Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
CONSTANTINA SAFILIOS-ROTHSCHILD
Department of Women's Studies, Agricultural University, De
Leeuwenborch, Wageningen, The Netherlands
The determinants of the ability of household members to adapt to social and economic changes can be best examined within the theoretical framework of a social and sexstratification model, supplemented by the conflict theory of family dynamics (Collins, 1975). While the social stratification theory is quite well known, it is important to spell out the nature of the sex-stratification theoretical model. The sex-stratification system has been defined as follows (Safilios-Rothschild, 1982a):
The sex-stratification system ranks the one gender, practically always the male gender, higher than the other - the female gender - and determines that men only will occupy major decision-making positions and will control the valued resources in the society. Such valued resources include a wide range of resources such as: wealth, income, credit, knowledge (in terms of literacy and education as well as information), technology, valued skills, valued income-generating activities, food, health, power, and prestige. The sexstratification system, that is, men's superior status leading to power over women and control of valued resources, is supported by laws and policies that spell out and legitimise men's privileged and dominant status and is justified by religious, traditional, moral, and/or pseudo-scientific ideologies and beliefs.
Moreover, it has been documented that sex stratification is institutionalized and pervasive and is being maintained and supported by widespread institutional mechanisms and sets of beliefs. Such mechanisms and beliefs help perpetuate the sex-stratification system even in the face of structural changes that ought to diminish, if not completely undermine, the strength of its operation. A striking example can be found in Lesotho, where about half (47.7 per cent) of the men are absent from the country working in South African mines at any time (and the percentage is much higher in mountainous rural areas). Even so. several mechanisms and beliefs aiming to diminish the impact of this absence are responsible for only a delayed and at best partial definition of the persisting pattern of male absence as a long-term manpower shortage necessitating structural dedifferentiation in many sectors. As a consequence, the process of structural dedifferentiation has taken a very long time and is by no means complete so as to allow especially rural women greater access to all types of services and resources, including political participation (SafiliosRothschild, 1985).
The combined framework of the two stratification systems helps to predict some important trends at the societal and the household level. The theoretical model predicts, for example, that in countries in which the class system tends to be closed and rigid, with few institutionalized avenues for social mobility, membership of a social class represents such an important status that it can neutralize the negative effects of the female (low) sex status (Safilios-Rothschild, 1980). A study in rural Kenya, for example, showed that elite women farm managers were visited as often by agricultural extension officers as were male farmers in elite jointly managed farms. Low-income women farm managers, on the other hand, were visited much less often than low income male farmers in jointly managed farms and less often than were elite women farm managers. The same trends, though even more accentuated, held true with regard to loans and access to agricultural training (Staudt, 1979). Thus, low-income women's access to resources is doubly curtailed by the compound effect of being a low-status member in both stratification systems.
The theoretical model can be used to make predictions for society at large and within the household. The more scarce the valued resources, the more the patterns of allocating these resources represent "rational" investment strategies that are thought to maximize the short- and long-term survival chances of the institution (i.e. society or household). The pervasiveness of a powerful sex-stratification system operating in practically all developing countries, coupled with an equally powerful social stratification system, is responsible for the fact that such rational investment strategies are usually male-biased. At the societal level, traditional patriarchal values assign to men the economic responsibility for women and children.' Men are thus given priority over women in terms of preferential access to paid employment, income-generating productive activities, and credit, in addition to better education and training opportunities. Such a pattern is clearly discerned in developing countries in which rapid rates of population growth are making scarce such key resources as paid employment, credit, training facilities, teachers, and land.
Similarly, gender (along with age) is a key distributional criterion at the household level. It can be used as a proxy for income-earning capacity, since women usually have lesser access to paid employment and income in general than do men. In many instances, women have no job or income-earning opportunities at all. Poor households, in which food and access to health care and education are scarce, allocate these resources on the basis of gender and age priorities (Safilios-Rothschild, 1980). By systematically favouring over women and girls (in descending order) adult men, young men, adolescent boys, and infant boys, all of whom have greater actual or potential access to income than do female household members, the chances for household survival are maximized. Alternatively, in households which are not constrained by scarcity of these resources, there is no need to make difficult allocation decisions: male and female members of all ages tend to have a greater probability of equal access to resources than in poorer households (SafiliosRothschild, 1979, 1980).
The sex-stratification theory predicts the important determinants of flexibility or rigidity a society, community, or household experiences in the face of important social and economic changes. The key determinant of the extent of flexibility or rigidity in response to these changes is the visibility of the changes in terms of their accurate documentation and the availability of the facts to policy-makers and programme designers and implementers. Institutional mechanisms and sets of beliefs play crucial roles in keeping "undesirable" changes invisible and, thus, in maintaining the sex stratification system intact (Safilios-Rothschild, 1985). Examples of significant changes that have remained invisible for a long time (and in a number of countries are still invisible) are the increasing incidence of female-headed households (legal and de facto), the feminization of smallholder agriculture, and the existence and importance of wives' incomes.
The sex-stratification system rests on the premise that women are economically dependent on men. The increased prevalence of female-headed households in many developing countries has been viewed as a challenge to this basic premise and a threat to the very fabric of patriarchal society. Sex-role stereotypes espoused by statisticians and interviewers, as well as male and female respondents, have led to methodologies and datasets that underestimate the prevalence of female-headed households, particularly of de facto rural female-headed households created by male migration (Safilios-Rothschild, 1982a, 1982b, 1982c; Youssef and Hetler, 1983). In Lesotho, for example, the official statistics have enumerated only the legal female-headed households (35 per cent) and not the other half (35 per cent) of de facto female-headed households (Safilios-Rothschild, 1985). Research undertaken in rural Kakamega, Kenya, showed that women whose husbands worked in Nairobi, and only visited for about one month per year, had diffculty admitting that they made all agricultural decisions by themselves. Such an admission would indicate that husbands no longer played a dominant role in the family and would shake the established sex-stratified order.
In addition, even when the appropriate data are collected, they are often not analysed and disseminated to policy-makers and programme designers and implementers. Through these mechanisms, the true prevalence of female-headed households remains underestimated and no structural changes are brought about to help these households become integrated into the development process on the same basis as male-headed households. Finally, even when the statistics concerning the prevalence of such households are collected and disseminated, the impact on policies and programmes is often negligible because of the existence of sex-stereotypic beliefs that tend to diminish the significance of the statistical trends. In Lesotho, for example, it is believed that men make all important agricultural decisions and do most of the heavy work in agriculture, despite the fact that they are present at home only for about one month in twelve, after a year's hard work in the South African mines. These beliefs are responsible for the fact that the women heads of 75 per cent of the households in most rural areas are not integrated into the mainstream of agricultural planning and programmes (Safilios-Rothschild, 1985).
In many areas of sub-Saharan Africa, male migration has led to the feminization of smallholder agriculture. The lack of societal adaptation to these changes, in terms of reorientation and redirection of agricultural services and resources towards women farmers, and the failure to provide women farmers with incentives, however, carries the heavy price of low agricultural productivity, lack of interest in adopting long-term land improvements and agricultural investments, and lack of food self-sufficiency for the household.
Women's income can also play a crucial role in the dynamics of intra-household allocation of resources. The visibility of this key variable is an especially important factor. In the context of low-income households (particularly of rural households), women's income, although usually very low, often represents as much as 40 to 50 per cent of the total household income. In-depth studies in various developing countries (e.g. Honduras) have revealed that, despite the crucial role that women's income often plays for family survival, in the traditional context of low-income households the size and importance of this income is usually not acknowledged by husbands. This is because such an acknowledgement would constitute an admission of the husband's inadequacy as a breadwinner, and would tend to undermine his unquestioned dominance (Safilios-Rothschild, 1988).
The maintenance of invisibility in this case is facilitated by the characteristics of women's income, particularly of rural women's income, and a number of mechanisms women (and men) use to minimize the significance of women's economic contributions. Their income is usually earned in very small sums from a variety of sources (such as selling eggs and chicken, foodstuffs, straw mats, or small quantities of crops). Moreover, it is unpredictable in that there is no regular amount and no set time when it is earned, as there would be with a regular pay cheque. Finally, since women in low-income households often use these small sums almost immediately to buy food or other needed household items, they do not always have the chance to develop the concept of an income that they earn and control. Interviewed rural women sometimes report that they do not earn an income, or that they earn less than is the case. This perception reduces the threat that a wife's earnings may represent to a husband with marginal income-earning capacity (Safilios-Rothschild, 1983, 1988). The combined class and sex theoretical model predicts that increases in women's income will tend to remain invisible if they are perceived as threatening to the image of men as breadwinners. Indeed, in-depth studies in rural Honduras have shown that when husbands do not earn a sufficient income to support their families properly, and when their wives earn almost as much as they do, the threat is diminished through a variety of mechanisms which minimize the importance of their wives' incomes. Although it is difficult to achieve the total invisibility of women's earnings, men grossly underestimate this income and thus are able more comfortably to retain power and authority. This distorted perception is reinforced by the characteristics of the income and by the women themselves, who understate what they earn partly in order not to threaten their husbands (Safilios-Rothschild, 1988).
Under these circumstances, women's income can easily remain invisible. Its characteristics in low-income rural households (e.g. irregularity, unpredictability, and small amounts from different sources) explain not only the fact that increases in women's earnings often pass totally unnoticed, but also such increases do not in any way affect the allocation of women's labour, time (especially for leisure), or decision-making power within the household. It is only when women earn a visible, stable, and substantial income from agricultural and market activities or from wages, and when husbands earn a secure income, larger than their wives', that the distribution of labour, time, and decision-making power is shifted according to the importance of the wives' economic contributions (Safilios-Rothschild and Dijkers, 1978; Safilios-Rothschild, 1988). What is unknown within this type of household context and needs to be researched is the extent to which invisible increases in women's income lead to positive changes in the intra-household allocation of food and/or health care, even when there is no reallocation of other inputs such as power or time and tasks.
When, on the other hand women's income becomes visible as a result of the shared earnings from a women's co-operative or group, or from participation in the formal employment sector, and the sums of money earned are both more regular and more substantial, the usual mechanisms for minimizing women's monetary contribution and maintaining its invisibility are no longer effective. In areas in which social structural conditions make it very difficult for men to be adequate breadwinners, women's access to income generated through such visible means is even more of a threat to men's status and identity. Powerful mechanisms are called into play to maintain the sex-stratification bias. In some cases, men have taken over the control of development projects that exclusively benefit women as soon as they show some potential for generating a substantial income (Bekele, 1982; Noble and Nolan, 1983; Sundar, 1981). In Mombasa, Kenya, for instance, it was found that in areas in which men had poorly paid, marginal occupations, the incomes women earned through group projects were controlled by the men, and the women were not able to acquire prestige or decision-making power as a function of such income (McCormack et al., 1986).
Within the context of a powerful and pervasive sex-stratification system, men cannot adapt to social and economic changes that significantly increase women's access to income, the key resource. Instead, under these circumstances, they usually resort to mechanisms that neutralize the impact of ongoing changes, and thus preserve the sex-stratification status quo. The degree of potential rigidity of adult male household members in response to women's increased and visible access to income becomes even more accentuated when all these mechanisms fail to bring about the desired outcome. Men have been found to resort then to force and violence: they may usurp the women's earnings, even if they have to beat them up, or they may become more domineering, authoritarian, repressive, and violent at home than before in order to ensure that the integrity of the existing patriarchal power structure is not in any way challenged by women's changing status (de Graft-Johnson, 1984).
This type of rigidity explains why women-focused income-generating projects in lowincome communities, where men have few opportunities to increase their own incomes, are usually doomed to failure; it also accounts for the fact that, even when women are able to substantially increase their own incomes, this change sometimes has little, if any, impact on the intra-household allocation of resources or power. It must be noted, however, that when men are able adequately to support their families, there is also considerable evidence of substantial male adaptability to women's increased income-earning. The men's willingness to change the intra-household allocation of resources under these circumstances is high. The more secure men are in their breadwinner role, the more adaptable they are to women's income-earning ability, probably because women's income does not represent a threat to their superior position as males (Safilios-Rothschild and Dijkers, 1978; Safilios-Rothschild, 1985). In Mombasa, Kenya, it was found that only in one community, in which husbands had a stable and sufficient economic base, were women able to translate their economic contributions into a valuable resource that earned them more decision-making power and more equality in the division of labour (McCormack et al., 1986).
The degree of rigidity or adaptability on the part of household members to social and economic changes may, in fact, differ according to the type of resource allocated and to the generation of the recipient. It is therefore possible that under the conditions specified above, wives, increased access to income and/or paid employment would promote rigidity with respect to changes in intra-household power allocation, but considerable flexibility and adaptation in terms of time allocation. Or, more importantly, these changes in wives' income-earning potential might prompt intra-household redistribution of resources such as food, health care, and education and training in favour of young girls, even if they do not favour adult women, or specifically wives. This pattern might be explained by the logic that women's increased access to income in creases the rationality of investments in daughters, in addition to sons, for maximizing household welfare (cf. Rosenzweig, this volume). Such an intergenerational reallocation in investment priorities does not pose as big a potential threat adult males as do changes that may affect their power relationships with adult women.
The class- and sex-stratification theoretical model must be supplemented by a conflict theory of family dynamics (Collins, 1975) in order to derive predictions and explanations of what determines the degree of rigidity or adaptability household members experience when faced with social and economic changes. An important premise of such a conflict theory is that families and households cannot be viewed as monolithic institutions in which the members all agree on the strategies and means to be used to maximize family or household welfare. Instead, individual household members have different interests and needs which are often incongruent and conflict. Household welfare is thus the balance achieved through a series of compromises and accommodations among individual household members according to their interests and needs, as determined by their bargaining power. This is a function of members' persona! characteristics (cf. Engle, this volume), their access to economic resources, the availability of alternatives, prevailing cultural norms, influential institutions, and macro-policies (Folbre, 1984; Jones, 1983; Fapohunda, 1987; Safilios-Rothschild. 1987). This reality of household dynamics is central to the understanding of rigidity and adaptability patterns.
The addition of a second wife to a household illustrates how conflicting interests between household members can lead to rigidity in some cases and adaptability in others, depending on the perceived costs and benefits to individual household members. Evidence from several sub-Saharan African countries in which polygamy is still widespread shows that first wives tend to accept the second or third wife because they perceive several advantages and not only disadvantages. This adaptability is predicated on the fact that first wives (particularly in rural areas) have a very heavy workload, and they welcome labour-sharing with the second or third wife (McSweeney, 1979). This is especially true when wives' land plots are large and productive enough to feed their own children and to yield a potential surplus. Thus, first wives adapt to a structural change that decreases their access to their husbands' love, sexual favours, labour, and income contribution because they gain in terms of time for their own sleep, labour, and a certain degree of autonomy in terms of being able to control the money they earn and to get involved in marketing away from home (Safilios-Rothschild, 1983). In some provinces and countries in sub-Saharan Africa, on the other hand, land is a scarce resource, and it is possible that additional wives become more threatening and less acceptable to first wives.
Macro-policies such as agrarian reform, population, and producer pricing policies have significant potential social and economic consequences for households and may have a differential impact on household members. Probably the least recognized and studied macro-policies in terms of their impact on household members are the producer pricing policies. In several sub-Saharan African countries (e.g. Zambia), such policies have been adopted, often favouring "men's crops," and thus creating considerable intra-household conflict regarding the mode of allocation of different household members' time and labour to agricultural production, as well as the allocation of resulting profits. Under these conditions, husbands may attempt to increase the size of land on which the favoured crop is cultivated, but are quite dependent on their wives' (and children's) labour for the successful outcome of this expansion. Wives, on the other hand, have little or no motivation to reallocate their labour so as to give their husbands more time at the expense of time for their own crops, the profits of which they control (Jones, 1983). There is, therefore, a source of built-in rigidity on the part of wives to adapt to macro-policies having potentially substantial social and economic consequences for intra-household allocation of time, labour, and income when their own status as independent farmers and producers is threatened. They are reluctant to change their status to that of unpaid or poorly paid family workers on their husbands' plots. The situation becomes even more complicated when official producer pricing policies encourage the cultivation of "men's" crops (cash crops) and also of "women's" crops such as sorghum, millet, and cassava. This is true in Zambia, where wives have an even greater vested interest in allocating their labour inputs to their own crops. The extent of women's rigidity or adaptability, as evidenced by their resistance or willingness to compromise and achieve a balance of labour-sharing between their own crops and their husbands', depends on several things.
A first important factor is the nature of the cost-benefit balance perceived by men and women as a result of changes in labour requirements, prices, and market demand for "men's" and "women's" crops and women's ability to control their crops and the resulting income. The types of incentives available for women's labour investment play a crucial role in the way conflicts are resolved. These incentives depend, on the one hand, on men's willingness to reward women's labour inputs to their crop production and, on the other hand, on women's ability to take advantage of formal markets and prices for their products without losing control over their own crops or income as they did in Cameroon (Jones, 1983; Guyer, 1978). Women's ability to take advantage of formal markets and prices may be curtailed by their not being members of the cooperatives which buy the crops, and, even more importantly, by their husbands, who may take over the formal market transactions of their wives' crops because now they involve greater sums of money. Husbands may be able to do so because they themselves are members of cooperatives.
Second, women's access to land and other assets independently of men and/or access to reliable income-generating activities can also influence considerably the extent to which wives are willing to adapt to the agro-policy changes. The lesser the women's access to land, assets, and income-generating activities, the greater their willingness to compromise by apportioning their labour between their husbands' crops and their own. On the other hand, in several sub-Saharan countries where women own their own land (via inheritance or outright purchase) on which they cultivate cash crops in addition to food crops, they may be under no obligation to provide their husbands with labour (Vellenga, 1977).
Finally, since marital dissolution is the limiting factor in this type of marital conflict (this is especially true in sub-Saharan countries), the degree of social acceptability of divorce, the probability of remarriage for women, and the viability of female-headed households, which in turn is determined by access to significant income-earning opportunities, further conditions wives' (and husbands') degree of rigidity or adaptability.
In conclusion, adaptability to social and economic changes on the part of household members generally appears to be determined by the extent to which these changes threaten their access to valued resources, especially those of power and prestige. Household members, therefore, tend not to perceive threatening changes or to minimize them as long as possible. Alternatively, they may react with rigidity to changes they cannot minimize or make invisible. For development policies and programmes it is crucial to determine what facilitates household members' adaptability to social and economic changes affecting their access to resources. This is particularly important since development interventions cannot directly interfere with household dynamics in order to obtain desirable outcomes in terms of intra-household resource allocation: they can only alter household members' access to resources, and thus indirectly affect such outcomes. Intra-household dynamics determine whether or not and to what extent changes in household members' access to resources lead to changes in intra-household allocation of resources.
The theoretical frameworks presented in this paper and illustrated with concrete examples from developing countries provide some important clues as to how development policies and programmes can increase household members' access to key resources and thus promote a more equitable distribution of resources within the household. The following findings seem to be the most important for development interventions.
NOTES
REFERENCES
Bekele, F. 1982. Report of Mission to DANA and Golgotta Settlement Schemes. Mimeo. Addis Ababa.
Collins, R. 1975. Conflict Sociology. Academic Press, New York.
Fapohunda, E.P. 1987. The Nuclear Household Model in Nigerian Public and Private Sector Policy: Colonial Legacy and Socio-political Implications. Devel. Change, 18: 281-294.
Folbre, N. 1984. Household Production in the Philippines: A Non-neoclassical Approach. Econ. Devel. Cult. Change, 32: 303-430.
de Graft-Johnson, K.E. 1984. Oral Presentation at the Inter-regional United Nations Seminar on Changing Families: Relevant Social Welfare Strategies, Moscow, USSR, 1984.
Guyer, J.I. 1978. Women's Work in the Food Economy of the Cocoa Belt: A Comparison. Working Paper 7. Boston University African Studies Center, Boston, Mass.
Jones, C. 1983. The Mobilisation of Women's Labor for Cash Crop Production: A Game Theoretical Approach. Am. J. Agric. Econ., 65(5): 1049-1054.
McCormack, J., M. Walsh, and C. Nelson. 1986. Women's Group Enterprises: A Study of the Structure of Opportunity on the Kenya Coast. World Education, Inc., Boston.
McSweeney, B.G. 1979. An Approach to Collecting and Examining Data on Rural Women's Time Use and Some Tentative Findings: The Case of Upper Volta. Working paper prepared for the Seminar on Rural Women and the Sexual Division of Labor of the Population Council, New York, 30 March 1979.
Noble, A.L., and M.F. Nolan. 1983. Sociological Constraints and Social Possibilities for Production of Goats in Western Kenya. Department of Rural Sociology, University of Columbia, Columbia, Mo.
Safilios-Rothschild, C. 1979. Access of Rural Girls to Primary Education in the Third World: State of Art, Obstacles and Policy Recommendations. Agency for International Development, Office of Women in Development, Washington, D.C.
-. 1980. The Role of the Family: A Neglected Aspect of Poverty. In: P. Knight, ea., Implementing Programs of Human Development, pp. 311-372. Staff Working Paper 403. World Bank, Washington, D.C.
-. 1982a. A Class and Sex Stratification Model and Its Relevance for Fertility Trends in the Developing World. In: C. Hohn and R. Mackensen, eds., Determinants of Fertility Trends: Theories Re-examined, pp. 191-202. Ordina Editions, Liege.
-. 1982b. Adequacy of Socio-economic Indicators for Monitoring and Evaluation of the lmpact of Agrarian Reform and Rural Development on Low-income and Disadvantaged Groups. Background document for the Expert Consultation on Socio-economic Indicators for the North African and Middle Eastern Region, organized by FAO in Tunis, Tunisia, 1982.
-. 1982c. Pertinencia de los indicadores socioeconómicos pare el seguimiento y la evaluación de los grupos de bajos ingresos y desfavorecidos. Background document for the Expert Consultation on Socio-economic indicators for the Latin American and Caribbean Region, organized by FAO in La Paz, Bolivia (ESH/ESS:ARRD/LA/4), 1982.
-. 1983. Family and Development in Sub-Saharan Africa. Paper commissioned by the Centre for Social Development and Humanitarian Affairs, Social Integration and Social Welfare Section, Vienna International Centre.
-. 1985. The Persistence of Women's Invisibility in Agriculture: Theoretical and Policy Lessons from Lesotho and Sierra Leone. Econ. Devel. Cult. Change, 33(2): 299-317.
-. 1987. Farming Systems and Gender Issues. Ministry of Agriculture and Fisheries of the Netherlands, The Hague.
-. 1988. The Impact of Agrarian Reform on Men's and Women's Incomes in Rural Honduras. In: D. Dwyer and J. Bruce, eds., A Home Divided: Women and Income in the Third World. Stanford University Press, Stanford, Calif.
Safilios-Rothschild, C., and M. Dijkers. 1978. Handling Unconventional Asymmetries. In: R. Rapaport and R.N. Rapaport, eds., Working Couples, pp. 62-73, Routledge & Kegan Paul, London.
Staudt, K.A. 1979. Class and Sex in the Politics of Women Farmers. J. Politics, 41 (2): 492512.
Sundar, P. 1981. Khadgodhra: A Case Study of a Women's Milk Cooperative. Soc. Action, 31: 79-88.
Vellenga, D.D. 1977. Differentiation among Women Farmers in Two Rural Areas in Ghana. Labour and Society, 2(2): 197-208.
Youssef, N.H. and C.B. Hetler. 1983. Establishing the Economic Condition of Womanheaded Households in the Third World: A New Approach. In: M. Buvinic, M.A. Lycette, and W.P. McGreevey, eds., Women and Poverty in the Third World, pp. 216-243. Johns Hopkins University Press, Baltimore, Md