CHARACTERISTICS OF MALE AND FEMALE HEADED HOUSEHOLDS
IN TWO CARIBBEAN ISLANDS: DOMINICA AND ST. LUCIA
Klaus de Albuquerque, Professor
Department of Sociology and Anthropology
College of Charleston, SC 29424
and
Jerome L. McElroy, Professor
Department of Business Administration and Economics
Saint Mary's College
Notre Dame, IN 46556-5001
(June, 1999)
CHARACTERISTICS OF MALE AND FEMALE HEADED HOUSEHOLDS
IN TWO CARIBBEAN ISLANDS: DOMINICA AND ST. LUCIA
There is a voluminous literature on the "family" and "household" in the Caribbean (see M.G. Smith, 1973; Barrow, 1996) dating back to the Herskovits' ethnographic research in the 1930s and 1940s (Herskovits, 1937; Herskovits and Herskovits, 1947). Influenced by structural-functionalist Anthropology (see Barrow, 1998), the dominant view at this time was that the nuclear family was the most elemental and functional form of social organization with the defining characteristic of co-residence (see Solien, 1960). It was assumed that a household should contain at least one nuclear family--a conjugal pair plus one child. However, in the Caribbean, social scientists discovered a multiplicity of "family forms," including multi-generational households with no nuclear families (see Henriques, 1953; Clarke, 1957; R.T. Smith, 1956; M.G. Smith, 1962). Faced with having to redefine family, these researchers chose to focus instead on the "household" or domestic group, defined by R.T. Smith (1956) as a "group of people occupying a single dwelling and sharing a common food supply."
Ethnographic studies and social surveys in the 1950s and 1960s revealed that many of these households were consanguineal--centered around the mother-child relationship and including maternal kin (grandmother, aunts, female cousins, nieces and nephews) (R.T. Smith, 1956; Clarke, 1957; Greenfield, 1966). Men, as husbands and fathers, were absent. These female headed households were labeled "matrifocal" or "matricentric" (Gonzales, 1970; R.T. Smith, 1973). Male marginality was linked to matrifocality (Rodman, 1971; R.T. Smith, 1973).
The ubiquity of matrifocality among the black lower class in the Caribbean has been explained as a legacy of the relative impermanence of males on slave plantations (Herskovits, 1941), high rates of male unemployment and underemployment, coupled
with poor community integration, characteristics, common in plantation societies (Beckford, 1972), labor migration resulting in long periods of male absenteeism (M.G. Smith, 1962; Kunstadter, 1963; Otterbein, 1965; Nag, 1971), and so on. Male absence/marginality was assumed to be dysfunctional to Caribbean
households (Simey, 1946; Matthews, 1953).
More recent research points to the benefits of matrifocal households. First, it suggests that they are economically more rational, as unemployed or underemployed husbands/fathers are a drain on family resources (Morrissey, 1998).(1) Second, male absence elevates the status of women in the home and the community (Rubenstein, 1987). Third, as both mother and father, a woman exercises more control over her children and has a greater impact on their socialization, which, given the legendary moral authority of West Indian matriarchs, has positive benefits for society (Senior, 1991). Fourth, social networks among female headed households tend to reassure and embolden women that they can go it alone, particularly, in a changing Caribbean with greater opportunities for female employment (tourism, assembly-type manufacturing, infomatics). Even if men are present and the putative heads of households, it is often women who bring in the steady income, feed and clothe the children, and pay their school fees. Thus, female headship is often underestimated in the Caribbean.
Research on the socio-economic characteristics of female headed households in the Caribbean point to a number of consistent findings (see Massiah, 1983; Barrow, 1996). First, female headed households tend to be larger than male headed households. Second, male headship rates are lower than almost any regional area of the world, given the well known fact that younger West Indian males have a general reluctance to engage in co-residential unions (Blake, 1961), while female headship rates are among the highest in the world. Third, male headed households have higher incomes than female headed households. It follows that the latter are much more likely to be impoverished (Massiah, 1983). Fourth, female heads are much more likely to derive income from informal activities, or if employed, tend to be congregated in low paying jobs in assembly-type manufacturing and infomatics (Pearson, 1993). Fifth, female heads are less well educated than their male counterparts. Sixth, the composition of female headed households is different from male headed households, with smaller nuclear components, and larger extended and non-related components.
Since most of these findings are based on field work in small rural communities or urban neighborhoods, they tend to suffer from cross level interpretive problems. In other words, what you see in the small, and generally lower class community, often does not apply to the nation as a whole. Furthermore, rapidly changing economic circumstances in the region have led to greater gender equality. This is reflected in: (1) a greater proportion of women than men employed in professional, technical, administrative, and managerial fields; (2) unemployment rates for women that are on par with that of men, and even lower in some cases (Dominica, St. Lucia, St. Vincent); (3) rapid gains in the incomes of women relative to men (parity in Dominica in 1991); and (4) educational performance and attainment superior to that of men (de Albuquerque and Ruark, 1998). Unfortunately, Caribbean feminist writings have not caught on to the remarkable changes that have occurred among Caribbean women in the 1990s (de Albuquerque and Ruark, 1998).
The Setting
Dominica and St. Lucia are mountainous volcanic islands that form part of the Windward island chain in the Lesser Antilles. Although larger in area (750 km2 vs. 620 km2), Dominica is considerably smaller in population (67,000 vs. 151,000). Both were originally settled by pre-Columbian Indians from South America who were virtually decimated by European colonization. The islands changed hands between the French and British several times until the Treaty of Versailles (1787) restored British control over Dominica, and the Treaty of Paris (1814) ceded control of St. Lucia to the British.
From slavery and the plantation system up to the present, agriculture has been the islands' economic mainstay (de Albuquer- que and McElroy, 1999). With the decline of sugar after emancipation, both countries became chronic labor exporters. Since World War Two, agriculture's importance has waned. Although banana exports created some measure of prosperity, the creation of the European Union in the 1990s has placed preferred access to United Kingdom markets in jeopardy. Economic diversi-
fication towards tourism has been more successful than efforts to establish light manufacturing exports (garments, assembly, info-
matics). Presently, St. Lucia is poised to become an internationally recognized mass tourism destination while Dominica is hinging its development on low density ecotourism.
The two islands are close socio-culturally because they share similar African/French-creole based cultural traditions and speak a mutually intelligible "Kweyol" (Creole). English is the official language. Both countries gained political independence from Great Britain in the late 1970s.
All data were derived from special tabulations of the 1991 censuses for Dominica and St. Lucia. For the purposes of this paper a "household" is defined as a "number of individuals living together in one dwelling, or part of a dwelling, who share cooking facilities, make common provisions for food, and eat some, if not all, meals together" (United Nations, 1973). Household components include family, nuclear (head, spouse, children), extended (relatives of head), and non-relatives such as boarders, wards, and foster children.
The majority of households in Dominica and St.Lucia are headed by males (Table 1). The number of male headed households in St. Lucia has remained constant between 1970 and 1991, with about 60 percent of the households headed by males and 40 percent by females. These figures obscure what was mentioned earlier, to wit, some men report themselves as heads of households when they are marginally involved, socially and economically, with the workings of the household. The high rate of female headed households tends to be a unique, and much remarked upon feature of Caribbean households. In the United States, in 1990, 11.7 percent of the households were headed by females (Weeks, 1994).
The average size of female headed households in St.Lucia is slightly larger than male headed households. In Dominica, there appears to be no difference in average household size, with the exception of 1970, when male headed households were larger. This calls into question the continuing perception by scholars, especially feminist scholars, that female headed households continue to be burdened by large numbers of persons, related and unrelated.
Male headed households, as expected, have higher median incomes than female headed households, but the income gender gap is narrowing (Table 1). In 1970, female headed households in Dominica had 65 percent of the income of male headed households. By 1991, this figure had risen to 89 percent. However, in St.Lucia, female headed household earnings only rose from 61 percent of the median income of male headed households in 1970 to 66 percent in 1991. Obviously, women have made greater gains on the income front in Dominica (see de Albuquerque and Ruark, 1998). The higher median incomes of male headed households can be partly accounted for by more full time employment, often involving two income earners (male head and spouse).
Headship rates by age and sex, for St. Lucia, are presented in Table 2. Compared to headship rates in other regional areas of the world (Latin America, Asia, Europe, North America, and the Pacific), in the Caribbean male headship tends to be only slightly lower, while female headship rates are three to four times higher (de Albuquerque and Tauge, 1995; United Nations, 1973). The former calls into question the 1960s research about the reluctance of Caribbean males to engage in co-residential unions. Younger males (15-34) in St. Lucia show headship rates similar to those in Latin America, Asia, Europe and North America. The remarkably high age-specific headship rates for women in St. Lucia underscores the degree to which many households in the region are organized around females.
Table 3 shows the economic activity of heads of households by sex. As expected, male heads are much more likely to be in the labor force than female heads--almost 2:1 in Dominica.
However, some female economic activity is obscured when women work in the informal sector and are counted as being economically inactive. Not surprisingly, labor force participation rates are much higher for male heads (80.3 percent in Dominica and 85 percent in St. Lucia) than for female heads (41.3 percent in Dominica and 51.4 percent in St. Lucia). Unemployment rates are quite low for both male and female heads, with male heads having slightly higher unemployment rates (Table 3). Again, this contradicts the widespread perception among Caribbean scholars of high unemployment rates among female heads. Admittedly, some of this unemployment is hidden--many women claim to be involved in home duties but if a job became available they would take it (de Albuquerque and Ruark, 1998).
In Dominica, 59 percent of the female heads of households declared themselves to be economically inactive, with 41 percent reportedly involved in home duties. While this explains the lower median income of female headed households, and would suggest that many of these households are characterized by poverty, in rural areas this is blunted since food is supplemented from home gardens and there is considerable inter-household transfers of food. In addition, income enters the household through informal economic activity (selling cooked foods or surplus produce at markets) and from periodic infusions of money from absent working spouses/partners or sons and daughters. A significant amount of income enters poorer households through remittances from spouses, sons, and daughters working abroad (Rubenstein, 1987).
Table 4 compares the occupations of female heads versus male heads. Interestingly, in both Dominica and St. Lucia, more female heads are involved in administrative, managerial, professional, and technical occupations, than male heads. De Albuquerque and Ruark (1998) also found more females than men employed in these prestigious occupations in Dominica, St. Lucia, and St. Vincent (1991), but noted from disaggregated data that females were employed at lower levels, i.e. nurses rather than doctors, Assistant to the Permanent Secretary of a Government Ministry, rather than Permanent Secretary.
The gender division of labor is clearly observable with the remaining occupations, except for unskilled labor. More female heads are employed in clerical and sales and service occupations, while male heads are involved in manufacturing and skilled craftsmanship. An unexpected finding is that there are almost twice as many female heads, than male heads, involved in unskilled labor ("Labourers and Others").
With respect to education, proportionately more female heads have completed primary and secondary school than male heads.(2)
However, when it comes to University, more than twice as many male heads have university degrees (Table 5). Given the fact that current enrollment at The University of the West Indies is, and has been for quite some time, dominated by females (de Albuquerque and Ruark, 1998), the next census is bound to show proportionately significant gains in female university graduates.
Table 6 examines differences in marital status between female and male headed households in St.Lucia. The 1991 census did not contain such data for Dominica. In all age groups, male heads of households are much more likely to be married than female heads--50 percent in the 35-44 age group versus 19 percent. Very low marriage rates for female heads are to be expected as their households are defined by the relative absence of men as spouses. Marriage rates for male heads are comparatively low when compared to other countries, and extremely low for female heads. The latter obscures the fact that many of these female heads are engaged in visiting (younger age groups) and common law unions. Female headedness does not necessarily imply that men as partners are absent. In the 45-64 age group, 26 percent of the female heads had a spouse or partner, and one would suspect that the majority of these were legal spouses (Henriques, 1953, Blake, 1961, and others note that West Indian men are more likely to marry as they get older).
The data clearly show that marriage rates increase with increasing age, a well documented fact in the region (Blake, 1961; Stycos and Back, 1964). Conversely, the percent of "never married" decreases with each successive age group. As expected, more female heads report being never married in all age groups, but as has been noted previously, this does not mean that men are absent from their lives. In fact, Caribbean women, like their men, have multiple partners during their lives. Stycos and Back (1964) note that five or more unions are not uncommon.
As expected, the percent of female headship that results from the death of a spouse increases with age--48 percent of the female heads aged 65 and above were widows, due obviously to differential (higher) male mortality.
Table 7 examines household components in St. Lucia by sex of household head. Such special census tabulations were not available for Dominica. As expected the "family" size component of female headed household is larger, primarily because there are more children and extended kin in these households. The nuclear components are roughly similar, with male headed households having significantly more spouses and fewer children. Male households also have fewer extended kin and non-relatives. The differences, however, are not as great as expected, especially when one examines percentages (82.3 percent of household size in male headed households is accounted for by the "nuclear" component versus 77.2 percent in female headed households). Even the much repeated observation that female headed households tend to be organized around maternal kin, and have a very much larger extended component, is not borne out in the data in Table 7 ("extended" components make up 20 percent of female headed households versus 15.6 percent of male headed households).
Our sense is that by 1991, the differences in household composition noted by many Caribbean scholars (see Blake, 1961; M.G. Smith, 1962) were not as prominent. One major difference
still remains, however, and that is the relative dearth of spouses or partners in female headed households.
The 1991 census data for Dominica and St. Lucia provides a macro look at the differences between male and female headed households. Far too much that has been written about the family and household in the Caribbean has focused on lower class households and small primarily rural communities. In addition, findings from these studies have sometimes been used to characterize all families and households. Absent are studies of upper and middle class families or households, or of urban underclass households in large metropolitan areas in the region.
There have also been many changes that have occurred to families and households since the heyday (1950s and 1960s) of research studies. Some of these changes have been noted, for example, men assuming greater responsibility as husbands and fathers (Barrow, 1998), but others uncovered in this study question the conventional wisdom.
First, female headed households in the 1990s are only slightly larger, or no different in size, from male headed households. Having data from two islands gives us greater confidence in the generalizability of our findings to other
English-speaking islands in the Eastern Caribbean with overwhelmingly black populations.
Second, the income gender gap is narrowing between male headed and female headed households, especially in Dominica. Nevertheless, female headed households continue to be more likely
to be poor.
Third, male age-specific headship rates, contrary to all expectations, tend to be only slightly lower than corresponding rates in other regions of the world. Female age-specific headship rates continue to be among the highest in the world, and this does define the Caribbean household as unique.
Fourth, as expected, male heads are economically more active than female heads, many of whom report being involved in "home duties." However, as we have noted women are much more likely to participate in the informal sector, and this may mask the economic activity of female heads.
Fifth, proportionately more female heads are employed in
administrative, managerial, professional, and technical operations than their male counterparts. This is an unexpected finding but it parallels the increasing occupational status of economically active women in the region (de Albuquerque and Ruark, 1998). Apart from this finding, female heads continue to be employed in sales and services and male heads in manufacture and craftsmanship.
Sixth, on the education front female heads tend to be better educated than male heads at the primary and secondary levels. Casual observation from around the region suggests they are closing the gap at the university level.
Seventh, marriage rates (St.Lucia) for men were much higher in 1991 (50 percent of the 35-44 age group and 65 percent of the 45-64 age group were married) than in previous studies, but still lower than in most regional areas of the world. Marriage rates for female heads, as predicted, were low, reaching a high of 26.4 percent in the 45-64 age group. In the 35-44 age group, 72 percent of female heads reported never being married.
Eighth, the data from St. Lucia also indicate changes in household composition with both male and female headed households being dominated by the nuclear component. Two continuing noticeable features of the nuclear component in female headed households is the dearth of spouses and the significantly greater number of children. The extended component of female headed households appears to have diminished significantly from earlier ethnographic studies that noted that these households were heavily peopled my maternal kin--grandmothers, aunts, female cousins, nieces, and nephews. The extended and non-related component in female headed households, however, still remains larger than in male headed households.
1. Morrissey forgets that unemployed and underemployed sons and
daughters are also a drain on household resources.
2. Caribbean females have been outperforming their male
counterparts in primary and secondary school for almost
two decades now. Universities in the region also show more
female registrants than male registrants (de Albuquerque and
Ruark, 1998).
REFERENCES
Barrow, Christine
1996 Family in the Caribbean: Theories and Perspectives. Kingston, Jamaica: Ian Randle Publishers.
1998 "Caribbean masculinity and family," pp. 339-358 in Christine Barrow (ed.), Caribbean Portraits: Essays on Gender, Ideologies and Identities. Kingston, Jamaica: Ian Randle Publishers.
Beckford, George L.
1972 Persistent Poverty. New York: Oxford University Press.
Blake, Judith
1961 Family Structure in Jamaica. Glencoe: The Free Press.
Clarke, Edith
1957 My Mother Who Fathered Me. London: George Allen and Unwin.
Commonwealth of Dominica
1993 Population and Housing Census 1991. Roseau: Central
Statistical Office.
1995 Special Tabulations of the 1991 Census. Roseau:
Central Statistical Office.
de Albuquerque, Klaus and Mariget Tauge
1995 Social Implications of Population Growth in Papua, New Guinea, 1980-1995. Weigani, Papua New Guinea, Institute of Applied Social and Economic Research.
de Albuquerque, Klaus and Sam Ruark
1998 "Men day done: are women really ascendant in the Caribbean," pp. 1-13 in Christine Barrow (ed.),
Caribbean Portraits: Essays on Gender, Ideologies and Identities. Kingston, Jamaica: Ian Randle Publishers.
de Albuquerque, Klaus and Jerome McElroy
1999 "The origins and socio-economic characteristics of the foreign born population of three Caribbean Societies: Dominica, St. Lucia and St. Vincent." International Migration Review (submitted).
Gonzalez, Nancie L.
1970 "Towards a definition of matrifocality," pp. 231-244 in N.O. Whitten Jr. and J.F. Szwed (eds.), Afro-
American Anthropology: Contemporary Perspectives. New
York: The Free Press.
Greenfield, Sydney
1966 English Rustics in Black Skin. New Haven: Yale University Press.
Henriques, Fernando
1953 Family and Colour in Jamaica. London: Eyre and Spottiswoode.
Herskovits, Mellville J.
1937 Life in a Haitian Valley. New York: Knopf.
1941 The Myth of the Negro Past. New York: Harper Brothers.
Herskovits, Mellville J. and F.S. Herskovits
1947 Trinidad Village. New York: Knopf.
Kunstadter, Peter
1963 "A survey of the consanguine or matri-focal family." American Anthropology 65:56-65.
Matthews, Dom Basil
1953 "Crisis in the West Indian family." Caribbean Affairs Vol. 9, Trinidad: University College of the West Indies.
Massiah, Jocelyn
1983 Women as Heads of Household in the Caribbean: Family Structure and Feminine Stratus. Paris: UNESCO.
Morrisey, Marietta
1998 "Explaining the Caribbean family: gender ideologies
and gender relations," pp. 78-90 in Christine Barrow
(ed.), Caribbean Portraits: Essays on Gender, Ideolo-
gies and Identities. Kingston, Jamaica: Ian Randle Publishers.
Nag, Moni
1971 "Pattern of mating behavior, emigration and contraceptives as factors affecting human fertility in Barbados." Social and Economic Studies 20:111-133.
Otterbein, Keith
1965 "Caribbean family organization: a comparative analysis." American Anthropologist 67:66-79.
Pearson, Ruth
1993 "Gender and new technologies in the Caribbean," pp. 287-295 in Janet Momsen (ed.), Women and Change in the Caribbean. Kingston, Jamaica: Ian Randle Publishers.
Rodman, Hyman
1971 Lower Class Families. New York: Oxford University Press.
Rubenstein, Hymie
1987 Coping with Poverty: Adaptive Strategies in a Caribbean Village. Boulder, CO: Westview Press.
Senior, Olive
1991 Working Miracles: Women's Lives in the English-Speaking Caribbean. Bridgetown, Barbados: UWI, Institute of Social and Economic Research.
Simey, T.S.
1946 Welfare and Planning in the West Indies. London: Oxford University Press.
Smith, Michael G.
1962 West Indian Family Structure. Seattle: University of Washington Press.
1973 "A survey of West Indian family studies," pp. 365-408 in Lambros Comitas and David Lowenthal (eds.), Work and Family Life: West Indian Perspectives. Garden City, New York: Anchor Books.
Smith, Raymond T.
1956 The Negro Family in British Guiana. London: Rutledge and Kegan Paul.
1973 "The matrifocal family," pp. 121-144 in J. Goody (ed.), The Character of Kinship. Cambridge: Cambridge University Press.
Solien, Nancie L.
1960 "Household and family in the Caribbean." Social and Economic Studies 9:101-106.
Stycos, J. Mayone and Kurt W. Back
1964 The Control of Human Fertility in Jamaica. Ithaca, NY: Cornell University Press.
St. Lucia
1993 Population and Housing Census 1991. Castries, Saint Lucia: The Statistics Department.
1995 Special Tabulations of the 1991 Census. Castries, Saint Lucia: The Statistics Department.
United Nations
1973 Methods of Projecting Households and Families: Manual
VII. New York: United Nations.
Weeks, John R.
1994 Population: An Introduction to Concepts and Issues.
Belmont, CA: Wadsworth.