SUSTAINABILITY OF STEEP LAND BEAN PHASEOLUS VULGARIS L.)
FARMING IN COSTA RICA:
AN AGRONOMIC AND SOCIO-ECONOMIC ASSESSMENT
BARBARA CAROL BELLOWS
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
u81VRZSITY OVFLODR UFRWR
Barbara Carol Bellows
This work is dedicated to those who taught me what sustainability means:
Teeboy, Lito, and Jopy, who killed and died trying to liberate the oppressed
Filipino peasant farmers.
The South-East Asian Social Rural Leadership Institute class of 1988, who put a
human face on sustainability.
Amor, Ramiro and other farmer innovators, whose contributions to agricultural
technology development are too often ignored.
The children of Costa Rica's small-scale farmers, whose future depends on
sustainable agricultural development.
A dissertation is the original work of a single author. But, behind the name on
the title page are the knowledge, efforts, and encouragement of professors, assistants,
friends, and family. This is especially true for on-farm studies. The success, or failure,
of these studies depends on the cooperation of farmers, neighbors, and farm laborers.
In acknowledgement of their roles as interpreters, teachers of much of my
knowledge about beans, chauffeurs, farm laborers, interview assistants, and friends, my
thanks and appreciation are given to Danilo Monge Granos and Jorge HernA Sandi
Camacho. Without them, my work would have been much more difficult and my stay
in Pejibaye much less pleasant.
Although the involvement of farmers in the experiments was limited, their
willingness to let me use their land, to respond to detailed interviews, and to provide
reactions to the experimental practices employed was essential to the success of this
study. My thanks are given to Santiago Jimdnez, Rodolfo Guadamos, Thederico Mora,
Antonio Quiros, Humberto Morin, Juan Carlos Hijos, Nemicio Mena, Francisco
Barboza, Manuel Zuniga, Rigoberto Zuniga, Tulio Castro, Martin Castro, Eliezer
Castro, Raul Mendez, Hugo Mora, Hipaulito Aguilar, Laurencio Campos, and Giovani
Camacho Quesada. Special thanks are given to Ramiro Morales, an innovative farmer
and cherished friend. He fulfills the classical description of the model farmer and holder
of indigenous knowledge systems. His own initiatives and experimentation provided me
with some of the basic concepts for my own experiments.
Appreciation is given to "Macho," the Pejibaye postman, who let me use the
postal scale for weighing my plant and erosion run-off samples. Thanks are also given
to Eric Gamboa and Carlos Dfas, the Pejibaye CNP and MAG representatives,
respectively, for orienting me to the farming community in Pejibaye. My love and
appreciation is also extended to Ing. Alice Zamora for her unflagging assistance in
helping me to establish my connections with various organizations in Costa Rica.
For the use of their laboratory facilities, I thank Ing. Oscar Acculia, Ing. Rafael
Salas, and Gissel Alvarado of the University of Costa Rica Center for Agricultural
Investigations. For insightful guidance regarding nitrogen mineralization studies and for
the use of her laboratory facilities, I extend my deepest thanks to Dr. Maria Julio Mazza-
rino of CATIE. I also want to thank Dr. Donald Kass, Dr. Roberto Dfas, and Jorge
Fustino of CATIE for their intellectual guidance.
For her moral and intellectual support, I am deeply indebted to the other "frijol
tapado Wisconsinite," Dr. Martha Rosemeyer from the Organization for Tropical Studies.
For introducing me to the topic of frijol tapado, for processing hundreds of my soil and
plant samples, and for her insightful feedback on the draft of the introduction and
socioeconomic chapters of my study, my sincere appreciation is extended to Dr. Judy
Kipe-Nolt of CIAT. Thanks are due also to Dr. William Jansson of CIAT and Ing. Eric
Borb6n of CIMMYT for their helpful suggestions in the preparation of interview
For uncomplainingly interupting her own research to run innumerable errands and
perform other tasks beyond the call of duty, I am grateful to Dr. Sylvia Coleman.
Although physical distance severely hindered our communication throughout my
stay in Costa Rica, I am indebted to the patient and insightful guidance provided to me
by the members of my dissertation committee: Dr. Brian McNeal, Dr. Ken Buhr, Dr.
Peter Hildebrand, and Dr. P.K. Nair. Special thanks are given to my major professor,
Dr. David Hubbell, who, despite my having wandered far afield from his specialty,
provided me with financial, administrative, and moral support.
My love and thanks to my parents, John and Florence Bellows, for their support:
emotional, financial, and spiritual. It is truly due to them that I could follow through on
Goethe's Challenge: "Whatever you can do, or dream you can, begin it."
Last, but not least, for his unpaid labor, ability to be the practice to my theory,
and his heavenly cooking, my never-ending thanks and love to my husband, Fernando
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ................................. iv
LIST OF FIGURES ............. ...... .................. xii
LIST OF TABLES ................................... xiv
KEY TO ABBREVIATIONS ............................... xvii
ABSTRACT ......................................... xviii
1 THE EFFECT OF TECHNOLOGY CHANGE ON THE SUSTAINABILITY
OF BEAN PRODUCTION IN COSTA RICA ..................
Sustainability ................................... 1
Sustainability: The Concept .................. ...... 1
The Implementation of Sustainable Agricultural Programs ...... 3
Agroecological Sustainability ....................... 6
Sustainability, Induced Technology Change, and Bean Farming
in Costa Rica ................................ 8
Bean Farming Methods in Costa Rica ................. ... 12
Introduction .................................. 12
Frijol Tapado ............................... 13
Frijol Espeque ................................ 18
The Transition From Frijol Tapado to Frijol Espeque ........ 20
The Assessment of Sustainability ....................... 23
Consequences of Degradation ....................... 23
Parameters for Assessing Sustainability ................. 25
Economic Measurements of Sustainability ............... 27
Assessing the Sustainability of Bean Growing Methods on
Steeply Sloping Lands in Costa Rica .................. 28
Research Objectives ................................ 30
2 PEJIBAYE DE PEREZ ZELEDON: A MICROCOSM FOR THE
ASSESSMENT OF THE EFFECTS OF TECHNOLOGY CHANGE ON
SMALL-SCALE FARMER PRACTICES IN COSTA RICA ......... 39
An Introduction to Pejibaye de Perez Zeled6n ............... 39
Reasons for the Choice of Pejibaye as a Study Location ....... 39
Description of the Study Area ....................... 41
Changes in Land Use and Bean Growing Practices ............. 42
Causes and Consequences of Land Use Intensification ....... 42
Government Programs to Increase Bean Production ........ 44
Adoption of Introduced Bean Varieties ................ 45
Bean Production Problems ........................ 47
Study Objectives .................................. 48
Methodology ....... ..... ................. ....... 49
Surveys ..... ............ .................. 49
Measuring the Economic Impact of Technology Change ........ 51
Statistical Analyses ............................ 52
Results and Discussion .............................. 52
Overview of Farming Practices ...................... 52
Tenants .......... ........ .................. 54
Farmer-Landowners ............................ 57
Frijol Espeque ................................ 58
Frijol Tapado ................................. 59
The Transistion from Frijol Tapado to Frijol Espeque ........ 61
Economic Aspects of Frijol Espeque and Frijol Tapado ....... 64
Conclusions ............ ................... ...... 66
Land Tenancy and the Adoption of Frijol Espeque .......... 66
Complementary and Competitive Interactions between Frijol
Tapado and Frijol Espeque ....... ................. 68
Limitations to Increasing the Profitability of Bean Farming ..... 69
Options for Increasing the Profitability of Bean Farming ...... 70
3 BEAN GROWING ON STEEP LANDS: RELATIONSHIPS BETWEEN
LAND USE, BEAN YIELDS, AND SUSTAINABILITY ........... 88
Sustainability of Introduced Technologies ................... 88
Soil Degradation .................................. 90
Soil Erosion ............... ................ 90
Intensity of Land Use .............. ... ........... 93
Burning of Residues .......... .................. 94
Consequences of Soil Degradation .................... 95
Field Assessment of the Effect of Technology Change on Soil
Degradation ....... .................... .
Study Design ..........................
Assessment of Agronomic Variables ............
Materials and Methods .......................
Field Analyses .........................
Statistical Analysis .......................
Results and Discussion .......................
Rainfall .... .........................
Variability Among Farms ..................
Parameters of Sustainability .................
Parameters of Degradation .................
Effects of Land Use Practices ................
.. . . . . 108
4 THE EFFECT OF LAND-USE INTENSITY AND PLANTING METHODS ON
BEAN YIELDS AND SOIL DEGRADATION .................. 124
Use of Modified Stability Analysis to Compare Frijol Tapado and
Frijol Espeque .................................. 124
Difficulties of Comparing Frijol
The Modified Stability Analysis
Tapado and Frijol Espeque .... 124
Method and On-farm Trials .... 126
Materials and Methods .......... .................... 130
Field Analyses ................................ 130
Statistical Analyses .............................. 134
Results and Discussion .............................. 135
Environmental Factors ........................... 135
Soil Analyses ................................ 135
Plant Analyses ................................139
Y ields ..... ... .. .. ............ .. ... ... 142
Soil Erosion .. ............... ..... .... .......145
Technology Adoption ........................... 149
.. . . . . 15 1
The Use of Modified Stability Analysis in the Analysis of
On-Farm Experiments ......................
Bean Planting Practices and Sustainability ..........
5 THE EFFECTS OF LAND FALLOWING, LAND PREPARATION
METHODS,AND PHOSPHORUS FERTILIZATION ON FRUOL ESPEQUE
BEAN YIELDS AND SOIL DEGRADATION ................. 173
Introduction ..................................... 173
Effects of Residue Burning and Mulching on Nutrient Availability and
Soil Degradation ................................. 174
Effect of Fallowing and Land Preparation Phosphorus Availability .... 176
Materials and Methods .............................. 178
Treatment Installation ............................ 178
Soil Analyses .................................. 180
Symbiotic Relationships with Bean Roots ............... 183
Harvest Analyses .............................. 183
Labor Inputs ................................. 183
Statistical Design .............................. 184
Results ............. ........................... 184
Rainfall ...... ..............................184
Nutrient Inputs from Mulches and Ash . ... 184
Soil Analyses ................................ 185
Mid-season Analyses ................. ........ 187
Harvest Analyses ............................. 188
Discussion ................ .. .................. 190
Land Degradation .............................. 190
Effect of Environmental Factors on Yields .............. .192
Nutrient Recycling and Nutrient Losses in the Soil System ..... 193
Estimate of Profitability of Bean Production ............ 194
Conclusions ............. ........................ 195
6 THE SUSTAINABILITY OF BEAN FARMING ON STEEP LANDS:
EVALUATION OF THE CONSEQUENCES OF TECHNOLOGY
INTRODUCTION ON THE AGROECOSYSTEM AND ON THE FARMING
COMM UNITY .................................... 218
Introduction .............. .................. ... .. 218
Trade-offs Between Productivity and Sustainability in Bean
Production ........... .......... ......... 220
Impact of Technology Introduction on Agroecosystem
Sustainability .......................... .. 221
The Development and Evaluation of the Sustainability of Agricultural
Recommendations ...... ......................... 223
Factors Affecting Sustainability ..................... 223
Developing Agricultural Recommendations .............. 224
Analyzing Agricultural Recommendations ............. 225
Analyses of Productivity and Degradation ............. 227
Reprioritizing Agroecosystem Values: Placing Equity
First ................. ...................... 229
LITERATURE CITED ................... .............. 233
A: INTRODUCTORY SURVEY ....................... 249
B: FRUOL ESPEQUE ECONOMIC SURVEY ............... 255
C: FRIJOL TAPADO SURVEY ....................... 258
D: DATA FOR EXPERIMENTS DESCRIBED IN CHAPTER 4 .... 264
E: DATA FOR EXPERIMENTS DESCRIBED IN CHAPTER 5 ..... 274
BIOGRAPHIC SKETCH ................... ............. 284
LIST OF FIGURES
1.1 Schematic representations of the four bean growth habits ........ ... 33
1.2 Bean growing areas in Costa Rica . ... ...... 34
1.3 Frijol tapado and frijol espeque planting methods . ..... 35
1.4 Agroeocological and socioeconomic interactions affecting bean
production on steep lands in Costa Rica . ..... 36
2.1 The location of Pejibaye de Per6z Zeled6n . ..... 74
2.2 Rainfall for 1990 at two elevations in Pejibaye de Per6z Zeled6n ..... 75
2.3 Deforestation of the Pejibaye de Per6z Zeled6n-1973 . .... 76
2.4 Calendar of agricultural activities and school semesters in Pejibaye de
Perez Zeled6n in relationship to rainfall . ..... 77
2.5 Common plants found in frijol tapado fallows . ..... 78
2.6 Frijol tapado and frijol espeque decision tree . ..... 79
2.7 Socioeconomic interactions affecting the sustainability of bean
production ............ ........... ..... .......... .. 80
2.8 Changes in the prices of beans and fertilizer and in the colon:dollar
exchange rate: 1975-1991. ................... .......... 81
3.1 Location of farms included in the study in relationship to Pejibaye 111
3.2 Rainfall, agricultural activities, and sampling times . .... 112
3.3 Soil losses due to erosion as affected by farm and sampling
period ................... ..................... 113
4.1 Location of farms included in the study in relationship to Pejibaye 155
4.2 Rainfall, agricultural activities, and sampling times . .. 156
4.3 Soil physical properties as affected by treatment . ..... 157
4.4 Bean yields as analyzed by ANOVA and MSA ............... 158
4.5 Classification of treatments based on the relationship between MSA
regression coefficients and treatment mean yields ............. 159
4.6 Relationships of fallow duration and phosphorus uptake to
environmental indices ................... ........... 160
4.7 Regressions of soil calcium, at planting, on soil losses due to erosion 161
4.8 Soil losses due to erosion: Analyses based on ANOVA and MSA .... 162
5.1 Location of farms included in the study in relationship to Pejibaye 198
5.2 Erosion runoff plot design ................. ......... ..199
5.3 Rainfall, agricultural activities, and sampling times . .. 200
5.4 Cumulative soil losses during the growing season . .... 201
5.5 Soil physical properties ........... ..... ............. 202
5.6 Time required to weed treatments at two weeks after planting and weed
infestation at harvest ................... ............ 203
6.1 Contradicting effects of government programs on bean production
sustainability ................. .................. 231
LIST OF TABLES
1.1 Comparisons of frijol tapado and frijol espeque ................. .37
1.2 Agroecosystem factors of social value for frijol tapado
and frijol espeque ..................................38
2.1 Summary of surveys conducted ......................... 82
2.2 Characterization of tenants and farmer-landowners in Pejibaye de
P6rez Zeled6n ..................................... 83
2.3 Input use in frijol espeque and frijol tapado . ..... 84
2.4 Plant species commonly found in frijol tapado fallows . .... 85
2.5 Comparisons of frijol espeque production systems: Pejibaye and
San Vito/Guagural ................... ............... 86
2.6 Economic aspects of frijol espeque and frijol tapado . .... 87
3.1 Physical description, prior land use, and land preparation methods
used on farms included in the study ...................... 114
3.2 Input use on farms included in the study . .... ....... 115
3.3 Soil characteristics of farms included in the study . .... 116
3.4 Yield, erosion, nodulation, and pest infestation data for farms
included in the study ................... ........ ..... 117
3.5 Soil phosphorus, microbial biomass, organic matter losses due
to erosion, and plant nutrient uptake at harvest. . 118
3.6 Economic data for farms included in the study . ..... 119
3.7 Regressions of soil, plant, and management factors on yields ... 120
3.8 Regressions of environmental and management factors on
net returns ....................... ..............
3.9 Multiple regressions of soil, environmental, and crop management
factors on yields ................................
3.10 Comparisons of land-use practices: t-test analyses ... .. ...
3.11 Effect of land-use practices: mean values for variables with significant
relationships in t-tests ..............................
4.1 Physical description, prior land-use, and predominant fallow species of
experimental farms ..............................
4.2 Soil characteristics of experimental farms at planting . .
4.3 Land preparation and planting treatments . . ... 16
4.4 Soil nutrients and organic matter at planting . ..... 166
4.5 Soil nutrients and organic matter at harvest . ..... 167
4.6 Nitrogen availability as determined by nitrogen mineralization
and microbial biomass ................... ..... ...... 168
4.7 Nodulation by rhizobia, nitogen uptake, and phosphorus uptake ... 168
4.8 Correlation coefficients, by treatment, between yields and soil,
environmental, and nutrient uptake factors ....... ......... 169
4.9 Analysis of variance of yield including environmental index
by treatment ................... ...... ........... 170
4.10 Correlation and regression coefficients for treatment yields
against environmental index ......................... 170
4.11 Correlation coefficients, by treatments, between erosion and
soil factors ...................................... 171
4.12 Analysis of variance of erosion including environmental index by
treatment .............. ........................ 171
4.13 Economic costs and returns from bare-field frijol espeque,
labranza cero, and frijol tapado ................... ..... 172
5.1 Physical characteristics, prior land-use, and predominant fallow
species of experimental farms ................... ...... 204
5.2 Soil characteristics of experimental farms . ..... 204
5.3 Soil physical characteristics . . . ... 205
5.4 Soil organic matter and pH at planting and harvest . .... 206
5.5 Phosphorus availability throughout the growing season . .. 207
5.6 Nitrogen mineralization throughout the growing season .... 209
5.7 Microbial biomass throughout the growing season . ..... 211
5.8 Midseason plant analyses ............................. 213
5.9 Harvest analyses .................................. 2 215
6.1 Agroecosystem factors of social value for frijol tapado, labranza
cero, and frijol espeque, as practiced on steep lands . ... 232
KEY TO ABBREVIATIONS
BNCR. ........... National Bank of Costa Rica
CATIE .......... Center for Tropical Agricultural Investigation and
CGIAR .......... Consultative Group on International Agricultural
CIAT ............ International Center for Tropical Agriculture
CIMMYT ......... International Center for the Improvement of Maize
CNP ........... National Production Council
FSR/E .......... Farming Systems Research and Extension
HYV ........... High Yielding Varieties
ICRAF .......... International Centre for Research in Agroforestry
IICA ........... International Institute for Agricultural Cooperation
LDC ....... ... Lesser Developed Countries
MAG ........... Ministry of Agriculture
MIDEPLAN ....... Ministry of Planning
ONS ........... National Seed Office
UCR ........... University of Costa Rica
USAID .......... United States Agency for International
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
SUSTAINABILITY OF BEAN (PHASEOLUS VULGARIS L.)
FARMING ON STEEP LANDS IN COSTA RICA:
AN AGRONOMIC AND SOCIOECONOMIC ASSESSMENT
BARBARA CAROL BELLOWS
Chairman: Dr. David H. Hubbell
Major Department: Soil Science
Changes in bean-production practices by farmers cultivating steep lands in Costa
Rica have impacted on agroecosystem sustainability. Frijol espeque, an input-based,
land-use intensive, dibble-stick-planted, bean-growing method, was introduced to replace
frijol tapado, a broadcast-planted, traditional, low-labor, shifting agricultural practice.
Socioeconomic impacts of the transition were assessed based on farmer surveys.
Agronomic impacts of the transition were determined using on-farm experiments.
Frijol espeque produced significantly higher yields but lower returns on
investment compared to frijol tapado. Clear-field and burned-field land preparation
practices, used in frijol espeque, resulted in soil degradation as indicated by significantly
greater soil losses due to erosion, depletion of plant nutrients, and increased soil bulk
density. Soil erosion was controlled, aggregate stability maintained, and plant nutrients
conserved in the mulch-based frijol tapado system.
Frijol espeque and frijol tapado were compared with labranza cero, an
intermediary bean production method. Labranza cero is planted in straight rows using
a dibble-stick, similar to frijol espeque, but has cut, fallow growth retained on the soil
surface as a mulch, similar to frijol tapado. Labranza cero produced similar yields to
clear-field frijol espeque while controlling soil degradation to an extent similar to frijol
Farmers have decreased the use of fallow rotations due to land-use competition.
Increasing land-use intensity resulted in significantly decreased phosphorus availability,
increased erosion, increased weed and web blight infestation, and decreased yields. For
labranza cero, crop residues alone provided an insufficient mulch layer to decrease soil
erosion losses, weed regrowth, or web blight infestation. Use of a grass fallow, in
combination with mulch-based frijol espeque, controlled soil erosion and increased the
availability of nitrogen and phosphorus for plant uptake.
Both agronomic and sociopolitical factors are responsible for the decreasing
sustainability of bean production on steep lands in Costa Rica. Agricultural development
programs need to address and assess the impacts of farmer adaptations of introduced
technologies and changing land-use practices on agroecosystem sustainability and equity
within the society. Assessment methods based on a combination of modified stability
analyses and ANOVA were shown to be effective in evaluating both the productivity and
the sustainability of the agroecosystem.
THE EFFECT OF TECHNOLOGY CHANGE ON THE SUSTAINABILITY
OF BEAN PRODUCTION IN COSTA RICA
Sustainability: The Concept
Agricultural sustainability is a concept that encompasses the interaction among
agriculture, household economics, the environment, society, and agricultural policies.
As a result of the complexity and the temporal aspect of this concept, definitions of
sustainability are often vague or contradictory. Many agricultural researchers and devel-
opment workers, frustrated in their attempts to define the concept, simply state that
agricultural sustainability is "understood intuitively."
Practically, sustainable agriculture is used to describe agricultural production or
development activities that can be adopted easily by small-scale farmers. From the
agroecosystem perspective, sustainable agriculture encompasses practices that have a
minimal negative impact on the environment, rely predominantly on nutrient cycling and
green manures for the maintenance of soil fertility, and promote systems diversity for
pest and disease control. From the perspective of production, sustainable agriculture
balances increased yields against risk investment rather than trying to obtain maximum
yields. Agroforestry, multiple cropping, and mixed crop-livestock systems are consid-
ered more sustainable than monocropping and sole cropping systems. The latter are
regarded as agronomically more fragile and economically riskier. From the perspective
of equitability, proponents of sustainable agriculture argue that sustainable agricultural
production by resource-poor farmers today, and by their children tomorrow, can exist
only if issues of land tenure, birth control, social security, economic development, and
natural resource exploitation are addressed.
Conway (1985) defined agroecosystem sustainability as "the ability of the system
to maintain productivity in spite of a major disturbance such as caused by intensive
stress or a large perturbation." In his analysis, sustainability is but one of four
properties of the social value of an agroecosystem. Besides sustainability, the other
properties are: productivity, the measurement of production; stability, the coefficient of
variation of productivity; and equitability, how production is shared. Although some
agroecosystems exhibit high levels of all four properties, Conway (1991) argues that, in
other situations, sustainability may need to be sacrificed to obtain higher levels of
productivity or equitability.
In their definition of sustainability, the Technical Advisory Committee of the
Consultative Group on International Agricultural Research (CGIAR) attempted to inte-
grate economic growth and productivity with sustainable land-use practices. They stated
that sustainable agriculture "should involve the successful management of resources for
agriculture to satisfy changing human needs while maintaining or enhancing the quality
of the environment and conserving natural resources" (CIMMYT, 1989). According to
this definition, sustainable systems may be difficult to implement. Although sustainable
systems are more productive in the long term, they may be less economically viable in
the short run (Barbier, 1989).
Sustainability also has been defined in terms suitable for formulating development
policies. The World Commission on Environment and Development (1987) report, also
known as The Brundtland Report, defined sustainable development as "development that
meets the needs of the present without compromising the ability of future generations to
meet their own needs." A member of the Brundtland Commission indicated that
achieving sustainable development requires a change in government policies and
expenditures (MacNeill, 1989). Policies should favor programs designed to protect crop-
lands, slow or reverse deforestation, slow population growth, increase energy-use
efficiency, develop renewable resources, and decrease the debt burden on developing
The Implementation of Sustainable Agricultural Programs
Interest in agricultural sustainability arose in response to what were perceived to
be unsustainable agricultural practices. In the United States, rising costs of fertilizer
inputs and heightened awareness of adverse effects of pesticides on non-target organisms
(Carson, 1964) resulted in a renewed interest in the use of green manures, multiple
cropping, and integrated pest control (USDA, 1980). In lesser developed countries
(LDCs), interest in sustainable agriculture arose in response to what was perceived to be
the unsustainability and inequity of the "green revolution" (George, 1977; Lappl and
The green revolution centered on the development and distribution of seeds of
maize, wheat, and rice exhibiting high yield response to agrochemical and labor inputs.
The introduction of these high yielding varieties (HYVs) significantly increased the
worldwide production of basic food grains and transformed several nations from
importers to exporters (Brown, 1970). On the farm level, however, marginal farmlands,
inability to afford inputs, low access to credit and markets, and insecure or usurious
tenure agreements hindered adoption of the green-revolution technologies by many
farmers (Roling et al., 1976; Fliegel, 1984). As a result, the impact of the green
revolution was to accentuate differences in the production capacities of rich and poor
farmers. Farmers with access to better land and agrochemical inputs could realize the
higher yield potential offered by green-revolution varieties. Resource-poor farmers
working marginal land and unable to apply fertilizers and pesticides were unable to
realize the higher yields of these input-dependent varieties.
To include in the development process those small-scale or resource-poor
farmers, who were unable to adopt the technologies of the green revolution, farming
systems research and extension (FSR/E) programs were developed. In contrast to
traditional transfer-of-technology extension methods, criticized as being researcher-
driven and commodity-oriented, the farming system approach was designed to be
interdisciplinary, whole-farm oriented, and based on farmer involvement (Shaner et al.,
1982). Farming system development is initiated by conducting informal surveys or
sondeos with farmers to determine their perceived needs and constraints (Hildebrand,
1981). A broad analysis of the farming system is conducted prior to conducting
experiments designed to satisfy the needs or alleviate the constraints expressed during
surveys. In this process, problems are identified and prioritized. The causes, both envi-
ronmental and socioeconomic, of the higher-priority problems are identified as are
potential solutions (Tripp and Woolley, 1989). Testing of the potential solutions is
conducted in conjunction with participating farmers under their constrained conditions.
Experiments are conducted on farmers' fields, often utilizing farmer labor and manage-
ment, rather than on experiment stations. By testing new technologies within the
variable environment of the farmers' fields rather than within the optimal environment
of the experiment station, appropriate recommendation domains for the technologies can
be identified (Harrington and Tripp, 1984; Chambers and Ghildyal, 1985). Farmers are
also involved in the evaluation process. As Rhoades and Booth (1982) noted, farmers
may not adopt recommended practices for reasons that may appear illogical to the
researcher but which are logical within the context of the culture or socioeconomic
environment of the farmer. Alternatively, farmers may not adopt an innovation
completely. Instead, they may conduct their own adaptive experiments until they have
developed a creative integration of the technology with their own traditional methods.
Proponents of the "farmer-first" development paradigm criticize the original
FSR/E methods as involving farmers but not being farmer-driven. Instead of incorporat-
ing farmers' opinions into researcher-designed development programs, farmer-first
developers seek to involve farmers, both men and women, in their own development
(Chambers and Jiggins, 1987a; 1987b). Designed specifically for resource-poor
farmers, this approach does not seek to replace traditional agricultural practices but
instead to use these practices as a basis for the development of new technologies (Roling
and Engel, 1989; Thrupp, 1989).
The policies of the green revolution, criticized as being non-egalitarian and not
conducive to sustainable program implementation, were also condemned as being
environmentally degrading. The wide-spread adoption of HYVs resulted in a drastic loss
of native germplasm (Cleveland and Soleri, 1989). Uncontrolled use of pesticides, many
of which are banned in the industrialized countries due to high toxicity against non-target
organisms, has caused pollution and health problems in LDCs. Soil degradation has
occurred when farmers adopted land-use intensive farming practices but were financially
unable to use fertilizers to replace nutrients which were removed due to land clearing
and harvest. Erosion has resulted when technologies developed for level lands are inap-
propriately practiced on steeply sloping lands (Eckholm, 1976; Lal, 1984a).
Traditional agricultural systems use fallows and mixed agricultural systems to
maintain soil fertility and to inhibit pest and disease infestations. Agroecologically, the
sustainability of these practices is based on nutrient recycling and system diversity.
When farmers replace traditional rotational systems with input-based systems, they
substitute chemical fertilizers and pesticides for the biological control mechanisms
provided by fallows and mixed cropping systems. In this process, the basis for system
sustainability shifts from environmental to economic. The resilience of the system is no
longer buffered by recycling processes combined with parasitic and mutualistic
interactions between system components. Instead, it is buffered by the ability of the
farmer to obtain the cash required for inputs.
The paradigm of agroecology was developed to study and develop low-input
management systems appropriate for use by resource-poor farmers. Agroecology recom-
mendations are designed to increase diversity among and within the enterprises of the
farm. System diversity enhances efficiency of nutrient recycling, moderates pest and
disease infestations, and increases efficiency in the use of moisture, sunlight, and
nutrients (Altieri, 1989).
To better assess the effects of introduced technology on the farming community,
proponents of agroecology subdivide the community into various sectors (de Janvry and
Helfand, 1990). Resource-poor farmers are considered members of subfamily farms.
Subfamily farms are farming households with resource bases insufficient to economically
support or to absorb the labor of the household. Involvement of one or more family
members in off-farm labor activities is required to supplement the income derived from
farming activities. Due to the reliance of this sector on crops and farming practices not
addressed by the green revolution, technology introductions have not increased the on-
farm contributions to the income of the subfamily farm. As a result of the marginaliza-
tion of the subfamily farm due to the introduction of input-based technologies, the family
becomes more dependent on off-farm sources of income.
In an attempt to assist the subfamily farming household to become more
economically competitive, Altieri (1989) recommended the formation of integrated,
community-based development programs aimed at conserving and efficiently using local
resources, minimizing dependency on external inputs, increasing the diversity of
agricultural practices in order to minimize risks, improving the nutritional intake of the
farm family, and ensuring that program benefits are equitably distributed among
community members. The low resource base of subfamily farms, however, may restrict
this sector from becoming economically competitive despite farmer-directed development
policies. To generate higher monetary incomes for this sector, de Janvry and Helfand
(1990) stressed that policies promoting land reform and economic opportunities in the
rural areas need to be pursued.
Sustainability. Induced Technology Change, and Bean Farming in Costa Rica
Recently, programs of agricultural development organizations have exhibited
increasing sensitivity to the concepts of sustainability. Many have incorporated the
development of sustainable agricultural practices into their mandates. As the transition
is made from researcher-directed to farmer-directed agricultural development, develop-
ment organizations are increasingly interested in comparing the impact of their previous
and current projects on issues of sustainability. Specifically, these organizations are in-
terested in assessing the trade-offs made between productivity, sustainability, and short-
term economic gains. For example, Torres (1991) asked, if productivity is given
priority over sustainability, what levels of stress can be resisted by the system before
sustainability is sacrificed?
The International Center for Tropical Agriculture (CIAT) has been involved,
since 1973, in research designed to increase the production of beans (Phaseolus vulgaris
L.) in Central and South America and the Caribbean. This work has included the
development of "genetically superior" cultivars of beans, examinations of cropping
systems, and studies on nitrogen fixation (CIAT, 1987; 1988a). In 1989, CIAT
requested a study in Costa Rica directed at assessing differences in the sustainability of
traditional, compared to introduced, bean production practices.
In Costa Rica, the common bean is an important component of the daily diet and
a primary ingredient in several traditional dishes. The annual per capital consumption of
beans is 10-13 kg, which provides approximately 10% of the dietary protein (van
Schoonhoven and Voysest, 1989). For resource-poor farmers who grow the majority of
Costa Rica's beans, bean production is both a source of income and of protein.
The traditional method of bean production in this country is the pre-Hispanic
practice of fijol tapado or covered beans (Patillo, 1965). Frijol tapado is a broadcast-
planted, land-use extensive practice formerly used throughout the monsoonal Pacific
Coast zone extending from Costa Rica to Ecuador. Slash-and-mulch farming practices
similar to frijol tapado are currently used in Costa Rica, Ecuador, Colombia, Peru, and
Brazil (H.D. Thurston, personal communication). This practice uses no commercial
inputs and requires low labor inputs.
To allow farmers to adapt to changes in relative resource scarcities, a multi-
institutional program was created in 1978 to research, evaluate, certify, and distribute
new varieties of bean seeds (Pachico and Borb6n, 1987a). The institutions involved in
this effort were the Ministry of Agriculture (MAG), the University of Costa Rica
(UCR), the National Production Council (CNP), the National Seed Office (ONS), and
CIAT. The coordinated efforts of these organizations resulted in the release of the
erect, determinant bush, or Type I black bean variety, Talamanca, in 1981 (Figure 1.1).
In 1982, a spreading, indeterminate bush, or Type II bean variety, Brunca, was released
for farmer use. These varieties were introduced to replace traditional seed varieties
characterized as Type III (indeterminate prostrate) and Type IV (indeterminate
In conjunction with adoption of these bean varieties, farmers began adopting the
frijol espeque planting method. Frijol espeque is characterized by planting beans in
straight rows using a dibble-stick or espeque. By planting their beans in straight rows
rather than by broadcasting, farmers are able to more easily control diseases and insects
using pesticides, and to manage soil fertility using fertilizers (Pachico and Borb6n,
1987a). The combination of the new planting method, introduced seed varieties, use of
commercial inputs, and increased labor inputs for weed control makes frijol espeque
more input-intensive than frijol tapado. The yield potential for frijol espeque is also
higher than that for frijol tapado. Based on farmer surveys, Pachico and Borb6n (1986)
reported average yields of 569 kg ha-' for frijol tapado planted with traditional varieties,
compared to yields of 941 kg ha"' for frijol espeque using introduced varieties.
The use of agrochemicals in frijol espeque allows farmers to use their land more
intensively. Despite the potentially higher yields of frijol espeque, frijol tapado can
provide almost three times the returns on capital investments compared to frijol espeque
(income/cost= 10.60 for tapado versus 3.76 for espeque) when the opportunity value for
land is low (CIAT, 1988a). A high, competitive, opportunity value for labor during the
period of bean growth also favors frijol tapado over frijol espeque. No labor inputs are
used for frijol tapado during the period between planting and harvesting. It is,
therefore, favored in areas where off-farm activities, particularly coffee harvests,
provide farmers with a secure source of income during the period of bean growth (-
Frijol tapado is planted on land that has been fallowed for 9 months to 5 years
(M. Amador, personal communication). This fallow period is used to regenerate soil
fertility and provide biological control of pests and diseases. In contrast, the adoption
of frijol espeque allows farmers to continuously crop a bean-maize rotation within a
Many resource-poor farmers have adopted the fundamentals of the frijol espeque
technology. Their lack of economic resources, however, often makes them unable to
apply the recommended rates of fertilizers and pesticides. Many farmers also plant frijol
espeque on highly erodible, steeply sloping lands. Although frijol espeque can produce
significantly higher yields than frijol tapado, a decrease in the sustainability of bean
production may occur concurrently with the transition from frijol tapado to frijol
The productivity of the frijol tapado system has also declined. Where land-use
pressure forces farmers to plant frijol tapado on land fallowed for less than 2 to 3 years,
this rotational agroforestry practice has been transformed into a crop-fallow or "agro-
shrubbery" system. The use of only minimal labor inputs and no chemical inputs in
frijol tapado results in a decrease in the productivity of the system as fallow length
Frijol espeque has been the focus of a major technology transfer and research
program. Recommended practices for this system have been described in research
reports (Corella, 1989; Hugo, 1989; Tapia et al., 1989) and in extension pamphlets
(Trece et al., 1989; CNP, 1990). Information regarding frijol tapado is much more
limited. On-farm assessments of frijol espeque or comparisons between frijol espeque
and frijol tapado have been limited to socioeconomic studies (Solano, 1980; Chapman et
al., 1983; Barrantes et al., 1986; Pachico and Borb6n, 1986). Some studies also contain
data that is confounded (Ospina, 1990) or incompletely described or justified (Shenk et
al., 1979; von Platen and Rodriguez, 1982). Due to the scarcity of agronomic research
information on frijol tapado or on farmer implementation of frijol espeque, a review of
the literature leaves as many questions unanswered as it answers.
Bean Farming Methods in Costa Rica
Bean production in Costa Rica is concentrated within the Central Mesa and in the
half of the country bordering on the Pacific Coast. Within this area, there are four
principal bean growing regions: the northern zone including Upala and Los Chiles, the
Nicoya Peninsula within the province of Guanacaste, the Central Mesa around San Jose,
and the southern zone from San Isidro de General to Buenos Aires (Figure 1.2). In the
northern zone, beans are grown on relatively level land by large-scale commercial
operators. In contrast, on the undulating to steeply sloping lands of the Nicoya
Peninsula, the Central Mesa, and the southern zone, bean production is primarily the
work of small holders.
In 1950, most beans grown in the El General Valley in the southern zone of
Costa Rica were grown using frijol tapado (Skutch, 1950). Between 1980 and 1987 the
percentage of land in bean production planted using frijol tapado decreased while the
percentage of land used for frijol espeque increased. In 1980, 80% of the land was
planted using frijol tapado, while frijol espeque was planted on only 18% of the land.
During the 1986-87 production season, the percentage of land planted to frijol tapado
had decreased to 49%, while frijol espeque was practiced on 38% of the land (CIAT,
1988a). One-third of the farmers interviewed by Borb6n (1984) in the San Isidro area
said they had converted from frijol tapado to frijol espeque during the ten-year period
from 1973 to 1983.
Frijol tapado is currently practiced within the regions of Upala, Guanacaste,
Acosta, San Isidro, and Cotq Brus. These regions, located along the Pacific Coast, all
have steeply sloping lands and a monsoonal climate.
Frijol tapado is a rotational agroforestry practice. Traditionally, and under
conditions of low land-use competition, a farm is divided into 3 or 4 lots. Each year,
frijol tapado is planted on one lot while the other lots remain in fallow. Woody species,
typical of the early stages of forest succession, dominate 3- to 4-year fallows used for
frijol tapado production. When land-use competition is high, frijol tapado is practiced
on the same lot each year. Pasture grasses, plantains, herbaceous plants, and ferns
dominate the short, 9 month fallows used for annual production of frijol tapado.
Frijol tapado planting method. To plant frijol tapado, a farmer initially cuts
narrow pathways through the fallow growth, following the contour of the land. The dis-
tance between pathways depends on the slope of the land, the height and thickness of the
fallow growth, and farmer preference. Typically, a farmer cuts pathways 6-8 meters
apart. From within these pathways, the farmer broadcasts bean seeds into the standing
fallow growth between the pathways. From the lower pathway, seeds are broadcast
upward into the lower half of the intervening standing fallow. Then, moving to the next
higher pathway, the farmer broadcasts seeds downward into the upper half of the
intervening area. Alternatively, the farmer cuts pathways 3-4 meters apart and
broadcasts seeds into the entire intervening area from the pathway below the area to be
planted (Figure 1.3a).
After the seeds are broadcast, the farmer cuts the remaining fallow growth with
a long machete (60-70 cm long). The cut residues cover the bean seeds. The height to
which the fallow is chopped depends on the growth habit of the beans. When erect, less
aggressive varieties are planted, weeds are cut close to the ground. When more
aggressive, climbing bean varieties are planted, tree trunks are left standing as trellises
(Araya and Gonzilez, 1987). Farmers prefer aggressive, indeterminate, Type III and
Type IV varieties of beans for frijol tapado production. These varieties are highly
competitive with weeds, since they can climb or grow over the regenerating fallow
The emergence of the germinated beans from beneath the mulch layer is
facilitated by coarsely chopping the cut fallow with a machete (Figure 1.3b). To further
offset the low emergence rate characteristic of this planting system and to make the
beans more competitive with weeds, sowing rates of 35-50 kg ha' are recommended.
This is one and one-half to two times the sowing rate recommended for frijol espeque
Soil fertility maintenance and pest control for frijol tapado. Following the
planting of a tapado bean field, the beans normally grow without any additional
management or labor inputs until harvest. Neither fertilizers nor pesticides are typically
applied to frijol tapado fields.
Natural fallows are used instead of commercial inputs to regenerate soil fertility
and decrease infestations by pests and diseases. In addition, nutrient recycling is
enhanced and weed regrowth may be reduced by maintaining crop residues on the soil.
This mulch cover also may protect the soil against nutrient losses due to erosion, caused
by the impact of intense tropical rainstorms. Galindo et al. (1983) reported lower inci-
dence of a splash-dispersed fungal disease, web blight [Rhizoctonia solanii, Kuhn,
anomer Thanatephorus sucumeris (Frank) Donk], in frijol tapado fields compared to
Under conditions of 1-2 year, grassy fallows, lack of herbicide application and
mid-season manual weeding may allow weeds to regrow during the growing season and
compete with the beans for nutrients, water, and sunlight. Weed regrowth also may
serve as a source of inoculant or as an alternative food source for diseases and pests of
Socioeconomic factors affecting frijol tapado. Frijol tapado is a low-input, low-
risk production system. It is used predominantly by farmers who have low access to
economic resources or who live in areas with low accessibility to markets. Frijol tapado
is also favored in areas where other on-farm or off-farm activities provide a high
opportunity value for labor during the period of bean growth (Ballestero, 1985). In the
coffee-growing area around San Vito, bean production activities occur simultaneously
with the highly profitable coffee harvest. By substituting the decreased weed growth and
disease incidence provided by fallows for labor inputs, farmers can obtain bean yields
per hectare close to those obtained using frijol espeque while being able to engage in the
The profitability of frijol tapado is dependent on a low opportunity value for land
use. Pachico and Borb6n (1986) reported that, of the farmers interviewed who owned
less than 50 hectares of land, less than one-half felt able to leave a portion of their land
in fallow. For farmers owning 10 hectares or less of land, only 16% were able to
maintain a fallow rotation. Farmers with small land holdings or with land accessible to
major transportation arteries use their land for the production of frijol espeque or other
high-value crops, such as coffee or vegetables. If they desire to plant frijol tapado, they
do so as tenants. Frijol tapado is planted on parcels of land that are more mountainous,
less accessible, and less valuable than lands used for more commercial crops.
Due to the availability of preferential loans for cattle production during the 1980s
(Annis, 1987) and the expansion of coffee production into San Isidro and Coto Brus
(Chapman et al., 1983), the availability of land to tenants for frijol tapado production
decreased. This resulted in the duration of fallow periods between croppings being de-
creased. The option of small-scale farmer landholders to plant frijol tapado has also
Frijol tapado research. Despite the social and economic importance of frijol
tapado in Costa Rica, few investigations have been conducted on this system. Reasons
for the lack of scientific interest include the belief that tapado bean production is not
responsive to inputs or improved management practices, and the reality that it is prac-
ticed in areas with low accessibility to extension agents (Araya and GonzAlez, 1987).
Additionally, the major economic advantages of frijol tapado are the minimal labor and
cash inputs required. Thus, the introduction of technologies to increase frijol tapado
yields through intensified management of the system may not be acceptable to farmers.
Researchers who have investigated frijol tapado suggest that nominal management
modifications could increase yields per hectare while retaining most of the advantages
of this system. Von Platen and Rodriguez (1982) obtained a 22% increase in frijol
tapado yields when 150 kg ha- of 10-30-10 NPK fertilizer and pesticides (500 g ha7'
Orthene 95% to control Diabrotica sp. and Aldrin 25% to control slugs) were applied by
broadcasting. Rosemeyer (1990) reported a significant increase in frijol tapado yields
when nitrogen and phosphorus fertilizers, at rates up to 35.1 kg ha' of N and 97.5 kg
ha- of P, were broadcast-applied. Neither Rosemeyer (1990) nor Araya and Gonzalez
(1987) obtained significant increases in yields per hectare due to rhizobium inoculation
of frijol tapado-planted traditional bean varieties.
Shenk et al. (1979) obtained an increase in yields from 77 kg ha' to 495 kg ha-',
while increasing labor costs by only 16%, due to dibble-stick planting rather than broad-
casting seeds. This indicates that bean emergence is more reliable when seeds are
dibble-stick planted rather than broadcast. It also may be indicative of the ease of
obtaining misleading results for frijol tapado. Araya and GonzAlez (1987) indicated that
the experience of the planter can significantly affect yields obtained from frijol tapado.
Frijol espeque planting method. Frijol espeque is characterized by planting beans
in straight rows using a dibble-stick (Figure 1.3c). Crop management activities are
facilitated by planting beans in straight rows. In contrast to frijol tapado, frijol espeque
employs commercial inputs and is less dependent on fallow periods to regenerate native
fertility. Recommended frijol espeque methods include planting certified seeds, two
applications of fertilizers, chemical or manual weed control during the first 30 days of
bean growth, and applications of agrochemicals to control insects and fungal diseases
(CNP, 1990). Unlike the majority of other Latin American countries, beans are grown
in Costa Rica principally as a sole crop rather than as an intercrop with maize. By
adopting frijol espeque, beans and maize can be grown as rotation crops on the same
plot of land during the two growing seasons per year.
Use of fire in the preparation of frijol espeque fields. When practiced on steeply
sloping lands or under conditions of low resource availability, frijol espeque is planted
without plowing. To facilitate production activities, minimize weed infestation, and
control against pest and disease infestations, the land is normally cleared before planting.
According to a farming systems study conducted by MAG personnel in an area near
Pejibaye de P6rez Zeled6n during the 1985 growing season, 74% of the farmers used
fire to clear their land in preparation for the bean planting season (Barrantes et al.,
1986). Alternatively, fields are cleared by grazing cattle on the fields during the dry
season or by piling cut fallow residues on the edges of the field.
A majority of the farmers interviewed by Barrantes et al. (1986) stated that they
burned their fields in an attempt to reduce slug (Vaginulus plebeius) damage. Slugs
were accidentally introduced into Central America in the early 1960s. Within ten years,
slugs have become a major pest problem throughout the wet, Pacific coast, bean-growing
region (Andrews, 1987). In central Honduras, one-half of the land devoted to bean
growing in 1974 was abandoned by 1979 due to uncontrollable slug problems (Rodrf-
guez, 1980). Slugs are nocturnal feeders which feed principally on young, succulent
plants. During the day, these soft-bodied mollusks shelter beneath residues to avoid
desiccation. Slugs can be controlled by using molluscacides (Andrews, 1985) or with
manual capture methods (Sobrado and Andrews, 1985). The method recommended by
MAG, however, as being most effective for slug control, is the removal of all residues
from the surface of the soil (MAG, 1988).
Frijol espeque planting on steeply sloping lands. Most small-scale farmers plant
beans on land having slopes of 50% to 70%. This is due to tradition, preference, and
displacement. Farmers traditionally associate sloping lands with bean production, since
frijol tapado is usually planted on sloping lands. Sloping lands are preferred for bean
production because, during periods of heavy rainfall, rainwater does not collect on the
fields. This decreases the susceptibility of beans to web blight and other fungal diseases.
Finally, for moderate to large-scale landowners, profits from cattle production can be
obtained at a lower risk, with fewer labor inputs, and with greater access to credit than
can profits from beans. The higher opportunity value for land in cattle production
compared to beans has resulted in the placement of most level lands into cattle produc-
The practice of clean cultivation on steeply sloping lands can encourage high
rates of erosion. In an attempt to develop less erosive, intermediary, agricultural prac-
tices, MAG specialists are currently conducting studies on the use of mucuna (Mucuna
utilis) as a cover crop and the planting of barrier rows of lemon grass (Cymbopogon
citratus) along the contour lines. They are also encouraging farmers to leave residues
on the soil surface during the growing season. This latter practice is called labranza
cero (G. Ramiro, personal communication).
The Transition from Frijol Tapado to Frijol Espeque
Adoption of frijol espeque. The objective of the coordinated extension program
has been to replace frijol tapado with frijol espeque. Since frijol tapado uses low labor
inputs and has a low response to agrochemical inputs, most researchers and extension
agents in Costa Rica treat this production method as a "primitive" system which should
be eradicated (Rosemeyer, 1991).
In the area of Pejibaye de Pdrez Zeled6n, the transition from frijol tapado to
frijol espeque was favored by an extension program that provided farmers with credit for
the purchase of introduced seeds and agrochemical inputs (Chapman et al., 1983). This
program also set the purchase price for beans by establishing buying stations during the
harvest season. Access to inputs and markets was also increased by the construction, in
the early 1980s, of an asphalt road between Pejibaye and the Inter-American Highway.
Although extension programs have been partially successful in increasing the
adoption of frijol espeque, frijol tapado is still practiced by almost 50% of the bean
farmers in Costa Rica. In the regions of Acosta and San Vito, frijol tapado accounts for
over 90% of bean production (M. Amador, personal communication; Rosemeyer, 1991).
Farmers may also use both practices, but on different parcels of land. Due to a
continuous cover of vegetation or mulch over the soil surface, frijol tapado provides
better control of web blight and soil losses due to erosion than does frijol espeque
planted on bare fields. Therefore, farmers may choose to plant level or gently undu-
lating lands to a frijol espeque / maize rotation, and to use their steeper lands for frijol
tapado production (Pachico and Borb6n, 1986; Monge et al., 1987).
The effect of frijol espeque adoption on the sustainability of bean production.
When land is relatively abundant, frijol tapado is the more environmentally sustainable
system. This is due to the regeneration of soil fertility and the stabilization of pest
populations during fallow periods. During the growing season, the maintenance of a
plant residue mulch over the soil surface decreases soil erosion and pest infestations. It
is also an economically sustainable system for resource-poor farmers. Due to its low
labor and agrochemical input requirements, farmers can use frijol tapado to diversify
their labor activities and minimize their economic risks.
As population pressure places increasing demands on land use, however, the
changing resource situation reduces the relative advantages of frijol tapado. Economic
and population pressures force farmers to use their land more intensively. This means
using shorter fallow cycles and/or converting forest lands into bean production.
Changing resource pressure also may persuade farmers to convert from frijol tapado to
When practiced on moderately level lands with recommended rates of inputs,
frijol espeque may produce stable production levels. When practiced on the steeply
sloping lands traditionally used for frijol tapado, frijol espeque may be unsustainable due
to loss of soil organic matter by erosion. When practiced by resource-poor farmers who
are unable to afford agrochemical inputs, soil nutrients are depleted since nutrients re-
moved through land clearing and harvest are not replaced through the use of fertilizers.
The literature indicates many significant differences between frijol tapado and
frijol espeque production methods (Table 1.1). Many questions, however, are left
unanswered regarding the adoption of frijol espeque and the relative sustainability of the
two systems. Do agronomic or socioeconomic factors have primary influence over
farmers' decisions to change bean production technologies? Why have farmers in some
areas of Costa Rica adopted frijol espeque while in other areas farmers continue to use
frijol tapado? Which planting method is most profitable and least risky for tenant
farmers? For farmers who use both planting methods, does adoption of introduced
varieties for frijol espeque result in an adoption of these varieties for frijol tapado? Can
frijol espeque be modified to produce a system which is both productive and sustainable?
The Assessment of Sustainability
Consequences of Degradation
Sustainability is often assessed by measuring its converse, non-sustainability or
degradation. Consequences of agroecological degradation include: soil erosion, loss of
soil organic matter, exhaustion of soil nutrients, desertification, deforestation, loss of ge-
netic biodiversity, build-up of pest and disease infestations, environmental pollution,
decreased yields, disruption of water supplies, reduced future availability of agricultural
inputs, and global warming (Liverman et al., 1988; Jodha, 1990; Sims, 1990; Harring-
ton, 1991). Aspects of socioeconomic degradation include overpopulation, a decreased
standard of living, malnutrition, loss of control of the resources of production, and
social unrest (Redclift, 1987; MacNeill, 1989).
Soil degradation is often cited as an important indicator of decreasing environ-
mental sustainability (Brown, 1987; IUCN, 1980). Soil resources can be degraded
chemically, physically, or biologically. Chemical soil degradation is indicated by
decreases in soil organic matter and cation exchange capacity, depletion of plant-
available nutrients, soil acidification, and aluminum toxicity (Logan, 1990). Examples
of physical soil degradation include soil erosion, increased soil bulk density, and
decreased water flow and infiltration (Hillel, 1982). Microbiological processes are
controlled by the availability of soil organic matter. Removal of organic substrates
results in decreased diversity of the microbial community and disruption of nutrient cy-
cles (Sims, 1990). Decreased soil organic matter and nutrients may also decrease the
activity of microorganisms responsible for soil aggregation (Elliott, 1986).
Intensification of agricultural land use can accelerate the soil degradation
processes. The replacement of forests by annual crops results in a considerable loss of
nutrients (Nair, 1984). In a forest ecosystem, nutrients are immobilized in plant
biomass. Removal of the biomass by burning enhances the availability of cations but
also causes nutrient loss due to nitrogen volatilization and post-bur nutrient leaching
(Ewel et al., 1981). Deforestation also can result in decreases in soil organic matter and
increases in soil bulk density (Adejuwon and Ekanade, 1987). Reductions in fallow time
allowed between successive croppings result in decreases in soil organic matter levels
and in associated nutrient availability (Greenland and Nye, 1959).
The practice of bare-land farming practices on steeply sloping land promotes soil
losses by erosion (Lal, 1984a). On soils having well-developed subsoils and no restric-
tions to rooting, crop productivity can be maintained if nutrients lost due to erosion are
replaced by fertilizers. On shallow or poorly developed soils, erosion can cause
irreversible soil degradation (Lal, 1985). Erosion-associated soil degradation causes
depletion of soil organic matter, disruption of microbiological processes, increased bulk
density, and depletion of plant-available nutrients.
Depletion of plant-available nutrients also results when nutrients are removed
from the land due to land clearing and harvest. These nutrients may be restored through
fallowing or fertilization. Adopting continuous cropping practices without application of
fertilizers results in a depletion of plant nutrients.
Parameters for Assessing Sustainability
The parameters used to describe degradation or loss of sustainability are
dynamic. They exhibit spatial dimensions and are interactive with other factors in the
environment. Furthermore, the definitions for economic, agricultural, and ecosystem
sustainability are often contradictory (Redclift, 1987). The sustainability of ecosystems
implies a balance between inputs and outputs while, from a development perspective,
economic and agricultural sustainability implies growth. In addition, confusion exists re-
garding what time frame and discount rates to use when comparing traditional and input-
intensive systems. The interactions, contradictions, and confusions associated with the
concept of sustainability make its assessment difficult. Despite these difficulties, at-
tempts have been made to develop criteria to assess sustainability, or its converse,
Criteria used for the assessment of sustainability must be both comprehensive and
discrete. Senanayake (1991) views sustainability as maintaining a dynamic equilibrium
between inflexible boundary conditions. Thus, determination of sustainability should
consider boundary conditions and internal dynamics. These factors are measured in ref-
erence to ground states of absolute sustainability, or to thresholds where sustainable
systems become irreversibly unsustainable. For example, Liverman et al. (1988)
emphasized that rates of soil erosion are only indicative of sustainability if they are
expressed in terms of a rate-to-stock ratio, or the percentage of topsoil lost during a
given period of time.
Conway (1991) suggests assessing agroecosystem sustainability according to five
characteristic responses of a system to shock. These responses include inertia, elasticity,
amplitude, hysteresis, and malleability. Inertia is the resistance of the system to change.
Elasticity is a measure of the time required for productivity to recover following a dis-
turbance. Amplitude is the maximum amount of shock or disturbance from which a
system can recover. Hysteresis measures the rate of recovery. Malleability estimates
differences in system steady states before and after a disturbance. These characteristics
of sustainability may be used to assess both the absolute and relative sustainability of a
system. They are also useful in differentiating between stability and sustainability.
For example, when land is brought into cultivation, the organic matter content of
the soil decreases rapidly, then stabilizes. The organic matter content of the soil, thus,
eventually could be characterized as stable, while the sustainability of the system could
be characterized as highly malleable. Inability of the cultivated soil to recover an
organic matter content characteristic of an undisturbed system indicates that the system
is inelastic and has a low level of inertia and hysteresis.
Several approaches to measuring sustainability have been proposed. Harrington
(1991) recommends using measurements that evaluate only the direction of change in
system sustainability and not the magnitude of such change. He suggests that assess-
ments of yield trends in relation to inputs applied may be acceptable if assessments also
take into account technology changes and resource degradation. Senanayake (1991)
suggests evaluating sustainability using an index that accounts for changes in external
inputs, energy ratios, power use, efficiency of solar flux use, erosion, and microbial
biomass. Hildebrand and Ashraf (1989) proposed using a linear program to assess
sustainability. Their linear matrix included an objective equation related to profit
maximization. This was combined with equations accounting for the proportion of the
land in fallow, and constraints related to the use of external chemical inputs.
Economic Measurements of Sustainability
In farming systems, agroecosystem sustainability and economic sustainability are
closely interrelated. Agronomic and economic risk aversion is sought by maintaining a
diversity of agricultural practices (Altieri, 1987). A transition from traditional to
commercial input-based farming systems may provide the farmer with higher per hectare
profits. The use of a commercial input-based system, however, requires farmers to have
access to cash or credit in order to purchase inputs and pay for laborers. When the poor
lose access to resources sufficient to fulfill their basic needs, often they are forced to
sacrifice long-term sustainability of their farming practices to obtain short-term survival.
The need to obtain cash to purchase inputs or to pay laborers may also result in
resource-poor farmers becoming increasingly dependent on usurious money lenders.
The potential ethical changes that occur concurrently with the transition from
traditional to commercial input-based agricultural systems require that economic
comparisons be based on more than cash outlays and cash returns. Valuation must be
based on the sacrifices the farm family makes to obtain cash for their farming
operations. Value must also be placed on the increased standard of living a more
productive farming method may provide.
The current economic system places a low value on natural resources (Redclift,
1987). This, combined with the ability of agrochemical inputs to dominate the environ-
ment, allows farmers to profitably produce a low-value crop while simultaneously
degrading soil resources. Since environmental resources are considered common-
property goods, their economic valuation is unclear (Barbier, 1987; Manning, 1987).
When low economic valuation is placed on natural resources, the extermination of
species or the irreversible degradation of soils may appear economically justified
(Tisdell, 1988). Consequently, the use of cost/benefit analysis to assess economic
sustainability necessitates a careful selection of discount rates for resource use and a sen-
sitive valuation of intangibles such as environmental or social benefits (Manning, 1987).
Assessing the Sustainability of Bean Growing Methods on Steeply Sloping Lands in
For small-scale farmers in Costa Rica, land use, input use, and bean-growing
practices are in a period of transition. Different farmers use different lengths of fallow
periods in their practice of frijol tapado. The number of frijol espeque and frijol maize
rotations a farmer plants before placing the land in fallow once more varies. Frijol
espeque land preparation practices also vary from bare-land to labranza cero. Thus, the
sustainability of bean growing practices cannot be assessed simply by comparing tradi-
tional frijol tapado with introduced frijol espeque.
Assessments of sustainability must also account for differences in productivity
and degradation process across environments. Land, labor, and input use employed
within each system and among different environments must also be considered. In
making economic comparisons among bean growing practices, the opportunity cost of
land fallowed for frijol tapado must be calculated. In addition, the opportunity value for
off-farm labor during the frijol tapado growing season must be estimated.
Although no comparisons have not yet been reported, it is supposed that soil loss
due to erosion is lower with frijol tapado than with frijol espeque (von Platen and
Rodriguez, 1982). Frijol espeque may be practiced without residues left on the soil
surface or, as in labranza cero, with residues on the soil surfaces. Thus, comparisons
of erosion losses and associated degradation processes should include labranza cero as
well as frijol tapado and bare-field planted frijol espeque. Assessments of erosion-
associated decreases in sustainability should include not only measurements of soil losses
due to erosion, but also assessments of soil organic matter, microbial activity, and soil
Pest and disease infestations are culturally controlled in frijol tapado by using
mulches and fallows. Retention of crop residues over the soil surface decreases the
incidence of web blight (Galindo et al., 1983). Weed infestations are suppressed fol-
lowing extended fallow (Ramakrishnan, 1988). In frijol espeque, weed and disease
infestations are controlled by using pesticides. Sustainability of pest control requires
both an assessment of infestation levels and a comparison of the direct and indirect costs
associated with each system.
The conversion from long to short fallow frijol tapado or from frijol tapado to
frijol espeque involves trade-offs between productivity and other agroecosystem prop-
erties: sustainability, stability, and equity (Table 1.2). Frijol tapado requires access to
sufficient land to practice a fallow rotation. Alternatively, tenancy rights to fallowed
land must be obtained and tenancy costs paid. Maintenance of system productivity in
frijol espeque depends on the ability of the farmer to replace nutrients removed and to
control pests and diseases. Input use, in turn, is affected by the access of farmers to
resources and by the balance between input costs and prices paid for beans. Increasing
system degradation may require farmers to apply increasing amounts of inputs and to
receive decreasing returns on their cash and labor investments in order to obtain stable
The research reported here was designed to assess the effect of technology
change on the agroecosystem sustainability and the economic sustainability on bean
production in Costa Rica. Since agroecosystem and economic sustainability are
intrinsically interrelated (Figure 1.4), farmer surveys were conducted in conjunction
with on-farm experiments. The transition from frijol tapado to frijol espeque was
examined as a prototype of the transitions from traditional to higher-input technologies
that are occurring throughout the developing world.
Specifically, the research was designed to answer the following questions:
1. How are bean yields, profits, and soil quality affected by the transition
from frijol tapado to frijol espeque?
2. Can methods intermediate between frijol tapado and frijol espeque be used
to produce a system which is both sustainable and productive?
3. Under what conditions should a transition from a traditional system not be
expected nor promoted.
4. How can system sustainability be assessed within on-farm trials?
The remainder of this dissertation is a detailed examination of the factors
affecting the sustainability of bean production on steep lands in the area of one small
town, Pejibaye de Perez Zeled6n, located in the southern zone of Costa Rica.
In Chapter 2, the study location is described in detail. Rationale for using this
location as a microcosm for technology-change processes throughout Costa Rica is
presented. Questions left unanswered in the literature regarding frijol tapado and the
transition to frijol espeque are investigated. Socioeconomic factors affecting technol-
ogy adoption and the sustainability of both frijol tapado and frijol espeque are also
In Chapter 3, the effects on sustainability of different frijol espeque land use
and land preparation practices are described. System sustainability was assessed
based on studies designed to investigate differences in farmer practices of frijol espe-
que. Interactions between frijol espeque practices, erosion losses, bean growth and
nutrient uptake, crop yields, and profits were investigated.
In Chapters 4 and 5, two on-farm experiments are described and discussed.
The first experiment was designed to compare frijol tapado with four frijol espeque
practices. The experiment was conducted on nine farms representing different lengths
of fallowing and intensities of land use. Interactions between factors affecting yields,
soil quality, and profitability were examined. The second experiment was designed to
compare three frijol espeque land preparation methods and phosphorus fertilization
within two environments: fallowed and continuously cultivated land. System sustaina-
bility in both experiments was assessed based on erosion losses, bean growth, nutrient
uptake, nitrogen mineralization, soil aggregate stability, crop yields, weed and web
blight infestations, and profitability.
Finally, Chapter 6 summarizes the work. The effects of technology change on
the sustainability of the environmentally and economically fragile agroecosystem of
steep lands are discussed. In addition, methods for evaluating the sustainability of
farmers' practices and introduced technologies are recommended.
Figure 1.la: Type I
Bean Growth Habit.
Figure 1.lb: Type II
Bean Growth Habit.
Figure 1.lb: Type IV
Bean Growth Habit.
Figure 1.lc: Type III
Bean Growth Habit.
Figure 1.1: Schematic representations of the four bean growth habits.
Source: Debauck and Hidalgo (1985)
Each dot represents approximately 100 hectares of beans.
Bean growing areas in Costa Rica.
Source: Schlotz (1983)
Figure 1.3a: Broadcasting frijol tapado
~,c~f, E~ 4..tpl
-a, JL .v4r
Chopping residues after
Figure 1.3c: Planting frijol
Figure 1.3: Frijol tapado and frijol espeque planting methods.
yLD OF LAND USE
Figure 1.4: Agroecological and socioeconomic interactions affecting bean
production on steep lands in Costa Rica.
Table 1.1: Comparisons of frijol tapado and frijol espeque.
Planting Season Veranero Inverniz
Land Use Intensity Rotation of one cropping Continuous cropping,
season to 1-5 years of possibly with fallow rotation
fallow every 2-5 years
Land Slope Steeply sloping Moderately sloping
Land Preparation No land preparation prior Residue chopped with
to planting machete then burned or piled
Seed Varieties Traditional Introduced
Bean Growth Habit Indeterminate Determinate
Planting Rate 35-50 kg/ha 20-30 kg/ha
Planting Method Broadcast Dibble stick planted
Input Use None Herbicides, fertilizers, and
Labor Use Only during planting and Land preparation, planting,
harvest input application, weeding,
Average Yields 569 kg/ha 941 kg/ha
Sources: Araya and GonzAles (1987), Pachico and Borb6n (1986) and von Platen and
Table 1.2: Agroecosystem factors of social value for frijol tapado and frijol espeque.
SOCIAL VALUE FRIJOL TAPADO FRUOL ESPEQUE
PRODUCTIVITY low yields high yields
low land-use intensity high land-use intensity
STABILITY diversity of income sources high labor inputs limits off-
fallows promote plant favors build-up of pest
diversity and stabilization populations
of pest and predator
populations increasing input use may be
needed to obtain stable
SUSTAINABILITY mulch layer protects soil clear-field planting
against erosion and encourages soil erosion, soil
compaction due to raindrop compaction, and soil drying
soil moisture conserved by
nutrient recycling enhanced loss of organic matter and
by mulches and fallows nutrients during land
clearing and burning
EQUITABILITY high returns on investment low return on investment
requires either ownership intensified land-use can
of sufficient land to fallow increase profits
or rental of fallowed land
does not require cash cash outlays for inputs may
outlays make farmers dependent on
usurious money lenders
Based on Conway (1991).
Sources: Pachico and Borb6n (1986), Araya and Gonz,1ez (1987), Ramakrishnan
PEJIBAYE DE PEREZ ZELEDON:
A MICROCOSM FOR THE ASSESSMENT OF THE EFFECTS OF
TECHNOLOGY CHANGE ON SMALL-SCALE FARMER PRACTICES
IN COSTA RICA
An Introduction to Pejibaye de Perez Zeled6n
Reasons for the Choice of Pejibaye as a Study Location
Pejibaye de Pdrez Zeled6n is located in the southern bean-growing region of
Costa Rica. During the 1940s, farmers abandoned the eroded and degraded lands of the
Central Mesa for this fertile land full of promise. Fifty years later, bean farming is no
longer considered profitable. Sons and grandsons of the original settlers are abandoning
farming and returning to the Central Mesa to seek employment in the urban sector.
In Costa Rica, bean production occurs within the Central Mesa and throughout
the half of the country bordering the Pacific Coast. From this vast area, Pejibaye de
P6rez Zeled6n was chosen for the focus of these studies because it is located in the
center of an area of prime importance to national bean production. More than one-third
of Costa Rica's beans are grown in the southern zone (Borb6n, 1984). Between 55%
and 65% of the beans produced during the first growing season are grown in the
subdistrict of P6rez Zeled6n (CNP, 1991). The northern zone of the country was reject-
ed as a study site, although it also provides important contributions to the national bean
production. The relatively level land around Los Chiles and Guanacaste, in the northern
zone, allows bean production to be highly mechanized and input-dependent. Frijol
tapado and frijol espeque, the bean-growing technologies that serve as the focus of this
study, are typically used by resource-poor farmers on steeply sloping lands. Bean
growing in Pejibaye is primarily the work of small-scale producers farming marginal
Pejibaye is an area in transition. From the onset of European immigration (in the
1940s) to the present, the population of the area has increased rapidly, due both to natu-
ral population increases and immigration. Construction, in the early 1980s, of an
asphalt road between Pejibaye and the Inter-American Highway increased economic
articulation between the center of Pejibaye and the urban sector of Costa Rica. Farmers
living in the interior areas surrounding Pejibaye are only minimally articulated with the
urban economy. This is because only rutted, seasonally passable dirt roads connect
Pejibaye with the interior farming areas. The difference in economic articulation
between areas close to Pejibaye and interior areas, no more than four kilometers from
the center of town, resulted in distinct differences in technology-adoption patterns
between farmers the two areas.
Since Pejibaye de P6rez Zeled6n represents a transitional area but also is
relatively accessible from San Jose, it has become the focus of various research studies.
The area has served as the focus of studies conducted by students from the University
of Costa Rica and the National University, as well as by researchers from the Instituto
Interamericano de Cooperati6n para la Agricultura (IICA), CIAT, Centro Agronomico
Tropical de Investigaci6n y Ensefianza (CATIE), MAG, and the Ministerio de
Planificaci6n (MIDEPLAN). Researchers from U.S. universities also have used
Pejibaye as a focus of their investigations.
The breadth of research previously conducted in this area provided background
for the current study. It also confirmed that Pejibaye de Perez Zeled6n serves as a
microcosm for demonstrating the effects of transitions in agricultural technologies on
resource-poor farmers. In this study, socioeconomic surveys were conducted in
conjunction with agronomic studies in order to understand constraints to adoption of
introduced agricultural technologies.
Description of the Study Area
Pejibaye de Perez Zeled6n is located in the southern Pacific zone of Costa Rica
between the El General River Valley and the Pacific Coast Range (Figure 2.1). The
subdistrict boundaries extend from San Marcos de Pejibaye to El Aguila de Pejibaye (9*
10' N Lat., 830 35' W Long. to 90 05' N Lat., 830 40' W Long.). Politically, Pejibaye
is located within the district of P6rez Zeled6n and the province of San Jose. According
to the 1990 census, it had a population of 10,112 (Minist6rio de Economia Industria y
Comercio, 1991). The elevation of the study area ranges from 400 meters in the
municipal center of Pejibaye to 1000 meters at the summit of the coast range. The area
experiences a hyperisothermic temperature regime. A mean annual temperature of 22*
C has been recorded at 950 meters (Barrantes et al., 1986). The soils in the area are of
colluvial origin. On steep lands, the association of Tropepts, Humults, and Ustalts
predominates (Corella, 1989).
The Pejibaye area receives approximately 2000 mm of rain annually, distributed
in a bimodal monsoonal pattern. The first, less-intense rainy season begins in April and
continues through July. An unpredictable, short, dry season occurs between late July
and early September. A second, more intense, rainy season begins in October and
continues through December. The intensity and duration of the rainfalls are moderated
by the area's location on the leeward side of the coast range. Between late December
and mid-April, this region endures an extended and distinct dry period. The 1990
monthly rainfall totals for two locations in Pejibaye are presented in Figure 2.2.
Two planting seasons per year are observed. Inverniz, the wet harvest season,
occurs during the first rainy season. Veranero, the dry harvest season, occurs during
the second rainy season.
Maize and beans are the primary grain crops grown in the Pejibaye area.
Relatively mild rainfalls and the existence of a mid-year dry season allows farmers to
grow frijol espeque during the inverniz season. Mild rainfalls deter the spread of web
blight, a fungal disease that is yield-limiting in this area. The mid-year dry season
allows farmers, in most years, to harvest and dry their beans following the inverniz
cropping season. Dibble-stick planted maize is grown during the veranero season as a
rotation with frijol espeque. Despite high opportunity values for land, many farmers
continue to plant frijol tapado during the veranero season.
Changes in Land Use and Bean Growing Practices
Causes and Consequences of Land Use Intensification
The first settlers arrived in Pejibaye in 1942. Rapid social and economic
development of the area began around 1952, resulting in Pejibaye being designated as a
legal district in 1966 (Sewastynowicz, 1986). Further commercialization of the area
became possible in the early 1980s with the completion of an asphalt road connecting
Pejibaye with the Inter-American Highway and San Isidro. Before construction of this
road, rutted road conditions and frequent bridge washouts often prevented travel during
the rainy season.
In the less than 50 years since the arrival of the first settlers, most of the primary
forests have been cleared for pasture use and crop production (Figure 2.3). The Land
Settlement Laws stimulated deforestation. These laws gave settlers title to land that
they had "improved" through land clearing and cultivation (Thrupp, 1990). Natural
population increases and in-migration also placed pressure on land use. Early settlers
purchased large tracts of prime agricultural land at low prices. Consequently, settlers
who arrived later encountered high prices and decreased availability of land and were
able to acquire only smaller tracts of more marginal land (Sewastynowicz, 1986). Even
more importantly, cattle production displaced small-scale crop producers. Throughout
the 1970s and 1980s the World Bank sponsored a farm loan program for cattle produc-
tion (Annis, 1987). As a result, the amount of land in pasture increased while the
amount of land used for crop production decreased (Censo Agropecurio, 1975; 1987).
The percentage of farms of greater than 50 hectares also increased during this period
while the percentage of farms with fewer than 5 hectares decreased (Chapman et al.,
In 1984, a farming systems study was conducted by MAG in Guagural de Buenos
Aires, located in the mountainous area adjacent to Pejibaye (Barrantes et al., 1986).
Based on interview results, 24% of the farmers in this area owned between 0.5 and 1.0
hectares of land, 39% owned between 1.1 and 5.0 hectares, and only 6% owned more
than 5 hectares of land. The remaining 30% of the farmers rented or sharecropped the
land they cultivated. Tenancy agreements described were both exploitive and temporary.
Tenants rented land for only one or two planting seasons. Sharecropping charges are
either one-third or one-half of the obtained yield. This indicates that the farmers who
farm sloping lands are those who are least able economically to implement capital-
demanding conservation practices.
Government Programs to Increase Bean Production
The government of Costa Rica has implemented numerous programs designed to
promote the production of basic grains and assist small-scale farmers. The most
significant programs have been the CNP, formed in 1956, and the National Basic Grains
Program, initiated in 1974. Acting as a mechanism for price stabilization, the CNP sets
the price paid to the farmer for the production of basic grains. The National Basic
Grains Program, funded in part by loans from the U.S. Agency for International
Development (USAID), was designed to develop concrete actions in order to assure a
substantial increase in the production of basic grains. To this end, an integrated
program of farm loans and technology transfer was established (Chapman et al., 1983).
Despite rhetorical emphasis on assisting small-scale producers, these programs,
in reality, served primarily to benefit medium to large-scale rice producers (Annis,
1987). Import-substitution policies were implemented. These policies provided subsi-
dies to producers of commodity exports; specifically, rice and beef. They were not
directed to, nor beneficial for, maize and bean producers.
In the area of Pejibaye, the effects of governmental programs aimed at increasing
bean production have been mixed. During harvest, the CNP establishes a bean and
maize-buying station in El Aguila de Pejibaye. This buying station regulates prices paid
for beans and maize.
Nonetheless, many farmers still prefer to sell their products to private grain
buyers. This is because the price actually paid by the CNP is usually lower than the
announced price. At the time of purchase, standardized deductions are made for
humidity and low quality. Private buyers do not deduct for these factors. Also, private
buyers usually provide the grower with sacks and they purchase the product at the farm
gate. This eliminates transportation costs between the farm and the CNP buying center,
which the farmer would otherwise have to pay (Chapman et al., 1983).
An inter-institutional technology change and bean production improvement
program for the subdistrict of Perez Zeled6n was initiated in 1982. The primary
emphasis of this program was to encourage farmers to adopt frijol espeque production
methods. Using the "training and visitation" extension method, MAG extension workers
organized farmer groups and presented training seminars. To assist farmers in adopting
this input-based production method, farm loans were made available through the BNCR
and seed varieties were introduced through the CNP (Chapman et al., 1983).
Adoption of Introduced Bean Varieties
As a consequence of this inter-institutional effort, farmers adopted the introduced
bean varieties, Talamanca and Brunca. By 1984, 80.8% of the farmers interviewed by
Pachico and Borb6n (1986) planted introduced bean varieties during the inverniz season,
while 73.3% planted these varieties, using frijol tapado, during the veranero season.
The determinant Type I variety, Talamanca, was preferred for frijol espeque, while the
indeterminate, Type II variety, Brunca, was preferred for frijol tapado production.
Adoption of the introduced bean varieties occurred simultaneously with adoption
of frijol espeque. The use of agrochemicals also increased substantially between 1980
and 1985. Of the farmers interviewed in Pejibaye, only 12% used fertilizers and 30%
used herbicides for bean production in 1982 (Chapman et al., 1983). By 1985, 73% of
the farmers from the adjoining region of San Isidro used fertilizers and 65% used herbi-
cides (Borb6n, 1984). Pesticides were used by 81% of these farmers.
Frijol espeque yields increased 35% due to adoption of introduced varieties while
frijol tapado yields only increased 25% (Pachico and Borb6n, 1987a). Average yields
reported by farmers planting Talamanca were 964 kg ha' for frijol espeque grown
during the inverniz season and 581 kg ha' for frijol tapado grown during the veranero
season (Ballestero, 1985). Farmers cited high yields, disease resistance, and erect
growth habit as reasons for adopting introduced bean varieties (Borb6n, 1984).
Adoption of introduced varieties for frijol tapado remains less than for frijol
espeque. Only 50% of the farmers living in a predominantly frijol tapado growing
region adopted introduced varieties, compared to 96% of the farmers living in a
predominantly frijol espeque growing region (Ospina, 1990). Non-adoption by frijol
tapado farmers may be due to farmer preference for traditional varieties in this planting
system. Farmers consider traditional varieties to be more resistant to adverse condi-
tions, and better adapted to local environmental conditions, than the introduced varieties
(Ballestero, 1985). Lack of adoption of introduced varieties in frijol tapado production
may also be due to a lack of interaction between frijol tapado farmers and extension
agents. Sterling (1981) noted that 26% of the farmers growing frijol espeque had
contact with extension agents while none of the frijol tapado farmers reported contact
with extension workers.
Bean Production Problems
Input-based cropping methods are designed to allow farmers to use their land
more intensively. During the 1950s, farmers in the vicinity of Zapote de Pejibaye
planted a crop-fallow rotation. Typically, this rotation consisted of two years of
cultivation followed by 6 years of fallow. By 1979, due to decreased farm size, over
50% of the farmers no longer felt able to include a fallow season in their cropping cycle
(Solano, 1980). Increased land use intensity also forced farmers to use more capital-
intensive methods. It also resulted in decreased sustainability, due to diminished
agroecosystem diversity and increased soil losses due to erosion.
Pest and disease attacks have a critical effect on bean production. Severe attacks
by slugs and corn rootworm (Diabrotica spp.) were noted by local farmers as the prima-
ry cause for losses in bean yields (Borb6n, 1984). Pathogens, especially web blight,
also reduce bean yields in this area. Other pathogen problems reported by farmers
include angular leaf spot (Isariopsis griseola), rust (Uromyces phaseoli), and anthrac-
nosis (Colletrichum lindemuthianum) (Ballestero, 1985). Weed infestations are also a
common problem locally (Barrantes et al., 1986).
Other production problems reported by farmers included lack of capital, inability
to obtain necessary hired labor, and either heavy rains or severe droughts (Barrantes et
Previous socioeconomic studies conducted in this area have focused on changes
in farming practices since the time of settlement (Solano, 1980), economic comparisons
of frijol tapado and frijol espeque production (Borb6n, 1984; Pachico and Borb6n, 1987a
and 1987b), economic impacts of adopting introduced bean varieties and frijol espeque
planting methods (Chapman et al., 1983; Ballestero, 1985; Ospina, 1990), and agro-
nomic problems experienced by frijol espeque producers (Barrantes et al., 1986).
The present study was designed to identify factors affecting the sustainability of
both frijol tapado and frijol espeque production on steep lands. The sustainability of
these systems is affected by a combination of social, economic, agronomic, and
ecological factors. This chapter is used to investigate the socioeconomic factors
affecting the sustainability of bean production practices. Interviews were conducted to
identify factors affecting the short-term profitability and the long-term sustainability of
bean farming. Additionally, factors affecting farmers' decisions to change technologies
were investigated. Based on results from the socioeconomic studies, experiments were
designed to test the agronomic sustainability of bean production. Technologies that take
advantage of the opportunities, while recognizing the constraints, within the agroeco-
system were recommended.
Three sets of formal interviews were conducted; initial base-line interviews,
economic surveys with frijol espeque growers, and economic surveys with frijol tapado
growers. Informal surveys and discussions were also conducted with key informants
including extension agents, agrochemical input dealers, and large landowners. A
summary of the surveys conducted is shown in Table 2.1.
Initial surveys. Initial interviews were conducted during March and April, 1990.
A random sample of 51 steep-land bean farmers from 10 barrios of Pejibaye de P6rez
Zeled6n were interviewed. All farmers lived within 8 kilometers of Pejibaye (30
minutes by motorcycle during the dry season). Questions were designed to obtain
information regarding land ownership, bean-growing practices, and attitudes concerning
soil-conserving agricultural practices. Agronomic questions focused on land-preparation
methods and the use of fertilizers, introduced seed varieties, pesticides, and herbicides.
Farmers were asked to describe agronomic problems experienced and changes imple-
mented in their agricultural practices. They were also asked to describe their perception
of the future for bean production in the area (Appendix A).
Costs and returns for frijol espeque production. In July and August, 1990,
following the inverniz harvest season, 31 frijol espeque farmers: 19 tenant farmers and
12 farmer-landowners were interviewed. Farmers were asked to list their production
costs. Specifically, questions were designed to determine the use of commercial inputs,
inputs of hired and family labor, tenancy costs, and marketing costs. Respondents also
were asked about bean production problems encountered, bean yields obtained,
marketing channels used, and off-farm labor activities (Appendix B).
Costs and returns for frijol tapado production. In December 1990 and January
1991, following the veranero harvest season, 28 frijol tapado farmers from the Pejibaye
area were interviewed. The Pejibaye area is relatively accessible, and introduced
technology-adoption is high. To obtain a comparison with frijol tapado use in less
accessible areas, where introduced technology-adoption is low, 28 farmers in Guagural
and San Vito were also interviewed. Farmers in these areas lived more than one hour
by motorcycle (during the dry season) away from a commercial center. Muddy roads
or swollen rivers often leave these areas isolated during the veranero growing season.
Frijol tapado farmers were asked questions similar to those asked of frijol
espeque growers. In addition, these farmers were asked to quantify profits from other
on-farm or off-farm labor activities in which they engaged during the veranero growing
season. Farmers were also asked questions to determine what factors affected their deci-
sions to continue planting frijol tapado (Appendix C).
Informal discussions were conducted with large land owners, tenant farmers,
extension agents, input vendors, and bank officials throughout the study period.
Landowners were asked about changes in the amount of land dedicated to bean produc-
tion, and about modifications of their tenancy and fallow-use practices. Tenants were
asked about the length of their tenancy agreements and about the form of tenancy
agreement they had with their landlord. Tenants were also asked if they would alter
their farming practices if they were farming their own land rather than sharecropped
land. Extension agents, input vendors, and bank officials provided information about the
services they provided to the farmers and about the farmers' use of their services.
These discussions were used to determine the extent of interaction between various
sectors of the agricultural community and to delineate likely changes in the farming
community and farming practices.
Measuring the Economic Impact of Technology Change
Bean production is only one of several economic activities for farmers in the
Pejibaye area. No attempt was made to compute the gross family income from all on-
farm and off-farm activities of the farm families interviewed.
The economics of bean production were calculated according to the following
Labor value was calculated based on the number ofjornales spent on each of the
following activities: land-preparation, planting, application of inputs, weeding, harvest-
ing, seed cleaning, and marketing. A journal is a farm-labor day, equivalent to approxi-
mately 6 hours. Each journal was valued at 350 Colones (107 Colones = $1 US, 1-1-
91). Family labor was valued at the same rate as hired labor.
Input costs included costs for purchased agrochemicals, seeds, sacks, and
marketing transportation costs. The value of seeds saved from the previous harvest was
based on the official CNP price paid for beans during the previous growing season. For
tenant farmers, input costs also included the value of the beans paid to the landowner as
rent. For landowners, costs of landownership and taxes were valued at 2000 colones per
hectare. This was roughly based on the land valued at 40,000 colones per hectare, 15
years for payment, and two payments per year. For most farmers, this value is
considerably higher than their actual expenses, since the land was acquired when land
prices were lower.
Gross income was based on the quintales of beans harvested multiplied by the
official CNP price per quintal. One quintal is equal to 45 kilograms.
Other economic factors calculated included net income, (cost of family labor was
not included as input costs), the ratio of gross income to input costs, and the ratio of net
income to jornales of family labor (the value of each journal of family labor).
Statistical differences between bean-production practices and economic indicators
were assessed using regression and t-test analysis methods of the statistical package SAS
Results and Discussion
Overview of Farming Practices
The farming community of the Pejibaye area consists of many small-scale
landholders and tenants living as neighbors to a few large landholders. All farmers
selected for interviews in this study planted a maize-bean rotation on steep slopes. Of
the 51 farmers initially interviewed, 18 (35%) planted frijol espeque as sharecroppers
while 33 (65%) owned the land they farmed. Over 70% of both tenant farmers and
farmer-landowners planted three hectares or less of beans (Table 2.2). Frijol tapado was
planted during the veranero season by 50% of the tenants and by 42% of the land-
Farming calendar. The calendar of activities of a farm family is determined
primarily by annual rainfall patterns and the production activities required for the crops,
trees, and livestock produced. The temporal interactions of the labor activities and
economic flows of the farm family are shown graphically in Figure 2.4. The availability
of high-value, off-farm or on-farm labor opportunities may compete with bean pro-
duction activities. Non-agricultural household expenses may not only force farmers to
sell produce, but may also prevent farmers from purchasing agrochemicals or labor
inputs. For many farmers, an important seasonal household cost is the purchase of
school uniforms and notebooks at the onset of the school year in March.
Alternative labor activities for tenant farmers. Tenant farmers divided their labor
time among bean production activities, work as day laborers, and harvesting coffee.
During the inverniz growing season, 65% of the tenant farmers worked as day laborers
on other farms. During the veranero growing season, 100% grew maize, 59% worked
as day laborers, and 88% engaged in coffee harvests (Table 2.2). Tenant farmers
harvested coffee either in the Pejibaye area or as temporary migrants to the Central
Mesa. Local coffee harvests extended from September to early December. The coffee
harvests of the Central Mesa extended from December to February. Women and
children as well as men participated in the local coffee harvests. Usually only men
migrated out of the area to pick coffee. Twenty-eight percent of the farmers who
planted beans as tenants produced coffee on their own land.
Alternative labor activities for landowners. Landowners did not engage in off-
farm employment, but instead divided their labor between bean production and other on-
farm activities (Table 2.2). During the inverniz growing season, landowners chopped
pastures, pruned and weeded coffee, and repaired fences in addition to frijol espeque
production. During the veranero growing season, agricultural labor activities of land-
owners included maize production, harvesting coffee on their own land, chopping pas-
tures, and caring for cattle.
Tenancy agreements. Tenant farmers grew beans according to two tenancy
arrangements; halves and thirds. According to the thirds agreement, the landlord
provides the tenant with the use of the land. In return, the tenant is responsible for the
costs of all commercial and labor inputs, and pays the landlord with one-third of the
crop yield. According to the halves agreement, as described by Barrantes et al. (1986),
the landlord provides the land and pays for half the costs of fertilizers, fungicides, and
insecticides. In return, the tenant is responsible for the entire cost of herbicides, half the
cost of fertilizers, fungicides, and insecticides, and for all of the labor costs. Crop
yields are split equally between the landlord and tenant.
Based on interviews with tenant farmers, practice of the halves agreement is quite
variable. Usually, instead of providing fertilizers, fungicides, and insecticides, landlords
only assist their tenants in the purchase of herbicides. Several tenant farmers com-
plained that they were not able to use agrochemical inputs to produce their crops because
landlords were unwilling to pay their half of the cost.
The thirds agreement was used by 71% of the tenants interviewed (Table 2.2).
It is the agreement preferred by most tenants and landlords. Tenants preferred this
agreement because it provides them with more freedom in their choice of farming
practices and is less costly. Landlords preferred it because it is less economically risky
in the event of a crop failure.
Duration of tenancy agreements. Tenancy agreements are generally of short
duration. Over 91% of the tenants interviewed held tenancy agreements for 5 years or
less (Table 2.2). One-third of the tenants had land use agreements for 2 years or less.
The short duration for lease agreements on steeply-sloping lands is due to perceived
decreases in production and profitability after two to three years of continuous cropping.
Following the expiration of tenancy agreements, leased land is either fallowed or placed
into other agricultural uses. Of the landlords interviewed, 70% indicated that, within the
past ten years, they had decreased the amount of land available to tenants for bean pro-
duction. Simultaneously, they had increased the amount of land in pasture.
As a result of the conversion of formerly sharecropped land to pasture use, 40%
of the tenants encountered difficulties in finding land to rent. More than two-thirds of
the tenants also had difficulties finding fertile land or land close to where they live.
Perceived decrease in the profitability of bean production. Most tenants said
their production costs have increased while yields have decreased. Reasons for
decreased profitability of bean farming most often noted were: overuse of land (65%),
high input costs combined with low bean prices (59%), and pest and disease infestations
(46%). All tenants indicated lack of land ownership as one of their greatest problems.
Having to give at least one-third of the bean yield to the landlord while also paying for
most or all of the production costs was the principal reason tenants gave for lack of
bean-growing profitability. One farmer noted, "due to the difficult economic situation
tenant farmers are facing today, I do not feel that we can do anything to improve our
The current economic situation may encourage tenants to use land-degrading
farming practices. Over 83% of tenants cleared land for frijol espeque production by
burning. Most of these tenants felt that fire "sterilizes" the land. They also admitted
that the farming practices they used on rented land abused or degraded the land. Half
of the tenant respondents said that, if they had their own land, they would not burn the
land but would instead use soil-conserving measures.
Despite the decreasing profitability of bean farming, most tenants felt that they
had no option but to continue farming. More than 83% of the respondents indicated that
they wished to quit farming and 70% did not want their sons to be farmers. They hoped
their sons would find less demanding and less risky work.
According to respondents, many tenant farmers already have left farming to look
for work in the urban sector or have moved out of the area in search of more fertile
lands. Typically, the sons who were continuing farming were described as those who
"do not like to study." That is, they were either disinterested in studying or unable to
do well enough in school to obtain permanent off-farm employment.
The out-migration of tenants appears to have created a circular problem. Several
large owners of more remote farms had difficulties in obtaining tenants to farm their
land. As a result, they increased the amount of land for pasture use and decreased the
amount of land they made available to tenants.
The majority of farmer-landowners owned small tracts of land and engaged in
diverse agricultural activities. Over one-half of the landowners owned less than 7 hect-
ares of land. Coffee was the major alternative crop for small holders (less than 7 ha)
since it has the potential to produce high economic returns per hectare. Conversely, due
to the low labor inputs required, larger landholders used a majority of their land for
cattle production. All landholders owning more than 14 hectares of land had some land
in pasture. Other commercially important crops produced by landowners included rice,
tomatoes, sugarcane, and cacao. Crops grown for home use included bananas, cassava,
and fruit trees.
Perceived decrease in the profitability of bean production. Similar to tenant
farmers, farmer-landowners felt that bean production was becoming less profitable.
Reasons given for the decreasing profitability of bean production included: high costs of
inputs combined with low prices paid for beans (82%), decreasing soil fertility (73%),
disease and pest infestations (64%), and lack of government assistance (64%). Due to
the perceived lack of profitability for bean growing, many landowners have placed much
of their former bean land into coffee, cattle, or vegetable production. Some landowners,
in desperation, have tried to extract themselves from debt by growing marijuana (R.
Floras, personal communication). Survey results also showed that the percentage of land
in bean production decreased as farm size increased. Although farm size ranged from
1.4 to 100 hectares, land devoted to bean production ranged only from 0.7 to 7.0
Agrochemical input use. Both MAG and CNP promoted the use of frijol
espeque. These organizations encouraged farmers to use certified seeds, fertilizers, and
pesticides in frijol espeque production. Recent MAG recommendations also encouraged
farmers to leave crop residues on the ground in order to reduce erosion losses. Due to
financial constraints, however, most farmers did not manage frijol espeque according to
extension recommendations. Although all landowners and tenants used bum-down
herbicides (Table 2.3), only 83% of the landowners used fertilizers. Less than 25% of
the landowners used a complete package of agrochemicals in their production of frijol
espeque. Of the tenant farmers, less than one-third used either fertilizers, fungicides, or
Two-thirds of the farmers used fire to clear their land prior to planting frijol es-
peque. Reasons given for land burning included decreased labor costs and a belief that
fire helps control pest and disease infestations, especially slugs.
Most of the farmers have adopted introduced seed varieties for frijol espeque
production (Table 2.3). Over 80% of all respondents used either Talamanca or Brunca
seed varieties. Although these are not the only varieties of introduced seeds, they are
the most recently released varieties and the varieties most commonly sold as certified
seed by the CNP. Only 18% of the landowners purchased seed from the CNP, while
none of the tenants interviewed used certified seeds. Farmers who did not purchase
seeds from the CNP used seeds saved from the previous harvest. Farmers often used a
combination of seed varieties. Almost one-half of the respondents used Talamanca or
Brunca in combination with either traditional varieties or previously released introduced
varieties. Although some farmers blended seed varieties, farmers usually selected
specific seed varieties depending on the slope, soil fertility, or fallow growth of each
portion of the field.
Labor inputs. Labor inputs for frijol espeque production included land clearing,
planting, application of fertilizers, herbicides, and pesticides, harvesting, threshing,
drying, and marketing. Family labor was generally limited to fathers and sons or
associations between brothers. Wives or daughters assisted in fertilization and harvest
activities according to 38% of the respondents. Hired labor was used by farmers to
assist primarily with land preparation and planting. One-fourth of farmer-landowners
used hired labor to assist with frijol espeque activities. Tenants were more than three
times as likely to employ hired labor as were landowning farmers. Almost 80% of the
tenant farmers used hired labor in combination with their own labor (Table 2.3).
Production problems. Web blight was reported as the major disease problem by
70% of all farmers interviewed, while slugs were mentioned as a pest problem by 51%.
The other major production problem, mentioned by 65% of the farmers, was environ-
mental; the sprouting of beans in the pods due to harvest-season rains.
Input use. Frijol tapado is a traditional planting method utilizing minimal
agrochemical and labor inputs (Table 2.3). Labor inputs for frijol tapado were required
only during the planting and harvest seasons. This allowed frijol tapado farmers to
devote their time to other on-farm and off-farm labor during the period of bean growth.
Agrochemical inputs typically are not used in the production of frijol tapado.
The introduction of slugs into the region, however, induced 35% of frijol tapado farmers
to apply molluscacides to their fields. The application rate used was between 10% to
25% of the recommended rate. According to farmers, slugs presented a greater problem
in annual frijol tapado fields than in fields that had been left in fallow for more than two
Preferred fallows for frijol tapado. Farmers preferred using land having an
easterly exposure for frijol tapado production. This orientation allows for the early
morning sun to dry the land, resulting in a decrease in the transmission of web blight
and other fungal diseases. Land that had been in fallow for at least two years was
preferred by 68% of the frijol tapado growers.
Preference was also given to land where slow-growing perennial plants dominated
the fallow growth. Many farmers mentioned three plants in the Compositae family; tora
(Verbesina sublobata), tuete (Verononia spp.), and paira (Melanthera aspera), as indica-
tors of soil fertility. Through discussions with farmers and observations of frijol tapado
fallows, a list of plants dominating fallows of various age was developed (Table 2.4).
Plant species and associations commonly associated with frijol tapado fallows are
illustrated in Figure 2.5.
Duration of fallow and tenancy. The length of the fallow used between
successive croppings of frijol tapado depended on the size of the landholding, competing
farming activities, the distance of the farm from markets, and the customs of the farmer.
Since frijol tapado is a land-use extensive farming method, only farmers who owned
more than five hectares of land were able to grow frijol tapado on their own land.
Farmers who owned smaller tracts of land grew frijol tapado as tenants.
Of the tenant farmers, 83% experienced difficulty in locating good-quality land
for tapado use. Despite expressed preferences for an extended fallow, most tenants
indicated that they were willing to "plant frijol tapado in whatever type of fallow they
Production problems. Production problems reported for frijol tapado were
similar to those reported for frijol espeque: slugs, web blight, and seed sprouting. Slugs
were mentioned as a problem in frijol tapado bean production by 44% of the farmers
interviewed, while web blight was a problem for 25% of the respondents. The problem
of web blight was less for frijol tapado compared to frijol espeque. Similarly, only
27% of the farmers mentioned sprouting beans as a frijol tapado production problem,
compared to 65% who noted it as a problem for frijol espeque. The combination of a
mulch cover and lower inoculum levels in frijol tapado fields was responsible for the
decreased incidence of web blight. The more favorable weather during the frijol tapado
harvest season compared to that of frijol espeque decreased problems associated with
The Transition from Frijol Tapado to Frijol Espeque
Based on farmer interviews, a decision tree (Gladwin, 1989) was developed
summarizing the major factors affecting farmers' decisions to use frijol tapado or frijol
espeque (Figure 2.6). The decision to plant frijol tapado was based on climatic factors,
land availability, availability of high-value off-farm employment during the frijol tapado
growing season, land slope, access to inputs, and farmer custom. Frijol tapado is
favored because of the availability of excess land, low access to economic resources and
financing, remoteness from sources of inputs and marketing centers, high opportunity
values for other on- or off-farm labor activities, and heavy rainfall. Frijol espeque is
favored because of high land-use competition, accessibility of inputs and markets from
the farmstead, and relatively more cash and labor availability.
The principal influencing factor was climatic. Farmers growing beans in
monsoonal areas practice frijol tapado during the veranero season. Heavy rainfall during
this growing season prohibits land-burning. The rainfall erosivity is also greater during
this growing season than during the inverniz season. By using frijol tapado, a mulch
layer is retained on the soil surface to protect against raindrop impact and the associated
soil losses due to erosion as well as the dissemination of fungal inoculum.
Due to the high opportunity value for labor during the veranero growing season,
frijol espeque is not commonly practiced during this season. Instead, the adoption of
frijol espeque is usually accompanied by a transition from using the veranero season to
using the inverniz season as the principal bean growing season. Similarly, the introduc-
tion of high yielding maize varieties resulted in a change in the primary maize growing
season from the inverniz to the veranero season. In contrast to the traditional varieties,
ears of introduced maize varieties are highly susceptible to rot if left on the stalk during
the rainy season. Farmers who engage in multiple on-farm and off-farm activities
double-over the maize stalk and do not harvest the ears until they are finished with more
urgent activities. Due to the extended dry period at the end of veranero, introduced
maize varieties can be left safely in the field at the end of the growing season. Thus,
genetic characteristics of the introduced varieties of beans and maize have contributed to
the reversal of bean and maize growing seasons.
Adoption of frijol espeque during the inverniz season is precluded in areas where
continuous rainfall at the end of this season present a high risk of crop loss due to bean
sprouting. Pejibaye is an example of an area with a moderated monsoonal climate,
while San Vito and Guagural experience continuous rainfall. Consequently, Pejibaye is
a high adoption area while San Vito and Guagural are low adoption areas.
Farmers in the Pejibaye area adopted frijol espeque as the primary mode of bean
production while continuing to use frijol tapado as a secondary source of beans and cash.
Frijol tapado was grown in this area primarily as a subsistence crop. Only one-half of
the farmers sold a portion of their frijol tapado crop. The remainder grew frijol tapado
for home consumption and to produce seed for frijol espeque. Since farmers used frijol
tapado to augment their seed stores for frijol espeque, over 60% of frijol tapado farmers
used introduced seed varieties either exclusively or in combination with traditional
Farmers also used frijol tapado as a crop producing method for bringing forested
or fallowed land into agricultural production. Consequently, only 38% of the parcels
used for frijol tapado were returned to fallow directly following the frijol tapado harvest
(Table 2.5). When used as means of land clearing, frijol tapado was planted concurrent
with clearing fallow growth during veranero. Following the dry season, weed regrowth
was cut, then the weeds and mulch residues were burned. The chopped residues were
burned to increase the availability of plant nutrients and to provide a clear field for
planting frijol espeque during the inverniz season.
Alternatively, land was used for a frijol tapado and pasture rotation. A field is
used for several years as an unimproved pasture, then placed into fallow for one to two
years to regenerate soil fertility and decrease weed growth. Prior to returning the land
to pasture use, the fallow is chopped concurrent with the planting of frijol tapado.
Farmers in the areas of San Vito and Guagural planted frijol tapado either
exclusively or as part of a rotation with frijol espeque, maize, and land fallowing. All
farmers in these areas continued to utilize a crop-fallow rotation (Table 2.5). Most
farmers produced frijol tapado for sale as well as for subsistence use. They also used
traditional seed varieties primarily. The preferred traditional varieties in both regions
were Chimbalo Rojo, Chimbalo Negro, and Vifia Blanca. These varieties are classified
as Type III and Type IV. Farmers who used traditional varieties noted the higher
resistance to pests and diseases of these varieties compared to the introduced varieties.
Farmers who grew frijol tapado for market grew Chimbalo Rojo, since a premium price
was paid for red beans. Only 8% of the farmers used introduced varieties exclusively.
Economic Aspects of Frijol Espeque and Frijol Tapado
Frijol espeque required a higher level of economic investment and provided
higher gross returns than frijol tapado (Table 2.6). Although differences were not
significant, the average jornales of labor reported for frijol espeque production were
almost double the labor inputs used for frijol tapado. The distribution of labor, through-
out the growing season, for the two bean-growing practices also differed. Frijol espeque
requires crop management activities while frijol tapado does not involve any labor inputs
during the growing season. Average costs for agrochemicals and seeds for frijol
espeque were over four times more than for frijol tapado. The average yields reported
for frijol espeque were more than two-and-one-half times those obtained from frijol
tapado. The highest yields reported for frijol espeque (1300 kg/ha) were more than
three times higher than the highest yields reported for frijol tapado (383 kg/ha). Near-
total crop losses, however, were reported for both planting systems.
Landowners spent significantly more on commercial inputs and obtained higher
yields for frijol espeque than did tenants. No significant differences in bean yields for
landowners compared to tenants were obtained for either frijol tapado or frijol espeque
when the planting methods were analyzed separately. When comparisons were made
across planting methods, yields obtained by landowners were significantly higher than
those obtained by tenants.
Due to the high rental costs, the value of each journal of family labor in frijol
espeque production was three times more for landowners than for tenant farmers. For
landowners, the value of each journal was over twice the amount that they could have
obtained working as a day laborer on another farm (350 Colones/ journal For frijol
tapado production, the value of their labor was slightly more than that paid to day labor-
ers. For tenant farmers, the net returns obtained per journal for both frijol tapado and
frijol espeque were less than the wage paid to a day laborer. No significant differences
in the value of each joral were obtained when comparisons were made between frijol
espeque and frijol tapado. For landowners, the average value of each journal of frijol
espeque labor was over twice that of each journal of frijol tapado labor. For tenants, the
value of each journal of frijol tapado labor was slightly higher than the value of each
journal of frijol espeque labor.
Land Tenancy and the Adoption of Frijol Espeque
Economic surveys results showed that the adoption of frijol espeque benefitted
landowners more than tenants. Landowners were able to apply more commercial inputs
to their frijol espeque fields due to their greater access to economic resources. Since
they incurred greater hired-labor expenses, the total costs incurred by tenants were often
higher than the costs incurred by landowners. Tenant farmers employed hired labor in
order to decrease the number of visits they needed to make to fields far from their
homes, or in order to free up their time so that later they could sell their own labor. As
a result of low yields, high labor costs, and high rental expenses, the net returns
obtained by the tenant farmers were less than one-half of the net profits obtained by
As the profitability of bean production declines, larger landowners have the
opportunity to transfer their land out of bean production and into more profitable
commodities; coffee or cattle. Depending on farm size and access to economic
resources, landowners also can increase the sustainability of bean production by adopting
accelerated fallows, labranza cero practices, or agroforestry techniques.
The options available to the one-third of the bean farming community who are
tenants are much more limited. Tenant farmers received significantly lower returns for
their bean growing activities, yet often did not have the option of changing to a more
profitable commodity. This was especially true for tenants living in areas with low
accessibility to markets. For this sector of the farming community, the most viable
alternative for increasing household incomes was to leave farming and obtain a perma-
nent labor position elsewhere. Although the majority of tenant farmers expressed a
desire to leave farming, they continued to engage in bean farming due to a lack of
permanent labor opportunities.
Tenants who remain in farming can try to sharecrop land that was recently
fallowed or that has relatively fertile soils. In this way, they increase their chances of
obtaining saleable yields without having to purchase fertilizers, insecticides, or fungi-
cides. The practice of extracting plant nutrients through farming without subsequent
replacement through fertilizer use can increase the profitability of bean production for
resource-poor farmers. It also enhances soil degradation and decreases sustainability.
Tenants can also use low-input, low-risk farming practices. For them, frijol
tapado represented an economic safety net for subsistence food source and increasing
household income. Although farm families were able to obtain relatively high profits
from coffee harvests, most frijol tapado farmers needed to both harvest coffee and grow
frijol tapado. The receipts from the coffee harvest paid for household expenses during
the veranero growing season, while yields from the frijol tapado harvests supported the
family during the extended dry season.
These results are in contrast to the results reported by Pachico and Borb6n (1986)
who concluded that frijol tapado was more important for large-scale producers than for
small-scale farmers. The present study indicates that frijol tapado was used for seed and
subsistence production by farmers with landholdings sufficient to place land in fallow.
As the profitability of bean production decreased, however, landowners removed their
land from frijol tapado use and placed it into coffee and cattle production. In contrast,
diminished land availability for frijol tapado production displaced tenants from farming
and caused increased out-migration to the Central Mesa in order to pick coffee, or to
Lim6n to work in the banana plantations.
Complementary and Competitive Interactions between Friiol Tapado and Friiol Espeaue
In the Pejibaye area, frijol tapado and frijol espeque exist as both complementary
and competitive forms of production. Over 40% of all bean farmers practiced both frijol
espeque and frijol tapado. The complementary nature of the two bean planting practices
was indicated by the use of the same varieties of seeds for both frijol tapado and frijol
espeque. Also, frijol tapado was used as a transition between land fallowing and frijol
espeque production. Landowners who rented land to tenants for frijol tapado produc-
tion, however, often were more interested in the land clearing accomplished by the frijol
tapado activities than in the bean yields they received. Consequently, much of this
cleared land was subsequently removed from frijol tapado production after harvest.
Limitations to Increasing the Profitability of Bean Farming
The actual frijol espeque practices used by small-scale farmers did not reflect
their knowledge about either soil conservation or input-use practices. Instead, it reflect-
ed their conservative use of available resources. This contradiction between knowledge
and practice was summed up by one farmer. He said, "There are many inequalities; we
want to use recommended methods but we cannot because we do not have sufficient
resources." Most frijol espeque farmers felt that pest problems could be decreased and
yields increased by not burning the land, and by using instead labranza cero or a regular
fallow rotation. Use of rotations, however, was considered by many small-scale
landowners as "a luxury." Similarly, many farmers wanted to adopt labranza cero
techniques but felt unable to do so because of the additional time required to plant within
residues compared to planting a cleared field.
Availability of cash, labor, and land resources for bean production was limited
by the need of the farm family to use these resources for other purposes (Figure 2.7).
Monies for chemical inputs competed with monies for food and school expenses
since the frijol espeque season begins at the end of the long, dry, "hungry" season and
concurrent with the beginning of the school year. The onset of school also prevented
children from assisting their parents in bean production activities. In addition, labor
availability for bean production, especially during the veranero growing season, was
limited by higher opportunity values for labor in other on- and off-farm activities. Land
availability for bean production was limited by the high opportunity value for land in
vegetable production and by farm loan programs promoting the production of cattle and
coffee. The farm size and the opportunity value for land use in other agricultural
activities also affected the duration and frequency of the fallows used.
Changes in political policies have affected the profitability of bean production.
In an effort to reduce government costs and decrease redundancy in government
services, the extension arm of the CNP was dismantled and the BNCR offices in small
towns, including Pejibaye, were closed in 1990. As a result, the rural population be-
came increasingly disarticulated from the urban market economy and farmers became in-
creasingly dependent on local vendors for loans and marketing outlets. Meanwhile,
rapid inflation resulted in a corresponding decrease in the price of beans while fertilizer
prices remained relatively stable (Figure 2.8). These factors further decreased the
margin of profit which farmers receive for their bean crops.
Options for Increasing the Profitability of Bean Farming
Farming systems and farmer-first methodologies were developed in an attempt to
direct development programs at the needs and constraints of the poorest sector of the
farming community. These methods stress the use of alternative organizational forms,
management techniques, and research methods appropriate to a people-centered
development and adapted to the context of the area of project implementation (Cham-
bers, 1986). Bureaucratic constraints, unfortunately, often have resulted in development
organizations becoming promoters of predetermined technologies in pre-identified target
areas rather than as catalysts for change in self-identified communities (Hellinger et al.,
Currently, several development organizations are involved throughout Costa Rica
in the promotion of agroforestry and accelerated fallow technologies designed to increase
the sustainability of frijol espeque production. In contrast, very few organizations are
concerned with the opportunities and constraints of frijol tapado production. Similarly,
few development or research programs have addressed the bean production problems of
The current lack of sustainability of bean production in Pejibaye is caused by a
combination of agronomic, economic, and political factors. Agronomically, intensifi-
cation of land use resulted in the adoption of land use practices inappropriate to the
steeply sloping terrain of this area. Economically, many farmers are unable to afford in-
puts required to make continuous cropping practices sustainable. Politically, govern-
mental policies benefitted producers of export commodities and larger landholders at the
expense of producers of subsistence crops and resource-poor farmers. Also, a history
of mismanagement or partial implementation of programs by MAG and CNP extension
offices discouraged farmers from associating with these organizations. Several farmers
said they felt that extension agents were incompetent or "spent all their time driving
around in expensive vehicles visiting the richest farmers."
Agronomic solutions to political problems are usually not as sustainable as
agronomic solutions to purely agronomic problems. Unfortunately, many of the
agricultural problems facing resource-poor farmers in the developing world have political
components. Small-scale bean farmers in the Pejibaye area are involved in a cycle of
decreasing sustainability. As population pressures and inequities of land tenure relations
increase, the availability of land decreases. This encourages small-scale farmers to use
intensive land use practices that are inappropriate for marginal lands. While these
practices may produce higher yields in the short-run, in the long-run they encourage soil
degradation and result in farmers having to apply increasing levels of inputs in order to
obtain decreasing yields. This results in a decline in both the economic and agroecolo-
gical sustainability of the farming practice.
As a result of the cycle of degradation in which small-scale and tenant farmers
must operate, recommendations that can be adopted by them may not offer long-term
sustainability. Economically viable options available to these resource-poor farmers
1. Planting frijol espeque on fallowed land using a minimum of fertilizers and
2. Planting frijol tapado
3. Using an intermediate technology that would be responsive to commercial
inputs but would require a minimum of labor expenditures during the
To test the sustainability of these options, agronomic experiments were con-
ducted. Labranza cero was chosen as the intermediate technology to be tested. It was
chosen because the mulch cover would maximize nutrient recycling within the system,
decrease soil losses from erosion, minimize the time required for manual weeding
activities, and protect against bean production losses due to web blight. Planting of
beans in a straight row rather than broadcasting seeds under a mulch cover should
increase the bean emergence percentage and allow farmers to manage their crops more
Experiments were designed to:
1. Compare farmers' implementation of frijol espeque practices in terms of the
effects of crop and land management practices on yields, economic
returns, soil nutrients, and soil losses due to erosion.
2. Compare the effects of land use intensification on yields and soil degradation
for frijol tapado, labranza cero, and clear-field planted frijol espeque
Throughout this series of agronomic experiments the following questions were
1. What are the trade-offs between decreased soil degradation and increased
2. What is the relationship between maximal yields and maximum profitability?
3. Which components of each technology affect its sustainability or its ability to
be adopted by resource-poor farmers?
The integrated investigation of agronomic and socioeconomic factors permits both
a critical evaluation of existing technologies and the recommendation of alternative
technologies to increase the sustainability of bean production on steep slopes.
Province of San Jose-` .
District of Perez Zeledon
de Perez Zeledon
Figure 2.1: The location of Pejibaye de Per6z Zeled6n.
Figure 2.2: Rainfall for 1990 at two elevations in Pejibaye de Perez Zeled6n.
Source: Instituto Costarricense de Electricidad (1991)
IJN FE MC I I I I I I I AUI I
JAN FE MARCII APRIL MAY JUNE JULY AUG SEPT OCT NOV DEC
Figure 2.3: Deforestation of Pejibaye de Perdz Zeled6n-
Source: Instituto Nacional Geografico.
0 O I
MARCH APRIL MAY JUNE JULY AUG SEPT OCT NOV DEC JAN FEB
INVERNIZ GROWING SEASON
VERANERO GROWING SEASON
Figure 2.4: Calendar of agricultural activities and school semesters in Pejibaye
de Perez Zeled6n in relationship to rainfall.
Tora (Yerbesina sublata)
A typical nine month fallow with
plantanilla (Heliconia spp.) and jaragua
(Hypahenia rufa) predominating
Figure 2.5: Common plants and plant associations found in frijol tapado fallows.
Guarumo (Cecropia spp.)
A typical woody two to three year
fallow used for frijol tapado production
Is there a high
Is land available to
Is the area subject
to monsoonal rains?
Opportunity value Is them a high opportunity value
activities other for labor activities other
bean growing? than bean growing?
I yes no
Is competition for I
land use high? Is the land
I steeply sloping?
Does the farmer like Docs
to plant frijol tapado? plant b
Arc drying facilities
available during invemiz?
espeque Docs not
means Does the farmer
have access to inputs?
Does the farmer like
to plant frijol tapado?
Plant both frijol tapado
and frijol espcque
is the land
Plant frijol espcque
or frijol tapado
Figure 2.6: Frijol tapado and frijol espeque decision tree.
SIZE OF FAR
S ACCESS TO MARKETS 0
A / SCHOOL
LAND BEAN LABOR / OFF-FARM
AVAILABILITY -" PRODUCTION" AVAILABILITY LABOR
Figure 2.7: Socioeconomic interactions affecting the sustainability of bean
CHANGE IN PRICE OF BEANS AND FERTILIZER
50 KG FERTILIZER
50 KG BLACK BEANS
1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
CHANGE IN COLON:DOLLAR EXCHANGE RATE
1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
Figure 2.8: Changes in the prices of beans and fertilizer and in the colon:dollar
exchange rate: 1975-1991.