That women throughout Africa tend to provide more
labor for agriculture than men—and almost always provide
more total labor—has implications for technology
adoption. Even if productivity is increased, women may
not be able to increase the number of hours that they
spend working. Simply comparing hours worked also does
not capture the issues of the type of work being done and
the energy expended. The value of time will vary across
seasons and tasks, thus people will be interested in saving
the time that is the most costly (Levi 1987). However, to
the extent that the tasks vary by gender and the value of
women’s time is lower, farmers may be more inclined to
adopt technologies that save men’s time.
Changes in the division of labor
Although tradition often specifies some tasks or crops as
women’s and some as men’s, these may change over time.
As early as 1928, Baumann noted that as the opportunities
for men to work outside of agriculture increased, women’s
involvement in agriculture also increased (Baumann 1928).
With new opportunities arising, gender divisions of labor
in many places are becoming less rigid (Kranz and Fiege
1983; Alwang and Siegel 1994; Donhauser et al. 1994;
Saito 1994). Many changes in the gender division of tasks
are related to increasing outmigration of men from
agricultural communities, as they seek higher earnings
elsewhere. As men leave the area, women take over many
of the traditionally male tasks. Joluo women in Kenya
reported that at the time they were interviewed, they
performed more tasks that were considered traditionally
“male” than when they were first married and moved into
the area (Pala 1983). In addition, when men move into
nonfarm activities, women may become more involved in
cash cropping.
Although the gender division of labor may be changing,
it does not appear that men are taking over women’s
agricultural activities, specifically, the production of food
for home consumption. When men move into activities
that are traditionally women’s, they usually are not
substituting their labor for their wives’ labor within the
household. Rather, usually some new opportunity has
arisen and activities that had been considered women’s
activities have become more productive or profitable. For
example, in Burkina Faso, women traditionally picked shea
nuts. Now that the sale of these nuts is profitable, men are
becoming involved in this activity, often with the
assistance of their wives (Zuidberg 1994).
The adoption of technology may also shift the gender
division of labor. For example, in Tanzania, one study finds
that men are becoming more involved in agriculture as use
of the plow becomes more widespread and as maize,
especially hybrid maize, is grown (Holmboe-Ottesen and
Wandel 1991). Similarly, the adoption of maize
technologies in Zambia has affected gender labor patterns
(Celis and Holleman 1991). In households that adopt new
technologies, men work more on crops and animals and
less on nonfarm tasks, while women spend less time on
crops and more time on post-harvest activities. Children
shift from tending crops to tending animals. The adoption
of technology increases the use of both household and
hired labor in all months.
Many studies seem to suggest that as technologies and
the social and economic environment change, the burden
on women increases (Kranz and Fiege 1983; Berio 1984;
Suda 1996). Even with good intentions, many of the
projects designed to decrease women’s workloads may not
produce the intended effects. The introduction of maize
mills is one such example. In Burkina Faso, one study
finds that as small commercial corn mills became available,
the time women spent hand-milling grain did not
decrease. Instead, they purchased commercially-milled
grain to supplement hand-milled grain only on occasions
when they otherwise would have gone without the meal
(McSweeney 1979). On the other hand, in Gambia, cereal
mills may positively impact women’s welfare by reducing
their workloads, saving both time and, more importantly,
their energy (Barrett and Browne 1994). The introduction
of mills, however, may have little effect if women do not
have the cash to use them (Freudenberger 1994). Similarly,
programs that require women’s labor inputs may not
succeed if the benefits do not accrue to the women
themselves. There is some question about how changes in
the division of labor affect the well-being of individual
household members. As women’s labor burdens and
responsibilities increase, their control over their labor and