|
LINKING SUSTAINABILITY, FOOD SECURITY, AND
IMPROVED WORKER LIVELIHOODS IN AN
ECUADORIAN AGROSOCIOECOSYSTEM
By
NORMAN E. BREUER
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
2003
LINKING SUSTAINABILITY, FOOD SECURITY, AND
IMPROVED WORKER LIVELIHOODS IN AN
ECUADORIAN AGROSOCIOECOSYSTEM
By
NORMAN E. BREUER
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
2003
Copyright 2003
by
Norman E. Breuer
To my wife Bea; and to my children, Norman III, Astrid, and Erik.
ACKNOWLEDGMENTS
My deepest thanks go to the Chair of my committee Dr. Peter E. Hildebrand. Over
the past 4 years, he provided wisdom, counsel, and example. At the same time, I would
like to express my appreciation to the other members of my committee. The support and
insights of Drs. Hugh Popenoe, Clyde Kiker, Nigel Smith, Janaki Alavalapati, and
Ram6n Espinel were instrumental in keeping the often wayward track of my
interdisciplinary wanderings on course.
Financial constraints have been the norm of my life during my graduate studies.
Many institutions contributed to alleviating this condition, allowing me to push forward
with my endeavor. I am indebted to the following for their financial support: Food and
Resources Economics Department and College of Agriculture and Life Sciences,
University of Florida; the Compton Foundation and the Center for Tropical Conservation
and Development Studies, UF; the Tinker Foundation; The Wong Foundation; Jos6,
Ver6nica and Michelle Breuer; and Mercedes, Francisco, and Bibiana Barriocanal.
So many people in Ecuador supported me that this expost listing is sure to be
plagued by inadequacies. I will start from the beginning, with the dynamic and sincere
Rafael Wong, whose meeting with Peter E. Hildebrand in early 2000 sparked the present
research. Dr. Lucila Pdrez, Executive Director of the Wong Foundations did everything
she could to provide me with adequate lodging and transportation. The following persons
at Reybanpac/Favorita Fruit Company provided comments, insights, and a helping hand:
Jorge Josse, Edgar Montero, Pedro Meza, Jorge Jativa, Mike Utley, Alexandra Jal6n, and
Fernando Torres, and others too numerous to list here.
Wong Foundation personnel at the Rio Palenque Science Center were the persons
with whom I interacted on a daily basis. For their encouragement and goodwill, I thank
each and every one of them. First and foremost is Leonardo Rodriguez. Venus Ar6valo
also helped me a great deal. Carlos Criollo introduced me to the area, its customs, and
people. I wish to express my appreciation to the small-scale farmers, plantation and
town-dwelling field workers, mothers, children, teachers and parents who opened their
minds (and very often, their hearts) to me. I was warmly received by the Paredes Crespo
family in Quito during my research sojourn in that beautiful capital city.
A man who has friends truly possesses a great treasure. I take this opportunity to
thank Matt Langholtz, an example of all good things meant by being American. Manuel
Avila kept me physically operating and mentally in tune with my homeland, Paraguay. I
especially thank my friend and colleague Victor Cabrera for being "un caballero" in
every sense of the word. I would like to thank Father John Gillespie and the community
at St. Augustine's Catholic Church in Gainesville for providing a structure for my
spiritual functions and needs.
I cannot help but put in a word of thanks for my parents, Norman and Sacha. I
have no words to express my appreciation for the love, support, patience, and tolerance of
my one-in-a-billion wife, Bea; and my unique children Norman III, Astrid, and Erik. I
have been blessed in many ways in Gainesville, and I am thankful for this.
I
I
I
TABLE OF CONTENTS
page
ACKNOW LEDGM ENTS................................. ............................................. iv
LIST OF TABLES.................................................................................. x
LIST O F FIG U R E S ........................................................................... xi
AB STRA CT ........................ ....................... .................................. xiii
CHAPTER
1 INTRODUCTION ........................................................... .. .......... .... .. .. .. 1
Background...................... ....... ......... ...................... 5
Problematic Situation .................... ......... .. .................... 6
Researchable Problem .. ............... ......... ............................ ................. 18
Research A rea ........ .................. .............................................. 18
O objectives .............................................................. ..................................... 21
H ypotheses ................... .................. ......... ............... ....-....-- ................ ....... 22
M methods .................................. .............. ............ ....... 23
A analysis ........................ ..... .............................. ..... .................... 26
O organization of the Study ..................................................................... 26
Significance.......... .......................................... 28
2 LITERATURE REVIEW: THEORETICAL UNDERSPINNINGS OF THE
STUDY............................. ................................... 30
Sustainability .................................... ...... ... .... .. ................... 3 1
Resilience: A Concept, and Indicator, and an Emergent Property.......................... 33
System s A pproach............................................................................. 38
Agricultural and Agroecological Systems............................................ 41
Banana Dominated Agroecosystems.................................... ........................ .....45
Environmental Aspects of Certification and Forest Fragments ...................... 46
Modeling as an Analytical Tool............................... ...... ........................ 53
Linear Programming............................................................. 55
Ethnographic Linear Programming ............................................................. .... 57
Participatory Linear Programming..................... .............................................. 58
Linear Programming and Agriculture ................ ........................ 59
Linear Programming Applied to Farm Household Analysis ............................ 61
viI
vii
Linear Programming and Sustainability........................................................... .. 63
Rural Development and Natural Research Management ................................... 67
Summary ....................................... : ....... ........................ ..................... 73
3 COMMERCIAL BANANA PLANTATION SYSTEM...................................... 74
Introduction..................... ......... .. ........................................................................ 74
Background ................................................................ 75
Banana Production in Ecuador............................................................................. 76
M methods ............................................. ... ............ .................... ...... ... 79
Importance of Bananas to the Ecuadorian Economy ........................................... 81
M markets ....................... .. ........................... ........ ......................... 83
Company Banana Haciendas ........................................... ......................... 84
Banana Production ................................................... 85
B anana C crisis ......................... .............................. ................................. 10 1
Strengths and Weaknesses in Commerce ......................................................... 103
M anagem ent............................... ..................................................................... 104
C certification ............................................................... ................. 107
Best- and W orst-Case Scenarios ....................................................................... 107
Linkages with the Agrosocioecosystem .................................................. 108
Sum m ary ................... .. ... .......... .. ............................. .... .... 111
4 PLANTATION WORKER COMPONENT..................................................... 112
Introduction ................................... ....... ........................................... ......... 112
M methods ....................................................................................................... 113
Background............. ................. ............... ...... ............................ 116
Food Security ..... ........... ......................... ............. 119
Types of W orkers................................ .... ... ............... ..... ............... 120
W hy they Cam e .................................................. .... ............... ............... 123
Life on the Plantation .................. ......... ..... ...... ........ ... 124
Women as Part of the Study Group: Some Qualitative Data................................ 127
Child Labor and Unionization................................. ............ 128
Life in Tow n ............................................. ........ ...................................... 131
A "Banana Worker Community?"....................... ... ......................... 135
Emigration, Security, Aspirations for the Future............................................... 139
Being Better off .................................. ........ ........ .. ......... 143
Summary and Discussion........................................ 146
5 SMALL-SCALE FARM LIVELIHOOD COMPONENT................................... 153
Introduction ............................ ................................. ..................................... 153
Complexity and Diversity within Livelihood Systems ................................... 156
M methods ............................................................ .. ................................ 157
Small-scale Farms in the Research Area .................. .................... 158
Livelihoods Aspects (off-farm )............................................... ........................ 160
Cropping System s .................................... .... ............... 164
I
viii
I
C rops........................ ............................................................................... 165
A nim al Com ponent ....................................................................................... 174
Opportunities/New Technologies........................................................................ 175 i
Analysis ................................... .................... .................... 179
Linkages to the Overall System .......................................................................... 179
Sum m ary ...................... ................................ .............................................. 181
6 ENVIRONMENTAL COMPONENT: CERTIFICATION AND THE RPSC
FOREST RESERVE ......................................... ........... ..................... 184
Introduction ................................................... ............................................... 184
M methods ............................................ ............................... .. .................... 185
Environmental Certification.......................................................................... 187
A Token Forest? .................... ............................. ...................... 200
A analysis ..................... ........ .............................................. 206
Linkages to the Agrosocioecosystem...................................................... 211
Resources and Lim itations.................................................................................. 214
LP A nalysis......................................................... ................. ......................... 2 15
Sum mary.................................... .................. .............................................. 215
7 ASSESSING RESILIENCE IN THE AGROSOCIOECOSYSTEM.................. 217 I
Introduction...................................................................... ................ 217
W orker Survey ................................................ ............................................ 2 18
H household M odels.............................................................................................. 220
Agrosocioecosystem M odel.............................................................................. 245
Science Center and Forest Reserve .............. ......................... 248
Sum m ary ................................................ .................. .......................................... 253
8 SUMMARY, FINDINGS, AND A VISION OF THE FUTURE......................... 255
Overview of the Study............ ... ... ............... ................................. 255
Summary of Findings ..................................... ........................ 257
Secondary Outputs ...................... ...................259
Vision for the Future ........................................................................ 261
Policy Implications ........................................... 267
Further R research ........................................ .................................................... 269
Lim stations of the Study ......................................... ... ............................. 270
APPENDIX
A SUMMARY OUTPUTS AND AGROSOCIOECOSYSTEM MODEL........... 273
B SMALL-SCALE FARM INPUT-OUTPUT DATA AND MODEL EXAMPLE. 298
C TOWN-DWELLING WORKER DATA........ .......................................... 311
I
I
ix
I
I LIST OF REFERENCES .................................................................................... 329
I BIOGRAPHICAL SKETCH............................................................................... 349
I
I
I
I
I
I
I
I
I
I
I
I
LIST OF TABLES
Table page
3-1. Banana area per province .................................................................................... 82
3-2. Banana exports from Ecuador ............................................................................. .... 84
3-3. Commercial banana data in Rio Palenque agrosocioecosystem area......................85
3-4. Fertilization on commercial banana plantations in Los Rios Province, Ecuador......91
3-5. Labor supply gaps calendar .................................................................................97
3-6. Banana prices at major export destinations and to grower......................................98
4-1. Categories, numbers, and percentages of workers surveyed.................................116
4-2. Mean responses to grouped perceptions questions..........................................150
4-3. T-test comparing food security and being better-off responses ............................150
4-4. T-test comparing food security and healthcare responses .......................................150
6-1. Plantation size and planted forest area .................................................................. 196
7-1. Mean responses to grouped perceptions questions.............................................219
7-2. Small-scale sample farm household data..............................................................228
7-3. Household composition of modeled small-scale farm households .........................229
7-4. Summary of small-scale farm model scenarios............................................... 230
7-5. Modeled town-dwelling worker household data .............................................235
7-6. Contribution of RPSC to agrosocioecosystem economic output...........................252
LIST OF FIGURES
Figure page
1-1. Map of Ecuador and study area..........................................................20
1-2. Schematic of the study area...............................................................25
3-1. Banana exports vs. total exports in Ecuador..............................................85
3-2. Commercialization scheme used by an Ecuadorian commercial banana company
that operates in the study area...........................................................105
4-1. Responses to 16 worker perception survey questions...............................147
6-1. Schematic of biological corridors and potential location of small-scale farms on
their borders............................................................................... 186
6-2. Conceptual scheme of environmental fruit certification.............................. 189
6-3. Map of the Rio Palenque Science Center and Forest Reserve.......................205
7-1. Resilience of 32 sampled small-scale farm households to 14 shock scenarios......231
7-2. Resilience of 32 sampled town-dwelling worker household to 14 shocks .........238
7-3. Effect of 15 scenarios on discretionary cash, average discretionary cash, and
threshold of resilience...................................................................243
7-4. Town-dwelling workers. Effect of 15 scenarios on discretionary cash, average
discretionary cash, and threshold of resilience.........................................243
7-5. Land area needed for resilient small-scale farms....................................244
7-6. Resilience of ASES under several scenarios...........................................249
7-7. Earnings and income distribution in the ASES.........................................249
7-8. Earnings and distribution in the ASES including negative scenarios...............250
7-9. Economic output in the ASES under seven positive scenarios.......................250
xii
7-10. Effect of different science center and forest reserve uses on the ASES............253
8-1. Banana plantations including natural areas, and a forest reserve in the
agrosocioecosystem as it currently exists.............................................262
8-2. Phase 1. Corridor between plantation No. 1 and the Rio Palenque forest. Small-
scale farms line and protect the natural corridors....................................263
8-3. Phase 2. Corridor between plantation No. 1 and plantation No. 2. Small-scale
farms line and protect the natural corridors...........................................264
8-4. Phase 3. Corridor between plantation No. 2 and plantation No. 3. Small-scale
farms line and protect the natural corridors............................................265
8-5. Phase 4. Corridor between plantation No. 3 and plantation No. 4. Small-scale
farms line and protect the natural corridors.........................................266
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
LINKING SUSTAINABILITY, FOOD SECURITY, AND IMPROVED WORKER
LIVELIHOODS IN AN ECUADORIAN AGROSOCIOECOSYSTEM
By
Norman E. Breuer
May 2003
Chair: Peter E. Hildebrand
Department: College of Natural Resources and Environment
Ecuador is the world's largest banana exporter. On the Ecuadorian Coast an
important part of the population live as limited-resource farmers or as a large class of
landless rural plantation workers. Much agriculture in Ecuador depends heavily on hand
labor. However, many people are migrating away from the country, due to economic
crises and other factors.
This study presents an assessment of the current situation in a selected
agrosocioecosystem, by studying its principal components, their resilience, and what
economic output they provide. The study also assesses the benefits of remaining a small
farmer, as an alternative to migration. There are a limited number of livelihood options
in the study area. People can be small-scale farmers; town-dwelling, salaried-plantation
workers; live and work on plantations; or migrate. This lack of opportunity creates an
unstable social situation. Four components or sub-systems were studied: commercial
xiv
banana plantations; town-dwelling plantation workers; small-scale farmers; and nature
reserves.
Analysis was undertaken using Ethnographic Linear Programming (ELP), which
uses qualitative and quantitative data to estimate systems outcomes under several
scenarios. Elicited data were used to construct models. Households were subjected to
sudden shocks, and those able to best respond were said to possess higher socioeconomic
resilience.
The study found that small-scale farmers are more socioeconomically resilient to
shocks than town-dwelling plantation workers. Transferring households from the town
labor supply to small-scale farms improves economic output and adds resilience to the
system.
A rural workers survey revealed that small farms are perceived as the safest, most
food-secure place to live. In contrast, the sample population perceived schools and
clinics to be more accessible in towns.
Finally, the study concluded a nature reserve could add more socioeconomic
resilience to the agrosocioecosystem if it were used as a center for the creation and
dissemination of knowledge among local resource-limited farmers.
CHAPTER 1
INTRODUCTION
Introduction
In Ecuador, as in many other countries in the developing world, agriculture is
extremely important. It is important for food security, for foreign exchange, and as a
generator of employment. For the last half of the twentieth century, agricultural research
was concerned with raising production per unit area of major crops. Great strides were
made in this area. During this time, Ecuador managed to become the world's largest
banana exporter, a large exporter of shrimp, and the home of a flower industry that rivals
its neighbor to the north.
Advances were also made in field crops, especially maize, rice, and soybeans.
Local farmers are proud of these advances. "We are heroes of agriculture," stated a
maize and rice farm manager. He was referring to the fact that Ecuadorian producers
manage to keep on going in spite of the fact that they receive no subsidies. The global
environment in which agricultural production occurs today is not unknown to locals. "In
America and Europe, people get paid to be farmers!" Centuries of local agricultural
production knowledge, enhanced by 50 years of modern production research have
resulted in better yields, but large numbers of poor farmers.
The type of agriculture practiced in a large portion of Ecuador depends heavily on
hand labor. Yet today, people are leaving Ecuador in droves: landless peasants,
marginalized farmers, and professionals. Although from the outside, looking at macro
numbers, this mass emigration may seem positive in its overall effects on the economy -
remittances are now the second source of income for the country after petroleum-local
effects are strikingly negative. The rip in the family and social fabric, and upheaval of
traditional roles and practices that takes place when adult males and teens leave their
communities are difficult to mend.
In hindsight (a luxury we possess as researchers), there were perhaps three very
important oversights made in agricultural research in the second half of the 20th century.
These were, looking at components of agriculture as individual and not connected;
separating the biophysical features of agriculture from the socioeconomic milieu in which
it takes place; and decoupling production from natural resource management.
This is a study on people's interaction with the ecosystems that surround them. It
incorporates theories and methodologies from many different fields with the aim of
improving sustainability; food security and rural worker livelihoods in a 13,300 ha study
area. Agroecosystems design is used in an attempt to test if the introduction of slight,
adoptable changes regarding worker living arrangements, can improve resilience and
overall stability of the agrosocioecosystem within which they are immersed.
Human societies have always interacted with the ecosystems around them.
History is rife with examples of migration, conflict, and conquest resulting in part from
overuse of the resource bases on which human populations relied-fertile soil, lumber,
minerals, game-that had been depleted in original homelands.
Reasons for these changes are still being studied. Perhaps populations had mined
the soil, or maintained sustainable livelihoods over many generations, only to then crash.
They may have been resilient for long periods of time, and then internal or external forces
suddenly-or slowly and painfully-rendered what may have been excellent livelihood
systems unsustainable.
In some areas of the world, when population grew out of all reasonable
proportion, and global climate events negatively affected food production; upheaval,
famine and death on an unimaginable scale followed at fairly regular intervals. What had
once been highly sustainable systems became unsustainable. Shocks and stresses caused
whole nations to move into new sets of controls, to disappear, or to be conquered and
subjected by others. Often the new sets of controls were sought-and found-in a new
geographical location.
On planet Earth there is no place left to go, no new frontier, no promised land.
Sustainability must be sought in situ. Ecologists have long known that the struggle for
availability of natural resources is played out in nonhuman populations through
territoriality and competition. Habitat loss is unsustainable for most nonhuman
populations because they have a less ample capacity to learn, adapt, and manage.
Humans can learn and adapt; they have in the past, and they can now. Participation and
empowerment may be the keys for the potential of this new adaptive process. The human
species is typically resourceful and resilient. The advent ofglobalization, however,
introduces a whole new series of unforeseeable factors into the livelihood decision-
making process. Shocks are greater and more frequent than ever before.
All of the human populations just described were essentially rural throughout
their histories. Today, the world remains more rural than most realize. Modem
interventions, modem trends ofglobalization, economic inequity and indebtedness, and
worldwide climate change have altered many traditional systems or pushed them to the
edges of their geographical and spatial sustainability.
Increasingly, population growth drives the need for greater food production in the
world. This increase in food production must be sustainable. The prevailing system,
which uses enormous amounts of external inputs, must be gradually transformed into one
that uses land, water, and biodiversity more intelligently. This is even more critical on
marginal lands where much developing country agriculture takes place.
At the same time, rural communities everywhere have become more complex.
The problem of landlessness obliges us to include the plight of town-dwelling rural
workers in our search for solutions. There is an interface in modern rural landscapes
where the periurban and the rural overlap. Comprehension of the structure and function
of these multi-faceted communities at community, landscape, subregional, and regional
level, is needed. Finally, the complexity of these "agrosocioecosystems" requires a
multidisciplinary approach to rural development and natural resource management.
Preparing for worst-case scenarios is sound management. However, design and
management of agroecosystems can move beyond this task to preemptive design. With
this approach, it is possible to reduce the incidences of internal worst-case scenarios and
their associated risks. None of the adjustments in agroecosystems design and
management delineated in this dissertation are set in stone. In order for continuously
growing success, there must be a process of permanent adaptive management. This
process should include multiple stakeholders in a learning process that feeds back into the
system to adjust and improve the delineated alternatives, in response to changing internal
and external conditions. This could allow for adaptive management in the future.
It was my intention to examine a specific agrosocioecosystem for resilience by first
understanding, then analyzing it through modeling techniques. The type of resilience
analyzed was social and economic (socioeconomic resilience, hereafter). Responses to
internal and external shocks exerted on the several components of the system may point
out trends that can later be used as a basis for policy with implications at the national
level. According to the Consultative Group for International Agricultural Research
(CGIAR) because of the biophysical and socioeconomic complexity of agroecosystems,
experimentation and hypothesis testing must increasingly be based on results of systems
analysis and modeling work; and so must extrapolation of results for location-specific
observations (Bouma et al. 1995).
Background
The total Ecuadorian population in 1970 was 5,970,000. In that year the
agricultural population was 3,201,000, which represented 53.6% of the total population.
In the year 2000-the most recent year for which data are available-the total population
of Ecuador was 13,184,000. The agricultural population for that same year was
3,480,000 (FAO 2002). As a percentage of the total population, in a pattern common in
the developing world, the agricultural population has dropped. Yet, when taken in
absolute numbers, nearly 300,000 more people are involved in agriculture today than 30
years ago.
Because of a series of economic crises, there is an urgent need for alternative
activities for small-scale farmers to produce cash without compromising food security.
Problems in the countryside are manifold.
The agrarian census of 2002 showed that cacao, rice, maize, bean, pineapple,
passion fruit, plantain, tomato, and bell pepper production have declined at an alarming
rate. The same is true for livestock production, including cattle, swine, and poultry.
Lack of credit has not allowed small-scale farmers to prepare their land for the rainy
season planting of short-cycle row crops. Furthermore, because of the crisis the country
is going through, many young people have emigrated to Europe and the United States.
During the economic chaos of the late 1990s, utility prices (especially the price of
natural gas with which most Ecuadorian families cook) were left to fluctuate freely. This
caused a rise in the price of the subsistence basket for both rural and urban families. As
there was not a concomitant rise in minimum wages, many families found themselves in
desperate economic straights. The direct result of this was mass emigration. It is
estimated that nearly one million left the country in the years 1999-2000 (SIISE 2002);
another 430,000 emigrated in 2001 alone'.
Three key elements justify the need for this study. First, large numbers of small-
scale, resource-limited farmers still exist in Ecuador and around the world. Second, the
current state of the Ecuadorian economy including the effects of globalization and
dollarization requires urgent attention. Finally, when the social implications of massive
out migration, both from the countryside to misery belts around large cities and to
overseas are taken into account, the need for the present study becomes apparent.
Problematic Situation
Until the 1970s, as much as 50% of Ecuador's foreign currency earnings came
from petroleum alone. During this time, major infrastructure projects were undertaken,
using loans from multilateral development agencies. In the 1980s, many of the loans
came to term, and Ecuador began to feel the squeeze of a growing service on its foreign
SThis notion is substantiated in an article from the Guayaquilan newspaper El Universo:
Beltrfn, B., 2002. Emigrantes confiesan que el trabajo en Espafia "no paga."
16 May 2002. Guayaquil, Ecuador
debt. Inflation began to rise at levels that were out of control during the mid to late
1990s. At the same time, the International Monetary Fund was pressuring for
liberalization and real exchange rates in Ecuador and other countries. As a response,
government controls were put in place to attempt to control the spiraling inflation. After
these efforts proved unsuccessful, and in a move widely considered to be one of
desperation, outgoing president Mahuad announced his intention to dollarize the
economy. President Noboa obtained the legislative authority for this initiative, which
was implemented beginning in March 2000 (O'Grady 2002, US Department of State
2000).
Gustavo Noboa became Ecuador's fourth president in four years. Newspapers
reported two people were killed and dozens injured on February 5th, 2001 when
Ecuadorian security forces clashed with Indian demonstrators protesting high gasoline
and transport costs. The Italy-sized nation of 13.2 million people was struggling to
consolidate political stability a year after a similar Indian uprising, backed by military
personnel, toppled the then president Jamil Mahuad. The increase in fuel prices
galvanized some 5,000 indigenous people to travel to Quito in protest. The South
American country is struggling through an economic crisis that saw it default on its debt
in 1999 and post the highest inflation in Latin American three years running. Only about
25 percent of Ecuadorians who want to work have a full-time job. The country's foreign
debt is equal to about 80% of its gross domestic product 2.
The government abandoned the Sucre as the national currency and adopted the
American dollar as a solution to its chronic economic problems. For Ecuador, adopting
the dollar was a way of imposing strict fiscal discipline, in effect, turning monetary
2 Reuters 2001.
policy over to the United States Federal Reserve Board. Other nations are considering
the shift for similar reasons.
But the transition appears to have been complicated by the country's poverty.
One of every eight Ecuadorians is illiterate, according to government figures, and in the
past those people have relied on the different colors in which denominations of the Sucre
were printed as a guide to the value of the notes.
Global Context
In a World Food Day globally televised interview on October 16t, 2002, Michael
Lipton, director of the Poverty Research Unit at Sussex University in England, described
the situation as follows. Seventy five percent of the 1.2 billion people living on less than
one dollar a day are in rural areas, where the economy is based on agriculture. Nearly
seven percent of the world's rural poor live in Latin America.
Whatever funds are still available for rural research will not contribute
significantly to poverty alleviation and the reduction in inequity, at least in the short term,
unless the rural poor are deliberately targeted. Essential steps include the creation of
employment for the land-poor and landless, increased production on small, medium, and
large-sized farms, provision of nearby input and output markets, and, recognizing where
the rural poor are mostly located, attention to regions of lower agroclimatic and resource
potential, not just the best.
In his Food Day interview, Lipton further pointed out that
Thirty million people need to escape poverty every year but only 10
million are currently doing so. The rate of poverty reduction needs to be
six times faster to meet the 2015 deadline. Cutting world poverty in half
requires a new focus on reviving agricultural development and responding
to the needs of rural populations. A global effort is required to give the
rural poor better access to land, water, technology, capital, and more open
markets. Land reform and new policies to combat bias against women and
girls, who constitute the majority of the rural poor-and whose poverty is
often reinforced through cultural and legal obstacles-are needed (World
Food Day Conference 2002).
Gordon Conway agrees with Lipton, that the principal reason why research into
agroecosystems sustainability is undertaken is the quest to relieve hunger and poverty.
The elimination of poverty and hunger is a goal within the reach of scientists. Conway
warns of the dire consequences of inaction:
Unless the developing countries are helped to realize sufficient food, employment
and shelter for their growing populations or to gain the means to purchase the food
internationally, the political stability of the world will be further undermined
(Conway 1997, p.11).
Conway states furthermore, that if the proportion of the population of the
developing countries deprived of an adequate diet remains the same; the number
undernourished by the year 2020 could be greater than 1.4 billion. An alternative
scenario, which explicitly addresses this objection, is for the developing countries to
undertake an accelerated, broad-based growth, not only in food production, but in
agricultural and natural resource development, as part of a larger development process
aimed at meeting most of their own food-production needs, including the needs of the
poor. Agriculture and natural resource use are inextricably related. This has been
recognized in the document that emerged from the Johannesburg conference on
sustainable development. In the agenda, signed by 180 countries, every mention of the
word agriculture is linked to the phrase natural resources (Bakiel 2002).
A new revolution has to start with the socio-economic demands of poor
households. Its goal is the creation of food security and sustainable
livelihoods for the poor. Success will not be achieved either by applying
modern science and technology, on the one hand, or by implementing
economic and social reform on the other, but through a combination of
these that is innovative and imaginative. It will require a concerted effort
on the part of the world application of new scientific and technological
discoveries in a manner that is environmentally sensitive and above all the
creation of new partnerships between scientists and farmers that will
respond to the needs of the poor (Conway 1997, p.42).
Ecuador is an economically unstable country. Overall economic stability in
Ecuador, as elsewhere, depends on a multitude of factors. This study focuses on one such
factor, which may contribute partially to success in rural development, a study of
socioeconomic resilience, as measured by workers own perceptions and simulation model
outcomes; and economic production in an agrosocioecosystem located in northern Los
Rios Province. This is a work on rural development and natural resource management.
Holistic Approach to Sustainable Agriculture and Natural Resource Management
In searching to improve the livelihoods of rural workers in a dualistic agricultural
system, perhaps neither the existence nor the profits of the commodity export sector need
be compromised. Potential threats created by export-oriented commercial agriculture
such as deficient biodiversity, monoculture, excessive chemical inputs and soil and water
degradation may be attenuated by an increase in diverse small-scale farms within an area
dominated by banana monoculture. It is likely that a holistic method is needed to
measure existing conditions and probable changes. It is suggested that using an
ecosystems property and indicator, resilience, on the broader socioeconomic plane may
be a useful focus for addressing sustainable development. This dissertation will deal with
a landscape-level agroecosystem in the Western lowlands of Ecuador.
In the study area, we have both large-scale, commercial farms as in the more
developed world, and small-scale, family agriculture. We also have, associated with the
large plantations, livelihoods of thousands of hand-laborers mostly living in small towns
in the area, and a small nature reserve which comprises-at least symbolically-the
environment, without which agricultural systems cannot be analyzed properly.
Currently, in the area surrounding the Rio Palenque Science Center and Nature
Reserve, there is heavy dependence on a very few species (banana, oil palm, rubber),
limited varieties, and foreign biotechnology (for example, banana clones from Israel).
The situation is similar to that of Costa Rica as described by Nicolas Mateo of the
International Service for National Agricultural Research (ISNAR): ". .. agricultural
expansion has resulted in poor management of the natural resources in most of the
country and very low value-added, and has created a dangerous dependence on a small
number of crops (in Bonte-Freidheim and Sheridan 1997, p.l 13)." Because commercial
and small-scale producers and the forest itself are stakeholders and components of the
system, the processes that occur in the entire system must be viewed in a holistic manner.
The approach of this study is fundamentally systemic. It is expected that the
system can be analyzed as a whole after understanding the fundamentals of each one of
its major components. Each component, i.e. commercial banana agriculture, small-scale
farms, the plantation worker livelihood, and the Forest Reserve is itself a system.
Although no one component or system can be a priori characterized as outweighing
another in importance, our principal interest lies in the small-scale farm systems. This
approach has its background in systems theory.
The systems approach is possibly the most appropriate for tackling the level of
complexity inherent in human interactions with agroecosystems. This point was heavily
stressed at the 17th Symposium of the International Farming Systems Association
(Norman 2002, Pretty 2002, Tanic and Dixon 2002).
Because development cannot be decoupled from natural resource management,
large and small farmers must seek ways to address issues that pertain to the care and
future availability of the natural resource base on which they depend. "Those who live
beyond the borders of the world's industrial economy subsist on nature's surplus-on
organic soil fertility for food, on stable hydrological cycles for water, and on forests for
fuel. Environmental degradation, consequently, has direct, tangible results: hunger,
thirst, and fuel scarcity. No line can be drawn between economic development and
environmental protection" (Agrawal 1989, p.1). The success of those who were able to
take advantage of the green revolution is noteworthy. It is at the same time a paradox
that on the very edges of the hugely successful commercial banana plantations in Los
Rios Province, Ecuador, plantation workers live in squalor (to the outside observer's
eyes), and near subsistence level, small-scale farming continues.
A World that Revolves Around Bananas
There are roughly 204,000 banana workers in the Ecuadorian population. If one
multiplies this by three dependents per worker, it means that 600,000 depend directly on
bananas. If one includes those who depend indirectly on bananas (transportation,
cardboard manufacture, plastic, etc.), the number reaches 2,000,000. Therefore, bananas
have a huge multiplier effect in the economy. Many small towns actually make their
living from bananas. Workers purchase their basic goods there.
Globalization and dollarization, however, may threaten the system:
We have leveled salaries. We pay more than what the government says.
Many companies are barely paying what the government says. We have
been confused more than anything by the dollar. A bag of potatoes, which
used to cost S 5,000 or 50,000, suddenly is worth USD 2.00. There is no
conscience of the value of the dollar. We should not rest on our laurels
like Argentina. Competitiveness was lost in Argentina, the same could
happen to us. This is my greatest fear, that one day we will be as
expensive as Central America [bananas] and we will not be able to
compete with Central American countries. People here don't understand
that the dollar doesn't vary in the States. The workers expect a 20 to 30%
pay raise. Inflation is currently in the high teens, and the year will
I
13
I
probably close at about 14% inflation. You cannot raise salaries this high
I because you would put yourself out of business. (a banana company
executive 2001).
I Because banana production and small-scale farmer production depend heavily on
their natural resource base, it is necessary to address problems in the area in a holistic
3 way. A mere economic analysis will most likely leave out the sources and sinks that
Ecological economists have worked into broader analytical schemes (Costanza 1991,
Pearce et al. 1990, Van Kooten and Bulte 2000). The existence of a primary forest in the
I research area is not a driver of the environmental approach to the study (which is
3 fundamentally agroecological), but it does serve as a potent reminder of the enormous
changes humans can induce upon their surroundings, and it may serve as a mirror for
I reflection as we move beyond the simply descriptive question of "what is going on here?"
I to the more critical one of "where will things go from here?"
Los Rios Province
I Agriculture is the principal source of production in the Province. On the southern
I flatlands rice, sugar cane, maize, oil palm, passion fruit, and papayas are the principal
crops. On ground that is somewhat higher in the northern half of the province, export
I cash crops such as cacao, banana, and plantains prevail. The province is ranked number
I one in the country in banana, cacao, dry bean, soya bean, and maize production. Los
Rios is ranked second in rice production. Cacao, known as the "golden kernel," was the
principal Ecuadorian export item from colonial times until the First World War. A large
I proportion of the country's palm oil is produced here and it ranks second only to
I Esmeraldas Province in this area. Obviously, Los Rios is one of the most privileged
provinces in Ecuador for agricultural production.
I
I
Cattle production increased during the1990s. Both beef and (lately) dairy are
produced. Fish, an important component of the human diet, were always plentiful in the
past. They are usually caught in artisanal fashion using hook and line, nets and seines.
However, the use of illegal substances and dynamite for fishing has greatly affected local
fish populations limiting the local population's access to them as a source of protein
(Explored 1999).
At the time of Spanish encounter the area seemed to be densely populated. Tribes
included the Babahoyos, Babas, Palenques, Mangaches, Ojivas, Quilchas, and Pimochas.
The local TsAchila (Colorado) indigenous people have all but vanished. Currently, there
are around 2,000 persons claiming this ethnicity (Becker 2000). The Tsichilas'
disappearance was caused initially by disease, but later by amalgamation with other
Ecuadorians, both Indoamerican and those of European origin (Juncosa 1988).
Most of the population of Los Rios Province is young. The demographic group
from 15-29 years old is the largest sector. A little over 62% of the population lives in
rural areas. The well-known phenomenon of rural to urban migration is very high in this
province, although no exact figures are available. Many people are also employed in the
so-called "service sector," which in many documents is a euphemism for the informal
sector of the economy.
The Local Environment
The boundaries of western Ecuador have been established as the Pacific Ocean to
the West, the Colombian border to the North, the Peruvian border to the South, and the
900 m contour line on the Andean mountains to the East. With this definition, western
Ecuador has a land area of approximately 80,000 sq km, about 1/3 of the total 263,000 sq
km of continental Ecuador. Most of western Ecuador consists of a series of penneplains
extending westward from the base of the abruptly rising western range of the Andes.
There is also a low range of coastal hills seldom exceeding 800 m in elevation (Cabarle et
al. 1989).
The midcoastal region of Ecuador is located about 1.50 of latitude below the
equator and around 790 West longitude. The average altitude is 300 m above sea level,
with an average annual temperature of 24.5 C measured between the 1990 and 1995
period (NfiHez Torres 1998). The general lay of the land is flat changing to rolling hills
towards the East. The region exhibits a marked diversity in regard to agroecological
variables such as moisture, soils, light intensity, altitude, and terrain relief (Jones 1987).
Most soils are of volcanic origin with high organic matter content. Soil texture is sandy
loam and loamy sand, highly permeable and porous, with low capacity for water
retention. The porosity of the soil is favorable for crops in their early phases. However,
it is a disadvantage during growth periods when a greater level of water retention and
chemical activity is needed in the soil (Nfifez Torres 1998).
Since aboriginal times, agriculture has sustained large populations and much of
the region is under intensive agro industrial use at the present time. The soils of western
Ecuador are much richer than those of other tropical lowland areas. These rich soils
make conservation in western Ecuador very different from that in most tropical regions.
Instead of the common outcome of a few years of crops followed by abandonment, in
western Ecuador the result of forest conversion is often productive and sustainable
agriculture (Cabarle et al. 1989).
The earth's seasonal oscillation produces annual north and southward shifts in the
cold Humboldt Current and the warm Panama Current, which affect the climate of
western Ecuador. Beginning about 10 south of the Equator, the climate becomes
increasingly humid northward, varying from an annual rainfall of 2,000 mm, in the
Quevedo area, to nearly 9,000 mm near the Colombian border (Cabarle et al. 1989). The
climate in the area is warm humid and the mean annual temperature is 230C. The area
was formerly (before wide scale deforestation) classified as tropical humid forest.
A relatively drier season occurs from June through November (called "verano" or
summer in the local Spanish). The hotter, though wetter period from December through
May is called "invierno" or winter locally. Rainfall can also vary considerably from year
to year. The moisture regime is a critical variable affecting cropping practices. There are
two cropping cycles in the study area. Soil moisture (and sunlight) in this part of
Ecuador, especially during the summer dry season, is a function of cloud cover and
rainfall (Jones 1987).
The Agrosocioecosystem
Four components or sub-systems are described and the linkages among them are
quantified and qualified. The next are the four components that exist in the research area:
Large-scale commercial banana plantations.
Plantation worker households who live in towns.
Limited-resource or small-scale farms and farmers.
A small private nature reserve-the "Rio Palenque Science Center."
A limited number of livelihood options exist (Le6n and Vos 2000, SIISE 2002,
Valverde pers. comm. 2002) for the mainly unskilled and poorly educated inhabitants of
the study area:
Be a small-scale farmer with occasional off-farm work.
Be a salaried-plantation worker and live in a dormitory town.
Live and work on the plantation.
Migrate to a large city or overseas.
This creates one of the most important problems in the area. There is poverty and
little opportunity among plantation workers and small-scale farmers, which creates an
unstable social situation (ExploRed 1999, Le6n and Vos 2000, SIISE 2002, Valverde
pers. comm. 2002).
The area is overwhelmingly dominated by agriculture, at all scales and by the
ancillary industries and services that support it. It is clearly an agricultural system.
Relationships among owners and workers, worker and worker, town dweller and small
farmer, etc., make up a social network, albeit a rather loose one, nevertheless a
community, in the Wilkinsonian sense. It is also then a social system. Westley et al.
(2002, p. 107) describe a social system as follows: "any group of people who interact to
create a shared sense of understandings, norms, or routines to integrate action, and
establish patterns of dominance and resource allocation. Like any system, it is dynamic,
meaning it is difficult to change any one part of it without considerable effects on other
parts."
Furthermore, it is a place where cultivation, manufacture, trade, and salaries link
components together; more than a social system it is then a socioeconomic system.
Finally, all activities whether in one remote corer of Ecuador or anywhere else on the
planet are beholden to and ultimately limited by the natural resources the Earth can
provide.
It is a complex interplay of living organisms, human beings, and the habitat that
surrounds them. It is an ecosystem as well. Ecosystems are defined as places on earth
that consist of biotic components (life) and abiotic or physical components (Carpenter
1998 in Westley et al. 2002, p.105). Agriculture, socioeconomics, and ecology are
interwoven in this agrosocioecosystem (ASES). Hereafter, in this study, the term
agrosocioecosystem and the acronym ASES refers to the study area.
The disciplines that describe and analyze these activities i.e. agricultural sciences,
sociology, economics, and ecology, are the lenses through which the agrosocioecosystem
will be described and analyzed. The particular fields within these sciences that apply to
our research are mentioned below.
Researchable Problem
Some plantation workers live on the plantations, others live in small local towns
and still others live on small-scale farms. It is not known which of these living
arrangements provides more socioeconomic resilience and thus more stability in the long
run for the agrosocioecosystem and better livelihoods for rural workers. Additional
infrastructure and amenities required to reduce out migration are also unknown.
Research Area
An area considered representative of where a major Ecuadorian banana production
and export company operates several important plantations was defined. This area was
measured using a hand-held GPS unit in March 2002. Waypoints were taken at the
extremes of the area's limits and the resulting polygon (see figure 1 map below)
contained 13,308 ha. The principal infrastructure feature is the Quevedo-Santo Domingo
highway, which bisects the study area north to south. Located in the northeast corer of
this area is the Rio Palenque Science Center and Nature Reserve. Because the research
was spawned by the interest of a local foundation to assess what benefits could be
extended from the forest reserve to nearby farmers, it was taken as a point around which
I
19
the study area would be defined. The topography immediately east and north of the
I reserve varies greatly as the foothills of the Andes begin there. These small hills are
Known as Sierra del Ila on local maps. This is why the study area fans out towards the
West and Southwest where topography similar to that of the station occurs, and waters
run towards the Palenque (Quevedo) River, as part of the Guayas Basin.
SMost jobs in the study area are on banana plantations. There is also some work
available on palm oil, rubber, aback (Musa textilis, grown for Manila hemp fiber), and
lately pineapple plantations. There are a very few agroindustries such as palm oil
B factories and a meat packing plant, which belongs to a supermarket chain, located near
I Patricia Pilar. Most everything, however, is related to agricultural production in one way
or another. Bananas are big but palm has also been around for more than 40 years.
SAccording to data from the Municipality of Buena F6, the research area contains
389 small-scale farms under 10 ha in size (mean = 3ha). Also located in the area are 513
medium-size farms between 11 and 99 ha in size (mean=12 ha), and 49 large properties
over 100 ha in size (mean= 10). The Parish, (an administrative unit below the canton or
county level) of Patricia Pilar has a population of 6,241 (SIISE 2002). The town of
Patricia Pilar proper holds around 4,500 inhabitants (Buena F6 Municipal Records).
SOther rural settlements in the area include the "recintos" (hamlets) of Los Angeles (pop.
S250), Valdez (pop. 100), and Fumisa (pop. 200), and several crossroads of 20-30 persons
each. Total farms in the area are 951. In summary, the population of the study area,
although very mobile, is roughly 5,000 in small towns and hamlets; 3,000 on small and
) medium farms; and 650 workers living on plantations or haciendas. The total population
of the study area is approximately 1,800 households, or 9,000 people. Of these, some
I
3,000, or nearly 32%, are banana plantation workers. Another 2,000 banana plantation
workers do not reside in the study area.
F Guayaql
Figure 1-1. Map of Ecuador and study area
The area includes five major banana plantations owned by the same company:
NG (483 ha), MC (116 ha), V (81 ha), SV (135 ha), and Z (548 ha). Also belonging to
the same agribusiness group are three other properties: ZC, a cattle ranch (approximately
600 ha), MA, a macadamia, rubber, and ornamentals plantation (80 ha), and the Rio
Palenque Science Center (140 ha), of which 90 ha is primary forest. Aside from this,
there are 19 banana plantations that are considered associated producers who are "loyal"
to the company but are individually owned, totaling nearly 1,500 ha. In summary, 16.4%
or 2,183 ha of the land in the study area are owned and operated by the above-mentioned
banana agribusiness group. Another 11.27% or 1,500 ha of banana plantations have
strong ties to the same commercial group. Together they make up 27.67% of the total of
13,008 ha. One company then directly or indirectly controls over /4of the region,
making their influence impossible to ignore in any serious study of the area.
Approximately 17% of the area is made up of small-scale or limited-resource
farms and nearly 46% is composed of medium-size farms. Urban areas including towns,
hamlets, and many populated crossroads make up less than 3% of the area. There are
also some 300 ha of protected areas, all on the banana company's lands, in the area.
These data are based on tax records and artificial divisions of farm size created by the
municipality of Buena F6 for tax purposes. Of the medium and large farms 40% are
banana, 40% palm and others include rubber, cattle, palm heart, malanga, pineapple,
abaca (manila hemp) and others.
Objectives
Study objectives were to describe, understand, analyze, and test alternatives
scenarios that might provide the greatest well-being for all humans at whatever level they
may be inserted into the system. Fundamentally, this means searching for the highest
possible quality of life for plantation workers and limited-resource farmers, without
compromising the profitability and growth of the banana export industry, of such vital
importance to the national economy of Ecuador.
Within this overall objective, specific objectives were to:
Assess resilience vis-d-vis economic crises, drought or flood, and sudden
household composition changes of limited-resource households, and town-
dwelling plantation worker households in the study area,
Determine the effects of several living arrangements on overall socioeconomic
resilience in the research area,
Address the issue of food security for both types of rural workers,
Assess the need for infrastructure or amenities needed to make rural living more
attractive, and
Assess the role of a science center and forest reserve as a component of an
agrosocioecosystem.
Hypotheses
Hypothesis 1.
Small-scale farm households are better able to survive stress and shocks than town-
dwelling plantation worker households. They are more socioeconomically resilient.
Hypothesis 2.
Transferring plantation workers living in town to small-scale farms would increase
overall economic output of the system and add to socioeconomic resilience in the study
area.
Hypothesis 3.
If redesigned rural worker living arrangements are to be successful and enhance
socioeconomic stability in the area, amenities and infrastructure, such as accessible
schools and clinics must be provided.
Hypothesis 4.
If a local science center and forest reserve is used to create and disseminate knowledge
to small-scale farmers in the research area over other possible uses, overall economic
output and socioeconomic resilience of the system will be increased.
Hypothesis 1 was assessed using a stated-preference, or perceptions survey among
workers. Evidence from this survey was reinforced by ethnographic and participatory
linear programming models. Hypothesis 2 was explored using a large matrix linear
program containing all four components (themselves systems) of the agrosocioecosystem.
Hypothesis 3 was explored using statistical analysis of a perceptions survey conducted
with rural workers in the study area. As the hypothesis states, simply rearranging the
place where rural workers live cannot enhance socioeconomic sustainability if the
principal stakeholders do not perceive available infrastructure and amenities as adequate.
Finally, hypothesis 4 was explored using the same large matrix LP described for
hypothesis 2.
Methods
The next questions guided the choice of methodologies employed for data
gathering and analysis:
What are the components of the system?
How do these components interact?
Is the system and each one of its components resilient?
How do people survive?
Who is better off(workers living in towns or small-scale farmers)?
Is there equitable distribution? and finally,
Who or what affects or is affected by the forest and certification?
In this work, the unit of study is referred to as an agrosocioecosystem. Some
elements of agroecosystem analysis (AEA) (Conway 1990, p.224) were used. AEA
draws on the concepts of agroecosystems, agroecosystem hierarchies, agroecosystem
properties and their trade-offs as a basis for analysis. Mathematical simulations and
statistical analysis will reinforce these analyses.
In order to address these questions a review of secondary sources was used, also a
modified Sondeo, interviews, local knowledge surveys, farming systems interviews, a
perceptions survey, ethnographic linear programming, participatory linear programming,
and statistical analysis. Further detail regarding methodologies used with plantation
workers can be found in Chapter 4, and those used with small-scale farmers or more fully
detailed in Chapter 5. Interviews were also conducted throughout the research period
extending from November 2000 to March 2002 with upper management of an important
fruit company, plantation field managers, town dwellers, the Rio Palenque Science
Center's biologists and managers, directors of a local foundation and personnel, parents
and teachers, and many others. The Sondeo, interviews and surveys, fundamental
Farming Systems methodologies, were invaluable for obtaining a great deal of
information to understand socioeconomic diversity and system structure and function
within a reasonable time frame.
The principal analysis tool used was the ethnographic LP model. These models
link ethnographic information to a quantitative analysis tool that allows multiple
objectives. The strength of the ELP is that it can incorporate demographic,
socioeconomic, ecological, climatic, production, and other data in one model. These
models use information gathered directly from producers and workers using participatory
I-- 93 -km
D
D
BaunaaPlantations
Other Plantations
Sciene Center and Natue Reserve
Smal-scale farm areas
Town
HamletK~rossroads
Figure 1-2. Schematic of the study area
methods. The models were calibrated with farmers, which is invaluable to understanding
the system. These models, which are themselves hypotheses, can be used to explore
possible shocks, to look at the mechanics of linkages, and to appraise new technologies.
The models provide acceptable conjectures if well constructed. Different outcomes under
several scenarios allow us to examine hypotheses. They are a rapid, low cost, effective
tool for ex ante prediction and hypothesis testing. Another strength of these models is
N
+
185 km
26 I
that they can be scaled up, that is, aggregated to the community or landscape level, and
thus simulate an entire agrosocioecosystem for predictive purposes. I
Analysis
Data gathered through the various procedures described, i.e. informal interviews,
surveys, secondary data, were used as inputs for linear program models. The limited- I
resource farm household was modeled and compared to a similar model of a plantation
worker household living in town. Both received similar shocks. Their resilience to these
shocks was contrasted to evaluate the first hypothesis. Additional similar, but not
identical scenarios were also run, and the results reported. Scenarios are described in
Chapter 7. A multi-year model was built encompassing all four components of the
system in order to explore the second hypothesis. Scenarios are described in Chapter 7.
A survey of worker's perceptions', with 82 samples was statistically analyzed both to test
the third hypothesis, and to add balance to the possible bias surrounding the banana
plantation system derived from extensive use of corporate interviews to describe this I
component of the agrosocioecosystem. These results and a discussion on them are given
in Chapter 4.
Organization of the Study I
The first chapter is an introduction to the background and problematic situation in
which the study is set. This chapter also deals with the objectives and hypotheses of the
dissertation, a general description of the study region and a more detailed definition of 3
the exact study area. The methodology for description and analysis is briefly described.
Finally, the justification for the study and its potential significance are discussed.
Chapter 2 is a literature review and presentation of the theoretical underpinnings of this
I
I
study. The next four chapters deal with each one of the components of the
agrosocioecosystem were defined in Chapter 1.
Chapter 3 is concerned with the export banana production system. It is described
and analyzed to understand internal dynamics, and how external forces, especially
changing global markets are affecting it. Chapter 4 refers to the workers who earn their
living on the plantations. After describing both the conditions on the plantations and
conditions in the small towns where most workers live, the results of a perceptions survey
conducted with these workers are shown. The aim of this was to capture both an emic
and an etic (Bernard 1995, p.238) side to our research. That is, one thing is what we have
learned about the workers from secondary sources including those who run the banana
company, and another may be the reality perceived by the workers.
Chapter 5 addresses the issue of the small-scale farmer component of the
agrosocioecosystem. The farming systems are first described and then analyzed using
ethnographic linear programming (ELP). The linkages that exist within this system are
explored and attention is given to the resilience of small-scale farm households to
economic, climatic, and social shocks. The economic shocks deal with inflation,
commodity prices and off-farm wages. The climatic disturbances refer to excessive rain
or droughts caused by El Nifio Southern Oscillation (ENSO) related phenomena. Social
shocks refer to sudden changes in household composition, such as those caused by
emigration or death (as opposed to natural family life cycles).
Chapter 6 deals with environmental certification and the Rio Palenque Science
Center and Nature Reserve, and natural areas on banana haciendas are described with
several objectives in mind. These include having a baseline knowledge of what the
28 i
natural ecosystem was like in the study area before the onset of modem agriculture; to I
understand the role of private conservation and development efforts; to seek linkages
between the existence of the forest and the availability of processes that enable improved
marketing of bananas; and finally, to look at the forest and its potential role as a site for
the creation of knowledge. I
In Chapter 7, the components that were dealt with individually in Chapters 3
through 6 are brought together for an overall analysis. In this chapter, data are analyzed
to test the several hypotheses. Linear programming allows us to understand the I
workings, dynamics, and complex mechanics of the system.
The final chapter summarizes findings from the study. Based upon these I
conclusions, it goes on to delineate potential changes based upon the above-mentioned I
search for economic and human development coupled with sustainability.
Significance
It is expected that the present study can contribute to our understanding of the I
biophysical and socioeconomic interface of the agrosocioecosystem. Ecuador is a
"wealthy country with poor people" (V. Bastidas pers. comm. 2002) (GDP = 19.3 billion;
population nearly 13 million). Nowhere is this more in evidence than in Los Rios I
Province. Nevertheless, with crisis may come opportunity. Useful information may
emerge from this study, which may serve as a reference for those working in agricultural
development and social policy. The highlighting of important inequities existent may aid
in fostering projects and programs for the area. Significant efforts of private corporations
and foundations in the areas of education, health care, and conservation, which currently
exist in the study area, may serve as important examples to others. I
I
I
On the other hand, local agricultural producers, large and small, may use data
from this work to improve their own livelihoods and those of others in their communities.
The weakness of formal institutions with regard to technology innovation and transfer,
credit and other basic issues, which are pointed out in the present work, may serve as a
stimulus for self-help and collaborative effort. The study may also allow farmers and
natural resource managers to reflect on their own roles in the future of the study area.
This work may be useful in shedding light on the current situation. It may aid in
delineating potential paths that may be followed in the search for improved resilience,
thus allowing for sustainable development and improved rural livelihoods in the area.
For Ecuador, the University of Florida and researchers elsewhere, it is hoped that
the work will provide not only a further step in agroecological and sustainability research,
but also a wider recognition of the need to find the intersection of the biophysical and the
socioeconomic in human livelihood systems. This dissertation seeks to create greater
awareness of the local human's role as stakeholders, but also as decision makers. It is
hoped that this may empower them towards a brighter future through their own efforts in
development and conservation.
CHAPTER 2
LITERATURE REVIEW: THEORETICAL UNDERPINNINGS OF THE STUDY
Compiling the literature for an interdisciplinary study such as this is a daunting
endeavor. Disciplinary, and often reductionist research, can often count on decades of
related, annotated work built up through a process of accretion. Work is available as a
single body of study in the form of condensed reviews and a handful of specific journals.
By contrast, studies that are interdisciplinary in nature call for a wide array of referential
material, commonly dispersed and often at first glance, unrelated. In order to convey the
double purpose of reviewing previous related research and assuring its inclusion within
the theoretical underpinnings upon which a new study is built, several areas of scientific
thought must be embraced, and then their intersection must be sought and explained.
This weaving of multiple fields into a common web that supports and helps set the tone
for a study that seeks to achieve the goals of ecological, economic, and social
improvement of a defined system and its principal subsets will be attempted in this
chapter.
The next areas are discussed in turn. Sustainability is described as a general
concept and resilience as a property and indicator of sustainability. A history and
relevant concepts of systems theory and the systems approach to scientific research is
given, including relevant research on banana dominated agroecosystems.
A brief outline of linear programming as a modeling technique and its application
to agriculture, livelihood systems, and sustainability research is also approached. As this
study entails many aspects of rural development and natural resource management, a
broad outline of several schools of economic development thought, along with current
thinking and pertinent studies follow. Finally, the ecological institutional aspects of
deforestation, private nature parks, and ecological or "green certification" are reviewed.
Sustainability
Sustainable agriculture is a relatively recent response to the decline in the quality
of the natural resource base with modem agriculture (Kelly 1995, Mclsaac and Edwards,
cited in Altieri 2001a, Pretty 2002). Discussion of agricultural production has evolved
from a purely technical one to a more complex one characterized by social, cultural,
political and economic dimensions. The concept of sustainability is controversial and
diffuse partly because of conflicting definitions and interpretations of its meaning.
However, it is a useful concept as it addresses concerns about agriculture, which
are the result of the coevolution of socioeconomic and natural systems (Jordan and
Hutcheon 1995, Reijntjes et al., cited in Altieri 200 b). A wider understanding of the
agricultural context requires the study of agriculture, the global environment, and social
systems, given that agricultural development results from the complex interaction of a
multitude of factors. It is through this deeper understanding of the ecology of agricultural
systems that doors will open to new management options more in tune with the objectives
of a truly sustainable agriculture (Altieri 2001a, Shiyomi and Koizumi 2001).
A general agroecological principle (which is widely applicable except, perhaps in
the case of wetland rice production) states: the greater the structural and functional
similarity of an agroecosystem to the natural ecosystems in its biogeographic region, the
greater the likelihood that the agroecosystems will be sustainable (Gliessman 1998,
Ruthenberg 1980). Agroecologists are now recognizing that intercropping, agroforestry,
and other diversification methods mimic natural ecological processes, and that the
sustainability of complex agroecosystems lies in the ecological models they follow. By
designing farming systems that mimic nature, optimal use can be made of sunlight, soil
nutrients, and rainfall (Pretty 1994).
With this, we are not suggesting that banana monoculture is unsustainable and
should therefore be eliminated. Indeed, the swidden type of banana production prevalent
before the introduction of the Cavendish variety was sustainable as long as there was
abundant forestland for shifting to new tracts of land, while old banana areas were left in
fallow to be "cleansed" of the disease. There is a good chance of conserving fertility
when perennial crops are grown. The frequency and intensity of cultivation is less than
in arable farming. Many perennial crops shade the soil and permit or require a permanent
cover of grass or leguminous vegetation. Some tree and shrub crops influence the soil in
the same way as the forest, and moreover permanent planting encourages the construction
of terraces, the control of watercourses, and other permanent land improvements. The
destruction of fertility in the cultivation of some perennial crops, like coffee in southern
Brazil or tea in Sri Lanka is very serious, but the fact remains that the damage is usually
less than in arable farming under similar conditions (Ruthenberg 1980).
High input Cavendish variety banana production however, may be less
sustainable due to a plethora of factors. These include low biodiversity, low nutrient
recycling, and excessive reliance on irrigation, low biological interactions-especially of
beneficial organisms-and low synergism among components. These lead to a
breakdown of key agroecological processes and services (Altieri 2001b).
It is not necessary (nor necessarily desirable) for the sustainability of a system, to
insist on the sustainability of all the system's components, because of substitution
I
I 33
possibilities among resources (Harrington 1992). This ignores, however, the normative
I question of whether it is desirable to have unsustainable subsystems within a larger
sustainable system. This is particularly an issue in a situation where an agricultural
system may be sustainable, but individual farms within it are not. While systems
sustainability is affected by the sustainability of lower system's levels, it is not
Necessarily dependent upon their sustainability. In other words, a system will not likely
i be sustained within a larger unsustainable system, though it can be sustainable even if
one or more of its components is not (Kelly 1995). A hypothetical illustration of this
point says that a resource-limited farm is not sustainable within a larger commercial
I plantation system that is unsustainable. Conversely, the commercial plantation system
may be sustainable even if the small-scale farms within it are unsustainable.
I Resilience: A Concept, an Indicator, and an Emergent Property
I Resilience is a key property of sustainability (Folke at al. 1998, Holling in Van
Kooten and Bulte 2000). Ecological resilience has been defined as the magnitude of
I disturbance that can be experienced before a system moves into a different state and a
I different set of controls (Holling 1973, 1986). Social resilience has been defined as the
ability of human communities to withstand external shocks to their social infrastructure,
I such as environmental variability or social, economic, and political upheaval (Adger
3 2000, Conway and Chambers 1992).
The sustainability we refer to in this work is intimately linked to the term
I development (Altieri 1990, 1994, 1995, Conway 1987, Spedding 1975). Both diversity
I and disturbance are parts of sustainable development and resilience. The interaction
between diversity and disturbance needs to be explicitly accounted for in an increasingly
I globalized and human dominated biosphere (Folke et al. 1998). Change and crisis are
I
34
part of the dynamic development of complex coevolving social-ecological systems
(Gunderson 1999). In keeping with Vandermeer and Perfecto 1995 (in Boucher 1999) 3
we engage the entire spectrum of disturbance, resilience, and sustainable development by
Beginning the political process of reorganizing socioeconomic-ecological U
systems by examining questions of food security'... as a mode of analysis,
examining food insecurity will cause us to deal with the entire complex
web of ecological, sociological, economic and political issues (1999,
p.96).
Thus, one of our principal theses is that small-scale polycultural agriculture may I
be an asset for sustainable development because those who engage in this activity may be
more resilient and confer some of this property to other levels of the systems hierarchy.
Shocks and stresses are emphasized to differentiate from the normal small disturbing
forces such as fluctuations in cycles in the surrounding environment (including physical,
biological, social, and economic variables that lie outside the agroecosystem under
consideration). Shocks can be external, that is, issues that are beyond the farmers'
control or exogenous. Internal shocks are directly associated with farming system I
operations and decision-making. These questions may or may not be associated with
externalities (Harrington 1992). The constancy of production within normal fluctuations
is known as stability, which is different from resilience (Conway 1985). 1
There are several foci in available definitions of sustainability. An I
environmentally sustainable agroecosystem maintains the resource base upon which it
1A nation is food-secure only if each and every one of its inhabitants is food-secure, that is, has access at all times to
the food required to lead a healthy and productive life. To achieve this, each individual or, in practice, each household
must grow sufficient food or be able to purchase the food from income earned either through selling agricultural
products or by engaging in agricultural or non-agricultural employment For urban dwellers, the only option is to
engage in non-agricultural employment, but for the vast numbers of rural poor, if they are not growing enough food to
meet their needs, they must have the means to purchase the food they require. For them, food security depends as
much on employment and incomes as it does on food. Production, and agricultural and natural-resource development
are crucial in both respects.
(Conway 1997, p.66).
I
depends, relies on a minimum of artificial inputs, manages pests and diseases through
internal regulating mechanisms, is able to recover from disturbances caused by
cultivation and harvest, and is productive over a long period of time without degrading its
resource base-either locally-or elsewhere (Gliessman 1998). Furthermore,
environmental sustainability itself seeks to improve human welfare by protecting the
sources of raw materials used for human needs and ensuring that the sinks for human
waste are not exceeded in order to prevent harm to humans.
Social sustainability involves elements including cohesion of community, cultural
identity, diversity, commonly accepted standards of honesty, laws, and others that
constitute the aspects of social capital least subject to rigorous management but essential
for social stability. Human and social capital, investment in education, health and
nutrition of individuals are now accepted as part of economic development, but the
creation and maintenance of social capital is not yet adequately recognized. It refers
especially to the ability to recover after shocks or stress (Conway and Chambers 1992,
Goodland and Pimentel in Shiyomi and Koizumi 2001, Paoletti 2001).
Another pertinent realm of sustainability is the economic one. The widely
accepted definition of economic sustainability is maintenance of capital. The amount
consumed in a period must maintain the capital intact because only the interest must be
consumed. Economic values are restricted to money; valuing natural intergenerational
capital, such as soil, water, air, biodiversity is problematic (Goodland and Pimentel 1998,
Paoletti 2001, Van Kooten and Bulte 2000).
Two salient commonalities regarding sustainability can be highlighted. First,
systems where sustainable development is to take place are typically diverse and
complex, that is, as interconnected as possible. Second, these systems and especially the
humans that affect and are affected by them must be able to recover from major changes
and reorganize themselves into new states of equilibrium2. We have chosen to examine
the system at hand with regard to the property of resilience. Resilience, if measured
correctly, may be one of the best measures of sustainability3.
To put it succinctly:
Human communities can be understood as comprising individuals, a
population, communities, and coalitions of communities that work
together or oppose each other on the basis of interest values. Like
agroecosystems, communities of place and of interest have emergent
qualities and are more than the sum of their parts. The interactions
between population and communities of interest and of place can enhance
or detract from sustainability. Sustainability is an emergent property of
the interactions between communities of interest and of place that includes
a healthy ecosystem, vital economies, and social equity (Flora 2001, p.1).
One of the bases on which this research builds is that resilience (to a point) can be
built into farm systems for sustainability. Sustainable agricultural systems will therefore
display the characteristics of a resilient system. The goal is, at least initially, to
understand which features will be more conducive to building resilience at the farm level
and thereby contribute to sustainable agriculture and livelihoods (Carpenter and
Gunderson 2001, Folke et al. 1998, Milestad et al. 2002).
Berkes and Folke (1998) hypothesized that successful resource management
systems will allow disturbances to enter on a scale that does not disrupt the structure and
2 A comprehensive review on the several definitions of sustainability, stability and diversity is given in
Wratten and Van Emden, in Ecology and Integrated Farming Systems. Also see Glen et al. 1995.
3 For a well-organized account of difficulties involved in measurement of sustainability in agricultural
systems, see L.W. Harrington. Measuring Sustainability. Journal of Farming Systems Research-Extension
(3) 1, November 1992, pp.1-18.
functional performance of the ecosystem and the services it provides. This capacity to
absorb and adapt to change in an active way includes these aspects:
Understanding cycles of natural and unpredictable events (Roling and Jiggins
1998);
Diverse and flexible on-farm and off-farm activities to stabilize the farm system
(Ellis 2000); and
Stewardship and socioecological management (Folke et al. 1998, cited in
Milestad et al. 2002).
Another useful element that may contribute to farm resilience is the degree of
self- organization between farms and with the "outside" world. This includes a limit to
dependence on external institutions for information, and a decrease in the level of
dependence on external inputs, rather relying on internal nutrient cycles (Capra 1996,
Pretty 1998). Finally, a component of resilience that reflects a learning aspect of system
behavior in response to disturbance is adaptive capacity. Feedback mechanisms, which
allow resource managers to receive signals, process and interpret them, and respond with
adequate changes in their management practices, are key (Berkes and Folke 1998,
Gunderson et al. 1995).
It is a basic tenet of systems theory that systems comprise subsystems that interact.
For systems directed by human, control is attempted by managing these sub-systems and
the interactions between them in an unpredictable environment (Dent and Anderson
1971, Tivy and O'Hare 1981). In this study, resilience of each component of the
system-which is itself a system-was analyzed, followed by analysis of the entire
system. Systems were examined by introducing shocks into linear program (LP) models
and quantifying the outcomes.
This study used an ethnographic linear programming (ELP) (Breuer 2000, Kaya et
al. 2000, Mudhara 2002, Thangata 2002) approach to better understand resilience in an
Ecuadorian agrosocioecosystem. The research is multidisciplinary in focus and
organization. Although theories, principles, and case studies from many different areas
are used as references and as a foundation upon which the current study is constructed,
the principal pillars are these:
Farming Systems Research, which is more thoroughly discussed in Chapter 4
(Collinson 2001, Hart 1980, 1981, Hildebrand 1986, Ruthenberg 1980).
Linear Programming (Cabrera 2000, Doyle 1990, Hazell and Norton, 1986, Kaya
et al. 2000, Mudhara 2002, Spedding et al. 1981).
Rural to Urban Migration, especially the argument that urban unemployment
might best be addressed by reducing the incentives to migrate to cities, by raising
rural incomes via a broad range of development programs (Harris and Todaro
1970, Hazell and Haggblade 1990, Todaro 1969).
Agroecosystems and Sustainability (Altieri 1987, Berkes and Folke 1998,
Conway 1985, Flora 2001, Gliessman 1998, Holling and Gunderson 2002).
Banana-Dominated Agroecosystems (Soluri 2001, Striffler 2002, Vandermeer
and Perfecto 1999).
Systems Approach
A system is an arrangement of components that function as a unit. Biological and
physical systems are open systems that interact with their environments, processing
inputs to produce outputs. Smuts pioneered the systems approach in biology with his
introduction of the concept of Holism in 1926. In the early 1930's, von Bertalanffy
formulated what he defined as a General Systems Theory (Hart 1980).
The systems approach has been applied to all biological disciplines, but is probably
most associated with ecology. In 1935, Tansley proposed the term Ecosystem. The
concept has been developed by many others, such as in the classic papers on trophic
levels by Lendeman and energy flow through ecosystems by H.T. Odum. Development
of the ecosystem concept into a larger ecological systems concept is probably most
associated with E.P. Odum (1971) and his Fundamentals of Ecology text and the energy
circuit approach of H.T. Odum. The former defines an ecosystem as follows:
An ecosystem is "any unit that includes all the organisms in a given area
interacting with the physical environment so that a flow of energy leads to
clearly defined trophic structure, biotic diversity, and material cycles
within the system." The flow of energy and cycling of materials
associated with ecosystems can be found in other ecological systems both
larger and smaller than ecosystems. In systems terminology, ecosystems
are subsystems of other systems as well as composed of subsystems. The
conceptual framework of ecology is based on the assumption that there
exists a series of hierarchically interacting systems from the universe to
the smallest subatomic particle (E.P. Odum 1971, cited in Hart, p.3).
Ecosystems, communities, and populations are probably the most common units
studied in ecology. Each hierarchical level is conceptualized as a system composed of a
set of subsystems. Interactions between two subsystems of the same system can be
defined as horizontal system integration. Horizontal system interaction can be
superimposed upon the vertical system interaction. This vertical and horizontal
ecological systems model can also be applied to the agricultural production process.
Since sustainable agriculture encompasses environmental, productive, economic,
and social concerns, its analysis is best served by a systems approach that accounts for
component interactions, and that can trace the consequences of an intervention through
the entire system. While in high input farming, interventions may disregard some
important system interactions and still obtain high yields; such disregard is not possible
in "sustainable" agriculture where these interactions are key to decision making. A
strong systems perspective is a distinguishing feature of sustainable agriculture (Carroll
et al. 1990, Conway 1985, Edwards 1995).
Potentially, the systems approach, in which the biological, economic, and social
aspects of a problem are examined in an integrated way, is very relevant to product and
resource decisions (Doyle 1990). Furthermore, an argument can be made for the
challenges posed by the interactions between agriculture and human social systems and
institutions. Spedding (1979) pointed out that agricultural systems lie at the intersection
of economics, the social sciences, and biology. He approached the complexity by
defining agroecosystems strictly in terms of their "purpose" and then defining
boundaries. However, a case can also be made, in light of our increased understanding of
environmental and social systems, for defining systems not only in terms of human
purpose but also by their unique structure and dynamics (Conway in Jones and Street
1990). In defining the agroecosystem under study, we follow both authors.
The nub of the systems approach is a belief that the whole is more than the
sum of its parts. This implies that an isolated study of the components that
make up the system is inadequate to understand the complete system.
This is because the separate parts are linked in an interacting manner and
it is the interactions and inter-relationships between the various
components that give the system its identity and organizational integrity
Thus, systems theory is primarily concerned with the systematic study of
interactions between different factors that make it up (Dent and Anderson
1971, Doyle 1990, Kiker 1999, Roundtree 1977, Spedding 1979).
In her work on adult education and systems thinking, Melinda Kiker (1999) traces
the origin of systems thinking to dissatisfaction among scientists in the 1950s regarding
the reductionist analytical process pervasive in research up to that moment. The systems
approach was a reaction against the simplicity of the self-contained laboratory
experiment because of its apparent failure to adequately address the problems of an
increasingly complex world. The author describes systems thinking, within which, a
system is a complex and highly interlinked network of parts exhibiting synergistic
properties-the whole is greater than the sum of its parts. Systems possess these basic
characteristics: a purpose; boundaries; emergent properties (properties as a whole entity
which cannot be deduced from their constituent parts); exist in nested hierarchies; have
internal processes of communication and control; go through transformation processes;
are to some extent able to adapt and survive; and are usually open to the environment
(Kiker 1999). Systems can be as different as living organisms, cosmological clusters,
corporate production and sales, agricultural and ecological systems. It is these last two
systems that concern us.
The agroecosystems concept involves multiple levels of organization. These are an
individual, an organism, a population (groups of similar organisms), communities
(groups of different organisms), and an ecosystem, which links the abiotic factors of an
environment and the communities of populations and organisms that occur in a specific
area (Gliessman 1998).
Agricultural and Agroecological Systems
Agricultural systems exhibit not only vertical hierarchical system interaction, but
also horizontal system interaction. Each hierarchical level is a functioning set of
subsystems with the outputs of some subsystems acting as inputs to others. While it is
possible to describe a global level agricultural system, from the point of view of
agricultural research and development, the geographic region is probably the largest unit
of interest.
A regional agricultural system includes all the farms in the geographic region; the
marketing, credit and information center; and the infrastructure that ties these regional
subsystems together. A region can be analyzed as a system with materials, energy,
money and information flowing into and out of the region and between subsystems
within the region. From an agricultural research point of view the farms within the
region are the most important subsystems and form the next lower hierarchical level
under the region.
A farm is also a system made up of subsystems. A farm system can be viewed
conceptually as a set of spatially definable areas in which crops, animal or both are
produced, and a homestead area where the farmhouse is located. The crop or animal
production areas form units, analogous to the ecosystem unit in ecology, and can be
defined as agroecosystems. The farmhouse area in which the farm family is fed and
clothed and the economic transactions and management decisions that occur on a farm
can be combined to form a socioeconomic subsystem of the farm system. The
socioeconomic subsystem and the agroecosystems interact to form a farm system
(Conway 1985, Hart 1980).
An agroecosystem is also a system made up of subsystems. As in the case of
natural ecosystems, an agroecosystem is composed of a biotic community of plants,
animals and microorganisms, and the physical environment in which the community
functions. Energy flows between trophic levels and materials are cycled. An
agroecosystem differs from a natural ecosystem in that at least one plant or animal
population is of agricultural value and that man plays an important management role.
Soil, crops, weeds, insects, and microorganisms can be defined as subsystems of crop-
dominated agroecosystems. Agronomic research has been done on all of these
subsystems, but crop systems and animal systems have received the most attention.
It is not always necessary or practical to use the entire hierarchy. Emphasis can
be placed at one level, as for example in the case of a cropping systems project. In
principal, however, it will always be necessary to study at least three levels: the unit of
interest, and the next higher and next lower levels. The next higher system must be
studied in order to measure the inputs into the system, and the next lower level must be
studied in order to understand how the system functions (Hart 1980).
Agroecosystems are, just as any other system, defined in terms of their
biophysical boundaries and flows. Furthermore, they are the result of human decisions
that derive from human goals. These are determined by the dynamics of human social
and economic cooperation and competition as embodied in a variety of human
institutions. The resulting system is thus as much a socioeconomic system as it is an
ecological system, and has both biophysical and socioeconomic boundaries.
Four properties have been defined for agroecosystems:
Productivity, which is the output of a valued product per unit of resource input.
This can be measured in yields (biomass) or by some more difficult to define
product such as social, economic, psychological or spiritual well-being; yields can
also be converted to some value such as Kcal.
Stability, which is the constancy of production in the face of small disturbing
forces arising from normal fluctuations and cycles in the surrounding
environment. Included in the environment are those physical, biological, social,
and economic variables that lie outside the agroecosystem under consideration.
Sustainability, which is the ability to maintain productivity in the face of a stress
or shock. After a shock, the productivity of the system may be unaffected or may
fall and then return to the previous level or trend, or settle to a newer level (or
new set of controls as Holling refers to it), or it may collapse altogether; and
Equitability, which is the evenness of distribution of the productivity of the
agricultural system among the human beneficiaries (Conway 1987, Holling 1986,
Odum 1970).
The importance of extending analysis of agricultural systems to the wider
environment and social systems within which agriculture is embedded has become
apparent (Conway 1990). Spedding et al. (1981) emphasized the ecological approach to
whole agricultural systems and not merely to their internal structure and functions.
Spedding also recognizes that the relationship is two-way: just as the health of agriculture
depends on the proper functioning of environmental processes, so does the health of the
environment depend on a respectful agriculture. Likewise, Conway (1990) recognizes a
two-way relationship between agriculture and social institutions. Whereas agriculture is
universally recognized as the provider of food without which hunger, illness, and warfare
would undoubtedly ensue, it also has more subtle relationships with human institutions.
Interdisciplinary research combines the systems approach, and allows for the
simultaneous pursuit of multiple goals, including sustainability, food security, and
improved livelihoods. A balanced pluralist approach, empirically based and with a wide
span in both political economy and physical ecology, is more likely to fit reality. Only
comprehensive, fully interdisciplinary approaches are likely to meet the challenge to
provide less-intensive farming systems. These systems need to be economically,
I
I 45
ecologically, and environmentally sound and sustainable in the long term (Chambers
I 1983, Jordan and Hutcheon 1995).
I Banana Dominated Agroecosystems
Although much research has been done on banana production starting in the early
I 20th century, three authors stand out for having treated the system of bananas and the
I systems of which commercial banana production is a subcomponent as a place where
human communities interface with agroecosystems. University of Michigan ecologist
I John Vandermeer and his colleague Ivette Perfecto wrote an important study looking at
I commercial banana production in Costa Rica and emphasizing the link among bananas,
logging interests and small-scale farmers. The thesis of their work was that these several
components, which had a rich history of working at odds at each other, might also work
3 together to prevent further rainforest destruction in that country (1999).
SThe introduction of a new commodity crop on the world market (bananas) that led
to an economic boom and then an agroecosystem bust in rural communities in Honduras
3 was described by John Soluri (2001 in Flora 2001). In his focus, decisions made outside
the community (such as that only a certain color, shape, and size of fruit-the Gros
Michel-was a banana), despite the wide variety of species available in Honduras,
Increased the monocultural-devastation cycle. The response of local communities,
government authorities, and transnational corporations to agroecosystem changes was
analyzed.
I Perhaps most pertinent to this study, Stephen Striffler studied banana culture in
3 southern Ecuador in the 20th century, from the standpoint of popular struggle and
agrarian restructuring in Ecuador. The author related how the Ecuadorian peasantry
I
gained and then lost control of the banana industry. Using history and anthropology,
Striffler concentrated on the human actors in the banana system and showed how,
although peasants were instrumental in dismantling foreign-owned plantation, they were
unable to handle the contract system of production, which had emerged in that country.
In order to nest his study, the author describes in detail the relations between
multinational fruit corporations, rural workers, agricultural production, and changes in
the system (including the environment) over the better part of the period 1900-1995
(2002)4.
Environmental Aspects of Certification and Forest Fragments
Importance of Coastal Ecuador for Biodiversity
A report "Research Priorities in Tropical Biology," commissioned by the National
Academy of Science highlighted the importance of coastal Ecuador in its summary and
recommendations. It stated it was important to give priority to areas containing the
richest biota, and biota in immediate danger of extinction. The authors suggested as areas
for emphasis the coastal forests of Ecuador, coastal southern Bahia and Espirito Santo in
Brazil, Eastern and southeastern Amazonia (Raven 1980).
Furthermore, they recommended that multidisciplinary efforts unfold on at least
two levels: (1) the investigation, in the tropical domain itself, of socioeconomic
dysfunctions, such as those produced by transplanted models of development, and (2) the
investigation of such dysfunctions in the broad perspective of national and international
linkages: historical, political, social and economic. Specifically, they emphasized that
knowledge of systemic characteristics of forest conversion demands research into the
4 Soluri (2001) also covers the role of multinational fruit corporations, and their interactions with local
communities in his Honduran agroecosystem study.
ways into which aboriginal, peasant, and urban-industrial societies perceive and relate to
the tropical milieus and regulate their use. The fact that these recommendations were
heeded little is evident in the fact that the RPSC forest is the only fragment of any
reasonable size extant in Los Rios Province, Coastal Ecuador.
Deforestation in Ecuador
In Ecuador, it is estimated that 340,000 ha are deforested every year. This
represents an annual loss of between 2.4% and 3.5% of total forest area. Ecuador is, in
relation to its size, perhaps the country with the highest plant diversity in the world, with
between 18,000 and 22,000 species of vascular plants (Ulloa and Jorgensen 1993,
Wunder 2001, 1994). Deforestation of the western foothills of the Andes is resulting in
the extinction of endemic species, the modification of ecological conditions, and the
interruption of evolutionary processes. Thus research into conservation in areas where
forest still exists is urgently needed (Cabarle 1989, Dodson and Gentry 1987).
Reserve Size
Conventional wisdom dictates that larger is better when it comes to nature
reserves. Economies of scale and proximity to other protected areas influence optimal
park size. Rapid expansion of private parks represents an important yet little understood
private-sector incursion into an activity long dominated by governments.
A study found that private parks require research into optimal reserve size so that
quality of protection takes precedence over quantity of land protected (Langholz et al.
2000). This point clearly applies to the Rio Palenque Science Center and Nature Reserve.
Langholz and others also noted that bequest value is particularly high for private reserve
owners. This issue is further discussed in Chapter 6.
Fragmentation
The Rio Palenque Science Center and Forest Reserve should more properly be
called the Forest Fragment Preserve. At only 90 ha in size, this unique forest could be in
danger from various negative effects that have been studied regarding similar forest
fragments. Thomas Lovejoy conducted the first neat, direct experimental test of effects
of fragmentation in the Brazilian Amazon.
Replicate isolates over a size range from one to 1,000 ha were studied prior to
isolation, and followed afterwards. The research team found that although bird diversity
is greatly decreased after the initial five months in which a "crowding effect" occurs,
these effects are not uniformly spread over bird species. Two important guilds are
particularly suppressed after isolation, army-ant, and mixed species ofinsectivores. The
distance by which fragments were separated from unaltered forests also made an
important difference, which varied among taxa.
Perhaps more disturbingly, 15 different species of pollinating insects would not
cross 100-meter cleared strips. This has both direct and indirect effects. The population
biology of at least 30 plant families was dramatically affected by reduced or lack of
pollination. Dung and Carrion feeding beetles responded similarly not crossing a 100-
meter barrier. The decomposition process is slowed. Mammalian extinction and survival
were an indication of the importance of both physical and biological edge effects.
Microclimate changes at the edges of newly isolated fragments were evident within days.
The more obvious ones are changes in light levels and in evaporative losses to dry winds
penetrating at the edge. Air temperature and relative humidity were both affected
(Lovejoy et al. 1984)5.
Other attempts to review the broader span of tropical forest fragmentation have
been undertaken. The studies reveal a relatively small set of causative factors that seem
collectively to explain loss of diversity attributable to tropical forest fragmentation.
These include:
Deforestation related disturbance;
Restriction of population sizes;
Reduced immigration;
Biotic and abiotic edge effects;
Higher order effects; and
Immigration of exotic species (Turner 1996).
According to a study published in Science, much of what remains of tropical
forest is being divided into small, isolated fragments that are unable to sustain their
original biodiversity. In many places, these are the last remnants of primary tropical
forest. Zones where forest edges are affected easily reach up to 1 km wide, and can
impact forest species and biological processes from 100 to 300 meters from the forest
edge. This means that forest fragments of up to 1,000 ha can easily be composed entirely
of edge-affected habitat (Gascon et al. 2000).
A final issue regarding rainforest fragment relates to seeds. Seeds that fall to the
ground in small fragments of tropical rainforest are three to seven times less likely to
5 For summary tables of edge effects in tropical forests fragments see Lovejoy et al. 1984.
sprout than those that fall in larger, continuous forest. Fragments are hotter and dryer,
and have more light penetrating the canopy to the forest floor than continuous forest do.
Those are not conditions that rainforest plants are adapted to, and new results
suggest that the seeds simply cannot survive. Furthermore, plants in fragments could
become inbred, which could make their seeds less likely to germinate in the first place.
Such inbreeding, coupled with edge effects, could push the reproduction rate so low that
the populations in forest fragments would eventually die away (Bruna 1999).
Many consequences of forest fragmentation are likely occurring within the RP
Forest Fragment Reserve. Urgent studies are needed in order to understand the site-
specific processes. Although this study does not address the ecological impacts through
zoological and botanical studies, it may contribute to mitigate negative effects, by
suggesting the creation of biological corridors between the reserve and natural areas on
banana plantations. This proposed connectivity might be useful in reducing patchiness
and allowing for a freer flow of species.
Environmental Certification
The history of environmental certification has its roots in different approaches to
forest certification. It is considered a remarkable social, economic, and historical
phenomenon. Within a relatively short time, environmental certification has solidified its
place as an integral tool for addressing a diversity of environmental issues and has
generated considerable controversy and debate. The combination of media attention,
public pressure, and market forces has created a web of interacting influences that can be
quite complex. However, the goal of certification is simple: its purpose is to ensure
management in accordance with a set of standards considered environmentally
appropriate, socially beneficial, and economically viable.
Certification recognizes that these management values have been achieved by
presenting a seal of approval that can be recognized by the public6 (Cabarle 1994, Kiker
and Putz 1997, Viana et al. 1996, Vogt et al. 2000). Other goals include "increasing
general consumer awareness of the environment; increasing consumer acceptance and
confidence; modifying consumer and manufacturer behavior; to improve the earth's
environmental quality; to increase market share; to provide product differentiation; to
provide an objective audit of the management of natural assets; to promote sustainable
management of these assets; and to demonstrate that natural resource management
provides sustainable economic, ecological, and social benefits" (Cabarle et al. 1995, cited
in Kiker and Putz 1997, p.38).
Kiker and Putz summed up the crux of environmental certification as follows.
Ultimately, the goal of certification processes must be to provide a means
for consumers to more precisely express their tastes, preferences, and
values in the marketplace. The hope is that consumers will respond by
purchasing the certified products and thereby provide greater financial
returns to natural resource managers using ecologically and socially sound
practices (1997).
Although the concept of sustainable forest management can be traced back
to the Middle Ages in Germany and France, the use of environmental certification
as a potential tool to deal with issues such as deforestation and massive pesticide
use can be traced to the 1992 United Nations Conference on Environment and
Development (UNCED), known as Rio 92. After this watershed event, the first
sets of principles, criteria, and indicators were released on how sustainable forest
management should be conducted and developed for forest certification. Some
6 For an in-depth methodological and ideological discussion of ecological certification from an institutional
ecological economics viewpoint, see Kiker, C.F. and Putz, F.E. Also see Figure 6-2, in Ch. 6 of this study.
authors also suggest that environmental certification is a consequence of the
growth of environmental awareness in the 1960s and early 1970s spurred by the
pivotal pesticide expos Silent Spring (Carson 1962).
In the 1980s, the public in developed countries became outraged by the high rates
of deforestation that were occurring and demanded immediate solutions to this problem.
For most people, this problem was recognized to be more acute in developing countries,
where it was also linked to a serious disrespect for the human rights of indigenous
groups. These many viewpoints were important catalysts for environmental groups, more
strongly in Europe, to campaign to the boycott of the purchase of tropical forest products
(Viana et al. 1996). While governments responded with solutions that were regulatory in
nature, whether unilateral or not, some companies began proactive measures to deal with
the public concern. NGOs that have played important roles in the entire spectrum of
certification activities include Conservation International, Friends of the Earth
International, Greenpeace International, The World Conservation Union, The Nature
Conservancy, World Wildlife Fund International, The World Resources Institute, and the
Rainforest Alliance (Vogt et al. 2000). National and international meetings to debate
environmental certification, green seal, trade-related intellectual property rights,
protection and tariffs, the scientific merit of certification, and a host of other issues, have
taken place in fora that range from the United Nations, the European Community, the
United States Congress, and local grass-roots organizations7 (Vogt and Fanzeres 2000).
7 For a study of the business aspects of certification, especially wood products, see Jenkins and Smith 1999.
The Business of Sustainable Forestry: Strategies for an Industry in Transition.
Modeling as an Analytical Tool
Mathematical modeling has been used to analyze the relationship between
ecological and socioeconomic parameters in farming systems (Dykstra 1984, 1994,
Ragsdale 1997 in Thangata 2002, Upton and Dixon 1994). The first step in a region,
farm, agroecosystem, crop or animal system study is the construction of a qualitative
model of the unit under consideration. In the context of this framework, model building
involves identifying the inputs and outputs of the system of interest, the subsystems of
the system, the circuitry connecting these subsystems and the goals and constraints
influencing the system. The next step is to begin to quantify the relationships
hypothesized in the qualitative model, and to construct a quantitative model of the
system. The precision required depends upon how the model will be used.
Models represent a simplification of reality. Simplification involves assumptions,
which in effect are best estimations as to the structure and function of the unit under
study. The validity of these assumptions and the potential of the conceptual framework
can best be evaluated by applying the model to reality and analyzing the results (Hart
1980, Shiyomi and Koizumi 2001).
Qualitative models developed vary with the ecological and socioeconomic
conditions of a specific region, farm, agroecosystem, or crop or animal system.
However, these systems have general inherent characteristics that make it possible to
outline general qualitative models for each level of the hierarchy. For the farming
system, Linear Programming has been shown to be one excellent way of combining
several types of data, including qualitative data, into a quantitative analysis tool. An LP
model that reflects actual farm behavior requires a large amount of data. The choice of
modeling technique depends on the problem and the available resources. Positive
I
54
programming can be a valuable addition to the farm modelers' arsenal (Cabrera et al.
2002, Hall 2001, Kaya et al. 2000). |
Inputs and outputs into the agroecosystems of the farm system can be
grouped into information, energy, and materials categories. Information
enters an agroecosystem in the sense that human, animal, or machine
energy enters an agroecosystem as part of a management plan. Farm
system research requires not only the construction of a qualitative model
describing these relationships, but also a quantitative model where real
numbers are assigned to the farm system inputs and outputs and the flows
between the subsystems of the farm. The primary objective of farm
system research would be to use the model to identify possible
modifications of an existing farm system or to design a new farm system.
The constraints upon this design process, such as labor availability,
nutrition requirements of the family, etc. would be determined before the
generation of a new farm systems begins. The regional system and the
socioeconomic subsystem of the farm are studied to identify
socioeconomic constraints. Agroecosystems are studied to identify the
physical and biological constraints. The performance of the crop system
within the agroecosystem is regulated by managing the agroecosystem,
hence the importance of the human element (Hart 1980).
Agroecosystem research has the ultimate objective of modifying either the
management of the agroecosystem, the crop system, the livestock system or others. This
requires analytical objectives in order to understand how the system functions (build
qualitative and sometimes quantitative models), and also experiments comparing
potential modifications with existing agroecosystems in specific areas (Hart 1980).
These tests can conveniently be undertaken after careful fieldwork, using different types
of linear programs (Breuer 2000, Breuer et al. 2002, Cabrera et al. 2002, Mudhara 2002).
A brief explanation of model behavior in ecological studies must be given in I
order to address the validity of said methodologies. The British meteorologist Lorenz
first noticed erratic model behavior in 1960. An early computer model created to predict
climate was hypersensitive to slight changes in initial conditions. This property is a
fundamental characteristic of a phenomenon that has been termed chaos. Fairly regular,
I
long-term (asymptotic) patterns typical of many economic models give way to quasi-
periodic solutions. Often, these patterns result from the competition among distinct and
incompatible periods in the systems' dynamics, as can occur for example, when an
inherently oscillatory system is driven (forced) by a process such as seasonal demography
that possesses a distinct periodicity. In ecological systems as elsewhere, complexity and
chaotic behavior arise because processes interact on multiple scales (Lorenz 1963,
Mandelbrot 1977, cited in Levin 1991, Schaffer and Kot 1985).
Finally, it should be mentioned that some scientists reject the very notion that
sustainability can be quantified, and thus modeled. The argument is that quantification
tends to distort the research process inducing the researchers to choose quantifiable (but
less relevant) variables at the expense of other non-quantifiable (but conceptually more
important) ones. They are especially skeptical of numerical modeling of biological
systems, arguing that the internal consistency of these models does not compensate for
their lack of realism. Although there is undoubtedly some truth to this, agroecosystem
models-which simulate biological and socioeconomic livelihood systems-produce
results which suggest that often it is possible to quantify and model complex biological
systems without unacceptable loss of realism (MacRae et al. 1989, McCall and Kaplan
1985, cited in Harrington 1992).
Linear Programming
Linear programming is a well-established analytical tool used from industrial
engineering to space exploration. It has often been applied to commercial farming in
developed countries. Although several analytical tools and experimental designs have
been used to test sustainability, i.e. resilience and food security, this study uses linear
programming. Without detracting from the merits of other methods, LP presents several
advantages when working with agriculture and natural resources.
In the most basic sense, an LP model is a tool for economic analysis based upon
constrained optimization. That is, it employs a mathematical algorithm, the simplex
method (Dantzig 1963), to search for the optimal solution to a problem of allocating
constrained resources-typically land, labor, and capital-to various alternative means of
production. LP models are mathematically based models of optimization. In the models,
a linear objective function is maximized or minimized subject to linear inequality
constraints. LP models have been used extensively to formulate farm plans (Hazell and
Norton 1986)8.
A linear program is an optimization model that solves a series of inequalities
subject to certain specified constraints. The general formula being:
Max (or Min): 1 = YiCjXj where j = 1...n
Subject to: liAijXj <= Ri where i = 1...m
And Xi >= 0
It is essentially a heuristic tool designed to simulate the natural or mental process of
resource allocation. It has been used extensively in applied psychology (Langholtz et al.
2001)9. Linear programming was at the core of the early work that eventually led to the
2000 Nobel Prize in economics for Daniel McFadden, of UC Berkeley and James
Heckman of the University of Chicago. They were lauded for the development of theory
8 Hazell and Norton (1986), McCarl and Spreen (1999), and Paris (1991) are recommended texts for
detailed, technical presentations of LP methodology as applied to agricultural production and farming
systems.
9 The study of resource-allocation behavior was particularly relevant to decision-making regarding
peacekeeping. See Langholtz et al. 2001. Other work in this area can be found in The Journal of
Organizational Behavior and Human Decision Processes.
and methods for analyzing discrete.choice. They had-useditto study household decision-
making processes of constituents with regard to federal policies that may affect them.
(www.almaz.com/nobel/economics/2000b.html). The basic tool of economic analysis of
small farm livelihood systems, since it concerns household decision-making is also linear
programming (Hildebrand and Sullivan 2002).
Ethnographic Linear Programming
In the late 1990s, a group of researchers at the University of Florida, who had
been using linear programming for economic analysis of small-farm livelihood systems,
began examining methods for obtainingbetter data. A novel way of understanding.
farming systems was the result. In this methodology, data are gathered ethnographically.
Researchers from several departments employed the technique at sites as diverse
as the Andes highlands, Zimbabwe, Mali, Malawi, Coastal Ecuador, and Eastern
Paraguay (Bastidas 2001, Breuer 2000, Grier 2002,Kaya et al 2000, Mudhara2002,
Thangata 2002). Early work was reported at a pivotal conference in Guelph, Canada
(2000). Breuer (2000, p.75) coined the term "ethnographic linear programming" in his
thesis on medicinal plant cultivation in limited resource livelihood systems in Paraguay.
Since this time interest has been growing in the use and application of ELP methodology.
ELP is a tool within FSRE, and applicable to integrated natural resource management
(INRM) and sustainable livelihoods analysis (SLA).
ELPs are a way of quantifying ethnographic data (mostly qualitative) and are both
descriptive and analytic. The descriptive models are developed to help researchers
understand the complexity and diversity of these systems, and ultimately to simulate the
systems. Once the simulations are validated, the models are used analytically to assess
differential adoption or rejection of potential technologies or the potential impacts on
households of proposed infrastructure of policy changes, or the effect of stresses and
shocks (Kaya et al. 2000). Farmers are asked questions about food, fiber, and medicinal
plant sources, cash production and handling, and constraints on land, labor, cash, and
infrastructure.
This process was used in research for the present study in November 2000, July-
August 2001, and February-March 2002. Research was conducted in farmers' and
workers' dwellings, fields, and at markets. The openness and hospitality of local farmers
allowed the researcher to spend extended periods of time on many farms and participate
in everyday activities. Both male and female farmers were approached and were able to
provide, through informal conversation and survey instruments, valuable information
needed in order to simulate the livelihood system in a computer mode.
Ethnographic linear programming is effective in evaluating various policy
instruments, and assists in uncovering farm household decision-making regarding land
planning and allocation. This is because it takes into account three important aspects
observed by Kuyvenhoven et al. (1999):
Available resource endowment (land, labor, capital).
Multiple objectives of the farm household (profit or end-year cash, home food
consumption, desire for education, etc.); and
Market conditions (prices, access, etc.).
Participatory Linear Programming
Ethnographic Linear Programming (ELP) may coincide with (though not
necessarily) Participatory Linear programming (PLP). In the latter technique, while
eliciting information from farmers, the researcher or extensionist-laptop in hand-
calibrates input coefficients and validates outcomes with the help of the farmer and his
family. PLP may allow for tailor-made.extension in-the near.future. In study of
potential use of improved climate forecasting among Florida livestock producers, Breuer
and others (2002, in review) used PLP to. calibrate.and.validate-livestock production
models. In developing-country research, both Paul Thangata (2002) and Maxwell
Mudhara (2002) in Malawi and Zimbabwe, respectively, have shown the value of taking
the laptop to the hut. PLP was used in the current study in Ecuador for the small-scale
farm and worker household models.
Linear Programming and-Agriculture
Linear programming (LP) based multi-year farm planning models are well
established (e.g. Loftsgard and Heady 1959 in Glen and Tipper 2001). Researchers-have
used linear programming models to determine capital dependent, steady-state cultivation
policies, optimal water allocation, cropping patterns, and the effectiveness of policy
instruments targeting farmers support (Caskie et al. 2001, Doppler et al. 2002, Haouari
and Azaiez 2001, Oweis and Hachum 2001).
Linear programming (LP) has been extensively used to model irrigation systems
in Australia and elsewhere. Heterogeneous farm-household models that capture social
and spatial interactions explicitly, and the individual choice of the farm household among
available production, consumption, investment, and marketing alternatives have also been
represented in recursive linear programming models. Adoption constraints were
introduced in the form of network-threshold values that reflect the cumulative effects of
experience and observation of peers' experiences (Berger 2001, Hall 2001).
The fact that in complex decision problems, some objectives are not well
quantified or are not introduced explicitly in optimization models is an apparent
limitation of models. Solutions that are nearly optimal, i.e. deviating less than a
predefined percentage from the optimal value of the quantified objective functions, can
constitute interesting alternatives for stakeholders. A study with a model developed to
design farming systems in the Netherlands, showed that nearly optimal solutions provide
relevant information to resolve complex decision problems (Makowski et al. 2001).
Peasant farm households in Malawi were modeled in a study designed to evaluate
the need to allow for embedded risk. Discrete Stochastic Programming (DSP) provided
more efficient solutions for problems with embedded risks. The study showed that labor-
scarce households adopt more tactical, sequential responses to uncertainty, especially
those households with limited access to capital and credit markets (Dorward 1999).
The impact of agriculture on land-use patterns, the environment and socio-
economics of a region, were incorporated into a farm-type linear programming sub-model
with environmental and socio-economic sub-models. In the environmental model, the
environmental factors and management practices for different land cover classifications
that were relevant to farm activity were defined. Management practice changes and the
land cover type of the 10 nearest parcels of land were used to determine the likelihood of
a parcel of land changing its associated vegetation type (Topp and Mitchell 2001).
A study of melina (known alternately as gomelina) (Gmelina arborea) and teak
(Tectona grandis) plantations on land suited for agriculture in the humid tropical Atlantic
lowlands of Costa Rica used a linear programming model that maximized regional
economic surplus, and the so-called technical coefficient generators that quantify input
and outputs of forestry and competing crop and pasture-based beef cattle production
systems. Results indicate that teak and Melina plantations are attractive land use options
while managed natural forest is not. The model also predicted that diversity in the
landscape would be reduced and changes in farm output would result in falling incomes
and employment across the region (Nieuwenhuyse et al. 2000).
These studies lie at the interface between purely agricultural production studies
using LP, and another group of studies, which apply LP to sustainability. A linear
programming model was constructed based on availability of various energy sources and
requirements of various human and agricultural activities in Tamil Nadu, India. The
model optimized allocation of land and energy. Since the model is adaptable to other
local conditions, it can have universal application for sustainable agricultural
development (Raja 1997). Terry Kelly (1995) used recursive linear programming in a
bioeconomic systems approach to sustainability analysis at the farm level. This study
presents how sustainability can be achieved and maintained.on a North Floridapeanut
farm. The value of this early study is in attempting to define measurable sustainability
indicators, which had confounded some earlier researchers, and then proceeding.to LP
analysis with viable outcomes.
Linear Programming Applied to Farm Household.Analysis
In farming systems LP models, the optimal solution is determined by the
objective function. This often involves a combination of maximizing.returns from
household production while meeting certain other household goals such as food security
(Bernet et al. 2001, Hazell and Norton 1986, Pannel 1997, Schilizze and Boulier 1997).
By producing a LP model that, albeit in a simplified manner, represents a real-
world livelihood system, it is then possible to run relatively inexpensive assessments of
both policy scenarios and project ideas generated by extension agencies and policy-
makers (Bernet et al. 2001, p.184). In this way researchers and extension agents can test
production alternatives in the virtual livelihood system which will give a rough prediction
of how the real world might respond-as seen through a change in livelihood strategies
(Hazell and Norton 1986, Pannel 1997, Schilizze and Boulier 1997).
Ellis (1993) observed that, in reality, smallholder farmers maximize "profits"
subject to trade-offs with other goals, resource constraints and performance of markets.
Further, he noted that the profits do not have to be in the form of money. Adjustment of
inputs and outputs should give the household a higher net income. In this case, profit can
be interpreted as end of year, or discretionary cash (after taking care of home
consumption).
LP models can handle multiple crop activities. Effects of changes in decisions on
the various crops can be determined (Singh and Subramanian 1986). Effects of
technology and changes in market variables on the objective function can be traced in the
LP model. Decisions on allocation of expenditure to different goods, and the allocation
of fixed and variable inputs to different production activities in the short run can be
incorporated. In any production cycle, a household objective function can be imposed
subject to given constraints. LP models are flexible to changes in assumptions, technical
coefficients and activities in the farming system. The marginal effects of such changes
can also be pursued.
In reality, farmers often consider several aspects of their welfare when making
these decisions (Sen 1985). This attribute can be addressed in LP models since profit
maximization does not need to be the only goal. Other objectives like minimizing labor
use can also be accommodated. Multiple objectives such as maximizing revenue after
satisfying subsistence requirements can be incorporated in the model.
Crop techniques that might be used by farmers can be introduced into the LP
matrix (Bade et al. 1996, Sicular 1986, Timmer et al. 1983). Activities are added to
existing crop and input activities to assess the likely reaction of farmers when presented
with new technologies, or shocks (Mudhara 2002). Linear programming (LP) is a cost-
effective way of understanding farm household responses to pricing policies, input use,
and shocks. It is useful for solving problems that deal with the allocation of limited
resources among various activities to achieve a particular objective.
Linear programming models in farming systems analysis differ from statistical
models. They are based principally upon a mechanistic rather than probabilistic
understanding of the system at hand (Bernet et al. 2001, Hazell and Norton 1986).
Linear Programming and Sustainability
Several studies have used linear programming to approach the issue of
sustainability. As the complexities of agroecosystems are incorporated into models over
multiple spatial and temporal scales, useful results are still obtainable. Of these, several
stand out. The use of linear programming (LP) models in the assessment of
socioeconomic sustainability prior to implementation can bring the planning process for
extension projects closer to the reality of small farmers (Bernet et al. 2001, Hazell and
Norton 1986, Kaya et al. 2000, Pannel 1997, Sulser 2001).
In Australia, modeling sustainable management at the regional level was
undertaken. In it, the authors explicitly attempt to reduce the complexity common
ecological systems models at different scales. The research showed that computer
optimization using linear programming can solve integrative forest resource models at a
resolution suitable for operational planning for forest regions where normally at least two
levels of planning would be applied (Turner et al. 2002).
In a similar study involving multiple scales a framework for integrated
biophysical and economic land use called sustainable options for land use (SOLUS), was
developed over a 10-year period of investigation in the Northern Atlantic Zone of Costa
Rica and encompasses scale levels that range from field to region. Biophysical and
economic disciplines are integrated and various types of knowledge, ranging from
empirical expert judgment to deterministic process models are synthesized in a systems-
analytical manner. Examples of application of SOLUS in the Northern Atlantic Zone of
Costa Rica show that introduction of alternative technologies may result in situations that
satisfy both economic and biophysical sustainability. On the other hand, negative trade-
offs were found among different dimensions of biophysical sustainability themselves
(Bouman et al. 1999).
This Costa Rica study was followed by another by the same team in which
economic and biophysical sustainability trade-offs in land use exploration at the regional
level were quantified. A methodology was presented for exploration of sustainable land
use options at the regional level by quantifying trade-offs between socio-economic and
biophysical sustainability objectives. The linear programming model maximizes regional
economic surplus subject to a flexible number of resource and sustainability constraints.
Though ample scope was found to exist for reducing environmental effects and
introducing sustainable production systems separately, pursuing both objectives
simultaneously considerably reduces economic surplus and agricultural employment.
Agricultural area can be decreased and forested area increased without severely affecting
the regional economic surplus (Bouman et al. 1998).
The issue of modeling multiple objectives for land use change was tackled
through interdisciplinary research in a 1999 study. The authors began from the premise
that sustainable development is not readily measurable, except as a compromise between
different parts of society, of which some may try to represent future generations of
mankind. To determine a sustainable development path in the relationship between
agriculture and its natural environment, a profound knowledge of this complex system
and its behavior under different socio-economic conditions was necessary.
A modeling system consisting of a set of hierarchically linked modules was
developed. The heart of the modeling system was a multiple-goal linear programming
model. Simulation of single farm models and regional models based on simultaneously
optimized farm types was possible. The modeling system appeared to be a highly
flexible tool with respect to the number and type of farms, sites, and production
techniques. Environmental objectives were included, and different levels of goal
achievement simulated (Zander and Kachele 1999).
Public policy is another area where linear programming proved useful in one
study for identifying ecologically and economically efficient and sustainable use of
agricultural landscapes. In farms and rural family households, basic decisions are taken
on the intensity of agricultural landscape use. Guiding landscape use in the right
direction is a difficult task for public decision makers because ecologically and
economically efficient and sustainable uses of agricultural landscapes are ill defined. The
author points out that this problem results from unknown values, unknown interrelations
between technology and nature, and unknown evolutionary processes in nature and in
societies. Thus, there is a broad area for speculation about optimal landscape use. In a
I
66
study of an agricultural landscape in Southern Germany, eight parts of a naturally
homogeneous region were analyzed using multicriteria valuation, linear programming
and trade-off analyses, the author shows that speculation can be minimized. Efficient
agricultural policy options were revealed for the simultaneous achievement of ecological
and economic efficiency and sustainability (Werner 1993). I
Another study from Costa Rica, a country that seems to be especially rich in I
sustainability studies, looked at the farm level trade-offs of intensifying tropical milk
production. Bottlenecks to intensification and how they affect choices between more and I
less sustainable practices were identified. A linear programming model of a highland I
dairy farm was constructed to simulate farmer decision-making. Significant economic
and environmental trade-offs were found. Land intensification was found not to be U
economic unless researchers can provide laborsaving devices appropriate for small farms I
or improve protein content of forages. Since the value of land for producing milk
exceeds clearing costs in nearly all models, compliance with land use controls to improve I
environmental sustainability is unlikely. Finally, it was found that targeting input and I
credit subsidies to inputs, which improve labor productivity and encourage
environmentally sustainable practices would mitigate economic and environmental trade-
offs in milk production (Griffith and Zepeda 1994).
In a similar study, linear programming was used to explore the sustainability of
beef cattle ranching in the humid tropics of Costa Rica. Breeding and fattening systems
were compared using N depletion as an indicator. The study attempted to break the
popular view of the "tropical forest-livestock connection" by showing that long-term N
I
I
mining can be avoided and cattle production can be sustainable even under humid
tropical rainforest conditions (Bouman and Nieuwenhuyse 1999).
Rural Development and Natural Resource Management
The history of development economics can roughly be divided into two periods:
the economic-growth-and-modernization era of the 1950s and the 1960s, when
development was defined largely in terms of growth in average per capital output; and the
growth-with-equity period since around 1970, when the concern of most development
economists broadened to include income distribution, employment, nutrition, and a host
of other variables. The prevailing view of agriculture's role in development changed
profoundly during these two periods (Eicher and Staatz 1984).
During the reconstruction of Europe through the Marshall plan, Americans and
researchers from other developed nations became aware that many countries were living
in a state ofunderdevelopment. This set off a series of projects basically aimed at
strengthening countries' economies through industrialization and infrastructure schemes.
Agriculture was not a focus of serious development research at this time. Even today, the
role of infrastructure is recognized by the president of the Rockefeller Foundation as
crucial to development.
Accelerated growth in agricultural output cannot be maintained without
adequate investments in rural infrastructure and in agricultural research
and extension. Indeed, without such investment the results of
liberalization policies may well fall short of expectations and set
governments against market-oriented approaches (Conway 1997, p.39).
Some enlightened economists, however, saw many positive aspects about
agriculture and its role in development. Johnston and Mellor (1961) for example,
argued that agriculture could provide five important contributions to structural
transformation: labor, capital, foreign exchange, food for a growing industrial
sector, and a market for domestically produced industrial goods.
Afterward, development economics moved to community development
projects, which assumed western values could be directly transferred to third
world communities. A major setback was a series of political disasters that struck
from the 1960s onward. These seemed to be somehow related to the stresses and
strains accompanying development and "modernization." This gave pause to
development specialists who had cultivated development economics impelled by
a vision of a better world. Most had presumed that all good things go together,
and many were shocked that economic growth entailed, not infrequently, a
sequence of events involving serious regression in other areas including a
wholesale loss of civil and human rights (Hirschman 1981).
Also beginning in the mid-1960s and continuing into the 1970s, scientists began
to apply technology and management to world agricultural systems. They began by
developing new strains of corn, wheat, rice, and other staple crops. Total grain
production increased dramatically even as world population continued to climb.
However, these potentially high-yielding varieties required intensive management, and
large amounts of external inputs. Farmers the world over were encouraged to plant these
new strains as monocultures, promising huge economic returns and increased regional
productivity.
This improvement in agricultural production can be linked to the proliferation of
irrigation, the more intensive use of fertilizer and plant protection chemicals, along with
the development of new crop varieties. These simultaneous developments resulted in
cropping systems that are capable of responding to higher levels of inputs and
management.
The higher crop yields attained during the first green revolution were most
readily achieved in robust soil areas. In those areas where agricultural
production expanded most rapidly, the rural poor benefited; but, overall, it
was the urban poor who were clearly the beneficiaries of the green
revolution. Technologies that will enhance the productivity and the
incomes of farmers living in poorer soil and resource locations are now
needed (Ruttan 1997, p.v).
Early enthusiasts for the tremendous increases in agricultural production met
strong early criticism by Frankel (1971), Griffin (1974), Lipton (1977), and others. They
argued that changes benefited mainly wealthy landlords and large farms in ecologically
favored areas, while they frequently impoverished small farmers and tenants, especially
in upland areas (Hayami and Ruttan 1985).
Dependency theorists argued that unequal terms of trade favored importation of
manufactured goods and the export of raw agricultural goods. These policies limited the
internal markets for consumer goods and led to the impoverishment of the mass of small
farmers (de Janvry 1981, Prebisch 1959, 1981). Johnston and Kilby (1975) showed that
broad-based agricultural growth was more effective than estate (plantation or hacienda)
production in stimulating demand for industrial products (whether these be in-country
produced or imported) and hence speeded structural transformation of the economy.
The period was pervaded by abstract theorizing at the macro level along with a lack
of attention to the need for technological change in agriculture and a lack of attention to
the biological, environmental, and location-specific nature of agricultural production
processes. Furthermore, there was little solid foundation based on empirical research at
the farm and village level. Recognition of these failures was an important element that
led to new approaches in agricultural research beginning in the 1970s (Collinson 2001,
70
Hildebrand 1986). Conway (1990) searching for approaches to address those "left
behind" by the Green Revolution, recognized the importance of the more subtle i
relationships between rural development and local knowledge and institutions as follows:
The large highly mechanized farms of the industrialized countries are a
consequence of government policies and, in particular, programs of
subsidies intended to ensure high standards of living for farmers. It is
also, though, a process that has led to considerable reduction in the size of
the farming community and a loss of the traditional family farm. In the
developing countries, similar trends are occurring on those well-favored
lands, which have been subject to the Green Revolution, with dramatic
changes in the nature and structure of traditional societies. Elsewhere, on
the more marginal lands where agriculture is relatively stagnant, there is a
growing appreciation of the role of village institutions and family
relationships in agricultural development (Gordon Conway 1990, p.207).
By 1970, it had also become apparent that urban industry in most countries could
not expand quickly enough in the short term to provide employment for the expanding I
labor force. Hence, the concern of development planners shifted to finding ways to hold i
people in the countryside (Eicher et al. 1970). The downside of this, especially in
regarding the "poles of development" projects, was an attraction of too many people to a I
particular region, thus overtaxing the natural resource base (Popenoe pers. comm. 2001). 3
Yet the need still exists, especially in Latin America, where landlessness is an
important issue (Barsky 1984, Binswanger and Landell-Mills 1995, de Almeida 1995, de
Janvry 1981, Form and Blum 1965). There are and will continue to be many limited-
resource small-scale farmers. Among the many reasons for this, one theory explains in
part, why this phenomenon occurs.
The different processes of wage determination in the traditional and
modern sectors give rise to "labor market dualism." Although the implicit
wage of workers in the traditional sector (based on their average
productivity) is not necessarily lower than the modern sector wage level
(based on marginal productivity), it generally is so, due to the low
opportunity cost of labor in the traditional sector. Large numbers of
workers remain in the traditional agriculture despite low wages, due either
I
to ignorance of better opportunities outside agriculture, or to their inability
to obtain a modem sector job despite wishing to do so, or to the costs of
moving being unacceptably high (including the cost of giving up the
relative security of remaining at home) in relation to the expected wage
premium (Todaro 1969 in Ghatak and Ingersent 1984, p.8).
In the early 1980s, World Bank economists began calling greater attention to food
security, nutrition, health, and other basic needs. Becker (1965) introduced the new
household economics theory that considers households as unified units of production and
consumption. Household modeling in which households have one utility function are
appropriate for smallholder farmers. In the new household economics theory, households
maximize utility subject to a resource and time constraint.
Schultz (1964), and Lipton (1977), in arguing that traditional farmers are efficient
allocators of available resources given their knowledge of technologies, set the stage for
Farming Systems Research and Extension (FSRE), which creates disequilibrium within
systems with new technology, thus allowing for reallocation to a higher level of
productivity (Hildebrand in Jones and Street 1990). The Farming Systems Research-
Extension methodology, for working at the microeconomic level with farmers, is a
practical focus to attain these ends. Further detail on the FSRE approach is given on
Chapter 5 of this dissertation. In the late 1990s, many FSRE concepts were incorporated
into projects that differed little fundamentally from those of earlier years, but received the
name of Natural Resource Management projects, a title more in keeping with current
funding attitudes.
Rural development now calls for increased agricultural output, but also for
sustainability and equitable distribution or social justice (Altieri 2000, 2001, Chambers
1983, Conway 1985). Years before environmentalists began calling for a sustainability
approach, two distinct schools of thought merged to form what has come to be known as
the human development paradigm. These were the basic needs approach oful Haq and
Streeten, on the one hand; and the human development paradigm of Indian economist
Amartya Sen on the other (Sen 1981, 1992, cited in Wise 2001, Streeten et al. 1981). The
latter uses the terms capabilities and functioning in the same sense that activities and
strategies are used in the present study. Capabilities and activities refer to the sets of
choices available to different individuals in groups within society. Functioning and
strategies refer to the options actually chosen by the individuals. Interdisciplinary
research is essential to this new notion of nonlinear development work (Flora pers.
comm. 2002).
Experience suggests that many, if not most, of the crucial questions for agricultural
development lie not in one province or another, but at their intersection. In the optimistic
words of Robert Chambers:
A massive shift in priorities and thinking is taking place, from things and
infrastructure to people and capabilities. Consonant with this shift, five
words, taken together seem to capture and express much of an emerging
consensus. These are well-being, livelihood, capability, equity, and
sustainability. The objective of development is well-being for all.
Livelihood security is basic to well-being. Capabilities are means to
livelihood and well-being. The poor, weak, vulnerable and exploited
should some first. To be good, conditions and change must be
sustainable-economically, socially, institutionally, and environmentally
(1997, p.9)'.
This study takes this human-centered approach. Sustainability is sought not as an
end in itself, but as a means of improving food security, local livelihoods, and overall
socioeconomic stability.
10 For greater detail on equity issues see Whose Reality Counts: Putting the first last, by Robert Chambers,
1997, pp.9-14. Also the various works of Amartya Sen and David Korten.
Summary
This literature has attempted to paint a picture of how interdisciplinary research
can relate sustainability and resilience, food security and improved livelihoods. It has
further delved into resilience as a concept, an indicator and an emergent property,
attempting to tie together some of the underpinnings and prevailing theories in the area.
We have described both historically and theoretically, why the systems approach is ever
more useful for description and analysis of complex and diverse systems. In an important
sub-theme, three major studies on banana-dominated agroecosystems, similar to the one
being analyzed in this dissertation are mentioned.
Because most hypotheses guiding this research are examined using mathematical
simulation models, the literature on modeling as an analytical tool, in general, and more
specific works on ethnographic and participatory linear programming were listed and
briefly described. Specific studies in which linear programming has been used for
agricultural and sustainability research were reviewed. Finally, a section on rural
development and natural resource management, seen in a broad overview of its evolution
from the 1950s to the present was presented. The literature review is summed up and
ends on an optimistic quote from Robert Chambers regarding the direction of future
development work. It is hoped that this work is a grain of sand on the path described by
the eminent Briton.
In the next chapter, the first component of the agrosocioecosystem will be
described. It refers to the commercial banana system in the research area. Chapters will
follow on the other three components of the agrosocioecosystem. Data will then be
brought together for analysis, conclusions, and discussion.
I
I
I
CHAPTER 3I
COMMERCIAL BANANA PLANTATION SYSTEM
Introduction I
A thorough analysis of the agrosocioecosystem under study could not be
undertaken without the inclusion of the most important component of the system, the
commercial banana plantation system. The size, extent, and reach of the banana |
haciendas into the lives of so many that reside in the study area are impossible to
overlook. The banana haciendas play a role similar to a manufacturing industry in the
sense that they provide the greatest amount of employment available in the area (Lewis
1954). They are industrial-type, or "hard systems." I
An industry that provides over 40% of agricultural exports for a country cannot be
dealt with lightly. Nor can research recommendations suggest doing away with the I
activity altogether with a notion that other source of production are better and should i
therefore replace commercial bananas. My work seeks to suggest ways in which some of
the aspects that may make small-scale agriculture agroecologically socially sustainable,
and economically more equitable can be incorporated into a larger system dominated by
commercial banana production.
The study hypothesized that small-scale farm households are better able to U
survive stress and shocks than town-dwelling plantation workers. If this were true, then
transferring a number of plantation workers living in local towns to small farms would
make the overall system more resilient. Furthermore, if amenities such as schools,
clinics, etc. were provided, it would reduce out migration, creating further worker 3
74
stability in the area. Additionally, the research suggested this could be achieved without
negatively affecting the commercial banana system.
One way to obtain these positive changes may be by establishing small-farm
homes where banana plantation workers can provide for a good percentage of their
nutritional needs. Plantation workers' salaries are low. Higher salaries would most
likely take Ecuador out of competition with other purveyors of the fruit, especially
Central America, where the three largest multinational banana corporations own nearly
70% of the land in production (Vandermeer and Perfecto 1999). It is useful (though
disheartening) to remember David Ricardo's 1820 "iron law" regarding wages: "Labor,
like all other things, which are purchased and sold, and which may be increased or
diminish in quantity, has its natural and its market price. The natural price of labor is that
price which is necessary to enable the laborers to subsist and perpetuate their race,
without increase or diminution (cited in Galbraith 1958, p.33)."
If wages cannot be raised, and yet, ways to improve the livelihoods of plantation
workers are needed, altering the condition of some of the workers from landless to landed
may be an option. With an open mindset, and a firm belief that enterprises should
provide the greatest societal welfare, a relatively simple restructuring of living
arrangements may lead to greater food security, well-being, and enhanced long-term
economic stability for the commercial banana sub-sector of the economy.
Background
Bananas are one of the most widely consumed fruits in the world. Yet, bananas
do not grow outside the tropics. Hundreds of millions of people in developed countries
are provided with bananas through the highly developed banana production system.
Although they are one of the most common fruits, bananas were considered to be exotic
and hard to obtain until the 20th century. It is only during the past 100 years, and
especially since World War II, that bananas have become a common food item in the
United States, Europe and Japan.
Countries with large populations that also have part of their territory in tropical
zones produce a great deal of bananas. India, Brazil, and Indonesia are huge producers of
the fruit, but most of what they produce is consumed within their borders. Another group
of countries, generally located within the tropics and with relatively small populations,
have a specialized banana production and export industry. These countries include the
Philippines (the largest supplier for the Japanese market), several West African countries
that supply part of Europe's needs, and especially the Central American countries and
Ecuador. A series of interesting conditions have converged to make Ecuador, above all
these, the largest banana exporter in the world.
The objectives of this chapter are to describe the commercial banana system as it is
practiced in Los Rios Province through the eyes of the people directly involved. An
increased understanding of the mechanics and subtleties of banana production and
commercialization is sought through extensive references to interviews. Land, labor,
capital, and management are explored as production resources. Agronomic issues and
cultural practices are described. Environmental certification is briefly discussed (further
discussion is in Chapter 6). A brief history of banana production in Ecuador, which
follows, is necessary for understanding this chapter and the overall dissertation.
Banana Production in Ecuador
Banana production in the study area is the industrial, or commercial agricultural
element of a dualistic agricultural system in the study site (Lewis 1954, Ruthenberg
1980) for much of the 20th century and into the 21st it has been that way. The other half
of the duality is near-subsistence farming. Small-scale farming has existed in Ecuador
from pre-Columbian times. Bananas, however, are not native to Ecuador. It was only
after the European encounter that the species was introduced (probably from the Canary
Islands then Panama) sometime in the 16th century (May and Plaza, 1958). Along with
plantains, bananas became a mainstay of slash and burn agricultural plots on the western
coast. But it was not until foreign interest began studying the possibility of production in
Ecuador that formal commercial plantations were installed.
Before the 1960s, bananas could not be cultivated on the same land for more than
10 years. This made their cultivation ideal for large companies. The United Fruit
Company for example owned vast tracts of land in Costa Rica and Panama until the
Panama disease' devastated production in both countries. This led to a search for
disease-free production sites. These included Guatemala, Honduras, Colombia, Mexico,
and eventually Ecuador. Ecuador was last on the list for a variety of reasons. American
companies were hesitant to get into Ecuador in part because of the rugged nature of the
coast and the lack of good transportation and port facilities (Striffler 2002).
It was out of this continual search for new lands and a cheap, docile labor force
that companies entered Ecuador and began exploring sites in the late 1920s. Cacao
production on the Ecuadorian coast had collapsed leaving large quantities of land and
labor available. Since Ecuador had never produced bananas on a large scale, Panama
disease was largely absent and workers had no experience in dealing with foreign banana
companies. In 1934, United Fruit (now known as Chiquita Brands) purchased a 100,000
' A fungal disease caused by Fusarium oxysporum f.sp. cubense. There are 4 races of this pathogen
recognized to date.
ha tract in southern Ecuador. They were not the first. Two Chilean companies, and one
Swedish operation had been working for some time, albeit on a small-scale (Larrea 1992,
Striffler 2002).
During the 1940s and 1950s, United Fruit built much infrastructure including a
system of railroads and network of roads and warehouses needed for housing, feeding,
and otherwise maintaining thousands of workers. "United Fruit was the largest producer
of bananas in Ecuador from the late 1940s to the late 1950s. They employed 2,000
workers and exported 80,000 stems/week (Striffler 2002, p. 13)." The many social and
political problems associated with the Central American model eventually led the
multinationals to adopt the alternate model of contract farming.
Aside from moving to new areas, multinational companies also spread their risk
by contracting to local producers. This was done on an experimental basis until the
1970s when it became the most common practice, especially in Ecuador. The principal
problem was a political one of land tenure. Confident that banana exports would
continue to flow from a variety of sources, companies could afford to wait out strikes,
repress workers, or simply abandon production, in certain places ibidd, 2002).
However, the relative absence of multinationals from direct production did not
make the Ecuadorian banana industry more independent. It simply made domestic
producers more vulnerable. Ecuador's role as the world's largest banana exporter has
often been attributed to excellent conditions and highly developed marketing skills. In
fact, however, the country functions as a secondary or reserve supplier (Espinel 2002,
Striffler 2002). Ecuadorian bananas are less expensive, and come into their peek during
the off-season for Central American fruit.
By some accounts, coastal Ecuador is not agroecologically an ideal place to
produce bananas. Although soil quality is very good, average yearly sunlight hours are
rather low (Nifiez Torres 1998), and the dry season is "too dry for too long," which
causes plant stress that takes a long time to recover (a commercial banana plantation
manager in the study area pers. comm. 2002). Furthermore, whereas the Gros Michel
variety grown in the first half of the 20th century required large quantities of land, the
newer Cavendish varieties need a very high level of inputs. Conventional wisdom
dictates that from 20 to 50 ha are essential to produce Cavendish. Because of the
advantages of economies of scale, even larger areas are better.
The use of national capital, labor force, initiative in production, and marketing,
have distinguished Ecuador from some other banana producing countries. Currently,
Ecuadorian companies grow 100% of Ecuadorian banana production, and national
companies market approximately 70% of output.
Methods
Banana production in the study area is a vast, complex, and highly organized
enterprise. Qualitative information was considered the most important data to be
gathered in the study. Production and export statistics are available on line and from a
variety of other sources. In order to attempt to grasp the main actors, issues, and drivers
of this component of the Rio Palenque agrosocioecosystem, methods pertaining both to
Farming Systems Research and Extension, and to quantitative and qualitative research
methods in anthropology (Bernard 1995) were employed. An initial transect was run to
have an overview of the presence of the banana sector in the research area. Interviews
were conducted with an important banana company's managers. These interviews were
held in a variety of settings, including company headquarters in Guayaquil, the main
commercial office in Quevedo, the hub of the northern production zone, and in vehicles
traveling between sites. Extensive interviews were held with the vice-president of a
major Ecuadorian banana corporation. The general manager and other top executives
were also interviewed. Questions were direct and answers were frank, clear and without
hesitation.
Farm visits, both on company-owned haciendas and those of independent
producers were used to gather data from an array of different sources. Ecuadorian and
Costa Rican agricultural engineers working as managers provided important insight,
especially regarding the comparison of the Ecuadorian banana system to that of their
home country. Hacienda foremen provided useful information regarding agronomic
techniques and cultural practices, labor costs and organization, and information on the
packing process and certification. Others interviewed on farms included packing gang
workers, field workers, cooks, office personnel, and drivers. Each one of these played
their part in filling in gaps regarding practices, customs and relationships. Another
important technique used was the identification of key informants (Bernard 1995,
Rhoades 1982).
These persons with whom the author came into contact as early as November
2000, proved to be crucial during subsequent field research in July-August 2001 and
February-March, 2002. As with any well identified and collaborative key informants, the
managers, biologists, social workers, several hacienda workers, and a local politician,
more than made up what they lacked in numbers by the breadth and scope of their
knowledge of the banana production system, from its history, nuts and bolts, to the nitty
gritty of buying and selling, and labor issues.
Finally, secondary data from a wide range of sources including Ecuadorian
government statistics, NGO reports, FAO statistics, traditional library sources and the
Internet were gathered. Numerical data corresponding to commercial banana production
are summarized in tables, and can be found in Appendix B of this dissertation. These data
will be used as input coefficients for the linear programming analysis that is conducted in
Chapter 7.
Importance of Bananas to the Ecuadorian Economy
United Fruit, Dole, and Del Monte, the three major multinational banana
companies, are not direct producers in Ecuador. Ecuador differs from other banana
producing countries, such as Costa Rica, Panama, Honduras, Guatemala, and even
Colombia, in that there is a wide diversity of landholdings in banana production (Espinel
2002).
Around 6,000 banana-producing units existed in Ecuador in 1998. These can be
divided into four categories according to plantation size. Nearly 80% of production is on
farms smaller than 30 ha. Therefore, the productive structure of the banana sector in
Ecuador depends largely on small and medium producers (SICA 1998). One of the top
executives at a banana company disputes this government information. In July 2001, he
confirmed consolidation that is going on in the banana sector. "There are many
producers that lately have invested in land. I would say that 20% have less than 50 ha;
60% are between 50 and 100 ha and 20% are larger than 100 ha, so, that sector is
growing" (a banana company executive pers. comm. 2002). The 2000 agrarian census
does nothing to clear up the disparity of data. The author however was able to gather
enough information through personal interviews to substantiate the interviewed
executive's consolidation theory, at least in the study.area.
I
82
Export banana marketing has strongly contributed to the country's economy in
1997, accounting for 35% of total exports. Contribution to total GDP was 3%, and 16% 1
to the agrarian GDP. Baana output in 1992 was 26 MT/ha on 135,000 ha, and 37
MT/ha from 127,000 ha in 1997. Ecuador has developed an integral banana industry,
including production, processing, shipping, and marketing. I
Table 3-1. Banana area per province I
Province Area (ha) No. Of Producers Average area/producer
EL ORO 44,101 2,656 16.6
LOS RIOS 44,994 841 53.5
GUAYAS 42,743 1,810 23.7
Source: Banana Unit, Ministry of Agriculture and Livestock (MAG), Quito
Prepared by the Servicio de Informaci6n y Censo Agropecuario (SICA) SICA-
IBRD/MAG Project Ecuador (www.sica.gov.ec) I
Los Rios Province is now the largest banana producer in Ecuador closely followed I
by El Oro, and Guayas (SICA 2001). A conversation with some men who considered I
themselves "old timers" in the town of Buena F6 shed further light on the changes that
had taken place in the Province during their lifetimes, including the switch from the Gros
Michel style of production to the more intensive Cavendish type: I
In the 1940s there was a lot of banana, but Panama disease killed the
incipient plantations in a very brief lapse of time. The first big company
was German, then Standard (Dole) came along. They used to buy bananas
in bunches. "This area was all palm, and is now banana (also passion fruit,
cacao, palm, which is processed here). The road was finished in 1958. I
remember well because big steamships used to come up to Quevedo from
Guayaquil. The last ones came in 1958. This is the story: first there was
cacao here (1880s-1920s). In the 30s and 40s this was all jungle (where
there had been large cacao plantations), but it was not all cacao. One 300
ha hacienda had perhaps 100 ha of cacao. The rest was forest. Cacao
plantations were not destroyed when bananas first moved in. Later palm 3
replaced much cacao. Bananas were planted on virgin forestland (starting
big in 1958 with the road). Then the disease came in the 1960s and it was
back to palm again, all this until 25 years ago when short cycle grains
I
became popular. Maize and soybeans had a good period from the mid
1980s to the mid 1990s. Now it is bananas again (a long-time Los Rios
resident and businessman 2002).
In order of importance, Noboa, Reybanpac/Favorita Fruit, and Pons all
have plantations over 100 ha (Espinel 2002). The agrosocioecosystem described
and analyzed in this dissertation is defined around an area where one of these
companies and its associated producers have a very strong presence.
According to Empresa de Manifiestos, total banana exports during 2001 were
3,990,427 MT, 5 percent lower than the year 2000 exports, which reached 4,221,599 MT.
Eighty four percent of banana exports are addressed to seven important markets, the most
important ones being the EU (34%), the U.S. (22%), and Russia (16%).
Markets
During the 1999-2001 period a change in market composition occurred. Thus, the
U.S. decreased from 31 percent to 22 percent, Russia increased from 6 percent to 16
percent, China fell from 4 percent to 3 percent, the Southern area of the South America
region remained in at 10 percent, and the EU rose from 16 percent to 20 percent.
Important market sales grew during the 2000-2001 period, the U.S. increased its
purchases by 14%. Other markets decreased; China 47%, Argentina 8%, and Chile 4%.
Comparing the years 2001 and 1999 data, some markets decreased-U.S. (-43%), the EU
(-6%), China (-34%), and Chile (-17%)-while other markets increased their
percentages, Russia (154%), Poland (6%), and Argentina (8%).
It is important to highlight that the sector's evolution was adversely affected by
climatic problems that reduced internal supply. The annual average price of a banana
box in the U.S. market improved during the year 2001, quoting an average price of USD
9.44 among all trade marks; i.e., 42% up from the year 2000 records; USD 6.65, and 53%
up from 1999.
The annual average prices of a banana box in the EU market also recovered,
registering an average quote of USD 14.81 for main trademarks, 9% up from the year
2000 that had registered USD 13.55. However, this quote did not surpass 1999 records.
Certain conditions were established for qualifying nontraditional operators, which
favors the participation of the Ecuadorian exporting companies. During 2000, the
Eastern Europe Market also recovered, quoting an annual average of USD 8.24/box, up
17% from the year 2000, USD 7.05; and 1% up from 1999.
Table 3-2. Banana exports from Ecuador
Year Area (ha) Exporting (MT) Exportable Yield
(MT/ha)
1994 124,417 3,307,624 26.58
1995 125,603 3,736,533 29.75
1996 127,140 3,842,442 30.22
1997 127,126 4,456,275 35.05
1998 138,230 3,848,059 27.83
1999 138,230 3,865,042 27.96
2000 143,961 3,947,002* 27.42
Source: Servicio de Informaci6n y Censo Agropecuario (SICA) SICA-IBRD/ MAG
Project Ecuador. through November
Company Banana Haciendas
The banana explosion in Los Rios has only been within the last 4-5 years. The
largest Ecuadorian company increased by 3,000 ha in four years. For the entire northern
district, the grand total for the same company is 5,485 ha and 4,070 workers. The south
(El Oro Province) comes in with just over 450 ha and some 400 workers, bringing the
grand total for the enterprise to nearly 6,862 ha and 5,800 personnel. The north is the
strongest production zone. Table 3-3 shows the numbers for the Agrosocioecosystem.
|
RSS
HIDE
