Waiting For Rain Agriculture And Ecological Imbalance In Cape Verde Mark Langworthy Timothy J Finan

Waiting For Rain Agriculture And Ecological
Imbalance In Cape Verde Mark Langworthy Timothy
J Finan download
https://ebookbell.com/product/waiting-for-rain-agriculture-and-
ecological-imbalance-in-cape-verde-mark-langworthy-timothy-j-
finan-51894298
Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.
Waiting For Rain Reflections At The Turning Of The Year 1st Edition
Bryna Jocheved Levy
https://ebookbell.com/product/waiting-for-rain-reflections-at-the-
turning-of-the-year-1st-edition-bryna-jocheved-levy-51369562
Waiting For Rain Reflections At The Turning Of The Year 1st Levy
https://ebookbell.com/product/waiting-for-rain-reflections-at-the-
turning-of-the-year-1st-levy-36426652
While Waiting For Rain Community Economy And Law In A Time Of Change
John Henry Schlegel
https://ebookbell.com/product/while-waiting-for-rain-community-
economy-and-law-in-a-time-of-change-john-henry-schlegel-47134768
Waiting For Raine Layla Dorine
https://ebookbell.com/product/waiting-for-raine-layla-dorine-49087608

Waiting For Raine Layla Dorine
https://ebookbell.com/product/waiting-for-raine-layla-dorine-51871894
Waiting For Raine Layla Dorine
https://ebookbell.com/product/waiting-for-raine-layla-dorine-51409350
Waiting For Raine Layla Dorine
https://ebookbell.com/product/waiting-for-raine-layla-dorine-58021514
Advanced Manual For Communication Laboratories And Technical Report
Writing For Wbut D Sudha Rani
https://ebookbell.com/product/advanced-manual-for-communication-
laboratories-and-technical-report-writing-for-wbut-d-sudha-
rani-10878812
Waiting For The Revolution Evan Smith Matthew Worley
https://ebookbell.com/product/waiting-for-the-revolution-evan-smith-
matthew-worley-45210380

Waiting for Rain

WaitingforRain
Agriculture and
Ecological Imbalance in
CapeVerde
Mark Langworthy
Timothy J. Finan
boulder
london

Published in the United States of America by
Lynne Rienner Publishers, Inc.
1800 30th Street, Boulder,Colorado 80301
www.rienner.com
and in the United Kingdom by
Lynne Rienner Publishers, Inc.
3Henrietta Street, Covent Garden, London WC2E 8LU
©1997 by Lynne Rienner Publishers, Inc. All rights reserved
Libraryof Congress Cataloging-in-Publication Data
Langworthy,Mark.
Waiting for rain : agriculture and ecological imbalance in Cape
Verde / Mark Langworthy and Timothy J. Finan.
Includes bibliographical references and index.
ISBN 978-1-55587-709-5 (hc : alk. paper)
1. Agriculture—Cape Verde.  2. Agriculture—Environmental
aspects—Cape Verde.  3. Natural resources—Cape Verde.
4. Agriculture and state—Cape Verde.  5. Food supply—Cape Verde.
6. Cape Verde—Rural conditions.  I. Finan, Timothy J.  II. Title.
S473.6.C25L36   1997
338.1'4'096658—dc21 96-44538
CIP
British Cataloguing in Publication Data
A Cataloguing in Publication record for this book
is available from the British Library.
Printed and bound in the United States of America
The paper used in this publication meets the requirements
of the American National Standard for Permanence of
Paper for Printed Library Materials Z39.48-1992.

Para o povo caboverdiano tão sofrido, tão amável

List of Tables and Figures ix
Acknowledgments xi
Map of Cape Verde xii
1Sustainability in a Fragile Environment: An Introduction 1
2Public and Private Resource Management:
A Methodological Approach 13
3 The Natural Resource Endowment 35
4 Characteristics of Cape Verdean Rural Society 53
5 Rainfed Agriculture 79
6 Irrigated Agriculture 101
7 Household Management Strategies 131
8 Agriculture and Policy Choice 159
9 Conclusions: The Future of Agriculture in Cape Verde 187
References 197
Index 204
About the Book 212
Contents
vii

Tables
3.1 Annual Precipitation on the Four Major Agricultural
Islands, 1970–1988 39
3.2 Groundwater Availabilities and Use on the Four
Major Agricultural Islands 43
3.3 Land Quality Characteristics for the Four Major
Agricultural Islands 46
4.1 Historical Record of Droughts, 1719–1996 61
4.2 Demographic Characteristics of Rural Population
on Selected Islands 62
4.3 Distributions of Rainfed Cropland on Santiago and
Santo Antão 70
4.4 Distributions of Irrigated Cropland on Santiago and
Santo Antão 70
4.5 Distributions of Rainfed Cropland by Landownership
on Santiago and Santo Antão 72
4.6 Distributions of Irrigated Cropland by Landownership
on Santiago and Santo Antão 72
5.1 Comparative Estimates of Cultivated Rainfed Land
on Selected Islands, 1978 and 1988 80
5.2 Yields of Corn and Beans, 1975–1990 84
5.3 Characteristics of Land and Labor Use in Rainfed
Agriculture on Santiago and Santo Antão 86
5.4 Representative Budgets for Rainfed Crops 90
5.5 Outputs and Net Values of Livestock 92
ix
Tables and Figures

5.6 Economic Returns of Representative Rainfed Systems
on Santiago and Santo Antão 93
5.7 Annual Net Returns of Livestock 97
6.1 Distribution of Irrigated Land on Santiago
and Santo Antão 102
6.2 Irrigated Land by Water Source 106
6.3 Distribution of Average Irrigation Intervals 108
6.4 Various Estimated Shares of Irrigated Area by Crop 109
6.5 Production of Major Irrigated Crops, 1971–1990 114
6.6 Crop Budgets for Major Irrigated Crops 120
6.7 Characteristics of Representative Farm Systems 127
7.1 Distribution of Rainfed and Irrigated Land by
Number of Adults in Household 134
7.2 Rainfed and Irrigated Landholdings 135
7.3 Average Household Labor Requirements and
Availability, Annual and Peak Labor Season 137
7.4 Regression Models of Household Labor Allocation 138
7.5 Characteristics of Irrigated Farms on Santiago by
Source of Water 143
7.6 Distribution of Irrigation Intervals Among Farmers
Within a Surveyed Irrigation Network, Santiago 144
7.7 Household Income on Santiago and Santo Antão
by Quintile 149
7.8 Characteristics of Surveyed Households on Santiago
and Santo Antão by Per Capita Income Quintile 152
8.1 Cropping Patterns and Net Revenues for Irrigation
Network Under Alternative Intervals 174
8.2 Simulations of Potential Increases in Agricultural
Incomes of Surveyed Households 177
Figures
1.1 Population and Agricultural Production Capacity
over Time 7
2.1 The Political Ecology of Cape Verdean Agriculture 15
7.1 Household Labor Demands and Supply 155
x Tables and Figures

This project has been a long labor of love, as endearing as it was enduring.
Over the dozen years of research and travel that generated this work, we
have benefited from the assistance and wisdom of many people who guided
us over unknown waters, maintained saintly patience in the face of our
disingenuousness, and provided the emotional support necessary to bring
this endeavor to its fruition.
In particular,we acknowledge Jimmy S. Hillman, project director of
the Foods Crops Research Project and retired head of the Department of
Agricultural Economics, and Carlos Véléz-Ibañéz, former director of the
Bureau of Applied Research in Anthropology (BARA) at the University of
Arizona. Both provided us with the motive and opportunity to finish this
long-term project. Of the many Cape Verdeans who so kindly shaped our ex-
perience, we offer special recognition to Horácio Soares, Elísio Rodrígues,
Milu Lobo, António Sabino, and other professionals who have dedicated
their lives to the challenge of rural development. In our research effort we
benefited from the friendship and knowledge of our manu Raul Varela
Semedo, Tito Andrade, and Vladmiro Andrade; we shall forever cherish
those exciting times spent in endless fieldwork. José and Zuleika Levy and
their respective families, especially Dona Milu, provided professional sup-
port and friendship.
In every rural community we visited over the years, local Cape
Verdeans took us in, shared their lives, and made us aware of their awe-
some strength in the face of adversity. Their friendship will always be the
richest reward afforded us. At the University of Arizona we received un-
erring assistance from our respective staffs—Connie McKay, Loretta Cos-
grove, Linda Phipps, and Maria Rodriguez. From our hearths we received
the quiet forbearance, encouragement, and warmth of Pat and Amélia. This
book is as much their accomplishment as ours.
—M. L.
—T.J. F.
xi
Acknowledgments

The Republic of Cape Verde, an Atlantic archipelago off the Guinea
coast of Africa, is small in both magnitude and population. Located 453
nautical miles due west of Senegal, the combined landmass of the 10 is-
lands and 2 islets amounts to around 4,000 square kilometers. The six is-
lands to the North (Santo Antão, São Vicente, Santa Luzia, São Nicolau,
Sal, and Boa Vista) constitute the windward, or barlavento, islands,
while the four southern islands (Brava, Fogo, Santiago, and Maio) make
up the leeward, or sotavento, islands. The population of the country is
currently estimated at about 380,000, and, as such, its population density
(95/km
2)is among the highest of West African nations (World Bank
1993b). Although it is perhaps a minor participant in the overall flow of
world economic and political affairs, Cape Verde nonetheless shares
many defining ecological characteristics with its Sahelian neighbors and
faces very similar challenges of resource management. The majority of
its population resides in rural areas and derives its livelihood, at least in
part, from agricultural activities. Andlike most of the Sahel, Cape Verde
is arid, its farming subject to the vagaries of highly unpredictable and
often miserly weather systems. Because of the common ecology and
shared history that tie the archipelago to the continent, an understanding
of Cape Verdean economy and society provides insights of a broader re-
gional significance.
Most evidence indicates that the islands were uninhabited until Por-
tuguese explorers, venturing south from Madeira, arrived in the late 1400s
and claimed them for the Crown. The motives for subsequent settlement were
founded more on the central geographic location of the islands than on the
richness of the local resource base. The proximity of Cape Verde to the
Guinea coast and its far vaster and more diversified set of resources provided
the Crown a dominant position in intercontinental trade and other commercial
advantages related to centralized warehousing and transshipment facilities.
1
1
Sustainability in a
Fragile Environment:
An Introduction

As a consequence, Cape Verde soon developed a dominant position in the
Guinea slave trade to Europe and the New World.
Agriculture, conversely, never emerged as a viable economic alterna-
tive. The insular colony failed to meet the expansionist objective of the
Crown, that of providing abundant agricultural land for Portugal’s impov-
erished peasantry—as Mozambique and Angola later did. During the early
years of settlement, commerce flourished, leaving agriculture to occupy
the role of a subsidiary subsistence activity. It is true that, on a minor
scale, the early colonists cultivated cotton, from which the utilitarian trade
good panos (dyed measures of cloth) was woven and exchanged for slaves.
They also raised livestock to supply slave ships with meat and hides. But
mostly the settlement farmed to feed itself. Despite the initial success of
the Guinea flesh trade and the rapid demise of agriculture, history,with its
ironic flair, would eventually determine other fortunes for Cape Verde.
Through time, the trade advantages of the archipelago were eroded by
changing political winds and innovations in sea travel, and as commerce
diminished, agriculture emerged as the primary economic activity of the
islands—somewhat by default.
Today,Cape Verde is an agricultural society in an arid environment,
and food production and provisioning constitute the most pressing and per-
sistent social concerns. As a rather extreme example, Cape Verde might be
considered a microcosm of the political ecology of the Sahel—a laboratory
where complex coping strategies are conditioned by a skein of sociopolit-
ical institutions and played out in a delicate and vulnerable environment.
Arapidly growing population—with limited opportunities outside agricul-
ture—struggles to eke out a subsistence livelihood by slowly exhausting a
fragile, increasingly degraded natural resource base. From the perspective
of the local rural household, survival is the central issue that drives short-
run decisionmaking. Government leaders also interpret their political suc-
cess in terms of policies that direct agricultural activity toward broader
goals of economic growth and national food security.Nonetheless, in this
unrelenting effort to achieve critical short-run household and national
goals, society runs the risk of undermining the productive capacity of the
system as a whole. It is possible that household need and national intent
have become incompatible with ecological reality, thus surpassing the sus-
tainable limits of the productive base.
The small scale of Cape Verde offers a dramatic fishbowl view of a
human ecological dynamic that has unfolded in large parts of Sahelian
Africa and elsewhere. In this regard, Cape Verde stands neither unique nor
alone; rather,it shares the critical food-security dilemma that affects the
entire region. There is widespread concurrence among both technicians
and inhabitants that environmental degradation has occurred and is occur-
ring throughout West Africa. The causes of this environmental decline,
2 Sustainability in a Fragile Environment

however, are hotly disputed and remain embroiled in a polemic armed with
single-factor explanations. At its core, the complexity of this controversy
contrasts the role of uncontrollable, cyclical phenomena such as climate
against the role of human agency. Most scholars agree on the existence of
patterned relationships among climatic events, rural population densities,
and environmental degradation; however, the lines of causality are fer-
vently debated. The frequent occurrences of severe drought over recent
decades together with the declining trend in average annual precipitation
throughout the Sahel have fed a global-warming hypothesis that suggests
widespread climate change and minimizes the human hand in the causal
process. Contrasting arguments assert that population growth and the ex-
pansion of agriculture into more marginal areas bear the responsibility for
increasing desertification, thus reducing the role of climate as the inde-
pendent causal factor.
Other participants in the debate adhere to more complex, multileveled
perspectives that examine the specific forms of human activity and their
respective environmental consequences (see Gorse and Steeds 1987:7–8).
This approach reasons back from the household and its decisionmaking
patterns to explore the external forces that determine differing strategies of
production, notably,the political pressures to expand cash cropping at the
household level. In their influential book on African famine, Franke and
Chasin (1980) interpret the ecological problem in the Sahel in terms of the
widespread adoption of inappropriate technologies related to the regional
relationship of dependence on the international economy. Within the con-
text of this argument, Ball (1976:520) directly attributes ecological degra-
dation to the colonial efforts to “integrate peasants and herders into a mon-
etarized, commercialized economy” primarily through national policies to
increase export and livestock production.
The most comprehensive and cogent positions on environmental
degradation in the Sahel draw eclectically upon the existing arguments.
Ecological deterioration is a complex problem that cannot be understood
solely in terms of natural events. As Boserup (1965) established, increases
in rural populations result in the intensification of land use (with shorter
fallowing periods), greater herd pressures on range vegetation, and the
overexploitation of fuelwood. Historical realities as well as national and
international forces further condition the strategies of agriculturalists and
pastoralists by shaping the incentives (and disincentives) that influence
their production decisions within the household. These decisions can ag-
gravate and prolong the negative consequences of drought and create a
downward spiral of environmental degradation that is difficult to reverse.
In effect, the most credible models of environmental degradation must
interrelate such macrolevel variables as climatic change, population
growth, and international economic trends with such microlevel variables
Sustainability in a Fragile Environment 3

as household production strategies, patterns of access to agricultural re-
sources, and technology choice. These models must account for potential
contradictions in goals between household and nation-state, as well as for
the inevitable fact that positive short-term returns may incur high long-run
costs in terms of resource depletion.
In Cape Verde, where food self-sufficiency is now remembered as part
of some near-mythic past, the problem of environmental degradation dis-
plays all these levels of complexity. The resource base is fragile and under
siege by an expanding population. Moran (1982), in a careful review of
colonial history, presents a well-supported argument that the exploitative
role of the Portuguese Crown in the formation of current agricultural pat-
terns, especially the establishment of maize as the staple crop, created an
agricultural economy inappropriate to local conditions. Today, soil conser-
vation experts are unanimous in their assessment of the negative impacts
of traditional rainfed agricultural practices (Lopes and Meyer 1993). To
accept this position is not to deny,however,that climatic patterns do show
adownward trend in rainfall while population growth has increased. For
our purposes here, whether the root causes of the current ecological crisis
lie in declining rainfall, historical patterns of resource allocation and use,
the incentives presented by the international economy,or the short-run
production strategies of the household itself, the critical enigma remains.
How do farmers and policymakers manage resources so as to assure both
the national food supply and the survival of the rural household when the
expansion of agriculture could engender irreversible damage to that very
resource base? At its core, the problem as we pose it is one of resource
management in a complicated sociopolitical and physical environment.
Beyond the Ecological Balance
At one level, this book is about Cape Verdean food production and the
welfare of farming households. As such, it offers a detailed and current de-
scription of Cape Verdean agriculture under both rainfed and irrigated
regimes. Based on several years of fieldwork, these descriptions document
current agricultural practices, variations in technical strategies of produc-
tion, patterns of input use, crop choice and rotation, and forms of indige-
nous knowledge employed by local farmers. Wealso estimate the sources
and magnitudes of rural incomes among representative rural households.
No other such comprehensive documentation of Cape Verdean agriculture
has been attempted since Ribeiro’s(1954) study of Fogo Island and Ama-
ral’s (1964) landmark volume on Santiago Island, both works of Por-
tuguese geographers based in Lisbon during the colonial period. At the
4 Sustainability in a Fragile Environment

same time, this book represents the first effort to document the impact on
agriculture of Cape Verde’s independence and the dramatic social changes
that occurred with this watershed event.
This book is also about agricultural sustainability and natural resource
management. Cape Verde has a resource endowment that defines a set of
binding environmental constraints (i.e., climate, soil, topography) on agri-
culture. Just as critically, however, social institutions—wrought in his-
tory—also determine the characteristics of farming. Individual households
gain access to existing resources and exploit them in accordance with two
categories of man-made rules: those established by local communities (i.e.,
“traditional institutions”) and those imposed by nonlocal forces (i.e., “pub-
lic institutions”). These sets of traditional and public rules shape resource
access and influence the patterns of resource management, which in turn
determine the short-run and long-run impacts on the environment. Not all
farmers enjoy equal access to agricultural resources, and this inequality re-
sults in significant forms of socioeconomic differentiation in rural society,
which inspires the hypothesis that different classes of households adopt dif-
ferent strategies of resource management. Thus, in this book, we intend to
describe the natural resource base available to farmers, emphasizing the na-
ture of physical constraints to production; then to document as systemati-
cally as possible the public and traditional rules that define resource access
and management, accounting for the variation due to social differentiation.
Always lurking throughout the following chapters is the shadow of the
Malthusian dynamic, which is readily evidenced in Cape Verdean history.
Over the first five centuries of colonial existence, the relationship of local
population to the resource base was primarily determined in strict biolog-
ical terms. Assailed by severe periodic droughts, population levels of the
islands oscillated in direct proportion to changes in the productive capac-
ity of the resource base. We learn from Carreira’s (1984) punctilious his-
tory of Cape Verde that the regular occurrence of pluriannual drought
caused the production levels of staple crops to dip below minimum sur-
vival needs. The subsequent mortality due to widespread starvation low-
ered the person/land ratio, thus restoring the ecological balance. When the
rains returned, agricultural production increased, and the population grew
rapidly. Although it is true that traditional sharing institutions mitigated
some of the regional consequences of drought by redistributing available
food stocks, overall food supplies were cruelly short, and stark Malthusian
principles ruled the day throughout most of Cape Verdean history.
Under the colonial rule of the Portuguese governor, little was done to
alleviate the horrors of drought. During these regular crises, the colonial
administration vacillated between inadequate gestures of assistance and a
laissez-faire approach consistent with reigning ideologies. Portugal itself
Sustainability in a Fragile Environment 5

suffered from chronic impoverishment and was without the financial means
to mobilize substantial mitigation activities. Moreover, long-distance com-
munications were notoriously slow, while starvation imposed its unforgiving
timetable. In the final accounting, however, historical reports suggest that
the Portuguese generally lacked the collective moral resolve to save Africans
from famine. So the rhythm of environmental crisis wrought its devastating
legacy of population control, even up to the mid-twentieth century.
Beginning with the 1970s, however, the historical tandem of drought
and decimation was severed. The wave of successful independence move-
ments throughout Lusophone Africa, the fall of the Salazarist regime in
Portugal, and the geopolitics of the Cold War all inspired a call for inter-
national intervention. As a result, drought-related starvation in Cape Verde
became intolerable to both the international donor community and the
many Cape Verdean emigrants residing in Europe and the United States.
While the political climate improved over this decade of change, the phys-
ical climate worsened. The archipelago began to suffer its most wide-
spread and extended drought—lasting, with only scarce interruptions, to
the present day. Shored up by food-aid assistance, the population has
nonetheless continued to grow since the 1970s, first at a 4 percent annual
rate, then falling in more recent years to 2.4 percent (World Bank 1993a).
Thus, in defiant reversal of the past Malthusian pattern, the Cape Verdean
population expands while the productive capacity of the agricultural re-
source base declines. As the title of this book suggests, Cape Verde has
moved into an ecological imbalance between sustainable production ca-
pacity of the resource base and the size of the population seeking to derive
its livelihood from agriculture.
This relationship, illustrated in Figure 1.1, casts its shadow over all
aspects of Cape Verdean economic and political life. Over centuries, the
population line (A) roughly accompanied the subsistence capacity line (B),
determined by the amount and distribution of rainfall. The short-run
changes in agricultural production capacity are due to variable rainfall pat-
terns, while the long-run decline is a function of soil loss and reduction in
soil fertility. After the last great catastrophe of midcentury (1946–1947)
and the onslaught of the protracted drought period of the 1970s, the re-
spective directions of the two lines have diverged, but not independently
of each other. When a rural population exceeds carrying capacity for a pro-
longed period of time, there arises the danger of a permanent reduction in
productive capacity that will not correct itself with the return of the rains.
And in Cape Verde, there are few rural employment opportunities outside
ofagriculture to absorb the excess rural population. The threat posed by
the continuing divergence of these two lines is that of irreversibility, which
raises critical and fundamental questions about household food security, na-
tional self-sufficiency and sovereignty,and social and political stability.
6 Sustainability in a Fragile Environment

Figure 1.1 P
opulation and Agricultural P
roduction Capacity o
v
er T
ime

Agricultural sustainability implies resistance in the face of change and
resilience under stress. Most scholars concur that sustainable activities are
those that meet economic objectives without compromising biological
ones. Thus, in theory, a sustainable food-production system utilizes a nat-
ural resource base without diminishing its availability to the next genera-
tion (see Repetto 1986; Beets 1990:16–18). As the complexity of the real
world asserts itself, we see that the sustainable systems may not meet do-
mestic food-security needs (Ruttan 1988), as is the case in Cape Verde.
The complex nature of sustainability ultimately requires difficult de-
cisions involving uncertain trade-offs, hence this book is also about food
and resource policy. It describes past and current policies that Cape
Verdean leaders have enacted to promote national and household food se-
curity in a fragile ecosystem. Here we focus primarily on two basic policy
approaches through which Cape Verde has sought to redress the ecological
imbalance. The first approach wagers a technological solution and seeks to
identify and encourage alternative practices that reduce environmental
degradation and increase the output efficiency of the resources currently
being utilized—primarily soil and water. To achieve this end, the govern-
ment has invested in research and extension activities and has played a
major role in the management of natural resources. The second approach
attacks the population side of the imbalance. The dilemma that Malthus so
aptly described for Cape Verdeans presents little choice but to stabilize
population growth by either reducing the birthrate or facilitating out-
migration from rural areas. A combination of reduction in the number of
rural people, improvements in the basic value of human capital through
education and training, and, over the long run, the development of non-
agricultural economic alternatives defines the ultimate strategy.
National self-sufficiency in food production is not a realistic goal for
acountry in which 90 percent of its foodstuffs is either imported or do-
nated by the donor community.On the other hand, the current ecological
imbalance constitutes a critical threat to economic and political stability,
and some level of sustainable agricultural production must eventually be
established in order to avoid disaster. The overarching challenge facing
policymakers in Cape Verde is how to narrow the gap between lines A and
B(in Figure 1.1) and to achieve a sustainable balance over the long run.
This imperative is depicted as lines P', P'', and P''', which represent the
vectors of possible policy interventions that would conceptually work to-
ward the establishment of a reasonable and sustainable system of produc-
tion. The policy alternative depicted by line P' establishes a sustainable
production level under current climatic conditions; line P'' represents a
policy initiative that promotes rapid agricultural growth over the short run,
but that ultimately results in the severe depletion of the natural resource
base. The third alternative, line P''', seeks a policy set that would move
gradually to lessen the food-deficit situation in a sustainable way.
8 Sustainability in a Fragile Environment

These alternative policy approaches will be discussed in the context of
the public role in agricultural resource management. Over the short run,
however, policymakers in Cape Verde face the urgent problem of feeding
the rural population. The lack of nonagricultural alternatives leaves more
people practicing agriculture than the resource base can support. At the
same time, the government understandably fears an uncontrolled exodus of
rural inhabitants to urban areas, which are already extended beyond their
ability to provide either employment or basic infrastructural services.
Moreover, the current short-run strategy that depends—almost entirely—
on international largesse and emigrant remittances is clearly vulnerable. In
essence, Cape Verdeans find themselves challenged to redefine the role of
agriculture, to relocate its place in the general economy,and to confront
the issue of sustainable resource utilization. But what are the specific pol-
icy issues that will shape the future of agriculture in Cape Verde? To ad-
dress this question, we need to discriminate between dryland and irrigated
farming.
Policy Issues in Dryland Agriculture
The majority of agricultural lands in Cape Verde are cultivated under a
rainfed regime, and only a small minority of farmers has any access to ir-
rigated lands. Numerous technical reports have directly associated farming
practices with environmental degradation (Freeman et al. 1978; USAID
1984; IFAD 1990; Lopes and Meyer 1993). The widespread dryland culti-
vation of maize has been implicated as a major cause of soil erosion and
high rainfall runoffrates. Soil and crop technicians argue that maize is an
inappropriate cultigen for the environmental conditions of Cape Verde,
which are better suited for more drought-tolerant and higher-producing
crops, such as millet and sorghum. In addition, current population growth
encourages the expansion of maize cultivation onto more marginalized
lands, particularly the steep, volcanic hillsides of the valleys. The conven-
tional argument denounces traditional production practices: farmers do not
adopt systematic measures of erosion control, such as contour planting,
and actually accelerate soil loss by removing all competing vegetation
from the sloping fields. Nevertheless, maize, which was introduced by the
Portuguese from Brazil, has been totally woven into the cultural fabric of
rural society and has been the preferred staple of consumption.
Under existing precarious conditions, yields are too low to assure
household food security,and the responsibility for provisioning the rural
population has fallen upon the public sector. Public work fronts are a com-
mon feature of rural Cape Verde and provide the employment that injects
cash into the rural economy and reduces the attraction of rural-urban mi-
gration. The public work system is financed by food aid that has been
Sustainability in a Fragile Environment 9

monetized through the private market system and made available to rural
customers. Thus, the major policy issues with regard to dryland agriculture
focus on strategies to reduce erosion, increase productivity, and, at least
for now, maintain a local safety net that keeps rural people on their farms.
Over the longer run, the questions become the sustainable size of the rural
population and the locations in which the excess population might be ab-
sorbed outside the rural areas.
Policy Issues in Irrigated Agriculture
On these arid islands, water is the critical resource not only for agriculture,
but also for rapidly expanding urban populations. The availability of water
for irrigation is highly variable and sparsely distributed, and the policy
problem focuses on the need to increase the use efficiency of this scarce
resource. Most experts agree that current groundwater supplies signifi-
cantly exceed current usage levels, which suggests that capital investment
in irrigation or reorganization of existing systems could increase high-
valued production.
Asecond issue addresses the equity of groundwater distribution. Poli-
cymakers perceive a critical trade-off in public versus private, or local, man-
agement of groundwater resources, in that larger,more efficient farms might
serve some national production goals but might negatively compromise the
situation of many rural households that rely heavily on their minuscule plots
of irrigated land. Because of this same concern for equitable access, share-
cropping tenure relations have been a target of policy interventions.
Athird policy dimension involves cropping patterns, because the ma-
jority of irrigated agriculture is devoted to the production of sugarcane for
the purpose of distillation into the traditional alcoholic spirit, grogue,
rather than to the food crops that would enhance national food security.In
an economy with such low levels of domestic food production, the cash
crop/food crop debate edges over toward one of national security and the
ability of the country to finance necessary food imports should food-aid
assistance be interrupted.
In the aggregate, these problematic issues project a particular vision
that guides policymaking in Cape Verde. Over the short-run, rural society
must be made viable, first to avoid the disasters of the past and, second,
to discourage rampant rural flight to ill-prepared urban centers. Over the
long run, Cape Verde seeks to wean itself from international donor depen-
dence and to discover its own sustainable basis for development both in
and out of agriculture. To discover what policy alternatives will effectively
achieve this vision constitutes the national challenge.
10 Sustainability in a Fragile Environment

The following chapters lay out the various pieces of the agricultural
resource-management and policy puzzle. Initially, we present a discussion
of the methodology that guides our analysis (Chapter 2), and we suggest
that this approach is readily adaptable to any agricultural society where
improved resource management requires a comprehensive understanding
ofexisting production strategies. Chapter 3 describes the natural resource
endowment of the archipelago, and Chapter 4 outlines the historical con-
text of agriculture in Cape Verde. Detailed current descriptions of dryland
(Chapter 5) and irrigated farming practices (Chapter 6) are followed by an
analysis of private resource-management strategies adopted by farmers
(Chapter 7). In Chapter 8, policy alternatives relevant to public resource
management are explored. Although this case study of the Cape Verde
archipelago highlights the problems and explores the management options
that have been tried or appear feasible, we, alas, must warn that the Cape
Verde story offers no facile solution to the essential dilemma that sets food
security and environmental sustainability in a contrapuntal relationship.
Sustainability in a Fragile Environment 11

The story line we establish throughout this book interprets the current
reality of Cape Verdean agriculture within both the physical and the policy
environment. In this chapter,we present the overall methodological ap-
proach that integrates the natural resource base, the agricultural activities
of rural households, and the impacts of national policy in Cape Verde. The
goal of the analysis is to identify those critical factors that influence the
management of productive resources at the disposal of rural households.
With an understanding of the incentives and constraints that direct agri-
cultural and nonagricultural activities, we can then assess the current wel-
fare of rural families in Cape Verde and identify the potential for public
actions to improve conditions of rural residents. Toward this end, we con-
struct a framework that depicts the relationships between the combination
of productive assets—labor,land, water,and capital—available to house-
holds and the outputs they are able to generate with those resources. As an
analytical basis, this framework allows us to estimate actual income lev-
els of households and to predict expected changes resulting from specific
policies designed either to increase household access to productive re-
sources or to improve the technologies that transform these resources into
valuable outputs. We then outline the types of information necessary for
ouranalysis and the data-gathering strategies that generated the empirical
foundation of this book.
The Political Ecology of Cape Verdean Agriculture:
AConceptual Framework
Borrowing from the recent social-science literature, we refer to the com-
plex relationships that link household strategies to the physical and insti-
tutional environment as the political ecologyof Cape Verdean agriculture
13
2
Public and Private
Resource Management:
AMethodological Approach

(depicted in Figure 2.1). In this conceptual diagram, we identify three es-
sential components: a specific resource endowment (the physical environ-
ment), a basic management unit (the household), and a set of institutions
that defines the means by which households can mobilize and utilize these
resources (the institutional environment).
1The overall resource base is es-
tablished by the geophysical, climatic, and biological conditions within the
archipelago. In Cape Verde, the critical aspects of the physical resource
base are the extreme scarcity and variability of rainfall, the sharp topo-
graphical relief that promotes erosion and the consequent low soil fertility,
and the small amount of total land available relative to the existing rural
population. In short, the agricultural productivity of land is very low and
extremely variable over time and location, and land is scarce relative to the
number of households that depend on agriculture for their livelihoods.
Households are the empirical units that manage the natural resource
base so as to assure livelihoods for their members. Following Netting,
Wilk, and Arnould (1984), households are residential units of variable
structure that perform a set of essential functions, including production, re-
production, and transference of resources.
2In contrast with polygynous so-
cieties of Sahelian Africa, the Cape Verdean household, as a social norm,
follows a rather straightforward, monogamous, nuclear pattern, although
the empirical reality reveals a high frequency of single-parent, women-
headed families. Although the structure of the residence unit may extend to
include a lineal or collateral relative, the production and decisionmaking
processes are concentrated around the household head, rather than dis-
persed among different household subunits, as with the Sahelian conces-
sion households. The private ownership of agricultural land is endorsed
under Cape Verdean social and institutional structures, and both law and
custom sanction the partible inheritance of family property,including land.
Family members constitute the core pool of labor services to work house-
hold lands. Thus, each household has at its disposal a mix of available re-
sources, which varies according to the composition of the family and the
set of assets that has been accumulated through ownership or other insti-
tutional forms.
A large body of literature analyzes the factors that may affect the pro-
duction and consumption decisions of agricultural households as economic
units.
3In the context of incomplete market integration, such decisions may
depend more on specific household characteristics (the combination of
available resources relative to the number of consumers in the household)
than on external market prices (Lopez 1986). For example, in developing
economies, households generally do not have ready access to markets for
all resources and commodities, so they articulate with markets in a com-
plex fashion. In the effort to meet domestic food needs, households decide
between consuming their agricultural production and marketing the output
14 Resource Management

Resource Management 15
Figure 2.1 The Political Ecology of Cape Verdean Agriculture

to acquire cash income for the purchase of necessary consumption items,
including food. Without markets to determine the prices of the whole
range of inputs and outputs, the relative values of cash and food depend
in part on household-specific characteristics that affect consumption pat-
terns and production possibilities. Perhaps the most important example of
incomplete market integration in rural Cape Verde is that of labor. Indi-
viduals have very limited opportunities to employ their labor in rural
areas. The two traditional alternatives have been the public work fronts
and emigration. Public work fronts do not operate throughout all of the
country, and many households do not have access to this opportunity sim-
ply because of their geographical location. In addition, the public work
program restricts the number of members from a single household that
may participate, and it closes down during the rainy season. Emigration,
on the other hand, is a very costly and risky employment strategy, espe-
cially for individuals without contacts in the country of destination. Thus,
particular household circumstances—namely,having friends or family cur-
rently abroad—strongly affect the attractiveness of this employment alter-
native. These household-specific differences in access to public work
fronts and emigration are not due to differences in market wage rates, but
they can significantly affect labor-allocation decisions in agricultural
activities.
Furthermore, farmers in many parts of the world commonly devise
complex complementary interrelationships among productive activities,
particularly between cropping and livestock production (McIntire,
Bourzat, and Pingali 1992). In Cape Verde, for example, many farmers use
corn straw to feed animals, and the value of the straw in some years sur-
passes that of the grain. This interplay between production and consump-
tion decisions arises when households are not freely able to buy or sell all
inputs, outputs, and factors of production in markets. In such an absence of
complete market integration, household resource-management strategies
depend on household characteristics and access to resources as well as on
relative prices.
It is thus necessary to understand the operational characteristics of all
the institutions, including markets, that regulate or define the transfer of
resources among agents within an economy. The particular terms that de-
fine transfers—for instance, forms of payment, forms of access—deter-
mine transactions’costs. Because of the various factors that either influ-
ence the prices faced by different producers or restrict producers from
participating in market transactions, agricultural systems exhibit signifi-
cant variation with respect to adopted technologies, proportions of differ-
ent inputs used in farming activities, and production responses to changes
in economic incentives.
4
16 Resource Management

According to our framework, households operate within an institu-
tional environment comprising all the socially sanctioned “rules of the
game” that regulate human interactions (North 1990:3). More specifically,
rural households acquire access to or dispose of available resources
through enabling institutions that define the conditions under which these
transactions may take place and the form of payment (if any) needed to ac-
quire use rights. Furthermore, these institutions define the distributional
patterns of use rights among specific groups or individuals within the com-
munity. For example, agricultural land in Cape Verde is transferred across
households by partible inheritance, and local customs define the specific
process by which such intergenerational transfers take place, namely, how
resources will be shared among offspring. In addition to inheritance, land
can also be transferred across households through market institutions—ei-
ther purchase, rental, or share tenancy agreements. Individual households
may mobilize one or several of these avenues to build their land endowment.
For analytical convenience, we divide the institutional environment
into local or private institutions, which are formed and monitored within
the rubric of community traditions, and state or public institutions, which
are embodied in the political, bureaucratic, and legal structures of the
nation-state. The distinction between these two forms is based on the no-
tion that the management of resources is more direct and personal in pri-
vate institutions, whereas public institutions require formal structures
(government bureaucracies) to undertake the role of resource stewardship.
Our conception of private institutions corresponds closely to Bromley’s
definition of autonomous and self-enforcing conventions, which are “au-
tonomous because they arise from the internal dynamic of . . . several
agents, and they are self-enforcing because no independent agent has any
incentive to modify the convention” (1991:90). Public institutions, in con-
trast, represent wider interests outside the local rural communities. Al-
though such public institutions as organized churches and international
assistance organizations also operate within rural Cape Verdean commu-
nities, our analysis focuses on state institutions because they have the most
direct impact on resource-management decisions of households. The ac-
tivities of other public institutions must be sanctioned within the national
legal system and, in fact, operate in close association with governmental
agencies.
The productive resources of rural households are governed by an over-
lapping network of private and public institutions. For example, much of
the daily management of irrigation networks is undertaken directly by the
farmers who share water from a specific water source, but some of their
activities are regulated by the national agency responsible for groundwater
management (and urban supplies). Similarly, household labor may be
Resource Management 17

exchanged with neighboring households through local institutional mech-
anisms or may be hired out through national or even international labor
markets. Land is also affected by a range of local customs, as well as by
national legislation defining legally sanctioned forms of land transfer. In
some cases, this intersection of influence among several institutional lev-
els can be complementary, as in the case of water management where pub-
lic and local institutions are highly integrated. In other cases, however,
state institutions clash strongly with the interests of rural households and
local institutions. In the early 1980s, land-reform legislation outlawed
sharecropping, but this land-transfer institution persists because it is a pre-
ferred mechanism for many landlords and tenants. Within this web of in-
teracting institutions that enables resource access and control, households
face complex management decisions, particularly when various institutions
impose conflicting constraints and incentives on resource use.
From where do such institutions come? In general, both public and
private forms have evolved over time through political negotiations among
interested parties seeking to address particular problems related to the dis-
tribution of resources among members of the community. Private institu-
tions reflect the interests of constituents within local communities as well
as adaptive responses to specific environmental conditions. Public institu-
tions, on the other hand, are shaped by outside (in many cases, even inter-
national) interests and may have little regard for the particular concerns
of members in local communities. Local institutions can adjust rapidly
when all members in the community can agree on a mutually advantageous
change; however,local institutions are often slow to respond in situations
where individuals have conflicting interests. Those who benefit under the
status quo will tend to resist any changes that are prejudicial to themselves
butimprove the overall management of the resource base within a com-
munity. By contrast, public institutions may totally ignore the interests of
local communities, or they may successfully address situations of conflict
among groups or individuals in an equitable manner.The specific institu-
tional environment operative at a given point of time represents an amal-
gamation of many diverse elements forged over a period of time and re-
flects shifts in relative power positions of different constituencies, as well
as the dynamics of resource management under changing circumstances.
Some of the institutions that govern rural life in Cape Verde are holdovers
from the colonial period and were fashioned when the agricultural re-
sources were less scarce in relation to the rural population, and they have
been adapted, with varying degrees of success, to current conditions.
Our analysis considers the institutional environment to be exogenously
determined for rural households at any given point in time. We do, how-
ever, recognize the adaptive and evolutionary nature of social institutions
(Barth 1966; Bailey 1969). Although households and local communities
18 Resource Management

must acknowledge the rules established by public institutions, they are
often able to devise actions that circumvent the effects of these external in-
stitutions. Consequently, for public institutions and policies to be effective,
they cannot diverge too radically from local interests, or avoidance behav-
iors will become widespread and efforts to ensure compliance will become
too costly and unwieldy. Our analysis helps to identify some of the long-run
pressures for change that will come to bear on the institutional environ-
ment, but we do not attempt to explicitly model these political processes.
In this analytical framework, households have an initial endowment of
resources that is determined by the individual circumstances of the house-
hold and its particular articulation with the institutional environment. At
any given point in time, individual households will allocate their existing
resource base among alternative productive activities. In order to have an
empirically tractable model, we assume that households make these deci-
sions to maximize total household income, which comprises cash income
(from agriculture and other activities) and the monetary value of agricultural
output consumed directly by the household. Income is generated either by
utilizing household resources in productive activities or by selling or hiring
out resources. Thus, household land and labor can be utilized to produce
agricultural and livestock outputs. Alternatively, land can be sold or rented,
and household labor can be employed off the farm to receive wage income.
5
Figure 2.1 illustrates the interrelationships among these three ele-
ments. The natural resource endowment of the islands forms the overall
productive base on which rural households earn their livelihoods. At the
same time, the institutional environment defines specific household re-
source endowments. In the figure, the household moves its resources in
one of two allocative directions—toward either agricultural or nonagricul-
tural activities. Under the agricultural category,we include crop and live-
stock activities, off-farm agricultural wage labor, and exchange labor. The
nonagricultural category is comprised of nonagricultural wage labor (pri-
marily public work front employment), artisan production, petty commerce
and other entrepreneurial activities, domestic maintenance, education, and
emigration. Household managers are assumed to make their decisions
based on their perceived trade-offs of the benefits and costs of the alter-
natives they face. This decisionmaking process determines for each house-
hold its allocation of available labor to agricultural, domestic, and off-farm
activities; specific crop choices and technologies employed in agricultural
production; investment levels in physical and human capital; patterns of
consumption and savings; and participation in community affairs. From
these allocative decisions, the family obtains some level of income. Over
time, households are assumed to undertake constant evaluation and modi-
fication of their resource-management strategies in order to enhance their
well-being.
Resource Management 19

At the same time, household resource-management strategies incur di-
rect environmental impacts that affect the sustainability of the resource
base, as Figure 2.1 suggests. In conceptual terms, households can ratio-
nally adjust their strategies if prevailing agricultural practices function to
degrade the production base, for example, by relying more heavily on non-
agricultural income sources. In rural Cape Verde, however, hard reality as-
serts itself in two ways. First, it is not clear that farmers consciously weigh
the trade-offs between short-term welfare and long-term sustainability.
Second, even when the consequences are known, the immediate lack of
opportunities outside agriculture severely restricts the household ability to
opt for long-run environmental benefits if they mean a reduction in agri-
cultural activity.Within limits, it is likely that households do alter their
patterns of resource management to the extent that they perceive and di-
rectly incur the long-run costs of their actions, but survival occupies the
position of priority for allocative decisions. Yet government leaders may
not share the same sense of urgency that households do, and their policies
reflect a priority concern with the environmental consequences and their
social costs.
In sum, a wide range of overlapping enabling institutions defines the
rules and conditions under which households may acquire, use, and trans-
fer agricultural resources. These institutions are neither deterministic nor
immutable but, rather,face continual pressures for change. The particular
allocative strategies that households pursue may themselves impose pres-
sures on the enabling institutions and their preservation through time. As
the dynamic between human intervention and the natural resource base un-
folds, the trade-offs between household production strategies and resource
sustainability may reduce the effectiveness of some institutions and in-
crease the importance of others. Changes in household behavior will have
interactive impacts on both the natural resource endowment and the insti-
tutional environment.
As one of its principal advantages, this framework explicitly recog-
nizes the variability across households with regard to access and use of
resources toward income generation. Rural households vary along measur-
able and patterned dimensions. Neither populations nor productive re-
sources distribute themselves evenly over the physical landscape, and all
individuals do not enjoy equal endowments of talent or equal access to op-
portunities. Thus, in the social environment of rural Cape Verde, we expect
tofind differences in scale, in forms of access to resources, in technology,
and in wealth. These differences establish the bases for patterned variation
in resource management—both directly in terms of farm household strate-
gies and indirectly in terms of the impact of public policies and institutions.
The first methodological challenge is to depict the existing variation
in a relevant, accurate, and interesting manner.We have chosen to employ
20 Resource Management

the concept of the representative household, or farming system, syntheti-
cally constructed on the criteria of access to land and water, cropping pat-
terns, and the type of household management structure—either jointly
managed or female managed. All policies, whether well-intentioned or in-
sidious, exact their impacts differentially on households, and we have at-
tempted to develop a classification of Cape Verdean farm households that
face significantly different “opportunity frontiers” and thus respond dif-
ferently to policy interventions.
The second methodological challenge is to explain the observed vari-
ation in such a way that identifies the critical variables affecting resource
allocation. If we first describe the range of strategies that households de-
vise within the context of Cape Verde’s resource endowment, it is then
necessary to discern why these particular strategies have been selected. At
this point, the analysis becomes useful to policymakers concerned with
both national food security and sustainability of the resource base.
The Economic Analysis of
Rural Cape Verdean Households
As our conceptual framework argues, the household is the operational unit
that makes management decisions about agricultural resources based on
the relative costs and returns of alternative management strategies. For this
analytical task, we construct and compare representative budgets of the
crop and livestock activities of discrete household types. The representa-
tive models incorporate information about the prices of inputs and outputs,
the production technologies utilized, and the resource endowment of each
household type. This approach constitutes an alternative to standard
econometric models that posit the uniformity of household response to
economic signals, namely,relative prices. Most empirical economic analy-
ses of allocative behavior assume that markets exist for all outputs, inputs,
and productive resources, and that all producers have identical access to
these markets and face the same prices. Under these assumptions, the de-
cisions of producers are driven only by relative prices between inputs and
outputs, and all producers will respond similarly to changes in relative
prices. Furthermore, all producers will use the same technologies and
apply the same relative proportions of factors to their production activities.
Econometric estimates of supply functions, profit functions, and cost func-
tions measure the relationships between quantities supplied to the market
and prices of alternative outputs and inputs. Fixed factors, such as land
and capital, may be included in the models, particularly in estimations of
profit or cost functions, but usually only as a measure of available capacity.
Thus these models do not assume different supply responses for producers
Resource Management 21

with different allocations of fixed factors; rather, econometric models typ-
ically compress the observed input/output decisions of all producers into
asingle average production function for the entire sample. As a result, dif-
ferences in responses across different types of farms are concealed.
6
It is precisely our interest in the variation among farm types that
makes this econometric approach less appropriate. In the Cape Verdean
context, the assumptions of free access to markets and identical prices
across producers do not accurately describe the agricultural reality. Many
resources are not allocated through markets, and the institutions that gov-
ern nonmarket transactions can impose different opportunity costs upon
different individuals. For example, the cost of irrigation water varies sig-
nificantly across farmers in Cape Verde, but not because of differences in
price. Water is not allocated through markets where individuals are able
to purchase desired amounts at any point in time. Rather, allocations are
either made through negotiations among members within a network, or, in
the case of publicly managed networks, a national irrigation agency deter-
mines patterns of distribution among the members. The physical capacities
of different irrigation systems and the particular institutional mechanisms
that schedule water deliveries to users create significant variations across
individual producers in terms of access to irrigation water over the course
of the growing season. Similarly, households face very different opportu-
nities with regard to off-farm employment. Households residing near an
operating public work project will have more ready access to this alterna-
tive than those households more distant from any public project. Likewise,
households with family members in other countries have much easier ac-
cess to emigration opportunities. These are very important differences
across households that cannot be captured by price differences alone, ei-
ther because markets (and thus prices) do not exist, or additional, nonprice
constraints restrict market participation.
RepresentativeHousehold Budgeting
Our analysis of household resource strategies incorporates more informa-
tion than just the relative market prices between inputs and outputs. In par-
ticular, this approach requires a quantitative understanding of variations in
technologies and access to resources across farms within the agricultural
sector,as well as a qualitative assessment of patterned variations in pro-
ducer access to markets and other resource transaction institutions. The
use of standard budgeting techniques allows us to depict this institutional
variation. Budgeting assumes a fixed-coefficient technology and requires
specific information about the magnitudes of the physical input/output co-
efficients. Market prices for all outputs, inputs, and factors of production
are also required. This framework may be used to evaluate differences in
22 Resource Management

technologies (input/output coefficients) and prices by constructing separate
budgets with appropriate changes. Budget results indicate the relative eco-
nomic incentives of alternative resource strategies and changes in those in-
centives that result from changes in prices or technologies.
7
A number of factors constrain the income-generating opportunities
available to Cape Verdean families. Most obvious among these are the lim-
ited quantity and quality of resources at their disposal. In addition to the
overall restricted land area accessible to most households, the geological
and climatic conditions on the archipelago influence the productivity of
both cropping and livestock activities on rainfed land. The scarce supply
of developed subterranean water sources poses limits on the total amount
of irrigated land. Moreover,institutional constraints determine resource
use rights of households, the mechanisms of resource transference, and
how returns from resources are distributed among households with over-
lapping ownership and usufruct claims. These institutions define and limit
the way in which households may use their available resources. The rep-
resentative household budgets explicitly incorporate these constraints by
specifying both the representative combinations of resources for the dif-
ferent household types and the associated factor opportunity costs.
The representative budgets provide a basis for the evaluation of alter-
native policy and resource-management initiatives. The budgets incorpo-
rate the impacts of the enabling institutional environment in Figure 2.1 in
two ways. First, the budgets that model representative household types
capture the variation in resource combinations resulting from specific pol-
icy decisions. Second, the prices included in the budgets may also reflect
the impact of institutional interventions, such as the cost of water. To the
extent that a given policy can be associated with a change in access to re-
sources, the adoption of a new technology,and specific pricing structures,
the budgets generate relevant economic consequences at the household
level. If information is available on the relative frequency of a given rep-
resentative household in the rural population, more aggregated impact es-
timates are possible. This use of budgets as a foundation for policy analy-
sis has an established tradition in development economics (Monke and
Pearson 1989).
Representative household types are interrelated systems of individual
activities. They are defined in terms of a specific mix of available re-
sources and agricultural production patterns. In our Cape Verdean exam-
ple, a given farm could be comprised of a certain amount of rainfed land
and, perhaps, an irrigated plot with a specific crop mix and yields that vary
with access to supplies of water.This farm also has a given livestock com-
ponent. All cropping and livestock activities are represented by individual
standardized budgets, which are combined in specific proportions to model
the representative household system of resource allocation. These models
Resource Management 23

thus explicitly incorporate the representative household’s access to a par-
ticular combination of resources, as well as the multiple and interrelated
activities undertaken by the household.
The representative households are assumed to share the overall objec-
tive of increasing net returns to household resources, or household income,
defined as the value of outputs minus the cost of all inputs and resources
that are not owned by the household. In particular, household labor and
owned land are not counted as costs in the calculation of net returns. Since
the household does not have to purchase these owned resources, the net re-
turns from all household activities can be interpreted as household income,
or the payment to family-owned factors.
8This analytical strategy allows us
to avoid the problematic valuation of household labor at current market
wage rates (Barlett 1980) and also provides a derivative estimate of the
value of the critical domestic labor (child care, food processing, etc.) that
supports productive activities but is seldom measured.
The budgets estimate family income from representative combinations
of cropping and livestock activities. However,Cape Verdean households
also allocate labor to nonagricultural sources of income. To reflect this re-
ality,the analysis also considers the dependence of rural households on
off-farm sources of income, primarily public work front labor, agricultural
wage labor, nonagricultural wage labor, emigration remittances, and pen-
sions. With this approach, any change in prices, in access to productive re-
sources, or in wages can be evaluated in terms of the impacts on household
income, that is, on family welfare.
Some qualifications must be acknowledged with respect to resource-
allocation decisions. First, the effects of risk associated with different
resource-management strategies require careful attention, since households
incorporate risk perceptions into their assessments of competing allocative
strategies.
9Poorer households are likely to be particularly adverse to any
strategy that has a significant probability of occasionally generating an un-
acceptable low return, even if, on average, expected returns are higher than
the traditional and less risky alternative strategy. Several authors have dis-
cussed differences in production decisions depending on the level of risk
associated with access to irrigation water over time (Easter 1977; Rei-
dinger 1974; Wade 1988). The tail-end users—those most distant from the
water source—are generally subjected to greater levels of risk in access
than are those closer to the water source. Bromley,Taylor, and Parker
(1980) cite case studies indicating that farmers with more insecure access
to water will select more drought-resistant crops with lower expected re-
turns and will reduce application rates of other inputs in comparison to
farmers with more reliable water supplies.
Time preferences may also vary systematically with income levels.
10
Poor households may be expected to discount possibly higher future income
24 Resource Management

more heavily than wealthier households, since the more vulnerable farm-
ers have fewer incentives to make investments with substantial delays in
payoffs. These poorer households are thus more likely to pursue nonsus-
tainable strategies, since they cannot afford to wait for long-term benefits
that could accrue from alternative resource uses. These differences in time
preferences are independent of the higher opportunity cost of external (in-
vestment) funds for poorer households, but the effects are the same: a re-
duction in the attractiveness of long-term investment.
The economics of the farm household, then, are modeled on synthetic
budgets that are in turn used to construct a typology of farms built around
specific combinations of resources. The analysis is static in the sense that
the results are tied to the specific assumptions about price and technical re-
lationships, as well as to the institutional environments that are changed
only exogenously to examine potential impacts of policy interventions. At
the same time, the budgets permit an effective comparison of the incen-
tives underlying alternative uses of resources. In the analysis, we employ
the budgets to establish a base, then alter the assumptions to examine the
impacts, and finally, interpret the results against the accumulation of data
acquired over a multiyear research period.
The Data-Collection Process
The focus on households as the unit of analysis exacts a heavy demand for
detailed and systematic information about the socioeconomic conditions
and behavioral strategies of rural families. The information set for this
study was collected in a series of field activities over a five-year period,
during which time a specific research framework was developed. Our
fieldwork methodology, which we call sequential rural research,is an it-
erative, mixed-methods approach that draws inspiration from the method-
ological innovations associated with the farming systems research para-
digm (Finan 1996). Under this strategy, a specific sequence of research
methods is employed, with each stage addressing a different dimension of
the research problem and building, in an interactive manner, upon the cor-
pus of knowledge accumulated in the prior stages.
The sequential rural research strategy is based on two basic principles.
First, some types of information are more costly or difficult to obtain than
others. For example, descriptive information, particularly that which can
be directly observed, such as the number of people in a household, is much
easier to obtain than information about the whole range of factors that in-
fluences households’ resource-allocation decisions. Second, the relative
benefits of different types of information depend on the extent to which
they address specific policy issues and offer a basis for continued empirical
Resource Management 25

research. For example, the distribution of individuals’ eye color is rela-
tively easy information to obtain, but it is not relevant to any policy deci-
sion. Distributions of age and sex within a population may not be directly
relevant to policy decisions, but these variables become extremely impor-
tant to the extent that household welfare is associated with the dependency
ratio or the gender of the household head.
The sequential rural research program in Cape Verde unfolded as part
ofafive-year Food Crops Research Project with the Cape Verdean Na-
tional Agricultural Research and Development Institute (Instituto Nacional
de Investigação e Desenvolvimento Agrário [INIDA]), funded by the U.S.
Agency for International Development (USAID). The overall objective of
the socioeconomic analysis component of the project was to identify the
most important constraints to increased agricultural production faced by
rural households. This information was then directed toward the establish-
ment of research priorities in the various departments of technical and
socioeconomic research in INIDA. The long life of the project facilitated
the development and implementation of this sequential rural research
approach.
The initial stage in this methodology was a series of rapid rural ap-
praisal (RRA) visits designed to acquire background information on farm
households within the project’s target regions.
11RRA techniques are often
recommended for situations where the existing knowledge base is mini-
mal—an accurate assessment of the Cape Verdean case. Here, the RRA ac-
tivities attempted to identify broad patterns of variation in several areas of
general information including: (1) household access to productive re-
sources and the range of agricultural and nonagricultural activities in
which those resources are employed; (2) technical information about agri-
cultural practices, including cropping calendars, input/output relationships,
and access to input and output markets; and (3) institutions that provide
opportunities for households to acquire additional resources or to sell and
hire out owned resources. The RRAs also provided invaluable access to
the local idiom of agriculture—that is, indigenous knowledge that ranges
from ethnobotanical terminology and local weights and measures to social
values and expectations.
The RRAs were initiated on the two most important agricultural is-
lands, Santiago and Santo Antão. The information obtained in these activ-
ities then provided the basis for a further series of data-collection exercises
designed to document the observed variation in a systematic and represen-
tative fashion. Major household surveys were organized and carried out
first on Santiago then on Santo Antão and São Nicolau. These surveys
were designed to simultaneously elicit information that is relatively easy
to obtain in a formal interview and provide a useful empirical basis for fur-
ther analyses. They focused on family demographic characteristics, access
26 Resource Management

to various types of agricultural resources, off-farm employment across
households on the two islands, and other factors thought to constitute con-
straints to the improved well-being of households.
The need for quantitative information about the distribution of re-
sources across households required a statistically representative sample of
rural households. On both Santiago and Santo Antão Islands, the research
team designed random samples of households stratified according to major
agroclimatic zones. In this two-stage sampling process, local communities
(localidadeson Santiago, zonason Santo Antão) were selected from com-
plete lists obtained from local representatives. Then local community lead-
ers provided the names of all families with agricultural resources, speci-
fying the sex of the head of the household and whether or not they had
irrigated land. A sample was then drawn from each of these household
lists.
The household surveys, using questionnaires administered in the
crioulolanguage, elicited three general types of information. First, an in-
ventory of household members was compiled, including specific informa-
tion on each individual’s age, sex, and level of education; time spent in
agricultural activities; emigration experience; and finally,off-farm em-
ployment. The total number of individuals in the household provided an
indication of the aggregate expenditure requirements necessary to meet the
needs of all family members, and the number of working-age individuals
provided an estimate of the supply of labor services available within the
household, which could then be compared with actual time spent in both
on- and off-farm activities.
The second category of information was an inventory of agricultural
resources to which the household had access in the year prior to the sur-
vey.This category included rainfed land, irrigated land, livestock, pur-
chased inputs, and water. Land area was broken down into owned, rented,
and sharecropped, providing an indication of household dependence on
different institutions of land access. For those households with irrigated
land, the average interval between irrigations and their source of water
were also obtained.
The third category concentrated on the utilization of family resources,
including cropping patterns on rainfed and irrigated land, the use of fam-
ilylabor in agriculture, the hiring out of family labor, and sales of crops
and livestock products. The surveys also inquired about the time needed to
perform basic domestic chores such as gathering fuelwood and securing
water for domestic use, both time-consuming activities. Labor allocation
was discriminated by gender and age when relevant.
The analysis of these survey data permitted the establishment of a basic
typology of representative farms based on access to resources, cropping
patterns, and gender of the household head. Since agricultural production
Resource Management 27

processes are complex and dynamic, involving biological, physical, and
social variables, the creation of a representative typology is a daunting ex-
ercise that requires careful judgment as well as accurate data. The essen-
tial difficulty was to discover a level of representativeness that adequately
reflected rural differentiation. At one level, production strategies vary
among all households, and it would have been possible (in fact, easier) to
establish a unique type for each household—clearly an unacceptable ana-
lytical solution. At the other extreme, collapsing all households into a sin-
gle representative model would have precluded the evaluation of differ-
ences across households—equally unacceptable. Thus, the analysis was
challenged to identify a defensible set of criteria with respective value
ranges that discriminated an accurate typology,but also controlled for rea-
sonable patterns of variation around any single type. As a pragmatic rule
of thumb, the variations within any type were judged to be of secondary
importance with respect to the overall analysis and to the evaluation of
factors that policymakers can influence.
The use of formal survey data to generate household typologies also
permitted preliminary identification of relational patterns between differ-
ent variables. For our sample of farmers in Cape Verde, analysis of the sur-
vey information suggested relationships between gender and cattle owner-
ship, cropping choices and source of irrigation water, farm size and tenure
mechanism, among others. Although the survey information can reveal
patterns and correlations among variables, this information alone seldom
provides an adequate causal explanation for these observed patterns. With
the survey,the researcher gains a certain level of understanding of the pat-
terned relationships among social facts, but an iterative process of verifi-
cation (or contradiction) through subsequent discussions with farmers is
necessary to more fully fathom the rural reality.
With the results of the formal survey and the household typologies in
hand, therefore, we identified a sample of farmers for periodic monitoring.
The sample was purposive in the sense that representatives of each house-
hold type were included. The sampling procedure was also self-selective
and participatory in the sense that we sought out households displaying a
willingness to participate in the research (after lengthy explanations of
project design and objectives). It is a basic research fact that some indi-
viduals have a better understanding of important processes and are better
able to communicate with outsiders (including Cape Verdean researchers).
For such deep and often introspective information, random selection of the
interview sample is often unsuccessful; therefore, we sampled households
that met a definitional set of characteristics and displayed an understand-
ing of the project, a desire to participate, and the ability to articulate
household strategies of production and well-being.
28 Resource Management

The technical assumption orienting this sampling procedure was that,
within each type category, the variation of response would be insignificant
for the kinds of information that were sought. For example, we assumed
that among farmers with a given pattern of irrigated crops, their explana-
tions of the relationship between cropping patterns and source of water
would be acceptably consistent and uniform.
Field interviews were undertaken with the farmers in this smaller sam-
ple to obtain the information necessary to construct activity budgets for
rainfed, irrigated, and livestock activities. Because this information re-
quires understanding of complex management processes, we employed
open-ended interviewing techniques that permitted follow-up questions
and the flexibility to follow particular lines of questioning tailored to the
respondents’particular areas of knowledge. This data-collection procedure
maximized the amount of local indigenous knowledge integrated into the
design of the crop budgets.
There are two advantages associated with this approach (Norgaard
1984). First, crop growth is influenced by an extremely large number of
variables in the farming environment. Weather and soil conditions and ap-
plication of inputs by the farmer all influence yields. The extreme com-
plexity of experimental designs to isolate the impacts of all these variables
prohibits this avenue of research in countries such as Cape Verde, which
have limited resources to devote to applied agronomic research. Neverthe-
less, farmers have vast experience in observing relationships between in-
puts and outputs. With this experience, they can make informed judgments
about, for example, the effects of varying the interval between irrigations
on crop yield, assuming all other conditions are normal. This indigenous
knowledge base of farmers provides a low-cost alternative to extensive ex-
perimental programs. Admittedly,farmer responses cannot be accepted un-
questioningly, but they provide the best guesses to use as working hy-
potheses until experimental data do become available.
The second reason for using farmers’estimates of technical relation-
ships is that the goal of the analysis is to understand the decisionmaking
process of farmers. This process draws on the knowledge base of the farm-
ers themselves. (We assume that the knowledge base of farmers is accurate
at least with respect to phenomena they directly observe.) Thus, understand-
ing farmers’ choices of alternative strategies requires in-depth knowledge of
their perceptions of trade-offs between alternatives. Experimental data about
input/output relationships are not available to most farmers in Cape Verde
(indeed, they are not available to researchers) and therefore are not directly
relevant to understanding farmer resource-management strategies.
Several assumptions are implicit in this technique for characterizing
agricultural production technologies. First, there are assumed to be general
Resource Management 29

similarities in production techniques across households, so that informa-
tion gleaned from one farmer can be generalized to other farmers in simi-
lar circumstances. Preliminary observations from the RRAs confirmed that
agricultural production practices are quite homogeneous in Cape Verde.
Crop budgets also incorporate assumptions of fixed-coefficient technolo-
gies and constant returns to scale. In order to model variations in produc-
tion techniques for a given crop, such as differing numbers of weedings
or applications of chemical fertilizers, separate budgets must be con-
structed that reflect these differences. For example, several separate bud-
gets were constructed for each of the irrigated crops, each with a different
irrigation interval and relevant variations in labor inputs and output yields.
The budgets with more frequent intervals have higher yields. The assump-
tion of constant returns to scale is appropriate for the technologies found
in Cape Verde, since there are no significant fixed costs to be averaged
over the quantity produced. The most important capital costs are the main-
tenance of erosion-control structures and the establishment of perennial
crops (e.g., sugarcane, bananas). These costs vary directly with the scale of
operations, so their unit costs are constant.
Analysis of the crop budgets provided insights into farmers’resource-
management decisions, but it also pointed to specific areas for further
fieldwork and analysis. In particular, our attempts to identify representa-
tive production techniques and crop budgets highlighted the importance of
access to irrigation water as a constraint for farmers with irrigated land.
This discovery prompted us to examine more closely the factors that de-
termine access to irrigation water.Adetailed survey of an irrigation net-
work was organized, with the objective of understanding in more detail the
operation of the water-scheduling institutions, the impact of farmers’ac-
cess to water,and the relationship to their strategies for utilization of irri-
gated land. This survey provided detailed information about the schedule
of water deliveries to all network members over the course of a year and
the cropping patterns of each member.Hence comparisons could be made
between water deliveries to individuals and their cropping strategies. Also,
detailed discussions with the network manager and many of the members
provided information on the operational minutiae of network management
and the constraints faced by individuals to secure water at a particular
time.
The final analysis incorporates information from the survey,the bud-
gets, and the ethnographic study of the irrigation networks into the con-
struction of the household models. Estimates of total household income
from all sources were made for the households in the survey sample. The
sample provided information about the quantity of resources available to
each household. Net incomes calculated from representative agricultural
activities—rainfed land, irrigated land, and livestock—were then attributed
30 Resource Management

to households on the basis of their available resources. Nonagricultural in-
come was estimated by multiplying the number of hours of off-farm work
by the average wage. These results provide a basis of comparison for the
welfare levels among different household types, as well as the potential
impacts of changes in resource endowment for specific households. More-
over, the aggregate effects of such changes can be estimated by reference
to the survey data and the relative frequency of a given representative
type.
While the multiyear research period provided us with the opportunity
to complete the sequence of tasks described above, the fact that field in-
formation spans a number of years also presented a problem of compara-
bility of results obtained from different years. Two important sources of
annual variation were accounted for in our analyses. First, because of the
great variation in growing conditions from year to year, the agricultural
production and productivity information obtained from any single year
cannot be used to extrapolate future conditions. For this reason, our esti-
mates of yields used in the crop activity budgets are intended to represent
normal or average growing conditions and are based on several years of
historical data as well as the recommendations of local agronomists and
farmers. By using these synthetic yield estimates, our calculations do not
refer to the specific conditions of a particular year, but represent what can
be expected in a normal year.
The second major factor to account for in comparing results from dif-
ferent years is inflation—the general tendency of all prices in an economy
to increase over time. Because of inflation, prices from different years are
not directly comparable unless the effects of inflation are removed. We
have used 1990 as the basis for all financial calculations. All prices are ad-
justed for inflation to be in real 1990 price levels. In 1990 the exchange
rate of the Cape Verdean escudo (CVE) was 80 to the U.S. dollar (USD).
In sum, the sequential rural research process consists of an ordered set
of information-gathering activities in which each succeeding round of
information collection is directed toward addressing issues or points of un-
certainty identified in previous analyses. In the case of Cape Verde, field-
work was initiated with RRAs in order to acquire a qualitative under-
standing of the problems facing rural households and to gather the
background information necessary to design an appropriate questionnaire
for a formal survey.The next step was to obtain information about the dis-
tribution of resources across rural households through a statistically rep-
resentative survey. This information, in turn, was used to define a small
number of representative farm types, and farms from each category were
selected for more in-depth surveying. Detailed information necessary to
define crop and livestock activity budgets was obtained through interviews
with farmers in this smaller sample. In this interactive research sequence,
Resource Management 31

the socioeconomic analysis of the different representative farm types
raised a number of more specific issues that came to direct further re-
search. As presented in the following chapters, three particular issues
formed central themes of this book: (1) the need to better understand the
institutional mechanisms by which farmers receive access to irrigation
water; (2) the importance of off-farm employment to household welfare
and the need to examine the factors that affect or limit household access to
off-farm work; and (3) the desirability of a more detailed analysis of the
economic incentives for corn and bean production from the perspective of
households given low reported returns to rainfed farming.
Summary
Our analysis of resource utilization by rural households is based on a
methodology that posits that households attempt to maximize the income
from their available resources. With this methodology, we are able to esti-
mate the incomes that households are able generate with their existing re-
sources given the available technologies to use these resources and the in-
stitutional environment that defines how households can acquire, dispose
of, or hire out the services of their productive resources. In addition, the
methodology permits analysis of the effects of alternative policies on
household incomes; thus, incentives for changes in resource-management
strategies can be inferred.
Rural households form the unit of analysis in our examination of the
agricultural resource-management strategies in Cape Verde. These house-
holds face a wide variety of constraints with regard to access to productive
resources for their own use and access to markets or other outlets for their
resources. Analysis must be directed toward understanding the variations
in access to resources among households; the ways in which these varia-
tions lead to different agricultural strategies, and ultimately,the overall in-
comes of households.
To obtain empirical information about rural households in Cape Verde,
asequential rural research approach was developed and applied on Santi-
ago and Santo Antão Islands. This methodological approach is iterative
and interactive, and it is particularly effective when little previous knowl-
edge of rural society is available. The particular form that the analysis
takes depends on specific characteristics of the agricultural sector being
studied and the questions being addressed; however, three categories of in-
formation can be used as a checklist at the beginning of any empirical ex-
ercise of agricultural systems: distribution of resources among households
(or relevant decisionmaking units); characterization of agricultural (and non-
agricultural) technologies that households can employ with their available
32 Resource Management

resources; and understanding the opportunities, or lack thereof, provided
by markets and other resource-transfer institutions to which households
have access. As more knowledge is obtained about a particular system, the
critical problems will fall into one or more of these areas. In the case of
Cape Verde, the key problems have been identified as the absolute lack of
access to productive resources for most households, inefficiencies in the
way that irrigation water is provided to farmers, and restricted access to
off-farm employment opportunities for the workers in rural households.
Notes
1. This framework is inspired by the seminal works of Binswanger and Rosen-
zweig (1986) and Binswanger and McIntyre (1987).
2. Werecognize that households are made up of individuals and that house-
hold strategies for the utilization of available resources are, in fact, outcomes of
some sort of an institutionally constrained bargaining process among household
members. Income and the products generated from the household resources are
similarly allocated to individual family members through complex institutional
processes. The methodological approach followed in this book abstracts from the
explicit analysis of allocations among individual family members to focus on
household-level resource-management strategies.
3. Chayanov (1966) was one of the first writers to focus on the factors that af-
fected resource-allocation decisions within peasant households. More recently,
Becker (1965, 1981) applied neoclassical economic production and consumption
theories to examine household decisions. Subsequently, many economists have
continued to develop this theoretical approach (Pollack 1985; de Janvry,
Fafchamps, and Sadoulet 1991). A number of authors have applied this theoretical
framework in quantitative analyses of production and consumption decisions
within agricultural households (Barnum and Squire 1979; Singh, Squire, and
Strauss 1986).
4. Berry and Cline (1979) find a strong inverse relationship between land pro-
ductivity and area farmed in a wide range of developing countries and argue that
this reflects differences in perceived prices of factors across different sizes of
farms.
5. It is not our intention to ignore the empirically observed scope of intrahouse-
hold bargaining or conflict. The meaning of consensus here is that either the bar-
gaining process has been successful and a compromise has been reached or that one
or another household member has imposed his or her will over the other member(s).
6. An exception to this is the analysis of short-run and long-run price re-
sponses of large, medium, and small dairy producers by Adelaja (1991), who found
that both short-run and long-run price elasticities were different across these three
groups. Another study, by Barnum and Squire (1979), of producers in northwest
Malaysia also incorporated farm characteristics into the model, in this case, farm
size and tenancy status.
7. Budget analysis, however, does not provide quantitative information about
the aggregate market-level response to such changes in prices. Econometric analy-
sis permits more general assumptions about technologies; in particular, it allows
for substitution and complementarity among inputs, as well as for economies of
Resource Management 33

scale. The specific input/output relationships do not have to be known indepen-
dently. However, a large number of observations on prices and market quantities is
needed in order to obtain statistic estimations.
8. To compare the economic efficiency effects of using resources in alterna-
tive activities within an economy, the opportunity costs of family-owned resources
must be included in the profitability calculations (Monke and Pearson 1989). How-
ever, as will be argued in later chapters, the opportunity costs of most resources be-
longing to rural families in Cape Verde—predominantly rainfed land and house-
hold labor services—have very low opportunity costs. Thus, economic profitability
calculations that incorporate the opportunity costs of family resources would not
diverge significantly from the net revenue figures.
9. Antonovitz and Green (1990) compare alternative means of incorporating
risk into econometric models. Based on empirical evidence from the U.S. fed beef
industry, they conclude that no single formulation is preferred to the others, and
thus expectations are heterogeneous rather than homogeneous.
10. Using panel data from households in the United States, Lawrance (1991)
finds that households with lower permanent incomes and education levels have sig-
nificantly higher rates of time discount than households with higher permanent in-
comes and more education.
11. The literature on RRA is rich and varied. In our research, we adopted the
principles expressed in Hildebrand (1981) and van Willigen and Finan (1990).
34 Resource Management

Cape Verde forms the westernmost part of the semiarid Sahelian swath that
traverses sub-Saharan West Africa. Its climate is characteristically Sahelian
and subject to periodic droughts of greater or lesser severity.At the same
time, Cape Verde is somewhat different from its West African neighbors in
terms of its geomorphology.The archipelago is volcanic in origin, its present
landform dating to the middle of the Tertiary period. The islands that
emerged with more recent volcanic activity (i.e., Santiago, Fogo, Brava,
Santo Antão, and São Nicolau) have a distinct topographical relief with peaks
rising to over 1,000 meters in elevation. Steep hillsides and, in some places,
nearly vertical walls shape valleys (known as ribeiras)that are narrow in
their upper reaches but then widen toward the sea. The major ribeiras with
their branches form large watershed systems that encompass many square
kilometers and are recognized as distinct residence zones. The vertical topog-
raphy of these islands increases their vulnerability to erosive processes
caused by rapid surface runoff and wind. However, these are also the islands
that accommodate agricultural activity.The remaining older islands (Maio,
Sal, Boa Vista, and São Vicente) have been worn flat through the millennia
and no longer have well-formed ribeiras. Consequently, their potential for
agricultural development is limited to isolated spots of irrigated farming.
Rainfed agriculture depends entirely on seasonal meteorological pat-
terns that are highly erratic and unpredictable in terms of intensity and
timing. Since Cape Verde has no stored surface water in the form of lakes,
rivers, or reservoirs, irrigated farming is also constrained by the amount of
annual precipitation that recharges groundwater aquifers. In contrast to
other parts of arid West Africa, Cape Verde has no vast rangeland that per-
mits extensive nomadic herding, virtually no wildlife to support hunting
activities, and little in the way of native vegetation that offers quality sup-
plements to local diets. Subsistence systems based on hunting and gather-
ing have never existed in the archipelago.
35
3
The Natural Resource
Endowment

The objective of this chapter is to describe the natural resource en-
dowment and to establish its fragility and vulnerability to population pres-
sure. Since our concern lies with Cape Verdean agriculture, we focus on
those resources directly related to agricultural production, primarily water
and soil.
1With regard to water, the relevant issues become its general
availability for farming and the efficiency of its use in irrigated farming.
In the case of soil, the sustainability of the resource is discussed in relation
to the threat of wind and water erosion. Central to the theme of this book,
we relate the variations in resource availability to the dynamics of resource
use in agricultural activities. The chapter also presents the official policy
concerns about the role of agriculture in the process of environmental
degradation and describes the corrective measures that have been designed
to conserve the resource base.
Rainfall and Climate
Climate is the most dominant yet unpredictable force in Cape Verdean
agriculture. For the majority of the population, the annual rainy season is
adramatic, somewhat awesome uncertainty in the rhythm of everyday life.
In any given year, the annual rainfall pattern determines a situation of
plenty or of hardship for poor families and sets the emotional tenor for the
year. As the rainy season unfolds during the summer months, families
begin to perceive what nature has dealt for that year. In the event of inad-
equate rains, there is apprehensiveness and a sense of urgency in that some
alternative subsistence strategy must be rapidly devised; when rains are
bountiful, there is collective relief.
2
The rains in the archipelago are extremely variable and highly unpre-
dictable in terms of both total quantity of precipitation and spatial and
temporal distribution. The advent of the rainy season in Cape Verde is de-
termined by the dominant meteorological phenomenon that produces an-
nual storms throughout all Sahelian West Africa. The intertropical conver-
gence (ITC), an atmospheric turbulence consisting of a moisture-laden,
unstable air mass, migrates northward from the south Atlantic bringing the
monsoons that deposit rainfall throughout the region (Amaral 1964:50–
53). Specific meteorological variables, such as wind, temperature, and
barometric pressure, determine the annual variations in the size, path, and
velocity of this migration, which in turn determine the length and the in-
tensity of the rainy season. Since the Cape Verdean archipelago lies some-
what north of the usual migration pattern, the islands are vulnerable to the
dramatic results of slight shifts in the normal weather pattern. Interannual
rainfall varies widely around the long-term average, and, in some years,
the islands are totally bypassed. Freeman et al. (1978) report that annual
36 The Natural Resource Endowment

precipitation levels in the capital city of Praia fell below the long-term av-
erage (250 mm) in 62 of the last 102 years. Of greater concern, the coun-
try recorded 58 official drought years from 1719 to 1947, that is, an aver-
age of one every four years (Moran 1982:71). This year-to-year variation
is not predictable, and farmers have few prior indicators—either traditional
or scientific—to orient their adaptive responses. In recent times, the avail-
able pluviometric data reveal a further disturbing reality: annual mean pre-
cipitation levels have been declining since about 1952 (Freeman et al.
1978:43; INGRH 1993:8). And, since the late 1960s, the archipelago has
been assailed by near-constant drought, spurring a reevaluation of what
constitutes a “normal” year of precipitation.
Another consequence of the ITC is the marked seasonality of rainfall
and the extreme intra-annual variation. Between 60 and 80 percent of Cape
Verde’s rainfall occurs between August and October, and precipitation be-
tween March and June is very rare. The actual number of days that regis-
ter precipitation is low (from about 20 days in the more arid regions to
about 50 days in the higher elevations). Nevertheless, individual storms
can be extremely violent with rainfall exceeding 100 millimeters per day
(Moran 1982:67). It is not uncommon for most of the annual rainfall to
occur in one or two stormy episodes so intense that older farmers unable
to read calendars often date life history events by reference to these par-
ticularly fierce periods. The concentration of rainfall within a very narrow
time span creates several problems for farmers. The uneven distribution of
moisture throughout the growing season can result in what is locally
known as the seca verde(green drought). Early rains summon agricultural
activity and promote strong initial vegetative growth, but the subsequent
absence of moisture during critical periods in the plant cycle inhibits fruit
or seed development. Thus, in some years, above-average annual precipi-
tation levels still do not prevent poor agricultural outcomes. At the same
time, torrential storms deposit more water than the ground can effectively
absorb, which results in the accelerated, high-volume flooding that uproots
young crops and unearths ungerminated seeds. Such downpours also exact
asevere long-run toll in terms of soil loss.
The vertical land-relief characteristic of the agricultural islands pro-
duces a pronounced orographic effect and wide spatial variations in rain-
fall during any given year. Whereas there are unpredictable interisland
variations in rainfall due to varying storm tracks, the regional differences
in annual precipitation on any island are significant and generally pre-
dictable. The prevailing winds in Cape Verde come from the north and
northeast, and the rain-bearing clouds release their moisture upon contact
with vertical landforms. Precipitation increases at a nearly constant rate
with increases in elevation, while the coefficient of variation around mean
annual rainfall decreases. Consequently,in most years, the orographic
The Natural Resource Endowment 37

effect dictates that the higher elevations facing the northeast will receive
more moisture. Conversely, all the coastal areas, as well as the ribeiras that
debouch toward the southwest, are characterized by greater levels of aridity.
Table 3.1 provides a quantitative demonstration of this orographic ef-
fect on the four principal agricultural islands from 1970 to 1988. On San-
tiago and Fogo, the leeward islands, annual averages can vary from about
600 millimeters at higher altitudes to less than 200 millimeters at lower
levels. The differences—while somewhat reduced—are also dramatic on
the agricultural windward islands. Spatial variation has a corresponding
impact on local vegetation and agricultural production potential, thus dis-
tinguishing three distinct agroclimatic zones: humid, subhumid, and semi-
arid. This standard classification is recognized in most agricultural plan-
ning and intervention strategies.
Other climatological characteristics show less variation relative to
rainfall. For example, temperatures average about 25 degrees Celsius dur-
ing the year,with an average maximum of about 34 degrees and an aver-
age minimum of approximately 16 degrees (Freeman et al. 1978:53). Al-
though prevailing winds blow from the north and northeast, a seasonal
wind (the harmattan)offthe African continent occurs between January
and May and carries sight-obscuring dust particles concentrated enough to
disrupt normal air travel schedules. Humidity and insolation are mostly
uniform throughout the year.
Considered in total context, the Sahelian climate of Cape Verde shapes
and constrains the nature of agriculture. First of all, Cape Verdean farmers
operate under conditions of extreme year-to-year risk. There seems to be no
proven method—neither scientific knowledge nor local lore—for accurately
anticipating a drought. Even in a wet year (on record), the rainy season may
be dangerously confined to one or two violent storms that wreak substantial
damage on the countryside but do not provide sufficient moisture through-
out the agricultural growth cycle. Second, there is highly variable local dis-
tribution of rainfall, not only from island to island and region to region, but
also from one side of a ribeira to another. Over the centuries, rural people
have developed, and continue to revise, local management strategies for
adapting to these conditions. Nonetheless, the downward trend in rainfall
since the 1960s has placed further pressure on an already precarious liveli-
hood by reducing the productivity of rainfed agriculture, limiting irrigated
agriculture, and lowering the quality of livestock grazing areas.
Water Resources
The geohydrology of Cape Verde is as complex as its varied landforms and
still awaits the systematic research necessary for a comprehensive evaluation
38 The Natural Resource Endowment

Random documents with unrelated
content Scribd suggests to you:

Section 4. Information about Donations to
the Project Gutenberg Literary Archive
Foundation
Project Gutenberg™ depends upon and cannot survive without
widespread public support and donations to carry out its mission
of increasing the number of public domain and licensed works
that can be freely distributed in machine-readable form
accessible by the widest array of equipment including outdated
equipment. Many small donations ($1 to $5,000) are particularly
important to maintaining tax exempt status with the IRS.
The Foundation is committed to complying with the laws
regulating charities and charitable donations in all 50 states of
the United States. Compliance requirements are not uniform
and it takes a considerable effort, much paperwork and many
fees to meet and keep up with these requirements. We do not
solicit donations in locations where we have not received written
confirmation of compliance. To SEND DONATIONS or determine
the status of compliance for any particular state visit
www.gutenberg.org/donate.
While we cannot and do not solicit contributions from states
where we have not met the solicitation requirements, we know
of no prohibition against accepting unsolicited donations from
donors in such states who approach us with offers to donate.
International donations are gratefully accepted, but we cannot
make any statements concerning tax treatment of donations
received from outside the United States. U.S. laws alone swamp
our small staff.
Please check the Project Gutenberg web pages for current
donation methods and addresses. Donations are accepted in a
number of other ways including checks, online payments and

credit card donations. To donate, please visit:
www.gutenberg.org/donate.
Section 5. General Information About
Project Gutenberg™ electronic works
Professor Michael S. Hart was the originator of the Project
Gutenberg™ concept of a library of electronic works that could
be freely shared with anyone. For forty years, he produced and
distributed Project Gutenberg™ eBooks with only a loose
network of volunteer support.
Project Gutenberg™ eBooks are often created from several
printed editions, all of which are confirmed as not protected by
copyright in the U.S. unless a copyright notice is included. Thus,
we do not necessarily keep eBooks in compliance with any
particular paper edition.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.
This website includes information about Project Gutenberg™,
including how to make donations to the Project Gutenberg
Literary Archive Foundation, how to help produce our new
eBooks, and how to subscribe to our email newsletter to hear
about new eBooks.

Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.
More than just a book-buying platform, we strive to be a bridge
connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.
Join us on a journey of knowledge exploration, passion nurturing, and
personal growth every day!
ebookbell.com

Waiting For Rain Agriculture And Ecological Imbalance In Cape Verde Mark Langworthy Timothy J Finan

About This Presentation

Slide Content

Tags

Categories

Download

Quick Actions

Statistics

Related Slideshows

Waiting For Rain Agriculture And Ecological Imbalance In Cape Verde Mark Langworthy Timothy J Finan

About This Presentation

Slide Content

Slide 1

Slide 2

Slide 3

Slide 5

Slide 7

Slide 8

Slide 9

Slide 11

Slide 13

Slide 14

Slide 15

Slide 17

Slide 18

Slide 19

Slide 20

Slide 21

Slide 22

Slide 23

Slide 24

Slide 25

Slide 26

Slide 27

Slide 29

Slide 30

Slide 31

Slide 32

Slide 33

Slide 34

Slide 35

Slide 36

Slide 37

Slide 38

Slide 39

Slide 40

Slide 41

Slide 42

Slide 43

Slide 44

Slide 45

Slide 46

Slide 47

Slide 48

Slide 49

Slide 50

Slide 51

Slide 52

Slide 53

Slide 54

Slide 55

Slide 56

Slide 57

Slide 58

Tags

Categories

Download

Quick Actions

Statistics

Related Slideshows

8-top-ai-courses-for-customer-support-representatives-in-2025.pptx

7-essential-ai-courses-for-call-center-supervisors-in-2025.pptx

25-essential-ai-courses-for-user-support-specialists-in-2025.pptx

8-essential-ai-courses-for-insurance-customer-service-representatives-in-2025.pptx

Know for Certain

PPT OPD LES 3ertt4t4tqqqe23e3e3rq2qq232.pptx