.9 4,632 33.1 418 424
1964-65 222.9 219.8 6,791 30.9 565 574
1965-66 190.8 185.4 5,886 31.7 545 554
1966-67 197.2 190.7 6,419 33.7 641 652
1967-68 196.3 190.6 6,889 36.1 704 717
1968-69 187.1 182.1 5,639 31.0 536 546
1969-70 160.1 153.5 5,470 35.6 525 535
1970-71 178.5 170.0 5,969 35.1 640 652
1971-72 199.6 189.9 6,389 33.6 623 635
1972-73 249.7 243.8 9,889 40.6 944 961
1973-74 265.5 257.6 8,604 33.4 809 824
1974-75 273.4 258.4 8,050 31.2 779 793
1975-76 298.7 284.7 10,807 38.0 1,032 1,050
1976-77 297.3 285.6 9,919 34.7 912 930
1977-78 297.2 282.3 9,029 32.0 877 894
1978-79 316.0 300.0 9,745 32.5 952 972
1979-80 332.1 316.8 10,608 33~~5 1,028 1,047
1980-81 339.1 320.7 10,623 33.1 1,103 1,121
1981-82 348.9 335.2 10,146 30.3 948 963
SOURCE: Florida Sugar Cane Leagi.e, 1982.
li~~uU.S~~~f .Il -dS51Sf115.15151 itS 551Sl
- 4;96,0 d ~f1W I- ffSSSlfUI USS 1dlf i
jnim -ff51 15 f SISf5 SSfiSf iS1151fS~
75 Si1i'fSSi1 SS SfiS ~S SSSfSflf
5555SS 15f 75 SS~f A~II i t',SISS551 SYI
the 1960-64 expansion phase. The largest producer and processor is U.S. Sugar, which owns two of the mills. Gulf and Western, a multinational corporation with extensive sugar operations throughout the Caribbean, is another major producer with its own mill. One of the remaining mills is owned and operated by Cuban-American capital, while a growers' cooperative also operates another.
The economic impact of the industry on the state and on the
producing area is considerable. In an industry-commissioned study, two economists calculated that sugar accounted (directly and indirectly through its multiplier effect) for 10 percent of all sales and 14 percent of all jobs in the local four county economy (Mulkey and Gordon, 1979:26). This same study also concluded that the money spent by the industry's migrant labor force generated between $3.2 and $5.4 million in income for local residents. At the state. level, sugar accounted for 11 to 12 percent of Florida's total agricultural sales for 1974 and 1975 (News Release of the Florida Sugar Cane League, July 5, 1979).
Perhaps even more than other agricultural commodities in the U.S., the profitability of sugar production is closely linked to the policies of the federal government. As both a cause and an effect of this condition, domestic sugar producers are organized into strong lobbies that operate at the state and national level. Florida growers are represented by the Florida Sugar Cane League. Since the repeal of the Sugar Act (which maintained a floor on the price paid domestic producers by controlling foreign access to the U.S. market through a system of country quotas), the principal objective of the sugar lobby has been to re-establish direct federal price guarantees. In the eyes of the industry the need for such legislation is justified by the volatile nature of world sugar prices since 1974. From a high of $.57 per pound in 1973-74 they plunged to $.06 in 1978 (Thompson, 1979:3).
Along with associations in the other producer states, the Florida Sugar Cane League lobbied strenUously for legislation to stabilize the market. In the words of the League's Vice President, "We are willing to take the top out of the Market .... At the same time, we want the bottom taken out (Dalton Yancey, n.d.,:4).11 In 1977 President Carter imposed import fees on foreign sugar and established a subsidized commodity loan program that, in effect, guaranteed a minimum price for domestic sugar (Johnson, 1979:549). Despite these steps, the industry fell into a second slump when the world price dropped to $.08 in 1979, and the goverment loan program proved insufficient to offset the comparatively high costs of U.S. producers. In Florida, for example, Johnson reports that costs were on the order of $.13-.14 per pound in 1979 (1979:54) In addition to the unstable price for their product, sugar producers face increasing competition from alternative, cheaper sweeteners. According to Thompson, starch sweeteners (such as corn) account for about one-fourth of per capita sweetener consumption in the U.S. (1979:83).
The domestic sugar industry has received additional assistance
from the Reagan administration. The Omnibus Farm Act of 1981 included a sugar price support provision that guarantees domestic producers around $.20 per pound through 1985. The President signed this legislation in tandem with two executive orders imposing import fees and increasing duties on foreign sugar CT11Q Sugar Bulletin, 60, No. 8 [January
1,1982):6]. In the spring of 1982, the President went even further by re-establishing import quotas (New York imes, May 5, 1982). All of the measures attest to the considerable economic significance and political influence of the sugar industry, the net result of which is to maintain prices to domestic producers at about $'020 per pound, while the world price slipped to under $.10 in mid-1982 (Miami Head October 11, 1982).
3. Harvestiang Tcniques
In the precarious and competitive production of sugar, nearly all Florida growers are unique in their reliance on manual harvesting. All other domestically grown sugar cane is harvested mechanically. That Florida growers rely on a large, legal migrant labor pool in the U.S. is a function of the peculiar nature of sugar cultivation in this state and special immigration law provisions for importing temporary labor.
Mechanization of the Florida harvest is precluded for technical and ecological reasons. The recumbent character of the cane makes it difficult to pick up and the soft muck soil will not sustain heavy equipment without destroying the ratoons which produce the following year's crop. Reinforcing these constraints on mechanization is the fact that the land is subject to rapid oxidation when exposed to the atmosphere.
Organic soil in the Everglades agricultural
area is subsiding at the rate of about one inch
per year. By the year 2000, it has been estimated that about 5000 thousand acres (or 87 percent of the total) of organic soil will be three
feet or less in thickness, and over 250,000 acres
will be less than one foot. The fate of agriculture on this soil is uncertain. Large tracts
of land that now are agriculturally productive
may be abandoned.... (Snyder,et.al., 1978:ii).
Sugar cane cultivation is particularly damaging to the muck soils, since it subjects them to conditions-- shrinkage, compacting, and oxidation-that accelerate their deterioration. If these problems are not resolved, sugar cultivation is likely to decline over the next 25 years. Experiments are currently underway to alternate cane with the cultivation of rice, a crop that retards oxidation through periodic
flooding of the fields.
Faced with technical and ecological constraints, most Florida
growers continue to find manual harvesting more profitable. The capacity to mechanically harvest cane has been developed and it is used on one of the large estates, having been adopted following a strike. But this technology, because it is less efficient from an overall production standpoint, is held in reserve in the event of disruption in the labor supply (Arias, 1976; Reubins, 1978:29). The Florida sugar cane industry thus depends on access to a stable, flexible supply of seasonal labor.
publicly defended the BWI arrangement. George Meany of the AFL-CIO declared in 1965 that, "It is well known that the BWI program improves the lot of domestic workers in regard to housing, worlanents compensation, and racial matters (.bid.-1978:21)."1
President Johnson brought an end to the bracero program in 1964~. As the number of Mexican legal migrants decreased, those entering Florida from the West Indies increased. This accelerated influx coincided with the emergence of large scale sugar cane cultivation around Lake Okeechobee. Of all the original employers of H-2 laborand it was utilized throughout Florida agriculture as late as the early 1970s --today only the Florida sugar industry and East Coast apple growers are able to regularly meet the qualifications to hire offshore seasonal labor (U.S. Department of Labor, 1982).
Under U.S. immigration law it is possible to import foreign
workers for seasonal labor. The H-2 category exists for that purpose. However, in an effort to protect U.S. workers from lower paid imported labor, Congress and the Department of Labor (DOL) have created a certification procedure that makes it difficult for the would-be employers to qualify for foreign workers.
The first step in gaining consent from the federal government to import labor is to demonstrate to DOL that there are no U.S. workers available to do the job in question. Once this condition is documented, DOL then advises the Immigration and Naturalization Service (INS) that a shortage of domestic workers exists, and INS grants approval for the importation of foreign workers for the specific jobs and employers that have been certified (McCoy, 192:3). This system is designed to protect the U.S. job market from what is viewed as unfair foreign competition.
Originally, certification was granted for up to three years on the basis of demonstrated labor shortages in a given region. Currently certification can only be given to the petitioning employers for pre-specified tasks, and is issued for one year or less (eight months in the case of sugar). This means that employers must be certified every year. Furthermore, DOL grants certification only after the employers have engaged in an exhaustive search, or "positive recruitment," for domestic workers.
Success in recruiting domestic workers depends not only on
advertising the job vacancies but also on wages and working conditions offered prospective employees. To prohibit West Indians (who would readily work for wages well below those attractive to American workers) from unfairly competing for U.S. jobs, the DOL has established what is known as the "adverse wage rate." This is a guaranteed hourly wage which should not only attract domestic workers but also, failing to do so, not adversely affect the wages of U.S. workers in related activities (i.e. agricultural workers). The adverse wage for a given season is dictated by DOL on the basis of studies, done by the Field Reporting Service of the Department of Agriculture, to-determine the prevailing wages for
farm workers.(I) In the case of sugar it has always been above the federal minimum wage but not high enough to attract many sufficient domestic applicants. In 1980-81 the adverse rate of $41.09 per hour compared with $3.10 for the federal minimum (State of Florida, Department of Labor and Employment Security, 1980).
Beyond offering potential employees an hourly wage in excess of prevailing agricultural rates, the sugar industry to be certified must provide living quarters, meals, and transportation to its cane cutters at reasonable rates. The conditions of employment are monitored by the DOL, working with its local and state counterparts.
In spite of its positive recruitment efforts, the industry does not secure sufficient domestic workers to harvest its annual crop. According to data presented in support of its case for DOL certification f or the 1979-80 season, over the preceeding six years the industry spent $109,041.27 (or an average of $18,174 per year) on advertising for domestic cane cutters. It spent an additional $413,778 ($7,297 per year) on transporting domestic recruits to the work place. The net result of all of this, however, was that =n =.n domestic worker lasted out the entire season (Morrison, 1979). The industry's long-standing contention that there are no workers in the U.S. who will cut cane appeared
strengthened the following year. Under pressure from Florida state officials, it sought to recruit workers from among the Cuban and Haitian refugees then entering the state. Even this apparently job-hungry labor pool produced only several hundred cutters for the 1980-81 season.(5)
Once the Labor Department certifies a shortage of domestic
laborers, the INS may then issue permission in the form of 1-94 forms for the workers in a quantity approximating the number requested by the industry. Under existing law, DOL certification is advisory and not binding. For 1980-81 there were 8,460 entries authorized (Interview with industry official, December 17, 1980). Although not the subject of this report, the related questions of why U.S. workers do not cut sugar cane, even in periods of high unemployment, and why the industry clearly prefers Caribbean labor are much debated. For our purposes here, however, it is sufficient to note that, with the granting of labor certification industry, representatives may then bring in cutters from abroad. In actual practice, the process of recruiting these workers begins well in advance of final certification.
3. Recruiting Caribbean M1grer
One of the strange idiosyncracies in this regard
is that the Jamaican who cuts in Florida will
not do so in Jamaica (Florida Sugar Cane
Because of low wages and high unemployment, the demand in the
Caribbean for participation in the sugar work force far exceeds the need for workers in Florida. For its part, the sugar industry seeks to restrict participation to those workers who have demonstrated their
reliability and productivity during previous seasons. The island governments, on the other hand, feel the need to keep the opportunity open to as many of their citizens as possible. The existence of these competing interests has led to the institutionalization of a recruiting procedure which balances the growers' preference for a stable, experienced work force with the desires of the Caribbean governments to spread the participation in the migratory stream around among their populations.
At present the program is limited to natives of the five
Commonwealth Caribbean islands: Jamaica, Barbados, St. Vincent, St. Lucia, and Dominica (see map and Appendix A). The roots of the arrangement date from its origin in 194~3. The mechanism for perpetuating it is an annual contract between representatives of the sugar industry and those of the West Indian governments and workers. For the industry the offshore labor program is managed by the Florida Fruit and Vegetable Association (FFVA) and U.S. Sugar Corporation, with FFVA representing most employers.
Representation of West Indian interests is more complicated., The interests themselves are diverse and occasionally divergent. The principal organization is the British West Indies Central Labour Organization (BWICLO). Its governing council the Regional Labour Board, is composed of the following: the Permanent Jamaican Secretary for the
Ministry of Labour, who serves as Chairman; two other' Jamaican government officials; the Permanent Secretary of the Ministry of Labour from Barbados; the Labour Commissioner of St. Lucia, representing the three remaining small islands; and a representative from one of the two Jamaican labor unions (depending on which party is in power). The BWICLO, with a permanent office in Washington, D.C., maintains 11 liaison officers in the sugar area. Although the Regional Labour Board and industry representatives meet annually to re-negotiate the contract, it remains essentially the same from year to year.
Several aspects of the arrangement are noteworthy. First, the
Department of Labor, because of its mandate to protect domestic workers, also sets wages and working conditions for H-2 participants. Second, Jamaica, which sends 80 percent of the labor force, dominates the BWICLO and Regional Labour Board. This situation is a source of criticism among the workers and officials of the four other islands.(6) Third, from the worker's perspective, the major criticism of the contract is that, after specifying hourly wages, the living and transportation arrangements, and working conditions, the contract is vague as to the piecework formula and other standards of productivity. According to some observers, this works to the disadvantage of the cane cutters. The
summary deportation of workers and subsequent blacklisting of individuals are often cited by critics, (Petrow, 1980 and NACLA, 1977). Worker complaints and grievances are channeled through the 11 BWICLO liaison officers.
The agreement between the employers and the BWICLO specifies that the industry recruiters may request no more than 60 percent of the work force from those who participated in previous seasons. The remaining 140 percent is drawn from the pool of "non-preferred" applicants. The
rLO IDATHE WEST INDIES
e x 10islands sending workers to Florida sugar industry.
IbS is T F I A N I I C
pupunhicANGUILLA faieJM IASo-to Domingo, INt SAB f s a AR BUDA~N.Ih
HODUA' jiS OANTIGUA
C A N '1 Il ii
N I CA R AG U BARBADOS
ARUBA 1810a dS
.0C0L0MGIA0 Miles 200
80* 10 10
proportional split between preferred and pool workers is an issue
regularly negotiated. Employers favor enlarging the former; island
governments press for increasing the number of pool workers.(7) The
percentage of preferred individuals has increased from 20 percent to the
current level (60 percent). They must select the remaining 40 percent
from a pool of non-preferred applicants.
The procedures for recruiting pool workers vary from island to
island. It is most elaborate in Jamaica where workers are recruited
through a complex system involving Members of Parliament (MPs), local
committees, and the Ministry of Labour. The Ministry assigns each
parish a quota of candidates for the non-preferred pool, and it requests
each MP to form a committee to identify candidates from his
constituency. On the basis of the committees' unanimous
recommendations, the Ministry sends those candidates a card which makes
them eligible to be interviewed by industry recruiters. Recently, to
discourage candidates from transferring or selling their cards, the
candidates are required to be fingerprinted (Interview with FFVA
official, July 20, 1982). The various steps in the selection process
have led to charges of undue political and economic pressures (Petrow,
1980; Palm Beach .ost, November 25, 1974). Once the pool of
eligible non-preferred workers is constituted, skilled industry
representatives go to Jamaica to select from it those whom they wish to
hire. This requires reducing the initial pool from approximately 10,000
to 7,000. It is done through personal interviews with each card holder
at Ministry of Labour pre-selection stations throughout the island. In
these interviews, the U.S. recruiters screen applicants on the basis of
their physical appearance, occupation, work history, and attitude.(8)
Preference is given to farm workers, especially with cane cutting
Following pre-selection by the U.S. recruiters, the would-be cane
cutters take medical exams in Kingston. They are also now screened by
the U.S. Consulate (Interview with consular official, September 10,
1982). Those who pass await final call from the growers. The length and
uncertainty of recruitment means that the individual worker may not know
whether he is going to Florida until several days before departing.
Some workers are not even called until mid-way through the harvest.
Travel arrangements, documents, and charter flights are handled for the
growers by a private firm.(10)
The recruitment of workers from the other four islands is less
complicated. It is also less formally integrated into the system of
political patronage compared to Jamaica. As in Jamaica, once a pool of
candidates is established, it is the sugar industry agents who make the
final selections among the non-preferred workers. The specific way in
which this is accomplished varies somewhat from island to island.
In Barbados, which supplies the second highest number of
participants -- approximately 600 in 1980-81 -- demand to participate in
the program is perhaps less intense than in other countries. This is
presumed to be related to the island's standard of living and wage
structure, both relatively high for the region (Brana-Shute and
Brana-Shute, 1981 and Appendix A). As the country develops, government
officials (Interview with Permanent Secretary for Labour, July 21, 1982)
anticipate declining participation. For now stateside employment fills an important need. It should also be noted that the Barbadian sugar industry, an important component of the economy, cannot find enough local workers to harvest its crop and must annually import them from the neighboring island of St. Vincent. (Interview with union offical, July 214, 1981). Because of overlapping seasons, local sugar workers must choose where they want to work. Barbadians interested in cutting cane
in Florida simply register at the Employment Exchange of the Ministry of Labour, which carries out a preliminary screening prior to the arrival of industry recruiters.(11)
St. Lucia and St. Vincent supplied 450 workers each for the 1981-82 season. Dominica, which in that year participated on an experimental basis, contributed 60. Under conditions of high unemployment, and in view of the low standard of living of these three less developed islands of the Eastern Caribbean, it is not surprising that the demand to go to Florida is great. A person who does so stands to earn several times the annual average local per capita income. This configuration of chronic conditions, combined with the fact that the islands had recently suffered the effects of a devastating hurricane, produce situations such as that in St. Lucia where there were 300 candidates for the 20 slots allocated to the district (Interview with local official, July 30, 1981). Not only are men from these islands eager to participate, but their governments are in need of foreign exchange, which it generates. (12)
Since 1974 the procedure for building the pool in St. Vincent is for ministers of government. Members of Parliament from the ruling party submit nominees to the Department of Labour. In Saint Lucia the selection process is entirely in the hands of civil servants, although they will consider recommendations from politicians.(13) Regardless of the nature of the selection process, it is clear that conditions internal to the less developed small islands lead to pressures on the governments to sustain the program.
14. Transportation, Livi.ng Arrangements.,.. Working CoQnitiQI2s -aid Wages
Recruitment into the pool of available workers and selection by
sugar industry representatives, both completed by early summer, do not guarantee work in the sugar fields. That depends upon a final call from the employers for specific numbers of workers. This request may not come until mid-fall. As a result some men do not go until the harvest is well underway. Although the industry estimates the number of men it will need each year, the actual size of the labor force depends on factors not determined until the harvest is about to begin. When a worker is summoned, he goes to a central location for transportation to Florida.
All workers are taken by charter aircraft to West Palm Beach and
then by bus to the sugar area. The contract guarantees free round trip air fare from Kingston for "all workers who complete 50 percent of the
contract period (State of Florida, Department of Labor and Employment
Security, June 4, 1980)." Non-Jamaican workers must make up the
difference equivalent to the fare from their island to Kingston.(14) In
all cases, the cost of the trip from Jamaica to Palm Beach is deducted
from the workers' paychecks and then reimbursed at the end of the
season. This practice is presumably designed to discourage the worker
from using the program for free passage to the U.S.
Each worker is assigned to a specified grower prior to departure.
Growers are contractually obligated to provide workers with free
housing. Three meals a day are also provided at a cost not to exceed an
agreed upon rate ($5.00 plus $.20 tax in 1980-81, State of Florida,
Department of Labor and Employment Security, June 4, 1980). Housing consists of barracks located in camps on or near the sugar estates.
Meals with West Indian menus prepared by native cooks are served in large mess halls, except for the mid-day meal which is served in the
fields during the 15 minute lunch break. The camps include offices, a
first aid clinic, commissaries and limited recreational facilities where
the occupants can watch TV or play dominoes. A larger infirmary is
adjacent to the BWICLO liaison officers headquarters in South Bay.
Serious medical cases are referred to area hospitals. A special program
offers workers night classes where it is possible to earn a high school diploma. Our study indicates that about 1 in 10 workers take advantage
of this opportunity.
Cutting sugar cane is physically demanding, dirty work. The
contract offers the [1-2 workers eight hours of work per day, six days
per week and guarantees them work for the hourly equivalent of
three-fourths of the work days of the contract period. The actual
number of hours worked per day was not information generated by our
survey. For their efforts they are assured an hourly minimum wage. To maximize their earning potential most workers choose to work on Sundays.
An individual's actual pay, however, is the function of a complex
Every worker's season begins with an eight-day training period
during which he is paid the minimum wage. In the words of the contract:.
Thereafter, workers will be paid on ~a task basis. The task will be based on the work which can be performed by the
average worker. Any time during the
- training period a worker demonstrates
that the worker is proficient by cutting a task in less than 8 hours, the worker will be assigned to the task rate if agreeable with the worker. In the event
With the training period finished and the cutting norm established, workers are encouraged to exceed it by a bonus system that allows them to earn additional wages by cutting more than their assigned task in an eight-hour day. For example, although the minimum wage rate was $4.09 per hour in 1980-81, the contract indicated that it should be possible for the "normal worker" to increase his average earnings to $4.29 per hour for the entire season. All workers are expected to cut an average of eight tons of cane a day. Stories of individuals who far exceed the minimum wage are common. The top worker in one camp reportedly earned a bi-weekly rate of $1000 (Interview with Office Manager, March 19, 1981). In addition to the promise of additional earnings, the task formula offers the inducement of being invited back as a preferred worker. On the other hand, it is criticized for giving excessive latitude to those who measure and assign tasks, and to the "ticket writers" who record the day's cut.(15) Workers who fall below the norm find themselves being sent back home and blacklisted from future seasons for breach of contract.0(6)
Life in the camps and field work is supervised through a system which relies heavily on West Indians as immediate supervisors. At the top of this hierarchy is the camp supervisor. Under his command and in charge of the field work are the field bosses, the lead men and the ticket writers. Most of these are West Indians. The field boss is responsible for the work in his field while it is the lead man who assigns each cutter his daily task or "row". At the end of the work day, the ticket writer records that day's cut. From the standpoint of
the worker there is sometimes confusion and controversy as to. the value assigned to the tasks. Similarly, disputes arise over the size of the annual bonus.
The responsibility of the 11 liaison officers is to represent the cane cutters when conflicts and grievances arise. As employees of the BWICLO, the liaison officers are to represent all workers equally. Yet, the predominance of Jamaicans among the liaison officers is a source of complaint. Non-Jamaican workers claim their interests do not receive equal treatment. In 1981-82 and 1982-83 steps were taken to incorporate non-Jamaicans into the liaison staff.
From each paycheck certain mandatory deductions are made. As
already mentioned, the employer removes funds advanced for the trip to Florida (at the end of the season this is returned to those who finish), an initial cash advance for subsistence and clothing, and the cost of meals. More importantly, 23 percent of gross earnings is deducted and sent back to the islands. Although the final disposition of these funds varies from island to island (for example on the small islands a portion goes to reimburse the BWICLO and government for the air fare advance), the largest share goes into an interest-free savings account for the worker to claim on his return. Two percent is claimed by the BWICLO for expenses and a special non-work related hospitalization plan for the workers.(17) The mandatory deduction, one-quarter of gross wages, represents a significant contribution to the islands' hard currency accounts, since it is redeemed to the workers on their return in local currency at the official exchange rate. Other deductions, of an optional nature, include savings accounts in local banks and regular
other countries in the surrounding area. Within the area itself there has always also been a flow of people among the islands that make up the West Indies (Marshall, 1982). Inter-island migration is especially salient in the chain of small states that comprise the Eastern Caribbean.
Although we do not have complete migration histories for each
individual in the sample, several items in the questionnaire (country and parish of birth; previous and current place of residence) provide a rough indication of the previous migratory patterns that characterize the H-2 labor force employed by the Florida sugar industry. The specific question addressed here is whether individuals who harvest cane in Florida are drawn from a population which is stable or geographically mobile at the place of origin.
In Table 2 the respondents are classified by country of birth and by country of residence at the time of the interview. As indicated by the large proportion of the sample that falls along the diagonal, the vast majority of the population of offshore workers live in the same country in which they were born.
An indication of the magnitude of intra-island migration (Table 3) is provided by the percent of workers who live in the same parish where they were born. In Jamaica, about two-thirds of the population lives in the place of birth. In other words, only about one worker in every three migrated from one parish to another. The data from Barbados and the small islands show a similar distribution.
A separate aspect of the migratory profile is the extent of rural to urban migration. A rough indication of this flow can be derived by comparing the proportion of the population that is of rural origin, with the proportion that declares a rural place of residence at the time of the interview. In Jamaica, 90.3 percent of the workers responded that they were born in the countryside, while 82.4 percent currently reside there. This combination implies a rural to urban movement of only about
7.9 percent of the population. In the case of Barbados and the small islands, the proportions are also quite small (6 percent and 3.9 percent, respectively), although the movement is in the opposite direction (from urban to rural areas).
To investigate the motivation for geographic mobility, respondents were asked the reason why they made the previous move. The results are in Table 4. With the exception of Barbados, a substantial proportion of migrants said that they moved to find a job or to change their occupation. Between 15 and 20 percent of the sample changed residence when they were young, accompanying parents. Similarly, a large proportion moved from one place to another either to get married, or to move in with a girlfriend. Although these findings confirm the salience of economic motivations for migration, they also suggest the importance of other factors, such as life-cycle changes, for shifting residence. These findings confirm other studies of Caribbean migration (see Marshall, 1982).
Data regarding the magnitude of migration are likely to
Country of Birth Jamaica Barbados Islands
Jamaica 99.0 ...
Barbados 1.0 93.0-Small Islands --- 7.0 100.0
Total 100.0 100.0 100.0
n 206 45 51
Resident in Have moved Total
Island of Residence Parrish of Birth From Parish of Birth* (n)
Jamaica 67.7 32.3 100
Barbados 63.4 36.6 100
Small Islands 67.4 32.6 100
underestimate the real values. The findings reported above are derived from comparing place of current residence with information on the respondent's birth place. This approach fails to classify as a migrant any individual who moves (once, or several times) in his lifetime, yet currently resides in the same parish or country in which he was born. In the absence of a complete migration history, it is not possible to determine the extent to which the results presented here may underestimate population movement.
Other difficulties are associated with the definition of what is a rural and what is an urban place. To live "in the countryside" is a common response that appears to reflect the conventional notion of a rural area. Beyond that the classification is potentially ambiguous. Individuals who said they lived in a city or a town were classified as urban dwellers, although the exact meaning of such terms may not have been consistent across respondents. Furthermore, the distinction between urban and rural is not a clear one in the West Indies where the physical boundaries that separate village and countryside are often blurred.
With these caveats in mind, we can make some fairly safe
generalizations about the migratory histories of the individuals who participated in the H-2 labor force in 1981. Our findings suggest that very few live in a country or parish different from the one in which they were born. Similarly, the degree of movement between rural and urban areas is small. Fran these data we can conclude that cane cutters are drawn primarily from a stable, predominately rural sector. The H-2 labor force does not appear to be recruited from the highly mobile Caribbean population.
The level of education achieved by an individual has a profound
effect on his overall life chances. Measures of educational attainment for the sample of workers can be derived from several questionnaire items. These include school attendance at the primary and secondary levels as well as participation in specialized training programs or apprenticeships for the acquisition of vocational skills.
Indicators of educational background are shown in Table 5. The results reflect a fairly low level of formal education. Nearly 13 percent of the labor force never attended school. For those who did, the average number of years of primary education is only 3.9. Further study beyond the primary grades is uncommon. Only 4.4 percent of the total sample were exposed to some schooling at the secondary level. These measures of educational background are approximately the same in Jamaica and in the small islands. Barbadians appear to be an exception. On the average, individuals from this island have more primary education (5.7 years), and a comparatively larger proportion attended secondary school (17.1 percent). -This is in keeping with Barbados' status as the
(% By Island)
Respondents Total Jamaica Barbados Islands
School 87.3 89.0 100.0 70.6
School 4.4 3.8 17.1 2.0
Training 43.2 38.6 61.4 64.0
in Florida 13.7 15.1 11.1 5.9
Educational Attainment by Previous Seasons Worked Mean years of
Previous Seasons Worked Primary Schooling
7 + 3.9
There is another aspect of the selectivity issue that is
associated, not with the recruitment of workers, per se, but with the
retention of individuals in the labor force from one season to the next.
As noted earlier in this report, the contract arrangement between the
sugar industry and the BWICLO permits employers to rehire 60 percent of
the labor force the following year (66.7% of our sample were preferred
workers). The criteria used by the industry to select "preferred"
workers includes a number of factors, the most important of which is
productivity. The question posed here is whether or not this yearly
selection process is selective of workers with high or low educational
To address this question the sample of H-2 labor force is
classified according to the number of previous seasons worked. If the
process of designating preferred workers is systematically selective
(in terms of education), we expect the mean years of schooling to change
as the number of past seasons increase. The results, shown in Table 6,
do not support this conclusion. Individuals that have worked two
seasons or less show a somewhat higher level of educational attainment
compared to those workers who have returned many times. This
difference, however, is relatively small, and is not statistically
significant. A more thorough analysis of this issue (using a regression
format to control for the respondent's age) does not alter the results
presented above. This would seem to indicate that worker productivity
(and therefore his value to his employer) is primarily a function of
experience cutting Florida cane rather than formal education.
3. Housing and Hme Ownership
Housing is an important indicator of a population's overall quality
of life. It can be assumed that the outright ownership of a home
indirectly reflects a degree of accumulation that is above that of an
individual who rents living space. The distinction between home owners
and renters thus provides a starting point for an analysis of the
socioeconomic conditions that characterize the BWI labor force at the
place of origin. In the Caribbean, where living arrangements often
assume complex forms, other criteria in addition to the owner/renter
distinction, are also important.
As shown in Table 7, the vast majority of the H-2 workers live in
houses, as opposed to flats or single rooms. This is especially true in
Barbados (95.6 percent) and in the small islands (92.2 percent). Only
in Jamaica does a relatively large percentage of the labor force reside
in flats or rooms (21.3 percent). Because the overwhelming majority of
the workers are from rural areas, where single family dwellings are
common, the large proportion of the labor force that live in houses is
Whether the individual was an owner or a renter, or if he had some
other arrangement with respect to housing, was determined by an
Type of Residence
(% By Island)
Respondent lives in Total Jamaica Barbados Islands
House 81.3 78.6 95.6 92.2
Flat 10.9 12.1 4.4 5.9
Room 7.8 9.2 --- 2.0
Total 100.0 100.0 100.0 100.0
Type of Housing Arrangement, by Age
Type of Housing
Arrangement Age of Respondent
Total 20-24 25-29 30-40 40 +
Own Home 46.9 22.6 17.5 42.6 81.9
Rents. 19.2 7.8 20.2 26.4 10.3
(From family) (2.3)
(From non-family) (14.3)
(Land, owns home) (2.6)
Other Arrangement 30.6 69.6 53.6 6.1
(Parents own) (15.8)
(other family member owns) (13.5) (Non-family member owns) 0.3)
Other 33.3 --- 8.7 2.0 1.7
Total 100.0 100.0 100.0 100.0 100.0
open-ended item included in the survey questionnaire. The results of the analysis are presented in Table 8. For the sample as a whole, 46.9
percent owned their own home. About one in every five workers paid rent of some kind, either for a house or flat, or, in some instances paid rent for the land on which the individual's home was built. A comparison between the various islands (not shown) reveals very few differences in the housing profile.
In addition to the distinction between owners and renters, the data presented in Table 8 reflect significant social dimensions of the housing question. About one in three workers (30.6 percent) neither pay rent nor do they own the place in which they live. Over 15 percent occupy a house that is owned by parents; 13.5 percent occupy facilities
that belong to other members of the family. For those who pay rent, only on rare occasions is it paid to persons who are related to the respondent (2.3 percent).
The pattern shown for the sample as a whole overlooks important
life cycle differences that exist with regard to housing arrangements. As showmn in the right-hand panels of Table 8, significant changes appear when we compare men of different ages. The proportion of home owners increases in the older ages, reaching 81.9 percent for those 40 years of age and older. On the other hand, rent-free occupancy of family owned property is characteristic of young men. The proportion who benefit from rent-free arrangement steadily declines at older ages.
These results indicate that workers, especially those who are in the younger ages (below 30), draw heavily for their support on the larger familial networks of which they are a part. More generally, the
information on housing arrangements confirms the importance of family ties in the Caribbean context.
B. Ocpto and giutrlAtvte
1. Occupational Bacgrfounid
A salient characteristic of Caribbean society is the multiplicity
of employment. In a given day, week, or month, an individual can be .
involved in any number of income generating activities (Brana-Shute and Brana-Shute, 1981). In the social science literature the frequent movement across job categories is sometimes referred to as "occupational multiplicity."
Distribution of Principal,
Second and Third Occupations Back in Islands
Occupation Principal Second Third
Farmer, Farm worker 63.8 37.1 25.9
Construction Related 19.8 32.9 19.6
Carpenters 4.8 4.8 3.6
Masons 7.3 9.2 9.1
Laborers 4.6 11.7 5.7
Othersi 3.1 7.2 1.2
Service related2 2.5 2.8 7.2
Drivers, Mechanics 3.3 9.3 5.7
Cane Cutters 1.7 2.5 7.2
Fishermen 0.4 1.5 4.3
Other3 8.5 13.9 30.1
Total 100 100 100
n of cases 251 141 28
% with a second occ'n 56.2 -% with a third occ'n -- 11.1
I Painter, Plumbers, Electricians, Cook
2 Petty sales, Clerks, Waiters
3 Bakers, Butchers, Tailors, Other
as to tlie other jobs that he may regularly perform.
The results generated by this approach are presented in Table 9. The percentage distribution shown in the first column demonstrates the importance of agricultural work, an expected finding since industry recruiters seek farm workers. For about two-thirds (63.8 percent) of the respondents, the primary occupation given was that of farmer or farm worker. Within the category of "primary occupation," the second most important jobs were those related to the construction industry. One in every five workers (19.8 percent) were involved in the building trade as either carpenters (4.8 percent), masons (7.3 percent), laborers (4.6 percent) or other miscellaneous tasks(3-1 percent).
Other occupations, such as drivers, mechanics, fishermen and
waiters are only marginally represented. It is worth noting that only 1.7 percent of the men interviewed on the sugar plantations in Florida declared cane cutting to be their primary job back home.
When asked about their second most important occupation, 37.1 percent of the sample said they were farmers or farm workers. As a source of secondary employment the importance of jobs related to construction increased to 32.9 percent. Finally, as shown in the last column of Table 9, farming and construction remain the principal occupations even for those who declare three regularly performed jobs.
The patterns of primary and secondary employment cannot be assumed to be the same across islands, as evident from the findings summarized in Table 10. Given the importance noted above of agriculture and construction-related jobs, the percentage distributions presented here are, for the sake of simplicity, restricted to these two work categories. The inter-island comparison shows Barbados to be a marked exception. On this island only 13.3 percent of the respondents declared agriculture to be their principal employment. The construction industry, on the other hand, is the major source of jobs in Barbados. The predominance of construction workers, and the relatively low number of farmers and farm workers in Barbados, is the opposite of what is found in the case of Jamaica and in the small islands.
Because the job categories presented in this section are based on the individual's self-report, the occupational data should be regarded with caution. It is often hard to know what separates one job description from another. Two workers, for example, may perform identical tasks, yet one may report his occupation as that of carpenter, the other simply as a construction worker.
These difficulties notwithstanding, the occupational distributions presented in this section provide some indication of the character of the population from which the H-2 workers are drawn. With the exception of Barbados, the findings confirm the predominance of farmers and farm-related workers. Jobs in the construction industry are also important, especially as a source of secondary employment. Farming and construction work account for 84.8 percent of the principal occupations, and 75.5 percent of the second jobs. The range of occupational backgrounds represented is, therefore, quite limited. Moreover, as
1.%, A -h i-11
underestimated. A recent AID (1978) report summarizes the part played
by the small farmer in the following way.
1. The small farmer produces most of the nation's domestic food
crop. This includes: legumes, vegetables, condiments, fruits, cereals,
plantains, potatoes, sweet potatoes, yams and other tubers. This
represented a value of J$115.O million in 1976 sand J$129.4 million in
2. Small farmers produce a substantial part of the nation's
-A foreign exchange that originates in agricultural exports. This includes
most of the coffee and cocoa, pimento and ginger, and smaller
percentages of bananas and sugar cane. Approximately 25 percent of the
value of agricultural exports in 1977 are estimated to have derived fram
the small-farm sector.
This section reports the results of several questionnaire items
which sought to probe into the extent and nature of farming activities of the H1-2 labor force, which is so heavily agricultural. Emphasis is given to a general description of the land tenure arrangements, to the
size of plots that the individual wns, or has access to, and to a rough
inventory of the crops cultivated and livestock owned. These data, it
should be emphasized, are principally intended to provide a general
profile of the labor force that seasonally migrates to Florida to harvest sugar cane. The agricultural background of these men is a
fundamental aspect of their social and economic circumstances at the
place of origin. The survey instrument, however, was not primarily
designed to be a sophisticated agricultural questionnaire. As a result,
less information is included in the data set than might have been the
case were the analysis restricted solely to this area of investigation.
We do not, for example, have reliable data on the quantity of
agricultural output, nor are we able to accurately assess the amount of
income specifically generated from agricultural activities. These
limitations notwithstanding, the information at hand does provide a general, if incomplete, characterization of the labor force that is
recruited in the West Indies. This agricultural profile of the workers
in the sugar program will, in turn, play a central role in the
evaluation presented in subsequent sections of the mechanisms by which
stateside employment influences the islands fran which the workers come.
Land is an essential element in agricultural production. Access to
and ownership of land is an important social indicator for those who
workers own or have access to less than one acre of land. The mean and
median farm size are 2.2 and 1.5 acres, respectively. With regard to
the size of plot, Jamaica and the small islands appear to have a fairly
similar pattern. Barbados is again an exception. For the few
individuals who do own farms, 82.3 percent have half an acre of ground.
The relatively small farm size in Barbados is reflected in the average
plot size of only .8 acres.
To say that the plots in the Caribbean cultivated by H-2 workers are "small" presumes that there is some relative standard to which the
data on size of plot is being compared. In other words, it is
necessary, for a more complete interpretation of these results, to place
*the findings from the survey of the H-2 workers into a broader context
of the agrarian structure of the societies fran which they are
recruited. This avenue of inquiry poses a number of difficulties that
are both of a conceptual and of an empirical nature. Secondary sources
of data, for example, may not employ the precise definition of "farm size" as that which we have used in this study. On a more practical
level, it should also be noted that the requisite information, which
also conforms to the same time period of the interviews that were
carried out in Florida, is not always available for the various islands
that participate in the labor program.
These drawbacks are less severe in the case of Jamaica where data from previous agricultural censuses offer useful reference points for
comparative analysis. The most recent figures that are available to us at this time only go through 1968. Although these data are out of date,
the pattern of land distribution tends, as a general rule, to remain fairly stable over time. The available census data thus provides an
approximate notion of how the farming background of cane cutters relates
to the larger context of Jamaican agriculture. Conclusions drawn fran
such an analysis must, necessarily, be treated as tentative.
The data on the distribution of the number and the acreage of farms by size groups in Jamaica are shown in Table 12. As noted above,
information from 19541 through 1968 reveals a fairly stable pattern, at least insofar as the distribution of cultivated farm land is 'concerned.
In 19541 about 13 percent of the agriculturally productive land was found
in farms that were less than 5 acres in size. The comparable figure in
, 1968 is 14.9 percent. At the other end of the distribution, it is clear
that well over half of the acreage (since 1961) is held by farms that
exceed 100 acres in size.
A separate, but related, indicator of the structure of Jamaican agricultural sector is the percentage distribution of farms by size of
plot. These data are presented in the lower panel of Table 12. The
estimates since 1954~ show a moderate,. hut nuiite mrkd, increase in the
Nature of Land Tenure Arrangements By Island
Tenure Arrangement Total Jamaica Barbados Is lands
Own 29.1 30.1 27.8 22.9
Leas e/Rent 34.4 34.9 33.3 31. 3
Family Land 28.7 27.4 27.8 37.5
Sharecrop 1.3 1. 1 5.6 2.1
Other 6.4 6.5 5.6 6.3
Total 100.0 100.0 100.0 100.0
Florida during the 1980-81 season. The sample contains individuals that
are of different ages, and with varying degrees of previous experience in the H-2 labor program. As noted in another section devoted to the
analysis of housing arrangements, one would expect the form of land
tenure to change according to the stage of the individuals' life cycle,
and/or as a result of the number of previous seasons that were worked
(two variables which are themselves intercorrelated). More
specifically, it is likely that young men, early in the stage of family formation and capital accumulation, would show a greater dependence on
family networks relative to older, more experienced workers. At a more
general level of analysis, changes in land tenure reflect the dynamic
~relationships that exist between the life cycle of the individual, the
character of his agriculture production, and degree of insertion into
and reliance upon the family system.
To investigate this hypothesis the form of land tenure, which is
presented in aggregate Corn in Table 13, can be broken down by the age of the respondent, and the number of past seasons worked in the United
States. These results (not shown) confir the expectations posited
above. The percentage of the H-2 labor force that owns land is 21
percent for those with less than two years of previous work experience
in Florida. As the nuber of past seasons increases, the proportion
owning land rises to nearly 40O percent (for those who have worked seven
or more seasons in the past). Conversely, the number who cultivate
family land is larger for newcomers to Florida (33.3 percent) compared
to those who have repeatedly engaged in stateside employment C13.3
The same patterns hold true for age. Older workers (40 years and
older) are more likely to own their own land (84.2 percent) compared to younger men (37.5 percent for those 20 to 24 years of age). Similarly,
47.5 percent of men 20 to 24 years old cultivate family property as
compared to only 10.8 percent of the individuals who are over 40 years of' age. As in the analysis of other variables, these findings confirm
the importance, especially among young men, of family netwoks in the
economic life of individuals in the Caribbean.
; ~~5. Ueo ao
If the forms of land tenure in the Caribbean are complex, so too
are the various mechanisms corrionly employed to cultivate land. These strategies are particularly important in the case of migrant laborers.
Tn ivd. l nh ar 3n thei hom country mus make allowance fourth
Selected Indicators of Rural Labor Practices and of the Impact of Emigration
A. Labor Exchanges in the Previous Year Percent
Exchange with neighbors 14.9
Exchange with friends 36.2
Exchange with family 5.0
B. Who tends to farm during absence
No one 5.2
Other family members 37.1
Paid workers 8.4
of the way in which individuals provide labor for their agricultural plots.
The responses to questions regarding the employment of labor are presented in Table 14. The practice of informal labor exchange, an arrangement that is widespread in the Caribbean, is clearly revealed by these findings. Two-thirds of the sample indicated that they did engage in some form of labor exchange during the previous year. These arrangements were most often carried out with friends (36.2 percent) or with neighbors (14.9 percent) who live in the same vicinity.
An important question that arises in this context is the way in which the emigration of farmers from the islands affects the labor process on the farm. It is sometimes argued that the absence because of migration of a principal worker from the household unit will have a positive effect on the rural labor market at the place of origin (see Griffin, 1976). These expectations are grounded on a set of inter-related assumptions, including the assumption that there is no labor surplus. Two key premises include the idea that absentee workers reduce the supply of labor on the islands, and that the money that is earned abroad is used to hire other laborers to take their places. According to this line of reasoning, stateside employment will have a positive effect in a double sense: by reducing the supply of labor-in the countryside the general wage level in the rural area is driven upward. Second, the influx of remittances permits the employment of individuals who would not otherwise be hired. The overall effect of these transformations would be arise in the income levels of the population. In the long run, the effect is to achieve a more equitable distribution of income.
The framework summarized above can be questioned on a number of grounds, especially the assumption that there is little excess labor. By whatever conceptual issues are at stake here (which we will not delve into at this point), the model nonetheless presumes that absent migrant workers in fact replace themselves with hired hands using funds that are, either directly or indirectly, related to the migrant's employment outside of his place of residence.
The findings presented in the lower panel of Table 14 are not consistent with this expectation. In a small and relatively insignificant proportion of cases (5.2 percent) no one at all takes over the tasks when the individual leaves the country. For the most part, however, it is the immediate or extended family that fills in the gap. According to nearly half of the respondents (41.8 percent), the responsibilities associated with the farm fall to wives and children. "Other family members" account for 37.1 percent of the labor needs. Taken together, this implies that, in about 79 percent of the cases it is the family that takes the place of the migrant worker while he is employed in the United States. Only 3.4 percent actually hired labor in their absence. These findings suggest that the effects on the rural labor market, assumed to-operate via the mechanisms noted above, may be less important than is sometimes assumed.
6. Agricultural Production, Consumption and Marketing
A final consideration in this section on agriculture is the nature
of the crops that are produced by workers who participate in the H-2
program. Three primary objectives serve to organize the presentation of
the data. The first is to provide a general profile of the types of
commodities that are produced. In addition, it is necessary to examine
inter-island differences in the types of agricultural output. A final
consideration is an attempt to discern the extent to which different
crops are produced primarily for sale or for consumption. The crops
most frequently cited by the respondents are presented in Table 15.
They include bananas, oranges, yams, cane, peas and vegetables.
Inter-island differences, at least insofar as such differences pertain
to the range of crops planted, are relatively minimal. All of the basic
foodstuffs noted above are produced, to a greater or lesser degree, in
each of the islands that form part of our sample.
When we turn our attention to the extent to which particular
agricultural crops are cultivated for consumption purposes, or for sale,
significant differences within and between islands do appear. Before
proceeding with a detailed analysis along these lines, it is important
to note the limitations that pertain to the data at hand. Under ideal
conditions one would want, for each individual in the sample, a precise
indication, not only of what he grows back home, but also information on
the proportion of the total crop that is consumed or marketed. Further
refinements would place a monetary value on the agricultural output.
Such a data set could identify the relative degrees to which a
particular crop is a source of income, a finding that would, in turn,
provide a quantitative assessment of the importance of farming to yearly
The questionnaire used in this study falls short of this ideal. We
do not, for example, have adequate information as to the precise
proportion of the agricultural output that is marketed or consumed.
After eliciting from the respondent information as to the crops that he
planted, subsequent items merely asked if these crops were grown
primarily for sale or for consumption. The intermediate response
("consume some/sell some) is ambiguous. Within this category we have no
way of knowing what the relative proportions (between selling and
consuming) might be. Our measure of consumption or marketing is
admittedly crude. The findings must, therefore, be considered as little
more than approximations.
A second point is the generalizability of our findings. Current
market forces undoubtedly have an effect on the distribution between
consumption and sale. Price fluctuations in a competitive marketplace,
government incentive policies, interest rates and other factors,
including the costs of transportation and of fertilizers, influence the
relative profitability of commodities from one year to the next. The
findings presented here refer primarily to 1980, the season prior to
when the interviews were carried out in Florida. These patterns are
probably relatively stable over time. However, it would be unwarranted
to generalize our findings to other years without first confirming the
stability of the various factors that affect the costs of production.
With these caveats in mind, the data presented in Table 15 do
provide us with a general, but nonetheless significant, insight into the
character and destination of agricultural production among H-2 workers.
The results presented in the first column, for example, clearly reveal
the importance of farm production for the maintenance and the
reproduction of the worker and of his household. With the exception of
sugar cane, somewhere between a third and a fourth of the respondents
indicated that all of the agricultural production was devoted
exclusively to their own consumption. In this regard, yams play an
especially important role. Of the farmers interviewed, 34.7 percent
cultivated yams only for their own use.
As units of both production and consumption, rural households, in
order to generate monetary income, also market some portion of their
surplus product. The importance of this aspect of the agricultural
sector in the Caribbean is revealed by the substantial proportion of
respondents who claim to produce various crops solely for the purposes
of sale. For the sample as a whole, bananas, oranges, cane, peas and
vegetables were of particular importance as commodities. The percentage
of respondents who engaged in the production of these items exclusively
for sale ranges from 28.7 percent in the case of yams to over 60 percent
for sugar cane.
The complexity of the information presented in Table 15, hinders
the ready identification of key patterns that exist across islands. To
simplify these findings, a simple index is constructed. The ratio of
the percentage of respondents who market all of crop X to the percent
that consume all of the same product provides a crude indicator of the
degree to which X is a marketed commodity among our population of H-2
workers. A high number would reflect, although in a very crude fashion,
the extent to which a given crop is produced solely for the purposes of
sale. For the sample as a whole (column one), it is clear that sugar
cane, with an index value of 3.9, is a product that is more often
marketed than yams (0.9), oranges (1.4) and peas (1.4).
The summary index is useful for inter-island comparisons. A
cursory examination of the values in parenthesis indicates that the
index of marketability is consistently lowest in the case of Barbados.
Bananas, oranges, yams and cane are produced there, just as they are on
the other islands. At the same time, not one respondent indicated that
these agricultural commodities were solely market-directed. The latter
does not imply that such products as bananas and oranges are never sold
(as indicated by the often substantial proportion of respondents who
fall into the intermediate category of consuming some/selling some).
However, the findings do reflect the fact that agricultural production
is less important to Barbadians as a source of monetary income. The
highest values, by contrast, are found in the small islands. Jamaica
falls somewhere between these two extremes.
These findings have implications for the objectives of this study
Consumption and Marketing Pattern for Selected Crops
Crop Total Jamaica Barbados Islands
BANANAS consume all 25.9* 28.1 75.0 11.8
cons./sell 35.8 35.4 25.0 38.2
sell all 38.4 36.5 --- 50.0
(Index) (1.5)** (1.3) (0.0) (4.2)
ORANGES consume all 28.1 36.4 66.7 33.3
cons./sell 32.3 31.8 33.3 50.0
sell all 39.6 31.8 --- 16.7
(Index) (1.4) (0.9) (0.0) (0.5)
YAMS consume all 34.7 35.8 62.5 22.9
cons./sell 36.6 35.8 37.5 42.9
sell all 28.7 28.5 --- 34.3
(Index) (0.9) (0.8) (0.0) (1.5)
CANE consume all 16.1 16.7 20.0 50.0
cons./sell 20.8 20.8 80.0 50.0
sell all 63.0 62.5 -- --
(Index) (3.9) (3.7) (0.0) (0.0)
PEAS consume all 28.1 28.0 40.0 --
cons./sell 32.3 31.7 40.0 100.0
sell all 39.6 40.2 20.0 ---
(Index) (1.4) (1.4) (0.5) (0.0)
VEGETABLES consume all 23.8 25.9 38.5 8.0
cons./sell 40.2 37.0 46.2 56.0
sell all 35.9 37.0 15.4 36.0
(Index) (1.5) (1.4) (0.4) (4.5)
* In percent
** Index = sell all/consume all
and more generally, for conceptualizing the way in which transnational migration affects sending societies. It is clear, for example, that the population of H-2 workers, although they share many characteristics in common, should not, for the purposes of analysis or for policy formulation, be treated as a homogeneous group. Moreover, the inter-island differences in agricultural production is one factor (among many others) which suggests that attempts to assess the impact of the H-2 program at the place of origin must be sensitive to the fundamental differences that exist between islands, and to the variability between individuals recruited to harvest cane in south Florida.
The questionnaire applied to the sample of West Indian cane cutters included several items designed to obtain information regarding the characteristics of the respondent's household back on the island. These data provide the basis for further examination of the role of stateside employment in the lives of those who participate in the H-2 labor force. Variables such as average household size, the dependency ratio and the number of other members who have jobs provide an indication of the importance of seasonal migration to the domestic unit of which the worker is a part.
The results shown in Table 16 indicate that workers belong to
households that, on the average, contain over 6 people. Nearly half of the members of the unit (3-0) are under 15 years of age; most of them are enrolled in school. Fewer than two other household members are employed outside of the household. Since the vast majority of cane cutters farm land in the Caribbean, it is likely that many of the individuals that form part of the domestic unit, including children, participate in productive activities of one sort or another.
A commonly used indicator of the degree of economic dependency in a population is the age dependency ratio. This measure is defined as the ratio of the combined youth and aged population (ages 0-14 and 65+) to the population in the intermediate, presumably economically active, age range (15 to 65), multiplied by 100. This value provides a rough indication of the degree of dependence on the economically active population, a measure that is useful for international comparisons.
The age dependency ratio varies widely from country to country. In 1970 developing nations such as Nigeria, Bangladesh and Brazil the ratio exceeded 100. On the other hand, in more developed countries such as Denmark and Japan the ratio was around 50. The differences are primarily determined by the proportion of children in a population, which, in turn, depends on fertility levels. Estimates derived from the survey suggest that the dependency ratio is around 90 for the population of West Indian cane cutters in 1980/81. Taken together, these characteristics of the workers' household indicate that they support a substantial number of people with their earnings from the U. S.
Household Size 6.3**
Number under age 15 3.0
Number in School 2.6
Number 15 to 64 years of age .*
Number over age 65 0.1
Number of other members with jobs 1.6
* Average values, total sample
**Includes the respondent
D. Ovrsa Eanig a Rmitace
1..Earning in .tte United States
Central to the study of the H-2 program is the analysis of workers'
wages. the wage earned in the United States provides the basis for
~estimating, if only in very approximate terms, the magnitude of money
and goods remitted to the Caribbean, a topic to be taken up in the next
Estimates of income earned, derived frcm individual responses to
questionnaire items, are subject to a number of problems. The
information is generated on a self-report basis, with no feasible way to
directly verify the accuracy of the responses given. The latter may be
subject to a number of potential sources of error. These include: lapses in memory, a deliberate under or overstatement of the true
figures and, possibly, the reluctance on the part of individuals to
respond at all to what may be considered a sensitive or private matter.
To mnmize these potential sources of bias, numerous steps were
taken. In the interest of enhancing the accuracy of the income data,
individuals were reminded, both of the anonymity of their responses, and
of their right to decline to answer any question that they might be
reluctant to address. To minmilze memory error, questions as to gross
and net income figures were restricted to the previous two week pay
The fortnight prior to the interview may, of course, not be
representative of a typical work period. Wages vary according to weather and other factors that affect worker productivity and earning potential.
To investigate for this possibility, a follow-up item was included in
the questionnaire. It asked whether the worker's pay during the
reference period was "above," "below" or "about the same" as his "usual
pay." Fifty percent of those interviewed indicated that it was about
the same. Those responding that it was higher or lower were evenly
split. The distribution of these answers suggests that, for the most part, the previous fortnight was not an atypical work period. Second,
the responses imply that the equal proportions of over and
underestimations of the average wage would tend, for the sample as a
whole, to cancel one another out.
Dataon e rigsA. drng4 the~ r tnigh4tL.4 n'n'4." pr.r t 4h irview*, n ae
How accurate are these figures? Ideally, individual responses to this question could be checked against the worker's actual pay receipt.
Under the conditions in which the interviews were carried out, such a direct method of verification was not possible. However, there are several indirect methods by which to evaluate these data. It should be noted, for example, that the overall refusal rate was very low (less than 2 percent). This does not, in and of itself, indicate that the answers given are necessarily correct. Nonetheless, a low refusal rate
does suggest that the question on earnings was not one that workers were predisposed to avoid outright.
In addition to an analysis of the refusal rate, the accuracy of the data on earnings can also be assessed by investigating the internal consistency of the average estimates presented. As discussed in Part I of this report, several standard deductions are withdrawn fron the worker's gross pay check. These include a $5.20 daily deduction for meals (or $72.80 for the two-week pay period). An additional 25 percent is deducted for the savings plan and BWICLO assessment. The average
gross wage estimated from the survey data shown in Panel A is $420. This implies that the 25 percent accounts for about $105. Taken together ($72.80 plus $105), these deductions amount to $177.80 for an average fortnight. If the survey estimates of earnings are consistent, then the mean gross wage ($420), minus the known deductions ($177.80), should come close to the average net wage reported by the individuals that were interviewed. The value generated by this procedure, $242.20 is very close to the estimate of net earnings ($245, as shown in Panel B).
Evidence that the fortnight prior to the interview (the reference period) was not atypical, a low refusal rate and the high internal consistency of the data indirectly suggest that the information on workers' wages generated by the survey can be accepted as reasonable estimates of actual earnings. However, for the reasons cited earlier in this sections, these data should be regarded as approximations of the true values.
The role of worker remittances is a key aspect of the seasonal
labor migration fron less to more developed countries, For the purposes of this section three issues are of central concern. The first objective is to estimate the magnitude of the resources that are sent
Workers' Gross and Net Incomes for the Previous Fortnight
Percentage Distribution of Respondents Small
A. Gross Earnings* Total Jamaica Barbados Islands
0-199 3.8 3.9 5.0 2.4
200-299 5.7 6.7 2.5
300-399 31.0 30.6 17.5 41.5
400-499 33.7 31.7 37.5 46.3
500-599 17.6 18.3 22.5 9.8
600-699 5.7 6.1 10.0
700 + 2.5 2.7 5.0
Total 100.0 100.0 100.0 100.0
Missing cases 13.4** 12.6 11.1 19.6
Mean Gross Wages $420 $419 $463 $405
Percentage Distribution of Respondents B. Net Earnings
0-199 26.1 25.4 25.0 31.9
200-299 45.1 45.7 32.5 46.8
300-399 22.6 22.8 30.0 17.0
400-499 4.0 3.6 10.0 4.3
500-599 1.8 2.0 2.5
600-699 0.4 0.5
7 0 0 + - --... ..
Total 100.0 100.0 100.0 100.0
Missing cases 5.2 4.4 11.1 7.8
Mean Net Wages $245 $245 $270 $233
* nU dollars
** Percent (includes refusals and missing data) 47
In the case of cane cutters who yearly come to Florida,. remittances to the place of origin are made in several ways. The contract agreement described in Part I stipulates that 23 percent of the worker's bi-monthly pay is automatically transferred to a bank account in the worker's hone island. This money, or part of it in the case of Barbados and the small islands, canl be retrieved at the end of the season. In addition to this automatic savings plan, workers on their own initiative send substantial amounts of money to family members and to friends via the mail. Wages are also used to purchase clothes and other items. Finally, at the end of the harvest, each individual returns to the Caribbean with a certain amount of cash in hand.
The total value of these monetary (savings plan, money orders, cash) and non-monetary resources (purchased goods) provides a rough estimate of the magnitude of the remittances made to the sending
countries. The analysis that follows estimates the amount of money transmitted through each of the various methods. It also explores some of the social aspects of these transactions by investigating the relationship between the respondents in the survey and the recipients of remitted funds.
a. Saving hlan
As noted above, the contract requires that 23 percent of the worker's gross wage be deducted fran each biweekly paycheck and deposited in a savings account on the island of origin. The question addressed here is the following: What is the total amount of money remitted to the Caribbean as a result of the savings program?
A rough estimate of this value canl be generated by taking, as a starting point of the analysis, the average gross wage earned the fortnight before the survey, or $420 (see the section on "Earnings"). Of this amount, 23 percent enters the savings account for each individual ($96.60). The product of this average value times 251 (the number of workers that were interviewed) indicates that about $24,246 was deposited for the workers in our sample. This value, however, applies only to the reference period, and not to the amount sent into the savings program during the entire season. The latter can be estimated indirectly, provided that certain assumptions are met. Responses to another item in the questionnaire, for example, indicate that the cane cutters who were interviewed had worked an average of 133 days, or about 9.5 pay periods. If the fortnight preceeding the interview is a typical work period (see the discussion in the section on
"Earnings"), the product of $24,246. times the average nunber of
the sample is representative (for a discussion, see the Appendix B),the survey findings can be generalized to the labor force as a whole. The results of this procedure indicate that the total remitted to the Caribbean via the savings plan was about $7,764,000.
Needless to say, this figure should be regarded only as an
approximation. With this caveat in mind, the findings do provide some idea of the order of magnitude involved. As subsequent analysis will show, the $7.5 to $8 million that is deducted from the workers' gross wages and automatically returned to the islands through the savings plan is the largest component of all remittances to the Caribbean by the H-2 labor force.
b. Maile Remittances
The questionnaire was designed to elicit information from the
respondent as to the average amount of money mailed to the islands, and the number of times that he had done so while employed in the United States. Since the interviews in Florida were carried out late in the season on the eve of their departure, responses to the question about the frequency of remittances should provide a reasonably accurate picture of the number of times that money orders were sent during the 1980-81 season. In addition to amounts and to frequencies, other items in the survey identified the person to whom the money was sent, and the way in which these funds were to be used back home. The results of the analysis are presented in Table 18. As shown in raw A, an average of about $71 was sent to the first recipient mentioned by the worker* wives, girlfriends and mothers. These individuals received money orders an average of about 7 times during the season (row B).
The pattern observed for the primary recipient changes
substantially when we analyze the amount, frequency and the destination of the funds sent to recipients 2, 3 and 4. The person receiving wages from the worker in Florida shifts from wives, girlfriends and mothers (the main receivers) to sons and daughters, sisters and brothers, and to more distant associates ("friends" and "others"). This change is accompanied by three others worth noting. First, the average amount sent steadily drops from an average of $71 for recipient 1 to a low of $15 for the fourth receiver. Second, the frequency of sending money orders also declines from 7 (recipient 1) to 3.3 times (recipient 4). Finally, as is clear from the data shown in row C, almost all (92.3 percent) of the workers in Florida mailed a portion of their wages to the islands at some point during their employment in the United States. However, the proportion of the labor force that sent money to persons other than the first recipient drops off sharply: 43.8 percent of the workers sent money to a second recipient; only 8.8 percent to a third person.
These findings provide some insight into the relative priorities that determine the amount, frequency and the destination of the wages that are remitted to the islands by mail. The primary importance of the immediate family is confirmed. Other, more distantly related
Average Amount and Frequency of Money Mailed Home, By Category of Recipient
1 2 3 4
A. Average Amount Sent $71 $38 $32 $15
B. Average Frequency 7.1 5.1 4.8 3.3
C. % of Labor Force 92.3 43.8 8.8 1.6
N of Cases 232 110 22 4
D. Recipient M
Wife 53.0 5.7 4.1 --Girlfriend 23.8 25.0 13.3
Mother 15.5 38.5 --- --Father 2.2 8.6 4.1
Son/Daughter .4 3.9 17.0 --Sister/Brother 1.7 4.3 16.4 --Friend .4 3.5 16.0 --Other 2.5 10.5 29.1 --Total 100.0 100.0 100.0 --*Insufficient number of cases
individuals, are shown to be of secondary importance. More generally,
these results underscore the notion that the question of remittances is
not simply a matter of the magnitude of the resources that are
transferred. Analysis of this phenomenon must also account for the
complex social networks within which these transactions occur.
To what use is this money put? According to the individuals
interviewed in the survey, most of the wages mailed to the islands goes
to support their families. About three-quarters of the respondents (76
percent) said that the principal purpose of sending money home was to
cover daily household living expenses. The other most cited reasons
include child support (6.6 percent) and the schooling of children (6.1
A final objective is to estimate the total amount of money remitted
to the Caribbean via the mail. A rough approximation can be obtained by
multiplying (for each individual in the sample) the average amount of
money mailed by the frequency of sending mail orders. The sum of these
values across individuals in the study indicates that a total of
$141,643 was sent to the Caribbean by workers interviewed in our sample.
If we generalize from the sample to the total labor force, these
findings suggest that the sum of all mailed remittances amounted to
approximately $4,774,000. Of this total, about 77.8 percent (or
$3,709,000) was sent to Jamaica; 13.8 percent ($656,000) was mailed to
the small islands; and 8.5 percent ($404,000) was destined for Barbados.
The desire to work in the United States for wages that exceed what
can be earned on the islands is unquestionably the principal factor that
motivates individuals to enter the H-2 labor force. A second reason is
the opportunity to purchase items which are then taken back to the
Caribbean at the end of the harvest. A substantial amount of money is
thus spent on goods to be taken home by cane cutters while they are in
the United States. The majority of these purchases take place in towns
that lie closest to the labor camps, although it is not unusual for
workers to travel to Palm Beach or to Miami for this purpose. It was in
response to this demand over the past 38 years that an entire business
district evolved in Belle Glade, Florida.
The analysis of the purchases made by the H-2 labor force while in
the United States can thus be seen in a double light. From the
standpoint of the merchants in south-central Florida the presence of
these workers constitutes a market for a variety of goods. On the other
hand, from the standpoint of the sending societies, the items bought
constitute a portion of the remittances that are sent back to the
The analysis of the workers' spending patterns involves at least
two dimensions: the value of the items that are purchased and the type
of things that are bought. The latter provides another indication of
the multiple ways in which the H-2 labor program influences sending
societies. For the purposes of this investigation a distinction can be drawn between "capital goods" and "consumer goods." Capital goods can be defined as the materials necessary for production, trade or commerce. These items, which include tools and equipment, are assumed to enhance the individual's productivity and his capacity to accumulate wealth back on the islands. Capital goods can thus be distinguished from other items which are consumed in order to maintain and to reproduce the worker and his family (e.g., clothes, food), or which serve only to elevate the quality of life (as through the "consumption" of televisions or radios). On an aggregate level, widespread importation of capital goods would presumably increase production, largely in agriculture and thereby constitute a significant contribution to the sending islands' development.
The impact of the H-2 labor program on the Caribbean is thus determined, in part at least, by the character of the consumption patterns of the workers involved. The relative proportion of expenditures allocated to capital or consumer items is one way to conceptualize this process. If most purchases are for capital goods, the remittances in kind are likely to have a greater economic impact on the household's capacity to produce and to accumulate than would be the case if consumer goods constitute the bulk of the individual's expenditures.
The distribution of the principal purchases made in the United States is presented in Table 19. The percentages indicate the proportion of the total sample that purchased a given item. The findings show that nearly all workers (97.4 percent) bought clothes and personal items. About 60 percent acquired entertainment equipment such as radios, tape players and televisions. Between 10 and 15 percent spent their wages on food, general household goods and appliances. Only a small proportion purchased tools (2.6 percent) and transportation equipment (2.0 percent).
The spending patterns revealed in Table 19 indicate a clear preference for consumer as opposed to capital goods. A closer examination of these items, howeVer, suggests that the distinction between the two is potentially ambiguous. Of all of the items that appear on the list of purchases only tools clearly fall.into the category of capital goods, as defined here. The accurate classification of other items depends on how they are used. A motorcycle used only for Sunday drives is quite different from one acquired for the purpose of petty commerce and trade. Similarly, entertainment equipment may not necessarily constitute a consumer good. Radios and televisions brought back to the islands for sale in a shop could contribute to the viability, even expansion, of a small retail business.
We have no way of knowing how a given item is actually used once the worker returns to his place of residence. However, a follow-up question in the survey did inquire as to the intended used of each item that was bought by the individual. Responses to this question indicate that the vast majority of the purchases were solely for personal and household consumption. These findings are consistent with the intended use of money mailed home, as shown in an earlier section. We therefore
Items Purchased in the United States
Item Percent Purchasing Item
Personal Items 97.4
Entertainment Equipment 59.4
Food/Household Goods 15.3
Transportation Equipment 2.0
NOTE: Personal items include: jewelry, consmetics, luggage
Entertainment equipment includes: radios, tape decks, stereos,
tapes, televisions, cameras
Food/Household goods include: food, sheets, mattresses,
plates, soap, kitchenware, curtains
Appliances include: mixers, stoves, refrigerators
Transportation equipment includes: bicycles, motorcycles
conclude that, despite the fact that the exact classification of
particular items may be ambiguous, the buying patterns of West Indian
workers emphasize consumer rather than capital goods.
A final consideration is the total (stateside) value of the items
that workers remit to the Caribbean. The sum of the amount of money
spent on purchases in the United States comes to $135,000 for those
individuals in the survey. To the extent that the workers interviewed
represent a random sample of the universe from which they were drawn
(see Appendix B), the spending patterns observed can be assumed to be
representative of all H-2 workers in the year of the survey.
Generalizing to the population of cane cutters in 1980-81 suggests that
the labor force as a whole spent approximately $4,551,000 on items that
were either sent or were carried back to the islands of origin.(20)
d. Cash in Hand
Compared to the other ways in which wages are remitted, the amount
carried back in cash is relatively small. The distributions shown in
Table 20 indicate that the vast majority (80.7 percent) carry $500 or
less on their return. Only 4.1 percent go back with more than $1,000 on
their person. The average amount, as shown in the lower panel, is $407.
For all individuals in the sample, the total sum of the money returned
in cash comes to $56,220. Using the same procedures applied in earlier
sections, this value, generalized to the labor force as a whole,
suggests that a total of approximately $1,895,000 was remitted in cash
to the Caribbean at the end of the harvest season. While small, this
particular remittance is of special significance to the worker, since it
is directly in his possession and is in the form of cash.
e. Remittances: Summary Tabulation
The analysis of the various forms by which resources are returned
to the islands, taken together, indicate that a total of about $19
million was remitted to the Caribbean during 1980-81 as a result of the
H-2 program. A summary of the respective amounts is as follows:
Savings Plan $7,764,000
Money orders 4,774,000
(value in USA) 4,551,000
Cash in hand 1,895,000
With regard to the accuracy of these estimates, it is important to
note several factors. First, it should be recalled that the figures are
Amount of Money Remitted to the Islands in Cash
Amount carried in cash*
A. 0 $205 31.4
$251 $500 49.3
$501 $1000 15.2
$1001 + 4.1
B. Mean $407
* U.S. dollars
derived from self-reports by individual workers interviewed in Florida. While the data are internally consistent, and appear to be free of systematic bias, they are subject to error (as noted in the various sections). Second, the sample size of 251 is relatively small thus increasing the confidence interval around each estimated value. When generalized to the labor force as a whole, the margin for error is magnified. For these reasons, and others that could be mentioned, the figures presented should be regarded, not as exact measurements, but rather as rough approximations of the true values.
E. Savings, Investment and Productivity
1. Capital Acuulto
Conventional perspectives on labor mobility typically assume that the migration of workers from less to more developed countries fosters
development at the place of origin. The mechanisms by which this is presumed to occur are many. The remittances of migrant laborers are viewed as instrumental in restoring a balance of payments, and in stimulating savings and investment. Upon their return, repatriates serve as agents of change by applying the ideas and the skills acquired abroad to establish farms, businesses and other enterprises conducive to increased production. By accelerating capital formation and technical
change on small peasant farms, migration is assumed to improve the distribution of income in rural areas. According to the equilibrium model, the international movement of labor thus leads to a gradual convergence in the levels of economic growth and social well-being (see Bohning, 1975; Rempell and Lobdell, 1978; Swanson, 1979).
Applied to the case of West Indian cane cutters, the perspective outlined above predicts that the wages earned in the United States provide the resources to start or to expand small businesses, or to buy land and equipment so as to increase agricultural output and productivity. While there is little doubt that this expectation is met in the case of individual workers the central question is the degree to which this phenomenon typifies the H-2 labor force. We have already
examined some evidence of the preference for consumer over capital purchases, the results of which challenge the equilibrium model.
Further analysis of our survey data provides additional insight
into the questions raised above. At the outset, it is important to note that the wages received harvesting sugar cane in the United States substantially exceed the general wage levels in the Caribbean. In Barbados and Jamaica the per capita income in 1980 was $2,685 and $1,406, respectively. As noted in another section of this report, H-2
workers earned an average gross wage of about $4,000 during the 1980/81 season. But do these resources enable the worker to accumulate sufficient productive assets to improve his income-generating capacity,
as the equilibrium model suggests?
A complete answer to this question would require a comparative
analysis with a control group of individuals who had never participated in the program. Also required are indices that adequately reflect the consequences of having had the experience of cutting cane in Florida. Our data meet neither of these conditions. Nonetheless, a comparative analysis .kijthi the sample of workers who were interviewed in the United States provides a rough indication of the effects of the program. More specifically, workers who have returned to harvest sugar cane more than
three seasons ("oldtimers") can be compared with "newcomers" (three seasons or less). With the appropriate statistical controls for age and
other variables, we would expect oldtimers to have accumulated more productive assets relative to newcomers.
orA comparison of the workers who have returned to Florida for four
more seasons with the rest of the labor force reveals few differences between the two groups in terms of capital accumulation. This can be seen, first, in the analysis of the number of acres that the worker owns or has access to back home. As shown in column 2 of Table 21, the mean size of farm for the sample as a whole is 2.3 acres. Using Multiple
Classification Analysis (MCA), it is possible to disaggregate the effects of other variables. Specifically, do oldtimers own larger plots of land compared to newcomers?
The figures shown in column 5 represent the deviation from the
grand mean (2.3) that is associated with the number of previous seasons worked. Newcomers own an average of 2.16 acres (2.3 0.14= 2.16). Oldtimers own 2.'44 acres (2.3 + 0.14 = 2.414). We can conclude that the differences between the two groups is relatively small, or about .3 acres. The deviations from the mean associated with island of origin can be interpreted in the same way.
The figures shown in column 5, however, are unadjusted estimates. That is, they do not account for the effects of other variables that could influence the number of acres owned. The geophysical differences between the islands are one set of factors that are important. Other variables include the age and education of the respondent. It is therefore necessary to estimate the deviations from the grand mean after removing the possible effects of other independent variables (island of origin) and covariates (age and education).
These findings are presented in column 6. Comparing the adjusted (col. 6) with the unadjusted (col. 5) deviations from the mean reveals very little change. We can conclude that, after controlling for other key variables, newcomers and oldtimers own about the same amount of land. Repeated participation in the H-2 labor program does not, therefore, appear to lead to the accumulation of agricultural resources.
Using land ownership as an measure of capital accumulation is open
Multiple Classification Analysis of the Effect of work Experience on Land Ownership and Earnings Dependent Grand Independent Deviation From the Mean
Variable Mean Variables N Unadjusted Adjusted'
(1) (2) (3) (4) (5) (6)
A. Acres 2.3 SEASONS2
Newcomers 95 -0.14 -0.15
Oldtimers 96 0.14 0.14
Barbados 5 -1.50 -1.50
Jamaica 159 -0.03 -0.04
Small3 27 0.40 0.50
B. Pay4 253.4 SEASONS
Newcomers 105 -21.7 -28.5
Oldtimers 110 20.6 27.0
Barbados 13 21.6 10.0
Jamaica 174 -0.5 -0.1
Small 28 -6.4 -4.3
Controlling for independent variables (SEASONS, ISLANDS) and covariates
(age of respondent, level of primary education).
2 Newcomers: those who have worked three seasons or less
Oldtimers: those who have worked more than three seasons
3 Includes: St. Lucia, St. Vincent and Dominica
4 Net Pay for the fortnight prior to the interview
to question in the Caribbean where the forms of land tenure are often
quite complex. Similar analysis were therefore carried out using a
number of other variables assumed to reflect the individual's assets.
These included home ownership, and the number of goats, cows and mules
that were owned by the respondent. In all cases the results were the
same; few if any differences were found between newcomers and
To argue that participation in the H-2 program fails to enable
individuals to accumulate productive assets is not to say that the
opportunity to work in the United States is unimportant to them. To the
contrary, nearly 100 percent of those interviewed expressed the desire
to return the following season. The reasons are not hard to find. As
noted in previous sections, analysis of numerous variables in the survey
demonstrates the importance of stateside employment in the lives of
individuals, and for the households and the extended kin networks of
which they are a part. These data, and the statements by the workers
themselves, indicate that the opportunity to work in south Florida
provides critically needed resources to those who are fortunate enough
to enter the seasonal labor stream. But the findings provide little
support for the hypothesis that individuals who regularly return are
able to accumulate a greater degree of productive assets compared to
others. Together with other data (on expenditures, the use of
remittances and mail orders), these findings suggest that, rather than
serving as a source of investment capital, the wages earned harvesting
sugar cane in Florida are primarily devoted to the maintenance and the
reproduction of the worker and his household back in the Caribbean.
Analysis of net wages earned the fortnight prior to the interview
indicates that previous experience harvesting sugar cane is associated
with greater worker productivity. As shown in column 5 of Table 21
(Panel B), newcomers earned about $22 below the grand mean; oldtimers
earned about $21 above it. This discrepancy widens after introducing
statistical controls for island of residence, age and education (compare
columns 5 and 6). Interpreting the adjusted deviations from the mean
(col. 6) we conclude that, net of controls, newcomers earned an average
of $225 and oldtimers about $280, a difference of $55.
These results are consistent with the analysis of other variables
that are related to worker productivity. The more experienced preferred
workers are less prone to injury (2.1 days per season vs. 5.9) and lose
fewer days on the job as a result of illness than do the less
experienced pool workers (1.5 vs. 2.4). These findings, taken together
with estimates of net earnings, suggest that individuals with extensive
previous experience are more skilled, less subject to accidents and
sickness, and are substantially more productive as compared to novices.
The system of migrant labor that has evolved (with its method of
selecting preferred workers from one season to the next) thus provides
the employers a continued supply of highly productive labor.
3. Inetmn lans
Given the magnitude of the resources that are remitted to the
islands (see "Remittances"), it is important to investigate the ways in which earnings are allocated at the place of origin. Respondents were asked how they planned to spend the money they earned in the United States once they returned home. This question refers primarily to the
income accumulated in the compulsory savings program and to the cash taken back in hand. Utilization of these funds is another indication of the importance of the H-2 program to the development of the five West Indian Islands. The list of activities and items targeted for investment were then collapsed into eight general categories. As shown in Table 22, 60 percent of the respondents interviewed intended to apply their resources to housing. This includes those who planned to build or add on to an existing structure, as well as individuals who intend to buy land or furnishings for a house. (Because each respondent could specify more than one investment, the columns in the table add up to more than 100 percent).
About 20 percent of the sample indicated that they wanted to buy land to farm, or intended to purchase farm equipment, fertilizers, or seeds. One in three planned to buy livestock. Just over ten percent expected to use their money to cover living expenses. These patterns differ somewhat by island, as shown in columns 2, 3 and 4. Livestock and farm-related investments are more salient in Jamaica and the smaller islands, but are less so in the case of Barbados. This pattern again confirms the non-agricultural origins of the Barbadian workers. Similarly the proportion of individuals who intended to allocate earnings to business ventures and savings is higher in the smaller islands compared to Jamaica and Barbados.
While there is some indication of investment of savings and cash in hand in farming and livestock, the most prevalent category by far is housing and living expenses. This pattern once again confirms the importance of consumer and subsistence spending from income generated by cutting cane.
III. SUMMA RY AND CONCLUSIONS
The principal objective of this study has been to provide an
empirical profile of the West Indian laborers who seasonally migrates to the United States to harvest sugar cane. In spite of the program's nearly four decades of existence, little is known about either the labor force or the impact of its seasonal migration on the participating workers, their dependents, and the sending islands. To address some of these issues, we administered a questionnaire to a random sample of 302 workers during M|arch 1981, at the end of the 1980-81 harvest. In the concluding section of this report, we summarize the principal findings of this study. A more thorough discussion of the theoretical and policy implications of the research must await subsequent analysis.
A. Prnc~a indig
Fran the analysis of the survey data we derive the following major conclusions :
1. West Indian laborers have been recruited to cut sugar cane in south Florida since 1943. The longstanding character of this international arrangement implies that offshore workers are not a transitory or cyclical part of the process of sugar production in the state. Unlike other migratory streams, that may come and go in response to different economic circumstances in the receiving area, the West Indian labor force (although it does fluctuate in size from year to year) can be considered to be an intrinsic feature of Florida's sugar industry, at least as it is currently constituted. Policy decisions, either to expand or limit the H-2 program, must take this into account.
2. Analysis of the survey results indicate that there are ninerous similarities between the workers that are recruited from Jamaica, Barbados and the smaller islands (St. Lucia, St. Vincent and Dominica). At the same time the economies and geophysical characteristics of each one are not identical. These dissimilarities are reflected in analysis of several key items used in the questionnaire, such as the importance of agricultural production and the occupational background of the respondent. These findings suggest that that the H-2 labor force cannot be regarded as a homogeneous entity, and policy decisions based on the aggregate impact of the program on the sending societies must be sensitive to this diversity.
3. With the exception of Barbados, the West Indian labor force is recruited primarily from the rural areas of the Caribbean. In general
given was that of farmer or farm worker. Within the category of "primary occupation," the second most important jobs were those related to the construction industry. In this regard Barbados is an exception (a higher incidence of construction workers and a relatively lower
proportion of "farmers").
5. Over half of those interviewed also declared a second occupation mostlyy in farming or farm related activities and in construction). These findings indicate that, although cane cutters do have more than one job back on the islands, the range of occupational backgrounds is relatively limited.
6. The characteristics of the H-2 labor force noted above should be considered in light of the recruitment process that is carried out every season. Studies of migration (both internal and international) have noted the "positive" selectivity of migration streams (migrants being better educated, or more "motivated" than non-migrants at the place of origin). Such an empirical generalization, however, applies to those population movements that are voluntary. In this instance, however, the institutional mechanisms by which workers are chosen has a marked impact on the "selectivity" of the pool of migrants. The predominance of individuals with farming backgrounds, for example, is the outcome of the criteria deliberately imposed by the employers. It follows that accepted generalizations as to the characteristics of migrants found in the literature on this topic do not necessarily apply inthis case.
7. Over a quarter of the H-2 workers own or have access to less than one acre of land. The mean and median farm size are 2.2 and 1.5 acres, respectively. Average size of farm is much smaller among cane cutters recruited from Barbados (0.8 acres), a finding consistent with the results on occupational and farming background noted above.
8. In addition to the single farm unit, owned and operated by the head of a nuclear family, there are many other formal and informal ways by which individuals gain access to land. The family, and the broader network of kin and friends, play an important role in the way land is divided and cultivated. In terms of land tenure arrangements, analysis of the survey data indicate that young men, early in the stage of family formation and capital accumulation, show a greater dependence on family networks (e.g., cultivating "family land") compared to older workers (who tend to own their plots). These findings suggest that analyses of the impact of remittances and other transfers to the islands by the H-2 labor force are likely to have different implications for men at different points in their life cycle.
9. The absence of the worker because of migration to the United
States has repercussions within the individual's household. In terms of tending agricultural plots, there is a reallocation of family labor. For the most part, it is the immediate or extended family that fills in for the departed worker. There is little direct evidence that remittances are used to hire third parties.
10. Farming activities clearly play a critical role in the
sustenance of the worker's household. Between a third and a fourth of the respondents indicated that all of their agricultural production was devoted exclusively to their own consumption (this varies somewhat by crop). This does not mean, however, that the entire H-2 labor force is drawn from "subsistence" farmers. On the contrary, a substantial portion market some or all of the foodstuffs produced. In this way it can be said that those who come to cut sugar cane in the United States also contribute to the production of foodstuffs for the island populations through their agricultural activities (and that of family members, as noted in 9 above).
11. Workers belong to households that, on the average (including the respondent), contain just over 6 persons. Nearly half of the members of the unit are under 15 years of age (most of whom are enrolled in school). Few other members are employed outside of the household. Taken together, these data indicate that seasonal migrants to the United States contribute directly to the support of a substantial number of people at the place of origin.
12. The principal reason that motivates individuals to seek stateside jobs as cane cutters is that the wages earned in Florida exceed that which can be earned in the Caribbean. A second reason is to exploit the opportunity to come to the United States in order to purchase items that are either unavailable on the islands, or which are more expensive there.
13. Workers in the sample reported an average gross biweekly wage of $420. This aniounts to about $4,000 per season, from which numerous deductions are made (some of which are returned to the worker when he gets back to the island; see 14, below). The average net wage was $245 a fortnight.
14. Remittances to the place of origin are made in several ways.
The contract agreement stipulates that 23 percent of the worker's pay is automatically transferred to a bank account in the worker's home island, where all or some (depending on the island) can be later retrieved in local currency. Workers on their own initiative also mail substantial amounts of money to family members and friends. They also purchase clothes and other items. Finally, at the end of the season, each individual returns to the Caribbean with a certain amount of cash in hand. Extrapolating from the sample to the total population of workers in 1980-81, we estimate that the West Indian work force remitted nearly $19 million (including the U.S. value of goods purchased) to the sending islands.
15. The primary recip ients of mailed remittances are wives,
girlfriends and mothers. Others, such as sons and daughters, sisters and brothers, assume importance as secondary recipients. These results underscore the point that the question of remittances is not simply a matter of the magnitude of the transfers involved. Analysis confirms the importance of the complex social networks within which these transactions occur.
16. About $4.5 million is spent by the labor force on purchases
that are returned to the island. Analyses of these spending patterns, together with other information on the destination and purpose of mailed remittances, indicate that the wages earned in the United States are used primarily for the maintenance and the reproduction of the worker's household. Although a snail percentage of the work force does invest in capital goods, the majority of the the purchases and remittances are for consumption purposes.
17. After statistically controlling for key variables (age,
education, island), the results indicate that repeated participation in the H-2 labor program does not appear to lead to the accumulation of productive resources (as measured by land and other assets) by the individuals involved.
18. Analysis of the net wages indicate that those individuals who
have repeatedly come to Florida to cut cane are systematically more productive than novices. Their wages exceed that of newcomers to the program, and they are relatively less prone to injury and to sickness. The method used by employers to select preferred workers from one season to the next (60 percent of the labor force in 1980-81) thus appears to provide the industry with a skilled and productive labor force.
The international seasonal transfer of labor analyzed in this
report brings a selected group of men from five West Indian islands to the United States where they are able to earn wages in excess of those available locally. There are three direct beneficiaries of the program. First, it benefits the participating workers and their dependents. Given the persistence of poverty and widespread unemployment in rural areas of the West Indies, there can be little doubt that these men and their households are better off because of the opportunity to cut sugar cane in Florida. It allows them to earn a steady income for at least five months of the year and to save a proportion for use at hone. It also opens up the possibility of purchasing consumer goods which not only ensures household survival but also enhances the quality of life of its members. A second major immediate beneficiary is the worker's home
island. Through a mandatory savings plan and mailed remittances, substantial amounts of hard currency are yearly infused into economies suffering from chronic balance of payments and foreign exchange deficits. Finally, the Florida sugar industry benefits fron the H-2 program. For the past 38 years it has regularly recruited a highly skilled, motivated seasonal work force for an industry which depends, in part, because of ecological constraints, on manual harvesting of its crop. The implications of the H-2 program in Florida sugar, for the long run development of both the participants and the West Indies are less nilear, hn~yever. Althnug oun r data-u are not cmplet o~n t-hi s maitteor,
Selected Indicators of Five Sending Countries Jamaica Barbados St. Lucia St. Vincent Dominica
(Sq. km) 11,424 431 616 389 750
(1981) 2,225,000 256,000 124,000 116,000 79,000
(Annual Average) 1.6% 0.4% 2.0% 2.3% 1.2%
(per 1000) n.a. 17.2 31.5 32.2 21.8
(per km) 585 201 298 105
Adult Literacy 86% 98% 80% 95% 80%
GNP $2,936.0 mil $511.2 mil $ 99.2 mil $ 50.3 mil $36.9 mil
(Year) (1980 GDP) (1978) (1979) (1979) (1978)
Rate (1978) -3.5% 5.1% 4.3% 14.6% 3.9%
GNP $ 1,352 $ 2,010 $ 790 $ 485 $ 459
Exports $ 979 mil $349.5 mil $ 65.6 mil $ 27.3 mil $ 9.4 mil
Imports $ 1,129 mil $372.7 mil $105.4 mil $ 52.4 mil $39.4 mil
Account $ -275 mil $ -4.7 mil $-31.6 mil $ -16.0 mil -$14.1 mil
External Debt n.a. $ 61.7 mil $ 13.8 mil $ 7.8 mil $12.1 mil
Ratio n.a. 2.9% 0.9% 9.8% 3.6%
Rate 31% 15% 13% 20% 23%
Sugar Quota 6954 508 389 567 42
Source: Caribbean/Central American Action. C/CAA's Caribbean Databook.
Washington, D.C.: Caribbean/Central American Action, 1981.
All currency figures in U.S. dollars.
The ntxnber of H--2 workers involved in harvesting sugar cane in
Florida fluctuates during the season. At the height of the 1980-81
season about 8,460 West Indians were employed. The vast majority (82.2
percent) were from Jamaica, with the remainder from Barbados (6.0 percent), St. Lucia (4.6 percent), St. Vincent.C6.7 percent), and
Dominica (0.5 percent). In March, 1981, at the end of the harveSt,oa
sample of men was drawn from the total remaining population of 144,41
workers. Since the distribution of this labor force by place of origin
was known from existing industry records, a quota system was employed to
generate a representative sample of the universe of workers,
Individuals to be interviewed were chosen from the roster of people
living in each labor camp such that the number from the five islands was
proportional to each island's weight in the total mid-March population.
This procedure generated a total of 251 cases. The composition of the
work force by island in mid-March did not differ significantly from that
during the rest of the season. The reduction in the nuber of workers occurred just before our interviewing began as the harvest concluded on
some of the farms.
A representative sample of this size has a major drawback. If
Barbadians are only 6 percent of the universe, a sample of 251 cases
implies a total of only 15 individuals from this island. Similarly, the
absolute number from St. Lucia, St. Vincent, and Dominica is also quite low (12, 17, 'and 1, respectively). The small number of cases restricts
the possibility of carrying out between-island comparisons. To allow
for this possibility, Barbadians and those from the smaller islands were oversampled so as to have approximately 50 cases in each subgroup. The total number of interviews carried out, therefore, was a representative
sample of the population (251), plus additional interviews with
individuals from the smaller islands, for a total of 302 cases. In the analysis presented here the findings for the labor force as a whole are
derived fron the random sample of 251 respondents. Comparative analyses
across islands make use of the total of 45 cases drawn from Barbados,
and the 51 cases fron St. Lucia, St. Vincent, and Dominica (collectively
referred to as "small islands").
: Because of possible differences associated with such factors as
camp size and the company that the individual works for, the sample was selected in such a way that it is representative of the distribution of
* the population by size of living quarters and by employer. Using this
procedure, 17 percent of the sample was drawn from small camps (less
than 200 workers); 49 percent from the medium-sized camps (210 599);
and 341 percent from large housing units (600 plus).
season, when the workers were preparing to leave the country, there were instances in which the individuals selected from the roster could not be located. Under these circumstances, respondents were chosen on an availability basis, attempting, whenever possible, to make substitutions randomly. The nature of the bias entailed in this process, if one exists, is not known.
In addition to the logistical problems of finding specific
individuals, interviews carried out at the end of the harvest raises other potential problems. By mid-March, when the survey was conducted, cane cutters had been in the field three to five months. The work is extremely arduous and often dangerous. Injuries are common, especially among the less experienced workers. These conditions suggest that, by interviewing late in the season, individuals who did not last out the full harvest are not included. Cane cutters who are uncooperative, or unproductive, or who are subject to severe illness or injury are sent home. The possibility therefore exists that we have oversampled individuals that may tend to be hardier, more productive and more highly skilled. When generalizing from the sample to the population of H-2 workers, the effect of overrepresenting the "survivors" may be to overestimate such things as net and gross earnings, or any other variables systematically associated with length of stay.
The magnitude of this bias is proportional to how many individuals are sent back early. Industry records suggest that, relative to the size of the total labor force, the number is small. On the basis of these considerations we conclude that there is the possibility of having oversampled more proficient workers, but that the tendency-is minimal and unlikely to significantly affect the results.
The problems associated with the fact that interviews were carried out at the end of the harvest are offset by a number of advantages. Much of the questionnaire, for example, is devoted to issues that concern earnings, remittances (amount and frequency) and to plans the individual may have to invest his earnings in his home country. Each of these variables (and others that could be mentioned) are best captured once the worker has completed the season and is about to return home. Information as to remittances, for example, is retrospective in nature, and can be obtained only after the respondent has been in the United States for several months. Moreover, goods to be taken back to the islands are typically bought, not at the beginning, but rather at the end of the harvest. Thus, questionnaire items that deal with purchases have greater reliability when the survey is carried out at this point in the work cycle. Similarly, individuals, on the eve of their departure from the United States, are especially cognizant of their future investment plans back home, thus enhancing the quality of the information on these topics.
I, P-Al. R.It th
Haitians did last through the entire season. This was not enough, however, to fill industry needs, and so permission was granted to import the BWI labor force.
6 Resentment of Jamaican domination was made explicit in ntnerous
interviews conducted with workers and government officials in Barbados, St. Vincent, and St. Lucia in July, 1981 and 1982. Specific criticisms were directed at Jamaican control of policy-making within the BWICLO, the absence of liaison officers from the three smallest islands in the Florida sugar camps, and extra transportation costs charged to workers from these islands.
7 According to the Permanent Secretary for Labour (Interview, July 21, 1982), Barbados is less committed to maximizing the percentage of pool workers. This position stems from the relatively high rate of job
abandonment among Barbadian pool workers, a development that led the employers to cut (as they are free to do) the Barbadian quota for this category of worker following the 1980-81 season. It was restored the following season following a decline in the AWOL rate among Barbadian pool workers. High rate of job desertion can be costly to the island governments since they must post bond on all pool workers. Request workers are bonded by their employers. Bond is forfeited if the worker fails to return home.
8 Among the applicants one U.S. recruiter automatically rejected were: tradesmen (because of the high cost of workmen's compensation incurred in the event of injury cutting cane), masons (too "muscle-bound"'), and fishermen (chronic backaches) (Interview with FFVA Recruiters, March 20, 1981). Critics charge that the recruiters select the "most docile, dependent, and dependable" candidates (Petrcw, 1980:18).
9 Among the procedures applied in the interview is what the workers refer to as the "hand test." The recruiter examines the applicants' hands, presumably to determine by the presence of calluses whether he is accustomed to hard physical labor.
10 The Meranda Corporation of Miami acted as the growers agent in these matters for many years. Beginning in 1982-83 this task will be assumed by the FFVA. (Dailv Gleaner, April 20, 1982).
11 The Barbadian Ministry of Labour Employment Exchange maintains a registry of individuals seeking work. If the job-seeker wishes to be
considered for either the H-2 program in the U.S., or a similar program in Canada, he so indicates on the Exchange formr. These forms are supposed to be renewed every month in order for the applicants' request to remain active. Once industry employers determine the Barbadian quota and notify the Ministry of hcow many they wish to interview, the latter
requested worker quota from Barbados was 303.
12 Data from the Ministry, of Labour in St. Lucia indicated that
stateside employment in the sugar industry generated $450,000 in foreign
exchange for this country during 1979-80.
13 According to the St. Lucian Labour Commissioner, the criteria
used to select workers fpr the pool are: 1) age (23-40 years old), 2) experience cutting cane, 3) lack of current employment, '4) dependents, and 5) physical fitness. Slots in the pool are allocated to districts
(which apparently coincide with parliamentary constituencies) by
population, rate of unemployment, and performance (Interview, July 29, 1981). A Member of Parliament from St. Vincent, where MPs monopolize nominations for the pool, claimed that he chooses men on the basis of
-~ "first come, first served," human considerations, and in marginal cases
on basis of political support (Interview, July 29, 1981). In 1982, this
procedure generated a pool of 600 applicants instead of the 300
requested by industry recruiters for St. Vincent; nevertheless, they
agreed to select from the entire 600, narrowing it down to 374 workers
in just several hours.
14 In the, case of Barbadian workers, the additional amount of $192
in 1980-81 was advanced to them by the BWICLO (going) and their
government (returning). A similar arrangement existed for the St.
Lucians. However, the Dominicans and Vincentians must pay even more
($148 plus $192 for the latter) since they must be transported to either
St. Lucia or Barbados for the flight to Florida.
15 One of the major criticisms made of the task formula, both by
the workers and labour officials on the islands, is that it is difficult
to comprehend. This breeds suspicions among the workers who sometimes
claim they are being cheated.
16 Management is entitled to verify a worker's productivity. If he
fails to achieve the required ninmun for three days, he can be sent
home for breach of contract. Breaching the contract, whether in this
manner, or by some other offense results in blacklisting the individual.
Recruiters come to the islands with a "blackbook" listing all
blacklisted workers. If an individual is listed in the book, he is
automatically excluded, according to BWI labor officials familiar with
recruiting practices. For their part, workers may try to circumvent
-~ these checks by false identities.
17 For some reason the two percent BWICLO deduction is sent to the
~islands converted to local currency then reconverted to U.S. dollars and
returned to the liaison office in Washington. The islands claim to lose
money on this double conversion.
4 2 t -..at~~.4 T..A n *.~nr~n .-n
$6-10 million of a $20 million payroll for a multiplier impact on local
sales of $12-20 million. They, thus conclude that "...local spending by
offshore workers generates between 3.2 and 5.4 million dollars in income
for local residents."
19 In addition to the predominant practice of awarding an across
the board bonus of six percent of gross seasonal wages, other methods
included: (1) a sliding scale that awards a higher percentage bonus to
the more productive (and higher paid) workers (Atlantic Company); (20 a
flat bonus of $10 per week for each week with no absences (Cane
Contractors); and (3) a fixed dollar bonus for each worker (Sugar Cane
Harvesting). (Information provided by FFVA official, June 22, 1982).
20 Although they do not have precise figures for local spending of
the offshore workers, Mulkey and Gordon's (1979:25) estimate of $6-10
million (if we subtract expenditures for personal consumption) tends to
confirm our figure of $4.5 million spent in goods remitted to the
21 As noted in the text, the measure of capital accumulation used
here should be regarded only as approximate indicators. From the
standpoint of a human capital perspective the conclusions presented
should be qualified. From the analysis of the remittances, spending
patterns and investment strategies it is apparent income earned in the
United States is devoted to educating workers' children. By upgrading
the skills of the second generation, it can be argued that this
"investment" contributes, in the long run, to raising the
income-generating capacity of the household. A test of this hypothesis
turns on whether the offspring of H-2 workers actually achieve greater
educational attainment than the sons and daughters of the rural
population at large. Our date set does not contain sufficient
information for such an analysis.
1978 The Small Farmer in Jamaican Agriculture: An Assessment of Constraints and Opportunities. Report of the Agricultural Sector Assessment Team of the Office of International Cooperation and Development to USAID, Jamaica. Kingston: USAID.
1978 The West Indies (BWI) Temporary Alien Labor Program:
1943-1977. A study prepared for the Subcommittee on Immigration of the Committee on the Judiciary U.S. Senate, 95th Congress, 2nd Session.
U.S. Department of Labor Employment Service. Division of Labor Certifications
1982 Labor Certification Granted for Temporary Foreign Workers (H-2s) in Agriculture and Logging Operations. Photo Copy (January 29).
n.d. Statement by Dalton Yancy, Vice President and General Manager, Florida Sugar Cane League.
1979 The Florida Sugar Industry: Its Past, Present Prospects. Unpublished paper.
Zepp, G.A. and Joe E. Clayton
1975 A Comparison of Costs and Returns for Hand Cutting and
Mechanically Harvesting Sugar Cane in Florida, 1972-73 Season. Economic Report to Food and Resource Economics Department, University of Florida, Gainesville (June).