Narrowing the Wheat Gap in Sub-Saharan
Africa: A Review of Consumption and
Production Issues*
Michael L. Morris and Derek Byerlee
CIMMYT, Mexico
Introduction
Over the past 3 decades, dietary patterns in sub-Saharan Africa have
undergone dramatic changes. Perhaps the single most remarkable development has been the rapid growth in consumption of wheat. Between 1960 and 1985, annual wheat consumption in sub-Saharan Africa
increased from less than 2 million tons to over 6 million tons, before
declining somewhat in the late 1980s. Underlying this sharp increase,
two trends can be discerned. First, more than two-thirds of the growth
in wheat consumption has been provided by imports; wheat production
has grown by less than 1 million tons and currently makes up only
about two-fifths of total consumption (fig. 1). Second, the rapid increase in consumption of wheat, a nontraditional food, has been
achieved at the same time that per capita consumption of traditional
food staples has been falling. Combined, these two trends suggest rapid
substitution of imported wheat for domestically produced food staples.
Wheat in sub-Saharan Africa is still a relatively minor food for
most consumers (per capita consumption averages around 10 kilograms per year) and an even less important crop for producers (annual
production exceeds 50,000 tons in only six countries). However, wheat
is economically and politically important because it constitutes a major
share of food imports and because it is consumed mainly in urban
areas. Largely for these reasons, the wheat economy of sub-Saharan
Africa has been the subject of numerous recent studies, both at the
regional level and at the individual country level. I
Many African countries today stand at a crossroads with respect
to wheat. They may choose to let recent trends continue, in which
case imported wheat will likely become a staple food for most urban
consumers. 2 Such a precedent already has been set in tropical Latin
© 1993 by The University of Chicago. All rights reserved.
0013-0079/93/4104-0006$01.00
ex>
w
-..J
c
l/l
~
.Q
C
g
I
FIG.
o
2
3
4
5
6
7
I
,
1965
1970
1975
1980
1985
1990
I.-Wheat production, consumption, and imports in sub-Saharan Africa, 1961-90
1960
Michael L. Morris and Derek Byerlee
739
America; in the Andean region, where production environments are
similar to many highland zones in Mrica, annual wheat consumption
now averages over 40 kilograms per capita, 91 % of which is imported.
Alternatively, if Mrican countries want to avoid a high level of dependence on imported wheat, appropriate policy decisions must be taken
now to manage wheat production and consumption.
This article reviews some of the options available to Mrican policymakers for managing wheat consumption in the short to medium
run and describes economic factors that influence the potential for
increasing domestic wheat production over the longer run. Policy reforms implemented during the 1980s to curtail wheat consumption
growth apparently have begun to take effect; however, efforts to build
up domestic production capacity generally have been less successful.
Recent studies indicate that wheat production in Mrica can sometimes
be efficient, but only under certain conditions that have not always
been recognized by policymakers. Consideration of both sides of the
wheat balance equation is important not only for designing efficient
and equitable policies but also for determining an appropriate level of
funding for future wheat research.
Historical Origins of Wheat in Africa
Wheat is a nontraditional crop in sub-Saharan Mrica that apparently
came into the region via several different routes. As early as the ninth
and tenth centuries, small quantities of wheat, arriving from the Mediterranean region with the trans-Sahara caravans or by sea along the
coast, were consumed by wealthy Arab traders in northwestern Mrica
and coastal eastern Mrica. 3 The European explorers, traders, and missionaries who began appearing in increasing numbers during the sixteenth century also brought with them a taste for bread made from
wheat flour. Much more recently, freed Mro-Brazilian slaves returning
to western Africa after 1835 introduced the habit of bread eating to
the local people. 4 Colonization provided a strong impetus to wheat
consumption, and by the end of the nineteenth century wheat flour
was a well-established import into all but the most remote interior
regions. s
Initially, imported wheat was a luxury food consumed mainly by
expatriates and a small number of wealthy Mricans. This situation
changed following World War II, when a number of wheat-producing
countries in Europe and the Americas adopted aggressive exportoriented agricultural policies to dispose of surplus production. As
world wheat supplies rose and international prices fell, Mrican countries began to import wheat in increasing quantities to make up for
shortfalls in domestic production of traditional staples, and bread soon
became a staple in the diet of many Mricans, especially in urban
areas. 6
Economic Development and Cultural Change
740
Although wheat consumption has a relatively long history in
sub-Saharan Africa, wheat production is much more recent (except in
Ethiopia, where wheat is a traditional crop). While there are reports of limited wheat cultivation in Chad and Nigeria dating from
the eighteenth century, extensive production began only during the
latter half of the nineteenth century, when European settlers in
eastern and southern Africa began cultivating the crop.7 Agroclimatic conditions proved most favorable for wheat in the highland
regions of Kenya, Tanzania, Zambia, and Zimbabwe, where it eventually became established on settler farms. Efforts were made subsequently to introduce wheat into other parts of the continent (e.g.,
the Sahel, western Africa, central Africa), but outside of a few highland areas, production generally has been possible only with irrigation.
Recent Wheat Production and Consumption Trends
For many African countries, independence ushered in a sharp rise in
wheat consumption. From 1961 to 1990, wheat consumption in subSaharan Africa increased at an annual rate of 4.2%, or 1.3% per capita
(table 1). Starting from a low base, the fastest growth per capita occurred in western Africa (2.9%), the Sahel (2.0%), and central Africa
(1.2%).8 Wheat consumption was already more established in eastern
and southern Africa by the early 1960s, so consumption per capita
grew more slowly in these regions (0.9% and 0.1%, respectively).
These aggregate statistics conceal considerable diversity between individual countries. The most dramatic increases occurred in Benin, Mauritania, Niger, Somalia, Swaziland, and Togo, where per capita consumption has risen by over 6% per year over the past 3 decades. In
contrast, in Burkina Faso, Burundi, Madagascar, Malawi, Namibia,
TABLE
1
WHEAT CONSUMPTION TRENDS IN SUB-SAHARAN AFRICA,
Sahel
Western
Central
Eastern
Southern
Total
1961-65
TO
1986-90
1961-65
Annual
Wheat
Consumption
(Millions
of Tons)
1986-90
Annual
Wheat
Consumption
(Millions
of Tons)
1961-65
Per Capita
Wheat
Consumption
(Kg)
1986-90
Per Capita
Wheat
Consumption
(Kg)
Consumption
per Capita,
Annual
Growth
.11
.27
.83
.31
2.45
.85
4.70
5.2
3.6
4.2
17.1
9.6
8.5
6.7
5.6
4.9
20.8
9.1
10.2
2.0
2.9
.25
.14
.98
.43
1.92
(%)
1.2
.9
.1
1.3
SOURCE.-Calculated from data reported in the AGROSTAT-PC computerized information system distributed by the Food and Agriculture Organization of the United Nations (Rome, 1991).
Michael L. Morris and Derek Byerlee
741
and Uganda, wheat consumption is still very low (less than 5 kilograms
per capita) and rising slowly, if at all.
During the period when wheat consumption in sub-Saharan Africa
was rising, total cereal consumption per capita actually decreased by
about 10% because cereal production failed to keep pace with rapid
population growth. Consequently, during the past 3 decades the share
of wheat among all cereals consumed rose from about 5% to over 10%,
meaning increased wheat consumption has come about chiefly from
substitution for other cereals, primarily sorghum, millet, and maize.
As wheat has grown in popularity among consumers, domestic
supply has not kept pace with demand. Despite widespread efforts to
promote local wheat industries, per capita wheat production has fallen
over the past 3 decades (table 2). Wheat is a significant crop (over
50,000 tons produced) in only six African countries, of which only
two, Tanzania and Zimbabwe, increased wheat production faster than
consumption from 1961 to 1990-and even these two countries experienced reductions in self-sufficiency levels during the 1980s.
Efforts to increase wheat production in Africa often have involved
development of large-scale "crash" production schemes based on imported technologies, high levels of purchased inputs, and mechanization. Despite extensive support from governments and foreign donors,
as well as considerable protection through price supports, input subsidies, and trade restrictions to keep out competing imports, most of these
schemes have not lived up to expectations. In view of the disappointing
production record, most of the increase in African wheat consumption
during the past 2 decades was supplied by imports, which have grown
at an annual rate of over 5% (table 3). Wheat self-sufficiency dropped
sharply, especially in eastern Africa, and currently only 39% of the
TABLE
2
WHEAT PRODUCTION TRENDS IN SUB-SAHARAN AFRICA.
Sahel
Western
Central
Eastern
Southern
Total
1961-65
Annual
Wheat
Production
(Thousands
of Tons)
1986-90
Annual
Wheat
Production
(Thousands
of Tons)
6.2
17.6
10.5
795.2
108.2
937.8
9.7
52.0
23.9
1,32\.4
419.2
1,826.1
1%1-65
TO
1986-90
1986-90
Per Capita
Wheat
Production
(Kg)
Production
per Capita.
Annual
Growth
(%)
1961-65
Per Capita
Wheat
Production
(Kg)
1.4
3.2
3.7
\.8
4.9
2.4
.3
.3
.3
13.8
2.4
4.2
.2
.4
.4
11.2
4.5
4.0
-1.1
.2
\.0
-1.1
\.9
-.5
Production.
Annual
Growth
(%)
SOURCE.-Calculated from FAO AGROSTAT-PC (1991) data base (diskette version) using loglinear regression.
Economic Development and Cultural Change
742
TABLE 3
WHEAT IMPoRT TRENDS IN SUB-SAHARAN AFRICA, 1961-65 TO 1986-90
Sahel
Western
Central
Eastern
Southern
Total
1961-65
Annual
Wheat
Imports
(Millions
of Tons)
1986-90
Annual
Wheat
Imports
(Millions
of Tons)
.11
.23
.26
.78
.28
.12
.19
.32
1.13
.43
2.88
.98
Imports.
Annual
Growth
(%)
1961-65
Per Capita
Wheat
Imports
(Ka)
1986-90
Per Capita
Wheat
Imports
Imports
per Capita,
Annual
Growth
(Kg)
(%)
4.9
6.4
S.3
4.S
9.6
4.6
6.3
2.1
2.9
1.2
4.7
-1.0
2.3
4.6
6.0
3.9
7.6
1.9
S.2
J..4
3.9
3.3
7.2
4.3
SoURCE.-Calculated from FAO AGROSTAT-PC (1991) data base (diskette version) using loglinear regression.
wheat consumed throughout the continent is locally produced (table
4).
The rapid increase in wheat consumption in sub-Saharan Africa
has been described as part of a pattern of changing food preferences
brought about by urbanization. 9 It is certainly true that in sub-Saharan
Africa, as elsewhere, consumption of bread and wheat-based pasta
products has increased because of demographic and economic changes
associated with urbanization (e.g., increased employment outside the
home, greater women's participation in the formal labor market, and
the rising cost of cooking fuel). Nonetheless, in many African countries
consumption of wheat also has been encouraged by policies that have
TABLE 4
WHEAT SELF-sumCIENCY LEVELS AND FOOD AID REcEIPTS IN SUB-SAHARAN AFRICA,
1961-65
Sahel
Western
Central
Eastern
Southern
Total
TO
1986-90
1961-6S
Wheat
Self-sufficiency
Ratio·
1986-90
Wheat
Self-sufficiency
Ratio·
1971-7S
Food Aid
Reliance
Ratiot
1984-89
Food Aid
Reliance
Ratiot
(%)
(%)
(%)
(%)
6
7
8
81
25
49
4
6
8
54
49
39
41
12
8
31
41
6
31
100
62
54
11
17
SOURCE.-Calculated from FAO AGROSTAT-PC (1991) data base (diskette version)
and Food Aid in Figures (Rome: Food and Agriculture Organization of the United
Nations, 1983, 1990).
• Wheat self-sufficiency ratio = domestic production/total utilization.
t Food aid reliance ratio = wheat food aid receipts/total wheat imports (discrepancies may arise because food aid flows are reported in marketing years, whereas trade
flows are reported in calendar years).
Michael L. Morris and Derek Byerlee
743
helped make imported cereals more available and more affordable than
domestically produced staples.
Four broad groups of policies can be distinguished that have been
especially influential in promoting increased wheat consumption in
Africa:
1. Monetary and fiscal policies frequently have led to exchange
rate overvaluation, which has reduced the cost of imported foods (e.g.,
wheat and rice) relative to domestically produced staples (e.g., white
maize, millet, sorghum, and roots and tubers). Many countries in Africa long maintained official exchange rates at several times their real
value, with pronounced effects on relative prices. For example, in
Nigeria from 1975 to 1984, the index of all food prices increased by
over 300%, while the price of imported wheat flour increased by less
than 150%.10 As a result of these divergent growth rates, wheat flour
became cheaper than locally produced maize. Exchange rate overvaluation similarly contributed to declining relative prices of bread in several other African countries where wheat consumption and imports
have grown most rapidly (e.g., Ghana, Sudan, Zimbabwe).
2. Explicit food subsidies have been instituted in many African
countries, ostensibly to protect the incomes of poor consumers from
the effects of rising food prices. These subsidies often favor imported
foods, consumed primarily in urban areas, especially wheat. In Sudan,
bread generally has been substantially cheaper than local staples based
on sorghum, providing a real incentive to substitute wheat in consumption patterns. II Other countries that have maintained subsidies on imported wheat include Benin, Cote d'Ivoire, Ethiopia, Mauritania, Somalia, Tanzania, and Zimbabwe.
3. Agricultural production policies frequently have discouraged
production of local food staples through low producer prices, deficient
input distribution systems, poor infrastructure, and inadequate research investment in food crops. 12 In the face of lagging and unreliable
domestic food production, it often has been easier for governments to
turn to imported cereals such as wheat to supply urban consumers,
especially those in coastal cities.
4. Food aid policies have made it easier for African governments
to maintain high levels of wheat imports and to finance consumer price
subsidies. From 1971 to 1975, food aid accounted for only 17% of total
wheat imports into sub-Saharan Africa; by the late 1980s, food aid
made up fully 54% of all wheat imports (table 4).13 In eastern Africa,
virtually all wheat imports now enter as food aid. These figures are
significant, because heavy reliance on food aid often appears to precede increased commercial imports; in countries where wheat is not a
traditional staple, greater amounts of wheat food aid received in the
past are associated with significantly higher levels of commercial wheat
imports today.14
744
Economic Development and Cultural Change
Policy Reforms Discouraging Wheat Consumption
Until very recently, these four broad groups of policies, together with
rapid urbanization of nearly 7% per year, were influential in promoting
steady increases in wheat consumption throughout much of sub-Saharan Africa. However, the situation now appears to be changing. Following the wave of structural reforms implemented during the 1980s,
many policy distortions now have been reduced or eliminated, and
on balance economic incentives have shifted against imported foods,
especially wheat.
At the macroeconomic level, exchange rates have been devalued
sharply in real terms in many countries, including some of Africa's
leading wheat importers. From 1984 to 1987, the real effective exchange rate index for all of sub-Saharan Africa fell by 60% (15% when
Nigeria is excluded). 15 In individual countries, the effect on the relative
prices of imported versus locally produced foods often has been dramatic. For example, in Nigeria, where the currency was devalued
sharply in 1988, imported goods now cost six times more relative to
local goods than they did in 1984. After falling in real terms for a
decade or more, bread prices in Nigeria have doubled since 1986 relative to prices of local foods. 16 Similar increases in the relative price of
bread in the wake of exchange rate adjustments have occurred in
Ghana and Tanzania.
Recent changes in trade policies also have discouraged wheat consumption. A number of countries have imposed ceilings on wheat imports, either by introducing physical quotas (e.g., Sudan) or by limiting
the amount offoreign exchange allocated to commercial imports (e.g.,
Zimbabwe). Nigeria has taken the most drastic step of banning wheat
imports entirely. The Nigerian ban has been only partially effective,
however, as wheat flour continues to be smuggled in from surrounding
countries. 17
Food policies favoring wheat consumption also have been revised.
Wheat consumer subsidies have been cut in many countries, exerting
upward pressure on wheat prices. There also has been some switch in
food aid toward rice, sorghum, and maize; today only a little over half
of food aid to Africa is in the form of wheat, compared to 75% for all
cereal food aid supplied to developing countries. In addition, many
countries are now implementing programs to increase the flour extraction rate from imported wheat and to mix wheat flour with that of local
staples for making bread. IS
These policy changes appear to be working. Total wheat consumption in sub-Saharan Africa fell from a high of 6.6 million tons in 1985
to 5.1 million tons in 1990, resulting in a dramatic 38% decline in
consumption per capita over 5 years. 19 During the same period, annual
wheat imports decreased by nearly half, falling from 5.6 million tons
to 2.9 million tons (much of this decline resulted from the total ban on
Michael
Michael L.
L. Morris
Morris and
and Derek
Derek Byerlee
Byerlee
745
745
wheat
wheat imports
imports enacted
enacted by
by Nigeria
Nigeria in
in 1985).
1985). Given
Given what
what appears
appears to
to be
be
a growing determination on the part of many governments to decrease
a growing determination on the part of many governments to decrease
reliance on imported cereals, the policy reforms will probably conreliance on imported cereals, the policy reforms will probably continue, and wheat consumption growth is likely to decelerate in the
tinue, and wheat consumption growth is likely to decelerate in the
1990s. The successes achieved thus far in reversing the long-standing
199Os. The successes achieved thus far in reversing the long-standing
trend
trend toward
toward increased
increased wheat
wheat consumption
consumption demonstrate
demonstrate that
that any
any comcomprehensive approach to reducing dependence on wheat imports must
prehensive approach to reducing dependence on wheat imports must
give
give at
at least
least equal
equal weight
weight to
to policies
policies that
that manage
manage wheat
wheat consumption,
consumption,
as
compared
to
policies
that
promote
wheat
production.
as compared to policies that promote wheat production.
The Potential for Wheat Production in Sub-Saharan Africa
The Potential for Wheat Production in Sub-Saharan Africa
While
While itit is
is encouraging
encouraging to
to see
see many
many African
African governments
governments taking
taking conconcrete
steps
designed
to
slow
wheat
consumption
growth,
crete steps designed to slow wheat consumption growth, the
the supply
supply
side of the wheat balance equation also merits serious consideration.
side of the wheat balance equation also merits serious consideration.
Evaluating the potential of domestic wheat production is a complicated
Evaluating the potential of domestic wheat production is a complicated
undertaking involving both technical dimensions (can the crop be
undertaking involving both technical dimensions (can the crop be
grown?) and economic dimensions (at what cost?). Unfortunately, in
grown?) and economic dimensions (at what cost?). Unfortunately, in
the past attention all too often has focused on narrow technical issues
the past attention all too often has focused on narrow technical issues
at the expense of broader economic considerations that in the long run
at the expense of broader economic considerations that in the long run
may
may prove
prove more
more decisive.
decisive.
Although only a small proportion of sub-Saharan Mrica is agrocliAlthough only a small proportion of sub-Saharan Africa is agroclimatically suited to wheat production, the total potential wheat-growing
matically suited to wheat production, the total potential wheat-growing
area
20 million
million hectares,
hectares, according
according to
to Food
Food and
and Agriculture
Agriculture
area (around
(around 20
is
many
times
larger
than
Organization
[FAO]
estimates)20
Organization [FAD] estimates)20 is many times larger than the
the current
current
area sown to wheat (slightly over 1 million hectares). This potential
area sown to wheat (slightly over 1 million hectares). This potential
wheat-growing area is characterized by a high degree of diversity.
wheat-growing area is characterized by a high degree of diversity.
Africa's
Africa's wheat-growing
wheat-growing environments
environments can
can be
be classified
classified into
into four
four major
major
types: (1) highland cool, wet areas, (2) midaltitude cool, dry areas, (3)
types: (1) highland cool, wet areas, (2) midaltitude cool, dry areas, (3)
21
midaltitude
midaltitude warm,
warm, humid
humid areas,
areas, and
and (4)
(4) low-altitude
low-altitude warm,
warm, dry
dry areas.
areas. 21
Table
Table 55 summarizes
summarizes the
the location
location of
of these
these environments,
environments, their
their relative
relative
sizes,
and
the
major
technical
and
economic
constraints
to
sizes, and the major technical and economic constraints to production.
production.
Technical constraints, especially diseases, are important in the wetter
Technical constraints, especially diseases, are important in the wetter
environments, while in the dry environments the main constraints tend
environments, while in the dry environments the main constraints tend
to
economic-for example,
example, the
the high
high cost
cost of
of irrigation,
irrigation, lack
lack of
of approapproto be
be economic-for
priate production technology, and competition from alternate crop and
priate production technology, and competition from alternate crop and
livestock enterprises. These economic constraints must be analyzed
livestock enterprises. These economic constraints must be analyzed
for each major environment in order to formulate a comprehensive
for each major environment in order to formulate a comprehensive
strategy for wheat research and development in Africa.
strategy for wheat research and development in Africa.
At
of promoting
promoting aa
At the
the individual
individual country
country level,
level, the
the desirability
desirability of
domestic
wheat
industry
depends
fundamentally
on
the
efficiency
of
domestic wheat industry depends fundamentally on the efficiency of
current (or potential) wheat production, since even noneconomic goals
current (or potential) wheat production, since even noneconomic goals
sometimes invoked in justifying development of local wheat producsometimes invoked in justifying development of local wheat production
tion capacity
capacity (e.g.,
(e.g., increasing
increasing national
national food
food self-sufficiency,
self-sufficiency, improving
improving
of
certain
disadvantaged
groups
the
welfare
of the
the population)
population) must
must
the welfare of certain disadvantaged groups of
take into account the economic dimensions of policy alternatives. The
take into account the economic dimensions of policy alternatives. The
~
0'1
130
IN SUB-SAHARAN AFRICA
Economic
Competing crop and livestock enterprises
Cost of irrigation, competing uses
of scarce water
Competition from high-value crops,
low wheat yields
Cost of irrigation, competing uses
of scarce water
Technical
Disease (Puccinia striiformis, Septoria tritici), weeds, soil acidity,
poor drainage
Disease (P. graminis, P. recondita)
Disease (Helminthosporium), soil
acidity
Heat stress, disease (P. recondita)
COUNTRIES
Burundi
Ethiopia
Kenya
Rwanda
Tanzania
Uganda
Zaire
Madagascar
Malawi
Zambia
Zimbabwe
Madagascar
Malawi
Zambia
Ethiopia
Malawi
Nigeria
Sudan
MAJOR PRODUCTION CONSTRAINTS
('ON~TRAINTS
SOURCE.-D. G. Tanner and M. Van Ginkel, "A Review of the Major Constraints to Wheat Production in Eastern, Central, and Southern Africa
and the Indian Ocean," in Proceedings of the Fifth Regional Wheat Workshop, Madagascar, 5-10 October, 1987, ed. M. Van Ginkel and D. Tanner
(Mexico City: CIMMYT, 1988).
Low altitude hot, dry
3
54
Midaltitude cool, dry
Midaltitude warm, humid
900
Highland cool, wet
AGROCLlMATIC ENVIRONMENT
WHEAT AREA
(Thousands of Ha)
MAJOR WHEAT PRODUCTION ENVIRONMENTS AND PRODUrTTON
TABLE 5
Michael L. Morris and Derek Byerlee
747
efficiency of wheat production usually is assessed by comparing the
economic returns that can be earned by growing wheat with the economic returns from alternative crop and livestock enterprises. While
theoretically simple, measuring efficiency tends to present practical
difficulties, since economic shadow prices must be estimated for production inputs and outputs to eliminate the effects of policy-induced
distortions affecting market prices.
One approach that has proven useful for assessing the relative
efficiency of production alternatives is domestic resource cost (DRC)
analysis. Like many conventional approaches, DRC analysis assigns
economic prices (import or export parity prices) to tradable outputs
and inputs. However, the DRC framework differs from most other
approaches in that it also assigns economic prices (opportunity cost
prices) to nontradable domestic resources, such as land, labor, capital,
and water. Because opportunity cost prices by definition reflect the
scarcity value of domestic resources (i.e., the value in their most
profitable alternate use), the profitability rankings generated by DRC
analysis provide direct empirical measures of relative efficiency among
production alternatives. 22
A number of recent studies have used DRC methods to examine
the economics of wheat production in selected African countries. 23 In
all of these studies, market prices (financial prices) were used to estimate the private profitability of wheat versus alternative crop and livestock enterprises (i.e., profitability from the point of view of the
farmer). The analysis was then repeated using economic prices to estimate measures of relative social profitability for all of the production
alternatives (i.e., profitability from the point of view of the nation). In
some cases, the studies also examined the efficiency of producing
wheat in different regions within the same country or using different
levels of production technology.
Table 6 summarizes findings from Kenya, Nigeria, Sudan, Tanzania, Zambia, and Zimbabwe. In all six countries, government policies
specifically designed to promote wheat production were found to increase private profitability. Typically these policies included producer
price supports for wheat, subsidies on fertilizer and crop chemicals
used in wheat production, and public-sector investment in irrigation.
In addition, most of the countries were found to have an overvalued
exchange rate, which while not targeted specifically at wheat nevertheless had the effect of significantly reducing the local cost of imported
inputs used in wheat production (e.g., irrigation equipment, tractors,
combine harvesters, fuel, lubricants).
But if wheat in sub-Saharan Africa is frequently a profitable crop
for farmers, supporting a domestic wheat industry may entail a considerable cost for the country. In the six country studies cited above,
when inputs and outputs were assigned opportunity cost prices re-
00
~
-..l
Subsidized credit ( + ), overvalued
exchange rate ( + )
Input subsidies ( + ), overvalued
exchange rate (+),
Input subsidies ( + ), official wage
rates (-)
Input subsidies (+), overvalued
exchange rate ( + )
Irrigation subsidies ( + ), overvalued
exchange rate ( + )
Irrigation subsidies ( + ), controlled
producer price ( + )
Overvalued exchange rate (+),
controlled producer price ( + )
Government Policies
Affecting
Private Profitability*
Irrigation costs, opportunity cost of water
Opportunity cost of labor
Positive
Variable
Yield level, irrigation costs
Consumption point
Opportunity cost of water, yield level
Irrigation costs, yield level
Field size
Critical Factors
Affecting
Social Profitability
Positive
Variable
Variable
Negative
Positive
Social
Profitability
SOURCEs.-For Kenya, J. Longmire and J. Lugogo, "The Economics of Small-Scale Wheat Production Technologies for Kenya," Working
Paper no. 01/89 (CIMMYT Economics Program, Mexico City, 1989); for Nigeria, J. D. Strycker and L. E. Brandt, "Domestic Resource Cost
Analysis," annex 6 in Nigeria Agricultural Sector Review (Washington, D.C.: World Bank, 1987); for Sudan, A. Saltar, Study of Cost of Production
and Comparative Advantage of Crops under Different Farming Systems in Sudan (Khartoum: Ministry of Finance and Economic Planning, UNDPI
IBRD Planning Assistance and Training Project, May 1982); for Tanzania, R. M. A. Loyns, C. Carter, G. Harris, and L. Loyns,· "Final Report of
the Benefit-Cost Study Team on the Tanzania-Canada Wheat Project" (submitted to the CIDA, 1986); for Zambia, D. J. Clements, "A Preliminary
Economic Analysis of Rainfed Wheat Production in Zambia" (subInitted to the CIDA, 1988); for Zimbabwe, M. L. Morris, "Comparative Advantage
and Policy Incentives for Wheat Production in Zimbabwe," Working Paper no. 02/88 (CIMMYT EconoInics Program, Mexico City, 1988).
* (+) = government policy increases private profitability; (-) = government policy decreases private profitability.
Positive
Mechanized
Zimbabwe:
Irrigated
Negative
Positive
Tanzania:
Rainfed
Positive
Animal draft
Mechanized
Sudan:
Irrigated
Negative
Rainfed
Mechanized
Nigeria:
Irrigated
Positive
Positive
Mechanized
Kenya:
Rainfed
Mechanized
Mechanized
Country
Private
Profitability
Zambia:
Irrigated
Power
Source
FACTORS INFLUENCING PRiVATE AND SOCIAL PROFITABILITY OF WHEAT PRODUCTION IN SIX AFRICAN COUNTRIES
TABLE 6
Michael L. Morris and Derek Byerlee
749
fleeting their true economic scarcity value, and when the real equilibrium exchange rate was substituted for the nominal exchange rate,
wheat production generally turned out to be unprofitable, except under
certain conditions (table 6). Although space limitations preclude a detailed review of all of the empirical findings, common elements can be
discerned in the conclusions. In particular, six key factors were found
to be influential in determining the economic efficiency of wheat production in Mrica.
1. Place of Wheat in the Farming System
The economic efficiency of wheat production often depends on where
wheat fits into the overall farming system, since the opportunity cost
of resources used in wheat production tends to be sensitive to competition from alternative crop and livestock enterprises. In irrigated zones,
if wheat is grown during months when it does not compete for land
and water with other crops, the social opportunity cost of these resources is low, and wheat production can be efficient (e.g., Zambia
and Zimbabwe). But if wheat is grown during months when it must
compete with export crops for land and water, the social opportunity
cost of these resources is high, and wheat production will be inefficient
unless extremely high yields are achieved (e.g., Sudan). Wheat being
a relatively low-value crop, it will usually not be able to compete
directly with cotton, tobacco, soybeans, and other high-value crops.
However, in irrigated areas wheat sometimes can be introduced
profitably in the rotation as an off-season crop, since this permits intensification of the farming system and leads to a reduction in the fixed
costs associated with irrigation infrastructure.
In rainfed zones, the social opportunity cost of land and water
resources is generally much lower, and wheat production may be efficient at more modest yields. However, when valuing land in particular,
it is important that noneconomic dimensions of opportunity cost be
taken into account. Since wheat is a nontraditional crop in much of
sub-Saharan Africa that can be grown in zones of relatively limited
rainfall, wheat promotion schemes sometimes have attempted to introduce wheat into areas that formerly were not cultivated. In Kenya and
Tanzania, failure to recognize that these areas previously were being
utilized by indigenous peoples as grazing land or hunting areas has led
to unanticipated conflicts, especially with nomadic pastoralists. 24
2. Cost of Irrigation
Many countries in sub-Saharan Mrica can grow wheat only by using
irrigation. But is it cost-effective to invest in irrigation infrastructure
for producing wheat? Empirical evidence from a number of Mrican
irrigation schemes suggests that even if wheat can be introduced in a
way so that it does not compete directly with alternative crops for
750
Economic Development and Cultural Change
resources, irrigation costs still may be a decisive factor determining
profitability. Much depends on the scale ofthe irrigation infrastructure,
as well as on the source of power used to lift or divert water.
Primarily to encourage domestic wheat production, large-scale irrigation systems have been constructed in Nigeria at a cost conservatively estimated at over US$20,000 per hectare. 25 Similar per hectare
investment costs for large-scale irrigation systems have been reported
in Niger, Mauritania, Senegal, and Sudan. 26 Assuming a real cost of
capital of 10% per year, in order to pay for themselves such systems
would have to generate net returns of at least US$2,000 per hectare
per year, or US$l ,000 per hectare per cycle if double-cropping is possible. Even under the best of circumstances, wheat is unlikely to generate net returns high enough to recover an initial investment of this
magnitude, meaning that it usually will not be cost-effective to construct irrigation systems primarily for wheat. It is not surprising that
the World Bank estimates that construction of large-scale irrigation
schemes to produce wheat in Nigeria is very unprofitable at present
costs and yields. 27 Similar conclusions have been reached regarding
large-scale irrigation projects in Mali, Niger, Senegal, Sudan, and elsewhere. 28
Although the high fixed cost of constructing irrigation infrastructure often is considered the primary obstacle to irrigated wheat production in Africa, another important cost component can be the variable
cost of the power needed to lift water. In Malawi, the diesel fuel alone
required for pumping water is estimated to cost the equivalent of about
1 ton per hectare of wheat. 29 In view of the high cost of power throughout most of sub-Saharan Africa, it is not surprising that the cost of
producing wheat in gravity-fed schemes is considerably lower than in
schemes that depend on pumps to lift water.
If large-scale pump irrigation schemes have for the most part
proved uneconomical, the experience of Zimbabwe suggests that less
costly irrigation technologies may offer improved prospects for efficient wheat production. Small-scale pump irrigation systems have been
constructed in Zimbabwe at a cost of about US$2,500-$5,000 per hectare. 30 In addition to being used during the summer cycle to supplement
natural rainfall on tobacco, cotton, maize, and soybeans, these systems are used during the dry winter months to grow wheat under full
irrigation. Because of the relatively low cost of these irrigation systems, Zimbabwe's wheat farmers have been able to make wheat a
profitable off-season crop (except in drought years when water is
scarce).
One final point should be made concerning the economics of irrigated wheat production. Where an initial capital investment in irrigation infrastructure already has been made (e.g., Mali, Nigeria, Senegal,
Sudan), there is a need to ensure its effective utilization, and to this
Michael L. Morris and Derek Byerlee
751
end a cropping pattern including wheat may be appropriate. For this
reason, it is important to distinguish between situations in which serviceable irrigation infrastructure is already in place, in which case
capital investment costs can be considered sunk costs, and situations
in which irrigation infrastructure has yet to be constructed (or has
been constructed but requires extensive rehabilitation), in which case
capital investment costs must be included as a fixed cost. Sudan and
Nigeria are examples of countries in which large-scale irrigation systems already have been constructed at great expense; although it
would be difficult to justify constructing additional systems of the same
type, efficient utilization of the infrastructure that is already in place
requires an intensified cropping system that may include wheat.
•
3. Choice of Production Technology
Wheat in sub-Saharan Africa is produced under extremes of technology ranging from labor-intensive methods requiring over 100 person
days per hectare (e.g., the small-scale systems of Burundi, Ethiopia,
Malawi, Rwanda, southern Tanzania, and Zaire) to highly mechanized
methods using less than 20 person days per hectare (e.g., the large
farms of Ethiopia, Nigeria, Sudan, northern Tanzania, Zambia, and
Zimbabwe). A critical issue in many countries involves selecting a
production technology embodying factor proportions that will maximize returns to scarce capital resources. At current factor prices, the
relatively capital-intensive technologies that characterize many existing wheat production schemes are not cost-effective unless extremely high yields are obtained. In large-scale mechanized production
schemes, the total foreign exchange cost of imported inputs (e.g., irrigation equipment, farm machinery, fertilizer, crop chemicals) frequently exceeds 50% of the foreign exchange savings realized by substituting locally produced wheat for imported wheat. Largely for this
reason, the Canadian International Development Agency (CIDA),
which in the past invested heavily in mechanized farms for wheat
production in Tanzania and Zambia, is now giving more emphasis to
production schemes based on labor-intensive technologies that appear
to be more efficient. 31
The question of optimal choice of production technology cannot
easily be separated from the question of scale. This has become particularly evident in a number of countries during efforts to implement
land reform. In the years following independence, some governments
in eastern and southern Africa implemented policies designed to break
up large farms (many of them former settler estates) and redistribute
the land to smallholders. Land redistribution often led to abandonment
of large-scale mechanized production technologies in favor of smallscale manual technologies, raising the question whether production
efficiency was being affected. J. Longmire and J. Lugogo examined
752
Economic Development and Cultural Change
the efficiency of alternative wheat production technologies in Kenya
under a range of field sizesY Labor-intensive production technologies
were found to be efficient on small fields (0.5 hectare and I hectare),
whereas mechanized production technologies were found to be efficient on large landholdings (over 4 hectares). These results suggest
that labor-intensive production technologies may be appropriate for
small-scale producers with restricted access to land, but that in the
absence of constraints on farm size, wheat production might remain
more efficient on larger landholdings where high levels of mechanization are feasible. 33
4. Wheat Yield Levels
In many African countries, wheat yields must be high in order for
wheat production to be efficient. This is particularly true for irrigated
wheat, given the considerable cost of installing irrigation infrastructure. Although the minimum yield level required to ensure the efficiency of wheat production varies from country to country, evidence
from several studies suggests that irrigated yields must reach 3-5 tons
per hectare for wheat to pay back irrigation costs and generate positive
social returns. For rainfed wheat, the threshold yield level is lower,
probably in the range of 2 tons per hectare in most cases.
Irrigated wheat yields of 3-5 tons per hectare are certainly not
beyond the realm of feasibility under high levels of management. In
central Zimbabwe, where temperatures during the growing season are
generally favorable for wheat, yields frequently exceed 10 tons per
hectare on experiment stations. 34 In Sudan, where temperatures during
the growing season are much hotter, yields of up to 5 tons per hectare
have been reported in demonstration plotsY However, achieving these
high yield levels in farmers' fields is more problematic. In many wheatgrowing zones in Africa, particularly hot zones where temperatures
exceed 35° during critical stages in the crop growth cycle (e.g., Sudan,
Nigeria), wheat yields are extremely sensitive to management practices, especially planting date, fertilization levels, and irrigation scheduling. Since wheat in these zones is a relatively new crop with no long
production history, many farmers still are learning how to manage the
crop. Farmers' lack of management experience is reflected in low and
variable yields. In one rainfed wheat project in Cameroon, yields averaged only 60 kilograms per hectare during the initial cropping season. 36
In Sudan's Gezira irrigation scheme, farmers' wheat yields currently
average less than 1.5 tons per hectare even after 10 or more years of
experience growing the crop.
Of course, low yields cannot be attributed entirely to farmers'
lack of experience. Because of deficient input distribution systems,
wheat farmers in many parts of Africa lack timely access to improved
seed, fertilizer, and other purchased inputs. Furthermore, those who
Michael L. Morris and Derek Byerlee
•
753
rely on irrigation water provided through large-scale, centralized irrigation schemes frequently experience disruptions in water supplies. The
fact that these constraints essentially are beyond the control of farmers
themselves has important implications for likely future growth in wheat
yields (and hence for the efficiency of wheat production). Although
farmers can be expected to improve their management skills as they
move along the learning curve, improving input delivery systems and
irrigation management may prove more difficult over the longer run.
5. Presence of Exchange Rate Distortions
Exchange rate distortions drive a wedge between financial and economic profitability of production activities that require large amounts
of imported inputs. Wheat production, especially large-scale mechanized wheat production, is one such activity. Studies of wheat projects
in Nigeria, Sudan, Tanzania, and Zambia have shown that the financial
profitability of large-scale mechanized wheat production can be exaggerated when overvaluation of the domestic currency leads to underestimation of the true economic cost of imported production inputs.J7 In
Nigeria and Sudan, DRC analysis indicates that when foreign exchange
is converted at a real equilibrium (shadow) exchange rate, at current
yields wheat production generates negative value-added for many
farmers (i.e., wheat production uses up more foreign exchange in the
form of imported inputs than it generates in the form of wheat to
substitute for imported wheat).38
Where exchange rate policies have been reformed to eliminate
distortions, the cost of mechanized wheat production often has increased dramatically, as occurred in Nigeria after the devaluation of
1987. Changing relative prices of labor and capital sharply decreased
the profitability of wheat production on large-scale pump-irrigation
schemes and provided strong incentives for small-scale production on
traditional fadama lands, which use less capital-intensive tube well
irrigation. 39
6. Location of Production and Consumption Points
A final key factor determining the efficiency of wheat production is
the location of production and consumption. Because internal transport and handling costs can be extremely high in Africa due to deficient
transportation infrastructure, high fuel costs, and long distances between supply and demand points, the spatial distribution of production
and consumption can drastically affect the relative cost of producing
wheat versus importing it.
The case of Tanzania illustrates this point (table 7). In 1988, the
landed price of imported wheat in the port city of Dar es Salaam was
about US$190 per ton, but shipping this wheat to the inland consumption center of Arusha added another US$60 per ton in transport costs. 40
754
Economic Development and Cultural Change
TABLE 7
EFFECT OF INTERNAL TRANSPORT COSTS ON WHEAT PRICES IN DIFFERENT LocATIONS IN
TANZANIA AND NIGERIA, 1987-1988
Price at
Port City
(Consumption
Point)
(US$ per Ton)
Tanzania:
Imported wheat
Local wheat
Nigeria:
Imported wheat
Local wheat
* C.i.f.
Internal
Transport
Cost
(US$ per Ton)
190*
230
+
210*
390
+
Price Inland
(Production and
Consumption Point)
(US$ per Ton)
60t
60t
+
250
170:1:
1l0§
1l0§
+
320
280 1
import price plus landing costs.
t Internal transport cost based on estimate in N. D. Frank and R. M. A. Loyns, "An
Economic Analysis of Small-Holder and Large-Scale Mechanised Wheat Production in
Northern Tanzania" (report submitted to the Natural Resources Branch, CIDA, Manitoba, March 1989).
:I: Local production cost based on estimate in R. M. A. Loyns, C. Carter, G. Harris,
and L. Loyns, "Final Report of the Benefit-Cost Study Team on the Tanzania-Canada
Wheat Project" (submitted to the CIDA, 1986).
§ Internal transport cost based on estimate in World Bank, Nigeria Agricultural
Sector Review, vol. 2 (Washington, D.C.: World Bank, 1987).
I Local production cost based on estimate in A. O. Ogungbile, "The Relative Priority and Economics of Growing Wheat in Nigeria," in CIMMYT, Wheats for More
Tropical Environments: A Proceedings of the International Symposium, September
24-28, 1984 (Mexico City: CIMMYT, 1985).
Hence the import parity price in Arusha was US$250 per ton (landed
price in Dar es Salaam plus internal transport cost). Assuming that
the cost of producing local wheat in the nearby Hanang region was
approximately US$170 per ton, this means that it was efficient to produce local wheat for consumption in Arusha. However, shipping local
wheat to Dar es Salaam would have added another US$60 per ton in
transport costs, raising its price to US$230 per ton in Dar es Salaam
(production cost plus internal transport cost). Therefore, it was efficient to produce wheat in the Hanang region to substitute for imports
in Arusha, but inefficient to ship the same wheat to consumers in Dar
es Salaam. A similar situation prevailed in Nigeria, where internal
transport and handling costs involved in moving wheat between the
coast and inland production points amounted to fully 50% of the import
parity price in the port city of Lagos (table 7).4'
These examples from Tanzania and Nigeria are not unusual in
Africa, where large distances and poor infrastructure often result in
high internal transport costs that may drastically affect the final retail
price. They illustrate the critical importance of choosing representative
production and consumption points in assigning economic (shadow)
prices to domestic and imported wheat.
Michael L. Morris and Derek Byerlee
755
Domestic Wheat Production and Food Security
,
Policymakers in sub-Saharan Africa are rightly concerned about the
rapid increase in wheat consumption and imports. If an undesirable
dependence on imported food is to be averted, effective policy actions
must be taken to remove policy distortions favoring wheat consumption, and possibly to increase domestic production. However, policies
designed to promote domestic production should be based on sound
economic reasoning, not merely on a desire to increase wheat selfsufficiency levels at any cost.
One example of the questionable policy recommendations that can
result from inattention to economics is the argument commonly made
in favor of supporting domestic wheat industries for food security reasons. Government policymakers in a number of wheat-producing countries in Africa have stated that they are reluctant to depend excessively
on imported wheat, because variability in world wheat prices exposes
consumers (or the state treasury) to sharp fluctuations in foreign exchange outlays.42 On closer examination, the food security argument
is not very convincing. Wheat is a marginal crop throughout most of
sub-Saharan Africa, with production subject to substantial year-toyear fluctuations because of a wide range of technical, economic, and
institutional constraints. Consequently, wheat production in Africa
tends to be highly variable from year to year, as evidenced by coefficients of variation (CV) around trend of 44% for Nigeria, 43% for
Sudan, 29% for Tanzania, and 30% for Zimbabwe during the period
1961-89. These figures compare with a CV around trend of 8%-10%
for production in major wheat-producing countries such as India, Pakistan, and Turkey, and with a CV around trend of 25% for the main
international reference price of wheat over the same period. Since the
world wheat market thus appears more stable than domestic production in Africa, attempts to increase self-sufficiency in wheat may increase rather than decrease variability in food supplies and prices. For
this reason, food security objectives probably should not be invoked
in justifying promotion of a domestic wheat industry.
Conclusions
What are the prospects for narrowing the wheat gap in sub-Saharan
Africa? This article has reviewed policies that in the past have encouraged wheat consumption, and it has described some of the factors that
determine whether or not domestic wheat production is likely to make
sense from an economic point of view.
On the demand side, policymakers in many African countries
seem to be serious about reversing long-standing policies that in the
past contributed to rapid growth in wheat consumption and imports.
Recent policy reforms (e.g., exchange rate devaluation and removal
of subsidies on bread and wheat flour) have succeeded in raising the
756
Economic Development and Cultural Change
real price of bread relative to the prices of other traditional staples,
slowing (and in some cases reversing) growth in wheat consumption.
If the policy reforms can be maintained, there is every reason to believe that the growth in wheat consumption will continue to slow down.
On the supply side, prospects for increasing wheat production are
more uncertain. Recent empirical work suggests that the efficiency of
wheat production in sub-Saharan Mrica depends on a complex set of
climatic, technical, economic, and institutional factors. Given modest
irrigation costs, high yields, and lack of competition for land and water
resources, irrigated wheat production sometimes can be efficient. But
where irrigation costs are high, yields are low, and wheat must compete with alternative crops for scarce land and water resources, the
economic argument in favor of a local wheat industry is not convincing. R. Hassan et al. have illustrated these relationships in figure 2,
which depicts the efficiency of irrigated wheat production in Sudan
under various combinations of wheat yield levels and relative prices.
Under actual farmer yields averaging less than 1.5 tons per hectare
and current world prices for wheat and cotton, irrigated wheat production in Sudan represents an inefficient use of scarce domestic resources. Assuming no change in the prevailing price ratio, a significant
increase in average wheat yields (to over 3 tons per hectare) would
be necessary to make irrigated wheat production socially efficient.
However, even if such a large yield increase were achieved, the efficiency of wheat production still could be undermined by a relatively
modest decrease in the world price of wheat relative to the world price
of cotton.
Under rainfed conditions, where wheat yields are generally low,
large-scale mechanized wheat production does not appear to be an
efficient use of resources, except possibly in Kenya. The challenge in
rainfed areas seems to be in developing appropriate small-scale technologies suitable for farmers with limited access to capital and land.
In both the irrigated and the rainfed cases, the spatial distribution of
production and consumption, along with internal transport costs, can
playa central role in determining the desirability of promoting domestic wheat production versus importing wheat.
It is important to emphasize that much of the evidence on the
economics of wheat production in Mrica remains inconclusive. Recent
studies clearly indicate that wheat production can represent an efficient
use of resources under certain conditions, but production efficiency
depends on numerous climatic, economic, and technical factors. Despite increasing amounts of applied research, policymakers frequently
still lack the information needed to formulate long-term strategies for
investing in wheat research and development. In particular, analyses
of the economics of wheat production often have not compared wheat
with alternative enterprises in terms of meeting the dual objectives of
•
"
-.I
\Jl
-.I
o
o
!
1989 cotton : wheat
price ratio
-----------------
o be4'
0.2
0.4
0.6
3
Wheat yield (t/ha)
2
Inefficient production
(negative value added)
4
r ,- - - - - - - - - - - - - - - - - - - - - - - - , . ,
5
FIG. 2.-Efficiency ranges for irrigated wheat production in Gezira, Sudan, under various
yield levels and relative price ratios, 1989.
u
~
i
Gl
'lii
Co
~
'C;
!!!
:p
0.8
1.2
758
Economic Development and Cultural Change
economic efficiency and food security. Furthermore, economic analyses have not always paid sufficient attention to how technological parameters may influence the efficiency of local wheat production, especially in the presence of exchange rate distortions that may be masking
true economic returns. This type of information is especially important
for establishing wheat research priorities. Hence there is an urgent
need for more in-depth economic analyses of the efficiency, current
and potential, of domestic wheat production to provide guidance to
researchers and policymakers to establish priorities for research and
development for the next 10-20 years.
Notes
* We would like to thank Laura Saad for valuable research assistance.
Helpful comments from an anonymous reviewer are also gratefully acknowledged.
1. W. M. Mwangi, "Sub-Saharan Africa Wheat Imports: Significance,
Extent, Determinants and Future Implications," East African Journal ofRural
Development 15 (1982): 143-63; M. M. Shah, G. Fischer, G. Kromer, and
K. S. Parikh, Africa's Growing Dependence on Imported Wheat: Some Implications for Agricultural Policies in Africa (Luxembourg: International Institute
for Applied Systems Analysis, 1984); M. B. Choudhri, Wheat Production Potential in Africa (Rome: Food and Agricultural Organization, 1987); A. Saltar,
Study of Cost of Production and Comparative Advantage of Crops under
Different Farming Systems in Sudan (Khartoum: Ministry of Finance and
Economic Planning, UNDP/IBRD Planning Assistance and Training Project,
May 1982); J. L. Stone, "Project Evaluation: A Case Study of the CanadaTanzania Wheat Project" (M.S. thesis, University of Guelph, 1982); G. Andrae
and B. Beckman, The Wheat Trap: Bread and Underdevelopment in Nigeria
(London: Zed Books, 1985); A. Blum, "Wheat in Malawi" (Malawi Ministry
of Agriculture, Lilongwe, August 1988); R. M. A. Loyns, C. Carter, G. Harris,
and L. Loyns, "Final Report of the Benefit-Cost Study Team on the TanzaniaCanada Wheat Project" (submitted to the Canadian International Development
Agency [CIDA], 1986); D. J. Clements, "A Preliminary Economic Analysis
of Rainfed Wheat Production in Zambia" (submitted to the CIDA, 1988);
M. L. Morris, "Comparative Advantage and Policy Incentives for Wheat Production in Zimbabwe," Working Paper no. 02/88 (CIMMYT Economics Program, Mexico City, 1988); J. Longmire and J. Lugogo, "The Economics of
Small-Scale Wheat Production Technologies for Kenya," Working Paper no.
01/89 (CIMMYT Economics Program, Mexico City, 1989); G. I. Mlay, G. C.
Ashimogo, A. E. Temu, and R. V. Kurwijila, "Wheat Production and Marketing in Tanzania: A Case Study" (FAO, Rome, 1989); R. Hassan, H. Faki, and
H. Obeid, "Economic Policy and Technology Determinants of the Comparative Advantage of Wheat Production in Sudan," CIMMYT Economics Paper
no. 5 (CIMMYT Economics Program, Mexico City, in press); P. T. Ngobese,
"The Economics of Large-Scale Wheat Production in Zimbabwe" (M.S. thesis, Virginia Polytechnic and State University, 1987).
2. The World Bank currently projects that by the year 2020, half of the
population of sub-Saharan Africa will live in urban areas. See World Bank,
Sub-Saharan Africa: From Crisis to Sustainable Growth (Washington, D.C.:
World Bank, 1989).
3. E. W. Bovill, Caravans ofthe Old Sahara (London: Oxford University
Michael L. Morris and Derek Byerlee
759
Press, 1933), cited in A. J. Youngs, "Wheat Flour and Bread Consumption in
West Africa: A Review with Special Reference to Ghana," Tropical Science
14 (1972): 235-44.
4. P. Kilby, African Enterprise: The Nigerian Bread Industry, Hoover
Institution Studies no. 8 (Stanford, Calif.: Stanford University Press, 1965).
5. Youngs.
6. Ibid.
7. G. C. Shrotriya and M. L. Dixit, "Wheat: New Hope for Nigerian
Harvest," African Farming (January/February 1987), pp. 20-23; M. H. Verlet
and M. J. Hauchecorne, "Wheat Cultivation at Lake Chad" (Bureau pour Ie
Developpement de la Production Agricole, Paris, n.d.); K. Briggs, "Wheat
Production and Development in Kenya," Agriculture and Forestry Bulletin 7
(1984): 27-32; Ngobese; P. L. Raikes, "The Historical Development of Wheat
Production in Northern Mbulu District, Tanzania," Economic Research Bureau Paper no. 70.11 (University of Dar es Salaam, 1970); Republic of Tanzania, Ministry of Agriculture, National Wheat Development Strategy: 19842000 (Dar es Salaam: Ministry of Agriculture and Livestock Development,
1984).
8. Aggregate statistics reported in this article were calculated from the
AGROSTAT-PC data base (diskette version) (Food and Agriculture Organization, Rome, 1991) using the following regional groupings: Sahel-Burkina
Faso, Cape Verde, Chad, Gambia, Mali, Mauritania, Niger, Senegal; western
Africa-Benin, Cote d'Ivoire, Ghana, Guinea, Guinea-Bissau, Liberia, Nigeria, Sierra Leone, Togo; central Africa-Burundi, Cameroon, Central African
Republic, Congo, Equatorial Guinea, Gabon, Rwanda, Sao Tome, St. Helena,
Zaire; eastern Africa-Djibouti, Ethiopia, Kenya, Somalia, Sudan, Uganda;
southern Africa-Angola, Botswana, Comoros, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Reunion, Seychelles, Swaziland, Tanzania, Zambia, Zimbabwe.
9. C. Delgado and C. Miller, "Changing Food Patterns in West Africa:
Implications for Research Policy," Food Policy 10 (February 1985): 55-62;
Andrae and Beckman; T. Reardon, T. Thiombiano, and C. Delgado, "Substitution by Urban Sahelian Consumers between Coarse Grains and Imported
Rice and Wheat: The Case of Ouagadougou" (submitted to the IFPRI Publications Review Committee, June 4, 1992); D. Byerlee, "The Political Economy
of Third World Food Imports: The Case of Wheat," Economic Development
and Cultural Change 35 (January 1987): 307-28; D. Franklin, M. P. Demousin,
and M. W. Haren, Consumption Effects ofAgricultural Policies: Bread Prices
in the Sudan (Raleigh, N.C.: Sigma One, 1982): P. Kilby, "Patterns of Bread
Consumption in Nigeria," Food Research Institute Studies 5 (1965): 3-18;
World Bank, Sub-Saharan Africa; J. Smith and K. O. Makinde, "Cereals
Policy in Nigeria: Banning Cereal Imports" (UTA, Ibadan, 1988).
10. Smith and Makinde.
11. J. S. Bickersteth, "Donor Dilemmas in Food Aid: The Case of Wheat
in Sudan," Food Policy 15 (June 1990): 218-26; Franklin et al.
12. World Bank, Sub-Saharan Africa.
13. Statistics on wheat food aid must be interpreted with caution. Discrepancies frequently arise between offical food aid data and trade data because
food aid flows are reported in marketing years, whereas trade statistics are
reported in calendar years.
14. D. Byerlee, "Wheat in the Tropics: Economic and Policy Issues," in
CIMMYT, Wheats for More Tropical Environments: A Proceedings of the
International Symposium, September 24-28, 1984 (Mexico City: CIMMYT,
1985); D. Byerlee and M. L. Morris, "The Political Economy of Wheat Con-
760
Economic Development and Cultural Change
sumption and Production with Special Reference to Sub-Saharan Africa," in
Southern Africa: Food Security Policy Options. ed. M. Rukuni and R. H.
Bernsten (Harare: UZ/MSU Food Security Project, 1988).
15. World Bank, Sub-Saharan Africa.
16. Smith and Makinde. Bread prices increased relatively less than the
prices of all imported goods because bread contains a domestic component
(e.g., labor for milling and baking) that was not directly affected by devaluation.
17. Ibid.
18. W. Mwangi, D. Rohrbach, and C. Mwila, "Commercial Utilization
of Sorghum and Millet in Zambia: An Approach to Food Security" (paper
presented at the Fifth Annual Conference on Food Security Research in Southern Africa, Harare, October 16-18,1989); B. Dendy and B. Trotter, "A Study
of the Constraints to the Adoption of Composite Flour Technology in Tanzania, Zimbabwe, Zambia, and Botswana," Post Production Systems Newsletter. no. 11 (1989); A. Siegel, Bakery Operation: Maiduguri Flour Mill (Edmonton: International Development Research Centre, 1976); "Prospects for
Increased Millet Consumption in Senegal," Food Outlook 5 (May 1984): 1516; D. Byerlee, "Experiences with Composite Flours in Bread Making in
Developing Countries: A Review of the Evidence" (CIMMYT, Mexico City,
1991).
19. Structural adjustment and policy reforms similarly have reversed the
increasing trend in wheat consumption in Latin America. See D. Byerlee and
G. Sain, "Relative Food Prices under Structural Adjustment: Preliminary
Findings from Latin America," Food Policy (August 1991): 74-84.
20. Choudhri (n. 1 above).
21. D. G. Tanner and M. Van Ginkel, "A Review of the Major Constraints to Wheat Production in Eastern, Central, and Southern Africa and
the Indian Ocean," in Proceedings of the Fifth Regional Wheat Workshop.
Madagascar. 5-10 October. 1987, ed. M. Van Ginkel and D. G. Tanner (Mexico City: CIMMYT, 1988).
22. For a description of the DRC framework of analysis, see E. Monke
and S. Pearson, The Policy Analysis Matrix for Agricultural Development
(Ithaca, N.Y.: Cornell University Press, 1989). See also M. L. Morris, "Determining Comparative Advantage through DRC Analysis: Guidelines Emerging from CIMMYT's Experience," Paper no. 01/90 (CIMMYT Economics
Program, Mexico City, 1990).
23. Clements (n. 1 above); Hassan et al. (n. 1 above); Longmire and
Lugogo (n. 1 above); Loyns et al. (n. 1 above); Morris, "Comparative Advantage and Policy Incentives for Wheat Production in Zimbabwe" (n. 1 above);
J. D. Stryker and L. E. Brandt, "Domestic Resource Cost Analysis," annex 6
in Nigeria Agricultural Sector Review (Washington, D.C.: World Bank, 1987).
24. C. Lane and J. Pretty, Displaced Pastoralists and Transferred Wheat
Technology in Tanzania. Gatekeeper Series, no. SA20 (International Institute
for Environment and Development, London, 1990).
25. Andrae and Beckman (n. 1 above).
26. Food and Agriculture Organization, "Irrigation in Africa South of the
Sahara," in FAO Investment Centre Technical Paper No.5 (Rome: FAO,
1986); Club du Sahel/CILSS, The Development of Irrigated Agriculture in
the Sahel (Ouagadougou: Club du Sahel/CILSS, 1980); S. Barghouti and G.
LeMoigne, "Irrigation in Sub-Saharan Africa: The Development of Public and
Private Systems," Technical Paper no. 123 (World Bank, Washington, D.C.,
1990); World Bank, Accelerated Development for Sub-Saharan Africa: An
Agenda for Action (Washington, D.C.: World Bank, 1981).
Michael L. Morris and Derek Byerlee
,
•
•
761
27. Strycker and Brandt.
28. C. Christiansen et al., Food Problems and Prospects in Sub-Saharan
Africa: The Decade of the 1980s, Foreign Agricultural Research Report no.
166 (Washington, D.C.: U.S. Department of Agriculture Economic Research
Service, 1981); World Bank, Accelerated Development for Sub-Saharan Africa; D. Byerlee and G. Varughese, "The Potential for Commercial Wheat
Production in Senegal: Some Issues" (CIMMYT, Mexico City, 1981).
29. Government of Malawi, "Wheat Production Development in Malawi"
(Ministry of Agriculture Planning Division, Lilongwe, 1982).
30. Morris, "Comparative Advantage and Policy Incentives for Wheat
Production in Zimbabwe."
31. Loyns et al. (n. 1 above); Clements (n. 1 above).
32. Longmire and Lugogo (n. 1 above).
33. In addition to examining the relative efficiency of alternative wheat
production technologies, Longmire and Lugogo also assessed the relative efficiency of wheat and maize. At prevailing 1987 factor prices, maize represented
the socially most efficient use of resources on small farms «4 hectares).
34. Agricultural Research Trust Annual Report (Harare: Agricultural Research Trust, various editions).
35. Hassan et al. (n. 1 above).
36. S. Franzel, personal communication (1989).
37. Andrae and Beckman (n. 1 above); Hassan et al.; Loyns et al.;
Clements.
38. Strycker and Brandt (n. 23 above); Hassan et al.
39. L. B. Olugbemi, "Major Constraints and Remedies to Wheat Production in Nigeria" (invited paper presented at the First National Conference on
Wheat Production, Processing, and Utilization in Nigeria, Maiduguri, January
29-February 2, 1990).
40. N. D. Frank and R. M. A. Loyns, "An Economic Analysis of SmallHolder and Large-Scale Mechanised Wheat Production in Northern Tanzania"
(report submitted to the Natural Resources Branch, Canadian International
Development Agency, Manitoba, March 1989).
41. World Bank, Nigeria Agricultural Sector Review, vol. 2 (Washington,
D.C.: World Bank, 1987).
42. C. K. amari, "Politics and Policies of Food Self-sufficiency in Tanzania," Social Science and Medicine 22 (1986): 769-82; Republic of Kenya,
"Economic Management for Renewed Growth," Sessional Paper no. 1 (Government Printer, Nairobi, 1986).