DO NOT REmnR r TAPLE
RESERVE DESK ou'I.'AL
TRAINING MATERIALS FOR AGRICULTURAL PLANNING
credit and marketing project
for smallholders in Swaziland
FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS
TRAINING MATERIALS FOR AGRICULTURAL PLANNING
SOCIAL ANALYSIS OF AN AGRICULTURAL INVESTMENT PROJECT
WITH EMPHASIS ON THE ROLE OF RURAL WOMEN:
A CASE STUDY
ON THE CREDIT AND MARKETING
PROJECT FOR SMALLHOLDERS IN SWAZILAND
Alice Stewart Carloni
FOOD AND AGRICULTURE ORGANIZATION OF THE
TABLE OF CONTENTS
I. BACKGROUND 2
1. The economy of Swaziland 3
2. The Rural Development Area Programme 4
3. The project 4
4. The case study methodology 5
II. THE SOCIAL SOUNDNESS OF THE IRRIGATED VEGETABLE AND
RICE PACKAGE 7
1. Women's role in production: who is responsible for
irrigated vegetable and rice cultivation? 8
2. Access to credit: can all scheme farmers qualify
for loans? 11
3. Employment generations who will provide the
additional labour for production? 13
4. Seasonal labour increases: can all scheme farmers
increase wet season labour inputs by 90%? 14
5. Changes in the crop mix: what is the incentive to
switch from green maize to summer vegetables? 16
6. What is the incentive to increase rice production? 16
7. Market constraints versus price disincentives:
is the lack of a market the main constraint for
expanding vegetable production or is it the price? 17
III. THE SOCIAL SOUNDNESS OF THE RAINFED MAIZE PACKAGE 18
1. Women's role in production: who is responsible
for rainfed maize production on Swazi Nation
2. Substitution of male labour; is increase access
to production inputs likely to increase output? 20
3. Credit, input use and sale of cops: will credit
facilitate production of a marketable surplus? 21
4. Consumption of incremental production: will farmers
sell the additional maize or will they eat it? 21
5. Maize marketing: it lack of a market the main
-constraint for production? 22
6. Will female farmers have access to agricultural
7. How will women's access affect the project's
social impact? 23
IV. CONCLUSIONS 24
1. The irrigated vegetable and rice package 24
2. The rainfed maize package 24
V. DESIGN ALTERNATIVES WHICH WOULD STRENGTHEN THS PROJECT 25
APPENDIX: Methodological Note on Social Analysis 29
Notes for trainers 37
The case study was developed to teach project planners in
developing countries how to take the role of women into account
at the time of project preparation. It illustrates how social
soundness analysis can be used to verify assumptions about how
farmers of different types will respond to the project strategy.
An actual project, the Smallholder Credit and Marketing project
in Swaziland. is analyzed to show how the achievement of the project's
objectives depends on recognition of women's role. Three factors
are emphasized the access of women to project inputs and services,
the ability of female farmers to mobilize the necessary labour and
the incentives for responding in the expected way.
The role of women in production is placed within the broader
context of male migration. Male migration, in turn, is examined in
the context of prices, returns and incentives.
The project is shown to have little probability of achieving
its objectives because the role of women was overlooked. Design
alternatives which might strengthen the project are suggested.
The Appendix includes a "Methodological Note on Social Analysis".
It explains where social soundness analysis fits in the project cycle,
where gender analysis comes in, and outlines a strategy; which would
permit project planners to test their assumptions about farmers
through the use of Rapid Rural Appraisal techniques.
Notes for trainers are also available. They suggest ways of using
the case study and identify possible exercises which trainees could
carry out. Utilization of the case study in other countries is also
In 1982, the Policy Analysis Division of FAO commissioned a
case study on the role of women and its implications for the preparation
of agricultural investment projects. The site chosen for the case
study was Swaziland, where a training course in project preparation
was under way for staff at the Ministry of Agriculture and Cooperatives.
The purpose of the case study was to develop materials which could be
used to train agricultural planners to take the rdle of women into
account during the process of project preparation.
Swaziland was thought to be an ideal choice for several reasons.
As part of the training exercise, participants in the course were to
replicate the preparation of an actual project. The project selected
was the "Credit and Marketing Project for Smallholders", prepared by
the FAO Investment Centre on behalf of the International Fund for
Agricultural Development (IFAD). Its objectives were to increase food
production and raise farm incomes in the traditional sector of Swazi
agriculture. .-Special attention was given to irrigated vegetable and
rice production, rainfed maize production and marketing support.
Vegetable, rice and maize production on Swazi Nation Homesteads
are typically women's work. Except for land preparation, practically
all agricultural operations are performed by women. Women do over 55%
of the work in agriculture, while men contribute less than 30%.
Opportunities for higher earnings in South Africa and in towns
have led to an outflow of adult males from subsistence agriculture.
At any given time, over half of the male labour force is absent from
the farm. Eighty percent of Swazi Nation Homesteads have at least one
wage-earning member. Those remaining on the farm are family members
with limited wage-earning potential, such as old people and children,
or high off-farm opportunity costs, such as women of childbearing age.
Twenty-one percent of rural households are headed by women. An
additional 21% have no adult males living at home. Women's responsibilities
are increasing as fewer men return home to help with ploughing for fear
of losing their jobs and children spend more time in school. Inadequate
land preparation and late ploughing in men's absence are known to be a
major cause of low yields.
The case study illustrates several principles of interest to project
planners. The first is the traditional division of labour by sex and
its implications for project design. The second is the impact of male
migration and off-farm employment on the traditional division of labour
and on farm production. The third is the impact of prices and markets
How will off-farm employment of men affect the viability of the
project? In particular, how will it affect labour availability and
incentives for intensification of production? What proportion of those
responsible for irrigated vegetable and rice and rainfed maize production
are women? Do all female farmers have access to project benefits? Is
the incidence of female farmers higher among the rural poor? If so,
will the project have a differential social impact because households
without adult males have less access to benefits?
1. The economy of Swaziland
The Kingdom of Swaziland is the smallest African country, with
a population of barely half a million. It is closely linked with
the economy of South Africa through its membership in the Southern
African Customs Union and the Rand Monetary Area.
Swaziland's development to date has been highly dualistic.
Although 85% of the GDP comes from the modern sector of the economy,
this sector employs less than a third of the economically active
population. Peasant agriculture continues to be the main source of
living for the majority of the population.
Like many developing countries, Swaziland exports agricultural
products and imports manufactured goods. The balance of trade has
usually shown a surplus, but in recent years higher fuel prices and
increased food imports have turned the balance into a marked deficit
in spite of increased sugar exports. The Government gives high
priority to reducing food imports through increased domestic production.
In keeping with the economy as a whole, the agricultural sector
is dualistic. During the nineteenth century, 44% of the land was
granted in concessions to European settlers as Title Deed Eard Land.
Today this land is largely owned by foreigners and companies engaged
in large-scale production of sugar cane, pineapples, citrus, cotton
and lumber for export.
The remaining 56% of the land was held in trust by the king
for future generations of Swazi Nation Farmers. It is used for cattle
and subsistence crops such as maize, sorghum, beans and groundnuts.
In comparison with Title Deed Farmers, Swazi Nation Homesteads are
very poor and their productivity is low.
In spite of this dualism, the two sectors of the economy are
interdependent. Farm income in the traditional sector is too low
to support a family without supplementary earnings from off-farm
work. Off-farm earnings are too low to support a family in town. In
the absence of unemployment compensation and retirement pensions, it
makes sense for off-farm workers to hold onto their land in the
traditional sector. In this way the traditional sector is functional
to the modern sector.
Those family members who can earn more in wage employment than
they can on the farm are drawn off into the modern sector. These
include educated people and males of prime working age. Women, children
and old people stay behind. When they can no longer find work, the
men return to the homestead.
In the meantime, the family members remaining on the homestead
subsidize low wages in the modern sector. They are forced to support
themselves because the husband's wages are too low. In the absence
of men, women bear the cost of supporting the aged, the disabled and
the young. There is a net transfer of resources from the traditional
to the modern sector.
2. The Rural Development Area Programme
The Rural Development Area Programme (RDAP), which was started
in 1971 with the financial assistance of the World Bank, the African
Development Bank, the Overseas Development Association and USAID, was
the first to give attention to traditional agriculture. Its objectives
are to raise the productivity of smallholder agriculture, improve
the standard of living of Swazi Nation farmers, reduce the rural-
urban income differential and move Swaziland toward self-sufficiency
in food production.
The main emphasis has been on raising the output of the main
staple crop-maize. The RDAP has also built small scale irrigation
schemes for intensive vegetable and rice cultivation. It has built
feeder roads and water supply systems. And it has provided supporting
services such as tractor hire, input supply, cooperatives, agricultural
extension and credit at subsidized rates. Nevertheless, the anticipated
effects on production have been slow to materialize. Maize self-
sufficiency-a high priority in the Third National Development Plan-
has not been achieved.
There are many problems. Agricultural extension services are
poorly organized. Government tractor hire services operate at a loss
and are unable to meet the demand for timely ploughing. Due to poor
maintenance and inefficient organization, the productivity of the
RDA irrigation schemes remains low.
Marketing arrangements for food crops remain inadequate. The
Swazi Milling Company, which is the sole buyer for maize, does not
reach small farmers in the traditional sector. There is no organized
market for vegetables, which continues to be dominated by small traders
and market women.
The Central Cooperative Union was forced to declare bankruptcy
because it was unable to recover its loans to small farmers in the
traditional sector. Although it has withdrawn from lending, the
cooperative movement has never recovered from the blow.
The sole source of smallholder credit is the Swaziland Development
and Savings Bank, which caters to larger farmers. It has little
incentive to lend to small farmers because of the high overhead cost
of delivering small amounts of money to large numbers of recipients.
Many constraints must therefore be faced if food production in
the traditional sector is to be increased and imports reduced.
3. The project
The overall aim of the Credit and Marketing Project for
Smallholders is to assist small, largely subsistence Swazi Nation
Farmers in fifteen Rural Development Areas (RDA's) to increase their
food production and incomes. The project has two main emphases:
vegetable and rice production on irrigation schemes and rainfed
maize production outside the schemes.
- 5 -
These objectives are to be achieved by: (a) encouraging farmers
to intensify their use of hybrid seed, fertilizer, insecticide and
tractor services by providing seasonal production loans; (b) improving
extension support, through collaboration with the Taiwanese Agricultural
Mission for irrigated vegetable and rice production and the USAID
Cropping Systems Research and Extension Project for rainfed maize and
other field crops; (c) providing loans for private tractor purchase
and rehabilitation of selected irrigation schemes; and (d) upgrading
marketing through provision of collection points, a Central Wholesale
Market and Produce Board and a market intelligence system.
Of the 54,000 homesteads on Swazi Nation Land, 489 farmers on
twelve selected irrigation schemes (0.9%) would benefit from the full
package provided by the project: rehabilitation of irrigation schemes,
equipment rental, intensive extension support, seasonal production loans
and marketing assistance.
An additional 3800 farmers growing maize on Swazi Nation Land (7%)
would benefit from a limited package: seasonal production loans. USAID
cropping recommendations and marketing support from the Central Cooperative
Union. Finally, 500 vegetable and fruit growers--mainly on Title Deed
Land-would benefit indirectly from the marketing component.
The cost of the project would be $ 7.7 million. Of these, $ 5.5
million would be loaned to the government by the International Fund for
Agricultural Development (IFAD) and $ 2.2 million in operating costs
would be borne by the government. Roughly half of the investment cost
would be for crop production (4% for irrigation and the rest for a
revolving fund) and half for institution building (construction of a
wholesale market, maize transit stores, buildings and installations
for the extension service, vehicles, tractors and equipment).
The entire cost of investment in crop production would be repaid
by the participating farmers. The full cost of rehabilitating irrigation
schemes would be charged to plot holders. Interest on seasonal production
loans would be increased from the current subsidized rate of 7.5% to
prime rate (14%). Charges for tractor hire would be increased from the
current subsidized rate (E. 16 per hour) to the break even rate (E. 30-32).
The rest--institution building-+would be repaid out of general taxation-
and receipts from the Customs Union with the Republic of South Africa.
The loan would be repaid to IFAD' at an interest rate of 4% over a period
of twenty years, including a grace period of five years.
4. The case study methodology
Rural development projects often fail because farmers do not respond
in the expected way. Sometimes project assistance does not even reach
the intended beneficiaries. Other times farmer response is poor because
the project strategy is inappropriate. These problems are compounded
when women are the main food producers. For their own protection,
governments should verify the assumptions about farmer response at the-time
of project preparation.
The technique used in preparing the case study is "social soundness
analysis". Rapid assessment methods are used to shed light on assumptions
of project planners about:
who is responsible for agricultural production;
the appropriateness of proposed technical solutions to the
needs and resources of the target group;
the incentives for their adoption, and
the spread of benefits to disadvantaged groups.
The object of analyzing the assumptions underlying the project
strategy is to delimit the information gathering to issues which make
a crucial difference for the achievement of project objectives. The
first step is to spell out the rationale for the project, its objectives
and strategy for achieving each objective. The next steps are to:
1. Identify assumptions about who is the target group for each
of the project components and who is expected to benefit in
what way. This includes assumptions about whether "farmers"
are men or women.
2. Identify criteria for receipt of project inputs and participation
in project activities to shed light on factors which may limit
the access of women or disadvantaged groups.
3. Examine documentation used to justify project assumptions,
identify data gaps and information needed to fill them.
4. Fill as many gaps as possible by collecting data from
5. Gather data in the field to fill remaining gaps; discuss
project assumptions directly with the intended beneficiaries.
6. Re-examine project assumptions in the light of findings and
show how restricted access of women or disadvantaged groups
to project inputs and activities or reduced incentives would
affect the rate of participation and the achievement of project
7. Suggest design alternatives more conducive to achieving
Most agricultural investment projects are prepared by teams of
international experts who fly into a country for approximately one month
to gather statistics and confer with government officials before flying
back to the main office to piece together a project proposal. They
usually spend no more than one week visiting field sites.
In the interest of realism, a conscious decision was made to adhere
to the same schedule in preparing the case study. Two weeks were allotted
for in-country preparation and meetings with government officials, '
followed by ten days for field work and four days for wrap-up meetings.
The point was to limit data-gathering to items which could be easily
collected by project preparation missions. One month was allowed for
report writing in Rome.
After reviewing existing studies and conferring with Ministry
officials, the decision was made to concentrate on the role of women
in irrigated vegetable and rice production. There was almost no
information on this subject, whereas better data already existed on
women's role in maize production than anything which could be gathered
in the allotted time period.
'Trbrthern RDA was selected as the site for data gathering,
because five of the twelve irrigation schemes included in the project
are located there. Interviews were also held with female farmers on
irrigation schemes in Central RDA.
Field visits were conducted with the assistance of Linda Vilakati.
manager of the UN Women-in-Development Project (based in Northern RDA)
and Thankful Dlamini, former Assistant Home Economics Officer for
Northern RDA. Their reputation and their network of contacts in the
project area proved invaluable in securing the full cooperation of
Field Officers, Credit Officers and farmers. A detailed description
of field work and data-gathering methods appears in the appendix.
The version of the project which is taken as a starting point
for the case study is the report of the preparation mission undertaken
on behalf of IFAD by the FAO Investment Centre. It has since beer
modified by the African Development Bank appraisal mission. The Podified
project was approved by IFAD in 1983. Owing to reshuffling of govern-
ment personnel after the death of King Sobhuza, as of December 1984,
the project was not yet operational.
In spite of these changes, the lessons drawn by the case study
remain valid. The principles are of interest to project planners
throughout Africa and not just in Swaziland. They include the division
of labour by sex and its implications for project design; the impact of
off-farm employment and male migration onr production and labour allocation;
and the impact of prices and markets on incentives.
II. THE SOCIAL SOUNDNESS OF THE IRRIGATED VEGETABLE AND RICE PACKAGE
According to the project preparation team. the highest potential
for increased food production in the traditional sector can be found on
the RDA irrigation schemes. These schemes, which were constructed
during the first phase of the Rural Development Area Programme. have
always been underutilized. Productivity remains low, due to poor
maintenance of irrigation works, inefficient management, inadequate
extension support and lack of market outlets.
The twelve irrigation schemes judged to have the highest potential
were selected for intensive support by the project. Five of these are
located in Northern RDA and the rest in other RDA's scattered around
the country. A total of 366 farmers hold plots on the twelve schemes.
Completion of irrigation works on two schemes would increase the
number of plots to 489. Irrigated plots are granted in usufruct by
traditional chiefs to Swazi Nation Homseteads. Although there is no
formal ownership, farmers holding plots are commonly referred to as
The project would provide loans for the rehabilitation of
existing irrigation works. Plot owners would be expected to take
out seasonal loans for the purchase of seed, fertilizer and insecti-
cides. Extension centres would be constructed and equipped. Swazi
extension workers would be trained in vegetable and rice production
by subject matter specialists provided by the Taiwanese Agricultural
To ensure a market for produce, the project would assist in
establishing a National Agricultural Marketing Board to regulate the
importation of vegetables from South Africa. A Central Wholesale
Market would be constructed on a main road near the capital. A
Market Intelligence Office would be established under MOAC supervision.
Extension centres would be equipped with radio transceivers to
enable them to communicate with the Market Intelligence Unit at the
Central Wholesale Market, to get the latest information on prices
and inform buyers when produce is available up country. Cold storage
would be provided at the Central Wholesale Market and collection points
for short term storage near the schemes. A fund would be established
to facilitate the Central Cboperative Union's purchase of tomatoes
All farmers on the schemes would be expected to participate.
Each would take out a seasonal production loan. In addition, plot
owners would be responsible for repaying their share of the cost of
rehabilitating the irrigation works. Discipline would increase and
farmers unwilling or unable to comply would lose their right to an
Labour requirements for irrigated agriculture would increase
by 89-95% during the wet season (summer) and by 10% in the dry season
(winter). Instead of the usual green maize, farmers would be instructed
to plant off-season vegetables during the wet season because of the
higher prices. As a result, farm incomes were expected to increase by
60-90 %. To verify the soundness of these assumptions, systematic
data gathering was undertaken on five irrigation schemes in Northern
RDA using~Iapid'Rural Appraisal techniques.
1. Who is responsible for irrigated vegetable and rice production?
During the dry season (winter), irrigated vegetables are grown
on all five schemes in Northern RDA. During the wet season (summer),
farmers on three schemes plant rice. On the remaining two schemes,
they plant green maize. The rice schemes are supervised by the
Taiwanese Agricultural Mission. The vegetable schemes are supervised
by Swazi agricultural extension officers.
IRRIGATION SCHEME DRY SEASON WET SEASON
Mashobeni vegetables green maize
Mavulandlela vegetables green maize
Mswati vegetables rice
Mgubudla vegetables rice
Mkovu vegetables rice/green maize
On the five schemes, over two thirds of the plot owners are women.
IRRIGATION SCHEME FEMALES MALES TOTAL
Mashobeni 60 4 64
Mavulandlela 8 8 16
Mswati 7 11 18
Mgubudla 5 10 15
Mkovu 6 15
TOTAL 86 42 128
Even when the formal plot owner is a male, women play an important
role in production of vegetables and rice. On the basis of farmer
interviews, discussions with extension officers and observation, the
division of labour by sex can be summarized as follows.
Irrigated vegetable production
Land preparation is by hired tractor, on both men 's and women's
Planting varies according to who owns the plot. Female plot
owners do their own planting. Male plot owners assiTt their wives.
Hired female labour is sometimes used when family labour is scarce.
Fertilizer application, insecticide spraying and irrigation are
done by whoever, man or woman, owns the plot. Ir spite of the heavy
weight of knapsack sprayers, women do their own spraying. Few women
own sprayers and most rent the equipment from male plot owners.
Weeding is exclusively women's work, whether the plot owner is
male or female.
Harvesting is done jointly by husband and wife when he is the
formal owner, but by the woman alone when she is a plot owner in her
own right. Hired female labour is sometimes used for the tomato
Sale of produce is the prerogative of the plot owner. Female
plot owners sell their own produce and male plot owners sell theirs.
Wives sometimes sell for their husbands in the husband's absence.
Irrigated rice production
The organization of rice production varies greatly between the
three schemes. On the oldest rice scheme (Mkovu), originally fifty
farmers were growing rice. Serious declines in yields have caused
34 farmers to drop out. Only fifteen are still cultivating the land,
and most of these grow green maize rather than rice during the wet
season. All production is organized on an individual basis by
Mgubudla scheme has been operating for several years. The first
time rice was grown, it was organized collectively. Since then eael-
participating household has organized its work individually.
Mswati is the newest rice scheme. Many production tasks are
organized collectively, under the supervision of the Taiwanese.
- 10 -
Land preparation is performed free of charge by the Taiwanese
Agricultural Mission the first time rice is grown. Thereafter, the
Taiwanese rent their power tillers to scheme farmers-men and women-
who operate the equipment themselves. Puddling in deep mud demands
great physical strength. Nevertheless, female plot owners operate the
heavy equipment without male assistance.
Transplanting is organized individually on Mgubudla and Mkovu
schemes. When the plot owner is a man, transplanting is a joint
activity of husband and wife. Then the plot owner is a woman, she does
it alone. Sometimes a daughter or daughter-in-law assists. On Mswati
scheme, transplanting is done collectively by the entire group of plot
owners, men and women, under Taiwanese supervision. The group finishes
one farmer's plot and moves on to the next.
Weeding is exclusively women's work, on men's plots as well as
Fertilizer application is the responsibility of the plot owner,
male or female.
Spraying is done by the individual plot owners on the two older
schemes. On Mswati scheme, where rice was grown for the first time,
the Taiwanese demonstrated the technique and a few farmers-both men
and women-were invited to try it.
Bird scaring requires the continuous presence of someone on the
scheme from dawn to dusk for about a month before the rice harvest.
The plot holder generally trades off with other members of the family
Irrigation requires someone's presence on the schemes every day,
seven days a week. The plot owner is usually responsible. The schedule
is determined by the scheme committee in consultation with the Taiwanese.
Harvesting is organized individually on the older schemes and
collectively on the newest scheme (Mswati). All family members contribute.
This is one of the seasonal peaks-for farm labour.
Post-harvest processing is mainly women's work, independently of
who owns the plot. Women carry crops from the fields to the concrete
drying floors provided by the project. Drying is their responsibility.
Threshing is done by the plot owner and his wife, or by the woman alone
in the absence of a man. Paddy is husked mechanically. There is no
tradition of hand pounding.
Marketing varies from scheme to scheme. All of the rice from
the newest scheme (Mswati) was sold unpolished to the parastatal
corporation, Tibiyo, through the Taiwanese. On the other schemes, small
quantities of rice have been polished for sale by individual farmers, to
cash in on the higher prices offered by storekeepers in the Piggs Peak
area in comparison with Tibiyo. Both men and women use the project rice
polishing equipment, but women handle the sales to shopkeepers.
In general, the "owner" of the plot, whether male or female.
provides the money for tractor hire and inputs, allocates labou.r sells
the produce and pockets the earnings. There are some differences depending
on whether the plot owner is male or female.
- 11 -
When the formal plot owner is a woman, she pays for tractor hire
and inputs out of her own savings, provides all of the work, belongs to
the farmers' association and attends scheme committee meetings. After
selling her produce, the shows her earnings to her husband,rbut the
decision how to spend the money is hers.
When the plot "owner" is a man, the situation is more complicated.
He pays for tractor hire and inputs, but someone else-a wife or
daughter-in-law--does most of the work. In one case, the six co-wives
of one of the chiefs cultivate adjoining plots, which are formally owned
by the husband. One part of each plot is "his" and the rest is "hers".
He pays for tractor hire and furnishes the seed for "his" part of each
plot. The wives pay for the rest. At the end of the season, he takes
all of the money, withdraws a share to cover his costs and other cash
expenses, and distributes the rest in six equal parts among his wives.
Other plot owners have fewer wives, but the same principle governs their
respective rights and obligations.
Decisions regarding which crops to plant, how much to spend on
inputs and when to apply fertilizer are made by the extension officers
who advise the farmers on production.
Socio-economic characteristics of plot owners
Because the scheme farmers were interviewed on the schemes rather
than the homestead, it proved difficult to get reliable information orn
their socioeconomic status. Field officers insisted that there was no
difference between farmers who have irrigated plots and their neighbours
who have none. One finding suggests that this may not be true.
Eighty percent of those interviewed were in polygamous marriages.
Other sources of data suggest that this is highly unusual (about 20% of
all marriages in Swaziland are polygamous). The incidence of polygamy
is higher among older men with greater wealth and influence. Because
irrigated plots are allocated by the traditional chief, it may be that
for clientelistic reasons plots have been given to older and more
influential homesteads. Older homestead heads are also less educated
and more attached to farming.
2. Can all scheme farmers qualify for loans?
All 489 farmers holdings plots on the twelve irrigation schemes
selected by the project are expected to take out seasonal production
loans. Otherwise the project's impact on output of vegetables and rice
will be lower and the cost per beneficiary higher. Full participation
is required to justify the investment.
The Swaziland Development and Savings Bank (SDSB) is the sole
source of institutional credit for smallholder agriculture. The Central
Cboperative Union's lending to primary cooperatives had to be dis-
continued because of widespread default. As a result of CCU's negative
experience, neither group lending nor crop lien programmes are acceptable
to Government. All loans are granted on an individual basis and require
tangible assets as collateral.
- 12 -
Land in the traditional sector is fairly evenly distributed.
However, since Swazi Nation Land is not individually owned, but is
held under customary tenure, the bank will not accept land as security
for loans. Therefore, most smallholder loans are secured by cattle.
Cattle has the disadvantage of being less evenly distributed than
land. According to a nation-wide survey, thirty-seven percent of all
Swazi Nation Homesteads own no cattle. Ownership is strongly associated
with age and wealth. Tribal elders control many cattle and many wives.
Poor homesteads and young men control few cattle. The requirement of
cattle as collateral has a built-in bias against the poor.
Because of their association with bridewealth, transactions
regarding cattle are strictly a male prerogative. Women do not own
cattle and wives have no decision-making power in their husband's
absence. Since two thirds of the scheme farmers are women, it is
important to investigate how collateral requirements would affect
their participation in the project.
Women's access to loans: SDSB figures
At the time when the case study was being prepared, the Swaziland
Development and Savings Bank did not keep separate figures regarding-
the share of agricultural loans disbursed to male versus female farmers.
On request, the credit officers went through their records on individual
borrowers to compile statistics.
Out of a total of 1430 homesteads in Northern RDA, loans were
issued during the 1980-81 season to 252 homesteads (17.6%); 96 of these
loans (or 37.8%) were granted to women. Thirteen of the loans were for
vegetables, seventy-two for maize, five for cotton and five for poultry.
This indicates that collateral requirements, per se, do not prevent
women from- obtaining loans.
On the other hand, out of 86 women holding irrigated plots, the
fact that there are only thirteen vegetable loans indicates that no more
than 15% have taken out loans expressly for this purpose.
According to the Branch Manager, loan recovery was poor when credit
officers worked with husbands, most of whom were only marginally involved
in farming. It improved dramatically when loans were made out in the
name of whoever-male or female--is responsible for farm management.
Women's repayment rate is better than that of men. The bank prefers to
work with women when they are the farmers.
Individual loans versus association loans
A different picture emerged from group meetings and interviews
with scheme farmers. Women's access to credit is excellent on the
rice schemes, but inadeuqate on the vegetable schemes. The explanation
lies in the different procedure adopted for negotiating loans.
On the rice schemes, identical loans are negotiated simultaneously
for all members of the scheme Farmers' Association, with the assistance
of the Swazi extension workers and their Taiwanese supervisors. Every
plot member-male or female-is a member of the association and all
members were able to obtain loans.
- 13 -
The loan for the entire Mgubudla Farmers' Association was
E. 3000 (about $ 3000 in 1982), or E. 200 for each of the fifteen
plot owners. As security for the loan, each individual plot owner
was mrquired to pledge three head of cattle. Although five of the
fifteen plot owners are women, the extension officer and the Scheme
Committee Representatives were able to persuade reluctant husbands to
sign the necessary papers to enable their wives to obtain loans.
Group members with cattle pledged them for those without. No one
At the end of the harvest, the Taiwanese set aside a certain
number of bags of rice from each farmer's fields to cover the cost
of the loan. All plot holders were able to repay without difficulty.
On the oldest rice scheme (Mkovu), the last association loan was
taken out during the 1980-81 season: 49 farmers participated. Because
rice yields were toolow to cover the cost of inputs, the loan has never
been repaid. Thirty-four farmers have dropped out of the rice scheme,
risking repossession of their cattle. The remaining fifteen farmers
are struggling to pay back the loan. Six of the fifteen are women.
Until the outstanding loans are repaid, none of the Mkovu farmers
will be eligible to participate in the project.
On the two vegetable schemes, each farmer who wants a loan must
apply for it individually. Under these circumstances, fewer women have
access to loans. Among the twelve farmers interviewed, both of the
men have individual loans, but only three of the ten women do. Two of
these women are wives of civil servants; one is a widow. None of the
other women had ever applied for a loan. Two would not be eligible
because their homestead has no. cattle. Two others insisted that their
husbands would never agree to let them take out a loan. Most women'
pay for inputs out of savings from the previous season's harvest, but
they would require a loan for the incremental inputs which the project
would expect them to purchase.
In spite of the bank's efforts to deal directly with whoever is
actually doing the farming, female plot holders and homesteads without
cattle remain at a disadvantage in qualifying for loans. Association
loans appear to be the best way to ensure that all scheme farmers have
equitable access to production inputs.
3. Who will provide the additional labour for production?
According to the project preparation team, labour intensive
agriculture is to be encouraged. Off-farm employment opportunities
are expected to decline in the near future, due to saturation of the
internal market and structural changes reducing the demand for unskilled
labour in the Republic of South Africa. The project expects scheme
farmers to increase their labour inputs into irrigated agriculture by
50--10% in the dry season and 89-95% for rice and vegetables in the
The unstated assumption is that scheme farmers are men. The
project is expected to generate additional employment for men, not women.
The preparation team was not aware that most scheme farmers are women.
- 14 -
The preparation report states that an income of E. 1,500 a year
is sufficient to keep men in farming. At the time, average incomes
of scheme farmers were estimated to be E. 1,440. With the project,
farm income is expected to rise to E. 2,600-2.750, which is judged to
be "sufficiently attractive to stem the migration of the more vigorous
and able workers."
Interviews with scheme farmers revealed that right noW irrigated
agriculture is nowhere near as profitable as project planners believe.
It is more profitable than other income-generating activities open to
women, but it is not profitable enough to be attractive to men.
Before turning to irrigated agriculture, the women earned cash
by selling handicrafts, beer, prepared foods and thatch grass. The
returns per manday were much lower. They no longer have time for these
activities, which they have given up to devote time to irrigated
The women were unanimous in saying that their husbands are
pleased with the choice, because their earnings from vegetables and
rice help the family. But when asked whether their husbands would be
willing to leave their jobs to help them on the irrigation schemes,
they reported that the men would never consider it.
These findings suggest that if labour inputs for irrigated
vegetable and rice production are to be increased, it will mean more
work for those providing the labour now. Since two thirds of the scheme
farmers are women and much of the work on men's plots is provided by
their wives, the success of the new package depends on whether women
have sufficient time and incentive to provide the additional labour.
4. Can all scheme farmers increase wet season labour inputs by 90% ?
Although irrigated crops are grown in both summer and winter,
labour inputs are unevenly distributed throughout the year. On the
vegetable schemes, most of the heavy work is done during the dry season
(winter); during the wet season (summer) less demanding crops are grown.
On the rice schemes, the workload is more even because wet season rice
is as demanding as winter vegetables.
In the opinion of the FAO preparation team, the potential of
irrigated plots is underutilized during the wet season. With the
project, labour requirements are expected to increase by 95% for wet
season rice production, 89% for wet season vegetable production, and
10% for dry season vegetable production. How realistic is this?
The dry season in Swaziland falls in winter, when activities in
the fields are at a standstill. The wet season falls in summer, which
is the peak season for rainfed agriculture.
Constraints on women's time are different from men's. In addition
to their year-round responsibility for fetching water and fuel. preparing
meals and caring for family members, the same women who have plots on
RDA irrigation schemes are responsible for smallstock and for growing
rainfed maize. Men's contribution to rainfed agriculture is limited to
ploughing at the onset of the rains. Men are chronically underemployed
and women are chronically overworked.
- 15 -
The advantage of irrigated vegetable production is that it gives
women a profitable way of generating income during the winter, when
work in the fields is at a standstill. During the rainy season, when
women are busy growing maize at the homestead, they have less time for
work on the irrigation schemes. One reason why they plant green maize
during the summer is that it requires less care than tomatoes and
The project calls for the opposite: a 10% increase in labour inputs
during the dry season when women are less busy and an 89-954 increase
during the wet season when women are already overworked.
During the summer, women get up at the crack of dawn. Together
with their children, between 4:30 and 8:00, they work in the maize fields.
Then she and her daughters fetch water from a nearby stream and she
prepares a thin porridge for breakfast. When the children leave for
school, she walks to the irrigation scheme, where she remains until 5 pm.
bringing a packed lunch. After returning to the homestead, she fetches
water, collects fuel, feeds the family, cleans up and goes to bed.
The peak season for rice transplanting and weeding overlaps with
the peak season for sowing and weeding maize at the homestead. Then,
when the rice forms a tassel, someone must be on the scheme from dawn to
dusk, to prevent bird damage. A second peak occurs at the time of the
rice harvest. It is followed immediately after by the maize harvest.
The women interviewed on Mgubudla scheme emphasized that rice
production leaves them no time for anything else. Maize production has
not suffered, but "during transplanting and harvesting, we must rush
with our hands and our feet and even our mouths when we eat. We are in
the fields from sun up to sundown. We no longer see our children. Our
children are doing poorly in school because they have no time to study.
They must help us in the fields. We are exhausted. Rice has ruined
our health. We could not possibly spend any more time growing rice."
If forced to choose between spending their time on maize versus
irrigated vegetables and rice, women give priority to maize. This makes
sound economic sense, in spite of the fact that earnings from vegetables
or rice may be twice those of maize. Maize is the staple of the Swazi
diet. If households fail to produce enough for their needs, they will
be forced to buy it. As long as the cost of buying maize is four times
that of growing it, it makes more sense to grow one's own than to buy
maize with the earnings from cash crops (Low 1981).
The expectation that all scheme farmers are able and willing to
increase labour inputs by 89-95% during the wet season is unrealistic.
The only households which can be expected to adopt the package are
those with a labour surplus. Because vegetable and rice production
cannot compete with off-farm employment for men, few homesteads have a
labour surplus. Female-headed households and wives of male migrants
are likely to be at a disadvantage. This will lower the rate of
participation and skew the project's social impact.
- 16 -
5. What is the incentive for farmers to switch from green maize to
At present, farmers on the two vegetable schemes grow tomatoes
and cabbage during the dry season (winter) and green maize during the
wet season (summer). The project would have them grow the following
Winter: tomatoes (40%), cabbage (30%). potatoes (10%).
onions (10%) and miscellaneous vegetables (10o);
Summer: green maize (40%), cabbage (16%). tomatoes (19%),
and miscellaneous vegetables (34%).
The justification for growing tomatoes, cabbage and other vegetables
during the summer is that these are more profitable off season. The
difficulty with growing them off season, according to the farmers and
Swazi extension workers, apart from labour constraints, is that the
high prevalence of disease during the rainy season makes their cultivation
If scheme farmers are to grow vegetables during the rainy season,
they will need to respray vegetables frequently, since daily rains wash
away insecticides. They will also need to buy more expensive varieties
of seed. Farmers-male and female-expressed reluctance to invest
large amounts of money in inputs when the risks of pest damage and disease
are high and the market returns are uncertain. They prefer to grow
green maize on their irrigated plots because corn on the cob brings
relatively high profits with low risk and much reduced labour requirements
in comparison with other vegetables.
6. What is the incentive for increasing rice production?
A weakness of the project rice package is the low level or returns
to labour. According to the preparation mission's own calculations.
the net return per labour day for rice is E. 6.06 in comparison with
E. 9.72 for green maize, and the net return per hectare for green maize
is almost double that of rice.
Interviews revealed that scheme farmers are not interested in
increasing rice production because they do not feel that the rewards
are proportionate to the amount of work. This attitude is more prevalent
on the older, established rice schemes than on Mswati, where rice was
grown for the first time last year.
In part, this change in attitude may reflect declining productivity.
Yields were good on all three rice schemes during the first season.
With each successive year they have declined. Yields on the new scheme
(Mswati) were double those on Mgubudla. Yields on the oldest scheme
(Mkovu) were too low to cover input costs ii 1980-81. Thirty-four
farmers dropped out and the remaining fifteen have switched to green
maize. The women interviewed on Mgubudla expressed a desire to do the
same or drop out altogether.
- 17 -
One reason for low returns is the high cost of land preparation.
The Taiwanese rent power tillers to scheme farmers at E. 12 per hour.
Farmers must operate the equipment themselves and provide their own
fuel. This appears excessive in view of the fact that RDA tractor
hire costs E. 16 per hour, including fuel and driver.
The project will provide additional power tillers to the Taiwanese-
supervised extension centres, for rental to farmers "at rates that cover
replacement cost after five years". An indication that present rates
are exorbitant is that ten hours of rental by each of the fifteen plot
holders on Mgubudla scheme would cover the entire cost of the smaller
Another disincentive is the low selling price. Currently most of
the rice produced in Northern RDA is sold unpolished to the parastatal
corporation, Tibiyo, through the Taiwanese. Tibiyo pays 31 cents for
a kilo of unpolished rice. After polishing, the same rice is sold to
the consumer for E. 1,29. Because the rice hulling machine provided
by the Taiwanese breaks 45% of the rice, Tibiyo will only accept
Rice is not a staple in the Swazi diet and although scheme farmers
eat it once a week on the average, they prefer maize meal. Alternative
market outlets are limited. Nevertheless, women have used the Taiwanese
equipment to polish small quantities of rice for sale in Piggs Peak,
where a bag of unbroken rice brings E. 1 per kilo. At fifty cents a kilo,
even the broken rice brings more than farmers receive from Tibiyo.
Dependency of scheme farmers on the Taiwanese for access to the
power tiller, rice polishing equipment and marketing outlets for rice
could be reduced through provision of loans to Farmers' Associations
for the purchase of power tillers and inexpensive rice hulling equipment.
Even with a 45% breakage rate, farmers' gross returns would be two and
a half times greater than they are at present. This would greatly
increase the incentive to grow rice. Although the project would provide
E. 362,700 worth of equipment (tractors, power tillers and vehicles)
to the Taiwanese-supervised extension stations, there is no mention of
rice hulling equipment.
7. Is the lack of a market the main constraint for expanding
irrigated vegetable production, or is it the price?
The project preparation team argues that scheme farmers would grow
more vegetables if they were assured a market. Farmers, on the other
hand, expressed a willingness to grow more only if the price is right.
Right now, vegetable growers are vulnerable to exploitation by buyers
because their produce is perishable and they have no way of transporting
it to market. They must wait for buyers to come to the scheme. Produce
is marketed individually and producers have little bargaining power.
Each product has a different set of buyers. For tomatoes there
are two categories of large scale buyers: Indian traders from Durban,
who are looking for fresh produce for South African markets, and the
Swazi Central Cooperative Union, which purchases Roma tomatoes at a
fixed price and sells them on contract to a cannery in South Africa.
In addition there are small scale traders from Mbabane and market women
who buy as much as they can carry by bus and sell it in rural markets
and in Piggs Peak.
- 18 -
The project would provide vehicles and assistance to the Central
Cooperative Union to enable them to buy more tomatoes. Scheme farmers,
on the other hand, prefer to sell to Indian traders because they can
get a higher price for their produce. Sale of tomatoes to CCU is seen
as a last resort because of the low price, unreliable collection and
long delays in paying for the produce. If the CCU price remains low,
provision of a market-in itself-is not likely to stimulate expansion
III. THE SOCIAL SOUNDNESS OF THE RAINFED MAIZE PACKAGE
The project's rainfed maize component would provide seasonal
production loans, extension advice and marketing support to 3800
Swazi Nation Farmers scattered throughout the entire country (or 7%
of all Swazi Nation Homesteads). The loans, which would be made
available to individual farmers through the Swaziland Development and
Savings Bank, would consist of purchase orders for tractor hire,
fertilizer, hybrid seed and insecticides. Cattle would be required
as collateral. Interest rates would be increased from the current
subsidized rate (7%) to prime rate (14% at the time of writing).
Because the inefficiency of existing RDA tractor hire services
has been a constraint for increased production, the project would also
provide a limited number of loans to private individuals for purchase
of tractors. The cost of tractor hire would be increased, in all
RDA's, from the current subsidized rate of E. 16 per hour to the
"break even rate", estimated to be about twice the current rate.
Recommendations for hybrid maize production in the four ecological
zones-Highveld, Middleveld, Lowveld and Lubombo--would be provided tp
the agricultural extension service through linkage with the USAID
Cropping Systems Research and Extension project.
The Central Cooperative Union (CCU) would receive support for its
input supply and marketing operations. Maize transit stores would be
constructed near junctions on the roads between Rural Development Areas
and central markets.
The assumptions of the project preparation team. regarding the
relationship between inputs, yields. output and farm income are the
with animal traction, a 45% increase in expenditure for
production inputs is expected to result in a 113% increase
in yields and output and a 123% increase in net income
with tractor ploughing, a 61% increase in expenditure for
inputs will result in a 60% increase in yields and output
and a 62% increase in hrm income.
All of the data used to verify these assumptions was drawn from
existing studies readily available in Swaziland.
- 19 -
1. Who is responsible for maize production on Swazi Nation homesteads?
All Swazi Nation Land is formally owned by the king, who holds
it in trust for the entire nation. Its use is governed by customary
tenure. Only upon marriage does a man receive land, in virtue of
having a wife to cultivate it for him. The size of the plot depends
on the family labour force and the number of mouths to feed. The more
wives, the more plots and the greater the complexity of homestead
In roughly 75% of Swazi homesteads. there is only one household
unit, which may consist of a husband and wife and their children. and
sometimes the husband's mother or unmarried sister or brother. The
entire group cultivates a single set of fields and consumes its meals
In the remaining 25% of Swazi homesteads, there are two or more
separate households, each with its own fields, granary and kitchen.
Each household grows its own food, using family labour, stores its
harvest separately and consumes its own maize. In addition to the
maize fields belonging to each individual household, certain fields--
referred to as "grandmother's fields"--belong to the homestead as a
whole. They are cultivated collectively by all of the households.
Their produce belongs to the homestead head, who keeps it in a separate
granary. It is used as a buffer stock to supplement the stocks of
any household which fails to produce enough for its consumption needs.
According to the traditional division of labour. men are
responsible for land clearance and ploughing and women for sowing.
weeding, harvesting and post-harvest processing. However, as children
spend more time and school and more men are working for wages off the
farm, women's contribution to production is increasing.
A nationwide survey on the role of women in Swaziland found
that the person primarily responsible for agricultural tasks is'
the following (Ministry of Health/UNICEF 1979):
TYPE OF ACTIVITY PERSON RESPONSIBLE
WIFE HUSBAND CHILDREN OTHERS
Land preparation 34.7% 54.6% 9.2% 1.5%
Fertilization 39.7 47.2 10.6 2.5
Ploughing 24.4 61.9 12.7 1.0
Planting 52.7 35.8 7.8 3.7
Hoeing 88.5 1.9 3.8 5.8
Weeding 91.0 0.0 3.8 5.2
Harvesting 92.4 1.3 0.4 5.9
Sorting/storing 88.7 6.0 1.0 4.3
Food preservation 96.4 0.0 0.0 3.6
These findings are corroborated by farm management surveys by the
RDA Monitoring and Evaluation Unit. In Northern RDA. females between
the ages of 15-64 years provide 55% of the labour inputs for local
varieties of maize and 49% of the labour for hybrid maize. Adult males
provide 31% of the labour for local maize and 36% for hybrid maize.
Boys and girls aged 10-14 provide 7.2% of the labour for local maize
and 5% for hybrid maize. Adult women also supply 88% of the labour for
sorghum, 40% for beans and 35% for the main cash crop-tobacco (RDA/MEU 1982).
- 20 -
2. Is increased access to production inputs likely to increase output?
The reports of the RDA Monitoring and Evaluation Unit are unanimous
in suggesting that input use has exceeded all targets: 85% of all RDA
households use tractor hire services; 70-80fo use hybrid maize. Yet
there is no evidence that output has increased accordingly. On the
contrary, maize production has stagnated.
In the search for an explanation, the Monitoring and Evaluation
Unit focuses on two factors-rural exodus and price policy. Although
male migration is a very old phenomenon, dating from early colonial
times, since the start of the RDA programme in 1970, 50% of the rural
labour force has left subsistence farming for other work. By now, 80%
of all rural homesteads have one or more members working off the farm.
Rural exodus has been aggravated by low agricultural prices,
which favour urban consumers to the detriment of traditional producers.
Agriculture cannot compete with off-farm employment. Farmers have
little incentive to grow crops for sale when they can earn cash more
quickly in towns or in the South African mines.
Formerly, male migration was seasonal. At ploughing time men
would return to the homestead to assist with farming. Today, increased
competition for jobs has increased the risk of leaving one's work even
temporarily. A survey of 207 migrant workers revealed that only 31%
return home to help with ploughing; 26% hired a tractor instead, but
the remaining 43% made no contributionito ploughing at home (De Vletter
As a result of death, divorce and desertion, 21% of all Swazi
homesteads are headed de jure by women. An additional 21% have no
resident adult male; these are headed de facto by women. Within the
project area, the incidence of female-headed households may be even
higher Sixty-three precent of the women interviewed in one survey
in Northern RDA were living and farming alone: 23% were widows, 30%
were wives of migrants and 10% were separated or unmarried (Andrehn 1977).
Although ploughing is traditionally men's responsibility, these women
had primary responsibility for ploughing in 59% of the cases.
The outflow of male labour from traditional agriculture explains
the high demand for RDA tractor hire services. At present, existing
services are unable to meet this demand. Inadequate land preparation
and late ploughing in the absence of men were identified by the M & E
unit as the largest single cause of low yields.
The project preparation team addresses this problem by providing
loans to private tractor operators. At the same time. it recommends
lifting the existing subsidy. This would double the cost of tractor
Increased availability of tractor hire services at times of peak
demand would clearly have a positive impact on production. But removal
of the subsidy would have the opposite effect. Already tractor hire is
the largest single expense faced by female farmers, ranging from 60-80%
of total production costs (Black-Michaud and Simelane,1982). If the
cost of tractor hire were to double, their production costs would
increase disproportionately in comparison with other farmers. Some
would be forced to reduce the area under maize to economize on tractor
hire expenses. Others will stop growing maize altogether, if off-farm
earnings permit them to buy as much as they need.
- 21 -
The main contribution of the RDA programme has been to maintain
existing levels of production in the face of a net outflow of 50% of
the labour force from farming. This has been achieved by substituting
one factor of production (capital) for another (male labour).
Farmers' use of tractor hire services has no effect on yields.
Use of hybrid seed and fertilizer, on the other hand, has raised
yields, but this has been offset by a reduction in the area under
cultivation. Neither has increased output.
Findings therefore do not support project planners' assumption
that use of modern inputs will automatically increase the total supply
of maize. On the contrary, they show that the relationship between
input use, productivity and output is not linear.
3. Will credit facilitate production of a marketable surplus?
The unstated assumption of the project preparation team is
that Swazi Nation Farmers need credit. Without it they cannot afford
to use modern production inputs, and without inputs, they cannot grow
a marketable surplus.
Findings from existing studies suggest that this is simply not
true. First, seasonal production loans are valued as a means of
financing inputs, but these inputs are used to substitute for absent
male labour rather than to increase the marketable surplus. Forty
percent of the households receiving seasonal production loans sell
no maize whatsoever. They repay their loans out of earnings from
off-farm wage emp6lyment.
Second, credit has been more attractive to households producing
a maize deficit than those producing a maize surplus. Households
which are primarily interested in growing maize for a profit prefer to
pay for their own inputs rather than borrow from the bank. Their
reason is that the borrowers are forced to sell their maize immediately
after the harvest when the price is lowest, in order to pay back their
loans. Those who do not borrow can make a profit, by selling late in
the season when prices are higher.(Sibisi 1981).
Third, eighty percent of Swazi Nation Homesteads already use
modern inputs. Adoption of hybrid maize, fertilizer and tractor
ploughing preceded and far outstripped the uptake of agricultural
The conclusion of a SDSB economist is that if agricultural loans
were not available, families would still purchase the same quantity of
inputs, paying for them out of wage earnings or savings from the previous
season. Rather than increasing input use, seasonal loans simply
facilitate the cash flow of rural households (Mercey 1982).
4. If farmers grow additional maize, will they eat it or sell it?
When the objective is to increase the supply of maize available
for sale to urban consumers, it is important to know whether farmers
grow enough to cover their own consumption needs. The project
preparation team notes that there is little evidence of severe under-
nutrition in Swaziland and surmises that current production is
sufficient to cover family consumption needs. Only in the poorest
- 22 -
geographic area does the farm model make allowance for families to
consume half of the incremental production.
Evidence from other sources suggest that the deficit is more
serious. A survey conducted by the FAO grain storage project revealed
that 6nly 20-30% of the homesteads in the Lowveld, 40% in the Lubombo
and 50-75% ir the Highveld and Middleveld produce enough maize to cover
their consumption needs for half a year (De Lima 1982).
Throughout the country, according to another source, 48% of all
homesteads supplement home production with purchases (De Vletter 1981).
Estimates of the proportion of households producing a surplus
often rest on the questionable assumption that all those selling maize
are surplus producers. On the contrary, most of the households selling
maize sell only a few bags and many of these have to purchase maize
later in the season. Many are forced to sell even when production is
below family consumption requirements, to repay loans and get cash for
children's school fees. In Northern RDA, 69% of the households
selling maize and 68% of those which do not sell experienced a seasonal
shortage (Andrehn 1977).
Farmers will therefore retain more maize for home consumption
and less will be marketed. This will reduce the project's impact on
5. Is lack of a market the main constraint for production?
There is abundant evidence that lack of a reliable market for maize
has been a disincentive to production. The average Swazi Nation
Homestead sells under ten bags of maize, but the Swazi Milling Company,
which is the sole buyer, only collects quantities in excess of thirty
bags. Furthermore. the company has been known to reject whole lots
because of weevil infestation. Farmers complain that the maize would
not have become weevil infested if the company had come to pick it up
when promised rather than a month later. Therefore, farmers welcome
the project's emphasis on marketing support through the Central
However, the homesteads most likely to produce a maize surplus
are not interested in selling through CCU because of the low price.
Those who can afford to purchase a pick-up truck (and most of the
Swazi Nation Homesteads dedicated to farming as a commercial enterprise
fall into this category) prefer to market their produce privately
rather than through the formal channels provided by the government.
A survey of forty surplus-producing homesteads found that only
two sell to the Swazi Milling Company, which buys maize at E. 10.26
per 50 kilo bag. Instead, they use their own transport to find better
markets. Local buyers pay E. 15 a bag, while buyers in the Lowveld
pay up to E. 18 late in the season (Sibisi 1981).
Provision of a market, in itself, is not likely to increase maize
production as much as a change in the price.
- 23 -
6. Will female farmers have access to agricultural extension?
On the RDA irrigation schemes, women's access to extension is
good. because all plot holders. male and female, come together in the
same place. It is prefectly acceptable for a male extension worker to
explain production techniques to a group of women in a public place.
Maize, on the other hand. is grown by individual farm families on
widely scattered homesteads. This introduces a bias against women. Not
only are individual farm visits time consuming, but it is less acceptable
for a male extension worker to visit a woman at the homestead when her
husband is not around. If she wants assistance, the burden is on her
to go to the extension centre. Since 42% of all homesteads have no
resident adult male and in another 20f1 tho head is gone during the day,
a large share of the target group is affected. For this reason the
feasibility of organizing farmers into mixed-sex groups for extension
delivery should be explored.
7. How will women's access affect the project's social impact?
According to a series of studies conducted by the ILO World
Employment Programme in the late 1970's, there is little indication
that male migration is a class-linked phenomenon associated with
poverty. On the contrary, households with migrants were found to have
more of everything: more education, more cash, more members of working
age and more likelihood of producing a maize surplus. Migrants are
mainly young men who have not yet accumulated enough cattle to marry
and set up an independent homestead. Poverty and wealth were interpreted
as cyclical rather than permanent. Wealth and influence build up over
the life cycle, only to decline when a man's sons set up their own
Women-headed households, on the other hand, fall at the two
extremes. Widows are found at the upper end of the age continuum
when the family's fortune has begun to decline, while wives of migrants
are found at the lower end of the age continuum, when the homestead has
few assets. Exclusion of women would skew benefits away from
The project's social impact will be a reflection of contradictory
forces. The package is attractive to maize deficit households and those
with a shortage of family labour, including widows and wives of migrants.
It is not very attractive to maize surplus households. But the categories
which would benefit most (i.e. widows and wives of migrants) cannot
qualify for loans without cattle.
An added complication is that loans will serve to hire tractors
to substitute for absent male labour rather than to increase input use.
It will facilitate the transfer df resources out of traditional agri-
culture and into the modern sector of the economy without affecting
maize output. This will benefit the economy as a whole, but it will
put a heavier burden on women.
- 24 -
A thorough understanding of male migration, the role of women
and producer incentives is crucial for project success. Bbth components
are weakened by project planners' incomplete understanding of these
1. The irrigated vegetable and rice package
Irrigated vegetable and rice production on RDA irrigation
schemes is primarily a female income-generating activity. Intensification
of production is expected to generate employment for men, but husbands
are not willing to give up their jobs to participate in irrigated
farming. The burden of increasing production will therefore fall on
All scheme farmers are expected to take out seasonal loans, but
collateral requirements discriminate against women, who are two thirds
of the plot owners.
The expected increase in labour inputs during the wet season
(90W) is unrealistic in view of women's responsibility for rainfed maize
production and domestic tasks.
The incentives for growing tomatoes, cabbage and rice during
the summer are less than anticipated. Farmers prefer green maize
because it requires less labour and returns are better than for rice.
The incentive to increase vegetable production is further reduced
by low farmgate prices, due to farmers' lack of bargaining power in
the face of buyers and recurrent gluts.
The incentive to increase rice production is low because of time
constraints, the low selling price for unpolished rice in comparison
with polished rice, and the lack of a local market.
Restricted access of female plot owners to loans, coupled with
overestimation of labour availability and incentives, can be expected
to lower the rate of farmer participation. The cost per beneficiary
will increase and the impact on output will be reduced.
2. The rainfed maize package
The expectation that provision of seasonal production loans for
the purchase of inputs, together with extension advice and marketing
support will stimulate the production of a marketable surplus is
unrealistic in the light of the experience of the RDA programme.
Increased input use has not led to increased maize output. but
merely to maintenance of existing levels of output in the face of an
outflow of 50% of the male labour force from farming.
The popularity of RDA tractor hire services is explained by
women's need to prepare land in the absence of men, who were traditionally
responsible for ploughing. Tractor use does not improve yields; it
merely substitutes one factor of production (capital) for another
Increases in productivity per hectare have been offset by a
reduction in the area planted. This in turn can be traced to labour
constraints, the high cost of tractor hire and difficulty obtaining
- 25 -
ploughing services at the critical time. Removal of the tractor
hire subsidy will hurt women, by increasing their production costs
disproportionately compared with other farmers. The area planted to
maize will be reduced if tractor hire rates double.
Farmers use seasonal production loans to finance the substitution
of absent male labour, rather than to increase the marketable surplus.
Forty percent of the households using loans sold no maize at all and
repaid loans out of off-farm earnings.
Surplus-producting homesteads have no interest in seasonal loans
or CCU marketing support. Project benefits will flow to maize
deficit households, but these will consume most of the incremental
production. The incentive to grow maize for the market is further
reduced by the low market price.
The income-distributional impact of the maize component is
likely to be positive, but the impact on the supply of maize for
urban consumers will be lower than anticipated.
V. DESIGN ALTERNATIVES WHICH WOULD STRENGTHEN THE PROJECT
The project's probability of success could be greatly increased
by introducing a series of minor modifications in the design of the
two components and in the policies of supporting institutions.
1. Measures to ensure women equitable access to credit
For the irrigated vegetable and rice component, the practice of
negotiating loans simultaneously for all members of the Farmers' Associ-
ation should be adopted on all schemes. Since this will not be possible
for maize loans, credit officers should be instructed to make out loans
directly in the name of whoever-man or woman-is responsible for
production. The possibility of adopting group liability in lieu of
cattle as security for loans should be explored.
2. Improving RDA tractor hire service without removing the subsidy
Given that the cost of tractor hire is already affecting the area
under cultivation,-It is in the interest of the economy as a whole to
maintain the current subsidy. The bottleneck of timely ploughing
during the peak season could be addressed in a more cost-effective way
by paying overtime to present RDA tractor operators during the critical
period than by providing loans to private tractor hire operators for
the purchase of additional tractors.
3. Revising rainy season vegetable and rice packages
In view of the heavy demands which rainfed maize production makes
on women's time, the wet season crop mix on irrigated plots needs to
be reconsidered. Green maize is more attractive to farmers because
it is less demanding, pays more per manday, has lower risks and a more
- 26 -
4. Providing loans for rice hulling equipment
Provision of inexpensive rice hulling equipment directly to
the scheme Farmers' Associations would allow farmers to process
their own rice to cash in on the higher price on local markets.
Gross returns would increase by 250%, greatly increasing the incer-`vo
to grow rice. Reduction of farmers' dependency on the parastatal,
Tibiyo, as the sole outlet for their produce is desirable.
5. Strengthening the bargaining power of vegetable growers
Incentives for increased vegetable production could be increased
by organizing scheme farmers to market their produce collectively,
to prevent buyers from playing one off against another. Farmers
would benefit from negotiating stable contracts with Indian vegetable
buyers from Durban, who offer higher prices for fresh produce than
the cannery in Malelane. Provision of transport facilities to farmers'
associations on the schemes would also increase their bargaining
power. Provision of cold storage facilities for perishable produce
near the irrigation schemes would further increase their bargaining
6. Maize prices
Direct price incentives may be a more cost-effective way of
increasing the marketable surplus of maize available for purchase in
urban areas than the alternatives considered by the project (credit,
input supply, extension and market support).
7. Increasing women's representation oD the irrigation Scheme Committees
Although over two-thirds of the plot owners on irrigation schemes
are women, they are underrepresented on irrigation scheme committees.
Some form of quota could be introduced to give women a greater voice
in project decision making.
8. Introducing labour saving technologies for women
Labour constraints could be eased by introducing appropriate
household technology to lighten women's work burden in domestic tasks,
thereby releasing time for greater participation in irrigated agriculture.
The "village technology" component of the UN "Income Generating Skills
for Women" project in Northern RDA has already developed expertise in
The most popular item with farm women is a device which collects
rainwater from roofs an: stores it in cement water jars, which can be
constructed on the farm with little cost. Since women on the schemes
spend two hours a day fetching water, the installation of water-
collecting devices could represent a considerable time savings, especially
during the rainy season when women's work load is heaviest. Prototypes
of maize hullers and grinders have also been developed.
Such technology could be introduced by linking the Credit and
Marketing Project with the Income-Generating Skills for Women Project.
The latter project's revolving fund could provide small loans to cover
the cost of the water-collecting devices.
- 29 -
APPENDIX: METHODOLOGICAL NOTE ON SOCIAL ANALYSIS
In their paper on "Human Factors in Project Work", Perrett and
Lethem of the World Bank make the following distinctions regarding
the use of social techniques at different stages of the project oyclet
Social assessment or diagnosis: consists of the collection,
organization and interpretation of social information about
the project population needed for refining the project concept
and designing and implementing the project.
Social pre-feasibility studies: build directly on any
existing social assessment and are intended to help in
elaborating the project concept and verifying its social
acceptability in a preliminary manner.
Social feasibility analysis: consists of an assessment of the
compatibility between the social and behavioral assumptions
or requirements built into the proposed project design and the
project populations' likely behaviour and response. It is
action oriented and feeds directly and explicitly into project
Social feasibility appraisals focuses on appraising the
compatibility of the project (i.e. its concept, physical
design and technology, implementation strategy, organization,
management and staffing) with the project populations, their
resources and social environment.
Impact prediction: involves social forecasting of the numbers,
and sometimes kinds of people who will end up benefitting
from a project or one of its component parts.
Social designs includes two types of activities-adapting the
project's physical characteristics and implementation strategies
to the sooio-cultural context in order to make them more
acceptable to the beneficiaries and proposing new elements or
activities to improve the project populations' response so
that it is closer to what is needed for successful implementation.
Impact measurement: systematically measures project progress
and impact as a part of the design of the project monitoring
and evaluation system.
The place of each technique in the project cycle is illustrated
in the diagram on page ii. The techniques which correspond to the
project preparation stage are "social feasibility analysis" and "impact
prediction". They are expected to lead to "social design". The
technique which corresponds to the appraisal stage is "social feasibility
appraisal". The present case study combines all four.techniques.
It focuses on testing assumptions about farmer behaviour ("social
feasibility analysis") and the distribution of project benefits ("impact
prediction"). It criticizes the FAO preparation team's farm models and
rates of return ("social feasibility appraisal"), and it suggests
modifications which would improve farmer response ("social design").
SOCIAL INPUTS INTO THE PROJECT CYCLE
OOLNTRY ECONOMIC WORK
COUNTRYY SECTOR WORK
+I~~ ~ -r i- :-
- 31 -
Ideally. verification of assumptions about relationships between
incentives. cropping patterns, labour allocation, output and returns
should be an integral part of project preparation. The sooio-eoonomist
should work closely with other members of the preparation team.
especially the economist and the agronomist, to ensure that technical
packages and farm models are realistic from the farmers' point of view.
Unfortunately, this did not happen with the Credit and Marketing
Project for Smallholders. Thus the only alternative was to criticize
the FAO preparation team's assumptions after the fact, to illustrate
to trainees how recognition of women's role would have led to different
Diagram 2 (page iv) outlines the steps used in preparing the
case study. Given (from mission reports) a project population and a
preliminary design, the first step was to make the social and behavioral
expectation of the project explicit so they could be verified. The
next step was to identify the characteristics which would determine
farmers' response and assess the compatibility of project design with
the intended beneficiaries. Design alternatives which would improve
the project's social impact are identified as a result of the process.
Gender-specific social soundness analysis
"Gender-specific social soundness analysis" is a newly-coined
term, which refers to analysis of sex-linked differences in the control
of land, labour and capital within the household. It is a sub-category
of "social soundness" or "social feasibility" analysis.
The first example of gender-specific social soundness analysis
was Ingrid Palmer's "The Nemow Case". "NEMOW", which is "WOMEN" spelled
backwards, is a case study designed for use in training courses for
project planners.. It is an ex-post evaluation of a hypothetical agri-
cultural project's impact on women. As part of the training exercise
the project's negative outcome is traced to planners' failure to
recognize the role of women.
The technique of ex-post analysis of the implications of rural
women's role for the success of actual field projects has since been
extended to cases in the Gambia (Dey 1981) and Cameroon (Jones 1982).
It has also inspired three sets of case studies for project planners,
published by FAO (Cartoni 1984), the Population Council and USAID
The application of gender-specific social soundness anaissia to
project design documents was illustrated for an agricultural project in
Nigeria (Burfisher and Horenstein 1982). A first attempt at synthesis
was made in a paper presented at the International Agricultural Economics
Meetings in Jakarta in August 1982 (Cloud and Overholt 1982).
The basic insight underlying gender-specific social soundness
analysis lies in its criticism of the conventional household concept
used by agricultural economists in their project-related work. Project
failure is traced to inappropriate assumptions about the organization'
of production and consumption within rural households.
SOCIAL FhAlillT IiY ANALYSIS ANDI SOCIAL. UESI( IN FRLoJcr R ltO
J)AaaSsS ttho. CharA.cer- (6) vicew bou. dne.u of
Ib .ta of the Project R cololne datilonl
PopulatollIts whul h will II II rl.u of Otllen
IetCermiIe Ithrlr leap une iltojct' 1i nli
rt the Expct atlonr Criteria
ot the IDeigners
I PUT U.L__ T__T O. T
Pujecl .an Expectations or the blllcien beteem no~llicatlotte (or. Eiucatlon (or (cthe, ialtrlb,,tlon of and Ieprooeent
(1) Ixistin Frelia.- ) Kke LplcitL the () (8) Idetity -.t -1 1
'ary Den of the Soc-l and Behvl (4)Idefy Incmpi- Alternatives or -1ir /on14i.at-- H- nd 1 Iu k
ldentlllctlato oei Project (or, of the Project Uelolge ond In tCertali Sltoatl*"o SI.llar) .1c iltlic iie.nflti unttder the ..l Ltkely BI~rl-I
So lal nd beltvIlortial
Problem Area in
------- PROCES S
Perrett and Lethem, Human Faotors in Project .Jork, a orld 3ank Staff Working Paper, No. 397 (June 1930).
- 33 -
In their project work, agricultural economists treat the
household as if:
its members pool land, labour and income in a single
decisions are made collectively,
male and female labour are fully substitutable, and
increases in household income automatically benefit women
The contribution of gender-specific social soundness analysis has
been to reveal the limitations of this approach.
Dey's study of farming systems in the Gambia shows that the
household is not always the relevant unit of production and consumption.
Some crops are grown by the whole compound. some by small groups
belonging to the same hearth and others by individual men and women.
There are many separate enterprises rather than one common enterprise
and wives are farmers in their own right, with full decision making
power over their own crops,(Dey 1981).
Because of the sex-typing of tasks, men do not do women's work
and vice versa. It is misleading to male and female labour in terms
of "man equivalents" because husband and wife do not pool labour
and substitutability is limited. When a project increases labour
for weeding and harvesting more than it does for land preparation,
the additional work burden falls more on women than men. Disaggragation
of labour requirements by gender gives a more realistic picture of
actual labour availability.(Burfisher and Horenstein 1982).
Women's incentives are greater when they control the harvested
product than when they are expected to provide free labour for their
husband's crops. In Cameroon, farmers' wives who did not control the
produce from irrigated plots neglected these in favour of rainfed
plots off the schemes (Jones 1982). Projects have failed because
project planners assumed that women would automatically provide
additional labour for crops controlled by their husbands.
In some parts of Africa sole responsibility for providing the
family food falls on women-4not because the husband is absent but
because tradition assigns this responsibility to her and other
responsibilities to the husband. Elsewhere men provide the staple
grains and women the complementary foods. Where the husband is not
expected to spend his income on food, concentration of cash earnings
in male hands could harm nutrition.
There is no indication that increases in the husband's income
are translated into a better standard of living for women and children.
Women need independent access to credit and extension for their own
farming enterprises, and they need an income of their own to enable
them to continue providing for their children in the customary way.
At the time when it was prepared, the case study on the Credit
and Marketing Project for Smallholders was the first to examine
gender-related issues within the process of project preparation.
- 34 -
Use of rapid rural appraisal techniques in project preparation
Project planners' need for simple, rapid and cost-effective ways
of testing their assumptions about farmers was recently addressed by
a conference on "rapid rural appraisal" held at the Institute of
Development Studies, Sussex University in 19$1. Its purpose was tn
identify short cut methods for generating reliable information about
farming systems in developing countries within the time limits typical
of project preparation and evaluation missions. Experienced sociologists,
anthropologists and economists were invited to share the tricks of
Among them was Ingrid Palmer, who contributed a paper on short-
cut ways of predicting the impact of agricultural projects on women.
Her conclusion was very simple: the best way for project planners to
find out how a project is likely to affect women is to ask them directly
(Palmer 1982). Unfortunately, this begs the question.
The problem, from project planners' point of view, is to know
how to separate questions which are essential from others which are
not essential, to learn how to ask them and to analyze the answers in
a way which is useful for improving project design.
The strategy adopted in preparing the case study on the Credit
and Marketing Project for Smallholders was to concentrate exclusively
on a single task-testing assumptions about:
who is doing the work in agriculture and whether farmers
are men or women;
whether all producers (men and women, rich and poor) are in an
equal position to participates
whether labour is available for intensification of production
in view of the sex-typing of agricultural tasks;
whether the technical solutions proposed by the project are
appropriate to the needs and resources of the intended target:
whether sufficient incentives exist Sfdr their adoption, and
whether benefits will spread to women and the poor.
Five distinct, but overlapping, operations were undertaken in
preparing the case study. The first step was to identify project
planners' expectations about farmers' resources an' behaviour. This
spelling out the rationale for the project, its objectives
and strategy for achieving each objective;
identifying who (men, women, rich or poor) is expected to
benefit from each component and in what way;
identifying the criteria for participation in each activity
and for receipt of inputs such as credit, training and extension:
i.e. sex, age, marital status, land ownership, education,
geographic location, labour availability, etc.
- 35 -
The final step in the desk review of project documents was to
examine documentation used to support planners' assumptions about
farmer characteristics and response and identify data gaps and
information needed to fill them. The checklist developed appears
in Appendix II.
This was followed by ailsystematic search for relevant information
from existing sources, including academic studies, farm management
surveys, farming systems research, monitoring and evaluation surveys
and studies by national women's units.
Preparation for field work involved deciding what information
was missing which could not be done without, where to go to gather it,
with whom, and how to go about it. Because there was no time for a
formal survey, heavy reliance was put on group interviews. The first
group meeting was held with village level extension- officers. They,
in turn, each organized a group meeting with farmers on their irrigation
Interviews with individual farmers were limited to no more than
4-6 per scheme. Instead of selecting them randomly, purposive sampling
was introduced. Certain categories of women of particular interest
were singled out (widows, married women with migrant husbands, women
in polygamous marriages, and women whose husbands were engaged full
time in farming). Field Officers were told beforehand what categories
of farmers we were looking for and they arranged for us to interview
them. The point of purposive sampling was to shed light on character-
istics which would determine farmers' response to the project.
Group meetings provided the general picture and individual
interviews permitted us to explore differences between homesteads
having different characteristics. A prototype interview schedule is
included in Appendix III.
Remaining time in the country was spent analyzing the findings
and discussing their implications with those responsible for implementing
- 37 -
NOTES FOR TRAINERS
Investment projects often fail because farmers do not respond
in the expected way. These failures should be avoided. The premise
underlying the case study is that it is possible to predict farmer
response in terms of three variables:
access to project assistance,
ability to intensify production, and
Two techniques are illustrated by the case study: social soundness
analysis and gender-specific social soundness analysis. The
"Methodological Note on Social Analysis" in the Appendix describes
them in detail.
The starting point is to make explicit the behavioral expectations
embedded in the project concept: who is supposed to participate (i.e.
men, women, wealthy, poor) and what are farmers expected to do (e.g.
plant new crops, take out loans, intensify weeding, use more fertilizer,
or market through the project). Once they are explicit they can be
verified. The exercise should be completed by trainees.
The next step is to outline a strategy for testing their soundness.
Preferably, the exercise should involve data gathering in the field.
Estimating farmer response
Access to project inputs and services is determined by the
outreach of delivery systems for credit and extension and by eligibility
criteria. The requirement that farmers must pledge cattle as collateral
for loans discriminates against women and households owning to cattle.
As an exercise, trainees could be asked to estimate how many farmers
could qualify for loans if all women and all households without cattle
Ability to participate is one level beyond access. Even if all
farmers were given access to credit and extension, not all of them
would have the necessary endowment of land, labour and capital to
increase production. Farm family labour is the limiting factor in most
Labour availability is influenced by the sexual division of
labour and by male migration. Women's labour availability is also
limited by their responsibility for domestic tasks.
As a training exercise, group members could be requested to
identify the tasks performed by men and women. The next step would be
to prepare disaggregated totals for male and female labour over the
calendar year "without" the project. Then to estimate how the project
would affect the work load of men and women.
Using data on male migration and household composition, trainees
should then compare the labour requirements with the actual supply and
discuss their findings.
- 38 -
Incentives to behave as expected by project planners are the
third level of analysis. Even if a farm family has the right resource
endowment to enable it to meet the additional labour requirements,
this does not necessarily mean that it is in their interest to invest
their resources in agriculture.
One way to find out whether farmers are willing to behave as
expected by project planners is to design a survey which asks them
whether they would be interested in planting tomatoes instead of green
maize, or in doubling their labour inputs or selling through the
Central Cooperative Union. If time is too short for a full-fledged
survey, Rapid Rural Appraisal techniques can be introduced (see the
Trainees should prepare an interview schedule for group meetings
with farmers and a more detailed schedule for selected interviews.
The rationale for using purposive sampling to get a feel for the
reactions of different categories of farmers should be thoroughly
After farmer interviews have been completed, attention should
focus on the application of economic concepts to incentives. Incentives
are determined by the direct returns to individual household members
from allocating their land, labour and capital in one activity in
comparison with other options. In Swaziland men and women have a
different off-farm earning capacity and this affects their incentives
to invest time and money in farming.
As an exercise, trainees could compare the per manday returns
from irrigated farming, "with" and "without" the project, with the
daily earnings of men and women in off-farm activities, to determine
whether both men and women or only women will be interested.
The control of crops and earnings from their sale is an important
determinant of incentives. If the wife does all the work, but cash
payment for rice is made to the head of household, the woman's incentive
is less than if the payment is made directly to her. Women may prefer
to market produce outside project channels for this very reason.
The case study contains three different examples of the effect
of price on producer incentives. The lack of incentive to grow
vegetables was traced to low farmgate prices. Perishability, gluts
and lack of collective bargaining were its cause. The lack of incentive
to grow rice was traced to circumstances which forced them to sell
their rice unprocessed to the parastatal. Low maize prices relative
to men's off-farm earning capacity caused rural exodus.
A household's incentive to grow a basic staple like maize for
its own consumption is influenced by the difference between the cost
of growing it and the cost of buying it on the market, while its
incentive to grow additional maize for sale is determined by the farmgate
price. In Swaziland it makes good sense to grow maize for home
consumption but little sense to grow it for the market unless you own
a truck and can sell it privately at higher prices.
Trainees could be asked to calculate how much a typical family
would save by growing their own maize and to predict how they would
respond if tractor hire costs were doubled or the maize price were to
be increased by a given percentage.
- 39 -
The final estimate of the rate of participation can be calculated
by subtracting first those excluded from access to project inputs and
services, then those unable to provide the necessary incremental labour
and finally those unwilling to participate due to lack of incentives
from the total number of farmers in the project area. This should be
done to compare the response to two alternative project designs.
The role of women
The case study shows that gender is an important factor
influencing access to project inputs and services (e.g. loans and
extension support), ability to comply with labour requirements (due
to the sex-typing of tasks) and producer incentives.
Female household heads are not the only women who need access
to loans in their own right. Married women have separate enterprises
from their husbands. They pay for inputs out of their own pockets.
Project planners should always examine whether loan eligibility criteria
or extension delivery systems restrict women's access.
Labour ayhilability is influenced by the sex-typing of tasks
and by male migration. Migration is a response to low producer
incentives in comparison with off-farm earnings.
Women's incentives to invest time and resources in farming-
differfrom men's. Women's incentives also differ depending on who
controls the crop and the earnings from its sale.
The applicability of these principles is not limited to Swaziland.
Women's importance for agriculture is not determined merely by the
high rate of male migration. Throughout Africa south of the Sahara,
married women are farmers in their own right having full decision-making
power over specific crops. Trainees in other countries should be
encouraged to make the same set of calculations using information from
their own area.
Although Swaziland is one of the African countries where migration
of off-farm employment has had the greatest impact on traditional
farming, practically all African countries face similar problems to
a greater'or lesser degree. Absence of young men affects the supply
of labour for a wide range of typically male tasks (brushing and felling,
digging irrigation channels, planting tree crops, soil conservation).
Off-farm earnings depress producer incentives. Trainees can be asked
to provide examples from their own country.
The types of analysis developed for Swaziland can be extended to
other countries, where low producer prices, the role of parastatals
as a marketing channel, lack of bargaining power and off-farm earnings
have a similar impact on farmers' incentives. The close linkages
between incentives, male migration and the role of women should also
be illustrated for other countries.
- 40 -
PROTOTYPE INTERVIEW SCHEDULE
After an initial group meeting with women on each irrigation scheme,
the team interviewed a small number of women who were purposely selected
to represent different categories of female farmers: widows, wives of
migrant workers, wives of full-time farmers, women whose husbands work
off the farm but live at home; women in monogamous unions, women in
polygamous unions. Interviews were used to get a rapid idea of how
marital status. type of union and the husband's absence or presence
affect women's role in production.
1. Respondent's characteristics
Name, age, marital status, number of wives in union, husband's
occupation, husband's place of work, husband's residence,
frequency of return
2. Homestead composition and labour force
Total number of people living at the homestead
Number eating from the same cooking pot
Number of cooking pots
Number of children aged 5-12 (girls, boys); in school?
Number of children aged 13-19(male. female); in school? working?
Number of adults aged 20-49 (male, female): working off the farm?
Number of adults 50 and over (male, female): working off the farm?
Total farm labour force available full time: (male. female)
Total farm labour force available part time: (male. female)
Number of persons absent from homestead: (male. female)
Reason for absence (work, schooling, other)
Frequency of return
3. Farm characteristics
Size of holding (at the homestead. on the irrigation scheme)
Who is the plot owner (self, spouse, other)?
Crops grown at the homestead
Crops grown on the irrigated plot (summer, winter)
4. Time use
What do you do in the morning before coming to the scheme?
(in the peak season? in the slack season?)
What time do you get to the scheme? (summer, winter)
What time do you leave? (summer, winter)
How long does it take you to walk to and from the scheme?
Do you do this every day of the week?
What is the busiest time of the year for women? men? children?
(crop, tasks performed by each)
5. Land preparation (irrigation schemes)
How do you prepare your plot for planting ?
If you use a tractor, where do you hire it?
How much do you pay?
Where do you get the money for tractor hire?
What problems do you have with land preparation?
6. Farm management (irrigated plot)
Who decides what and when to plant?
- 41 -
Who is responsible for planting (rice, vegetables)?
Who else helps?
Do you use any hired labour?
Where do you get the seed?
How much do you pay for it?
Where do you get the money?
Who is responsible for irrigation?
How frequently do you get water and for how long?
Who determines the schedule?
Who opens/closes the irrigation channels?
How is the Scheme Committee elected?
How many women are members?
What problems do you have with irrigation?
How often do you weed? (rice, vegetables)
When do you weed?
Who does the weeding?
Who else helps?
What problems do you have with weeding?
9. Fertilizer use
Do you use fertilizer on your rice? vegetables? (type)
Where do you buy it?
How much did you pay for it?
Where did you get the money?
Who applies it?
When do you apply it and how often?
10. Pest control
Do you use insecticide?
Where did you buy it?
How much did you pay for it?
Where did you get the money?
Who applied it?
Where did you get the sprayer?
How much did you pay to rent it?
(for rice): Who is responsible for bird scaring?
Who else helps?
What problems do you have with pests?
11. Harvesting and post-harvest processing
Who is responsible for harvesting? (rice, vegetables)
Who else helps?
Do you ever use hired labour? If yes, when?
Who pays for hired labour?
Where do you get the money?
Who is responsible for drying rice? Threshing?
Who carries the crop to the drying floor?
- 42 -
Who does the rice/vegetables belong to?
Who sells it?
How much did you sell last year? (specify crop)
Here did you sell it?
Who bought it? (specify crop by crop)
How much did they pay?
What problems do you have selling your produce?
How much (total) did you make from each crop last year?
Was it a good year? (why or why not)
13. Control of income/expenditure
What did you do with the money you made?
Did you show it to your husband?
What did you spend it on?
Who decides how to spend it?
What is your husband's attitude toward your vegetable/rice growing?
14. Other income sources
What other sources of cash income do you personally have?
What did you do for cash before you started growing vegetables/rice?
Do you still engage in those activities? (give reason)
What other sources of cash income does your family have?
Do you receive cash remittances from absent family members?
If so. how much and how often?
About what share of the family cash comes from your own vegetable/
rice growing activities?
15. Other farm activities
Who is responsible for maize production at the homestead? Who
Who decides how much to plant and when?
Who supplies the seed?
Do you use a tractor? If so, who pays for it?
Where do you get the money for tractor hire?
Do you use fertilizer?
If so, who pays for it?
Where do you get the money for fertilizer?
Does irrigated agriculture ever conflict with maize farming?
If so, at what times of year specify crop, task)?
If there is not time for both, which crop gets priority?
Has maize production ever suffered? (explain)
Has rice or vegetable production ever suffered? (explain)
How has irrigated agriculture affected the time available for
your children and other family responsibilities?
How has rice growing/vegetable growing affected your health?
16. Interest in increasing production
W would you like to grow more vegetables? Rice? (specify reason)
Would you grow more if you had a market?
Would you grow more if you could get a better price?
Would you have enough time?
Could you afford to pay for inputs if the cost doubled?
Would you like to grow more maize?
Does your homestead need to buy maize?
Does it sell any maize?
Would your husband give up off-farm work to spend more time in
- 43 -
17. Willingness to change crop mix
What crops do you grow in the summer rainy season?
Would you be willing to switch to tomatoes, cabbage and vegetables?
Have you ever tried to grow these crops in the summer?
If so, what happened?
Could you manage to spend twice as much time on your summer
vegetable/rice crops as you do right now? (explain)
Have you ever applied for an agricultural loan?
If so, where did you apply? (SDSB, CCU, etc.)
Did you actually receive a loan?
If so, specify for what crop and amount
Did you repay it on time?
Did you ever borrow money from other sources (traders, etc.)?
If so, what were the terms? Do you still owe money?
If you have never taken out a loan, would you like to?
Why or why not?
oes your homestead have cattle?
would your husband agree to pledge his cattle to secure your loan?
19. Extension contact and group participation
Where do you get your information about how to grow vegetables/rice?
About maize and other crops at the homestead?
How many times has the extension officer visited your plot on the
schemes? How many times has he visited the homestead?
Are you a member of the scheme farmers' association?
In your husband a member?
Are you a member of the scheme committee? Is your husband?
When a meeting is held, which one of you attends?
Does he tell you what goes on at meetings?
If there is a cooperative on your scheme, are you a member?
Is your husband?
DO .OT REOVE STAPLE
RESERVE DLE3 Oiii;GAL