Conference Paper No. 9
Maize in Eastern and Southern Africa:
“Seeds” of Success in Retrospect*
by Melinda Smale and T.S. Jayne

Paper presented at the NEPAD/IGAD regional conference
“Agricultural Successes in the Greater Horn of Africa”
Nairobi
November 22-25, 2004

*This paper has been previously published as EPTD Discussion Paper No. 97 by the International Food Policy
Research Institute.

ABSTRACT
This synthesis revisits the “maize success story” in Sub-Saharan Africa, drawing
selectively from an extensive published literature about maize seed technical change and
related policies. The review focuses on the countries of Kenya, Zimbabwe, Zambia and
Malawi, where maize is most important in the food economy, and refers to the period when
maize became a dominant food croup through the 1990s.
The term “success” is equivocal in this case, both because of the difficulty of
establishing the appropriate counterfactual and because some of the policies that contributed
to success inone period later led to decline. While the “seeds” themselves were the result of
innovative, successful maize breeding, boom periods in maize production were episodic and
the public investments in the controlled markets that bolstered them were not fiscally
sustainable.
Since maize will remain a crucial part of the food security equation even while the
agricultural economies of the region diversify, continued investments in both maize research
and market institutions, some of which must be public, are essential. The most vital question,
however, is where the domestic political pressure to support these investments will originate
– an issue related to governance.

ii

ACKNOWLEDGEMENTS
Significant contributions acknowledged from staff of the national Ministries of
Agriculture in case study countries, CIMMYT, IFPRI, Rockefeller Foundation, Egerton
University/Tegemeo Institute, and Michigan State University’s activities in Eastern and
Southern Africa, including the Food Security Cooperative Agreement. This paper has also
been published by IFPRI as EPTD Discussion Paper No.97.

iii

TABLE OF CONTENTS

1. Background ................................................................................................................. 1
2. Maize becomes a dominant foodcrop (1900-1965) .................................................... 4
3. Expansion of smallholder maize (1965-1900) ............................................................ 8
4. The period of uncertainty (1990s) ............................................................................. 13
5. Outcomes .................................................................................................................. 19
6. Summary ................................................................................................................... 23
7. References ................................................................................................................. 30

iv

MAIZE IN EASTERN AND SOUTHERN AFRICA:
“Seeds” of Success in Retrospect
M. Smale and T.S. Jayne*
“The next stage in fact may well be to change maize in Kenya from a peasant
subsistence food crop to one of the major raw materials of a country poor in many other
raw materials and mineral resources. Land could be released for other cash crops in
rotation with grazed pasture, leading to increased soil fertility instead of exhaustion
from maize monoculture. New secondary industries could be generated to stimulate the
country’s economy. In fact, it is possible to envisage maize improvement causing an
agricultural revolution as happened in the USA Corn Belt after the introduction of
hybrid maize...”
M.N. Harrison, Chief maize breeder, Kenya (1970: 54)

“Inability to pay for seed and inability to sell product in a stable (and honest) market is
the chief impediment to adoption of improved (and properly adapted) varieties, either
private or public, in many developing countries….in Africa I fear that what was
happening yesterday may not be a good predictor of what is happening today.”
D.H. Duvick, (previously) Pioneer maize breeder, pers. comm.(March 13, 2002)

I.
1.1.

BACKGROUND

Definition of success

Though maize research and development was the most frequently
cited success story among experts surveyed by Gabre-Madhin and Haggblade (2001), even a
casual review of past work reveals that the term “success” is in this case equivocal. The story of
maize seed research and development in Sub-Saharan Africa has been described as a “maize
revolution” (Eicher and Byerlee 1997), a “qualified success story” (Eicher 1995), a “stop-and-go
revolution” (Howard and Mungoma 1996), a “delayed green revolution” (Smale 1995), an
“obscured revolution” (Gilbert et al. 1993), a “failure” (Kydd 1989), and a “cause of peasant
impoverishment” (Page and Chonyera 1994). Each appellation refers to a specific period in
time, location, and definition of success. Defining success is therefore fundamental.
There is little doubt that maize seed technical change has occurred in Sub-Saharan
Africa. African smallholders have readily used well-adapted, modern maize types1 in a number
of locations and time periods, as documented historically by maize researchers and social
scientists (the most recent broad compilation by Byerlee and Eicher 1997). Though percentages
vary by year, in the late 1990s, an estimated 47 percent of all maize area in Sub-Saharan Africa
and 58 percent in Eastern and Southern Africa was planted to modern maize (Morris 2001).
Modern maize is estimated to represent more than three-quarters of maize area in Kenya,
1

In this chapter, modern maize types refer to cultivars generated through adaptive on-station research, including
either improved open-pollinated varieties or hybrids.

1

Zimbabwe, Swaziland, Lesotho, and South Africa. There is no doubt that as the land frontier
diminishes in many parts of the continent, sustainable development in societies that remain
primarily agrarian will require productivity gains rather than area expansion.
In this chapter, while we revisit some of the actors and technical ingredients that
composed research “successes,” we also seek to highlight the significance of the policies that
promoted them. In particular, we ask whether seed technical change and associated policies
enhanced (1) maize productivity; (2) farmer incomes; (3) equity; and (4) sustainability.2 Using
these criteria of “success”, we draw purposively and selectively from a vast published literature
about maize seed technical change, institutional development, and policy change in Sub-Saharan
Africa. Viewing the case from a perspective that spans much of the 20th century, we seek to
identify the factors that contributed to episodes of “success” as well as those that may have
provoked periods of uncertainty and decline.
Defining the appropriate counterfactual for measuring success based on these criteria is
exceedingly difficult, for three major reasons. First, the rate of growth in observed maize yields
understates actual yield gains. Without continued infusions of improved seed, maize yields are
likely to have declined over time due to pests and declining soil fertility, which would have
made it even more difficult to meet food consumption requirements as the population grew
(Gilbert et al. 1993). Second, separating yield changes due to genetic yield potential from those
associated with changes in management practices is a methodological challenge, given their
interactions, especially over larger geographical scales. Some practices, such as plowing in
Zambia, have had negative long-term effects on the soil resource base (Haggblade and Tembo,
2002). Others, such as contour ridging in Malawi (Kettlewell 1965), protected the soil. Third,
the rapid uptake of improved maize seed in Africa has involved a complex interaction between
technological, institutional, and policy innovations. Technical change cannot be analyzed
separately from changes in the institutional and policy environment, and it is nearly “impossible
to separate the relative influences of technology, human capital, and institutional innovations
with much accuracy” (Bonnen 1990: 263).
1.2.

Scope

The documented history of maize in Africa spans the 20th century, but our focus is the
second half, when the establishment of formal planting breeding institutions on the continent
resulted in the systematic release of improved open-pollinated maize varieties (IOPVs) and
maize hybrids. We have defined three key historical phases. During the 1900-1965 period,
maize became the dominant food crop in case study countries. Smallholder maize production
expanded during the second period, after Independence, which occurred in distinct calendar
years and global economic contexts in each case study country. We label as “uncertain” the third
period, from 1990. The past decade has been characterized by a combination of unfavorable
weather conditions, declining public investments in agricultural research, subsidy reductions,
and erratic policies, which appear to have precipitated a decline in maize productivity (national
maize yields and production) in case study countries.
Our case study countries are Kenya, Malawi, Zimbabwe, and Zambia, though we also
identify cross-cutting themes drawn from Nigeria, Ghana, and Ethiopia (Box 1). Selection of
countries reflects the relative importance of maize in the food economy, as measured in terms of
the percent of total cereal area, per capita consumption, extent of direct use as a staple food
(Table 1). Among the largest maize producers in Eastern and Southern Africa, maize occupies
75 percent or more of cereal area only in Kenya, Malawi, Zambia and Zimbabwe. The largest
2

We use these terms as defined in chapter 1.

2

maize producers in Western Africa are Nigeria and Ghana, though the average percentage of
maize used for human consumption is considerably lower there than in most of eastern and
southern Africa. The average per capita consumption of maize as food is over 94 kg/yr in
Kenya, over 100 kg/yr in Malawi, Zambia, and Zimbabwe, and a mere fraction of that level in
Ghana and Nigeria.
1.3.

The Challenge of Maize

There are several properties of maize as a crop species that influence the nature of seed
genetic change and therefore the likelihood of achieving any impacts on the welfare of
smallholder African farmers through yield increases. These are aptly summarized by Morris
(1998 and 2001) and Pandey (1998), on which the following is based.
Maize is predominantly a cross-pollinating rather than a self-pollinating species. When
rates of cross-pollination are high in a crop, genetic material is exchanged when pollen flows
among neighboring plants. Unless carefully controlled, all of the maize plants in a given field
will differ from the preceding generation and from each other. When maize self-fertilizes, the
progeny often have undesirable traits, but when it cross-fertilizes, some demonstrate significant
yield advantages relative to their parents (termed “hybrid vigor”). Maize is the world’s most
widely grown cereal, cultivated across a range of latitudes, altitudes, moisture regimes, slopes
and soil types, with the simplest to the most mechanized production technologies.
High rates of cross-pollination means that the attribution of yield increases in farmers’
fields to specific introductions of improved seed (either of improved open-pollinated varieties or
of hybrids) is more difficult than it is for other major cereals like rice or wheat. The advantages
of F1 seed of maize hybrids can degenerate rapidly when farmers save the seed and replant it,
though evidence suggests that in some cases advanced-generation hybrids significantly
outperform the variety that the farmer was growing previously—depending on the type of hybrid
and the control that serves as the basis for comparison (Morris, Risopolous and Beck 1999).
Due to market imperfections and cash constraints, African smallholders often “recycle” F1 seed.
Hybrid vigor in maize means three very important things for African smallholder
farmers. First, it means that to sustain the yield increases they seek, they are reliant on a seed
industry in a way that neither the rice nor wheat farmers of Asia’s green revolution ever have
been. On the other hand, a hybrid-based maize sector also requires large-scale commercial seed
enterprises, whose profits can be sustained only by strong seasonal demand by farmers for
renewing their seed (Tripp 2001). Furthermore, temperate maize germplasm is not easily adapted
to the non-temperate environments of the developing world—so that the gains achieved by
private companies in the U.S, Europe, and some parts of China are not easily transferred to many
of the smallholder farmers who produce in the wide range of microclimates and technologies
found in the developing world (Morris 2001).

3

II.

MAIZE BECOMES A DOMINANT FOODCROP (1900 - 19653)

There is no evidence of maize cultivation in Africa until the 16th century (Miracle 1966),
when it was introduced from the Americas to Africa along the western and eastern coasts,
gradually moving inward as a ration with the slave trade.
Before 1965, the rise of maize production in all of the case study countries was propelled
to a greater or lesser extent by five driving factors: (1) the agronomic suitability of maize (2) the
British starch market; (3) milling technology; (4) the integration of Africans into the settler wage
economy; and (5) market and trade policies promoted by settler farm lobbies.
2.1 Agronomic Suitability
Millet and sorghum were the staple foods of Africans at the beginning of the century.
However, maize was regarded as “eminently suitable” (Anthony 1988) for mixed farming by
pioneers because 1) it required less capital investment and technical skill than did cotton and
tobacco (and could therefore be produced by newly arrived novices), and 2) it gave higher
returns to land than other indigenous cereals under reasonably favorable conditions, though not
throughout the entire range of ecological conditions (Weinmann 1972; Harrison 1970: 22).
Maize’s transition to a major crop in Kenya occurred during World War I, when disease in millet
led to famine and millet seed was consumed rather than planted. In Zimbabwe, white settlers
began producing maize as early as the 1890s (Byerlee and Heisey 1997). Howard (1994) reports
that maize was a staple in Northern Rhodesia’s Luapula Province by the end of the 1700s,
though it is fairly certain that “there are no large areas where maize was more than a minor food
at the beginning of this century” (Miracle 1966: 156). Maize did not dominate in Malawi until
well into this century (Williamson 1956).
Long term weather patterns and soils maps suggest that much of Malawi’s growing
environment has long been favorable for rainfed maize (Heisey and Smale 1995, p.5), as are a
number of environments in Zambia, Kenya, and Zimbabwe. Returns to labor are likely to have
been higher for maize. Maize is protected from bird damage by its leafy covering, while the
exposed grain of sorghum and millet requires labor time for scaring birds. When off-farm wage
employment became important for Africans, these agronomic features of maize provided further
advantages compared to small grains (Low 1986).
2.2

The British Starch Market

Until the 1920s, export volumes exceeded human consumption as the major destination
of the maize produced by settlers in Kenya and Zimbabwe. Ironically, the preferences of today’s
African consumers for white as opposed to yellow grain color began with the influence of the
British starch market during these years.
As early as 1911, increasing demand and price premia were evident for white maize in
the British starch market, apparently because North American producers of yellow maize had a
decisive transportation cost advantage in supplying Britain (Masters 1994). Since the British
starch market provided a premium for white maize, local legislation was passed in some parts of
3

Prior to the transition to majority rule in 1966 and 1980, respectively, Zambia and Zimbabwe were known as
Northern and Southern Rhodesia, respectively. Malawi was known as Nyasaland. For simplicity, this chapter refers
to these areas by their current names.

4

the region requiring that only white maize be accepted for export. Though both white and yellow
maize varieties of maize were grown, settler farmers were informed by the Secretary of the
London Corn Exchange that exports required better grading and uniformity (Weinmann, 1972,
pp. 19-20). Farmers discovered that when yellow and white maize were grown in close
proximity and cross-pollinated, the grains of the progeny was mixed in color, rendering it
unsuitable for export. The Rhodesian Maize Authority passed a resolution in 1923 stating that
the introduction of yellow maize varieties into the territory posed “a vital danger to the maize
growing industry.” In Zimbabwe, the Maize Act (1925 to 1970), enabled growers to petition the
Government to restrict the growing of maize in their area to a specific variety and color (Jayne et
al 1995).
Emphasis was also given to the soft dent-type maize favored by the British starch market,
as this was easier to process and less injurious to industrial roller mills (Kydd 1989). By 1920,
both smallholder and commercial farmers in Zambia and Zimbabwe had largely replaced their
white flint cultivars with improved white dents, though these may also have been higher-yielding
(Jayne et al 1995).
2.3.

Integration of Africans into settler wage economy

The grafting of mines, plantations, and cattle enterprises onto the local economy
expanded the demand for food in the country. Eventually the domestic demand for maize grew
as Africans left their farms to work on settler farms, in mines or industrial plants, particularly in
Kenya, Zambia and Zimbabwe (Mosley 1983; Jansen 1977). Food consumption preferences
were influenced by the rations that employers used as in-kind payments. Diets adapted in a selfgenerative process, as “people got used to what they consumed"(Shopo, 1985).
By the 1920s, maize accounted for approximately 80 and 60 percent of the settler
cropped area in Zimbabwe and Zambia, respectively, and provided in-kind payments for the
rapidly increasing African labor force. By contrast, Malawi had the smallest extent of land in the
former British Central Africa but the largest concentration of peoples (Pachai 1973). An exporter
of African labor to the mines of the Rand and farms of Zambia and Zimbabwe (Tindall, 1968),
Malawi had few areas with elevations high enough to attract a European population easily
decimated by malaria (in contrast to Zimbabwe) and no rich mineral deposits (in contrast to
Zambia). Consequently, Malawi never developed an organized settler constituency, a distinction
that has led to a different trajectory in maize research and maize consumption preferences.
2.4.

Settler Lobbies, Market Control, and Trade

The threat of competition from African farmers led colonial administrations to design
and implement of controlled marketing systems from the 1930s. Evidence from Kenya, Zambia,
and Zimbabwe indicates that African farmers were capable of generating maize surpluses at
prices below the production costs for most settler farms, and that settler farmers successfully
lobbied colonial legislatures for protection (Jayne et al 1995). Initially, in both Kenya and
Zimbabwe, settler consumer interest groups such as animal feeders, mines, and plantation
farmers strongly opposed protection for settler maize farmers on the grounds that it would raise
the cost of maize (Keyter 1975; Mosley 1975). Over time, however, settler maize producers’
interests were increasingly well represented in the colonial legislatures. Catalyzed by slumping
world agricultural markets during the 1930s depression, the colonial governments passed the
Maize Control Acts of the 1930s in Zimbabwe and Zambia, and the Native Produce Ordinance
in Kenya in 1935.
5

These marketing acts (1) created state crop-buying stations in European farming areas
without parallel investments in African farming areas; (2) enforced a two-tiered pricing scheme
with higher prices for settler farmers than for native Africans; and (3) established restrictions on
grain movement from African areas to towns, mines and other demand centers. From 1935, the
combination of maize legislation, land evictions, and fiscal policies weakened Africans' position
in food marketing relative to that of settler farmers in Kenya, Zambia, and Zimbabwe (see Figure
1 for evidence in Zimbabwe; see also Mosley 1983; Jayne et al 1995).
The rise of centralized, state maize marketing boards also encouraged the development of largescale, concentrated grain milling industries, using roller mill technology first employed on a
large scale in Kenya, South Africa, and Zimbabwe in the 1950s. Controls on private maize
movement provided the licensed roller millers (that produced a refined and more expensive type
of meal) with a de facto monopoly on maize meal sales to cities and grain-deficit rural areas
once local supplies were exhausted (Jayne et al 1995).
This system of market regulation was effective in achieving its principal objectives.
Settler maize production expanded and producers earned prices that generally exceeded export
parity. Kenya and Zimbabwe (except during World War II) remained reliably self-sufficient in
maize. Through discriminatory pricing made feasible by controls on marketing, the cost of
supporting settler maize production was paid largely by African farmers and consumers rather
than European taxpayers, making the system fiscally sustainable (Jayne and Jones 1997).
Opposition to agricultural price supports was accommodated by selective consumer subsidies.
The stability of the policy and pricing environment (and the limited competition faced by
industrial mills) contributed to rapid growth in commercial agriculture, the demand for maize
research and later, the adoption by commercial farmers of hybrid maize varieties developed by
national research systems.
2.5.

Milling technology

African maize production received a boost with the introduction of the hammer mill in
the 1920s (Shopo 1985). Hammer mills gave a processing cost advantage to maize over small
grains, since maize could be dumped into the hopper for grinding, while millet and sorghum
husks required de-hulling first.
The prevalence of large-scale, industrial processors in Kenya, Zambia and Zimbabwe
contributed to preferences, particularly in urban areas, for dent types. The removal of the germ
and pericarp makes refined meal look whiter, last longer, and taste sweeter than whole meal.
Hammer-milled, whole meal remains the primary staple food in the grain self-sufficient rural
areas of these countries. Though they also consume whole meal in rural areas, Malawi’s
population has long expressed a strong preference for maize porridge made from “ufa woyera”
(white, almost iridescent, flour). The fact that Malawi’s population remained predominantly—
and densely—rural may explain the persistence of other processing methods.4 Maize flour in
rural areas of Malawi is processed with a combination of hammer-mill technology and several
stages of pounding by hand, soaking, washing, and drying (Kydd 1989, Ellis 1962).
2.6.

Early maize research

4

The appropriate technology depends in part on the opportunity cost of women's labor, since grain processing has
been viewed as a woman's task in most of Southern Africa (Bagachwa 1992).

6

The investments by colonial government and settlers themselves in the maize research
that radically transformed maize production in Kenya and Zimbabwe began as early the 1930s,
“soon after the news of the great success of hybrid breeding in the USA began to be widely
known” (Harrison 1970: 28). In 1919, commercial farmers in Zimbabwe founded the Maize
Breeders Association (Rusike 1998). Zimbabwe’s maize breeding program, initiated by H.C.
Arnold in 1932, was the first outside of the U.S. to produce double-cross hybrids for commercial
use, releasing Southern Rhodesia-1 (SR-l) in 1949. The team then turned their attention to
single-cross hybrids, which have a greater yield advantage and are more uniform in the F1
generation, though their seed is more costly to produce. In 1960, they released the first
commercially grown single-cross hybrid in the world, SR-52 (Rusike 1998: 306).5 “The
combination of SR-52 seed, fertilizer and improved agronomic practices increased maize yields
46% over Southern cross, the most common improved local variety” (Eicher 1995). SR-52
diffused rapidly and widely among commercial farmers, becoming one of the most popular
hybrids in the region and a parent of many others.
Before independence, Zambia relied on its Federation partner Zimbabwe for improved
maize germplasm, and the importation and use of SR52 (as well as other hybrids) is said to have
doubled commercial farmer yields from 1949-53 levels of 1.3 t/ha to 2.7 t/ha 1959-63 (Howard
1994). The preferences of Malawi’s smallholder farmers and the superior on-farm processing
rates of Malawi’s flint maize types relative to dents were recognized by Malawi’s first maize
breeder, R. T. Ellis (1954, 1962), who developed several synthetics as well as a flint hybrid
LH11 before his resignation in 1959.
The first scientific maize research program in Kenya began in 1955 in Kitale, the center
of maize production in the White Highlands. The first modern maize type released in Kenya,
Kitale Synthetic II (an IOPV released in 1961), was based on inbreds developed from the Kenya
Flat White complex of farmers’ selections. M.N. Harrison, chief Kenyan maize breeder, felt
“the need to widen the genetic base of the Kitale program,” and while “nothing of value” had
come from earlier testing of U.S. Corn Belt, European, South Africa, Rhodesian, and Australian
materials, “the great diversity of center-of-origin material from similar ecological conditions to
those of East Africa, close to the Equator with a wide range of altitude, had never been
tried”(Harrison 1970: 38). In 1958 he returned from a trip to Mexico and Colombia, funded by
Rockefeller Foundation, with exotic breeding material (Anthony 1988). After screening 124 test
top-crosses of these materials with Kitale Synthetic II, the outstanding result was a cross with an
unimproved Ecuadorian landrace (Ecuador 573). Released in 1964, the varietal hybrid (Hybrid
611) made from Kitale Synthetic II and Ecuador 573 was the basis of all hybrids developed since
by the national (Hassan et al. 1998). The yield advantage of this unique varietal hybrid over
Kitale Synthetic II was 40 percent (Gerhart, 1975: p. 5). Varietal hybrids have lower seed costs
than conventional hybrids, with lesser loss of yield advantages when recycled. Released on the
eve of Independence, H611 diffused among large- and small-scale farmers in the high potential
areas of western Kenya “at rates as fast or faster than among farmers in the U.S. Corn Belt
during the 1930s-40s “(Gerhart, 1975: 51).

5

Breeding records indicate that Pioneer first released single crosses as named hybrids in 1961, though small U.S.
companies mostly likely released them before 1960 (Duvick, pers. comm.).

7

III. EXPANSION OF SMALLHOLDER MAIZE FROM 1965 [TO 1990]
By the time of transition to majority rule in Eastern and Southern Africa, maize became
the cornerstone of a “social contract” that the post-independence governments made with the
African majority to redress the neglect of smallholder agriculture during the former colonial
period (Jayne et al 1999). The controlled marketing systems inherited by the new governments
at independence were viewed as the ideal vehicle to implement these objectives
The benefits of controls for settler farmers during the colonial period generated the belief
that the same system could also promote the welfare of millions of smallholders if it were simply
expanded (Jenkins 1999). The social contract meant that governments were responsible for
ensuring food self-sufficiency in white maize at a cheap price. Because the world market for
white maize is thin, domestic production shortfalls often necessitated imports of yellow maize,
an inferior substitute that connotes agricultural policy failure (Jayne and Rukuni 1993).
Each of the case study countries achieved impressive episodes of maize production
growth, driven by interacting innovations in technology, policies, and institutions. We mark
these growth episodes as: 1965-1980 in Kenya; 1970-1989 in Zambia; 1980-89 in Zimbabwe,
and 1983-93 in Malawi. Some of the ingredients of these episodic successes retain continued
importance, while others could not be sustained, and may have contributed to the uncertainty of
the 1990s.
Good germplasm through innovative breeding was the essential ingredient of each
episode of success. Though each has other, unique ingredients, all share: (1) complementary
investments in agronomic research, extension, seed distribution systems, and rural infrastructure;
and (2) institutions to coordinate grain marketing with seed, fertilizer and credit delivery. This
second ingredient, however, was implemented through controlled pricing and marketing systems
that incurred large subsidies and treasury costs, and which eventually contributed to fiscal crisis.
The points reported below synthesize findings from a set of detailed theses and studies analyzing
the causes of seed technical change in smallholder maize production during the relevant time
periods (Gerhart 1975; Karanja 1990; Hassan 1998); Zimbabwe (Rohrbach 1988; Rusike 1995;
Eicher and Kupfuma 1997); Zambia (Howard 1994; Howard and Mungoma 1996); and Malawi
(Smale 1994; Heisey and Smale, 1995).
3.1. Maize Germplasm
There is little doubt about the quality of the germplasm products released by Kenya’s
maize research program in the early years. Research breakthroughs were accomplished first and
foremost because of the capacity and continuity of the Kenya maize program staff, which was
“probably unmatched in any national research program in Africa” (Gerhart 1975), a program
which remained public. From its inception in 1955 to 1973, the maize breeding program had
only two directors, M.N. Harrison and F. Ogada. During those initial years, the Rockefeller
Foundation and USAID facilitated the exchange of germplasm between continents as well as the
sharing of new research experience concerning hybrid genetics. (Gerhart, 1975; Harrison 1970;
Anthony 1988). The four Kenyan maize programs (Kitale, Embu, Katumani, and Mwtapa) have
since released a succession of hybrids and improved open-pollinated varieties, and one of the
greatest achievements has been the release of a range of materials to suit the nation’s diverse
agroclimatic conditions (Hassan 1998; Karanja 1990).
8

Similarly in Zimbabwe, “small teams of highly motivated and well-paid local scientists”
devoted their entire careers to maize research (Eicher and Kupfuma 1997: 35-37), backed by the
revenues earned and contributed by commercial farmers. The second major episode of maize
seed technical change in Zimbabwe occurred following independence in 1980 among
smallholder farmers of the previously “communal” areas, largely based on R200, R201 and
R215. Suitable for production on sandy soils in low rainfall areas, these short-season maize
hybrids were bred by Nelson for settler farmers seeking to diversify from tobacco exports when
the Unilateral Declaration of Independence (UDI) was declared in 1965. Though three-way
crosses are in general relatively uniform and intended for annual purchase and higher levels of
management, they may still perform well relative to unimproved types when grown with fairly
low levels of management. Like SR52, the R-200 series were later popular among smallholders
in neighboring countries. Not until the change to majority rule in 1980, however, did
smallholders acquire widespread access to the germplasm and complementary inputs. Earlier,
smallholder use of modern maize would have occurred spontaneously as a result of native
Africans’ interaction with commercial, settler farmers, but not as part of a systematic, nationwide effort. Estimates in the literature suggest that between 20 and 40% of Zimbabwe’s
smallholders in communal areas grew maize hybrids during the 1970s (Rohrbach 1988;
Kupfuma 1994
The germplasm produced by the Zambian maize program from the mid-1970s through
the early 1990s included an impressive array of 10 double and three-way crosses bred by
Ristanovic, and 2 flint-type, early-maturing, IOPVs bred by Gibson (Howard 1994; Howard and
Mungoma 1996). In all but the most difficult growing environments, the hybrids outyielded local
(and improved) open-pollinated varieties even without fertilizer. Unlike the single cross SR52,
they were double and three-way crosses, so their yield advantages lost with recycling were not so
great. Meanwhile, the very early maturing, drought-tolerant, flinty IOPVs were used as an early
food source as green maize (Howard 1994). As in the Kenya and Malawi cases and to a far lesser
extent the Zimbabwe case the advances depended on access to both international breeding
expertise and germplasm collections.
Malawian smallholders waited much longer for suitable hybrids. One reason was
discontinuity in staffing and funding—accompanied by shifts in emphasis between hybrids and
IOPVs (Kydd, 1989). Another constraint was scientific, since local flint materials were too tall,
regional breeding had focused on dents, and exotic flint germplasm with suitable characteristics
was not easy to identify. Malawi did not have the dense urban populations to feed that
preoccupied Zambian leaders, and neither the settler farmers nor Malawian smallholders had a
political voice (Smale 1995). In 1990, B.T. Zambezi, E.M. Sibale, and W.G. Nhlane released
the national program’s first adapted, semi-flint hybrids, MH17 and MH18. These were topcrosses of Malawian lines and flint populations from CIMMYT. Three aspects of the semi-flint
top-crosses made them a success. First, semi-flint, top-cross hybrids were well suited to
production by smallholders who process and consume their grain on farm and recycle their
hybrid seeds (Zambezi et al. 1997; Benson 1999). Farmer evaluations (Smale et al. 1993) and
experimental results demonstrated that that these semi-flint hybrids could be processed on the
farm like local flint types (Box 2). Second, trial and demonstration results showed that there
were only minor differences in yield among the various Malawi hybrids, so that yield was not
sacrificed for grain texture. Third, analysis of trial results and extensive demonstration data for
three of the major maize producing zones (representing 70% of Malawi’s maize area) during the
1989/90-1991/92 seasons showed that unfertilized hybrid maize yields were higher than those of
9

unfertilized local maize, even during the worst drought year in decades (Heisey and Smale
1995).
3.2 Complementary investments
Harrison (1970) and Gerhart (1975) both attributed successful diffusion of improved
maize seed in Kenya not only to good maize germplasm, but to a sound agronomic research,
effective linkages of research to extension, a strong commercial seed enterprise, and the
coordination of input and maize marketing. The Kenya Seed Company (KSC) was formed by
large-scale farmers in 1956 but entered into an agreement with the Kenya government to produce
and distribute maize seed in 1963 (Karanja 1990), employing stockists who were “small-scale
African storekeepers selected for their location, reputation, and interest.” The approach,
modeled on the marketing of Wilkinson razor blades, was “every stockist an extension agent”
(Gerhart 1975: 9). Credit was used only at the discretion of individual stockists. Like Harrison,
Gerhart (1975) emphasized that the Kenya Maize Research Program had never been isolated
from the other components of a national maize program. These included, as early as the 196070s, a relatively dense transport network, crop research, marketing boards offering guaranteed
prices, and an extension service. In 1965 the extension services reputedly planted 5,000
demonstrations, followed by hundreds of fertilizer demonstrations. Gerhart’s survey revealed
that extension agents were the first source of information about hybrid seed, followed by other
farmers.
The second major phase in Zimbabwe’s maize revolution (Eicher 1995) occurred from
1980 to 1988, following the end of the civil war in 1980. Much of the institutional foundation
that fueled the 1980s smallholder growth episode was inherited from the colonial period. Private
sector maize seed production began in 1940 when settler farmers established the Seed Maize
Association of Southern Rhodesia, which successfully met the needs of its clients (Rusike,
1995). In 1979, the Seed Maize Association was converted to the Seed Maize Cooperative,
administered by the powerful Commercial Farmers Union, and organized as a cooperative in
order to transfer monitoring and inspection costs from the government to members, as well as to
avoid taxes and profit distributions (Rusike, 1995). Seed Co-op was created in 1983, and
continued to enjoy autonomy from the government at the same time that it benefited from tax
breaks, exclusive rights to some seed types, and access to subsidized credit (Rusike, 1998).
When the civil war ended in Zimbabwe, the new government attempted to expand the input
delivery, credit, and marketing programs that had previously served only large-scale producers
(Blackie, 1990; Rohrbach 1988). Rohrbach (1988) attributes the tripling in smallholder maize
production that occurred from 1980 to 1988 to five factors: (1) the ending of the independence
war in 1979; (2) an increase in the use of hybrid maize seeds from about 40% in 1979 to 98% by
1985; (3) an increase in state crop buying stations serving smallholder areas, from five in 1980
to 148 in 1985, thus reducing the costs and risks associated with producing maize for the market;
(4) an eight-fold rise in in-kind credit allocated to smallholders between 1979 and 1986, which
stimulated fertilizer use and maize yields; and (5) an associated response by private input
suppliers to the increased demand for farm inputs.
In addition to maize germplasm, Howard (1994) and Howard and Mungoma (1996)
identified as driving factors Zambia’s land surplus, favorable weather, and the heavily
subsidized state-led system of credit, input, and maize marketing support to smallholders.
SIDA, which funded maize research from the 1980s, also funded the establishment of a seed
industry, including Zamseed and the Seed Control and Certification Institute. Zamseed
10

produced all seeds as a commercial company, though the government and SIDA owned large
shares (Cromwell 1996).
Several structural factors other than germplasm contributed to sharply rising smallholder
adoption rates in Malawi from the mid-1980s until 1993, as indicated by the fact that use of
hybrids with dent grain texture rose in the years preceding the release of the semi-flint, topcrosses. Primary among them was a supply of quality commercial seed, which had been a
binding constraint in the early years of Malawi’s maize breeding program. National Seed
Company of Malawi (NSCM) took responsibility from ADMARC in 1978, and Cargill acquired
most of the equity of NSCM in 1988, with a more aggressive approach to seed production,
procurement and marketing. Price ratios were also favorable for the use of hybrid seed and
fertilizer during the late 1980s in Malawi (Smale and Heisey 1994: 699). Seed prices also dipped
in that period, likely in response to large inventories held by NSCM after years of low use of
hybrid seed.
3.3. Coordinated input and grain markets
The objectives of the newly independent governments (stated above) were addressed in
all four case study countries through similar mechanisms. These mechanisms enhanced the
profitability of growing maize by smallholders, stabilized net returns in a variable rainfed
growing environment, and encouraged maize consumption by urban and rural consumers.
First, the expansion of state marketing infrastructure in smallholder areas facilitated the
disbursement of credit and subsidized inputs to smallholders by state agencies designed to
recoup loans through farmer sales to the marketing boards (Rohrbach 1988; Howard 1994;
Jabara 1984; Kaluwa 1992; Smale 1994). Second, both direct and indirect input and credit
subsidies supported maize production. For example, fertilizer subsidies in Zambia averaged 60
percent of landed cost by 1982—with 90 percent of fertilizer used on maize (Williams and
Allgood 1991). In addition to the direct subsidy, an expanded network of cooperative marketing
depots reduced the transactions costs of selling maize in remote areas. In Malawi, ADMARC
(Agricultural Development and Marketing Corporation) assumed an all-encompassing
responsibility for delivering inputs for the production of the staple food, marketing of maize
output, stabilizing maize prices through maintaining storage facilities, and transporting maize
into food-deficit areas during the hungry season. Third, pan-territorial pricing brought
smallholders in remote areas into production for state markets and assured a national crop but
distorted regional advantages. Finally, in some cases and periods, as in Zimbabwe from 1980
through 1998, maize producer prices were guaranteed by the state at well above export parity
prices.6 At the same time, most governments subsidized the retail price of industrial maize meal
to consumers, thereby raising the demand for domestic production under a policy of maize selfsufficiency.
However, in all four countries, marketing board costs escalated as the scale and
complexity of their activities increased. Losses consisted of two types: those which government
forced on the board by mandating it to carry out activities that were unprofitable but fulfilled
“social” functions like buying maize at above-market prices in remote areas (which encouraged
maize production expansion), and those related to operational inefficiency (which probably had
little effect on smallholder maize production). Pan-territorial pricing was particularly
burdensome in Zambia and Zimbabwe, since it raised the share of grain delivered to the boards
6

This conclusion flows from the GMB’s annual maize trading account statements showing losses on maize exports
to regional neighbors during this period.

11

by smallholders in remote (but often agronomically high-potential) areas where transport costs
were high (Bryceson 1993, GMB 1991). Stockpiling white maize, a consequence of government
preoccupation with maize self-sufficiency, was also costly (Buccola and Sukume 1988; Pinckney
1993). Operational inefficiency varied across countries, though it appears to have been greatest
in Zambia and Kenya (Amani and Maro 1992; Scott 1995; Bates 1989). Allegations of
corruption were widespread, even in Zimbabwe’s relatively efficient Grain Marketing Board
(CSM 1986). In some cases, the treasury costs of state marketing operations were so large that
they affected rates of inflation, interest, and currency exchange, especially in Zambia during the
1980s (Jansen and Muir 1994).7
Other distortions resulted at the farm level. Packages were often of fixed size and input
proportions that were not optimal for all farmers. Farmers had little choice concerning seed
source or type, and grew what was available in that year from a single source, while often seed
was effectively rationed since certain quantities were reserved for credit club members
(Cromwell and Zambezi 1993; Smale et al. 1991). Farmers near urban demand centers who
were implicitly taxed through pan-territorial pricing resorted to parallel markets. Declining
volumes through the state marketing channels further exacerbated the boards’ trading losses. In
turn, the increasing proportion of maize sales by smallholder farmers, generally on poorer land
with less reliable rainfall, increased the instability of marketing board purchases and sales, and
hence of the fiscal demands made by the marketing system. Controls on private grain
movement, imposed by the earlier colonial governments, were continued in all countries even
after independence. These controls also prevented grain from flowing directly from surplus to
deficit rural areas (Jayne and Chisvo, 1991; Odhiambo and Wilcock 1990). Official restrictions
on private trade and weak market infrastructure often made it easier for surplus farmers to sell to
the Boards rather than their deficit neighbors a few kilometers away (Jayne and Chisvo 1991).
Finally, a growing body of empirical evidence pointed to the controlled marketing systems as
suppressing or imposing additional costs on parallel trading and processing channels that often
served the interests of both producers and consumers more effectively than the official state
apparatus (Rubey 1995; Amani and Maro, 1992).
A major difference in implementation of pre- and post-independence maize policy was
the method of supporting food production growth. The colonial governments in Zimbabwe and
Kenya in particular drew the resources to support a relatively small, privileged group of settler
farmers largely from consumers and African farmers, through discriminatory pricing schemes.
By contrast, the post-independence governments drew the resources primarily from the treasury
when they attempted to lavish the same level of support to a much wider African farmer and
consumer constituency. In most cases, the treasury and aid donors bore most of the cost of
expanding marketing services to smallholders, and the system was not fiscally sustainable.8

7

For example, Kenya’s National Cereals and Produce Board accumulated losses equal to 5% of GDP in the 1980s.
And Zambia’s National Agricultural Marketing Board’s operating losses were roughly 17% of total government
budgets in the late 1980s (Howard and Mungoma 1994).
8
During this phase of marketing policy, donors helped cover the operating losses of the state marketing system, as
in Malawi. Donors also provided extensive support to the expansion of government bulk grain handling facilities,
which yielded little direct benefit to smallholders but added substantially to the marketing board costs (Jones 1994).

12

IV.

THE PERIOD OF UNCERTAINTY (1990s)

We have identified four major factors that contributed to lower rates of growth in maize
production and maize yields during the 1990s: (1) weather instability (2) contraction of state
subsidies and market support (3) declining investment in public agricultural research, combined
with uneven progress in liberalizing the maize seed industry, and (4) crisis-motivated maize
policies.
While the first factor constituted an “exogenous shock” for all actors involved, the
remaining three are evidence that governments “reneged” on the social contract of the postindependence period, choosing short-term panaceas over longer-term policy solutions. The
1990s brought much tighter fiscal constraints on government social activities, due largely to the
treasury deficits accumulated during the maize boom periods and to donors’ unwillingness to
continue supporting them. As explained further below, tighter fiscal constraints during the
1990s led to a new series of government and donor programs that still served important political
ends but provided less commitment to the long-term public investments needed to sustain rural
productivity over time.
4.1.

Weather

Maize yields are considerably more variable in Southern Africa, and particularly in
Zimbabwe, than in other parts of the world (Byerlee and Heisey 1996). But weather during the
past decade probably has accentuated yield instability. Meteorological data from 1914/15 to
2000/01 in Zimbabwe indicates that five of seven stations received at least 15% lower
cumulative seasonal rainfall over the last decade than in the previous eight decades. Drought
beset the country in 1992/93, 1994/95, and 1997/98. The early 1990s in Malawi were
characterized by frequent drought or mid-season dry spells, with the 1991/2, 1993/4 and 1994/5
seasons all being poor for maize. In particular, the drought in 1991-92 devastated maize
production in Zambia, Zimbabwe, and Malawi.
4.2.

Contraction of state marketing systems

Fiscal crises and increased donor leverage over policy pushed the grain marketing
systems of Eastern and Southern Africa toward liberalization in the mid-1980s. After first trying
to strengthen the performance of state marketing boards in the 1960s and 1970s, donors and
international lending agencies began promoting the reform of food marketing and pricing as a
central component of structural adjustment programs in Africa. Donors had lost patience with
phased and partial reform programs that were increasingly seen as propping up costly and
otherwise unsustainable pricing and marketing policies rather than facilitating reforms (Jones
1994). Political economy models (e.g., Bates 1981) further suggested that state interventions in
agricultural markets, while ostensibly designed for rural development or to correct for market
failures, were in fact designed to serve the interests of a dominant elite composed of bureaucrats,
urban consumers, and industry.9 The framework of policy-based lending within which market
reforms have occurred in each of these countries has strongly influenced the path of reforms, and
has expanded external leverage over domestic agricultural policy through aid conditionality.
9

Later political economy analyses also highlighted the importance of dominant rural elites (e.g. Toye 1992). These
rural elites have continued to reap income transfer benefits from the structuring of agricultural policies even as state
marketing programs have been contracted (Jayne et al 2002).

13

A major cause of the stagnation in smallholder maize production in Zimbabwe was that
the costly marketing policies – for credit, inputs, and maize – could not be financially sustained
from the tax base. Zimbabwe’s Grain Marketing Board maize operations required government
subsidies of $30-90 million annually in the mid-1980s, roughly five percent of GDP. By 1990,
almost 80 percent of smallholder recipients of state credit were in arrears (Chimedza 1994). The
maintenance of high maize prices to sustain surplus production also put pressure on government
to cushion the impact on consumers by subsidizing the price of maize meal. Unable to tackle its
mounting fiscal crisis, the Government of Zimbabwe reluctantly approached international
lenders for financial support. The terms of these agreements involved a cutback in the activities
that contributed to the fiscal deficit, maize marketing and credit policies being among them. As
a result, Grain Marketing Board buying stations in smallholder areas were reduced. Even though
20 additional depots were established between 1985 and 1991, the number of rural collection
points in smallholder areas declined from 135 in 1985 to 42 in 1989, to 9 in 1991. The real
value of seasonal AFC credit disbursed to smallholders declined steadily since 1986. By 1999,
the state allocated less credit to smallholders than at independence in 1980. Smallholder use of
fertilizer in the 1990s has become more volatile compared to the 1980s, although it has not
declined on average. Yet smallholder production has not since exceeded the levels reached in
1985 (Figure 2).
Kenya’s Cereal Sector Reform Program began in 1988, and entailed a similar cutback in
state grain depots. Because of the small size and dispersion of such maize sales, small traders
tend to be more efficient and cost effective in handling smallholder sales than NCBP, whose
activities are largely concentrated in maize-surplus, large farm areas (Karanja 1990: 34).
Surveys of rural households conducted in 22 districts in 1997 and in 2000 indicate generally
broad support for the liberalization process, except in the several districts where government
policies traditionally have been designed to support maize prices (Jayne et al 2002). The broad
rural support for maize market reform appears to be mainly driven by the majority of rural farm
households who resided in maize deficit areas, who were directly hurt by the former controls that
restricted direct inter-district trade in maize and therefore made them dependent on the more
expensive industrial maize meal through the official marketing system to satisfy their residual
grain consumption requirements (Argwings-Kodhek 1999). Lower risks and costs of maize in
the post-reform period have facilitated shifts in cropping patterns to other crops that provide
higher returns to land and labor. However, the reform process continues in a halting manner.
The Kenyan government has several times reversed its course in the liberalization process, and
has since 1999 resumed limited marketing board purchase of grain in politically important areas
at fixed support prices, coupled with tariffs on maize imports. Policy uncertainty has risen in the
reform period.
In Zambia, reforms did not begin in earnest until 1993 (Howard and Mungoma, 1996).
Price controls were removed on all commodities with the exception of maize meal, and some
restrictions on foreign exchange, imports and exports were lifted. From 1991 to 1995, interest
rates rose from 50 to 120 percent over 4 months. Throughout the 1990s, the volume of fertilizer
distributed on credit through government programs with very low recovery rates declined as
donors withdrew their earlier support. The maize marketing board, NAMBOARD, was
abolished in 1990 and replaced by a combination of private trading networks and a government
organized cooperative system of crop marketing. Neither of these successor marketing networks
had the ability to continue subsidizing maize production in remote areas through pan-territorial
pricing. The private sector has apparently responded vigorously to liberalization in areas of
reasonably well-developed infrastructure, particularly in cotton, according to surveys in districts
14

of Central and Eastern Province (Kahkohnen and Leathers, 1999).10 Yet for maize, the fertilizer
nutrient-to-maize grain price ratios rose sharply after the decline of NAMBOARD, particularly
in the areas that formerly benefited from the transport subsidies inherent in pan-territorial
pricing. Farmers were caught between rising input and financial costs, lower producer prices,
and a contraction of market support in remote areas. As a result, there was a discernible shift in
cultivated area from maize to crops such as cassava, groundnuts, and sweet potato (in the more
remote northern and northwestern areas of the country) and to cotton (in the more
commercialized areas). Maize remains the main food crop in most parts of the country, but its
role as a commercial cash crop has fallen sharply (Figure 3).
Despite liberalization efforts during the late 1980s in Malawi, the ex-parastatal
ADMARC continued to enjoy a monopsonist/monopolist status given that Asian traders were
restricted from selling in rural areas and the weak economies of neighboring countries limited
cross-border trade opportunities (Christiansen and Southworth 1988), and producer-consumer
price margins were often so narrow that it was not worthwhile for large traders to participate
(Kandoole and Msukwa 1992). As part of the donors’ structural adjustment program in the
1980s, fertilizer subsidies were gradually reduced, but when internal strife in Mozambique
blocked the cheapest transport route the subsidy removal plan was abandoned. Subsidies were
again removed after the change in government. In 1995, nitrogen-to-grain price ratios rose by
four times their earlier levels, though the fertilizer subsidy at its highest never exceeded 20% of
the fertilizer cost (Mann 1998). This occurred for several reasons: (1) subsidies were removed;
(2) the devaluation of the Malawi kwacha boosted fertilizer prices disproportionately to maize
prices; (3) world fertilizer prices rose; and (4) private fertilizer dealers were requiring substantial
risk premiums to hold and transport fertilizer in an inflationary economy with uncertain demand
(Conroy 1997; Diagne and Zeller 2001; Benson 1997). Maize prices followed export parity
while fertilizer prices reflected full import costs, even though Malawi often imports maize. Since
most fertilizer in Malawi is used on maize (and the remainder on tobacco), the removal of
implicit subsidies in the form of over-valued exchange rates had a strong negative effect on
fertilizer use. Depreciation of the real exchange rate also raised nitrogen to grain price ratios
because fertilizer is imported (Minot, Kherallah and Berry, 2000; Heisey and Smale 1995).
These factors, along with shifts in relative prices of competing crops, may have contributed to
the moderate decline through the 1990s in the percentage of cropped area devoted to maize by
small-scale farmers (Figure 4).
None of the African governments in this study were able to find an internal solution to
the large deficits to which their agricultural policies in the 1970s and 1980s contributed. Nor
would external donors agree to make loans without redressing the sources of states’ treasury
deficits. Fiscal crises were the driving force behind the acceptance of structural adjustment. Yet
an analysis of the implementation of maize marketing reform over the 1990s concludes that
many states have actually pursued a de facto continuation of former control policies, albeit in a
more contracted way that has left out most smallholder farmers while continuing to provide
benefits to politically important farming interests (Jayne et al., 2002). In each of the four case
study countries, the main state maize marketing agency continues to exist, sometimes in a
reconstituted form, and remains a major player in the market. This contrasts markedly with
10

Since liberalization, the mill-to-retail maize marketing margins within cities has declined, though
farm-to-wholesaler marketing margins between cities has increased. Some of the main causes of higher
margins were inadequate transportation infrastructure high transaction costs caused by weak contract
enforcement, impacted information, and policy uncertainty caused by continued government distribution
programs especially in more recent years (Kahkohnen and Leathers 1999: 21; Govereh et al 2002).

15

cereal market reforms elsewhere in Africa where state grain marketing agencies are now largely
defunct (World Bank 1994).
Moreover, there have been few initiatives to support private trading activity and to
develop the public institutions required for a privatized marketing system to function effectively
(Jayne and Jones 1997; Kherallah et al., 2002). To some extent, the removal of government
controls on grain trading appears to have mitigated the adverse effects of declining state
marketing subsidies associated with structural adjustment. Lower grain processing costs made
possible through liberalization have reduced the wedge in some countries between producer
prices and consumer prices (Rubey 1995; Jayne and Argwing-Kodhek 1997). In some cases,
price spreads between surplus and deficit regions have declined (Nyoro et al 1999). Evidence
indicates that the number of private traders serving smallholders has increased, especially in
high-potential areas, within several years after the initiation of partial food market reform
(Amani and Maro 1992; Kaluwa 1992; Nyoro et al 1999). However, in contrast to other regions
in Africa where liberalization helped to dismantle agricultural taxation, the transition from
controlled to market-oriented systems in Eastern and Southern Africa has reduced state
expenditures on agricultural marketing investments and subsidies. In Zambia, for example,
maize and fertilizer subsidies, in constant 1998 kwacha, declined from an average of 250 million
kwacha per year in the mid to late 1980s to under 50 million kwacha per year since 1993.
4.3. Maize seed industry
The genetic advances needed to sustain or enhance maize yields depend on past
investments in maize research, with lagged effects. In Kenya, public maize research expenditure
declined in real terms from 1980 through 1990. USAID withdrew its technical assistance in
1977, but in 1987, the government asked it to resume support. The scientific and institutional
cooperation that created the maize success story of the 1960s and 1970s collapsed in the 1980s,
as a combination of fiscal crises and donor pressure weakened public financial support for
research, extension, and credit (Hassan and Karanja 1997). In Zimbabwe, Eicher (1995) warned
of declining government commitment to maintaining the continuity of research funding and
scientific leadership, based on a declining real budget for agricultural research from 1987.
Public sector funding of agricultural research declined by 39% in real terms between the periods
1975-1985 and 1990-2000.
Karanja has contended that the yield potential of successive maize seed releases in Kenya
continued to rise but the rate of increase declined (Karanja 1990). In plant breeding, yield
breakthroughs of the type obtained when Kitale Synthetic II was crossed with Ecuador 573 are
periodic rather than routine; “the yield advantages offered by early releases were the cream to be
skimmed off the milk”(Harrison, cited in Gilbert et al. 1992:29). In all case study countries, a
more disturbing problem is that smallholders are likely to be even farther from realizing yield
potential, for reasons not confined to weather. One reason is that the genetic advances offered by
breeding research have not been matched by agronomic practices and efficient support services
for smallholders, many of whom are located in marginal areas (J. Ransom, pers. comm.2002;
Hassan et al. 1998; Rusike and Eicher 1997). As adoption of improved maize moves into more
marginal areas the effects on national yield levels are also numerically marginal (Byerlee and
Heisey 1996). Simply put, “there is little to no advantage of using a hybrid if the yield potential
of your soil is less than 1 ton”(J. Ransom, pers. comm.). The secular decline in soil fertility in
the intensive maize systems of this region has been aggravated by slackened fertilizer use and
the abandonment of traditional methods of soil regeneration as populations rise (Lynam and
Hassan 1998). With the removal of subsidies, high nutrient-grain price ratios, linked to high
16

transportation costs, have eroded the profitability of fertilizer use (Morris, Tripp, and Dankyi,
1999; Heisey and Mwangi; Minot, Kherallah and Berry, 2000). In addition, pest and disease
problems have worsened with intensification. Advances in yield maintenance rather than yield
potential may hold the greatest promise for this region (Lynam and Hassan 1998; De Vries and
Toenniessen 2001), though costs of resistance breeding are relatively high, and yield “savings”
from this type of breeding are not always perceptible to farmers.
At the farm level, seed supply problems have accompanied the patchy, incomplete
process of seed market liberalization. On one hand, it is reasonable to ask why public sector
companies in eastern and southern Africa persist in emphasizing hybrid maize, which could be
provided as well or better by private companies (Tripp 2001). On the other hand, growth in seed
sales in Kenya slackened considerably in the 1980s and has recently declined (Karanja 1996; de
Groote, pers. comm. 2002), apparently provoked by inefficiencies and seed quality problems.
More recent publicly-bred releases have diffused more slowly than did their predecessors
(Karanja 1990; Hassan et al. 1998; Bourdillon et al. 2002). In Zimbabwe, a number of private
companies compete in the seed market, but the government maintains a firm grip on variety
release and seed importation, which can be counterproductive (M.Banziger, pers. comm., and
Tripp (2001). Seed Co-op has remained an important actor due to its research capacity, a
diversified asset base, and collaborative technical agreements with a range of multinational
companies (Rusike 1998). Though private investment may be compensating for declining public
research budget and there is no apparent slowdown in the rate of variety release, companies do
not target the specific needs of marginal areas in Zimbabwe, such as IOPVs and stress tolerance
(M. Banziger, pers. comm.).
Liberalizing the market in Malawi only exposed its deficiencies, especially in the maizesurplus producing areas of the North, where population densities are low. Since the most
common hybrids in Malawi yielded more than local maize without fertilizer at the seed prices
that prevailed through the early 1990s, it made economic sense for farmers to grow hybrids even
if they could not apply fertilizer (Heisey and Smale 1995; Benson 1999). However, the seed-tograin price ratio nearly tripled between 1989 and 1997, and even if farmers could access the cash
to purchase a hectare of hybrid maize, it is unlikely that smallholders in many environments
would have enjoyed the yield advantage required to pay the seed costs (Benson 1999).
Furthermore, official price ratios in no way incorporated the transactions difficulties experienced
by smallholders. Though a number of key steps were taken, in a 1995 report on the retail trade
in agricultural inputs, Tsoka concluded that policy, legislative, and regulatory frameworks were
not conducive for retailers’ participation in the marketing of agricultural inputs. In 1997 as
compared to 1990, farmers purchased their seed from a range of retail sources, though
ADMARC remained the major source of seed (Smale and Phiri 1998). In 1998, Monsanto
bought the controlling share of NSCM. Recent studies in the Central Region suggest that while
the use of agents in maize seed distribution has extended coverage, smallholders pay almost 20
percent more in retail outlets than at the depot (Nakhumwa, 2002). Although there are several
companies operating in Malawi, much of the Monsanto-Malawi seed sales have been marketed
through the supplementary inputs programs (see Box 3), as Cargill-Malawi’s were previously
channeled through the smallholder credit system. All company representatives interviewed by
Nakhumwa (2002) reported that seed sales are constrained by the low disposable income of
farmers, a unpredictable growing environment, and a stagnating market.
4.4.

Crisis-motivated Maize Policies

17

Malawi provides an illustrative example of the erratic policies of the 1990s that have
directly affected the maize germplasm supplied to smallholders, and consequently, yields (Box
3, Figure 5). Disruptive policy initiatives, sometimes pushed by donors, appear to have
exacerbated the deepening poverty that has resulted from failure to redress the long-term
structural problems in the economy (Kydd and Christiansen 1982; Lele 1990; Sahn and
Arulpragrasam 1991) and those factors more specific to the 1990s, described above.
In June of 1993, Malawians voted to change their government to a multiparty system
from an oppressive, single-party dictatorship. Following this historic decision, many other
changes occurred in the organization and delivery of inputs. While it became obvious that the
promise of seed technical change in maize production would require addressing long-term
structural factors —investing in physical infrastructure to reduce transfer costs, strengthening
institutions to mobilize rural financial systems, continued investments in agricultural research
and extension, and a stable agricultural policy environment—much of the cumulative learning
from years of maize research in Malawi appears to have been discarded in favor of short-term,
crisis-motivated solutions. Initiated during a food crisis in 1996-7, though originally conceived
as a technology-based plan that was cheaper than importing maize, the Starter Pack (SP) and
Technology Input Program (TIP) have functioned as a relief effort, since the research input into
the package they extend is minimal. In effect, a farm input subsidy system with controlled maize
prices seems to have been substituted by in-kind transfers with full price liberalization, though
the social welfare implications of the switch are unclear (Levy et al. 2000; Levy and Barahona
2001; Sibale et al. 2001; Box 3). Similar crisis-motivated approaches have substituted for longterm investments in public goods both Zimbabwe and Zambia (Rohrbach et al 2002).

18

V.
5.1.

OUTCOMES

Cumulative adoption

Cumulative adoption rates for modern maize types are high in Kenya and Zimbabwe, and
remain relatively high in Zambia [65%-100%], relative to Malawi where adoption stands at
roughly half that level (43%)] (Table 2). Ranges in the data shown for 1996 reflect differences
between national program and seed sales estimates. As a result of continued infusion of these
materials through farmer purchases and seed distribution schemes, and of the involuntary
mixture of germplasm types through cross-pollination in farmers’ fields, it is difficult to estimate
reliably the areas under improved seed.
5.2.

National maize productivity

During the episodes of success as we have defined them, growth rates in maize yields
were positive in all case study countries and a sizeable component of the positive growth in
maize production was attributable to yield (Table 3). Growth rates in maize yield and production
were negative in the 1990s in each country except Malawi, where a few years of favorable
weather and continued infusions of improved maize seed and fertilizer through free input
distributions seems to have had an impact11. Malawi’s successful period, relatively brief and
delayed, overlaps the 1990s. However, estimated growth rates mean little when coefficients of
variation are so high. Maize yields are most variable in Zimbabwe and considerably more stable
in Kenya than in any of the other countries. Furthermore, changes in actual aggregate maize
yields underestimate the true achievements in maintaining yields against a naturally deteriorating
germplasm and soil resource base (Gilbert et al. 1993). In none of the case study countries has
actual maize production per capita kept pace with population over the past 40 years (FAOStat).
Even so, seed technical change clearly promoted these countries’ ability to feed themselves at a
lower cost to consumers than otherwise would have been the case.
5.3.

Farmer income

Net farm income from using maize hybrids and fertilizer depended not only on stable,
controlled input/output ratios (particularly the nitrogen to grain price ratio), but also on direct
and hidden subsidies that reduced that transactions costs faced by smallholders. The illustrative
partial budgets shown in Table 4 suggest that the returns to land in maize cultivation declined
dramatically between the episode of success and the subsequent period, considering the costs of
credit, seed and fertilizer after the collapse of the smallholder credit scheme, subsidy removal,
and devaluation. After subsidies on seed, fertilizer, and especially those on credit were removed,
and in a year that real nitrogen and seed prices reached their zenith relative to the real output
price, returns to land and labor were in the negative range for fertilized hybrids, and incentives
11

In Malawi, advanced-generation hybrid seed is likely to have gradually replaced local maize
and F1 hybrid seed. Survey estimates suggest that in the major maize-producing zones between
1990 and 1997, local maize as a percent of total estimated maize area declined from 85 to 60
percent, while the percent planted to F1 seed was similar (12 and 10, respectively). From 1990
to 1997, however, only a scant 7% of farmers were able to grow F1 seed in each season (Smale
and Phiri 1998). The estimated area in recycled hybrid seed rose from 2 to 30 percent, however,
with IOPVs still representing only 1 percent. This pattern is likely to have become generalized
for the nation from 1998 to 2001 with the subsequent annual infusions of small quantities of
hybrid seed with Starter Pack and TIP schemes.
19

shifted toward the cultivation of unfertilized local maize. For many smallholders, these figures
in no way capture the sheer physical difficulty of obtaining the inputs when the marketing
system changed.
Static, partial budgets reveal only a snapshot in time, however. Smale et al. (1991), Heisey
and Smale (1995), Benson (1997), and Diagne and Zeller (2001) reported partial budgets and
probability distributions of net returns constructed over a period of a decade in Malawi, with
large numbers of yield observations from farmer surveys, on-farm trials and demonstrations. The
relative profitability of fertilized and unfertilized hybrid seed relative to the local benchmark
varied by the growing season, whether or not subsidies were in effect, and whether the
household was a net buyer or consumer of maize. Detailed analysis of more then 1,900 trials
conducted nationwide in 1995-6 demonstrated that in that season, the most profitable
recommendation for farmers in most areas of Malawi was to apply no fertilizer at all; fertilizer
use at any level only made economic sense when maize was produced for home consumption
(because of wide price bands) and the household had other cash-earning activities to finance its
purchase (Benson 1997). Practically speaking, because maize is a food crop, smallholder credit
has not been available for use in purchasing maize seed and fertilizer since the mid-1990s. In
Malawi at least, instability in input-output price ratios seems also to have wreaked havoc with
smallholder farm incomes.
5.4.

Equity

On one hand, detailed adoption studies in case study countries demonstrated that during
the episodes of success, maize hybrids and fertilizer were adopted by some of the poorest,
smallest maize producers in the world -- turning on its head the stereotype of the commercial
hybrid grower (Gerhart 1975; Howard 1994; Smale 1992; Hassan et al. 1998). For example, in
the major maize-producing regions of Malawi in 1990, an estimated 36 percent of farmers with
an average farm size of 1.2 ha planted F1 maize hybrids. As is commonly demonstrated in
microeconomic models of adoption, the likelihood of using hybrid seed differed by household
assets such as total landholding, human capital variables such as farming experience and past
extension contact, and physical factors such as the agroecological zone. Access to extension and
credit linked to input use and output markets, often through club membership, was a major
explanatory factor. Access to oxen was significant in areas where farmers used animal traction,
such as Zimbabwe and large parts of Zambia (less so in Kenya and Malawi).
In their 1995-1996 survey of 400 households in five districts of Malawi where
microcredit schemes operated after the collapse of the smallholder credit program, Diagne and
Zeller (2001) found that when households chose to borrow they realized lower net crop incomes
than non-borrowers—for several reasons—though most they related to growing hybrid maize
with high rates of fertilizer application. The finding of the greater downside risk (higher
cumulative probabilities of lower net margins in the negative range) of fertilized hybrid maize is
consistent with some of those reported earlier for major maize-producing areas (Smale et al.
1991).
By 1991-92, 60 percent of the total small/medium maize area was planted to Zambian
improved maize types (Howard 1994), with adoption rates differing by agroecological zone.
Adopters were, as expected, those with larger farm sizes, larger household sizes (labor stocks),
and those who used animals or machinery rather than hoes. They had more formal education
and had been visited by extension agents. The majority lived close to service centers or roads,
20

received credit for maize and sold maize output, and were dependent on local rather than
regional depots.
In Zimbabwe, as in the case of Kenya, the largest gains in smallholder maize production
and sales were concentrated among farmers in areas with better rainfall and those owning more
land and livestock. Based on detailed Grain Marketing Board data on maize purchases from
1985-1991, Jayne and Rukuni (1993) report that 10% of the smallholder farmers accounted for
90% of the maize sold during this period to the Grain Marketing Board by smallholder farmers.
Recent analysis using nationally-representative survey data in Zambia (Tembo et al,
forthcoming) has shown that in 1999/2000, 10 percent of farmers defined as small and medium
scale (under 25 hectare farms) accounted for 78 percent and 93 percent of the maize produced
and marketed, respectively, by these sectors. Jayne et al (2002) found similar results in Kenya.
The main attributes distinguishing these farmers is (a) location in “high-potential” areas with
long growing seasons suited to the long-maturing hybrid maize varieties released in the 1970s;
(b) relatively large farms by smallholder standards (in the over-five hectare range).
In Malawi, the greatest problem affecting the use of improved maize seed and fertilizer
since 1993 has not been the input-output ratio or the profitability of the technology (since most
farmers are net consumers) but cash flow and the continued erosion in the effective purchasing
power of rural households with successive devaluations, inflation and other macroeconomic
changes. Until the purchasing power of the 60 percent of households in Malawi that are maizedeficit is raised, “higher maize prices translate into increased misery” (Benson 1997:20). Based
on her longitudinal research in the Shire Highlands of southern Malawi, Peters (1996) argued
that market liberalization provided new opportunities (through tobacco and maize sales) that
have disproportionately benefited the better-off households, while the poorest 25% experienced
a relative worsening in income and food security. By retaining more maize, selling more labor,
and increasing the budget share of purchased maize, the poorest quartile of the households
suffered a decline in food security.
Changes in the marketing system had shifted the spatial locus of hybrid maize
production, given the relative land-abundance and lower road density of the the Northern
Region. Between 1990 and 1997, the percent of farmers growing F1 hybrid seed dropped from
40% to 18% in the villages of the Northern Region, where farmers had previously produced a
surplus and grown the crop for cash (Smale and Phiri 1998). Although the farmers in the villages
of Blantyre District, Southern Region were using F1 seed, they farmed such small areas and used
such small amounts that they could not market a surplus. The farmers most consistently growing
hybrid maize over the intervening years from 1990 to 1997 period were found in the villages of
Kasungu District in Central Region, where tobacco and other cash crop opportunities were
consistently better.
In Zambia, following price liberalization, “nearly full regional and seasonal
differentiation of maize and other crop prices occurred” to reflect market conditions and
transactions costs (Howard and Mungoma 1996). National maize area contracted relative to the
expansion of the 1980s, particularly in the drier and more remote regions. Crop mixes have
diversified according to agroecological endowments and distances from transport infrastructure.
Exports have doubled in value and become more diversified, including sugar, cut flowers and
specialty crops. On the other hand, many farmers are now “left out” of market processes.
5.5.

Consumption
21

There were undoubtedly widespread benefits to consumers during this maize production
“boom” period, in Kenya, Zambia and Zimbabwe. Rapid rates of population growth would have
resulted in a costly import bill had there been no maize production expansion during this period
—which would have added to the nation’s external debt (Gilbert et al. 1993: 34). However, the
controlled marketing systems during the boom period did not always allow the decline in real
producer prices during the late 1980s to translate into lower consumer prices (Jayne and Jones
1997). In fact, real prices of packaged maize meal during the late 1980s in Zimbabwe remained
relatively constant, despite falling real producer prices, as industrial millers benefited from their
regulated oligopoly position in the official marketing system (Jayne et al 1995). Moreover,
detailed farm-level studies in Zambia and Malawi were unable to demonstrate a link between
adoption of hybrid maize and improved nutritional status of household members (e.g.,
Siandwazi, Bhattarai, and Kumar 1991; Diagne and Zeller, 2001).
5.6. Fiscal sustainability
In Zimbabwe, Kenya, and Malawi, published analyses suggested that public investment
in maize research paid off (ranging from 43 to 64 percent, depending on assumptions and time
periods), even when curtailed by a small market, as in Malawi (Kupfuma, 1994; Karanja 1996;
Smale and Heisey 1994). However, these findings overestimate returns since they do not account
for the allied investment costs in market infrastructure, credit and input distribution programs,
and pan-territorial pricing policies that encouraged the uptake of this technology. Costs were
indeed high. Both the National Cereal and Produce Board (NCPB) in Kenya, and the Grain
Marketing Board in Zimbabwe incurred losses equivalent to five percent of their countries’ GDP
during parts of the 1980s (Jayne and Jones 1997). Zambia’s input and output marketing
subsidies were equivalent to 17 percent of the total government budget in the late 1980s
(Howard and Mungoma 1996).
Only Howard’s (1994) analysis for Zambia explicitly includes the costs of a full range of
investments leading to hybrid maize adoption by smallholder farmers. Marketing costs
accounted for roughly 59% of the total costs of all investments, in contrast to the seed research
investments, which were only 3% of the total. Extension and other service provision programs
accounted for the remaining 38%. The rate of return on maize research was highly favorable
when the costs of marketing were not included. Yet after including all investments (research,
extension, seed and marketing), the average rate of return to maize research in Zambia was
negative over the 1987-91 period.

22

VI.

SUMMARY

We have argued that from 1900 to 1965, 1) the agronomic suitability of the crop, 2) the
British starch market, 3) milling technology, 4) the integration of Africans into the settler
economy and 5) market and trade policies established through agricultural settler lobbies explain
to a large extent why white maize became the dominant food staple in the case study countries
and the preferred staple of today’s African consumers. Furthermore, many of the institutions
established during the colonial period, as well as colonial investments, generated benefits for
smallholder maize production in the post-independence period. They spurred the maize breeding
successes in Kenya and Zimbabwe before 1965. Of the two, however, only Kenya’s was a
smallholder maize revolution. Zimbabwe’s brief smallholder maize revolution followed about
twenty years later.
In our case study countries, we conclude that the maize “seeds” produced—that is, the
quality of plant breeding and agronomic research that accompanied it—were an unquestionable
success. In retrospect, the “maize success stories” were largely a phenomenon of the 1970s and
1980s. During the 1990s, all four countries experienced absolute declines in maize production,
with the exception of Malawi, where production was supported through free input distribution
programs. This underscores the obvious conclusion, also stated by Byerlee and Eicher (1997:
251) that seed genetic change is a necessary but not a sufficient condition for improving the
welfare of African smallholders. In the last decade since the data assembled in their work were
compiled, the necessary investments in germplasm have further declined while the investments
in the institutions required to translate germplasm advances into improved income, including
seed and grain markets, have faltered. The public investments in state-controlled, coordinated
input and output markets were not fiscally sustainable and generated social costs by directing
more resources to maize production and improved maize in areas where farmers may not have
chosen to grow it otherwise. The dismantling of the systems that supported these episodes of
success generated an uncertain decision-making environment for farmers who were already beset
by unfavorable weather conditions. Estimating the true counterfactual, or the social costs of not
having pursued either the research or supporting investments would require not only a general
equilibrium framework, but knowledge of yield levels in the absence of improved germplasm
and the partial effects of genetic as compared to all other intervening physical, economic and
social factors.
Maize successes in the future will continue to depend on strategic crop improvements
such as those targeted to relieve specific environmental and disease problems and enhance the
stability of net returns to farmers, but also on enabling these advances to release land for
alternative uses and diversify the income sources for farmers, regions, and nations. The
continued development of seed markets and a realistic understanding of the farmers who are the
actors in those markets remain critical. Two maize breeders in two time periods, M.N. Harrison
in 1970 and D. Duvick in 2002, have already delivered these messages.
A third, vital point concerns whether states will make the investments needed to support
more efficient private systems of input delivery, finance, and commodity marketing, not only for
maize but for a range of crops that may offer higher returns to farming in the changing
environment of Africa’s rural areas. Because most parts of Africa are experiencing increased
land pressure and limited potential for area expansion, population growth is causing a decline in
land/labor ratios and farm sizes are declining. According to national surveys, about 50 percent
of the small-scale agricultural households in Zambia, Mozambique, Kenya, Rwanda, and
Malawi own less than 1.5 hectares of land (Jayne et al 2002). Maize is a relatively low value-to23

bulk crop. It can provide consistently high returns per unit of land in only a relatively small
proportion of smallholder farming areas in the region. Given reasonable assumptions about
future productivity improvements, it is unlikely that maize can provide the net revenue on the
millions of farms that are 0.5-1.0 hectares or smaller to generate substantial crop income growth,
especially in the semi-arid areas. Overcoming rural poverty will require achieving much higher
returns on small farm sizes. This will require the same type of sustained support of institutional,
policy, and technology systems for a wide range of other crops (e.g., sugar in the lowlands of
western Kenya, tea in the highlands, cotton in semi-arid parts of Zimbabwe, Zambia, and Kenya,
and horticulture where water control is possible and where transport infrastructure to export
markets can be developed. Allied investments in education and skill development, market
infrastructure, and agricultural research programs for a more diversified set of crops will all be
critical.
However, even if successes are achieved in moving toward higher-return cropping
patterns, maize will remain a crucial part of the food security equation in two ways: first, as a
purchased commodity for satisfying the food requirements of a more diversified rural economy,
and second, as a cash crop in areas where it is agro-ecologically suited to provide high returns
(e.g., Kenya’s North Rift, Zimbabwe’s Mashonaland maize belt, parts of Zambia’s Central and
Southern Provinces). Overall however, rising land constraints will progressively encourage
farmers to shift toward crops providing high returns to scarce land – provided that the marketing
and institutional infrastructure allows them to do so. This will be manifested in a shift from the
strategy food self-sufficiency strategy to a comparative advantage strategy.
While such a shift will be crucial for poverty alleviation for millions of small farms in
these countries, particular in semi-arid areas, this outcome is not assured. Governments have an
important role to play in building the institutions to reduce the transaction costs and physical
costs of trade, supporting agricultural research and development, education and extension, and
investing in the development of reliable and low-cost food, finance, and input marketing systems
to serve rural areas, so that households can benefit from the higher incomes afforded by a wellfunctioning commercialized agricultural sector while still ensuring their access to food
(Dorward, Kydd, and Poulton, 1998).
Lastly, can a local constituency be formed to successfully stake a claim on public
resources over the long run to support agricultural research, marketing institutions, and other
kinds of growth-promoting public goods? These issues underscore the strong connection
between agricultural development and governance. The early success of the maize industry in
Zimbabwe and Kenya can be largely attributed to the strength of the institutions built by settler
farmers, which provided a constituency to encourage sustained public and private support for the
sector. Today, farm lobbies are generally weaker, more fragmented, and reflective mainly of
large-scale farmer interests. Representation has always been weak for smallholder farmers,
particularly in semi-arid regions where their welfare is closely tied to the reliability and
efficiency of maize markets where they purchase maize as consumers. A crucial issue is how
the key growth- and equity-promoting investments in agricultural research, infrastructure, and
market institutions can be financed. Perhaps most importantly, where will the domestic political
pressure for these public investments come from?

24

Box 1. Cross-cutting themes: Episodes of success in Ghana, Nigeria, and Ethiopia
The cases of maize in Ghana and Nigeria are distinct from those of Eastern and Southern
Africa in important ways. First, until recently, the germplasm focus has been different,
emphasizing open-pollinated varieties instead of hybrids. About 75% of all improved maize
types released in West Africa as a whole from 1965 were open-pollinated synthetics and
composites rather than the hybrids that have dominated in ESA (Manyong et al. 2001). Second,
and more importantly, the role of maize in the food economy means that while it is less
important as a source of food, it is extensively traded and has been attractive to farmers as
commercial crop. In neither case was the political economy of the country oriented toward the
production of maize as a wage good.
Maize falls far behind roots and tubers as a source of calories in Ghana, but farmers
identify it as an important source of cash income (Morris, Tripp and Dankyi 1998). At least
half of Ghana’s maize production enters markets, which have long been controlled by the
private rather than the public sector (Tripp and Marfo 1997). Recent evidence suggests that
seed markets do not appear to function as well as the grain market, however. Farmers rely to a
large extent on other farmers for seed information and often cannot name the variety they grow
(Morris, Tripp, and Dankyi 1999). Since the beginning of the national maize breeding program
in Ghana (Ghana Grains Development Program) in 1979, most releases have been improved
open-pollinated varieties, though several quality-protein maize (QPM) hybrids were released
from 1997. The emphasis on OPVs reflected the fact that, to maintain the yield advantages of
hybrids farmers must purchase fresh seed every cropping season and are therefore dependent
on a seed industry. However, farmers who grow IOPVs are also advised to replace their seed
after several seasons, and seed quality must in any case be assured. After the collapse of the
parastatal Ghana Seed Company, the NGO Sasakawa-Global 2000 was instrumental in
establishing a system for the production and marketing of OPV maize and cowpea seed, which
Tripp (2001) concluded has been one of the more viable examples of project-initiated smallscale seed production to be found in Africa.
Based on farmers’ qualitative judgments, Morris, Tripp, and Dankyi (1999) found that
the adoption of improved maize types has been associated with significant farm-level yield
increases and noticeable increases in farm income earned from maize sales. They estimated
high benefit-cost ratios, and the tradability of the crop in Ghana as well as farmers’ responses
indicate that these benefits might actually have been earned. As elsewhere, however, they
found that the adoption of improved seed and fertilizer is affected indirectly by factors beyond
the control of researchers—including agricultural extension service, the inputs distribution
system, and the policy environment. With the privatization of fertilizer imports and
distribution, combined with the elimination of the fertilizer subsidy during the 1980s and
1990s, the nitrogen-to-maize price ratio rose exponentially and uptake fell.
Nigeria inherited the strongest research system in West Africa at the time of
independence (Gilbert et al. 1993: 54), and TZB, the first highly successful IOPV, was bred
during the 1960s. Maize expansion was initially driven by its potential as a source of cash
through sales to established southern markets—the taste for maize developed only gradually as
it was mixed into tuwo, marginally improving protein content, and was used for feed or
brewing beer (Gilbert et al. 1993: 61). While maize had been a minor crop through the mid1970s, it displaced cotton and groundnut as a cash crop and sorghum and millet as a food crop
in the 1980s, making the Northern Guinea Savannah one of the most intensified areas in West
Africa. Significant determinants of this change were exogenous to farmers’ decisions: (a) an
improved transport system; (b) the release of and improved maize variety (TZB) that was well
adapted to an ecology that was naturally suited for growing maize because of high solar
25

radiation and low night temperatures, and had a high grain yield response to modern rates of
nitrogen application; (c) high levels of fertilizer subsidy (dating back to the 1950s), though
fertilizer was not widely used until the adoption of improved maize (Smith et al. 1994). At
that time the net margin per hectare of improved maize was six times the net margin for the
traditional food crop mixture of sorghum and millet, and maize was seven times as profitable
as the traditional cash crop, cotton (Balcet and Candler 1981). A budget from Norman,
Simmonds and Hays (1982) showed that the returns to labor for maize in the early 1980s were
superior to other crops in the system (Table 5 ). Survey results indicated that cash earnings
from maize were a primary source of agricultural inputs for use on other crops, substitute for
underdeveloped credit markets and generating income multipliers (Smith et al. 1994).
Several years after the first study, Smith et al. (1997) recognized that yield advantages
came at a fiscal cost. The fertilizer subsidy was around 85% of the real cost of delivering
fertilizers to farmers, and the fiscal cost of the subsidy was 32% of federal government
agricultural expenditure and 3.7% of total government expenditure. Oil revenues enabled the
suppression of the nitrogen-to-maize price ratio to levels as low as 0.9-1.9 from the mid-70s
through the 1970s and 1980s. Fertilizer subsidies led to “fiscal problems, inefficiencies,
corruption, and unsustainable predominance of cereal cropping….”(Smith et al., 1997: 123).
They concluded that although high-yielding, fertilizer responsive maize can make a major
contribution to food production in Nigeria and other areas in Africa, it can do so only in a
limited number of areas. “Production systems in Western Africa, even intensive ones, require
diversity to be sustainable” (Smith et al., p. 124, 1997) since they are inherently complex and
heterogeneous.
The recent Ethiopian experience demonstrates the pitfalls of successfully pursuing a
production-oriented strategy without sufficient concern for market-related factors. The
government extension services did such a good job of scaling up their half-hectare extension
plots during the late 1990s (reaching 5 million farmers in 2000) that farmers produced bumper
harvests in the high potential zones. Insufficient local demand for maize, combined with the
simultaneous sales of large harvests by many farmers in order to meet financial and social
obligations, led to plummeting prices and had serious repercussions on farmer welfare. In the
subsequent season, fertilizer use declined and famers reverted to local maize or planted
second-generation hybrid seed (D. Tanner, pers. comm., and Lemessa, 2001).

26

Box 2. Farmers’ perceptions
At the outset of the 1991/92 season, the Malawi Ministry of Agriculture, CIMMYT, and
National Seed Company of Malawi undertook a farmer evaluation survey in which 150 farmers
in clusters of villages located in Central Mzimba, Kasungu and Blantyre Districts compared the
yield, maturity, processing and storage quality of newly released semi-flint hybrids MH17 and
MH18 to those of their own local maize and the previously released, dent hybrids MH12 and
NSCM41. Each participating farmer was provided with a 5-kg sample of two hybrids and
asked to grow them with usual practices. Enumerators laid yield sub-plots to measure their
yields and interviewed them later in the season regarding their perceptions.
In the drought year that followed, farmers harvested highest yields with MH18, even
when they applied no fertilizer to either the hybrid or their own local check. Their statements
conveyed their surprise about the new hybrids. MH18 was often called “the local seed you
gave us,” as compared to the “real” dent hybrids, NSCM41 and MH12. One farmer said “you
have given us back the local maize we used to grow.” Because flint type and pounding
efficiency were long associated with local maize, farmers called seed that satisfied those
conditions “local.” Others insisted that the semi-flint hybrids were local maize and that they
would keep the seed (Smale et al. 1993).
Several of the participating farmers later traveled to Lilongwe to discuss their
experience on the radio with Ministry of Agriculture field staff, B.T. Zambezi, who bred the
hybrids, and representatives from NSCM, who sold them. Alesi Matope, from Chiradzulu in
Blantyre District, reported “I have been farming for a long time and I used to plant local maize
and NSCM41. In 1989, the field technicians brought MH17 and MH18, which I planted.
Although there was little rainfall, I had good yields compared to local maize and the hybrid
matured earlier. When I harvested this maize, the advisor came and told me that I should store
a little of each because a visitor would come from Lilongwe to see me, and to see the yield,
storage and poundability of these hybrids, and to see which seed produced more bran. I did not
apply any chemical to prevent weevils. NSCM41 was the first to be attacked by weevils. When
pounded, MH17, MH18 and local maize produced little bran. The nsima made from this type
of hybrid is very tasty and its colour is pure white. The roasted green maize is even better, and
of course when shelling from the cob, it shells very easily and one does not have to struggle
with the task.” Jaleki Kapinga from Kasungu reported that “when setting off on my visit to
Lilongwe I was told to tell the truth about the MH17 and MH18 hybrid seed by all other
farmers in my area. ‘They should breed plenty of seed.’ That is what my fellow farmers told
me to tell you…We do not want to find problems in buy MH17 and MH18 hybrid seed in our
markets.” Lyson Mbewe from Central Mzimba said that his wife insisted on keeping the seed
from the harvest, even though he told her it was only an introduction. This photo of Mayi
Jumbe, another farmer from Thyolo in Blantyre District, seated on the top of her maize harvest
after the drought year, was later displayed on one of the annual reports of the Maize
Commodity Team.

27

Box 3. Crisis-motivated maize policy in Malawi
During the worst drought year of 1991/92, an amnesty was granted to on smallholder loans.
Credit expansion in 1992/93 brought in new, less credit-worthy borrowers, and loan repayment
problems were aggravated by the election year promises of politicians to forgive loans.
Massive defaulting on smallholder loans and a liquidity crisis ensued in 1994, leading to the
collapse of the smallholder credit scheme.
From April 1995, in recognition of the need to diversify crop production, prices of all
inputs and crops except maize were fully liberalized and the extension service began promoting
other cross and activities, including cassava and sweet potatoes, and grain legumes to serve in
crop rotations and better nutrition. In maize, the government implemented a price ban system
to dampen extreme variation in maize availability and prices. The difference between import
and export parity prices is great in Malawi, however, and the band therefore had to be narrower
to cushion both consumers and producers from risk of destabilizing price swings (Mann 1998).
In the 1996-7 season, grain reserves were insufficient to release into the market, and
government, which retained regulatory control over imports, imported maize too late (as also
occurred in 2001). Private traders extracted rents by buying rationed maize and selling it at
even higher prices outside ADMARC gates. Ironically, liberalization was blamed for the
shortages and high prices, although these effects arguably were more attributable to continued
regulatory barriers blocking private importation of maize. Government’s attempts to defend
ceiling prices without importing the supplies necessary to do so only exacerbated the
perception of exploitation by private traders. There were reports of maize theft of green maize
from fields and lynch mobs despite security measures, and widespread disillusionment from
liberalization. Emergency imports were costly and high maize prices created inflationary
pressures (Mann 1998). Until the purchasing power of the 60 percent of households in Malawi
that are maize-deficit is raised, “higher maize prices translate into increased misery” (Benson
1997:20).
In response to a crisis situation, the Starter Pack Initiative was proposed. The rationale
for the Starter Pack schemes that were first implemented in 1998/99 was to use the “best bet”
technology (based on the area- and objective-specific fertilizer recommendations of the Maize
Productivity Task Force) to “jump-start” maize production for all smallholders. As originally
conceptualized, then, the program built on the technical knowledge of the maize research
program to refine a package distribution system that was cheaper and more effective in
relieving the food security problem than costly food imports in a dysfunctional maize
marketing system. The plan was to differentiate starter packs from a free inputs program,
containing no more than enough seed and fertilizer for 0.1 ha, and in addition to hybrid maize
seed, seed of other crops, with “well formulated and presented educational materials.” The
program was perceived as a long-term development program (5-10 years) rather than a quick
fix—and by early calculations, paying for itself in terms of both incremental maize output and
other benefits from rotations and intercrops. Since F2 seed of Malawi’s hybrids still yield more
than local maize, benefits would even accrue from farm-saved seed of each year’s allotment.
This approach was considered more effective than input price subsidies, reaching the poorest
rural households more effectively than would credit programs. Starter Packs with hybrid
maize and other seed were distributed to nearly all Malawian smallholders (3.5 million) in the
1988-99 and 1999-2000 growing seasons, and in the 2000-01 growing seasons, the scheme was
replaced by the Targeted Inputs Programme, designed to deliver IOPV maize and other seed to
1.5 million of the poorest smallholders.
28

Practical considerations of the SPI and TIP schemes proved to be different than its
original conceptualization—at least with respect to germplasm. Though the daunting task of
distributing inputs was accomplished, the logistic challenges seem to have precluded any
precise targeting of the maize varieties. The amount of maize seed distributed in the first year
(1998-99) was about 150% of the country’s certified seed production in the previous year)—
leaving many packages filled with grain from ADMARC stocks rather than seed. Since some
of the hybrid maize seed was sourced from South Africa, Zambia and Zimbabwe (likely dent)
due to insufficient supplies of MH18, MH17 and NSCM41, it is likely that much of it did not
perform well. The scheme disturbed the distribution of recommended varieties by sending
them to the wrong destinations (Longley, Coulter, and Thompson 1999: 105). Not surprisingly,
the incremental maize yields attained were fairly low. Furthermore, though there is quite a
range of hybrids offered by the three major competitors in Malawi—on paper, at least (Ngwira,
Kabambwe and Nhlane, 2002)—sales have been thwarted by seed provision schemes
(Nakhumwa 2002).
When harvests were exceptionally good in several seasons, donors apparently turned
coat and began to complain about the welfare nature of the SP scheme (Nation, August 16,
2000). The 2000-01 program, called the Targeted Inputs Programme, provided for roughly half
the number of beneficiaries (1.5 million) as the previous two schemes, and distributed 0.1packs of fertilizer, IOPV maize seed through a voucher system targeting the poorest half of the
smallholder population as part of Malawi’s National Safety Net Strategy (NSNS). Both of
these aspects—the scaling down and move towards “more sustainable inputs” were seen as
part of an “exit strategy” from Starter Pack.
The TIP scheme no longer aimed to promote the use of hybrid maize, on the grounds
that IOPV seed was thought to be more “sustainable.” The justification for this perception is
difficult to find, for at least one of several reasons. First, in the past, yield advantages had been
quite low with existing Malawi IPOVs, though Masika and some recent introductions through
CIMMYT-Zimbabwe appear promising (Ngwira, Kabambe and Nhlane 2000). Second,
optimal seed replacement for IOPV maize varieties is 3 years as compared to annually or biannually with hybrids, but since there is no incentive for private companies to produce IOPV
seed, this seed supply must be generated through other means. In Malawi, the government
failed to pay farmers contracted under the smallholder seed multiplication scheme last year,
causing farmer protests and loss of confidence.
There were problems in implementing the TIP, including a major mismatch between the
rains and delivery of packages. The estimated direct production impact was negligible
compared to earlier years. Levy and Barahona (2001) report that while some skeptics argued
that the 1998-99 and 1999-2000 national output was solely the result of good weather, the
methodology isolated production impact and indicated that (1) SP1 contributed around onequarter of total maize produced by smallholder farmers in 1998-99; (2) SP2 contributed 1530% of total maize produced by smallholder farmers in 1999-2000; (3) TIP was expected to
contribute only 5% of total maize produced by smallholder farmers in 2000-01. They estimated
that only 5 percent of smallholder farmers had enough food to last from the 2000 harvest until
the 2001 harvest. National grain stocks were sold in 2001, and famine ensued in 2002.

29

REFERENCES
Amani, H.K.R. and W. Maro, "Policy to Promote an Effective Private Trading System in Farm Products
and Farm Inputs in Tanzania," in J. Wyckoff and M. Rukuni (Eds.), Food Security Research in
Southern Africa: Policy Implications, (Harare: University of Zimbabwe, 1992).
Aquino, P., F. Carrion, R. Calvo and D. Flores. (2001). Selected maize statistics. Part 4 of the
CIMMYT Maize Facts and Trends. Mexico, D.F., Mexico.
Anthony, C. G. (1988). Mechanization and Maize: Agriculture and the Politics of Technology
Transfer in East Africa. New York, Columbia University Press, 1988.
Argwings-Kodhek, G., 1999. Background Document for the Maize/Wheat Sector Stakeholder
Consultation, Policy Issues Facing the Maize Sector in the North Rift,” Tegemeo Institute
Working Paper, Nairobi. Balcet, J. C. and W. Candler. (1981). Farm technology adoption in
Nigeria. World Bank,
Department of Western Africa Projects, Agricultural Division, Washington, D.C.
Bates, R.H., Markets and States in Tropical Africa: The Political Basis of Agricultural Policies
(Berkeley: University of California Press, 1981).
Bates, R.H, Beyond the Miracle of the Market: The Political Economy of Agrarian Development in
Kenya (Cambridge: Cambridge University Press, 1989).
Benson, T. 1997. The 1995/96 Fertilizer Verification Trial-Malawi. Economic Analysis of Results for
Policy Discussion. Action Group I, Maize Productivity Task Force, Chitedze Agricultural
Research Station, Malawi.
Benson, T. 1999. Validating and strengthening the area-specific fertilizer recommendations for hybrid
maize grown by Malawian smallholders. Maize Commodity Team, Chitedze Agricultural
Research Station, Malawi.
Bonnen, James T. 1990. Agricultural Development: Transforming Human Capital, Technology and
Institutions. In Agricultural Development In The Third World, ed. Carl K. Eicher and John M.
Staatz. Baltimore: Johns Hopkins University Press.
Blackie, M. J. Maize, food self-sufficiency and policy in east and southern Africa, Food Policy,
Vol. 15 (October 1990), pp. 838-394.
Bourdillon, M., P. Hebinck, J. Hoddinott, B. Kinsey, J. Marondo, N. Mudege, and T. Owens. 2002.
Impact of Agricultural Research on Poverty Reduction: Assessing the Impact of HYV Maize in
Resettlement Areas of Zimbabwe. Systemwide Poverty Impact Draft Report. IFPRI.
Bryceson, D., “Urban Bias Revisited: Staple Food Pricing in Tanzania,” in Real Markets: Social and
Political Issues of Food Policy Reform, Hewitt de Alcantara, C. (ed.), London: Frank Cass, 1993.
Buccola, S., and C. Sukume, "Optimal Grain Pricing and Storage Policy in Controlled Agricultural
Economies: Application to Zimbabwe," World Development, Vol 16, No. 3 (1988): pp. 361-71.

Byerlee, D., and C.K. Eicher (eds.) 1997. Africa’s Emerging Maize Revolution. Lynne
Rienner, Boulder.
Byerlee, D. and P. W. Heisey. (1996). Past and potential impact of maize research in sub-Saharan
Africa. Food Policy 21(3): 255-277.
Byerlee, D. and P.W. Heisey. (1997). Evolution of the African Maize Economy. In Eicher, C.K.,
and D. Byerlee, Africa’s Emerging Maize Revolution. Boulder, Lynne Rienner.
Chimedza, R. "Rural Financial Markets," in M. Rukuni and C. Eicher (Eds.), Zimbabwe's Agricultural
Revolution (Harare: University of Zimbabwe Press, 1994), pp. 139-152.
Christiansen, R.E. and V.R. Southworth. 1988. Agricultural pricing and marketing policy in Malawi:
Implciations for a development strategy. Paper presented at the Symposium on Agricultural
Policies for Growth and Development, sponsored by the Government of Malawi, Mangochi,
Malawi.
CSM (Christian Science Monitor), 1996. “Grain Scandal Could Hurt Zimbabwe’s Food Exporting
Efforts,” April 30, 1986, pg. 11.
Conroy, A. 1997. Examination of policy options facing Government in the event of a shortfall in national
maize production. Discussion paper. Lilongwe, Malawi: Ministry of Finance.
Cromwell, E. (1996). Governments, Farmers and Seeds in a Changing Africa. CAB International
and Overseas Development Institute, London.

30

Cromwell, E. and B.T. Zambezi. 1993. The Performance of the Seed Sector in Malawi.
London, U.K.: Overseas Development Institute.
DeVries, J. and G. Toenniessen. 2001. Securing the Harvest: Biotechnology, breeding
and Seed systems for African crops. CABI.
Diagne, A. and M. Zeller. 2001. Access to Credit and its Impact on Welfare in Malawi. IFPRI
Research Report 116.
Dorward. A; Kydd, J., and Poulton, C. (Eds.), 1998. Smallholder Cash Crop Production Under Market
Liberalization, New York: CAB International.
Eicher, C. K. (1995). Zimbabwe’s Maize-Based Green Revolution: Preconditions for Replication.
World Development, V. 23 (5): 805-818.
Eicher, C. K. and B. Kupfuma. (1997). Zimbabwe’s emerging maize revolution. In Eicher and
Byerlee.
Ellis, R. T. (1954). An Investigation into the Types and Relationships of the Maize in Nyasaland
Occurring in Peasant Cultivation. D. Phil thesis. London, U.K., University of London.
Ellis, R. T. (1962). The food properties of flint and dent maize. The East African Journal 24: 251253.
Gabre-Madhin, E., and Haggblade, S. (2001). Successes in African Agriculture: Results of an Experts
Survey. Washington, D.C.: International Food Policy Research Institute.
Gerhart, J. (1975). The Diffusion of Hybrid Maize in Western Kenya. Ph.D. thesis, Princeton
University. Abridged by CIMMYT. CIMMYT, Mexico.
Gilbert, E., L. C. Phillips, W. Roberts, M. T. Sarch, M. Smale, and A. Stroud. (1993). Maize
Research Impact in Africa: The Obscured Revolution. USAID, Division of Food, Agriculture
and Resources Analysis, Office of Analysis, Research and Technical Support,
Bureau for Africa.
Govereh, J. Jayne, T., Nijhoff, J., Haantuba, H., Ngulube, E., Belemu, A., Shawa, J., Banda,
A.,Donovan, C., and Zulu, B. 2002. Developments in Fertilizer Marketing in Zambia:
Commercial Trading, Government Programs, and the Smallholder Farmer. Working Paper 4,
Lusaka: Zambia Food Security Research Project.
GMB (Grain Marketing Board) (1991). Response from the Management of the Grain Marketing Board
to the Structural Adjustment Programme,” (Harare: Planning Unit, Grain Marketing Board,
1991).
Harrison, M. N. (1970). Maize Improvement in East Africa, in C.L.A. Leakey, (ed.). Crop
Improvement in East Africa (Farnham Royal, England: Commonwealth Agricultural
Bureau), pp. 27-36.
Hassan, R. M. and D. D. Karanja. (1997). Increasing maize production in Kenya: technology,
institutions and policy. In Eicher and Byerlee.

Hassan, R.M. 1998. Maize Technology Development and Transfer: A GIS Application
for Research Planning in Kenya. CABI.
Hassan, R. M., K. Njoroge, M. M. Njore, R. Otsyula, and A. Laboso. (1998). In Hassan (ed.)
Hassan, R. M., M. Mekuria, and W. Mwangi. (2001). Maize Breeding Research in Eastern and
Southern Africa: Current Status and Impacts of Past Investments Made by the Public and
Private Sectors 1966-1997. Mexico, D.F., CIMMYT.
Heisey, P.W. and W. Mwangi. 1996. Fertilizer Use and Maize Production in Sub-Saharan Africa.
CIMMYT Economics Working Pazper 96-01. Mexico City, Mexico: CIMMYT.

Heisey, P.W. and M. Smale. 1995. Maize Technology in Malawi: A Green Revolution in the
Making? CIMMYT Research Report No. 4. Mexico, D.F.: CIMMYT.
Howard, J. A. (1994). The Economic Impact of Improved Maize Varieties in Zambia. Ph.D. thesis,
Michigan State University, Department of Agricultural Economics, East Lansing, Michigan.
Howard, J. A., and C. Mungoma. (1996). Zambia’s Stop-and-Go Revolution: the Impact of
Policies and Organizations on the Development and Spread of Maize Technology. MSU
International Development Working Paper No. 61.
Jabara, C. (1984). Agricultural Pricing Policy in Kenya, World Development, 13 (5): 611-626.

31

Jansen, D. (1977). Agricultural Policy and Performance in Zambia: History, Prospects and Proposals
for Change. Berkeley: University of California, Institute of International Studies.
Jansen, Doris, and Kay Muir, "Trade, Exchange Rate Policy and Agriculture in the 1980s," in M. Rukuni
and C. Eicher (Eds.), Zimbabwe's Agricultural Revolution (Harare: Univ. Zimbabwe Press, 1994).
Jayne, T., Mukumbu, M., Chisvo, M., Tschirley, D., Zulu, B., Weber, M., Johansson, R., Santos, P., and
Soroko, D. (1999). Successes and Challenges of Food Market Reform: Experiences from
Kenya, Mozambique, Zambia, and Zimbabwe. International Development Working Paper 72,
East Lansing, Michigan State University.
Jayne, T.S., and M. Chisvo (1991). Unravelling Zimbabwe's Food Insecurity Paradox, Food Policy, 16
(5): 319-329.
Jayne, T.S., and Mandivamba Rukuni (1993). Distributional Effects of Maize Self-Sufficiency in
Zimbabwe: Implications for Pricing and Trade Policy, Food Policy, 18 (4): 334-41.
Jayne, T. S., L. Rubey, D. Tschirley, M. Mukumbu, M. Chisvo, A.P. Santos, M. T. Weber, and P.
Diskin. (1995). Effects of Market Reform on Access to Food by Low-Income Households:
Evidence from Four Countries in Eastern and Southern Africa. International
Development Paper No. 19. Michigan State University, Department of Agricultural
Economics, East Lansing.
Jayne, T.S. and S. Jones (1997). Food Marketing and Pricing Policy in Eastern and Southern Africa: A
Survey. World Development, 25 (9): 1505-1527.
Jayne, T.S. and G. Argwings-Kodhek (1997). Consumer Response to Maize Market Liberalization in
Urban Kenya. Food Policy, 22 (5): 447-458.
Jayne, T.S., J. Govereh, A. Mwanaumo, J. Nyoro, and A. Chapoto (2002). False Promise or False
Premise: The Experience of Food and Input Market Reform in Eastern and Southern Africa.
World Development, November 2002.
Jayne, T.S., T. Yamano, M. Weber, D. Tschirley, R. Benfica, D. Neven, A. Chapoto, and B. Zulu. 2001.
Smallholder Income and Land Distribution in Africa: Implications for Poverty Reduction
Strategies. International Development Paper 24, East Lansing, Department of Agricultural
Economics, MSU.
Jayne, T.S., T. Yamano, J. Nyoro, and T. Awuor. 2002. Do farmers really benefit from higher maize
prices? Balancing rural interests in Kenya’s maize pricing and trade policy. Tegemeo Working
Paper #3, Tegemeo Institute/Egerton University, Nairobi.
Jenkins, Carolyn. 1997. The Politics of Economic Policy-Making in Zimbabwe. Journal of Modern
African Economies 575-602.
Jones, S.P., "Privatization and Policy Reform: Agricultural Marketing in Africa" (Oxford: Food Studies
Group, University of Oxford, 1994).
Kandoole, B.F., and L. Msukwa, "Household Food and Income Security Under Market Liberalization:
Experience from Malawi," in J. Wyckoff and M. Rukuni (Eds.), Food Security Research in
Southern Africa: Policy Implications (Harare: University of Zimbabwe, 1992).
Kaluwa, B., "Malawi Food Marketing: Private Trader Operation and State Intervention," in J. Wyckoff
and M. Rukuni (Eds.), Food Security Research in Southern Africa (Harare: Univ. of Zimbabwe,
1992).
Karanja, D. D. (1990). The Rate of Return to Maize Research in Kenya: 1955-88. M.Sc. thesis,
Department of Agricultural Economics, Michigan State University, East Lansing, Michigan.
Karanja, D. D. (1996). An Economic and Institutional Analysis of Maize Research in Kenya.
Michigan State University International Development Working Paper No. 57. Department of
Agricultural Economics, Michigan State University. East Lansing.

Kettewell, R.W. 1965. Agricultural Change in Nyasaland: 1945-1960. Food Research
Institute Studies 5: 229-285.
Keyter, Carl, Maize Control in Southern Rhodesia: 1931-1941: The African Contribution Toward White
Survival, Local Series 34 (Harare: Central African Historical Association, 1975).
Kherallah, M., Delgado, C., Gabre-Madhin, E., Minot, N., and Johnson, M. (2002). Reforming
Agricultural Markets in Africa, Baltimore: Johns Hopkins University Press., 2002.
Kupfuma, B. (1994). The Payoffs to Hybrid Maize Research and Extension in Zimbabwe: An
Economic and Institutional Analysis. M.Sc. thesis, Michigan State University, East
Lansing, Michigan.

32

Kydd, J. 1989. Maize Research in Malawi: Lessons from Failure. Journal of
International Development 1(1): 112-144.
Kydd, J. and R. D. Christiansen. (1982). Structural change in Malawi since independence:
Consequences of a development strategy based on large-scale agriculture. World
Development 10,5 (1982): 377-396.
Lele, U. (1990). Structural adjustment, agricultural development and the poor: Some lessons from
the Malawian experience. World Development 18(9): 1207-1219.
Lemessa, Dechassa. 2001. Maize price crisis badly hits farmers in Western Ethiopia. Field Assessment
Report. Mimeo. United Nations Development Programme, Emergencies Unit for Ethiopia
(UNDP-EUE).
Levy, S., C. Barahona and I. Wilson. (2000). 1999-2000 Starter Pack Evaluation Programme Main
Report. Statistical Services Centre, University of Reading.
Levy, S., and C. Barahona. (2001). Main report of the monitoring and evaluation programme.
2000-2001 targeted inputs programme. Calibre Consultants, Reading, and Statistical
Services Centre, University of Reading.
Longley, C., J. Coulter, and R. Thompson. (1999). Malawi Rural Livelihoods Starter Pack
Scheme, 1998-99 Evaluation Report. ODI, London, August 1999.
López-Pereira, M. A. and M.L. Morris. (1994). Impacts of International Maize Breeding Research
in the Developing World, 1966-1990. Mexico, D.F., CIMMYT.
Low, A. 1986. Agricultural Development in Southern Africa: Farm Household Economics and the Food
Crisis. London: James Curry.

Lynam and R.M. Hassan, 1998. A new approach to securing sustained growth in Kenya’s maize
sector. In Hassan (ed.) 1998.
Mann, C. (1998). Higher yields for all smallholders through “best bet” technology: the surest way
to restart economic growth in Malawi. Soil Fertility Network Research Results Working
Paper 3. CIMMYT, Harare.
Manyong, V. M., J. G. Kling, K. O. Makinde, S. L. Ajala, and A. Menkir (in press, 2003). Impact of
IITA germplasm improvement on maize production in West and Central Africa. In R.E. Evenson
and D. L. Gollin (eds.), The Impact of International Agricultural Research: Varietal
Improvement and its Effects on Productivity. CABI.
Maredia,M. J., Howard, and D. Boughton. (1999). Increasing Seed System Efficiency in Africa:
Concepts, Strategies, and Issues. MSU International Development Working Paper No. 77.
Masters, W. A. (1994). Government and Agriculture in Zimbabwe. Westport, CT, Praeger.
Miracle, J. P. 1966. Maize in Tropical Africa. Madison, University of Wisconsin Press.
Minot, N., M. Kherallah, P. Berry. (2000). Fertilizer Market Report and the Determinants of
Fertilizer Use in Benin and Malawi. MSSD Discussion Paper No. 40. IFPRI.
Morris, M. L. (1998). Maize in the developing world: Waiting for a Green Revolution. In M.L.
Morris (ed.), Maize Seed Industries in Developing Countries. Lynne Rienner, Boulder.
Morris, M. L., R. Tripp, and A. A. Dankyi. (1999). Adoption and Impacts of Improved Maize
Production Technology: A Case Study of the Ghana Grains Development Project.
Economics Program Paper 99-01. Mexico, D.F.: CIMMYT.
Morris, M. L., J. Risopoulos, and D. Beck. (1999). Genetic Change in Farmer-Recycled Maize
Seed: A Review of the Evidence. Economics Working Paper 99-07. CIMMYT, Mexico,
D.F., Mexico.
Morris, M. L. (2001). Assessing the benefits of international maize breeding research: An
overview of the global maize impacts study. Part II of the CIMMYT 1999-2000 World
Maize Facts and Trends. Mexico, D.F.: CIMMYT.
Mosley, P., The Settler Economies: Studies in the Economic History of Kenya and Southern Rhodesia
1900-63 (Cambridge: Cambridge University Press, 1983).
Mosley, P., "Maize Control in Kenya 1920-1970," (Bath: Center for Development Studies, Univ. Bath,
1975).
Nakhumwa, C. (2002). Assessment of Marketing and Utilisation of Improved Maize Seed in
Malawi: Cases of Lilongwe and Salima ADDs. SOUTHERN AFRICAN DROUGHT AND
LOW FERTILITY PROJECT (SADLF). Internal Report.
Ngwira, P., V. Kabambe, G. Nhlane. (2002). Characteristics of Maize Varieties Released in

33

Malawi. Chitedze Agricultural Research Station, Lilongwe, Malawi.
Norman, D., E. Simmonds, and H. Hays. (1982). Farming Systems in the Nigerian Savanna.
Boulder, Co: Westview Press.
Nyoro, J. M. Kiiru, and T. Jayne (1999). Maize Market Liberalization in Kenya. Paper presented at the
Third Workshop on Agricultural Transformation in Africa. June 27-30, Nairobi, Tegemeo
Institute, Egerton University.
Odhiambo, Mark, and David Wilcock, "Reform of Maize Marketing in Kenya." In M. Rukuni, G.
Mudimu, and T. Jayne (eds.), Food Security Policies in the SADCC Region (Harare: Univ.
Zimbabwe, 1990).
Pachai, B. 1973. Land policies in Malawi: An examination of the colonial legacy. Journal of Modern
African History 14(4): 681-689.
Page, S. L. J. and P. Chonyera. (1994). The promotion of maize fertilizer packages: A cause of
household food insecurity and peasant impoverishment in high rainfall areas of Zimbabwe.
Pandey, S. (1998). Varietal development. In M.L. Morris (ed.), Maize Seed Industries in
Developing Countries. Lynne Rienner, Boulder.
Peters, P. (1996). Failed magic or social context? Market liberalization and the rural poor in
Malawi. Development Discussion Paper No. 562, Harvard Institute of International
Development. But hers is a small sample from the Shire Highlands.
Pinckney, Thomas C., "Is Market Liberalization Compatible with Food Security? Food Policy, Vol. 18,
No. 4 (1993), pp. 325-333.
Rohrbach, D.D. 1988. The Growth of Smallholder Maize Production in Zimbabwe: Causes and
Implications for Food Security. Ph.D. diss, E. Lansing, Michigan State University.
Rohrbach, D., I. Minde, M. Maredia, and J. Ndjeunga, forthcoming. Harmonizing Seed Trade Policies
in Southern Africa. Food Policy, June 2003.
Rubey, L., 1995. "Maize Market Reform in Zimbabwe: Linkages between Consumer Preferences, Small
Enterprise Development and Alternative Market Channels," Ph.D. diss., E. Lansing: Michigan
State Univ.
Rusike, J. (1995). An Institutional Analysis of the Maize Seed Industry in Southern Africa.
Unpublished PhD Thesis. Dept. of Agricultural Economics, Michigan State University.
Rusike, J. and C. K. Eicher. (1997). Institutional innovations in the maize seed industry. In Byerlee
and Eicher.
Rusike, J. (1998). Zimbabwe. In M.L. Morris (ed.), Maize Seed Industries in Developing
Countries. Lynne Rienner, Boulder.
Sahn, D. and J. Arulpragasam. (1991). The stagnation of smallholder agriculture in Malawi. Food
Policy: 219-234.
Siandwazi, C., S. Bhattarai, and S. K. Kumar. (1991). The effects of technological change on food
consumption and nutritional status in Eastern Province. In R. Celis, J.T. Milimo and S.
Wanmali (eds.), Adopting Improved Farm Technology: A Study of Smallholder Farmers
in Eastern Province, Zambia. IFPRI.
Scott, Guy, "Agricultural Transformation in Zambia: Past Experience and Future Prospects," paper
presented at Agricultural Transformation Workshop, Abidjan (East Lansing: Michigan State
Univ., 1995).
Shopo, Thomas. 1985. The Political Economy of Hunger in Zimbabwe. Working Paper No. 2. Harare:
Zimbabwe Institute of Development Studies.
Sibale, P. K., A. M. Chirembo, A. R. Saka, and V. O. Lungu. (2001). Food Production and
Security. Module 1. of the 2000-2001 targeted inputs programme. Calibre Consultants,
Reading, and Statistical Services Centre, University of Reading.
Sasaki, N., “Maize Market Liberalization, Seasonal Price Storage, and Private Sector Storage,” in
Towards 2000: Improving Agricultural Perfornance, (Nairobi: Egerton Univ./Policy Analysis
Matrix, 1995).
Smale, M. 1992.
Smale, M., with Z. H. W. Kaunda, H. L. Makina, M. M. M. K. Mkandawire, M. N. S. Msowoya,
D. J. E. K. Mwale, and P. W. Heisey. (1991). Chimanga Cha Makolo, Hybrids and
Composites: An Analysis of Farmer Adoption of Maize Technology in Malawi, International
Maize and Wheat Improvement Center (CIMMYT) Economics Working Paper 91/04,

34

Mexico, D.F.: CIMMYT.
Smale, M., Z. H. W. Kaunda, H. L. Makina, M. M. M. K. Mkandawire. (1993). Farmers'
Evaluation of Newly Released Maize Cultivars in Malawi: A Comparison of Local Maize,
Semi-Flint and Dent Hybrids. Harare: International Maize and Wheat Improvement
Center (CIMMYT).
Smale, M. and P. W. Heisey. (1994). Maize Research in Malawi Revisited: An Emerging Success
Story? Journal of International Development 6 (November-December 1994): 689-706.
Smale, M., and A. Phiri, with contributions from G. A. Chikafa, P. Heisey, F. Mahatta, M. N. S.
Msowoya, E. B. K. Mwanyongo, H. G. Sagawa, and H. A. C. Selemani. (1998).
Institutional Change and Discontinuities in Farmers’ Use of Hybrid Maize and Fertilizer in
Malawi: Findings from the 1996-7 CIMMYT/MOA Survey. Economics Working Paper 9801. CIMMYT.
Smale, M. (1995). Maize is life: Malawi’s delayed green revolution. World Development 23(5):
819-831.
Smith, J., A. D. Barau, A. Goldman, and J. H. Mareck. (1994). The role of technology in
agricultural intensification: The evolution of maize production in the Northern Guinea
Savanna of Nigeria. Economic Development and Cultural Change 42(3): 537-554.
Smith, J., G. Weber, M.V. Manyong and M.A.B. Fakorede. (1997). Fostering sustainable increases
in maize productivity in Nigeria. In Eicher, C. K., and D. Byerlee, Africa’s Emerging Maize
Revolution. Boulder, Lynne Rienner.

Tanner, D. 2002. Agronomist, East Africa Region, CIMMYT-Addis Ababa. Personal
communication. February 19.
Tindall, P.E.N. History of Central AFrica, London, Longman Green and Company, 1968.
Tembo, G. forthcoming. Crop Marketing Patterns in Zambia: Implications for Pricing
Policy. Working Paper, Lusaka, Food Security Research Project-Zambia.
Toye, J. (1992). Interest Group Politics and the Implementation of Adjustment in Sub-Saharan Africa.
Journal of International Development, Vol 4(2), 183-198.
Tripp, R. and K. Marfo. (1997). Maize technology development in Ghana during economic decline
and recovery. In Eicher and Byerlee.
Tripp, R. (2001). Seed Provision and Agricultural Development. Overseas Development Institute
and James Currey, London and Oxford.
Weinmann, H. 1972. Agricultural Research and Development in Southern Rhodesia, 1890-1923.
Department of Agriculture Occasional Paper 4. Harare: University of Zimbabwe.

Williams, L.B., and J.H. Allgood. 1990. Fertilizer Situation and Markets in Malawi.
Muscle Shoals. Alabama, International Fertilizer Development Center.
Williamson, J. (1956). Useful Plants of Malawi. Zomba: Government Printer.
World Bank, Adjustment in Africa: Reforms, Results and the Road Ahead (N.Y.: Oxford Univ. Press,
1994).
Zambezi, B. T., F. K. Nyondo, G. Nkhono, G. F. Mbingwani, and T. R. Chakuta. (1997).
Evaluation of recycled maize hybrids at three levels of nitrogen in Malawi. In Proceedings
of the 5th Eastern and Southern Africa Regional Maize Conference, June 3-7, Arusha,
Tanzania. Nairobi, Kenya: CIMMYT.

35

Table 1 -- Selected production and consumption statistics for maize case study countries, 1997-99
Country
Kenya
Malawi
Zambia
Zimbabwe

Production 1997-99
(‘000 t)
2255
1826
818
1710

Maize area % of total
cereal area
79
89
78
75

Average per capita consumption of
maize as food (kg/yr)
94
148
140
122

Average % of maize used
in human consumption
91
82
85
76

42
34

75
55

Ghana
1015
52
Nigeria
5419
22
Source: Aquino et al. (2001). Based on FAOSTAT.

36

Table 2 -- Percentage of national maize area in improved maize for case study countries, 1990, 1996 and 2000
Country

1990
1996
1999
Hybrids
All MVs
OPVs % Hybrids
All MVs OPVs % Hybrids
All MVs
%
Kenya
8%
62
70 7.5-9.5
62-64
71-73
2%
85
87
Malawi
3
11
14 1-4.4
13-33
14-37
4
39
43
Zambia
5
72
77 0.6-0.7
18-22
19-23
3
62
65
Zimbabwe
0
96
96 0-4.5
82-91
82-96
9
91
100
Sources: Hassan et al. (2001), CIMMYT (2001); Manyong et al. (2002). Malawi Ministry of Agriculture,
Lopez-Pereira and Morris, 1994, Smale and Phiri (1997). Figures for 1999 in Malawi include free input distributions
OPVs %

37

Table 3 --National maize area, yield and production growth rates during and after episodes of success
Area
Average Growth
('000 ha) rate (%)
Episodes of success
Kenya (1965-1980)
Zimbabwe (1980-1989)
Zambia (1970-1989)
Malawi (1983-1993)

1330
1293
774
1253

1.86
-0.43
-3.07
1.92

Yield
Average Growth
(mt/ha) rate (%)
1.31
1.47
1.59
1.10

1.44
2.21
4.92
1.18

Coefficient
of variation
0.09
0.38
0.18
0.23

Production
Average Growth
(mt)
rate (%)
1757970.6
1905928.6
1167970
1379579.8

3.30
1.77
1.85
3.10

Decline (1990-2000)
Kenya (1990-2000)
1420
-0.12
1.65 -1.32
0.12
2341964
-1.45
Zimbabwe (1990-2000)
1313
2.52
1.29 -2.75
0.36
1013827
-0.23
Zambia (1990-2000)
635
-1.56
1.60 -0.84
0.29
1722154
-2.41
Malawi (1994-2000)
1282
0.75
1.35
3.67
0.24 1753398.3
4.42
FAOSTAT data used for purposes of internal consistency.
Coefficient of yield variation adjusted for trend (Cuddy and Della Valle, 1978).
Notes: 1992 excluded for Zimbabwe, Zambia and Malawi because of extreme values.
In Malawi, with several years of good rainfall and infusions of free germplasm through
Starter Pack and other initiatives, growth rates are positive even during the decline (see Figure 2)

38

Table 4 -- Illustrative partial budget for smallholder maize producer in Malawi, during and after the episode of success

Yield (kgs/ha)
1
Producer price(MK/kg)
transport and
harvesting costs/kg
Gross returns(MK/ha)
2
Seed costs (MK/ha)
3
Fertilizer
4
Credit charges
Variable cost
Returns to land (MK/ha)
Returns to land (USD/ha)
5
Returns to labor (MK/day)
Returns to labor (USD/day)

1991
1996
Fertilized
Unfertilized
Fertilized
Unfertilized
Hybrid Maize Local Maize
Hybrid Maize Local Maize
--constant 1991 Malawi Kwacha (MK)-2774
745
2774
745
0.27
0.27
0.24
0.24
0.04
0.04
0.04
0.04
638.02
37.5
196.35
28.06
261.91
376.11
137.27
6.07
2.21

171.35
6.5
0
0
6.50
164.85
60.16
3.23
1.18

561.88
77.28
358.98
174.50
610.75
-48.88
-17.84
-0.79
-0.29

150.90
10.53
0
0
10.53
140.37
51.23
2.75
1.00

Source: Yield, harvesting, labor, wages from survey data summarized in Smale et al. (1991), for 420 farmers
in major maize-producing regions of Malawi; 1996 seed prices from survey data summarized in Smale
and Phiri (1997).Fertilizer cost per hectare from Benson (1997).
1
The producer price is used here although the consumer price is relevant for most Malawian smallholder
farmers, who are deficit producers of maize. Using the consumer price would increase
2
25 kgs/ha constant seeding rate
3
At recommended rates of N-P-K. (96:40 nutrient kgs/ha)
4
Credit was virtually unavailable for smallholder maize production in the 1996 survey, and interest rates
charged by the Malawi Rural Finance Corporation ranged from 40-54 percent.
5
6-hour days, 62 person-days for hybrid maize, 51 person-days for unfertilized local maize.
Modal rural wage in 1991 of MK 1.3/day assumed not to have changed in real terms in 1996.
In 1996, MK 15=USD 1; in 1991, MK 2.74= USD 1.

39

Table 5 -- Returns to Labor, Daudawa Village, S. Kastina, 1973-4
Crop

Improved maize
Improved sorghum
Local sorghum
Improved cotton
Local cotton

Labor input
Overall return
without traction
to labor
(days/ha
526.3
400.5
240.9
516.9
526.3

(N/hour)
0.33
0.17
0.21
0.07
0.04

Return to labor;
peak labor period
(June-July)
(N/hour)
1.05
0.53
0.81
0.25
0.17

Source: Norman, D., Simmonds, E., and Hays, H. 1982. Farming Systems
in the Nigerian Savanna. Boulder, CO: Westview Press. Cited in Gilbert (1993).
Note: cotton prices were low on the world market at that time.

40

Figure 1. Per Capita Grain Production in African Communal Lands, Zimbabwe 19141994.

Source: computed from data in Annual Reports of the Chief Native Commissioner (presented in Mosley 1983);
Ministry of Agriculture (1995).
Note: The spline function generating the trend estimates was:
Yt = b + b1(TRENDt) + b3D(TRENDt – I) + b4D(TRENDt - I)2 + vt
where Yt is grain output per capita in Zimbabwe's communal lands in year t; TREND is a time trend; I is a constant
equalling the value of the time trend variable at independence in 1980; D is a dummy variable taking on a value of
zero before independence in 1980 and 1 otherwise. The term t is assumed to be independent and identically
normally distributed with mean zero and constant variance. The estimated average annual change in per capita grain
output is b1 from 1914-1979 and b1 + b3 + 2b4(TREND-I) from 1980-1994. This specification allows for nonlinearity in the trend after independence. F-test results rejected the assumption of linearity at the .001 significance
level. Model estimation produced the following results (t-statistics in parentheses):
Yt = 308.7 - 1.85(TRENDt) + 32.8(TRENDt-I) - 2.21(TRENDt-I)2
(18.3) (-4.32)
(3.99)
(-3.77)

R2 = 0.23
F = 8.99
DW = 2.21

Results indicate that average annual change in grain production per capita was - 1.85 kgs from 1914-1979 and -1.12
kgs when evaluated at the mid-year of the post-independence period (1987).

41

160

2000

120

1500

80

1000

40

500

0
1970

1975

1980

GMB buying stations
Maize yields

1985

1990

1995

kgs per hectare (maize yield)
Z$ * e7 (Agric. expenditures)
Z$ * e6 (AFC credit)

buying stations

Figure 2. Public expenditures supporting smallholder maize yields in Zimbabwe
during episodes of success and decline, 1970 – 2000 (monetary values are in constant
2000 Zimbabwe Dollars)

0
2000

Ministry of Agric. expenditures
AFC credit

notes: GMB buying stations include main depots and collection points; Ministry of Agriculture
expenditures include research and development, extension services, and allied services; AFC credit is the
value of seasonal credit disbursed by the Agricultural Finance Corporation.
Sources: Agricultural Finance Corporation, Agricultural and Economic Review and Board papers, Harare
office; Ministry of Lands, Agriculture and Rural Resettlement, 2001. The Agricultural Sector of Zimbabwe:
Statistical Bulletin – 2001, Harare.

42

Figure 3. Maize Area and Maize Marketed by the Small-scale and Medium-scale
Farming Sector, Zambia, 1980-1999.
1400

1400000
1200000

1000
1000000

800
600

800000

400
600000

200
0

400000
84

86

88

maize sales

90

92

94

maize area

96

total cropped area

Source: Crop Forecast Surveys, Central Statistics Office, Lusaka.

43

98

hectares

000 tons marketed

1200

Figure 4 –
Changes in Percentage Area Distribution among Maize and Competing Major Crops, Malawi, 19822000
100%

80%

60%

40%

GROUNDNUT/TOBACCO
CASSAVA/SWEET POTATO
SORGHUM/MILLET

20%

MAIZE

0%
1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

Source: Malawi Official Crop Estimates, 1982-2000

44

1992

1993

1994

1995

1996

1997

1998

1999

2000

Figure 5 –

Hybrid maize diffusion, maize production, exogenous shocks and policy changes in Malawi, 19812001
2

60

1.6

Election year
promises
government

Collapse of
smallholder
credit scheme

SP and TIP
programs

50

1.4

40

1.2
1

30

0.8
0.6

20
Worst drought
year

Devalution and
removal of
fertilizer subsidy

0.4

Reduced TIP
substitution of
IOPV seed

10

0.2
0

0
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Maize yields
Maize production
Official crop estimates
Seed company estimates

Data sources: FAOSTAT, National Seed and Cotton Milling, Malawi Ministry of Agriculture (unpublished data), Nakhumwa (2002).
Data sources: FAOSTAT, National Seed and Cotton Milling, Malawi Ministry of Agriculture (unpublished data), Nakhumwa (2002).

45

Percent maize area in hybrids

Maize yields and production per capita (mt)

1.8