Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Gender and agricultural systems in North-Eastern Ghana

The Kusasi farming system has been described as a predominantly male farming system. All household members old enough to work are expected to labour on the compound or household farm, and it is from this farm that household members are fed for most of the year. The produce is controlled by the household head, who allocates it to each married woman on a regular basis for preparing the family's meal.

Men and women also cultivate private plots and have complete rights of disposal over the produce from these. Men's fields in Zorse, however, tend to be bigger than those of women, which average only about 0.25 ha, as compared with an average of 0.8 ha for men's fields. As a result, men's income from their private fields tended to be higher than that of the women. Women's fields had a greater range of cash crops, including cotton, kenaf, fibres and dry season vegetables. Small livestock could also be reared by women, which was used as a source of cash to meet emergencies.

Women's fields were smaller than those of men because of the difficulties women face in getting access to land. The Kusasi land tenure system is characterised by communal ownership, with individual lineages or families, headed by men, owning portions. Thus, theoretically each member of a kinship group, male or female, has rights to land by virtue of membership of the group. However, in practice the authority to decide on land allocation is delegated to male lineage heads or household heads with women's access dependent on the goodwill of the male members of the household or lineage. Second, most of the land allocated for farming is inherited under a patrilineal system of inheritance, with the majority of women deriving their use rights from their husbands. More than half (56.5 per cent) of the women who acquired land obtained it from their husband's kin group, 30.7 per cent from their own kin group and 12.8 per cent from other sources.

The average age of women who acquired land through their own efforts was 54 years, suggesting that as women grow older they gain more independence.

The gender division of labour

The division of labour on the farm has become more gender neutral in response to shortages of male labour. Traditionally, the division of farm labour was gender specific, with men in charge of land clearing, ploughing, weeding and some elements of harvesting. Women mainly planted, harvested, transported the crops home and threshed the millet (table 15.1). However, over time, due to net male outmigration, women have taken on new tasks and now weed on all three types of farms and are involved in some initial land clearing on their private plots, although men still cut down the trees. Men have access to the unremunerated labour of their wives and often of all women in the household on their private plots, but women do not have automatic access to their husband's labour on their private plots. Often women use their own labour or obtain labour from male members of their kin group or in-laws, but have to provide them with food and drink. The ability to command labour is, therefore, crucial in determining the household's viability.

Table 15.1 Division of Farm Labour among the Kusasi

Farm type Crops grown Female tasks Male tasks
Household farms Millet Planting, weedinga, harvesting, transporting crops, processing crops Clearing fields, planting, weeding, harvesting
Men's private fields Millet, sorghum, rice, groundouts Planting, weedinga, harvesting, transporting, processing Clearing fields, planting millet, weeding, harvesting
Women's private fields Groundnuts, rice, cowpeas, milleta Clearing fieldsa, planting, weedinga, harvesting, transporting, processing Clearing fields, weeding, harvesting

Source: Fieldwork, Zorse, 1991.
a. Activities which have undergone changes in the sexual division of labour.

The type of crop grown was also traditionally gender specific. Men had greater responsibility for the main staple crops of millet and sorghum, which were grown on the household farm and controlled by the household head. Millet could also be grown on men's private plots. Women were not allowed to grow millet on their private fields, although their labour was crucial for its growth on the household farm and they were thus completely dependent on men for the staple food. Thus, by controlling rights to certain basic crops, men kept women dependent on them. Women were, however, responsible for supplementing the diet with the produce from their private fields, groundnuts, rice and legumes.

However, as it became more profitable for men to concentrate on cash crops and as yields of millet from the household farms declined because of erratic rainfall and soil impoverishment, women began growing millet on their private plots to make up for the shortfall in household millet production. Crops from women's private fields are traditionally viewed as the "hunger crop," to be used when the supply of millet from the household granary is used up. Thus, women's agricultural production is very important for the survival of the household.

The introduction of cash cropping in Zorse appears not to have altered substantially the gender division of labour in agriculture, but it has led rather to an intensification of women's work. Women are both working harder by taking on additional tasks such as weeding and losing resources by having to use their own private land for millet production for the household. The gender division of labour has, therefore, become more gender neutral and flexible in response to cash cropping, the shortage of male labour and declining yields because of environmental degradation.

Structural adjustment and its impact on health, nutrition and consumption patterns

The follow-up survey of 124 households in 1991 revealed gender differentiated changes in health and nutrition, food consumption, education and women's workloads, which are discussed in the subsequent sections.

Even before the introduction of SAPs, the quality of life in northern Ghana, in comparison with the other regions, was low. With 19 per cent of the nation's population, the northern sector accounted for only 5 per cent of public health service attendance. The SAP-forced cuts in government expenditure on health care, the introduction of user charges at hospitals, the withdrawal of supplementary feeding programmes and increases in food costs further exacerbated the problem. The impact of these measures on the rural poor in north-eastern Ghana was that health services became expensive and inaccessible to many families. Attendance at the maternal and child welfare clinic in Bawku hospital, for example, dropped by 15 per cent between 1984 and 1991, and attendance at the three hospitals in the region dropped from 104,447 in 1986 to 91,518 in 1991 (Ministry of Health 1991).

With the food price rises associated with the SAP measures, women as mothers and care takers of the family have found it very difficult to afford to buy the required food to supplement declining yields from household farms. As a result, the general level of nutrition and health of many families declined in the region. Diseases related to dietary deficiencies increased, with anaemia and malnutrition becoming the commonest diseases reported at the Bawku district hospital in 1991. The incidence of child malnutrition in the region also increased from 52 per cent in 1986 to 70 per cent in 1990 (Ministry of Health 1991).

In Zorse, the data point to a decline in health status during the seven year period. Infant mortality, which can be used as an indicator of the health status and quality of life, increased in Zorse from 160 to 450 deaths per thousand live births, while infant and child mortality showed a smaller increase from 190 to 220 deaths per thousand live births. The high increase in infant mortality may be linked to the large number of undernourished women, who are at greater risk of giving birth to low birth weight babies, who in turn face a higher risk of mortality, and to the higher health charges which led to reduced attendance at maternal and child welfare clinics.

Many households made some changes to their diets in response to rising food prices and declining yields from household farms. Fifty-two per cent of the women were of the opinion that both the quantity and quality of the food they consumed had deteriorated since 1984, because of either a lack of money to buy or invest in growing items or the need to sell crops and consume only the lowest value products, such as millet. Intra-household food distribution, which favours men and boys in Zorse, is such that when households have to reduce food consumption, it is likely that the consumption by women and girls is reduced more than that by men and boys.

Table 15.2 Distribution of Women by Level of Education, Zorse (%)

Level of education 1984 1991
No education 93.4 85.0
Primary school 2.2 12.4
Middle school 3.6 1.8
Post-middle 0.8 0.8
Total 100.0 100.0

Source: Field surveys, 1984 and 1991.

Changes in educational status

The adjustment policies of the introduction of school and book user fees, and housing and feeding charges at secondary schools and universities, worsened the already limited access to education in the region, for girls in particular. Primary school fees of 500 cedis per term were introduced in Zorse in 1985. In addition, the parents had to provide school uniforms, a chair, a table, books and stationery, which were previously provided free by the government, with the exception of uniforms. These expenses were estimated to amount to about the equivalent of 10 pounds sterling in 1991. The government has argued that fees at the primary school level are modest, but with an average yearly income of about the equivalent of 12 pounds in 1991 for women in Zorse, the cost of educating one child equals about a year's income, thus making education inaccessible to low income families.

Although very few women had any formal education in Zorse, the number doubled over the seven year period from 7 per cent in 1984 to 15 per cent in 1991 (table 15.2). In spite of the increasing number of women with some education, the level of education achieved was lower in 1991. Of the 7 per cent who had some formal education in 1984, 66 per cent had more than a primary school education, compared to only 17 per cent of the 34 per cent of women in 1991. These figures indicate a growing drop-out rate for girls, particularly after primary school. National data for the Upper East region indicate a drop of 6 per cent in school enrolment rates between 1984 and 1991 (Statistical Services 1989).

Thus, even though the nutritional, health and educational status of the population in the region has always been low, erratic rainfall conditions, declining soil fertility and the effects of adjustment policies have reinforced the deteriorating circumstances of many households.

Table 15.3 Income-earning Activities of Women, Zorse, 1984 and 1991 (%)

Activity 1984
(N = 250)
(N = 226)
Farming 25.2 36.7
Trading 38.6 20.4
Beer brewing 9.6 8.4
Food processing 3.2 9.7
Handicraft production 4.8 7.1
Formal sector 4.2 4.4
None 14.4 13.3
Total 100.0 100.0

Source: Fieldwork, Zorse.

Changes in income-generating activities

SAP policies of retrenchment of workers from the public sector meant increasing unemployment and an influx of retrenched workers into the formal sector at the national level. In Zorse, with only 2 per cent of women engaged in the formal sector, there was very little change in the number of women engaged in an income earning activity between 1984 and 1991. However, there were changes in the types of income generating work over the seven year period, with more women undertaking farming and food processing rather than trading in 1991 (table 15.3).

This may suggest that women are cashing in on the higher prices of crops as a result of SAP policies, as found in a study among women in southern Nigeria (Guyer and Idowu 1991). In Zorse, however, the incentive to move into farming was not the higher producer prices of food crops, since these are produced at a higher cost, but survival. As most women farmers put it, "at least with farming, one's children would not go hungry."

Changes in women's time use

The changes outlined above have had a number of important implications for women. In their multiple roles as mothers and wives, producers and community managers, women have been most affected by the adjustment policies in terms of increasing workloads and conflicting demands on their time (Moser 1992). Time use studies can be used as a measure of the degree of involvement in different types of work, and help to reveal the daily and seasonal fluctuations in the demand for labour and trade-offs between different categories of work. The assumption is that the amount of time used for work can provide a reliable measure of the workload of various household members. In Zorse, the sexual division of labour is noticeable from the age of six onwards and gradually becomes more rigid with age, with women being in charge of almost all domestic chores.

Table 15.4 Average Daily Time Spent on Reproductive Work by Women, Zorse, 1984 and 1991

Activity Wet season Dry season
1984 1991 1984 1991
(N = 250) (N = 226) (N = 250) (N = 226)
Household maintenance
Cooking 1.5 1.7 2.3 2.0
Child care 1.6 1.2 1.4 1.1
Water collection 1.4 1.1 1.9 1.2
Cleaning/washing 1.1 0.6 0.8 0.8
Total housework 5.6 4.6 6.4 5.1
Kitchen gardening 1.5 1.4 0.6 0.4
Total 7.1 6.0 7.0 5.5

Source: Fieldwork, Zorse, 1984 and 1991.

Table 15.4 shows that cooking, including food preparation, is the most time consuming domestic chore and also the most gender specific task. It was the activity which revealed the least equitable distribution between members of the household. Of the 180 married women living with their husbands, 72 per cent did the cooking themselves and the remainder had some help, mainly from daughters. In the wet season, cooking is done only in the evenings, but in the dry season, when food is more plentiful, a morning meal may be cooked as well. Cooking and food preparation are laborious tasks involving the pounding of vegetables and dried fish in a mortar and the winnowing of millet and sorghum before taking the grain to the mill for grinding into flour.

Collecting water takes about an hour of women's time each day on the average, but this time varies with family structure and size and with the season. The main sources of water in Zorse are a borehole situated at the centre of the village, a well and a dam located about 6.4 km to the south of the village. Water is carried in wide enamel bowls which weigh about 25 kg when full.

Firewood collection also takes about an hour each day. The main types of fuel used are dried millet and sorghum stalks and wood obtained from gathering fallen twigs and branches from the nearby forest reserve or from surrounding farms. After the harvesting of the millet grain, the stalks are left to dry on the fields and later gathered and transported to the compound. These are stockpiled and serve as the main source of fuel from November to about March. When the millet stalks run out, and the twigs and branches from the forest reserve are no longer available, women have to walk long distances, sometimes even going into neighbouring Burkina Faso to gather fuelwood. Women usually organise these trips in groups about once a week and return carrying bundles of wood weighing up to 30 kg.

Cleaning, washing clothes and utensils and other housework are the least time consuming activities. The mud huts do not entail much cleaning and sweeping, and this is normally shared out among the female children in the household. Kitchen gardening is a wet season activity. Each married woman cultivates a garden on the land directly behind her hut, growing mainly vegetables for the family's meals. Women also keep small livestock, mainly poultry, sheep, pigs and goats in the compound.

Women's time use and seasonality

The marked seasonality of rainfall that characterises north-eastern Ghana means that the availability of resources changes throughout the yearly cycle and has a profound impact on the types of activities for both men and women in northern Ghana. The seasonal nature of rainfall means that for six months, from about May to October, everyone's time and energy go into producing enough millet to feed the family and to earn some cash for the whole year. At the same time, women must take care of and maintain their households. Like rural women in the developing world generally, the women of north-eastern Ghana are involved in a "zero-sum game," a closed system in which time or energy devoted to any new effort must be diverted from another activity, the effect of which is reflected in table 15.4 in relation to seasonal time use as well as a comparison between 1984 and 1991. Cooking and food preparation time increases in the dry season, when there is enough food for two meals a day. However, the increase was less in 1991 than in 1984 because of other competing demands on women's time. Cleaning the house and washing clothes also take more time in the dry season, as it is a dusty period with trees shedding their leaves, thus demanding more sweeping and dusting. Clothes have to be taken to the water source for washing rather than using rainwater, as can be done in the wet season. Water collection time increased in the dry season in 1984 as a result of the drying up of the dam and wells. The construction of a borehole in Zorse in 1985 increased the availability of water in both seasons in 1991.

Time spent in gathering fuelwood, on the other hand, is less in the dry season, because of the availability of millet and sorghum stalks stored in the compound. Time use declined from 1.2 hours in the wet season to 0.7 hours daily in the dry season in 1991. In the dry season, 62 per cent of the women used millet stalks as their main source of fuel, compared to only 9.6 per cent in the wet season. This intensive use of millet stalks gathered from the grain fields leaves the soil bare and exposes it to sheet and gully erosion, thus contributing to the degradation of the environment. Kitchen gardening also takes very little of women's time in the dry season, as it is dependent on rainfall.

In terms of seasonal time use between 1984 and 1991, there was a smaller increase in time spent on household maintenance activities in the dry season in 1991 than in 1984 (table 15.4). This was compensated for by an increase in the time spent on income earning activities (table 15.5), with more women undertaking farming and food processing.

Total average daily hours of work between 1984 and 1991 fell from 11.3 hours to 10.6 hours in the wet season, while dry season work hours increased from 11.7 to 12.4 hours. The total daily average has, therefore, not changed between the two years, but the balancing of time between women's various roles has been adjusted. The findings confirm those from Ecuador during a period of economic crisis (Moser 1992). Time spent in Zorse in reproductive work and social duties fell in both seasons between 1984 and 1991, but work hours in income earning activities increased by almost 1 hour in the wet season and by about 3 hours in the dry season (table 15.5). Thus, in 1984, whereas in the dry season work hours in household maintenance and in social duties increased and productive work hours decreased, the opposite prevailed in 1991. Reduced time in reproductive work and social duties in the dry season was used to earn more income in 1991. This suggests the dry season was used to generate and diversify income, as has also been noted in Burkina Faso (Reardon et al. 1988). The traditional view of the dry season as a period of rest, when ceremonies such as funerals and festivals can be organised, is changing. As household incomes fall with deteriorating environmental conditions and economic restructuring, women are making more use of the dry season to generate additional income, thus increasing their workloads.

Table 15.5 Total Average Daily Time Use of Women, Zorse, 1984 and 1991 (hrs/day)

Activity Wet season Dry season
1984 1991 1984 1991
Reproductive work 7.1 6.0 7.0 5.5
Household maintenance 5.6 4.6 6.4 5.1
Kitchen gardening 1.5 1.4 0.6 0.4
Social duties 0.9 0.5 2.4 1.7
Productive work 3.3 4.1 2.3 5.2
Total time use 11.3 10.6 11.7 12.4

Source: Fieldwork, Zorse, 1984 and 1991.

When asked if they thought they were working harder in 1991 than seven years earlier, 60 per cent of the women thought they were, while 32 per cent thought their workload had not changed. As described by Lariba,

We work more than our mothers in the past. Our mothers in the past only busied themselves with preparing food for the household. After that they sat down and relaxed. They only had to grind the millet and plant on household farms. Our mothers did not weed. Today, we work from sunrise in our domestic tasks and in addition go and work on the farm. We weed and in the evening we have to go home and prepare the evening meal for the family. This is what we do daily from sunrise to sunset. Our mothers in the past never did this. We work and work and we are tired and poor.


This paper illustrates how macroeconomic adjustment policies have combined with environmental degradation to make women in Zorse, a small savanna village in northern Ghana, more vulnerable to impoverishment. The effects of adjustment - rising food prices, increases in the cost of social services and declining real and household incomes, combined with declining and erratic rainfall conditions and deteriorating soil fertility - have meant increasing workloads and falling standards of living. There is a visible process of impoverishment taking place in Zorse, which is reflected in deteriorating health and nutritional status, changes in the pattern of food consumption, a growing school drop-out rate for girls, and an intensification of women's workloads, as shown by increasing work hours, especially in the dry season, and the intensification of labour inputs in agriculture. In spite of increased workloads, women in Zorse appear to be losing access to land and cash income, as their private land is now used for cultivating the staple crop millet for household consumption.

The division of labour on the farm has become more gender neutral as women take on tasks traditionally reserved for men in response to declining household incomes. At the same time, household duties remain gender specific and men have not taken on new responsibilities.

In response to the question of whether men or women are more affected by the crisis, almost all women thought themselves to be in a worse position relative to men, as they felt that women were more concerned with and responsible for household welfare. Gender inequalities in access to land, credit, capital and labour increased women's difficulties.

By ignoring the sexual division of labour in work and intra-household distribution of resources, as well as women's triple role, macroeconomic policies of structural adjustment appear to have given rise to greater gender inequalities and placed heavy burdens on rural women. By doing so, they have increased pressure on the resources of the environment of the poorest people and thereby exacerbated the downward trend of population pressure and environmental deterioration.


  1. A 1988 study found that the percentage of the "hard core poor," defined by the World Bank as the population falling below a per capita yearly income of 18,562 cedis, or approximately 20 pounds sterling, has been on the increase in Ghana since the mid-1980s (Boateng et al. 1990).
  2. A survey in the Upper East region (National Food and Nutrition Board 1986) found that 30 per cent of males and 50 per cent of females were underweight for most of the year, but during the ''hungry season" this imcreased to 49 per cent of men and 63 per cent of women. The number of low birth weight babies has also been on the increase since the mid-1980s, more than doubling from 7.3 per cent in 1988 to 19.5 per cent in 1991 (Ministry of Health 1991).
  3. Data for time spent collecting fuelwood were unavailable in 1984 and as a result are omitted from the 1991 data presented in tables 15.4 and 15.5.


Bening, R. 1975. Colonial development in northern Ghana, 1898-1950. Bulletin of the Ghana Geographical Association 17: 659-79.

Benneh, G. and E. Gyasi. 1993. Farming and land degradation in stressed environments: the case of the Upper East region. Paper presented at the Second Congress of African Geographers, Rabat, Morocco, 19 April 1993.

Boateng et al. 1990. A poverty profile of Ghana, 1987-1988. SDA Working Paper, No. 5. Washington, DC: World Bank.

Commonwealth Secretariat. 1989. Engendering Adjustment for the 1990s: Report of a Commonwealth Expert Croup on Women and Structural Adjustment London: Commonwealth Secretariat.

Cornia, G., R. Jolly and F. Stewart, eds. 1987. Adjustment with a Human Face: Protecting the Vulnerable and Promoting Growth, Vol. 1. Oxford: Clarendon Press.

Government of Ghana. 1987. A Programme of Structural Adjustment: Report Prepared for the Fourth Meeting of the Consultative Group on Ghana. Paris.

Guyer, J. and O. Idowu. 1991. Women's agricultural work in a multi-modal rural economy: Ibarapa District, Oyo State, Nigeria. In: C. Gladwin, ea., Structural Adjustment and African Women Farmers, 257-80. Gainesville, FL: University of Florida Press.

Ministry of Health. 1991. Upper East Regional Health Services Annual Report, 1991. Bolgatanga: Ghana.

Moser, C. 1992. Adjustment from below: low-income women, time and the triple role in Guayaquil, Ecuador. In: H. Afshar and C. Dennis, eds., Women and Adjustment Policies in the Third World, 87-116. London: Macmillan.

National Food and Nutrition Board. 1986. The Nutrition of Adults in Northern Ghana: A Report of the National Nutrition Survey of the National Nutrition Board. Accra: Ghana.

Reardon, T., P. Matlon and C. Delgado. 1988. Coping with household level food security in drought affected areas of Burkina Faso. World Development 16: 1065-74.

Toye, J. 1991. Ghana case study. In: P. Mosley, J. Harrigan and J. Toye, eds., Aid and Power: The World Bank and Policy-Based Lending, Vol. 2, Case Studies. London: Routledge.

UNICEF. 1987. Adjustment policies and programmes to protect children and vulnerable groups in Ghana. In: G. Cornia et al. (1987), pp. 93-125.

Contents - Previous - Next