<%BANNER%>

Seeds of Gold

Permanent Link: http://ufdc.ufl.edu/UFE0042269/00001

Material Information

Title: Seeds of Gold the Impact of Biodiversity on Cacao Production Decisions of Small Landholder Households in Northwestern Ecuador
Physical Description: 1 online resource (87 p.)
Language: english
Creator: Blare, Trent
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: agroforestry, biodiversity, cacao, ecuador, household, shadow, wage
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Our study of Ecuadorian smallholder cacao farmers revealed that these farmers chose to use a farming method that is not most profitable, which appears to contradict the classical economic model of profit maximization. Many farmers still grow a less profitable variety of cacao, cacao nacional, in an agroforestry system instead of the more productive hybrid variety, cacao CCN-51, which is planted with few other crops and trees and uses many chemical inputs. Farmers must receive additional nonmarket value from cacao nacional such as ecological values, which induces them to raise cacao nacional. Our study uses a household model and shadow prices to capture t these nonmarket values in order to understand the full value of each production method to these smallholder farmers. Our study builds on the current models of shadow wages by not only including the biodiversity value but also showing the impact of segmented markets such as f fair trade, organic, or other markets on household income. Furthermore, the shadow wage takes into account the effect of imperfect substitution of family and market labor to better understand a household's labor decisions. The agroforestry production of cacao provides the households with ecological services as well as a source of additional income and food. When the additional value from biodiversity is included in the shadow wage of traditionally produced cacao nacional, the shadow wage for cacao nacional is greater than that for cacao CCN-51. This shadow wage difference would predict that smallholder farmers would chose to raise cacao nacional in order to maximize household utility. Thus, smallholder Ecuadorian households are acting as rational actors when choosing to raise cacao nacional as this production system maximizes household utility.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Trent Blare.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Useche, Maria Del Pilar.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042269:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042269/00001

Material Information

Title: Seeds of Gold the Impact of Biodiversity on Cacao Production Decisions of Small Landholder Households in Northwestern Ecuador
Physical Description: 1 online resource (87 p.)
Language: english
Creator: Blare, Trent
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: agroforestry, biodiversity, cacao, ecuador, household, shadow, wage
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Our study of Ecuadorian smallholder cacao farmers revealed that these farmers chose to use a farming method that is not most profitable, which appears to contradict the classical economic model of profit maximization. Many farmers still grow a less profitable variety of cacao, cacao nacional, in an agroforestry system instead of the more productive hybrid variety, cacao CCN-51, which is planted with few other crops and trees and uses many chemical inputs. Farmers must receive additional nonmarket value from cacao nacional such as ecological values, which induces them to raise cacao nacional. Our study uses a household model and shadow prices to capture t these nonmarket values in order to understand the full value of each production method to these smallholder farmers. Our study builds on the current models of shadow wages by not only including the biodiversity value but also showing the impact of segmented markets such as f fair trade, organic, or other markets on household income. Furthermore, the shadow wage takes into account the effect of imperfect substitution of family and market labor to better understand a household's labor decisions. The agroforestry production of cacao provides the households with ecological services as well as a source of additional income and food. When the additional value from biodiversity is included in the shadow wage of traditionally produced cacao nacional, the shadow wage for cacao nacional is greater than that for cacao CCN-51. This shadow wage difference would predict that smallholder farmers would chose to raise cacao nacional in order to maximize household utility. Thus, smallholder Ecuadorian households are acting as rational actors when choosing to raise cacao nacional as this production system maximizes household utility.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Trent Blare.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Useche, Maria Del Pilar.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042269:00001


This item has the following downloads:


Full Text





SEEDS OF GOLD: THE IMPACT OF BIODIVERSITY ON CACAO PRODUCTION
DECISIONS OF SMALL LANDHOLDER HOUSEHOLDS IN NORTHWESTERN
ECUADOR


















By

TRENT BLARE


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2010

































2010 Trent Blare





























To Clemencia Quishpe, mi mama Ecuatoriana









ACKNOWLEDGMENTS

I would like to especially thank my advisor Dr. Pilar Useche for assisting me from

the proposal, survey construction, data analysis, model development, and thesis writing.

I am very thankful to the Tropical Conservation and Development Program within the

Center for Latin American Studies for funding my field research from June through

August 2009. I owe a great debt to those who helped me in Ecuador collect the data as

well as house me and support me during my time in Ecuador. They also provided me

with immeasurable support and insight while I analyzed the data. I especially would like

to thank the Fundaci6n Acci6n Social Caritas, the provincial governments of Santo

Domingo de los Tsa'clias and Pichincha, Jose Chuquirima, Mauricio Eraz, Clemenica

Quishpe, and Betty Cuellar. I also would like to thank Jhonny and Eduardo Alvarez for

helping me understand some particularities of the data during the final weeks that I

wrote and defended my thesis.









TABLE OF CONTENTS

page

A C KN O W LED G M ENTS ...................... .. ................................ ......................................... 4

LIST O F TA BLES ......... ................ ..................... ...... ............... 7

LIS T O F F IG U R E S .................................................................. 8

LIST OF ABREVIATIONS ................................. ............................ ... ....... 9

A B S T R A C T ...................................................... 10

CHAPTER

1 IN T R O D U C T IO N .................................... .................................................................... 12

Cacao's Im portance to the Tropics ................................................................... 14
Cacao's Im portance to Ecuador ......................................................... ........... 15

2 V A LU E O F B IO D IV E R S ITY ........................................................................ 19

Ecological Im portance.............................. ............... 20
Pollution Control ................................. ........................... .... ....... 21
S o il E n h a n ce m e n t ........................................................................ 2 1
Carbon Sequestration ................................. .......... ............... ............... 22
Household Use Value of Biodiversity .................. ...................... 23

3 THEORETICAL FRAMEWORK ....................... ... ................... ... 25

M modeling Specialty/Fair Trade M markets ............................................ ............... 25
Current Conceptualization of Household Shadow Price Models ........................... 28
Framework for Valuing Biodiversity ...................... .... .......... ........ 29
Countervailing Market Imperfections ................................... ........ 32
Imperfect Segmented Product Markets .............................................. ... 33

4 EMPERICAL IMPLEMENTATION ......... .......... ......... ............... 34

R e s e a rc h S ite ............... ............................................... ............... 3 4
S survey D issem nation ......... .............. ........................ ............................... 37
Characteristics of Households that Produce Each Variety of Cacao .................. 41
Parcel Characteristics ................................................................ 42
Production and Profitability Values ......... ................ .................... ........... ...... 43
P e rce ptio n of P rofita b ility ......... .............. ......... ................................. 4 7
A access to S specialty M markets ......... ............ .............................. ............... 49
Likelihood to P lant C acao ......... ............. .................................. ............... 50
Perceptions of Environmental Factors ............... .............. ............ .......... ..... 50









Estim action ........................................................................... ......................... 52
Specific Estimated Equations ............................................ ............... 53
Definition of the Variables............................. 54
Results of the Estimated Equations ...................... ...................... ............... 57
S um m ary of the R esults...................................................................... ......... 6 1

5 C O N C LU S IO N .......................................... ................. ............... 79

New Conceptual Framework......................... ..... ............................. 79
Impact on Current Research ...... .... ...................... ............... 80
Need for Further Research ...................... ........................ 81

LIST OF REFERENCES ....................................................... 83

B IO G RA PH IC A L S K ETC H ............. ................. ................. .................... ............... 87




































6









LIST OF TABLES

Table page

4-1 Household characteristics ... ... ................................................ ............... 63

4-2 Parcel characteristics ...... .. ..... ......... ... ........... ............... 64

4-3 Yield, costs, revenue, and profits per hectare .............. ..... ............... 65

4-4 Yield, costs, revenue, and profits per hectare equivalent .............. ............... 68

4-5 Perceptions of profitability ............ ............. ................... 70

4-6 Environm mental perceptions ... .... ............................................... .................. 73

4-7 Variables included in production and biodiversity regressions ........... ........... 75

4-8 Coefficients of the biodiversity equation .............. ..... ........ ........... .... 76

4-9 Coefficients in the production equation.................................... ..................... 77

4-10 Shadow wage components by market..... ................................ 78

4-11 Shadow wage by variety and market........ ......................... .. .. ............ 78









LIST OF FIGURES

Figure page

4-1 Counties of Santo Domingo de los Tsa'chila, Puerto Quito, and Quininde......... 62

4-2 Perceptions of profitability ....... .... .............. ...... ................... 70

4-3 Likelihood to plant cacao ............................................................................. 71

4-4 Cacao variety planting preferences .............. ..... ........ .. ............... 72

4-5 Perceptions of biodiversity......................................................... 74











BIOFASCA

CORPEI

FASCA

FOCs

MRSLC

UOCAQ


LIST OF ABREVIATIONS

Cacao processing center ran by FASCA

Corporaci6n de Promoci6n de Exportaciones e Inversiones

Fundaci6n Acci6n Social Caritas

First Order ConditionsMPL Marginal Product of Labor

Marginal Rate of Substitution between Leisure and Consumption

Uni6n de Organizaciones Campesinas del Cant6n Quininde









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

SEEDS OF GOLD: THE IMPACT OF BIODIVERSITY ON THE CACAO PRODUCTION
DECISIONS OF SMALL LANDHOLDER HOUSEHOLDS IN NORTHWESTERN
ECUADOR

By

Trent Blare

August 2010

Chair: Pilar Useche
Major: Food and Resource Economics

Our study of Ecuadorian smallholder cacao farmers revealed that these farmers

chose to use a farming method that is not most profitable, which appears to contradict

the classical economic model of profit maximization. Many farmers still grow a less

profitable variety of cacao, cacao national, in an agroforestry system instead of the

more productive hybrid variety, cacao CCN-51, which is planted with few other crops

and trees and uses many chemical inputs. Farmers must receive additional nonmarket

value from cacao national such as ecological values, which induces them to raise

cacao national. Our study uses a household model and shadow prices to capture t

these nonmarket values in order to understand the full value of each production method

to these smallholder farmers. Our study builds on the current models of shadow wages

by not only including the biodiversity value but also showing the impact of segmented

markets such as f fair trade, organic, or other markets on household income.

Furthermore, the shadow wage takes into account the effect of imperfect substitution of

family and market labor to better understand a household's labor decisions.









The agroforestry production of cacao provides the households with ecological

services as well as a source of additional income and food. When the additional value

from biodiversity is included in the shadow wage of traditionally produced cacao

national, the shadow wage for cacao national is greater than that for cacao CCN-51.

This shadow wage difference would predict that smallholder farmers would chose to

raise cacao national in order to maximize household utility. Thus, smallholder

Ecuadorian households are acting as rational actors when choosing to raise cacao

national as this production system maximizes household utility.









CHAPTER 1
INTRODUCTION

Farmers through the world face multiple decisions from which crops to grow to

where to market their products. Although these decisions have traditionally been

conceptualized as choosing the option that is most productive and profitable, recent

research has shown that farmers include many other criteria when making production

decisions. Hilderbrand (2002) demonstrated that a livelihood model is a better method

to describe production these decisions. The livelihood system includes household

consumption and leisure needs as well as needs and production constraints. These

competing objectives may lead the household to make decisions that contradict the

popular notion of the rational profit maximizing decision maker. Even though the

household may not be maximizing profit, these households are making rational

decisions to maximize their total welfare or utility. Bechetti and Costantino (2008) found

that household welfare is very important to production decisions. They found that

farmers chose to participate in organic markets not just for the price advantage. They

also participated in these markets because of the additional welfare benefits that the

additional income provided as well as could be obtained directly from the production

such as a nutritious, safe food source.

An example of the importance of multiple objectives in a farmer's production

decisions is in the production of cacao in Ecuador. Many Ecuadorian smallholder cacao

farmers produce a variety of cacao that is less productive and profitable, which counters

traditional economic theory. Ecuadorian farmers include other factors besides profits in

their decisions. Ecological factors are very important in their decision process. These

traditional production systems use agroforestry methods and provide households with









not only economic benefits but also a source of food and medicine (Bentley, Boa, and

Stonehouse 2004). These additional benefits only occur to households that raise cacao

in diversified or biodiverse production systems. Thus, Ecuadorian cacao farmers take

into account this cropping biodiversity and all its benefits when making production

decisions.

This paper develops a household model that includes biodiversity as a part of the

decision criteria in Ecuadorian cacao producers' production decisions. Past work has

examined shadow prices to examine market failures especially in the labor market (Le

2009; Arslan and Taylor 2009; Jacoby 1993). This work adds a new aspect by

estimating a shadow wage and income that takes into account the additional value for

biodiversity that is not captured in the market. The shadow wage is derived from a

household model where a rational household chooses to maximize its utility in the

presence of market imperfections, which may be different than the choice the

household would make if such constraints did not exist. The distinctive characteristics

of this analysis reveal that such market imperfections are associated with environmental

goods, such as the ones resulting from intercropping of different crop varieties, rather

than with transaction costs (de Janvry, Fafchamps, and Sadouelt 1991) and constraints

in the labor (Jakoby 1993; Skoufias 1994; Le 2009), land and credit markets (Eswaran

and Kotwal 1995; Carter and Yao 2002).

The shadow income for agroforestry cacao production was found to be higher than

that for commercial cacao production system. This situation shows that the relevant

imperfections is associated with over-employment, rather than with under-employment

(Feather and Shaw 1999) of family labor as has been assume by most papers in the









household models literature. Thus, a rational household would chose to use the

traditional rather than the commercial system.

The production of Ecuadorian cacao provides a great case study to examine the

effect of biodiversity on smallholder household utility and production decisions for

several reasons. First, all households growing cacao participate in the cacao market as

self-consumption is very close to zero. They use market prices as signals for the

production decisions. Thus, this example allows for the disaggregation of the market

imperfections in the cacao cash crop market from the total of the market imperfections

to determine the impact of biodiversity on the shadow wage. Second, the manner in

which traditional cacao is raised in diversified agroforestry systems provides a

particularly good contrast to the manner in which modern varieties of this product are

grown. This crop is important to the economy of many developing countries including

Ecuador and other countries throughout the world that are cradles of biodiversity

(Bentley, Boa, and Stonehouse 2004; Brush, Taylor, and Bellon 1992). Finally, the

unique cacao specialty market that has developed in Ecuador for its traditional cacao

production provides a good example to study the impact these premium markets on

smallholder households.

Cacao's Importance to the Tropics

The tropics including Asia, Africa, and Latin America are known for their high

levels of poverty and biodiversity. The impact of cacao production in these regions is

large with 7.42 million hectares being devoted to this crop in 2007 that produces over

five metric tons of cacao valued over 6.75 billion US dollars (Food and Agriculture

2009). Indeed, cacao is the second most important cash cop in the tropics (Steffan-

Dewenter et al. 2007). An interesting fact about cacao is that much of its production is









achieved by small landholders who produce between 70% and 90% of the world's

cacao (Dahlquist et al. 2007). The average cacao farm size in Nigeria is just 1.7

hectares, 4.5 hectares in Ecuador, and 2.8 hectares in the Ivory Coast. These small

landholders can be very productive. For instance, farmers in Sulawesi, Indonesia have

been able to produce 2000 kg of cacao per hectare per year which exceeds production

levels that can be achieved by commercial cacao producers (Rice and Greenberg

2000). So, cacao is important in the development plans for this region because of its

ecological impact and importance to small landholders.

Cacao's Importance to Ecuador

One of these tropical countries heavily dependent on cacao production is Ecuador.

During the late 1800s and early 1900s, cacao was Ecuador's most important export with

large expanses of the coastal region devoted to the production of this crop. During this

period, cacao was referred to as "Pepa de Oro" or "Seeds of Gold." In fact, cacao

production helped the new country finance its war for independence against Spain.

However, the crop was nearly destroyed by two fungi, monilla and witch's broom. The

demand for cacao collapsed during the Great Depression. This economic crisis forced

the large cacao haciendas to be divided into parcels that were sold to small landholders.

The land reforms of the 1960s further divided these large farmers so that today most of

Ecuador's cacao is raised on small farms (Vasco 2010). Today, 90% of cacao is

produced on landholdings with less than 50 hectares and over 30% is produced on

farms smaller than 10 hectares (Coporaci6n de Promoci6n de Exportaciones e

Inversiones 2009).

Even though cacao production may not be as large of a component of the

Ecuadorian economy as it was 100 years ago, cacao continues to be important to









Ecuador. In fact, 12% of the country's economically active population is involved in

cacao production, employing 600,000 Ecuadorians (Coporaci6n de Promoci6n de

Exportaciones e Inversiones 2009). Furthermore, Ecuador controls 60% of the market

of the cacao used to produce gourmet, dark chocolates production and is one of only

three countries where this high quality cacao can be grown. The variety raised in

Ecuador is referred to locally as cacao national or cacao arriba in international markets

(Asociacion Nacional de Exportadores de Cacao 2007).

The need for economic development in Ecuador is indeed acute. Forty-six percent

of Ecuadorians live below the poverty line. Poverty is especially concentrated in the

rural areas of the country (Central Intelligence Agency 2009). The promotion of cacao

production is seen as an avenue to help alleviate this rural poverty. Because of the

importance of cacao to Ecuador's economy and especially to small landholders, the

Ecuadorian government along with local and international development organizations

has begun to advocate cacao and especially cacao national as an economic

development strategy.

Most of the cacao raised throughout the world is raised using hybrid varieties such

as the CCN-51 variety grown in Ecuador. The advantage of cacao CCN-51 over cacao

national is that it is more productive. However, cacao CCN-51 does not produce the

high quality cacao that is produced by cacao national. Cacao CCN-51 is used in lower

quality chocolates which garner lower prices (El Cacao Volvi6 Ser la Pepa de Oro

2007). Yet, cacao CCN-51 has been promoted to small landholders over the past 15 to

20 years as a superior option to cacao national because of its production advantages

(Bentley, Boa, and Stonehouse 2004). During the past 5 years, some farmers have









obtained access to specialty where they receive premiums for cacao national and

organically produced cacao (Bentley, Boa, and Stonehouse 2004). Because of these

new markets, Ecuadorian farmers are being encouraged to switch from producing

cacao CCN-51 to cacao national. Indeed, the price premium for cacao national has

expanded from nearly 20% to over 60% between 2004 and 2007 (El Cacao Volvi6 Ser

la Pepa de Oro 2007). In fact, the Ecuadorian export institution, the Corporaci6n de

Promoci6n de Exportaciones e Inversiones (CORPEI), now even requires farmers to

separate these two varieties to meet chocolate makers demand for cacao national.

In addition to the quality differences, these two varieties of cacao are raised using

different farming methods. The traditional system used to raise cacao national includes

a diverse array of crops and retention overstory trees to enhancing nutrient cycling

(Bentley, Boa, and Stonehouse 2004). Cacao is commonly planted with plantains

and/or bananas. Many farmers also plant oranges or other citrus fruit, or avocado in

their cacao plantains. Farmers include many timber species such as laurel on their

cacao parcels. One of the most popular tree species included in the cacao national

agroforestry system is locally referred to as guaba. The guaba tree produces a long

pod that contain a sweet pulp that surrounds its seeds and is an important nitrogen

fixer, providing the system with a natural fertilizer. Rather than using chemical fertilizers

and pesticides the traditional cacao production system uses mechanical methods to

control diseases and pests, such as pruning. This approach to production varies greatly

from the commercial method. Cacao CCN-51 is not as susceptible to sun damage and,

therefore, is not grown in the traditional agroforestry system but in parcels that contain









few other crops and trees besides cacao CCN-51. Cacao CCN-51 is also raised using

more agrochemicals than cacao national.









CHAPTER 2
VALUE OF BIODIVERSITY

Biodiversity has not traditionally been considered a component of smallholder

household's production decisions as it does not fit into the traditional profit maximization

models. Many of the benefits of biodiversity are not received as cash income. There

are positive externalities not realized in the market such as the aesthetic value that a

farmer realizes from their biodiverse farms. In addition, there are benefits that may

have monetary value but this value is not received directly and/or is not realized until

sometime into the future such as enhanced soil fertility, which leads to greater

production. Yet, some products of biodiversity do have a market value. For instance, a

diversified cropping system produces a variety of crops that can be brought to the

market. All these biodiversity values may influence farmers' willingness to receive lower

prices for their products as the value they receive through the positive externalities of

biodiversity may compensate for the lower price. Indonesian cacao farmers said that

they would be willing to accept a price lower than full compensation for the smaller

yields of traditionally raised cacao to grow cacao in an agroforestry system. These

farmers would be willing to receive just a 0.36 U.S. dollars premium per kg for their

cacao grown in an agroforestry system (Steffan-Dewenter et al. 2007).

Chapter 2 examines the current research on the ecological impacts of the

biodiverse cacao national agroforestry system. The values are often difficult to capture

in traditional markets. Yet, they would likely be included in a smallholder producers

production decisions as they impact the sustainability of their livelihoods. The second

section examines some of the use values such as meeting consumption needs that

these biodiverse systems provide households.









Ecological Importance

The diversified cropping system of traditional agroforestry cacao systems such as

that of the cacao national have been shown to enhance the diversity of native plants

and animals (Reitsma, Parrish, and McLarney 2001). Both the above ground and below

ground biological diversity of cacao agroforests have been found to be greater than that

of commercial cropping systems and similar to that of long term fallowed fields but still

less than primary forests (Duguma 2001). Cacao agroforestry systems in Indonesia

were found to contain forty percent of the plant and insect species found in natural

forests. This level of biological diversity can only be found when the cacao is planted in

a diversified agroforestry system. Costa Rican cacao agroforestry systems were

shown to have a diversity of bird species statistically significantly similar to that of a

natural forest. Natural forests were shown to have 144 bird species compared to 131 in

cacao agroforestry systems (Reitsma 2001). This potential to create a biologically

diverse habitat is especially prevalent in deforested areas. However, the monoculture

nonshaded cacao production practice which is the common method used in raising

cacao CCN-51 have not shown to provide similar levels of biodiversity (Steffan-

Dewenter et al. 2007).

These biodiverse cacao agricultural system can provide many of the

environmental benefits that natural forests provide. While the world's tropical

rainforests are being destroyed, one of the critical objectives of the cacao agroforestry

practice is to replicate the natural forest environment so that the beneficial qualities of

the tropical forests can be achieved in these agroforestry systems so that the loss of

natural forests can be mitigated. The protection of native plant and animal species is

particularly important to Ecuador as it is one of the most biodiverse countries in the









world with many plants and animals threatened with extinction due the loss of habitat

(Conservation International 2007). To slow this destruction, many environmental

groups are promoting the production of cacao national as a way of protecting this

habitat. The additional ecological benefits these biodiverse systems provide include

pollution and runoff control, improved soil quality through nutrient cycling and organic

matter accumulation, and carbon sequestration.

Pollution Control

As compared to raising crops in intensive monoculture production, utilizing a

cacao agroforestry system has shown to help prevent further degradation of water

resources. In addition, the use of cacao agroforestry systems can ameliorate soil

erosion which is critical in the sustainability of agricultural activity. The shade trees

utilized in the cacao agro forestry practice provide several ecological benefits. One

benefit is that the shade trees prevent some of the leaching of pesticides and other

chemicals into groundwater. The root zone of the trees can prevent these chemicals

from reaching the water supply. In addition, shaded farmer have lower run off rates

and, thus, less soil loss. The tree root zone and the natural litter form the shade trees

slow the runoff so less soil is removed. The canopy also protects the soil from the full

impact of rainfall slowing down the raindrops as they fall so as to protect the soil from

being loosened and washed away (Beer et al. 1998).

Soil Enhancement

The dense planting of trees and plants in cacao agroforestry systems provide a

large level of litter and organic matter to be recycled back into the soil to maintain its

fertility and long run sustainability. Total biomass in cacao agro forest systems in

Cameroon was higher than that in food crop fields with 305 metric tons per hectare









versus only 85 metric tons per hectare. Yet, this was much less than biomass in

primary forests, 541 metric tons per hectare, and long-term fallow, 460 metric tons per

hectare. These agro forestry systems proved to have larger amounts of organic matter

and higher levels of calcium and magnesium in the soil than secondary forests

(Duguma 2001). Cacao agroforestry systems have been shown to on average add 10

Mg of plant litter per hectare per year (Gamma-Rodriguez 2009). One study found that

soil organic matter increased by twenty one percent in a ten year period under pruned

Erythrina peppigiana and by 9% under laurel trees in fields that were converted from

sugar cane production to a cacao agroforestry fields (Beer et al. 1998). Furthermore,

soil decomposition rates and the abundance of soil arthropods were found to be greater

in shaded cacao agro forestry systems as compared to systems that provide less shade

(Steffan-Dewenter et al. 2007).

Carbon Sequestration

Cacao agroforestry systems can be a useful tool in the sequestration carbon in the

effort to control global warming. In fact, these agroforestry systems seem to be able to

store carbon at a similar rate of that of natural forests. Soil organic carbon stock in

these systems is not significantly different from that of forests in the first 100

centimeters of soil with a rating of 302 mega grams of organic carbon per hectare

(Gama-Rodrigues 2009). Another study found that cacao agroforest systems have

shown to contain 62% of the carbon stock found in primary forests (Duguma 2001).

The shade trees in the cacao agro forestry system store between 14 and 52 Mg of

Carbon per hectare in the aboveground woody biomass. In fact, this agroforestry

practice stores between 10 and 50 Mg of carbon more per hectare in the litter and soil

organic matter than annual crops. Therefore, cacao agroforestry system could prevent









the release of 1000 Mg of Carbon per hectare that would occur by planting annual crops

(Beer et al. 1998). In Cameroon mature agroforestry systems have been found to fix

150 metric tons of carbon per hectare for mature trees and between 111 and 132 metric

tons of carbon per hectare for younger cacao agroforestry systems. Even though these

levels of carbon sequestration are lower than those found in natural forests at 307

metric tons per hectare, much more carbon is sequestered in a cacao agroforest than

can be achieved in fields that only plant annual food crops (Rice and Greenberg, 2000).

The carbon that is sequestered has been proven to be a well protected form of carbon

so that it cannot be easily released to the environment (Gama-Rodrigues 2009).

Household Use Value of Biodiversity

In addition to the environmental services that cacao agroforestry systems, such as

that used to produce cacao national, provide, this production method allows these

households to meet their consumption needs. Many other products such as plantains

and fruit that are important to the diet of Ecuadorian farm families are included in these

biodiverse, traditional cacao production system. Even if a household may be able to

earn much more cash income for cacao CCN-51, exclusively pursuing this alternative

may not be the best alternative for a family as they would not be able to meet their

subsistence needs because cacao CCN-51 is generally raised in a monoculture system

without plantains or fruit trees. This tradeoff between working for cash income and

providing subsistence was shown in a study in Brazil nut gathers in the Brazilian

Amazon. Household consumption needs were not met as smallholder farmers

dedicated their time to collecting Brazil nuts instead of growing subsistence crops (Ros-

Tonen et al. 2008). Cash income may not be very important these Ecuadorian









households as many of them do not depend on the purchasing of goods in the market to

meet their family's needs.

The use value of these biodiverse production systems is shown in a study of

Cameroon agroforestry systems. This research found that smallholder farmers received

the highest level of net present value when a cacao agroforestry system included shade

trees that provide an economic benefit such as fruit and/or timber trees. This system

had a higher net present value than more commercial systems that had a few shade

trees of one variety. These additional values provide an economic incentive to raise

cacao in an agroforestry system (Duguma 2001). These diversified cacao, fruit, and

timber agroforestry systems were found to have greater financial returns than

plantations using conventional, leguminous trees. These systems also provide more

consistent income from the multiannual harvest of cacao and other fruit trees in the

cacao agroforestry system (Beer et al. 1998). Thus, smallholder households that raise

cacao national in such a diversified agroforestry system obtain additional consumption

and use values along with the nonmarket values that would make it the preferred

production method for Ecuadorian smallholder producers..









CHAPTER 3
THEORETICAL FRAMEWORK

Modeling Specialty/Fair Trade Markets

The unique market for cacao national is similar to fair trade markets in that in both

fair trade and specialty cacao national markets farmers receive a premium for their

product. Thus, the work developed to explain how the fair trade markets influence

smallholder farmers' production decisions can be applied to this situation of the

specialty cacao national market. A model built by LeClair (2002) and critiqued by

Hayes (2008) shows how the premium paid to participants in the fair trade markets

affects labor allocations in the goods for this market. The model explains how a

household would supply more labor to the production of the fair trade good due to the

fair trade price premium. He models a household's income from production as

Y = (P r)Q (3-1)

where

Y is household income,
P is the market price for the good,
r is per unit costs, and
Q is the quantity of the good produced.

The model is then transformed for the income of a household that participates in the fair

trade market.

Y = (P(1 + a)- r) Q (1 + b) (3-2)

where

a is the premium paid for the fair trade market or the specialty cacao market
b is the additional output caused by the labor increase

This model of the fair trade market can be transformed for shadow income (y*)

which is developed later in the next chapter. The shadow income includes the









additional value from biodiversity that is not included in the cash income the household

receives. Thus, the shadow income represents the true value from production for the

household. By the model developed by LeClair, shadow income in the standard market

where no premium is given for cacao is

y* = PcQc + PoQo + V total costs (3-3)

where

Pc is the price for cacao in the standard market,
Qc is the quantity of cacao produced,
Po is the price for the other crops on the farm,
Qo is the quantity of other crops produced on the farm, and
V is the value that household obtains from the biodiversity externality.

Our model alters LeClair's in order to reflect an agroforestry situation and

specifically the cacao agroforestry system. Our model is applicable to many

smallholder farmers as they commonly practiced diversified cropping systems that

provide additional ecological benefits. Our model includes the income from cacao as

well as the other agricultural products raised in the diversified system. The value the

household recognizes for biodiversity (V) is also included. In our model the cost

function is calculated according to land area. In a diversified cropping system, a farmer

cannot disaggregate the inputs utilized for one crop from another. The entire production

costs must be subtracted from the whole system rather than calculating the costs for

each unit of production for each corp. Furthermore, the application of inputs by

smallholder farmers is considered in terms of amount applied per unit of land. So, the

costs would be measure in terms of units of land than by the quantity of the product

produced.









The shadow income for the specialty market that includes the price premium (a)

and the production increase (b) from added labor would be

y* = (Pc(1 + a) Qc(1 + b)) + PoQo + V- total costs (3-5)

An important distinction between our model and LeClair's model is that the model

includes the value of biodiversity. These ecological impacts must be included to better

understand how smallholder household's production decisions are made. In addition,

the inclusion of this factor is important in the evaluation of the fair trade and specialty

markets since environmental improvement is the objective of many of these marketing

schemes in order to better understand its full impact. LeClair argued that the premium

could induce participants in fair trade markets to dedicate labor to these products thus

increasing the supply and lowering the price of the good in the general market for all

producers including those who do not participate in the fair trade markets. However, by

including the value of biodiversity in our model, an additional positive benefit from fair

trade and specialty markets must be included in agricultural markets. The premium

could induce farmers to produce goods that enhance biodiversity which would be a

great societal benefit, which contour LeClair's conclusions.

LeClair and Hayes' analysis is at a macro level that examines how the premium

from fair trade or specialty markets affect market prices and labor supply in the entire

system. The model does not examine the effects for an individual household's

participation in specialty markets. In addition, a need exists to understand the impact of

these specialty markets at the individual household level. A household level analysis

allows for an understanding of how these markets affect the household's production

decisions. Shadow prices provide a means to analyze the impact of these markets at









the household level while including the impact that biodiversity on production decisions.

The shadow wage for the sale of goods inside and outside the specialty or fair trade

markets can be compared to determine which market provides more value for the

smallholder household. In the following section, the household model is used to build a

shadow wage to examine value that smallholder households place on the various

planting options to make production decisions.

Current Conceptualization of Household Shadow Price Models

The household model provides a better explanation than the traditional profit

maximization models for the production decisions of small landholder households. The

household model as developed by Barnun and Squire (1979) integrates the household's

consumption and production decisions with the objective of maximizing household

utility. The model depicts consumption and production decisions as not being made

separately; they are made simultaneously. Unlike commercial farmers who view their

farms as a business in order to earn income to purchase products for their household

needs, smallholder farmers do not separate these consumption and production

activities. A smallholder household's consumption needs influences its production

decisions, which would not be captured in a traditional model that does not consider

these consumption and production objectives simultaneously.

A household will choose to maximize its utility through consumption and leisure.

Given a budget constraint, the value of time is compared by family members to the

value of other goods to determine how much to "enjoy life" or how much to work in order

to be able to consume goods. While the work that the family undertakes off the farm is

generally valued at a market wage with money that is earned used to purchase needed

goods, the work on the farm can be valued in monetary or nonmonetary terms. In









particular, many of the positive attributes or amenities generated from existing

biodiversity in the farm cannot be traded in markets.

Our theoretical model is built upon the model of Le (2009), who developed a

shadow wage and income using to explain imperfections in the labor market. The

household model to build shadow wages, which is the marginal product of labor (MPL),

was first used by Jacoby (1993) to explain the wage that the labors perceived. This

work was furthered by Arslan and Taylor (2009), who develop a shadow price to explain

the additional value that Mexican households place on traditional maize varieties that

are not reflected in market prices. Our model expands on this work by adding a value

for environmental factors to the shadow wage and income. Specifically, the value of

biodiversity as perceived by Ecuadorian farmers is a component of the shadow wage

and income these farmers realize.

Framework for Valuing Biodiversity

Our model is based on a farm household that maximizes a utility function of leisure

(/) and consumption (C); U(C, I; D) where D is a vector of household characteristics

(e.g. preferences, number of children). C includes goods that are produced at home (v)

or purchased in the market (c); C = c + v. The labor supply (h) is the total stock of time

for the household (T) minus leisure (I), which is divided among two activities farm labor

(L), market labor (M); T- I = L+ M = h

The production function for the farm is F(L, z; A). z is a vector of variable inputs

(e.g. hired labor, fertilizer, insecticides etc.) and A is a vector of quasi fixed assets (e.g.

land). The production function F for the diversified cacao agroforestry system is

specified as F(L, z; A) = Q(L, z; AQ) + V(L,z; Av), where Q is the cacao production and V

is the subsistence and environmental benefit derived from the production method. The









production function takes this form since cacao is commonly grown jointly with other

crops or trees (such as fruit trees, cassava, and plantains) for self-consumption. Since

the two types of crops are grown in the same parcel, farmers do not account for the

amount of inputs used for each crop. Rather, purchased inputs oriented toward

producing the cash crop have the external effect of benefitting the production of

intercropped crops and trees.

Several authors have shown that when markets are imperfect or missing,

profitability variables that are commonly taken as exogenous become endogenous.

Specifically, they become functions of household characteristics and preferences

Eswaran and Kotwal (1995) explained the case of imperfect credit markets. Carter and

Yao (2002) discussed imperfect land rental markets. This situation is due to the fact that

the farm household problem becomes inseparable. Households' production decisions

are not independent from household consumption decisions. In this inseparable

environment, AQ, is an endogenous variable in the cacao production function.

The household maximization model can be summarized as

Max U (C, I; D) (3-6)

subject to

pQ(L, z; AQ) + V(L,z; Av) = pzZ + wM + R + V(L,z; Av) C

and

M>M

where

p is the farm output price,
pz is a vector of input prices,
w are the market wages, and
R is other income.










The Langrangean of the problem can be expressed as

X = U (C, T-h; D)+ Al[pQ(L,z;AQ) pzZ + wM + R + V(L, z; Av) C] + A2(M-M)(3-7)

The first order conditions (FOCs) are

aU/ac + A, = 0 (3-8)

where

Ai = MUC

-aU/aL + A1[p (aQ/aL) +9V/9L] = 0 (3-9)

In the FOCS the p(aQ/aL) equals MPL.

p (aQ/aL) +aV/aL = (aU/aL)/(aU/aC) = wi* (3-10)

wi* is the shadow wage, which includes the additional household production

values that occur to the household outside of the market. The shadow wage is the

marginal rate of substitution between leisure and consumption (MRSLc)

(au/aL)/(au/aC) = MRSLC (3-11)

Equation 3-11 reflects the fact that the household will choose to invest time working on

the farm until the point that labor's total marginal product equates the MRSLC.

The FOCs in Equations 3-7, 3-8, 3-9, and 3-10 include the value. The contribution

of biodiversity to the model is evident when Equations 3-7, 3-8, 3-9, and 3-10 are

compared to Equations 3-12, 3-13, and 3-14, which represent the standard

methodology employed to determine the shadow wage in the household literature:

Ai = au/aC = MVC (3-11)

-8U/aL + Alp (8Q/aL) = 0 (3-12)

) = p(aQ/aL) = (aU/aL)/(aU/ac) wi* (3-13)









Previous studies ignore the biodiversity value include in the second term of

Equation 3-9. This omission leads to a bias in the estimated shadow wages, when

there are externalities derived from the production method. An important consideration

is that & in Equation 3-13 will be equal to the market wage w when labor markets are

perfect (M > M) but will be lower than w when these market are imperfect and the labor

constraint binds.

Following other household studies (Le 2009), the labor market failure is modeled

as one of limited demand for labor. When this constraint binds, &, the production part of

the shadow wage of labor, is below market wage. However, an important highlight is

that under other types of labor market imperfection a can be higher than the market

wage. For example, when family labor and hired labor are not perfect substitutes, then

the household is limited in its ability to provide farm family labor by the number of

household members. In this case, when the marginal productivity of family labor on the

farm is very high, the shadow wage of labor may be higher than the market rate.

Finally, even when the marginal value of the cacao productivity of labor is lower

than market wage, i.e. & < W, there is still potential for the total shadow wage of labor to

be above market wage. This situation occurs when there are positive benefits derived

from biodiversity in the production system.

Countervailing Market Imperfections

A positive effect of biodiversity, aV/8L on shadow wages (Equation 3-9), can be

seen as a countervailing effect, market failures, which may offset the difference

between & and w derived from labor market imperfections. The value from biodiversity

would increases the shadow wage high enough so that higher benefits are realized for









farm work relative to market work, which could explain the lack of participation of farm

households in the labor market. This hypothesis is explored in the estimations in

Chapter 4.

The term reflecting marginal benefits of biodiversity from labor used in production,

aV/aL, is difficult to measure because of both the multidimensionality of V (composed of

self-consumption, medicinal plants, shade for the cacao, education of children,

continuation of cultural identity) and the missing markets for some of its components

(such as cultural or traditional values). In this analysis, a proxy measure is used (see

Chapter 4), which allows for the estimation of a lower bound for the shadow wage.

Imperfect Segmented Product Markets

Another practice that is used in the literature in order to simplify the analysis is the

normalization of the market price p, by dividing all variables expressed in dollar terms by

an index that reflects the market price of produced goods. The problem with this

strategy in a setting of imperfect or segmented markets is that farmers often face

heterogeneous prices, depending on whether they have access to markets or not. In

this case, this type of imperfection arises because of the existence of specialty markets

with limited demand. Thus, separate shadow wages are estimated depending on

whether farmers have access to specialty markets or not.

The estimation of the shadow prices are created using data collected from

smallholder farmers in northwestern Ecuador. A description of the study site, survey

methods, and results are included in Chapter 4. This data is then used to calculate the

shadow prices according to this theoretical framework, which considers the value of

biodiversity and segmented specialty markets.









CHAPTER 4
EMPERICAL IMPLEMENTATION

Research Site

Fifty households in the Ecuadorian counties of Santo Domingo de los Tsa'chila,

Puerto Quito, and Quininde were interviewed in July and August of 2009 about their

production methods, access to alternative markets, impressions of the profitability of

their operations, perceptions of environmental quality, and satisfaction with their well

being. A stratified sampling method was utilized that included participants from the

three counties as well as from various socioeconomic classes and ethnic groups. The

sample included mostly small and medium sized farming operations with 31 (62%) of

the farmers surveyed having less than 10 hectares. Some of the most marginalized in

Ecuadorian society either through ethnic discrimination or poverty are included in the

survey. Thirteen indigenous Tsa'chila households, who until 100 years ago were the

only inhabitants of the Santo Domingo area, were included in the sample along with four

Afro Ecuadorian households. The rest of the participants in the survey are mestizos of

mixed indigenous, Afro Ecuadorian, and Spanish ethnicities. This stratified sampling

method allows the survey to address the concerns raised in a previous case study by

Nelson and Galvez (2000) that found that the most marginalized populations had little

access to the alternative cacao markets.

The counties of Santo Domingo de los Tsa'chila, Quininde (Rosa Zarate), and

Puerto Quito are located in the northern coastal region of Ecuador (Figure 4-1). These

three counties are located in three different provinces. Puerto Quito is in the Pichincha

province, Quininde in the Esmeraldas province, and Santo Domingo in the recently

created Santo Domingo province. Although there are several jurisdictions, the









economic and cultural activity of the region centers on the main trade city of Santo

Domingo, a bustling frontier city located 150 kilometers southwest of the Ecuadorian

capital, Quito. Within the last fifty years, this region was settled as the Ecuadorian

government encouraged residents of the mountainous region of the country to inhabit

these less populated areas. This region continues to receive immigrants from other

areas of Ecuador and Colombia. In fact, city of Santo Domingo is Ecuador's fasting

growing city growing with an annual population growth rate of 3.7%. Due to this

continuing influx of immigrants and transient population, the true population of Santo

Domingo is unknown. However, it was estimated in 2007 to have 330,000 permanent

residents and over 500,000 people living in the province at any one time. In 2007,

Quininde was estimated to have 150,000 habitants; Puerto Quito was estimated to have

17,200 residents. While Santo Domingo is mostly an urban county with over 70% of its

residents living in the city, Quininde and especially Puerto Quito are largely rural

counties (Provincia de Santo Domingo de los Tsachilas 2010; Municipalidad de Puerto

Quito 2009; and Municipalidad de Quininde 2009).

Besides the immigrants that moved to the area, there is a small but prominent Afro

Ecuadorian and indigenous population in the area. The region of Santo Domingo was

originally inhabited by the indigenous Tsa'chila nation that currently live in seven

communities around the city of Santo Domingo. This group was traditionally hunters

and gathers. Today, they have become farmers due to destruction of the forests where

they used to hunt. Although there are only 3,000 Tsa'chila in the Santo Domingo area,

they have had a large impact on the region and especially in the rural areas because of

their historical significance and because of the large amount of land located within their









communities (Gobierno de la Provincia de Pichincha 2003). There is a large Afro

Ecuadorian population in the county of Quininde. The province of Esmeraldas is well

known for its Afro Ecuadorian residents many of whom are descendants of slaves

during the Spanish Colonial period. Slaves had originally had been brought to Quininde

from Colombia to harvest rubber in the 1800s and stayed in the area after they were

freed. In fact, the Afro Ecuadorians were some of the first residents of this region

(Municipalidad de Quininde 2009).

This region is a particularly poverty stricken area of Ecuador. Little data is

available on the economic status of this region because of the transient nature of the

population. High poverty rates are especially evident in Santo Domingo and Quininde

were little of the city lacks access to running water and very few of the streets of these

major population centers are paved. The rural areas have even less access to these

basic services although almost everyone has been connected to the electrical grid

within the last five years. The Tsa'chila and Afro Ecuadorian populations are

particularly marginalized and often much poorer the rest of society. The symptoms of

poverty are clearly evident in these populations as they lack to means to access

adequate healthcare, education, and nutritional sources. These communities have

abnormally high rates of infant mortality, tuberculosis and malaria infection, juvenile

malnutrition, as well as low literacy rates as just more than half of members of the

Tsa'chila nation are able to read and write (Gobierno de la Provincia de Pichincha

2003).

These three counties straddle and the equator so that the region has a very warm

subtropical. These counties were once covered by lush tropical forests rich in









biodiversity. This climate is also conducive to raising many varieties of tropical crops

including plantains, passion fruit, cacao, African Palm, organs, pineapple, and coffee.

Now, much of this forest has been lost as the settlers removed the trees to build their

towns and cities and farms. Because of this rich biodiversity and extensive

deforestation, many endemic species are threatened with extinction. This threat has

promoted this region to be included within the Tumbes-Choc6-Magdalena biodiversity

hotspot (Conservation International 2007).

Agriculture is the main economic driver in these three counties. Even the urban

province of Santo Domingo, is heavily dependent on the trade of agricultural goods.

Plantains and cacao are the main crops in the Santo Domingo area with many other

tropical crops grown such as malanga, cassava, and passion fruit. Quininde has

become well known for the production of African Palm on large plantations. Much of

this county's economy depends on the harvesting and process of the seeds of this palm

for palm oil. However, most of the smallholder farmers are dependent on cacao

production as they lack extensive landholdings to raise African Palm. The county of

Puerto Quito is composed mostly of smallholder farms and has the advantage of being

at a higher altitude than Santo Domingo and Quininde so that besides growing many

the tropical crops they are able to raise some cooler weather crops. Many households

in the three counties also raise livestock. Almost every farm has a few chickens and

many have pigs. Beef cattle are raised in all three counties while many farmers from

Puerto Quito have dairy operations.

Survey Dissemination

Three groups of farmers were included in the survey. The first group includes

farmers who sold their cacao to a processing center run by a local nongovernmental









organization, Fundaci6n Acci6n Social Caritas (FASCA) in Puerto Quito. This

organization is the social arm of the Catholic Diocese of Santo Domingo. The

processing center called BIOFASCA uses the cacao purchased from the farmers to

make cacao paste which is then exported to high quality chocolate producers in Spain

and Italy. BIOFASCA purchases cacao in baba, in wet form, before it is fermented or

dried so that it can complete the processing process to highest standards. Normally,

the middleman purchases the cacao from the farmers once it has been dried and

fermented. The middleman then discounts farmers when they do not complete these

processes to the highest quality standards. The farmers receive a better price for their

cacao from BIOFASCA as their product is not disconnected. In addition, FASCA pays

for the organic certification costs for 147 farmers and then pays a premium for this

organically certified cacao. Workers at the plant purchase the cacao directly from the

farms so that the farmers do not have to pay transportation costs. An additional benefit

of the FASCA cacao project is that it also provides technical assistance to any farmer

who sells cacao to them.

I accompanied the plant's agronomist as visited farms in the region. Farmers

would call him and ask for his help, so we never knew which farms we would visit in a

given day which made the sampling of the farmers random. In addition to the technical

trips with the plant manager, I interviewed a few farmers as they came to the plant to

sell their cacao. I was able to extensively interview 10 farmers in the Puerto Quito

region which represents 20 percent of the study. I also had the opportunity to talk in

depth with the plant manager, the director of FASCA, and workers at the plant about









their thoughts and feelings about BIOFASCA as well as the cacao market and organic

and fair trade markets.

I visited several of the communities around the city of Santo Domingo de los

Tsa'cahila. I had served as Peace Corps Volunteer in the area from 2005 until 2008.

During that time, I lived in an indigenous Tsa'chila community. Thus, I had the unique

opportunity to interview s 15 Tsa'chila households. I also visited a few communities of

the other communities I had previously assisted to plan a cacao conference in 2007. I

stayed with a farm family in the village of Roca Fuerte for several days as a farmer took

me to interview his neighbors. The interviews in the Santo Domingo area provided me

with opportunity to talk with farmers who were not members of farmers associations nor

had access to the organic and fair trade markets as the farmers. As Santo Domingo

has by far the largest population of the three counties involved in my sample, the

number of surveys conducted in this province is much greater than that of the other two

with 30 farmers being interviewed or 60% of the total of those surveyed.

The final group of farmers that I visited was a group that participated in a larger

farmers' cooperative near the city of Quininde that represents an association of nine

smaller farmers' associations throughout the county of Quininde officially called Rosa

Zarate. The larger farmers' cooperative, the Union de Organizaciones Campesinas del

Cant6n Quininde (UOCAQ), is building a cacao collection center similar to BIOFASCA

process plant that control the drying and fermentation processes in order to so produce

the highest quality cacao for export. They are currently working with Nestle, which has

offered to pay them up to 35 U.S. dollars more per quintal (100 pounds) for organic,

cacao national if all the quality standards are met. This premium is quite large since









cacao generally does not sell for more than 100 U.S. dollars per quintal. In addition,

Nestle is working with farmers to raise improved cacao national trees developed

through the assistance of French agronomists. Besides selling to Nestle, the

cooperative is also examining the possibility of selling their cacao in fair trade markets.

While in the Quininde area, I traveled to several villages with the director of

UOCAQ and an Ecuadorian agronomist, who works for Veterinarians and Agronomists

without Borders. This international NGO has been supporting UOCAQ in the

construction of the collection center as well as providing the farmers with technical

advice as they work to become organically and fair trade certified. The agronomist

provided the farmers with production and marketing advice as well as listened to their

concerns about the project. I was able to interview the farmers from the various

farmers' cooperatives as we visited their farms. I never knew which farmers I would be

interviewing until I arrived in each community. I interviewed 10 farmers in the Quininde

area, which represent 20% of those surveyed.

The results of the survey are included in the following sections. The first section

describes the basic characteristics of the households in the survey, the second includes

the characteristics of the parcels of cacao national and cacao CCN-51, the third

examines the profitability of each variety, the fourth describes the farmers perception of

profitability of cacao, the fifth describes farmers access to specialty markets, the sixth

farmers likelihood to plant more cacao, and the last descriptive section details farms

environmental perceptions. The last section is devoted to the creation of equations to

the estimate the shadow wage and income.









Characteristics of Households that Produce Each Variety of Cacao

There are distinct differences between those households that that raise cacao

national and those that raise cacao CCN-51 (Table 4-1). Households that produce

cacao CCN-51 are wealthier than households that raise cacao national. Household

incomes were recorded on a scale form one to nine. Household's that listed an income

of one earned less than 100 U.S. dollars per month and those who said their income

was over 800 U.S. dollars were listed with a nine. The average rating of total household

income for families that raise cacao CCN-51 was 4.87 with a median of five while

households that raise cacao national had an average score of 2.44 with a median of

three. As households in both categories are statistically similar in size, household

incomes can be compared on a one to one basis. So, there is a difference of at least

200 U.S. dollars between the total income that most households that raise cacao CCN-

51 and those that raise cacao national. This difference is especially large in Ecuador

as the minimum wage in Ecuador in 2009 was 218 U.S. dollars per month and the

poverty line was 56 U.S. dollars per person per month. Not only is average total income

statistically different between the two types of households, but average farm income is

also statistically different although the difference is smaller. Households that raise

cacao CCN-51 on average earn nearly 100 U.S. dollars more in farm income than those

farmers who raise cacao national.

Other indicators of household wealth reveal a significant difference between

households that raised cacao national and those who raise cacao CCN-51. Those

households that raise cacao CCN-51 have larger homes and farms than households

that raise cacao national. CCN-51 households have farms that are on average 28.44

hectares compared to 13.08 hectares for families that raise cacao national. In addition,









households that raise cacao CCN-51 are more likely to own a car and have running

water as compared to farmers who raise cacao national. Fifty percent of CCN-51

households own a car as compared to just 15.15% of households that raise cacao

CCN-51. Twenty-five percent of households that raise cacao CCN-51 have running

while just 6.25% of cacao national households have this service.

Parcel Characteristics

The parcels where cacao national is raised are larger, have older trees, and have

a richer diversity of other crops and trees than parcels planted with cacao CCN-51

(Table 4-2). Cacao national trees are statistically significant older than cacao CCN-51.

As cacao national has been raised in Ecuador for over 150 years and cacao CCN-51

for only 10 years, the cacao national parcels would be expected to be older. The

average age of cacao national is 10.94 years old and for cacao CCN-51 is 3.94 years

old. Although households that raise cacao CCN-51 have larger farms, these

households plant their cacao on smaller plots that are more densely planted. The

average parcel size for cacao CCN-51 parcels is only 2.57 hectares while cacao

national plots have an average size of 4.50 hectares with cacao CCN-51 trees being

planted at a density of nearly 700 trees per hectare while cacao national trees were

planted at a density of about 500 trees per hectare. The fact that cacao national

farmers plant their plots with fewer cacao trees allows them to plant other crops and

trees dispersed between their cacao trees. In addition, cacao national requires more

shade than cacao CCN-51 so more shade trees must be planted in these traditional

cropping systems.

The survey results confirm the observation that the cacao national parcels are

raised in diversified agroforestry systems while cacao CCN-51 is raised in less diverse









systems. Farmers, who participated in the survey, were asked how many varieties and

which varieties of crops and trees they planted in their cacao parcels. This study

considered that the more variety of species of plants that were included in a parcel the

more biodiverse the parcel. Both native species as well as introduced crops were

included in this biodiversity measure. Cacao national farmers plant over two varieties

of perennial crops with their cacao and at last one other tree variety while cacao CCN-

51 are most likely to plant only one other perennial crop with their cacao. Thus, cacao

national is raised in more biodiverse cropping systems that may provide additional

ecological benefits.

Production and Profitability Values

Our survey results reveal a clear difference between cacao national and cacao

CCN-51 in the use of labor and other inputs, yield, and profits from each production

system. Households that raise cacao national are more dependent on family labor and

utilize more labor overall for the production of cacao. Comparing the two cropping

systems on a per hectare basis does not reveal the true difference between the two

systems as cacao CCN-51 is planted much more densely than cacao national. In order

to make an equal comparison, the two varieties need to be compared as if they were

planted at the same density per hectare. A useful comparison can be made if the parcel

area is converted to the density recommended by the agronomist at BIOFASCA, 625

trees per hectare. The hectare equivalent makes for a better comparison than on a per

hectare basis for costs and profits between the two varieties. A parcel with less a tree

density of less than 625 cacao trees per hectare would have a hectare equivalent

smaller than the actual hectare size while a parcel with more than 625 trees would have

a hectare equivalent value larger than the number of hectares in the parcel. Basically,









the hectare equivalent designation allows each variety to be compared on a tree for tree

basis. This term, hectare equivalent, is utilized throughout this study to reference this

conversion. Table 4-3 lists the production costs, yield, revenue and profit on a per

hectare basis while Table 4-4 lists these variables on a hectare equivalent basis.

The traditional cacao national production is much more labor intensive, less

productive, and less profitable than cacao CCN-51. There is not a difference in the

amount labor utilized on cacao CCN-51 production systems and the amount utilized on

cacao national production systems when compared by labor use per hectare with an

average of 33.62 days per hectare used to produce cacao CCN-51 and 33.96 days per

hectare to produce cacao national. However, there is a significant difference in labor

use when compared by hectare or equivalent with cacao national production systems

using 63.77 days of labor per hectare equivalent while the cacao CCN-51 production

system uses 28.75 days of labor per hectare equivalent. This difference in labor use by

hectare equivalent is a better measure of input use as it compares an equal number of

trees. In particular, there is a significant difference between the days of family labor

used with 37.72 days of labor per hectare equivalent in the cacao national production

system compared to an average of 16.16 days per hectare equivalent to produce cacao

CCN-51. No significant difference was found to exist in the use of hired labor for the

production of cacao CCN-51 and cacao national.

A large difference exists in the use of other inputs such as fertilizers, insecticides,

and herbicides between cacao national and CCN-51 production systems. The cacao

CCN-51 system uses an average of 51.50 U.S. dollars in other inputs per hectare

equivalent compared to 28.78 U.S. dollars per hectare equivalent for cacao national









parcels. Households that raise cacao CCN-51 substitute other inputs for labor

especially herbicides. Weeding is one of the most labor intensive agricultural activities

of the households as it involves going to the fields and cutting the thick weeds with a

machete. There is a clear difference in amount of labor used for weeding. Cacao

national production uses 29.14 days of labor per 625 trees and cacao CCN-51

production utilizing 9.24 days of labor per 625 trees. CCN-51 production uses

chemicals to control these weeds while farmers who raise cacao national are use much

more labor to control their weeds.

The market price for hired labor does not properly represent the cost the

household realizes for utilizing this labor. First, many of the household members would

not be able to participate in these markets such as woman and children. With few

options for employment, the opportunity costs of many members of the household

would be much less than the market wage. As many household members do not have

any other option for employment, their opportunity cost would be near zero. Second,

the household is really paying itself for this work. In order to try and capture the real

cost for family labor, the costs and profits were calculated both as if family labor were

valued at the market rate and at zero to examine the differences in the two values. The

true cost to family labor is likely located somewhere between these two values. When

family labor is valued at the market rate, the total costs for raising cacao national is

much higher than that for cacao CCN-51. On average cacao national costs 564.61

U.S. dollars per hectare equivalent to grow while cacao CCN-51 costs on average

304.40 U.S. dollars per hectare equivalent. When family labor is valued at zero, then

there is no significant difference in the cost to raise either variety, cacao national









production costs average 258.40 U.S. dollars per hectare equivalent and cacao CCN-51

averages 165.14 U.S. dollars per hectare equivalent.

Most farmers measure their production on per hectare basis which reveals that

cacao CCN-51 is much more productive than cacao national. In fact, higher yields are

the main reason why cacao CCN-51 is touted as a preferred alternative to cacao

national (Melo 2009). The study confirmed this perception that cacao CCN-51 has

significantly higher yields. The farmers that participated in oyr study revealed that the

average yield for cacao CCN-51 between September 2008 and August 2009 was 16.41

quintals of cacao per hectare compared to 8.97 quintals of cacao per hectare for cacao

national. These values are very near the BIOFASCA's estimates of annual cacao

production for cacao national and cacao CCN-51 which is 8 quintals for cacao national

and 16 quintals for cacao CCN.51. However, this measurement of yield is not a true

measure of cacao production on these farmers' fields as the tree density is much

different for each variety. A much better measure would to compare the yield a tree by

tree basis, which our study does by using the established hectare equivalent

conversion. Using this measure average yield for cacao CCN-51 during this time period

was 14.80 quintals per hectare equivalent compared to 11.66 quintals for cacao

national per hectare equivalent. These values are not significantly different.

Households receive significantly larger revenue for cacao CCN-51 than for cacao

national. In standard markets where farmers do not receive price premiums for cacao

national, households who raised cacao CCN-51 received on average 1,393.97 U.S.

dollars per hectare equivalent for their cacao in September 2008 through August 2009

while the average revenue for cacao national was significantly smaller at 834.88 U.S.









dollars per hectare equivalent. Households that raise cacao national and had access to

markets, which pay a premium for this variety, received 1,060.86 U.S. dollars per

hectare equivalent. The average revenue for organic cacao producers was 1,026.95

U.S. dollars per hectare during the year studied. Cacao CCN-51 also proved to be

significantly more profitable than cacao national in both the traditional commercial

market and when a premium was received for cacao national. There was a significant

difference between the revenue of cacao CCN-51 and cacao national even when the

value of family labor was discounted. The average profit for cacao CCN-51 was

1,223.84 U.S. dollars per hectare equivalent and the average profits for cacao national

are 608.65 U.S. dollars per hectare equivalent and 677.90 U.S. dollars per hectare

equivalent in the specialty markets.

Surprisingly, the few households that raise organic cacao are not profitable. When

family labor input is included, the organic farmers had an average loss of 182.32 U.S.

dollars per hectare equivalent and only an average profit of 233.57 U.S. dollars when

family labor is not valued. This result is surprising as organic cacao has a higher

market value than cacao sold in either the traditional commercial market or the premium

paid for cacao national. Farmers stated that the average organic cacao price ranged

from 90 to 110 U.S. dollars per quintal. This price is much higher than the price the

farmers received in the specialty markets, 75 to 100 U.S. dollars per quintal, and in the

standard markets, 60 to 90 U.S. dollars per quintal. Smaller yields and greater labor

costs would make organic cacao much less profitable.

Perception of Profitability

One section of oursurvey asked farmers to rate the profitability of their cacao

parcels from very good to very bad for their cacao national and cacao CCN-51 parcels









and their farmers as a whole. The questions were asked with a time element as

farmers rated the profitability of these three entities six years ago, three years ago, now,

and three years in the future. A time element was included in order to try and measure

sustainability. These scores were converted from one to five with the very good ratings

given to a five and the very bad identified with one. These values were then averaged

to determine the farmers' relative perceptions of profitability of each activity (Table 4-5

and Figure 4-2). Farmers on average gave significantly higher profitability ratings for

both varieties of cacao than for their entire farm. Although the differences in

perceptions of profitability for cacao national and cacao CCN-51 are not significantly

different, the results show that famers believe that cacao national was more profitable

than cacao CCN-51 six and three years and that it will be in the future. However, they

see cacao CCN-51 as being more profitable now. This observation may be due to the

fact that the average age of cacao CCN-51 in the survey is only about four years. Since

cacao provides its first harvest at about three years of age, cacao CCN-51 was not

productive during this period. As cacao CCN-51 is more productive than cacao

national, farmers rate this variety as more profitable. They may believe that cacao

national prices will be higher in the future as new markets develop for this variety.

Thus, they gave this variety a high profitability rating for the future. As households are

more dependent on family labor for production cacao national which may have less

opportunity costs than market labor, farmers may view cacao national as less costly,

thus, more profitable. Furthermore, as the market has responded by providing price

premiums for cacao national and organic cacao, farmers seem to believe that they will

receive much higher prices for cacao national in the future so that it will be more









profitable than cacao CCN-51. They seem to believe that the price advantage will make

up for the production differences.

Access to Specialty Markets

Our survey included questions about farmers' access to specialty cacao markets.

The farmers were asked if they had access to markets were they are paid premiums for

cacao national or markets for organic cacao. If they lacked access to these markets,

they were then asked if they would like to have access to these markets and what

factors inhibited them from having access to the more lucrative markets. Only 26% of

households had access to the premium markets for cacao national. Of those

households that did not have access to these markets, 42% of households would like to

have access, 22% are unsure if they would like to access these markets, and only 6%

were not interested in having access in these markets. Of those who did not have

access, 43% were in the process of trying to gain access to the market, 35% were

unaware as to how to access this market, 17% claimed that they did not produce

enough cacao to sell the markets, and 5% thought that there was no demand for in this

market. Even fewer households had access to organic markets, 6% of farmers having

access to these markets. Yet, 58% of households would like to have access to organic

markets. Of these farmers that would like to have access 79% listed that cost and

knowledge about the markets prohibited their entry into this market. These results

complement the results of the case study by Nelson and Galvez (2000) which found that

many cacao farmers lacked access to fair trade markets. They found that the most

remote, marginalized households lacked access to these more lucrative markets.









Likelihood to Plant Cacao

A good indicator of household long term expectations for cacao is household's

intentions to plant cacao. The farmers that participated in our survey were asked if they

would like to plant more cacao trees, which variety they would like to plant, and what

factors inhibited them from planting cacao (Figures 4-3 and 4-4). Overall, they had high

expectations for cacao as 88% of families would like to plant cacao and of those who

would like to plant cacao 65.9% would like to plant cacao national. This result appears

to contradict the traditional theory that households act as profit maximizing rational

actors. A profit maximizing household should not choose to plant cacao national

instead of cacao CCN-51 as cacao CCN-51 is more profitable. These planting

preferences reveal that smallholder household's decisions are based not only on profit.

These households would appear to receive additional value from cacao is captured in

the market, which would induce to plant cacao national.

Perceptions of Environmental Factors

The section of our survey that examined environmental perceptions was

conducted in the same format as that of the profitability perceptions. The farmers rated

these perceptions over the past six years and into the future. A one to five scale of

measuring the responses from very good to very bad was also used. The farmers were

asked to rate their perceptions of water quality, soil fertility, and biodiversity of their

entire farms as on their fields of cacao national and cacao CCN-51 (Table 4-6 and

Figure 4-5). Farmers clearly perceived cacao national parcels to have better soil

quality over the entire period than the entire farm and that of cacao CCN-51. Cacao

CCN-51 was perceived to have better soil quality than the farm now and the future.

One can question if farmers plant cacao in parcels that have higher soil quality than the









rest of their crops. Then, the higher soil quality ratings would be due to planting

decision rather than if cacao does indeed improve the soil quality. This question is

refuted when the ratings for cacao CCN-51 are observed and the age of the trees are

considered. The farmers in our survey, as well as those who were early included in the

study by Bentley, Boa, and Stonehouse (2004), had recently planted cacao CCN-51.

They had planted this variety of cacao within the last ten years and the majority had

planted it within the last five years. The farmers rated the soil quality of parcels planted

with cacao CCN-51 to be worse than that of the entire farm six and three years ago

when the seedlings would have recently been planted which would disapprove that

cacao is planted in better soil. Farmers may have chosen to plant cacao in poorer than

average soil. The reality that cacao CCN-51 parcels have higher soil quality ratings

than the farm now and the future while have lower ratings in the past would

demonstrate that cacao may improve soil quality.

For biodiversity perceptions, cacao national has a higher average rating than the

farm in all time periods and cacao CCN-51 has nearly equal ratings as the entire farm.

The farmers are nearly unanimous in stating that they had witnessed a large loss of

native plants and animals on their farms over the last few years. However, the loss is

rated to be less for cacao and statistically significant less for cacao national. These

findings support the ecological research that claim that cacao agroforestry systems

protect and even enhance biodiversity in comparison to other cropping methods. Yet,

the farmers believe that the loss of plants and animals is so grave that the agroforestry

system may not provide enough benefits to prevent a general decline biodiversity on the









farms. Cacao agroforesty systems appear to a second best solution as compared to

native forests in controlling the loss of native plant and animal species.

The descriptive statistics from our survey as well as the research on the ecological

impact of cacao agroforestry systems reveal the importance of nonmarket values in the

production of cacao. Smallholder households utilize these values in making production

decisions. The shadow wage and income as constructed in Chapter 3 provide a solid

framework for analyzing these additional values that are included in a smallholder

producer's decisions. The following section uses our model in Chapter 3 to calculate a

shadow wage and income for cacao producers in Ecuador that includes the value of

biodiversity and accounts for the segmented specialty markets.

Estimation

In order to estimate the shadow income of Cacao Nacional, the shadow wage

defined in Equation 3-9 are used. The budget constraint can be transformed to a linear

budget constraint while allowing the household to still arrive at the optimal choice

(Jacoby 1993, Skoufias 1994, Le 2009). The transformed maximization problem is:

max U(C, I; D) (4-1)

subject to

C w*h = y*

where y* is the shadow income

y*= p pzZ + wM + R + P- w*h (4-2)

pQ describes the market access while captures longer term value of household

production including the value the household places on the ecological benefits of

biodiversity. w*h provides the shadow income from work.









Solving the maximization problem and after estimating w*and y*; the optimal labor

supply function can be estimated:

ho= (w* y*; A) (4-3)

In our study, the focus is on the difference in shadow income across households

that have differential access to specialty markets and who face different labor

constraints (Equations 3-9, 4-1, 4-3). First, w* is estimated using the method of

Skoufias(1994) but extending it with an estimated value for biodiversity. A Cobb

Douglas production function for cacao and biodiversity (Q(L, z; AQ) and V(L, z; Av)) is

used to estimate them in logarithmic form in order to make them additive and to recover

estimates of the input elasticities. Multiplying the labor elasticity in each equation (,/LQ,

,/LV) by the corresponding predicted average product of labor (O/L, VIL), gives us the

marginal product of labor (MPL). In order to obtain w*, MPL is multiplied by the price to

be received for the cacao, p.

As the household's utility increases by both incremental income as well as benefits

from biodiversity, its decisions are taken by jointly considering these benefits. For this

reason, the equations that predict cacao production and biodiversity are solved

simultaneously. By discovering the additional value that a household realizes through

biodiversity, the shadow wage, which captures this nonmarket additional value as well

as the production value from cacao, can be estimated.

Specific Estimated Equations

Biodiversity = 3o + 31 Log family labor + 32 Nacional*Log family labor + 33 Log
hired labor +34 Log other inputs + 35 Flat + 36 Hilly + 37 Insecure land rights + 38
Years land owned + 39 Ethnicity + 310 Household size + E1 (4-4)









Production = ao + al Log family labor + 02 Nacional*Log family labor + 03 Log
hired labor +a4 Log other inputs + a5 Flat + 06 Hilly + 07 Insecure land + a8 Years
land owned + 09 Ethnicity + aio Household size + E2 (4-5)

Definition of the Variables

The variables in Table 4-7 were included in both regression equations that predict

the number of varieties of cacao as the descriptive statistics, past research, and/or field

experience suggest that these variables may be significant factors in determining cacao

production and amount of biodiversity in these cacao parcels. As in any production

function, labor and other inputs are needed to produce a good. These variables are an

important component of the regression.

Labor is divided into family and hired labor in order to examine the effect of the

shadow wage and to determine the substitutability of family labor and hired labor. If

hired labor and family labor have the same MPL which the coefficients represent, then

they would be perfect substitutes and in a perfect market the MPL for both family and

hired labor would equal the wage. All the labor coefficients are expected to be positive

values for both equations as more labor would lead to higher production. They would

not be positive if the household was at the point of maximum output with diminishing

marginal production, which would be very unlikely. A similar conclusion can be made

about the variable for the other inputs. This coefficient would be expected to be positive

for production although it might not be for the biodiversity equation. The production of

cacao CCN-51 is more input intensive than cacao national as shown by the survey

results. Since cacao national is produced on more biodiverse parcels than cacao CCN-

51, a negative correlation would be expected to between input use and biodiversity.

The labor value for cacao national is included in order to determine if a difference

exists in the effect of the labor applied to cacao national on biodiversity and on









production as compared to that of the other labor variables. Since cacao national

production systems are more biodiverse than CCN-51 cropping systems, the coefficient

of family labor in cacao national production in the biodiversity regression would be

expected to be larger than the coefficient for family labor for all cacao production. As

cacao CCN-51 has higher yields than cacao national, the family labor and hired labor

coefficients would be expected to be larger than the coefficient for family labor for cacao

national alone in the production regression.

The variables flat and hilly are included in the regressions. The steep slopes

would be expected to be the most biodiverse parcels. Farmers tend to plant the flat

areas first as they are easier to maintain and harvest. To prevent erosion the farmers in

this area often leave the steepest slopes forested as the steep areas are often on

stream and river banks. Thus, the steeper areas would be more likely to be forested

and thus more biodiverse than other areas. The coefficient for flat would be expected to

be a large negative number while the coefficient for hilly will be a smaller negative

number.

The variable for insecure land rights definitely would be expected to have an

impact on the biodiversity regression. Households who fear that their property rights

may be removed would only invest in crops with short harvests such as annual crops

instead of trees, which do not provide a revenue for several years. Thus, the insecure

land rights variable is expected to have a negative coefficient in the biodiversity

equation. The variable for the initial land size is a measure of the wealth of the

household. A household's wealth would determine its ability to invest in its farm in order

to increase production. A positive coefficient would be expected for this variable in the









production equation. The length of time that someone owned property would influence

the biodiversity equation. The longer a household has owned land the more it would be

expected to invest in trees and other crops with long term return. So, they would be

expected to have larger biodiversity coefficients. Since cacao becomes more

productive over time, the coefficient for the length of ownership of the parcel would be

expected to be positive for the production regression.

Ethnicity is an important variable in the equation. This variable is included as a

dummy variable distinguishing between those households that do and do not have a

member of the Tsa'chila nation as its head. There is a significant population of

indigenous Tsa'chila who farm large areas of the region. The Tsa'chila have a distinct

culture from the rest of the population. They believe that all plants and animals have

spirits and need to be protected. This belief would suggest that they would be more

likely to protect the various varieties of plants and animals on their land. In addition,

they use many medicinal plants that they grow in their fields which would add the

diversity of plants in their fields. So, the ethnicity coefficient in the biodiversity

regression would be expected to be a positive but may be negative in the production

regression as Tsa'chila farmers do not have as much training in modern agricultural

practices.

Finally, the household size variable would be expected to be an important variable

in the equations. The coefficient for these variables would be expected to be positive

for both the biodiversity regression and the production regression. The household

would need to produce more to provide for the needs of additional household members









and would be more likely to plant additional crops in its plots to provide for its

consumption needs.

Results of the Estimated Equations

The estimated coefficients for the variables determining production of cacao and

related biodiversity are presented in Tables 4-8 and 4-9. The biodiversity equation

predicts the number of species of crops or trees in a parcel while the production

equation predicts the amount in quintal of cacao produced. The most important input

used for production of cacao on these farms appears to be family labor, whose log has

a coefficient of 0.35 in the cacao production equation. Family labor, though, does not

have a differential impact by type of cacao, whether national or CCN-51, on the cacao

production equation. However, it does have a differential impact on the biodiversity

equation. The interaction coefficient between family labor and a dummy for cacao

national is not significant in the production equation, while it is in the biodiversity

equation. The effect of labor on biodiversity production is significant for cacao national,

but not for cacao CCN-51. This conclusion is inferred from the positive significance of

the interaction term on biodiversity and the lack of significance of the family labor term

that is not interacted with cacao national on biodiversity. Hired labor and other inputs

have smaller coefficients, which are not significant for either type of production.

Land and several aspects related to land ownership are also important

determinants of both biodiversity and cacao production. For example, the surface of the

land, whether flat, hilly or steep, has a significant impact on both equations, although

opposite signs. Flatter land influences productivity in a positive manner, relative to steep

or hilly areas of land. The opposite is true for biodiversity production. Individuals

confined to produce on steeper plots have a higher number of varieties in their farms.









This result confirms the observations that farmers leave the steeper areas forested to

prevent erosion. A similar effect is observed through the initial amount of land that a

farmer started cultivating. This coefficient suggests that less wealthy farmers have more

biodiversity. Less wealthy farmers may be more dependent on their farms for

subsistence and need to plant additional varieties in the plots to meet these needs. The

number of years that a household has owned their land is positively correlated with both

cacao production and biodiversity. Farmers who own their land longer have had more

time to invest in their land in long term crops, enhancing crop diversity.

Our results support the conservation literature, which suggests that insecure

property rights have a strong negative effect on biodiversity. However, insecure property

results did not significantly affect cacao production. The result may be due to the fact

that the sample is composed mostly of low income farmers with small noncommercial

farms. The difference in access to credit because of land insecurity (Eswaran and

Kotwal 1995), which affects productivity, is minimal relative to other farmers with land

title. Those households with land title have very little extensions of land. The lack of

land for collateral would also limit these farmers access to credit.

Finally, the variable indicating whether the household head belongs to the

indigenous Tsa'chila group or not, indicates that households with this type of ethnic trait

have more biodiversity but less cacao production, relative to other households in the

sample. This results supports the hypothesis as Tsa'chila for cultural reasons chose to

leave their land forested and also would produce less because they have less

knowledge about modern agricultural practices. The final variable included in the

regressions is the household size. Larger households would need more resources to









sustain themselves. Thus, a larger household would be expected to plant more

plantains and other crops needed to provide the household with food.

Shadow Wage Results

From the family labor coefficients in the previous section, the shadow wage is

calculated for a day of family labor as described in the estimation section. Our results

are presented in Table 4-10. The terms in the first column, WQ* and wv*indicate the two

different components of our shadow wage (Equation 3-9): WQ*= p(aQ/aL) and

wv*=aV/aL. These two terms are not added directly, since the two dependent variables

have different units. The first shadow wage represents the value of an extra quintal of

cacao that will be produced with an extra day of labor, which depends on which market

it is being sold on. The second is the value of the amount of an extra-variety that can be

produced with an extra day of labor. Column two presents the marginal value of labor

for all cacao and column three the marginal value that is to be added (or subtracted if

the coefficient is negative) if the variety is national. In the table, N.S. stands for values

that are not significant. Pv is the hypothetical value of the respective variety of crop

planted in the parcel along with cacao; and 0.01 is the marginal productivity of labor

when the cacao variety is national, the traditional variety.

Column two reveals that the only the shadow wage component which applies for

all cacao production is higher than the mean of the wage obtained in the market (8.33

U.S. dollars). This shadow wage can be higher than the market wage when hired and

family labor are not substitutes and family size limits the available family labor.

However, it is still in the range of high market wages (the range of all daily wages is

between 7 U.S. dollars and 12 U.S. dollars). A family selling their cacao in the specialty









market will obtain about 1.5 U.S. dollars more per day of labor than a family who does

not have access to this market (11.90 U.S. dollars compared to 10.40 U.S. dollars).

The value of biodiversity or of adding an extra variety consists of multiple market

and nonmarket components. Some of these, such as self-consumption or production of

substitutes for medicinal plants may be easier to estimate; but others, such as

ornamental or ritual use, are at best difficult to estimate. In this sense, the variable Pv is

"hypothetical". The third column of Table 4-10 shows that there is a "biodiversity

premium" that is added to the shadow wage of family labor if cacao national is

produced. However, a lower bound can only be created for this value based on the

market price of goods for self-consumption, which will apply for specific cases. For

example, if cacao national is intercropped with cassava, producing 50 quintals per

intercropped hectare, at 7 U.S. dollars per quintal, the marketable MPL for this variety

would be .49. The total shadow wage of intercropped cacao would be about 3 U.S.

dollars higher. If cacao national is intercropped with oranges, the shadow wage would

be at least 1.10 U.S. dollars higher (for 4000 Units/ha at 0.04 U.S. dollars per unit).

Added to this shadow wage would be benefits like shade, soil enhancement,

ornamental and cultural value, medicinal and ritual uses, etc.

Based on our analysis, Table 4-11 shows that the shadow price for a day of work

in the production of cacao CCN-51 is equivalent to a high market wage. Furthermore, a

day of work for producing cacao national may be higher than a high market wage once

the additional biodiversity value is included. The shadow wage is particularly higher than

the market wage for cacao national producers who have access to specialty markets

because of the cash premium they receive. Through the analysis of the shadow wage,









which includes both production and biodiversity values, cacao national provides more

value to smallholder households than cacao CCN-51.

Summary of the Results

Our survey of cacao producing households in northwestern Ecuador produced a

result that would seem to contradict conventional economic theory. Households would

prefer to raise cacao nacaional instead of cacao CCN-51 even though cacao CCN-51 is

more profitable. Our results also revealed that the cacao national production system

provides some benefits that are not captured in the market such as soil enhancement

and the protection of native species of plants and animals. The shadow prices provide

a manner to include these additional values that the household obtains form cacao

production. The shadow wage is higher for cacao national that it is for cacao CCN-51

and much higher when farmers sell in specialty markets. The shadow wage difference

demonstrates that these households receive more value from cacao national than

cacao CCN-51. The households act rationally when they chose to plant cacao national

instead of cacao CCN-51.








_RE P 1BLI CA DEL


Figure 4-1. Counties of Santo Domingo de los Tsa'chila, Puerto Quito, and Quininde


E C UA DOT









Table 4-1. Household characteristics
Cacao CCN-51 Cacao national
Total income per month1
Mean 4.87 3.44t
Median 5 3
Number of observations 15 34

Farm income per month1
Mean 3.81 2.85t
Median 3 2
Number of observations 16 34

Percentage indigenous 31.25 29.41
Number of observations 16 34

Family size
Mean 4.75 4.09
Median 5 4
Number of observations 16 34

Farm size (ha)
Mean 28.44 13.08t
Median 11 8
Number of observations 16 34

Percentage that own a car 50.00 15.15ttt
Number of observations 16 33

Percentage with running water 25.00 6.25t
Number of observations 16 32

Rooms in the home
Mean 4 3.25"
Median 4 3
Number of observations 16 32

tSignificant difference at the 10% level
" Significant difference at the 5% level
" Significant difference at the 1% level









Table 4-2. Parcel characteristics
Cacao CCN-51 Cacao national
Parcel area (ha)
Mean 2.57 4.50"
Median 2.00 3.00
Number of observations 18 51

Tree density (trees/ha)
Mean 702.78 502.18t
Median 670 600
Number of observations 18 51

Tree age (years)
Mean 3.94 10.49tt
Median 2.00 6.00
Number of observations 18 51

Slopea
Mean 1.83 1.82
Median 2.00 2.00
Number of observation 18 51

Number of annual crops per parcel 0 0.25t
Mean 0 0
Median 18 51
Number of observations

Number of perennial crops per parcel
Mean 1.33 2.65ttt
Median 1.00 3.00
Number of observations 18 51

Number of trees varieties per parcel
Mean 0.61 1.08t
Median 0 1.00
Number of observations 18 51

tSignificant difference at the 10% level
t Significant difference at the 5% level
ttt Significant difference at the 1% level
a 1:Plain, 2:Hilly, 3:Steep









Table 4-3. Yield, costs, revenue, and profits per hectare
Cacao CCN-51 Cacao national
Total labor (days/ha)
Mean 33.62 33.96
Median 24.85 28.27
Number of observations 18 48

Family labor (days/ha)
Mean 18.67 22.72
Median 14.52 18.00
Number of observations 18 49

Hired labor (days/ha)
Mean 14.95 10.87t
Median 11.33 8.67
Number of observations 8 49

Planting costs with family labor (USD/ha)
Mean 55.71 43.00
Median 28.30 21.00
Number of observations 13 29

Planting costs without family labor
(USD/ha) 47.65 18.85"
Mean 21.57 10.12
Median 13 30
Number of observations

Other input costs (USD/ha)
Mean 55.88 21.84ttt
Median 48.8 7
Number of Observations 18 51

Total costs with family labor (USD/ha)
Mean 355.99 296.91
Median 295.13 256.11
Number of Observations 18 48

Yield (quintal/ha)
Mean 16.41 8.97"
Median 9.41 7.29
Number of Observations 18 49









Table 4-3. Continued
Cacao CCN-51 Cacao national
Revenue in the standard market
(USD/ha) 1419.37 648.90tt
Mean 652.80 572.71
Median 13 31
Number of observations

Revenue in the specialty market
(USD/ha) Na 824.41
Mean 600
Median 15
Number of observations

Revenue in the organic markets (USD/ha) Na 562.60
Mean 165.00
Median 3
Number of observations

Profit in the standard market with family
labor (USD/ha) 1066.54 361.31
Mean 263.06 261.74
Median 13 30
Number of observations


Profit in the standard market without
family labor (USD/ha) 1233.21 543.34t
Mean 495.07 513.00
Median 13 31
Number of observations

Profit in the specialty market with family
labor (USD/ha) Na 644.63
Mean 511.60
Median 14
Number of observations

Profit in the specialty market without
family labor (USD/ha) Na 762.20
Mean 600.55
Median 14
Number of observations









Table 4-3. Continued
Cacao CCN-51 Cacao national
Profit in the organic market with family
labor (USD/ha) Na 745.91
Mean 501.67
Media 4
Number of observations

Profit in the organic market without family
labor (USD/ha) Na 870.87
Mean 614.03
Median 3
Number of observations
tSignificant difference at the 10% level
" Significant difference at the 5% level
" Significant difference at the 1% level









Table 4-4. Yield, costs, revenue, and profits per hectare equivalent
Cacao CCN-51 Cacao national
Total labor (days/hae)a
Mean 28.75 63.77t
Median 25.89 41.93
Number of observations 18 48

Family labor (days/hae)a 16.16 37.72tt
Mean 13.95 22.5
Median 18 49
Number of observations

Hired labor (days/hae)a
Mean 12.59 25.01
Median 11.81 11.19
Number of observations 18 49

Labor for weeding (days/hae)a
Mean 9.24 29.14"
Median 9.17 15
Number of observations 18 51

Other labor input costs (USD/hae)a
Mean 51.50 28.78t
Median 36.97 8.77
Number of observations 18 51

Total costs with family labor (USD/hae)a
Men 304.40 564.61t
Median 289.60 327.21
Number of observations 18 49

Total cost without family labor (USD/hae)a
Mean 165.14 258.40
Median 171.70 121.03
Number of observations 18 49

Revenue in the standard market (USD/hae)a
Mean 1393.97 834.88t
Median 569.92 656.25
Number of Observations 13 31

Revenue in the specialty market (USD/hae)a
Mean Na 1060.86
Median 750.00
Number of Observations 15









Table 4-4. Continued
Cacao CCN-51 Cacao national
Revenue in the organic markets (USD/hae)a
Mean Na 1026.95
Median 1246.09
Number of observations 4

Profit in the standard market with family labor
(USD/hae)a 1081.10 285.15t
Mean 382.20 382.20
Median 13 30
Number of observations

Profit General Market without family labor
(USD/hae)a 1223.84 608.65t
Mean 345.45 569.92
Median 13 31
Number of observations

Profit specialty market with family labor
(USD/hae)a Na 677.90
Mean 472.50
Median 15
Number of observations

Profit in the specialty market with family labor
(USD/hae)a Na 677.90
Mean 472.50
Median 15
Number of Observations

Profit in the organic market without family labor
(USD/hae)a Na -182.32
Mean 288.39
Median 4
Number of Observations

Profit in the organic market without family Labor
(USD/hae)a Na 233.57
Mean 496.37
Median 4
Number of observations
a hae = hectare equivalent. Each parcel was converted to the amount of hectares it would
contain if the trees were planted at a density of 625 trees per hectare.
tSignificant difference at the 10% level
" Significant difference at the 5% level
" Significant difference at the 1% level











Table 4-5. Perceptions of profitability
Variable Name CCN-51 parcels Nacional parcels Entire farm


Profitability 6 years agor 3.00 3.57 2.90tt



Profitability 3 years agor 3.25 3.44 2.76ttt



Profitability nowr 3.28 3.09 2.67tt


Profitability 3-5 years in 3.92 4.11 3.46tt
the future

r Farmers scored the variables on a scale from one to five with one being very bad and five
being very good.
t Difference between cacao national and the entire farm is significant at the 10% level
t Difference between cacao national and the entire farm is significant at the 5% level
ttt Difference between cacao national and the entire farm is significant at the 1% level


* Total Farm

* Cacao Nacional

* Cacao CCN-51


0.00 -,


I


6 Yrs Ago


I1


3 Yrs Ago


I


Now


Next 3 Yrs


Figure 4-2. Perceptions of profitability


4.50

4.00

3.50

3.00

LA
S2.50
r-
01 2.00
t-
.C
4-,
. 1.50

1.00

0.50








Would You Like to Plant More Cacao?









E Yes
E No
U No Response


Figure 4-3. Likelihood to plant cacao








Which Variety Would You Prefer to
Plant?


4.5%


* Nacional
* CCN-51
* Both


Figure 4-4. Cacao variety planting preferences









Table 4-6. Environmental perceptions
Variable name CCN-51 parcels Nacional parcels Entire farm

Soil quality 6 years ago' 3.00 4.36aaa 3.76ttt

Soil quality 3 years ago' 3.00 4.25aaa 3.45ttt

Soil quality now' 4.00 4.24 3.10ttt

Soil quality 3-5 years in the future' 4.00 4.38a 3.07ttt

Biodiversity 6 years ago' 2.00 3.95 3.65

Biodiversity 3 years ago' 1.67 2.95aa 2.64

Biodiversity now' 2.50 2.42 2.04t

' Farmers scored the variables on a scale from one to five with one being very bad and five
being very good.
a Difference between cacao national and CCN-51 is significant at the 10% level
aa Difference between cacao national and CCN-51 is significant at the 5% level
aaa Difference between cacao national and CCN-51 is significant at the 1% level
t Difference between cacao national and the entire farm is significant at the 10% level
t Difference between cacao national and the entire farm is significant at the 5% level
tt Difference between cacao national and the entire farm is significant at the 1% level






4.5
4
3.5
3
2.5
2
1.5
1
0.5
0-


6 Yrs Ago 3 Yrs Ago
Figure 4-5. Perceptions of biodiversity


Now


Next 3 Yrs


* Total Farm
* Cacao Nacional
* Cacao CCN-51


Fl


C]









Table 4-7. Variables included in production and biodiversity regressions
Variable name Description
Biodiversity Number of additional varieties of plant species in the
parcel

Production Additional quintals of cacao produced

Log of family labor Log of the days of family labor for all cacao production

Nacional log of family Interaction term of a dummy variable for cacao national
labor with the log of the days of family labor

Log of hired labor Log of the days of hired labor for all cacao production

Log of other inputs Log of cost in US dollars other inputs besides labor for all
cacao production

Flat parcel Dummy variable indicating of the parcel where cacao is
planted in a parcel with a flat terrain

Hilly parcel Dummy variable indicating if the parcel where cacao is
planted neither a flat nor steep terrain

Insecure land Dummy variable to represent if the household has secure
property rights (insecure property rights are those defined
as communal land, squatted land, or inherited land)

Initial land Hectares of land in household's possession when the
household began farming

Years land owned Number of years that the household has the parcel of
land

Ethnicity (head of Dummy variable if the head of the household is
household) indigenous Tsa'chila

Household size Number of members in the household









Table 4-8. Coefficients of the biodiversity equation
Variable name Coefficient Standard Error


Log of family labor 0.09 0.07

Nacional log of family labor 0.15" 0.05

Log of hired labor 0.05 0.04

Log of other inputs 0.03 0.03

Flat parcel -0.64tt 0.19

Hilly parcel -0.40tt 0.19

Insecure land rights -0.49tt 0.15

Initial land size (ha) 0.00t 0.00

Years land owned 0.01t 0.00

Ethnicity (head of household) 0.67t 0.19

Household size -0.05t 0.03

Constant 0.52 0.22
F(11, 32) = 20.94
Probability > F = 0
R-squared = 0.7518
Number of observations = 44
SSignificance at the 10% level
" Significance at the 5% level









Table 4-9. Coefficients in the production equation
Variable Name Coefficient

Log of family labor 0.35t

Nacional log of family labor -0.03

Log of hired labor -0.07

Log of other inputs 0.06

Flat parcel 0.91tt

Hilly parcel 0.73tt


Insecure land rights

Initial land size (ha)

Years land owned

Ethnicity (head of the household)

Household size

Constant
F(11, 32) = 7.99
Probability > F = 0
R-squared = 0.6302
Number of observations = 44
SSignificance at the 10% level
" Significance at the 5% level


0.15

0.00

0.01t

-0.92t

-0.03

0.48


Standard Error

0.16

0.04

0.10

0.07

0.19

0.23


0.27

0.00

0.00

0.28

0.07

0.54









Table 4-10. Shadow wage components by market
All Cacao


Nacional


wQ*(in USD per day of family 10.35 N.S.t
labor) in Standard Market

wQ*(in USD per day of family 11.93 N.S.t
Labor) in Specialty Market

wv*(in varieties per day of family N.S.* Pv*0.01a
labor)

t N.S. means the value is not significant
a Pv is the value of the additional variety in the parcel


Table 4-11. Shadow wage by variety and market
Price in US dollars per day
of labor
Shadow value of labor for cacao national sold in > 10.35
standard markets

Shadow value for a day of labor for cacao national sold > 11.93
in specialty markets

= 10.35
Shadow value for a day of labor for cacao CCN-51 sold
in standard markets









CHAPTER 5
CONCLUSION

New Conceptual Framework

A need exists to better conceptualize the production decisions of small landholder

households. Many models of smallholder farmers' production decisions have begun to

include factors besides profit maximization such as subsistence needs and overall

household welfare. The environmental impact of production methods are also important

factors to smallholder households as these variables impact the sustainability of their

farms. In addition, these factors provide the household with aesthetic and other values.

A biodiversity measurement is one way to capture this ecological value as the research

on agroforestry cropping systems have shown that these diversified cropping systems

provide many ecological services important to smallholder households. These

biodiverse systems also provide many monetary benefits such as additional incomes

sources and meeting consumption needs. The biodiversity factors have proven to be

an important consideration in the production decisions of smallholder cacao producers

in Ecuador as these farmers would prefer to raise cacao national which provides many

ecological and other benefits that are not obtained in cacao CCN-51 parcels.

In order to analyze the importance of biodiversity in the production of cacao, we

created a model that disentangles the effects of family labor for biodiversity and for the

production of cacao by analyzing shadow prices. Shadow wage and shadow income

value biodiversity which has not been included in past models of shadow wages and

labor. Our model also takes into account of the effect of segmented markets.

Segmented markets exist in the Ecuadorian case by certification requirements and

limited demand in specialty cacao national markets. Our modeling technique is useful









for the study of other crops that participate in segmented markets such as in the fair

trade markets, not just the cacao national specialty market.

The shadow wage in this model is the sum of the traditional measure of the

shadow wage, household's MPL for cacao production, and the new variable, the MPL of

biodiversity. Our analysis of the shadow prices for cacao in northwestern Ecuador

reveals the households receive a higher shadow income for planting cacao national

than for planting cacao CCN-51. Only a lower bound for the shadow wage could be

developed as the aesthetic and other nonmarket values of biodiversity were not

captured in the survey. The magnitude of the difference between the market and

shadow wages depends on the crops planted in the cacao parcels. The shadow prices

also show that family and market labor are not perfect substitutes. Family labor

receives a very high value that would induce family members to work on the farm

instead of participating in the labor markets. The shadow wage for cacao production is

at the upper bounds of the labor market and the shadow wage for cacao national sold

in specialty markets is even higher. Thus, smallholder households in Ecuador would be

able to maximize the value of their labor by devoting to production of cacao national.

Impact on Current Research

Our analysis of shadow wage extends the theoretical work of Jacoby (1993) and

Skoufias (1994). Our model adds to this work by showing the importance of including

externalities, such as the benefits from biodiversity, in computing shadow wage in

segmented markets. By examining segmented markets, our analysis of shadow income

enhances the work on fair trade of LeClair (2002) and Hayes (2008) by examining the

impact of these markets at the household level. The determinants of market access in

these specialty and fair trade markets were also examined. Our surveys revealed that









many farmers were excluded from these markets. Our model could help create a better

understanding of the failings of these methods in reaching smallholder households. Our

model of shadow prices has shown to be effective in trying to capture the non market

benefits that farmers realize in their production and take into account in making copping

decisions.

Need for Further Research

Our study leaves several areas to be examined to better understand smallholder

household's production decisions. First, this research was not able to fully explore the

impact of risk on these decisions. Research on agroforestry has shown that these

diversified cropping systems are less risky than monoculture production methods as a

household has additional income sources to meet its needs (Reitsma 2001). The

household welfare impacts of growing diversified, organic production systems compared

to monoculture commercial production systems such as between cacao national and

cacao CCN-51 need to be examined. The former production system is argued to be

more ecologically sustainable but the human welfare impact also needs to be examined

to truly understand the impact to total welfare of each production system. Finally,

methods need to be utilized to give a value to the non market benefits for biodiversity.

Only by placing a value on these factors can shadow wage be crated that includes all

the externalities rather than creating a lower bound for this wage as was included in our

study.

These additional nonmarket values are especially important in developing

countries where many households are dependent on their farms not only for cash

income but also for meeting subsistence needs. The Ecuadorian cacao case provides a

good example of the need for the inclusion of additional values in the household









decision model. When comparing the shadow wages, the traditional production

methods for cacao national proved to best production decision when the value for

biodiversity was included in analyzing the smallholder household's production decision.

The shadow wage and income that includes biodiversity as a value in the price provides

an additional method for determining the true value of a production method to

smallholder households.









LIST OF REFERENCES


Arslan, A., and J.E. Taylor. 2009. Farmers' Subjective Valuation of Subsistence
Crops: The Case of Traditional Maize in Mexico. American Journal of
Agricultural Economics 91:956-972.

Asociacion Nacional de Exportadores de Cacao. 2007. Manuel del Cultivo de Cacao
para Productores. Guayaquil, Ecuador.

Auld, G., L. H. Gulbrandsen, and C.L. McDermott. 2008. Certification Schemes and the
Impacts on Forests and Forestry. Annual Review of Environmental Resources
33: 187-211.

Bardhan, P. A. 1999. Development Microeconomics. New York: Oxford University
Press.

Barnum, H. N., and L. Squire 1979. An Econometric Application of the Theory of the
Farm-Household Journal of Development Economics 6: 79-102.

Becchetti, L., and M. Costantino. 2008. The Effects of Trade on Affiliated Producers:
An Impact Analysis of Kenyan Farmers. World Development 36: 823-842.

Beer, J., R. Muschler, D. Kass, and E. Somarriba. 1998. Shade Management in Coffee
and Cacao Plantations. Agroforestry Systems 38: 139-164.

Bentley, J., E. Boa, and J. Stonehouse. 2004. Neighbor Trees: Shade, Intercropping,
and Cacao in Ecuador. Human Ecology 32: 241-258.

Brush, S. B., J. E. Taylor, and M. R. Bellon. 1992. Technology Adoption and
BiologicalDiversity in Andean Potato. Agriculture. Journal of Development
Economics 39: 365-387.

Butler, R. A., and W.F. Laurence. 2008. New Strategies for Conserving Tropical
Forests. Trends in Ecology and Evolution 23: 469-472.

Carter, M. R., and Y. Yao. 2002. Local Versus Global Separability in Agricultural
Households Models: The Factor Price Equalization Effect of Land Transfer
Rights. American Journal of Agricultural Economics 84: 702-715

Central Intelligence Agency. 2009. Factbook. Website
https://www.cia.gov/library/publications/the-world-factbook/geos/ec.html.
Accessed 30 November 2009

Conservation International. 2007. Biodiversity Hotspots. Website.
http://www.biodiversityhotspots.org/xp/hotspots/tumbes_choco/Pages/defau lt. asp
a. Accessed 1 October 2009









Coporaci6n de Promoci6n de Exportaciones e Inversiones. 2009. Cacao. Ecuador
Calidad de Origen. Quito, Ecuador.

Dahlquist, R. M., M. P. Whelan, L. Winowiecki, B. Polidoro, S. Candela, C. A. Harvey, J.
D. Wulfhorst, P. A. McDaniel, and N. A. Bosque-Perez. 2007.. Incorporating
Livelihoods in Biodiversity Conservation: a Case Study of Cacao Agroforestry
Systems in Talamanca, Costa Rica. Biodiversity and Conservation 16: 2311-
2333.

de Janvry, A., M. Fafchamps, and E. Sadoulet. 1991. Peasant Household Behavior
with Missing Markets: Some Paradoxes Explained. The Economic Journal 101:
1400-1417.

Duguma, B. J. 2001. Smallholder Cacao (Theobroma cacao Linn.) Cultivation in
Agroforestry Systems of West and Central Africa: Challenges and Opportunities.
Agroforestry Systems 51: 177-188.

El Cacao Volvi6 Ser la "Pepa de Oro." 5 February 2009. El Universo (Guayaquil,
Ecuador) pp. 23.

Eswaran M., and A. Kotwal. 1995. Why Poverty Persists in India. New York: Oxford
University Press

Feather, P., and W. D. Shaw. 1999. Estimating the Cost of Leisure Time for
Recreation Demand Models. Journal of Environmental Economics and
Managemen 38: 49-65.

Food and Agriculture Organization of the United Nations. Statistical Database. Website
http://faostat.fao.org/default.aspx Accessed 20 March 2009

Gamma-Rodrigues, E.F. 2009. Carbon Storage in Soil-Size Fractions under Cacao
Agroforestry Systems in Bahia, Brazil." Working Paper, University of Florida,
School of .Forest Resources and Conservation.

Gobierno de la Provincia de Pichincha. 2003. Plan de Desarrollo Participativo 2002-
2012. Quito: Gobierno de la Provincia Pichincha.

Gomez Tovar, L., L. Martin, M.G. Cruz, and T. Mutersbaugh. 2005. Certified
Agriculture in Mexico: Market Connections and Certification Practices in Large
and Small Producers. Journal of Rural Studies 21: 461-474.

Hilderbrand, P. E. 2002. Global Research Challenges: Including Small Holders in Rural
Development. First Henry A. Wallace Inter-American Scientific Conference on
Globalization of Agricultural Research. Turrialba, Costa Rica.

Hayes, M.G. 2008. "Fighting the Tide: Alternative Trade Organizations in the Era of
Global Free Trade" A Comment. World Development 36: 2953 2961.









Jacoby, H. G. 1993. Shadow Wage and Peasant Family Labour Supply: An
Econometric Application to the Peruvian Sierra, Review of Economic Studies 60:
903-921.

Le, T. L. 2009. Shadow Wages and Shadow Income in Farmers' Labor Supply
Functions. American Journal of Agricultural Economics 91: 685-696.

LeClair, M. S. 2002. Fighting the Tide: Alternative Trade Organizations in the Era of
Global Free Trade. World Development 30: 949 958.

Melo, C. J. 2009. Agroecological Assessment of Ecuadorian Cacao Production
Systems. Working Paper, Florida International University, Department of
International Relations.

Municipalidad de Puerto Quito. 2009. Website.
http://www.puertoquito.gov.ec/index.php?option=com_content&view=article&id=5
5<emid=58. 11 March 2010.

Municipalidad de Quininde. 2009. Website. http://www.quininde.gov.ec/. 11 March
2010

Neilson, J. 2008. Global Private Regulation and Value-Chain Restructuring in
Indonesian Small Holder Coffee Systems. World Development 36: 1607-1622.

Nelson, V., and Galvez, M. 2000. Social Impact of Ethical and Conventional Cocoa
Trading on Forest-Dependent People in Ecuador. Working Paper. University of
Greenwich, Natural Resources and Ethical Trade Programme.

Perfecto, I., J. Vandermeer, A. Mas, and L.S. Pinto. 2005. Biodiversity, Yield, and
Shade Coffee Certification. Ecological Economics ,54: 435-446.

Provincia de Santo Domingo de los Tsachilas. 2010. Information General. Website
http://www.gptsachila.gov.ec/index.php?option=com_content&task=view&id=16&
Itemid=33. Accessed on 11 March 2010.

Reitsma, R., J. D. Parrish, and W. McLarney. 2001. The Role of Cacao Plantations in
Maintaining Forest Avian Diversity in Southeastern Costa Rica. Agroforestry
Systems 53: 184-193.

Rice, R. A., and R. Greenberg. 2000. Cacao Cultivation and the Conservation of
Biological Diversity. Ambio 29: 167-173.

Ros-Tonen, M. A., T. V. Andel, C. Morsello, K. Otsuki, S. Rosendo, and I. Scholz,
2008. Forest-related Partnerships in Brazilian Amazonia: Three is More to
Sustainable Forest Management than Reduced Impact Logging. Forest Ecology
and Management 1256: 482-1497.









Skoufias, E. 1994. Using Shadow Wages to Estimate to Labor Supply of Agricultural
Households. American Journal of Agricultural Economics 76: 215-227.

Steffan-Dewentera, I., M. Kesslerc, J. Barkmannc, M. M. Bosa, D. Buchorig, Stefan
Erasmih, H. Fausth, G. Geroldh, K. Glenke, S. R. Gradsteind, E. Guhardjai, M.
Harteveldd, D. Herteld, P. Hohna, M. Kappash, S. Kohlerh, C. Leuschnerd, M.
Maertensj, R. Marggrafe, S. Migge-Kleiank, J. Mogeai, R. Pitopangl, M.
Schaeferk, S. Schwarzem, S. G. Spornd, A. Steingrebek, S. S. Tjitrosoedirdjoi, S.
Tjitrosoemitoi, A. Tweleh, R. Weberh, L. Woltmannk, M. Zellerm, and T.
Tscharntkea. 2007. Tradeoffs between Income, Biodiversity, and Ecosystem
Functioning during Tropical Rainforest Conversion and Agroforestry
Intesification. Proceedings of the National Academy of Sciences, Vol 104.
Available at www.pnas.org/cgi/doi/1 0.1073/pnas.0608409104

Vasco, J. S. Breve Historia del Cultivo de Cacao en Ecuador. Ministerio de
Agriculture, Ganaderia, Acuacultura y Pesca en Ecuador: Website
http://www.sica.gov.ec/cadenas/cacao/docs/historia_cacao.htm. Accessed 11
March 2010









BIOGRAPHICAL SKETCH

Trent Blare was raised on a ranch in western South Dakota where he first gained a

passion for agriculture and the rural lifestyle. While in high school, Trent was very

active in 4-H and FFA. The activities in these organizations enhanced his interest in

agribusiness. So, he decided to pursue a degree in Agricultural Economics at the

University of Nebraska- Lincoln where he was an honors graduate and Chancellor's

Scholar. He had the opportunity to participate in two undergraduate research

opportunities while at the University of Nebraska. He first worked with political science

professor Dr. David Rapkin to examine the impact of the United States' position in trade

negotiations in the World Trade Organizations. He continued this work with his adviser

Dr. E. Wesley Peterson to build a model that explained the impact on American small

grain producers of a new trade deal in the World Trade Organization.

One of the highlight of Trent's life and professional development was his service in

the Peace Corps. Trent lived in a Tsa'chila indigenous community for three years. He

worked with the community on an eco tourism project as well as helped improve their

production and marketing of cacao. This experience spurred Trent to continue his

studies in Food and Resource Economics at the University of Florida with a focus on

agricultural development. While at the University of Florida, Trent has been active in

the Tropical Conservation and Development Program, which provided him with funding

to pursue his research on Ecuadorian cacao production in the summer of 2009. Trent

plans to continue his work in agricultural development and especially his research on

cacao in Ecuador as he begins a PhD in Food and Resource Economics at the

University of Florida in August of 2010.





PAGE 1

1 SEEDS OF GOLD: THE IMPACT OF BIODIV ERSITY ON CACAO PRODUCTION DECISIO NS OF SMALL LANDHOLDER HOUSEHOLDS IN NORTHWESTERN ECUADOR By TRENT BLARE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTI AL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Trent Blare

PAGE 3

3 To Cle mencia Quishpe mi mama Ecuatoriana

PAGE 4

4 ACKNOWLEDGMENTS I would like to especially thank my advisor D r. Pilar Useche for assisting me from the pro posal, survey construction, data analysis model development, and thesis writing. I am very thankful to the Tropical Conservation and Development Program within the Center for Latin American Studies for funding my field research from June through August 2009. I owe a great debt to those who helped me in Ecuador collect the data as well as house me and support me during my time in Ecuador They also provided me with immeasurable support and insight while I anal yzed the data. I especially would like to thank the Fundacin Accin Social Caritas, the provincial go v ernments of Santo Quishpe, and Betty Cuellar. I also would like to th ank Jhonny and Eduardo Alvarez for helping me understand some particularities of the data during the final weeks that I wrote and defended my thesis.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABREVIATIONS ................................ ................................ ............................... 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 ................................ ................................ ......... 14 ................................ ................................ .............. 15 2 VALUE OF BIODIVERSITY ................................ ................................ .................... 19 Ecological Importance ................................ ................................ ............................. 20 Pollution Control ................................ ................................ ............................... 21 Soil Enhancement ................................ ................................ ............................ 21 Carbon Sequestration ................................ ................................ ...................... 22 Household Use Value of Biodiversity ................................ ................................ ...... 23 3 THEORETICAL FRAMEWORK ................................ ................................ .............. 25 Modeling Special ty/Fair Trade Markets ................................ ................................ ... 25 Current Conceptualization of Household Shadow Price Models ............................. 28 Framework for Valuing Biodiversity ................................ ................................ ........ 29 Countervailing Market Imperfections ................................ ................................ 32 Imperfect Segmented Product Markets ................................ ............................ 33 4 EMPERICAL IMPLEMENTATION ................................ ................................ .......... 34 Research Site ................................ ................................ ................................ ......... 34 Survey Dissemination ................................ ................................ ............................. 37 Characteristics of Households that Produce Each Variety of Cacao ...................... 41 Parcel Characteristics ................................ ................................ ............................. 42 Production and Profitabili ty Values ................................ ................................ ......... 43 Perception of Profitability ................................ ................................ ........................ 47 Access to Specialty Markets ................................ ................................ ................... 49 Likelihood to Plant Cacao ................................ ................................ ....................... 50 Perceptions of Environmental Factors ................................ ................................ .... 50

PAGE 6

6 Estimation ................................ ................................ ................................ ............... 52 Specific Estimated Equations ................................ ................................ ........... 53 Definition of the Variables ................................ ................................ ................. 54 Results of the Estimated Equations ................................ ................................ ........ 57 Summary of the Results ................................ ................................ .......................... 61 5 CONCLUSION ................................ ................................ ................................ ........ 79 New Conceptual Framework ................................ ................................ ................... 79 Impact on Current Research ................................ ................................ ................... 80 Need for Further Research ................................ ................................ ..................... 81 LIST OF RE FERENCES ................................ ................................ ............................... 83 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 87

PAGE 7

7 LIST OF TABLES Table page 4 1 Household characteristi cs ................................ ................................ .................. 63 4 2 Parcel characteristics ................................ ................................ ......................... 64 4 3 Yield, costs, revenue, and profits per hectare ................................ .................... 65 4 4 Yield, costs, revenue, and profits per hectare equivalent ................................ ... 68 4 5 Perceptions of profitability ................................ ................................ .................. 70 4 6 En vironmental perceptions ................................ ................................ ................. 73 4 7 Variables included in production and biodiversity regressions ........................... 75 4 8 Coefficients of the biodiversi ty equation ................................ ............................. 76 4 9 Coefficients in the production equation ................................ ............................... 77 4 10 Shadow wage components by market ................................ ................................ 78 4 11 Shadow wage by variety and market ................................ ................................ .. 78

PAGE 8

8 LIST OF FIGURES Figure page 4 1 la, Puerto Quito, and Quininde ......... 62 4 2 Perceptions of profitability ................................ ................................ .................. 70 4 3 Likelihood to plant cacao ................................ ................................ .................... 71 4 4 Cacao variety planting preferences ................................ ................................ .... 72 4 5 Perceptions of biodiversity ................................ ................................ .................. 74

PAGE 9

9 LIST OF ABREVIATIONS BIOFASCA Cacao processing center ran by FASCA CORPEI Corporacin de Promocin de Expo rtaciones e Inversiones FASCA Fundacin Accin Social Caritas FOCs First Order Conditions MPL Marginal Product of Labor MRS LC Marginal Rate of Substitution between Leisure and Consumption UOCAQ Uni n de Organizaciones Campes inas del Cantn Quininde

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for th e Degree of Master of Science SEEDS OF GOLD: THE IMPACT OF BIODIVERSITY ON THE CACAO PRODUCTION DECISIONS OF SMALL LANDHOLDER HOUSEHOLDS IN NORTHWESTERN ECUADOR By Trent Blare August 2010 Chair: Pilar Useche Major: Food and Resource Economics Our study of Ecuadorian smallholder cacao farmers revealed that these farmers chose to use a farming method that is not most profitable, which appears to contradict the classical economic model of profit maximization. M any farmers still grow a less profitable variety of cacao, cacao naci onal, in an agroforestry system instead of the more pr oductive hybrid variety, cacao CCN 51, which is planted with few other crops and trees and uses many c hemical inputs. Farmers must receive additional nonmarket value from cacao nacional such as ecological values, which induces them to raise cacao nacional Our study uses a household model and shadow prices to capture t these nonmarket values in order to understand the full value of each production method to these smallholder farmers Our study builds on the current models of shadow wages by not only including the biodiversity value but also showing the impact of segmente d markets such as f fair trade, organic, or ot her markets on household income Furthermore, the shadow wage takes into account the effect of imperfect substitution of

PAGE 11

11 The agroforestry production of cacao prov ides the households with ecological services as well as a source of additional income and food. When the additional value from biodiver sity is included in the shadow wage of traditiona lly produced cacao nacional, the shadow wage for cacao nacional is greater than that for cacao CCN 51. This shadow wage difference would predict that smallholder farmers would chose to rais e cacao nacional in order to maximize household utility. Thus, smallholder Ecuadorian households are acting as rational actors when choosing to raise cacao nacional as this production system maximizes household utility.

PAGE 12

12 CHAPTER 1 INTRODUCTION Farmers throug h the world fac e multiple decisions from which crops to grow to where to market thei r products. Although these decisions have traditionally be en conceptualized as choosing the option that is most productive and profitable, recent research has shown that farmers include many other criteria when making production decisions Hilderbrand (2002) demonstrated that a livelihood model is a better method to des cribe production these decisions. The livelihood system include s household consumption and leisure needs as well as nee ds an d production constraints. These competing objectives may lead the household to make decisions that contradict the popular notion of the rational profit maximizing decision maker. Even though the household may not be maximizing profit, these household s are making rational decisions to maximize their total welfare or utility. Bechetti and Costantino (2008) found that household welfare is very important to production decisions. They found that f armers chose to participate in organic markets not just fo r the price advantage. They also participated in these markets because of the additional welfare benefits that the additional income provided as well as could be obtained directly from the production such as a nutritious, safe food source. An example of t decisions is in the production of cacao in Ecuador. Many Ecuado rian smallholder cacao farmers produce a variety of cacao that is le ss productive and profitable which counter s traditional econo mic theory. Ecuadorian farmers include other factors besides profits in their decisi ons Ecological factors are very important in their decision process. These traditional production systems use agroforestry methods and provide households with

PAGE 13

13 not only economic benefits but also a source of food and medicine (Bentley, Boa, and Stonehouse 2004). These additional benefits only occur to households that raise cacao in diversified or biodiverse production systems. Thus, Ecuadorian cacao farmers take into acc ount this cropping biodiversity and all its benefits when making production decisions. This paper develops a household model that i ncludes biodiversity as a part of the decision criteria in Ecuadorian cacao production decisions P ast work has e xamined shadow prices to examine market failures especial ly in the labor market (Le 2009; Arslan and Taylor 2009; Jaco by 1993). T his work adds a new aspect by estimating a shadow wa ge and income that take s into account the additional value for biodiversit y that is not captured in the market. T he shadow wage is derived from a household model where a rational household chooses to maximize its uti lity in the presences of market imperfections, which may be different than the choice the household would make if such constraints did not exist The distinctive characteristics of this analysis reveal that such market imperfections are associated with environmental goods, such as the ones resulting from intercropping of different crop varieties, rather than with tr ansaction costs (de Janvry, Fafchamps, and Sadouelt 1991) and constraints in the labor (Jakoby 1993; Skoufias 19 94; Le 2009), land and credit markets (Eswaran and Kotwal 1995; Carter and Yao 2002). The shadow income for agroforestry cacao production was fo und to be higher than that for com mercial cacao production system. This situation shows that the relevant imperfections is associated with over employment, rather than with under employment (Feather and Shaw 1999) of family labor as has been assume by mos t papers in the

PAGE 14

14 household models literature Thus, a rational household would chose to use the traditional rather than the commercial system. The production of Ecuadorian cacao provides a great case study to examine the effect of biodiversity on smallhold er household utility and production decisions for several reasons. First, a ll households growing cacao p articipate in the cacao market as self consumption is very close to zero. T hey use market prices as signals for the production decisions. Thus, this example allows for the disaggregation of the market imperfections in the cacao cash cr op market from the total of the market imperfections to determine the impact of biodiversity on the shadow wage. Second, the manner in which traditional cacao is raised in diversified agroforestry systems provide s a particularly good contrast to the manner in which modern varieties of this product are gro wn. T his crop is important to the economy of many developing countries including Ecuador and other countries throughou t the world that are cradles of biodiversity (Bentley, Boa, and Stonehouse 2004; Brush, Taylor and Bellon 1992). Finally the unique cacao specialty market that has developed in Ecuador for its traditional cacao production provides a good example to stud y the impact these premium markets on smallholder households The tropics including Asia, Africa, and Latin America are known for their high levels of poverty and biodiversity. T he impact of cacao production in these regi ons is large with 7.42 million hectares being devoted t o this crop in 2007 that produces over five metric tons of cacao value d over 6.75 billion US dollars (Food and Agriculture 2009 ). Indeed, c acao is the second most important cash cop in the t r opics (St effan Dewenter et al. 2007). An interesting fact about cacao is that much of its production is

PAGE 15

15 achieved by small landhold ers who produce between 70% and 90% cacao (Dah lquist et al. 2007). T he average cacao farm size in Nigeria is just 1.7 hectares, 4.5 hectares in Ecuador, and 2.8 hectares in the Ivory Coast. These small landhold ers can be very productive. For instance, farmers in Sulawesi, Indonesia have been able to produce 2000 kg of cacao per hectare per year which exceeds production levels that can be achieved by commercial cacao producers (Rice and Greenberg 2 000). So, cacao is important in the development plans for this reg ion because of its ecological impact and importance to small landholders. One of these tropical countries heavily dependent on cacao production is Ecuador. large expanses of the coastal region devoted to the production of this crop. During this per iod, In fact, c acao production helped the new country finance its war for independence against Spain. However, the crop was nearly destroyed by two fungi, monilla and T h e demand f or cacao collapsed during the Great Depression. This economic crisis forced the large cacao haciendas to be divided into parcels that were sold to small landholders. The land reforms of the 1960s further divided these large farmers so that today most of (Vasco 2010 ) Today, 90% of cacao is produced on l andholdings with less than 50 hectares and over 30 % is produced on fa rms smaller than 10 hectares ( Coporacin de Promocin de Exportaciones e Inversiones 2009). Ev en though cacao production may not be as large of a component of the Ecuadorian economy as it was 10 0 years ago, cacao continues to be important to

PAGE 16

16 Ecuador In fact, 12% cacao production empl oying 600,000 Ecuadorians (Coporacin de Promocin de Exportaciones e Inversiones 2009) Furtherm ore, Ecuador controls 60% of th e market of the cacao used to produce gourmet, dark chocolates production and is one of only three countries w here this high qu ality cacao can be grown The variety raised in Ecuador is referred to locally as cacao nacional or cacao arriba in international markets (Asociacion Na cional de Exportadores de Cacao 2007) The need for economic development in Ecuador is indeed acute. F orty six percent of Ecuadorians live below the poverty lin e Poverty is especially concentrated in the rural areas of the co untry (Central Intelligence Agency 2009) The promotion of cacao production is seen as an avenue to help alleviate this rural pove rty. Because of the Ecuadorian government along with local and international development organizations has begun to advocate cacao and es pecially cacao nacional as an econom ic develo pment strategy. Most of the cacao raised throughout the world is raised using hybrid varieties such as the CCN 51 variety grown in Ecuador The advantage of cacao CCN 51 over cacao nacion al is that it is more productive However, cacao CCN 51 do es not produce the high quality cacao that is produced by cacao nacio nal. C acao CCN 51 is used in lower q uality chocolates which garner lower price s ( El Cacao Volvi Ser la Pepa de Oro 2007) Yet c acao CCN 51 has been pro moted to small landholders over the past 15 to 20 years as a superior option to cacao nacional because of its production advantages (Bentley, Boa and Stonehouse 2004) During the past 5 years, some f arm ers have

PAGE 17

17 obtained access t o spe cialty where they receive premiums for cacao nacional and organically produced cacao (Bentley, Boa, and Stonehouse 2004) Because of these new markets Ecuadorian farmers are being encouraged to switch from producing c acao CCN 51 to cacao nacional. Indeed, t he price premium for cacao nacional has expanded from nearly 2 0% to over 60% between 2004 and 2007 ( El Cacao Volvi Ser la Pepa de Oro 2007) In fact, t he Ecuadorian export institution, the Co r poracin de Promocin de Exportaciones e Inversiones (CORPEI), now even requires farmers to separate thes e two varieties to meet chocolate makers demand for cacao nacional. In addition to the quality differences, these two varieties of cacao are raised using different farming methods. The traditio nal system used to raise cacao nacional includes a diverse array of crops and retention overstory trees to enhancing nutrient cycling (Bentley, Boa, and Stonehouse 2004) Cacao is commonly planted with plantains and/or banana s. Many farmers also plant orange s or other citrus fruit, or avocado i n their cacao plantains. F armers include many timber species such as laurel on their cacao parcels. One of the most popul ar tree species included in the cacao nacional agroforestry system is locally referred to as guaba. The guaba tree produces a long pod that contain a sweet pulp t hat surrounds its seeds and is an important nitrogen fixer, providing the system with a natural fertilizer. Rather than using chemical fertilizers and pesticides the traditional cacao production system uses mechanical methods to control diseases and pes ts, such as pruning. This approach to production varies greatly from the commercial method. Cacao CCN 51 is not as susceptible to sun damage and, therefore, is not grown in the traditional agroforestry system but in parcels that contain

PAGE 18

18 few other crops a nd trees besides cacao CCN 51 Cacao CCN 51 is also raised using more agrochemicals than cacao nacional

PAGE 19

19 CHAPTER 2 VALUE OF BIODIVERSITY Biodiversity has not traditionally been considered a component of smallholder househol d o es not fit into the traditional profit maximization models. Many of the benefits of biodiversity are not received as cas h in come T here are positive externalities not realized in the market such as the aesthetic value that a farmer realize s from their b iodiverse farms. In addition, t here are benefits that may have monetary value but this value is not received directly and/or is not realized until somet ime into the future such as enhanced soil fertility, which leads to greater production. Yet, some prod ucts of biodiversity do have a market value. For instance, a diversified cropping system produces a variety of crops that can be brought to t he market. All t hese biodiversity values may lower price s for their pro ducts as the value they receive through the positive external ities of biodiversity may compensate for the lower price. Indonesian cacao farmers said that they would be willing to accept a price lower than full compensation for the smaller yields of tradit ionally raised cacao to grow cacao in an agroforestry system. These farmers would be willing to receive just a 0 .36 U S dollars premium per kg for their cacao grown in an agroforestry system (Steffan Dewenter et al. 2007). Chapter 2 examines the current research on the ecological impacts of the biodiverse cacao nacional agroforestry system. The values are often difficult to capture in traditional markets. Yet, they would likely be included in a smallholder producers production decisions as they impact t he sustainability of their livelihoods The second section examines some of the use values su ch as meeting consumption needs that these biodiverse systems provide households.

PAGE 20

20 Ecological Importance The diversified cropping system of traditional agrofore st r y cacao systems such as that of the cacao nacional have been shown to enhance the diversity of native plants and animals (Reitsma, Parrish, and McLarney 2001). Both the above ground and below ground bio logical diversity of cacao agroforests have been foun d to be greater than that of commercial cropping systems and similar to that of long term fallowed fields but still less than primary forests (Duguma 2001). Cacao agroforestry systems in Indonesia were found to contain forty percent of the plant and insec t species found in natural forests. This level of biological diversity can only be found when the cacao i s planted in a diversified agro f orestry system Costa Rican cacao agro forestry systems were shown to have a diversity of bird species statistically significantly similar to that of a natural forest. Natural forests were shown to have 144 bird species compared to 131 in caca o agroforestry systems (Reitsma 2001). This potential to create a biologically diverse habitat is especially prevalent in defo re sted areas However, the monoculture non shaded cacao production practice which is the common method used in raising cacao CCN 51 have not shown to provide similar levels of biodiversity (S teffan Dewenter et al. 2007). These biodiverse c acao agricultural s ystem can provide many of the environmental benefits that natural forests provide. While the rainforests are being destroyed, one of the critical objectives of the cacao agroforestry practice is to replicate the natural forest environment so that the beneficial qualities of the tropical forests can be achieved in these agroforestry systems so that the loss of natural forests can be mitigated. The protection of native plant and animal species is particularly important to Ecuador as it is o ne of the most biodiverse countries in the

PAGE 21

21 world with many plants and animals threatened with ext inction due the loss of habitat (Conservation International 2007). To slow this destruction, many environmental groups are promoting the production of cacao n acional as a way of protecting this habitat The additional ecological benefits these biodiverse systems provide include pollution and runoff control, improved soil quality through nutrient cycling and organic matter accumulation, and carbon sequestration Pollution Control As compared to raising crops in intensive monoculture production, utilizing a cacao agroforestry system has shown to help prevent further degradation of water resources. In addition, the use of cacao agroforestry systems can ameliorate soil erosion which is critical in the sustainability of agricultural activity. The shade trees utilized in the cacao agro forestry practice provide several ecological benefits. One benefit is that the shade trees prevent some of the leaching of pesticid es and other chemicals into groundwater. The root zone of the trees can prevent these chemicals from reaching the water supply. In addition, shaded farmer have lower run off rates and, thus, less soil loss. The tree root zone and the natural litter form the shade trees slow the runoff so less soil is removed. The canopy also protects the soil from the full impact of rainfall slowing down the raindrops as they fall so as to protect the soil from being loosen ed and washed away (Beer et al. 1998). Soil Enh ancement The dense planting of trees and plants in cacao agroforestry systems provide a large level of litter and organic matter to be recycled back into the soil to maintain its fertility and long run sustainability. Total biomass in cacao agro forest sy stems in Cameroon was higher than that in food crop fields with 305 metric tons per hectare

PAGE 22

22 versus only 85 metric tons per hectare. Yet, this was much less than biomass in primary forests, 541 metric tons per hectare, and long term fallow, 460 metric tons per hectare. These agro forestry systems proved to have larger amounts of organic matter and higher levels of calcium and magnesium in the soil than s econdary forests (Duguma 2001). Cacao agroforestry systems have been shown to on average add 10 Mg of p lant litter per hectare per y ear (Ga mma Rodriguez 2009 ). One study found that soil organic matter increased by twenty one percent in a ten year period under pruned Erythrina peppigiana and by 9% under laurel trees in fields that were converted from sugar cane production to a cacao agro forestry fields (Beer et a l 1998). Furthermore, soil decomposition rates and the abundance of soil arthropods were found to be greater in shaded cacao agro forestry systems as compared to systems that provide less shade (St effa n Dewenter et al. 2007). Carbon Sequestration Cacao agroforestry systems can be a useful tool in the sequestration carbon in the effort to control global warming. In fact, these agroforestry systems seem to be able to store carbon at a similar rate o f that of natural forests. Soil organic carbon stock in these systems is not significantly different from that of forests in the first 100 centimeters of soil with a rating of 302 mega grams of organic carbon per hectare (G ama Rodrigues 2009 ). Anot her st udy found that cacao agro forest systems have shown to contain 62% of the carbon stock found i n primary forests (Duguma 2001). The shade trees in the cacao agro fo restry system store between 14 and 52 Mg of Carbon per hectare in the aboveground woody bioma ss. In fact, this agroforestry practice stores between 10 and 50 Mg of carbon more per hectare in the litter and soil organic matter than annua l crops. Therefore, cacao agro forestry system could prevent

PAGE 23

23 the release of 1000 Mg of Carbon per hectare that w ould occur by pla nting annual crops (Beer et al. 1998). In Cameroon mature agroforestry systems have been found to fix 150 metric tons of carbon per hectare for mature trees and between 111 and 132 metric tons of carbon per hectare for younger cacao agro f orestry systems. Even though these levels of carbon sequestration are lower than those found in natural forests at 307 metric tons per hectare, much more carbon is sequestered in a cacao agroforest than can be a chieved in fields that only plant annual foo d crops ( Rice and Greenberg, 2000). The carbon that is sequestered has been proven to be a well protected form of carbon so that it cannot be easily releas ed to the environment (Ga ma Rodrigues 2009 ). Household Use Value of Biodiversity In addition to the environmental services that cacao agroforestry systems such as that used to produce cacao nacional, provid e, this production method allows these households to meet their consumption needs. Many other products such as plantains and fruit that are importan t to the diet of Ecuadorian farm families are included in these biodiverse, traditional cacao production system. Even if a household may be able to earn much more cash income for cacao CCN 51, exclusively pursuing this a lternative may not be the best alte rnative for a family as they would not be able to meet their subsistence needs because cacao CCN 51 is generally raised in a monoculture system without plantains or fruit trees This tradeoff between working for cash income and providing subsistence was s hown in a study in Brazil nut gathers in the Brazilian Amazon. Household consumption needs were not met as smallholder farmers dedicated their time to collecting Brazil nuts instead of growing subsistence crops ( Ros Tonen et al. 2008) Cash income may no t be very important these Ecuadorian

PAGE 24

24 households as many of them do not depend on the purchasing of goods in the market to mee The use value of these biodiverse production systems is shown in a study of Cam eroon agroforestry systems. This research found that smallholder farmers received the highest level of net present value when a cacao agroforestry system included shade trees that provide an economic benefit such as fruit and/or timber trees. This system had a higher net present va lue than more commercial systems that had a few shade trees of one variety. Th ese additional values provide an economic incentive to raise cacao in an a groforestry system (Duguma 2001). These diversified cacao, fruit, and timber agroforestry systems were found to have greater financial returns than plantations using conventional, le guminous trees. These systems also provide more consistent income from the multiannual harvest of cacao and other fruit trees in the cacao agroforestry system (Beer et al. 199 8 ). Thus, smallholder households that raise cacao nacional in such a diversified agroforestry system obtain additional consumption and use values along with the nonmarket values that would make it the preferred production method for Ecuadorian smallholder producers

PAGE 25

25 CHAPTER 3 THEORETICAL FRAMEWORK Mod eling Specialty/Fair Trade Marke ts The unique market for cacao nacional is similar to fair trade markets in that in both fair trade and specialty cacao nacional markets farmers receive a premium for their prod uct Thus, the work developed to explain how the fair trade markets influence can be applied to this situatio n of the specialty cacao nacional market. A model built by LeClair (2002 ) and critiqued by Hayes (2008 ) shows how the premium paid to participants in the fair trade markets affects labor allocati ons in the goods for this market. The model explains how a household would supply more labor to the production of the fair trade good due to the fair trade price premium. income from product ion as Y = ( P r ) Q (3 1) where Y is household income P is the market price for the good r is per unit costs and Q is the quantity of the good produced The model is then transformed for the income o f a household that participates in the fair trade market Y = ( P (1 + a ) r ) Q (1 + b ) (3 2) where a is the premium paid for the fair trade market or the specialty cacao market b is the additional output caused by the labor increase This model of the fa ir trade market can be transformed for shadow income ( y *) which is developed later in the next chapter T he shadow income includes the

PAGE 26

26 additional value from biodiversity that is not in cluded in the cash income the household receives. Thus, the shadow inc ome represents the true value from production for the household. By the model developed by LeClair, shadow income in the standard market where no premium is given for cacao is y* = P C Q C + P O Q O + V total costs (3 3) where P C is the price for cacao in the standard market Q C is the quantity of cacao produced P O is the price for the other crops on the farm Q O is the quantity of other crops produced on the farm and V is the value that household obtains from the biodiversity externality Our mo del alters to reflect an agroforestry situation and specifically the cacao agroforestry system. Our model is applicable to many smallholder farmers as they commonly practiced diversified cropping systems that provide addi tional ecological benefits Our model include s the income from cacao as well as the other agricultural products raised in the diversified system The value the household recognizes for biodiversity ( V ) is also included In our model the cost function is calculated according to l and area. In a diversified cropping system, a farmer cannot disaggregate the inputs utilized for one crop from another. The entire produc tion costs must be subtracted from the whole system rather than calculating the costs for each unit of production for each corp Furthermore, the applicat ion of inputs by smallholder farmers is considered in terms of amount applied per unit of land. So, the costs would be measure in terms of units of land than by the quantity of the product produced

PAGE 27

27 T he shadow income f or the specialty market that includes the price premium ( a ) and the production increase ( b ) from added labor would be y* = ( P C (1 + a ) Q C (1 + b )) + P O Q O + V total costs (3 5) An important distinctio n between our model and l includes the value of biodiversity. These ecological impacts must be included to better the inclusion of this factor is important in the evaluation of the fair trade and specialty markets since environmental improvement is the objective of many of these marketing schemes in order to better understand its full impact. LeClair argued that the premium could induce participants in fair trade markets to dedicate labor to thes e products thus increasing the supply and lowering the price of the good in the general market for all producers including those who do not participate in the fair trade markets. However, b y including the value of biodiversity in our model, an additional positive benefit from fair trade and specialty markets must be included in agricultural markets. The premium could induce farmers to produce goods that enhance biodiversity which would be a great societal benefit which contour LeCl at a macro level that examines how the premium from fair trade or specialty mark ets a ffect market prices and labor supply in the entire system The model does not examin e the effects for a n participation i n specialty markets. I n addition, a need exists to understand the impa ct of these specialty markets at the individual household level. A household level anal ysis decisions Shadow prices provide a means to analyze the impact of these markets at

PAGE 28

28 the household level while including the impact that biodiversity on production decisions. T he shadow wage for the sale of goods inside and outside the specialty or fair trade marke ts can be compared to determine which market provides more value for the smallholder household. In the following section, the household model is used to build a shadow wage to examine value that smallholder households place on the various planting options to make production decisions. Current Conceptualization of Household Shadow Price Models The household model provides a better explanation than the traditional profit maximization models for the production decisions of small landholder households The h ousehold model as developed by Barnun and Squire (1979) integrates consumption a nd production decisions with the objective of maximizing household utility. The model depicts consumption and production decisions as not being made separately ; they are made simultaneously. Unlike commerci al farmers who view their farm s as a business in order to earn income to purchase products for their household needs smallholder farmers do not separate these consumption and production activities. A smallh older decisions which would not be captured in a traditional model that does not consider these consumption and production objectives simultaneously. A household will choose to maximize its utility t hrough consumption and leisure. G iven a budget constraint, the value of time is compared by family members to the to be able to consume goods. While the work that the family undertakes off the farm is generally valued at a market wage with money that is earned used to purchase needed goods the work on the farm c an be valued in monetary or non monetary terms In

PAGE 29

29 particular, many of the positive attributes or amenities g enerated from existing biodiversity in the farm cannot be traded in markets Our theoretical model is built upon the model of Le (2009), who developed a shadow wage and income using to explain imperfections in the labor market The household model to buil d shadow wages, which is the marginal product of labor (MPL), was first used by Jacoby (1993) to explain the wage that the labors perceived. This work was furthered by Arslan and Taylor (2009) who develop a shadow price to explain the additional value th at Mexican households place on traditional maize varieties that are not reflect ed in market prices. Our model expands on this work by adding a value for environmental factors to the shadow wage and income. Specifically, the value of biodiversity as perce ived by Ecuadorian farmers is a component of the shadow wage and income these farmers realize. Framework for Valuing Biodiversity Our model is based on a farm household that maximizes a utility function of leisure ( l ) and consumption ( C ); U ( C l ; D ) where D is a vector of household characteristics (e.g. preferences, number of children). C includes goods that are produced at home ( v ) or purchased in the market ( c ); C = c + v The labor supply ( h ) is the total stock of time for the household ( T ) minus leis ure ( l ) which is divided among two activities farm labor ( L ), market labor ( M ); T l = L + M = h The production function for the farm is F ( L z ; A ). z is a vector of variable inputs (e.g. hired labor, fertilizer, insecticides etc.) and A is a vector of q uasi fixed assets (e.g. land). T he production function F for the diversified cacao agroforestry system is specified as F ( L z; A ) = Q (L z; A Q ) + V ( L,z; A V ), where Q is the cacao production and V is the subsistence and environmental benefit derived from t he production method. The

PAGE 30

30 production function takes this form since cacao is commonly grown jointly with other crops or trees (such as fruit trees, cassava, and plantains ) for self consumption. Since the two types of crops are grown in the same parcel, f armers do not account for the amount of inputs used for each crop. Rather, purchased inputs oriented toward producing the cash crop have the external effect of benefitting the production of intercropped crops and trees. Several authors have shown that whe n markets are imperfect or missing, profitability variables that are commonly taken as exogenous become endogenous. Specifically, they become functions of household char acteristics and preferences Eswaran and Kotwal (1995) explained the case of imperfect c redit ma r ket s. Carter and Yao (2002 ) discuss ed imperfect land rental markets This situation is due to the fact that the fa rm household problem becomes inseparable are not independent from household co nsumption decisions In this in separable environment, A Q is an endogenous variable in the cacao production function The household maximiza tion model can be summarized as Max U ( C, l; D ) (3 6 ) subject to pQ ( L, z; A Q ) + V ( L,z; A V ) = p z z + + R + V ( L,z; A V ) C and > M where p is the farm output price, p z is a vector of input prices, are the market wages, and R is other income.

PAGE 31

31 The Langrangean of the problem can be expressed as L = U (C, T h; D)+ 1 [ pQ (L,z;A Q ) p z z + M + R + V(L, z; A v ) C ] + 2 ( M) (3 7 ) The first order con ditions (FOCs) are + 1 = 0 (3 8 ) where i = MUC + 1 [ p V L ] = 0 (3 9 ) In the FOCS the p ( ) equals MPL p V L = ( )/( ) = i (3 10 ) i is the shadow wage, which includes the additio nal hous ehold production values that occur to the household outside of the market. The shadow wage is the marginal rate of substitution between leisure and consumption (MRS LC ) MRS LC (3 11) E quation 3 11 reflects the fact that the househo ld will choose to invest time working on total marginal product equates the MRS LC The FOCs in E quations 3 7, 3 8, 3 9, and 3 10 include the value. The contribution of biodiversity to the model is evident when Equatio ns 3 7, 3 8, 3 9, and 3 10 are compared to E quations 3 1 2, 3 13, and 3 14, which represent the standard methodology employed to determine the shadow wage in the household literature: i = = MVC (3 11 ) 1 (3 12 ) = p ( ) = )/( ) i (3 13 )

PAGE 32

32 Previous studies ignor e the biodiversity value include in the second term of E quation 3 9 This omission leads to a bias in the estimated shadow wages, when there are externalities derived from the pro duction method. An important consideration is that in E quation 3 13 will be equal to the market wage when labor marke ts are perfect ( > M) but will be lower than when these market are imperfec t and the labor constraint binds. Following other household studies (Le 2009), the labor market failure is modeled as one of limited demand for labor. When this constraint binds, the production part of the shadow wage of labor, is below market wage. However, an important highlight is that under other typ es of labor market imperfection can be higher than the market wage. For example, when family labor and hired labor are not perfect substitutes, then the household is limited in its abili ty to prov ide farm family labor by the number of household members. In this case, when the marginal productivity of family labor on the farm is very high, the shadow wage of labor may be higher than the market rate. Finally, even when the marginal value of the caca o productivity of labor is lower than market wage, i.e. < W, there is still potential for the total shadow wage of labor to be above market wage. This situation occurs when there are positive benefits derived from biodiversity in the production system. Countervailing Market Imperfections A positive effect of biodiversity, on shadow wages (E quation 3 9 ), can be seen as a counterv ailing effect market failures which may offset the difference between and derived from labor market imperfec tions. The value from biodiversity would increases the sha dow wage high enou gh so that higher benefits are realized for

PAGE 33

33 farm work relative to market work, which could explain the lack of participation of farm households in the labor market. Th is hypothesis is explored in the estimation s in Chapter 4 T he term re flecti ng marginal benefits of biodiversity from labor used in production, is difficult to measure because of both the multidimensionality of V (composed of self consumption, medicinal plants, shade for the cacao, education of children, continuation of cultural identity) and the missing markets for some of its components ( such as cultural or traditional values). In this analysis, a proxy measure is used ( see Chapter 4 ), which allows for the estimation of a lower bound for the shadow wage. Imperfect Segmented Product Markets Another practice that is used in the literature in order to simplify the analysis is the normalization of the market price p by dividing all variables expressed in dollar terms by an index that reflects the market price of produced goods. The problem with this strategy in a setting of imperfect or segmen ted markets is that farmers often face heterogeneous prices, depending on whether they have access to markets or not. In this case, this type of imperfection arises because of the existence of specialty markets with limited demand. Thus, separate shadow w ages are estimated depending on whether farmers have access to specialty markets or not. The estimation of the shadow prices are created using data collected from smallholder farmers in northwestern Ecuador. A description of the study site, survey methods and results are included in Chapter 4. This data is then used to calculate the shadow prices according to this theoretical framework, which considers the value of biodiversity and segmented specialty markets.

PAGE 34

34 CHAPTER 4 EMPERICAL IMPLEMENTATION Research Site Puerto Quito, and Quininde were interviewed in July and Aug ust of 2009 about their production methods, access to alternative markets, impressions of the profitability of th eir operations, perceptions of environmental quality, and satisfaction wi th their well being A stratified sampling method was utilized that included participants from the three counties as well as from various socioeconomic classes and ethnic groups. T he sample include d mostly small and me dium sized farming operations with 31 (62%) of the farmers surveyed having less than 10 hectares. S ome of the most marginalized in Ecuadorian society either through ethnic discrimination or poverty are included in the households who until 100 years ago were the only inhabitant s of the Santo Domingo area, were included in the sample along with four Afro Ecuadorian households The rest of the participants in the survey are mestizo s of mixed indigenous, Afro Ecuadorian, and Spanish ethnicities. Th is stratified sampling method allows the survey to ad dress the concerns raised in a previo us case study by Nelson and Galvez (2000) that found that the most marginalized populations had li ttle access to the alternative cacao markets. Puerto Quito are located in the northern coastal region of Ecuador ( Figure 4 1) These three counties are located in three different prov inces. Puerto Quito is in the Pichincha p rovince, Quininde in the Esmeraldas province and Santo Domingo in the recently created Santo Domingo province Although there are several jurisdictions, the

PAGE 35

35 economic and cultural activity of the region center s on the main trade city of Santo Domingo, a bustling frontier city located 150 kilometers southwest of the Ecuadorian capital, Quito. Within the last fifty years, this region was settled as the Ecuadorian government encouraged residents of the mountainou s region of the country to inhabit these less populated areas. This region continues to rece ive immigrants from other areas growing city growing with an annual population growth rate of 3.7% Due to this continuing influx of immigrants and transient population, the true population of Santo Domingo is unknown. However, it was estimated in 2007 to have 330,000 permanent residents and over 500,000 people living in the province at any one time. In 2007, Quininde was estimated to have 150,000 habitants; Puerto Quito was estimated to have 17,200 residents. While Santo Domingo is mostly an u rban county with over 70% of its residents living in the city, Quininde and especially Puerto Quito are largely rural counties (Provincia de Santo Domingo de los Tsachilas 2010; Municipalidad de Puerto Quito 2009; and Municipalidad de Quininde 2009). Besides the immigrants that moved to the area there is a small but prominent Afro Ecuadorian and i ndigenous population in the area. The region of Santo Domingo was originally inhabited by the communities around the city of Santo Domingo. This group was traditionally hunters and gathers Today, they have become farmers due to destruction of the forests where they have had a large impact on the region and especially in the rural areas because of their historica l significance and because of the large amount of land located within their

PAGE 36

36 communities (Gobierno de la Provincia de Pichincha 2003) There is a large Afro Ecuadorian population in the county of Quininde. The province of Esmeraldas is well known for its Afro Ecuadorian residents many of whom are descendants of slaves during the Spanish Colonial period Slaves had originally had been brought to Quininde from Colombia to harvest rubber in the 1800 s and stayed in the area after they were freed. In fact, th e Afro Ecuadorians were some of the first residents of this region (Municipalidad de Quinind 2009). This region is a particularly poverty stricken area of Ecuador. Little data is available on the economic status of this region because of the transient na ture of the population. High poverty rates are especially evident in Santo Domingo and Quininde were little of the city lacks access to running water and very few of the streets of these major population centers are paved. The rural areas have even less access to these basic services although almost everyone has been connected to the electrical grid particularly marginalized and often much poorer the re st of society. T he sympt oms of poverty are clearly evident in these populations as they lack to means to access adequate healthcare, education, and nutritional sources. These communities have abnormally high rates of infant mortality, tuberculosis and malaria infection, juvenile malnutrition, as well as low literacy rates as just more than half of members of the (Gobierno de la Provincia de Pichincha 2003) These three counties straddle and the equator so that the region has a very warm subtropical These counties were once covered by lush tropical forests rich in

PAGE 37

37 biodiversity. This climate is also conducive to raising many varieties of tropical crops including plantains, passion fruit, cacao, African Palm, organs, pineapple, and coffe e. Now, much of this forest has been lost as the settlers removed the trees to buil d their towns and cities and farms Because of this rich biodiversity and extensive deforestation many endemic speci es are threatened with extinction. This threat has pr omoted this region to be included within the Tumbes Choc M agdalena biodiversity hotspot (Conservation International 2007) Agriculture is the main economic driver in these three counties. Even the urban province of Santo Doming o is heavily dependent on the trade of a gricultural goods. Plantains and cacao are the main crops in the Santo Domingo area with many other tropical crops grown such as malanga cassava, and passion fruit. Quininde has become well known for the production of African Palm on larg e plantations. Much of economy depends on the harvesting and process of the seeds of this palm for palm oil. However, most of the smallholder farmers are dependent on cacao production as they lack extensive landholdings to raise African Pal m The county of Puerto Quito is composed mostly of smallholder farms and has the advantage of being at a higher altitude than Santo Domingo and Quininde so that besides growing man y the tropical crops they are able to raise some cooler weather cro ps Ma ny households in the three counties also raise livestock. Almost every farm has a few chickens and many have pigs. Beef cattle are raised in all three counties w hile many farmers from Puer to Quito have dairy operations. Survey Dissemination T hree groups of farmers were included in the survey The first group includes farmers who sold their cacao to a processing center run by a local nongovernmental

PAGE 38

38 organization, Fundacin Accin Social Caritas (FASCA) in Puerto Quito. This organization is the social arm of the Catho lic Diocese of Santo Domingo. The processing center called BIOFASCA uses the cacao purchased from the farmers to make cacao paste which is then exported to high quality chocolate producers in Spai n and Italy. BIOFASCA purchases cacao in baba in wet form, before it is fermented or dried so that it can complete the processing process to highest st andards. Normally, the middleman purchases the cacao from the farmers once it h as been dried and fermented. The middleman then discount s farmers wh en they do not complete these processes to the highest quality standards. T he farmers receive a better price for their cacao from BIOFASCA as their product is not disconnected. In addition, FASCA pays for the orga nic certification costs for 147 farmers a nd then pays a premium for this organically certified cacao Wo rkers at the plant purchase the ca cao directly from the farms so that th e farmers do not have to pay transportation costs An additional benefit of the FASCA cacao project is that it also pro vides technical assistance to any farmer who sell s cacao to them would call him and ask for his help, so we never knew which farms we would visit in a given day which made the s ampling of the farmers random. In addition to the technical trips with the plant manager, I interviewed a few farmers as they came to the plant to sell their cacao. I was able to extensively interview 10 farmers in the Puerto Quito region which represent s 20 percent of the study. I also had the opportunity to talk in depth with the plant manager, the director of FASCA, and workers at the plant about

PAGE 39

39 their thoughts and feelings about BIOFASCA as well as the cacao market and organic and fair trade markets. I visited several of the communities around the city of Santo Domingo de los During t hat time, oppo rtunity to interview s 15 households I also visited a few communities of the other communities I had previously assisted to plan a cacao conference in 2007. I stayed with a farm family in the village of Roca Fuerte for several days as a farmer took me to interview his neighbors T he interviews in the Santo Domingo area provided me with opportunity to talk with farmers who were not members of farmers asso ciations nor had access to the organic and fair trade mar kets as the farmers. As Santo Dom ingo has by far the largest population of the three counties involved in my sample, the number of surveys conducted in this province is much greater than that of the other two with 30 farmers being interviewed or 60% of the total of those surveyed. The fin al group of farmers that I visited was a group that participated in a larger Zrate Cantn Quininde (UOCAQ), is building a cacao collection center similar to BIOFASCA process plant that control the drying and fermentation processes in order to so produce the hi ghest quality cacao for export. They are currently working with Nestl, which has offered to pay them up to 35 U S dollars more per quintal (100 pounds) for organic, cacao nacional if all the quality standards are met. This pr emium is quite large since

PAGE 40

40 cacao generally does not sell for more than 100 U.S. dollars per quintal. In addition, Nestl is working with farmers to raise improved cacao nacional trees developed through the assistance o f French agronomists. Besides selling to Nestl, the cooperativ e is also examining the possibility of selling their cacao i n fair trade markets While in the Quini n de area, I traveled to several villages with the director of UOCAQ and an Ecuadorian agronomist, who works for Veterinarians and Agronomists without Border s. This international NGO has been supporting UOCAQ in the construction of the collection center as well as providing the farmers w ith technical advice as they work to become organica lly and fair trade certified. The agronomist provided the farmers with production and marketing advice as well as listened to the ir concerns about the project. I was able to interview the farmers from the various s as we visited their farms. I never knew which farmers I would be interviewing until I arri ved in each community. I interviewed 10 farmers in the Quininde area, which repres ent 20% of those surveyed. The results of the survey are included in the following section s The first section describes the basic characteristics of the household s in the survey the second includes the characteristics of the parcels of cacao nacional and cacao CCN 51, the third examines the profitability of each variety, the fourth describes the farmers perception of profitability of cacao, the fifth describes farmers acce ss to specialty markets, the sixth farmers likelihood to plant more cacao, and the last descriptive section details farms environmental perceptions. The last section is devoted to the creation of equations to the estimate the shadow wage and income.

PAGE 41

41 Char acteristics of Households that Produce Each Variety of Cacao There are distinct differences between those households that that raise cacao nacional and those that raise cacao CCN 51 (Table 4 1). Households that produce cacao CC N 51 are wealthier than hous eholds that rais e cacao nacional. Household incomes were recorded on a scale form one to nine. Household of one earned less than 100 U S dollars per month and those who said their income was over 800 U S dollars were listed with a nine. The average rating of total household income for families that raise cacao CCN 51 was 4.87 with a median of five while households that raise cacao nacional had an average score of 2.44 with a median of three. As households in both categories are statistically similar in size, household incomes can be compared on a one to one basis. So, there is a difference of at least 200 U S dollars between the total income that most households that raise cacao CCN 51 and those that raise cacao nacional. This difference is especially large in Ecuador as the minimum wage in Ecuador in 2009 was 218 U S dollars per month and the poverty line was 56 U S dollars per person per month. Not only is average total income statistically different between the two type s of households, but average farm income is also statistically different although the difference is smaller Households that raise cacao CCN 51 on average earn nearly 100 U S dollars more in farm income than those farmers who raise cacao nacional. Other i n dicators of household wealth reveal a significan t difference between households that raised cacao nacional and those who raise cacao CCN 51. Those households that raise cacao CCN 51 have larger homes and farms than households that raise cacao nacional. CCN 51 households have farms that are on average 28.44 hect ares compared to 13.08 hectares for families that raise cacao nacion a l. In addition,

PAGE 42

42 households that raise cacao CCN 51 are more likely to own a car and have running water as compared to farmers w ho raise cacao nacional. Fifty percent of CCN 51 households own a car a s compared to just 15.15% of households that raise cacao CCN 51. Twenty five percent of households that rais e cacao CCN 51 have running while just 6.25% of cacao nacional households h ave this service. Parcel Characteristics The parcels where cacao nacional is raised are larger, have older trees, and have a richer diversity of other crops and trees than parcels planted with cacao CCN 51 (Table 4 2 ) C acao nacional trees are statistical ly significant older than cacao CCN 51. As c acao nacional has been raised in Ec uador for over 150 years and cacao CCN 51 for only 10 years, the cacao nacional parcels would be expected to be older. The average age of cacao nacional is 10.94 years old and for cacao CCN 51 is 3.94 years old. Although households that raise cacao CCN 51 have larger farms, these households plant their cacao on smaller plots that are more de nsely planted. The average parcel size for cacao CCN 51 parcels is only 2.57 hectares while cacao nacional plots have an average size of 4.50 hectares with cacao CCN 51 trees being planted at a density of nearly 700 trees per hectare while cacao nacional trees were planted at a density of about 500 trees per hectare. The fact that cacao na cional farmers plant their plots with fewer cacao trees allows them to plant other crops and trees dispersed between their cacao trees. In addition, cacao nacional requires more shade than cacao CCN 51 so more shade trees must be pla nted in these tradition al cropp ing systems. The survey results confirm the observation that the cacao nacional parcels are raised in diversified agroforestry systems while cacao CCN 51 is raised in less diverse

PAGE 43

43 systems. Farmers, who participated in the survey, were asked how ma ny varieties and which varieties of crops and trees they planted in their cacao parcels. This study considered that the more variety of species of plants that were included in a parcel the more biodiverse the parcel. Both native species as well as introd uced crops were included in this biodiversity measure. Cacao nacional farmers plant over two varieties of perennial crops with their cacao and at last one other tree variety while cacao CCN 51 are most likely to plant only one other perennial crop with th eir cacao. Thus, cacao nacional is raised in more biodiverse cropping systems that may provide additional ecological benefits Production and Profitability Values Our survey results reveal a clear difference between cacao naci onal and cacao CCN 51 in the use of labor and other inputs, yield, and profits from each production system. Households that raise cacao nacional are more dependent on family labor and utilize more labor overall for the production of cacao. Comparing the two cropping systems on a per hectare basis does not reveal the true difference between the two systems as cacao CCN 51 is planted much more densely than cacao nacional. In order to make an equal comparison the two varieties need to be compare d as if they were planted at the same den sity per hectare. A useful comparison can be made if the parcel area is converted to the density recommended by the agronomist at BIOFASCA, 625 trees per hectare. The hectare equivalent makes for a better comparison than on a per hectare basis for costs and profits between the two varieties. A parcel with less a tree density of less than 625 cacao trees per hectare would have a hectare equivalent smaller than the actual hectare size while a parcel with more than 625 trees would have a hectare equivalent value larger than the number of hectares in the parcel. Basically,

PAGE 44

44 the hectare equivalent designation allows each variety to be compared on a tree for tree basis. This term, hectare equiva lent, is utilized throughout this study to reference this conversi on Table 4 3 lists the production costs, yield, revenu e and profit on a per hectare basis while Table 4 4 lists these variables on a hectare equivalent basis The traditional cacao nacional production is much more labor intensive, less productive, and le ss profitable than cacao CCN 51. There is not a difference in the amount labor utilized on cacao CCN 51 production systems and the amount utilized on cacao nacional production systems when compared by labor use per hectare with an average of 33.62 days pe r hectare used to produce cacao CCN 51 and 33.96 days per hectare to produce cacao nacional. However, there is a significant difference in labor use when compared by hectare or equivalent with cacao nacional production systems using 63.77 days of labor pe r hectare equivalent while the cacao CCN 51 production system uses 28.75 days of labor per hectare equivalent. This difference in labor use by hectare equivalent is a better measure of input use as it compares an equal number of trees. In particular, the re is a significa nt difference between the days of family labor used with 37.72 days of labor per hectare equivalent in the cacao nacional production system compared to an average of 16.16 days per hectare equivalent to produce cacao CCN 51. No significan t difference was found to exist in the use of hired labor for the production of cacao CCN 51 and cacao nacional. A large diff erence exists in the use of other inputs such as fertilizers, insecticides, and herbicides between cacao nacional a nd CCN 51 produc tion systems. T he cacao CCN 51 system uses an average of 51.50 U S dollars in othe r inputs per hectare equivalent compared to 28.78 U S dollars per hectare equivalent for cacao nacional

PAGE 45

45 parcels. Households that raise cacao CCN 51 substitute othe r input s for labor especially herbicides. Weeding is one of the most labor intensive agricul tural activities of the households as it involves going to the fields and cutting the thick weeds with a machete. There is a clear difference in amoun t of labor used for weeding. Cacao nacional production uses 29.14 days of labor per 625 trees and cacao CCN 51 production utilizing 9.24 days of labor per 625 trees. CCN 51 production uses chemicals to control these weeds while farmers who raise cacao nacional are use much more labor to control their weeds. The market price for hired labor does not properly represent the cost the household realizes for utilizing this labor. First, many of the household members would not be able to participate in these mark ets such as woman and children. With few options for employment, the opportunity costs of many members of the household would be much less than the market wage. As many household members do not have any other option for employme nt their opportunity cost woul d be near zer o. Second, the household is really paying itself for this work. I n order to try and capture the real cost for family labor, the costs and profits were calculated both as if family labor were valued at the market rate and at zero to examine the difference s in the two values. The true cost to family labor is likely located somewhere between these two values. When family labor is valued at the market rate the total costs for raising cacao nacional is much higher than that for cacao CCN 51. On average cac ao nacional costs 564.61 U S dollars per hectare equivalent to grow while cacao CCN 51 costs on average 304.40 U S dollars per hectare equivalent. When family labor is valued at zero, then there is no significant difference in the cost to raise either v ariety, cacao nacional

PAGE 46

46 production costs average 258.40 U S dollars per hectare equivalent and cacao CCN 51 averages 165.14 U S dollars per hectare equivalent. Most farmers measure their production on per hectare basis which reveals that cacao CCN 51 is m uch more productive than cacao nacional. In fact, higher yields are the main reason why cacao CCN 51 is touted as a preferred alternat ive to cacao nacional (Melo 2009 ). The study confirmed this pe rception that cacao CCN 51 has significantly higher yield s The farmers that participated in oyr study revealed that the average yield for cacao CCN 51 between September 2008 and August 2009 was 16.41 quintal s of cacao per hectare compared to 8.97 quintal s of cacao per hectare for cacao nacional. These values a production for cacao nacional and cacao CCN 51 which is 8 quintal s for caca o nacional and 16 quintals for cacao CCN.51. H owever, this measurement of yield is not a true measure of cacao production on t different for each variety. A much better measure would to compare the yield a tree by tree basis, which our study does by using the estab lished hectare equivalent conversion Using this measure average yi eld for cacao CCN 51 during this time p eriod was 14.80 quintal s per hectare equivalent compared to 11.66 quintal s for cacao na cional per hectare equivalent. These values are not significantly different. Households receive significantly larger revenue for cacao CCN 51 than for cacao nacional. In standard markets where farmers do not receive price prem iums for cacao nacional, households who raise d cacao CCN 51 received on average 1 393.97 U S dollars per hectare equivalent for their cacao in September 2008 through August 2009 while the average revenue for cacao nacional was significantly smaller at 834.88 U S

PAGE 47

47 dollars per hectare equivalent. Households that raise cacao nacional and had access to markets, which pay a premium for this variety received 1 060 .86 U S dollars per hectare equivalen t. The average revenue for organic cacao producers was 1 026.95 U S dollars per hectare during the year studied. Cacao CCN 51 also proved to be significantly more profitable than cacao nacional in both the tradition al commercial market and when a premium was received for cacao nacional. There was a significant difference between the revenue of cacao CCN 51 and cacao nacional even when the value of f amily labor was discounted. T he avera ge profit for cacao CCN 51 was 1 223.84 U S dollars per hectar e equivalent and the average profits for cacao nacional are 608.65 U S dollars per hectare equivalent and 677.90 U.S. dollars per hectare equivalent in the specialty markets. Surprisingly, t he few households that raise or ga nic cacao are not profitable. When family labor input is included the organic farmers had an average loss of 182.32 U S dollars per hectare equivalent and only an average profit of 233.57 U S dollars when family labor is not valued. This result is s urprising as organic cacao has a higher market value than cacao sold in either the traditional commercial market or the premium paid for cacao nacional. Farmers stated that the average organic cacao price ranged from 90 to 110 U S dollars per quintal. T his price is much higher than the price the farmers received in the specialty markets, 75 to 100 U.S. dollars per qu intal, and in the standard markets, 60 to 90 U.S. dollars per quintal Smaller yields and greater labor costs would make organic cacao much less profitable. Perception of Profitability One section of our survey asked farmers to rate the profitability of their cacao parcels from very good to very bad for their cacao nacional and cacao CCN 51 parcels

PAGE 48

48 and their farmers as a whole The questions were asked with a time element as farmers rated the profitability of these three entities six years ago, three years ago, now, and three years in the future. A time element was included in order to try and measure sustainability. These scores were conver ted from one to five with the very good ratings given to a fi ve and the very bad identified with one. These values wer e then averaged (Table 4 5 and Figure 4 2 ) Farmers on average gave significantly higher profitability ratings for both varieties of cacao than for their entire farm Although the differences in perceptions of profitability for cacao nacional and cacao CCN 51 are not significantly different, t he results show that famers believe that cacao nacional was more profitable than cacao CCN 51 six and three years and that it will be in the future. However, they see cacao CCN 51 as being more profitable now This observation may be due to the fact that the average ag e of cacao CCN 51 in the survey is only about four years. Since cacao provi des its first harvest at about three years of age, cacao CCN 51 was not produc tive during this period. As cacao CCN 51 is more productive than cacao nacional farmers rate this va riety as more profitable. They may believe that cacao nacional prices will be higher in the future as new markets develop for this variety. Thus, they gave this variety a high profitability rating for t he future. As households are more dependent on fami ly labor for production cacao nacional which may have less opportunity costs than market labor, farmers may view cacao nacional as less costly, thus, more profitable Furthermore, a s the market has responded by providing price premiums for cacao nacional and organic cacao, farmers seem to believe that they will receive much higher prices for cacao nacional in the future so that it will be more

PAGE 49

49 profitable than cacao CCN 51. They seem to believe that the price advantage will make up for the production diff erences Access to Specialty Markets Our specialty cacao market s The farmers were asked if they had access to markets were they are paid premiums for cacao nacional or markets for organic cacao. If they lacked access to these markets, they were then asked if they would like to have access to these markets and what factors inhibit ed them from having access to the more lucr ative markets. Only 26% of households had access to the premium markets for ca cao nacional. Of those households that did not have access to these marke ts, 42% of households would like to have access, 22% are unsure if they would like to access t hese markets, and only 6% were not interested in having access in these markets. Of those wh o did not have access, 43% were in the process of trying to gain access to the market, 35% were unaware as to how t o access this market, 17% claimed that they did not produce enough cacao to sell the markets, and 5% thought that there was no demand for in this market. Even fewer households had access to organic markets, 6% of farmers having access to these markets. Yet, 58% of households would like to have access to organic markets. Of these farmers that woul d like to have access 79% listed that cost an d knowledge about the markets prohibited their entry into this market. These results complement the results of the case study by Nelson and Galvez (2000) which found that many cacao farmers lacked access to fair trade markets. They found that the m ost re mote, marginalized households lacked access to these more lucrative markets.

PAGE 50

50 Likelihood to Plant Cacao intentions to plant cacao. The farmers that participated in our survey wer e asked if they would like to plant more cacao trees, which variety they would like to plant, and what factors inhibited them from planting cacao ( Figure s 4 3 and 4 4 ) Overall, they had high expec tations for cacao as 88% of families would like to plant c acao and of those who would like to plant cacao 65.9% would like to plant cacao nacional. This result appears to contradict the traditional theory that households act as profit maximizing rational actors A p rofit maximizing household should not choose t o plant cacao nacional instead of cacao CCN 51 as cacao CCN 51 is more profitable These planting These households would appear to receive additional value from cacao is captured in the market which would induce to plant cacao nacional Perceptions of Environmental Factors The section of our survey that examined environmental perceptions was conducted in the same format as that of the profi tability perceptions. The f armers rated these perceptions over the past six years and into the future. A one to five scale of measuring the responses from very good to very bad was also used. T he farmers were asked to rate their perceptions of water quality, soil fertility, and bi odiversity o f their entire farms as on their fields of cacao nacional and cacao CCN 51 (Table 4 6 and Figure 4 5 ) Farmers clearly perceive d caca o nacional parcels to have better soil quality over the entire period than the entire farm and that of cacao C CN 51. Cacao CCN 51 was perceived to have better soi l quality than the farm now and the future. One can question if farmers plant cacao in parcel s that have higher soil quality than the

PAGE 51

51 rest of their crops. Then the higher soil quality ratings would be due to planting decision rather than if cacao does indeed improve the soil quality. This question is refuted when the ratings for cacao CCN 51 are observed and the age of the trees are considered The farmers in our survey as well as those who were ea rly include d in the study by Bentley, Boa, and Stonehouse (2004) had recently p lanted cacao CCN 51. They had planted this variety of cacao within the last ten years and the majority had planted it within the last five years. The farmers rated the soil q uality of parcels planted with cacao CCN 51 to be worse than that of the entire farm six and three years ago when the seedlings would have recently been planted which would disapprove that cacao is pla nted in better soil. F armers may have chosen to plant cacao in poorer than avera ge soil The reality that cacao CCN 51 parcels have higher so il quality ratings than the farm now and the future while have lower ratings in the past would demonstrate that cacao may improve soil quality For biodiversity percept ions, cacao nacional has a higher average rating than the farm in all time periods and cacao CCN 51 has nearly equal ratings as the entire farm The farmers are nearly unanimous in stating that they had witnessed a large loss of native plants and animals on their farms over the last few years. However, the loss is rated to be less for cacao and statistically significant less for cacao nacional. These findings support the ecological research t hat claim that cacao agroforestry systems protect and even enha nce biodiversity in comparison to other cropping methods Yet, the farmers believe that the loss of plants and animals is so grave that the agroforestry system may not provide enough benefit s to prevent a general decline biodiversity on the

PAGE 52

52 farms Cacao agroforesty systems appear to a second best solution as compared to native forests in controlling the loss of native plant and animal species. The descriptive statistics from our survey as well as the research on the ecological impact of cacao agroforestry systems reveal the importance of nonmarket values in the production of cacao. Smallholder households utilize these values in making production decisions. The shadow wage and income as constructed in Chapter 3 provide a solid framewor k for analyzing thes e additional values that are included in a smallholder The following section uses our model in Chapter 3 to calculate a shadow wage and income for cacao producers in Ecuador that includes the value of biodiversity and accounts for th e segmented specialty markets Estimation In order to estimate the shadow income of Cacao Nacional, the shadow wage defined in E quation 3 9 are used. The budget constraint can be transformed to a linear budget constraint while allowing the household to st ill arrive at the optimal choice (Jacoby 1993, Skoufias 1994, Le 2009). The trans formed maximization problem is: max U ( C, l; D ) (4 1) subject to C = y* where y* is the shadow income y* = p p z z + M + R + *h (4 2) p describes the ma rket access while captures longer term value of household production including the value the household places on the ecological benefits of biodiversity. provides the shadow income from work.

PAGE 53

53 Solving the maximization problem and after estimating and y* ; the optimal labor supply function can be estimated: = ( ) (4 3) In our study, the focus is on the difference in shadow income across households that have differential access to specialty markets and who face different lab or constraints (Equations 3 9, 4 1, 4 3 ). First, is estimated using the method of Skoufias(1994) but extending it with an estimated value for biodiversity. A Cobb Douglas production function for cacao and biodiversity ( Q ( L, z; AQ ) and V ( L, z; Av )) is used to estimat e them in logarithmic form in order to make them additive and to recover estimates of the input elasticities. Multiplying the labor elasticity in each equ ation ( ) by the corresponding predicted average product of labor ( / L / L ), gives us t he marginal product of labor ( MPL ). In order to obtain MPL is multiplied by the price to be received for the cacao, increases by both incremental income as well as benefits from biodiversity, its decisions are taken by jointly considering these benefits. For this reason, the equations that predict cacao production and biodiversity are solved simultaneously. By discovering the additional value that a household realizes through biodiversity, the s hadow wage, which capture s this nonmarket additional value as well as the production value from cacao can be estimated Specific Estimated Equations Biodiversity i = 0 + 1 Log family labor + 2 Nacional*Log family labor + 3 Log hired l 4 Log other i 5 Flat + 6 Hilly + 7 Insecure l and rights + 8 Years land owned + 9 Ethnicity + 10 Household size + 1 (4 4)

PAGE 54

54 Production = 0 + 1 Log family labor + 2 Nacional*Log family labor + 3 Log hired l 4 Log other i 5 Flat + 6 Hilly + 7 Insecure l and + 8 Years land owned + 9 Ethnicity + 10 Household size + 2 (4 5) Definition of the Variables The variables in Table 4 7 were included in bo th regression equations that predict the number of varieties of cacao as the descriptive statistics, past research, and/or field experience suggest that these variables may be signif icant factors in determining cacao production and amount of biodiversity i n these cacao parcels. As in any production function labor and other inputs are needed to produce a good. T hese variables are an importan t component of the regression. Labor is divided into family and hired labor in order to examine the effect of the sh adow wage and to determine the substitutability o f family labor and hired labor. If hired labor and family labor have the same MPL which the coefficients represent, then they would be perfect substitutes and in a perfect market the MPL for both family and hired labor would equal the wage. All the labor coefficients are expected to be positive values for both equations as more labor would lead to higher production They would not be positive if the household was at the point of maximum output with diminis hing marginal product ion, which would be very unlikely A similar conclusion can be made about the variable for the other inputs. This coefficient would be expected to be positive for production although it might not be for the biodiversi ty equation. Th e production of cacao CCN 51 is more input intensive than cacao nacional as shown by the survey results. Since cacao nacional is produced on more biodiverse parcels than cacao CCN 51, a negative correlation would be expected to between input use and biodi versity. The labor value for cacao nacional is included in order to determine if a difference exists in the effect of the labor applied to cacao nacional on biodiversity and on

PAGE 55

55 production as compared to that of the other labor variables. Since cacao nacio nal production systems are more biodiverse than CCN 51 cropping systems, the coefficient of family labor in cacao nacional production in the biodiversity regression would be expected to be larger than the c oefficient for family labor for all cacao producti on. As cacao CCN 51 has higher yields than cacao nacional, the family labor and hired labor coefficients would be expected to be larger than the coefficient for family labor for cacao nacional alone in the production regression. The variables flat and hil ly are included in the regressions. The steep slopes would be expected to be the most biodiverse parcels. Farmers tend to plant the flat areas first as they are easier to mai ntain and harvest. T o prevent erosion the farmers in this area often leave the s teepest slopes forested as the steep areas are often on stream and river banks. Thus, the steeper areas would be more likely to be forest ed and thus more bio diverse than other areas. T he coefficient for flat would be expected to be a large negative numbe r while the coefficient for hilly will be a smaller n egative number. The variable for insecure land rights definitely would be expected to have an impact on the biodiversity regression. Ho useholds who fear that their property rights may be removed would o nly invest in crops with short harvests such as annual crops in stead of trees, which do not provide a revenue for several years Thus, the insecure land rights variable is expected to have a negative coefficient in the biodiversity equation. The variable for the initial land size is a measure of the wealth of the in order to increase production. A positive coefficient would be expected for this variable in the

PAGE 56

56 production e quation. The length of time that someone owned property would influence the biodivers ity equation. T he longer a household has owned land the more it would be expected to invest in trees and other crops with long term return. So, they would be expected to have larger biodiversity coefficients. Since cacao becomes more productive over time, the coefficient for the length of ownership of the parcel would be expected to be positive for the production regression. Ethnicity is an important variable in the equa tion. This variable is included as a dummy variable distinguishing between those househ olds that do and do not have a There is a significant population of e a distinct culture from the rest of the population. They believe that all plants and animals have spirits and need to be protected. This belief would suggest that they would be more likely to protect the various varieties of plants and animals on their land In addition, they use many medicinal plants that the y grow in their fields which would add the diversity of plants in their fields. So, t he ethnicity coefficient in the biodiversity regression would be expected to be a positive but may be negative in the production modern agricultural practices. Finally, the household size variable would be expected to be an important variable in the equations. The coefficient for these variables would be expected to be positive for both the biodiversity regression and the production regression. T he household would need to produ ce more to provide for the needs of additional household members

PAGE 57

57 and would be more likely to plant addit ional crops in its plots to provide for its consumption needs. Results of the Estimated Equations The estimated coefficients for the variables determining production of cacao and related biodiv ersity are presented in T ables 4 8 and 4 9 The biodiversity equation predicts the number of species of crops or trees in a parcel while the production equation predicts the amount in quintal of cacao produced. The most important input used for production of cacao o n these farms appears to be family labor, whose lo g has a coefficient of 0.35 in the cacao production equation. Family labor, though, does not have a differential im pact by type of cacao, whether nacional or CCN 51 on the cacao production equation. However, it does have a differential impac t on the biodi versity equation. The interaction coefficient between fami ly labor and a dummy for cacao n acional is not significant in the production equation while it is in the biodiversity equation T he effect of labor on biodiversity produ ction is significant for ca cao n acional, but not for cacao CCN 51. This conclusion is inferred from the positive significance of the interaction term on biodiversity and the lack of significance of the family labor term that is not interacted with cacao nacional on biodiversity H ired labor and other inputs have smaller coefficients, which are not significant for either type of production. Land and several aspects related to land ownership are also important determinants of both biodiversity and cacao production. For example, the surface of the land, whether flat, hilly or steep, has a significant impact on both equations, although opposite signs. Flatter land influences productivity in a positive manner, relative to steep or hilly areas of land. The opposite is true for biodiversi ty production. Individuals confined to produce on steeper plots have a higher number of varieties in their farms.

PAGE 58

58 This result confirms the observations that farmers leave the steeper areas forested to prevent erosion. A similar effect is observed through t he initial amount of land that a farmer started cultivating. This coefficient suggests that less wealthy farmers have more biodiversity. Less wealthy farmers may be more dependent on their farms for subsistence and need to plant additional varieties in th e plots to meet these nee ds. T he number of years that a household has owned their land is positively correlated with both cacao production and biodiversity Farmers who own their land longer have had more time to invest in their land in long term crops, enhancing crop diversity Our results support the conservation literature which suggests that insecure property rights have a strong negative effec t on biodiversity. However, insecure property results did not significantly affect cacao production The res ult may be due to the fact that the sample is composed mostly of lo w income farmers with small non commercial farms. T he difference in access to credit because of land insecurity (Eswaran and Kotwal 1995), which affects productivity, is minimal relative to o ther farmers with land title. Those households with land title have very little extensions of land. The lack of land for collateral would also limit these farmers access to credit. Finally, the variable indicating whether the household head belongs to the indigenous Tsa'chila group or not, indicates that households with this type of ethnic trait have more b iodiversity but less cacao production, relative to other households in the sample. asons chose to leave their land forested and also would produce less because they have less knowledge about modern agricultural practices The final variable included in the regressions is the household size. Larger households would need more resources t o

PAGE 59

59 sustain themselves. Thus, a larger household would be expected to plant more plantains and other crops needed to p rovide the household with food. Shadow Wage Results From the family labor coefficients in the previous section, the shadow wage is calculat ed for a day of family labor as described in the estimation section. Our results are presented in T able 4 10 The terms in the first column, w Q and w V indicate the two different component s of our shadow wage (E quation 3 9 ): w Q *= p ( ) and w V = T hese two terms are not added directly, since the two dependent variables have different units. The first shadow wage represents the value of an extra quintal of cacao that will be produced with an extra day of labor, which depends on which market it is being sold on. The second is the value of the amount of an extra variety that can be produced with an extra day of labor. Column two presents the marginal value of labor for all cacao and column three the marginal value that is to be added (or subtracted if the coefficient is negative) if the variety is n acional. In the table, N.S stands for values that are not significant. Pv is the hypothetical value of the respective variety of crop planted in the parcel along with cacao; and 0.0 1 is the marginal productivity of l abor when the cacao variety is nacional, the traditional variety Column two reveals that the only the shadow wage component which applies for all cacao production is higher than the mean of the wage obtained in the mark et (8.33 U S dollars). This shadow wage can be higher than the market wage when hired and family labor are not substitutes and family size limits the available family labor. However, it is still in the range of high market wages (the range of all daily w ages is between 7 U S dollars and 12 U S d ollars). A family selling their cacao in the specialty

PAGE 60

60 market will obtain about 1.5 U S dollars more per day of labor than a family who does not have access to this market (11.90 U S dollars comp ared to 10.40 U S dollars). The value of biodiversity or of adding an extra variety consists of multiple market and nonmarket components. Some of these, such as self consumption or production of substitutes for medicinal plants may be easier to estimate; but others, s uch as ornamental or ritual use, are at best difficult to estimate. In this sense, the variable Pv is The third column of Table 4 10 wage of family labor if cacao n ac ional is produced. However, a lower bound can only be created for this value based on the market price of goods for self consumption, which will apply for specific cases. For example, if cacao nacional is intercropped with cassava, producing 50 quintals p er intercropped hectare, at 7 U S dollars per quintal the marketable MPL for this variety would be .49 T he total shadow wage of int ercropped cacao would be about 3 U S dollars higher. If cacao n acional is intercropped with oranges, the shadow wage wou ld be at least 1.10 U.S. dollars hi gher (for 4000 Units/ha at 0.04 U S dollars per unit) Added to this shadow wage would be benefits like shade, soil enhancement ornamental and cultural value, medicinal and ritual uses, etc. Based on our analysis, Tab le 4 11 shows that the shadow price for a day of work in the production o f cacao CCN 51 is equ ivalent to a high market wage. Furthermore, a day of work for producing cacao nacional may be higher than a high market wage once the additional biodiversity val ue is included. The shadow wage is particularly higher than the market wage for cacao nacional producers who have access to specialty market s because of the cash premium they receive Through the analysis of the shadow wage,

PAGE 61

61 which includes both production and biodiversity values, cacao nacional provides more value to smallholder households than cacao CCN 51. Summary of the Results Our survey of cacao producing households in northwestern Ecuador produced a result that would seem to contradict conventional e conomic the ory. Households would prefer to raise cacao nacai onal instead of cacao CCN 51 even though cacao CCN 51 is more profitable. Our results also revealed that the cacao nacional production system provides some benefits that are not captured in the market such as soil enhancement and the protection of native species of plants and anim als. The shadow prices provide a manner to include these additional values that the household obtains form cacao production. The shadow wage is higher for cacao nacion al that it is for cacao CCN 51 and much higher when farmers sell in specialty markets. The shadow wage difference demonstrates that these households receive more value from cacao nacional than cacao CCN 51. The households act rationally when they chose t o plant cacao nacional instead of cacao CCN 51.

PAGE 62

62 Figure 4 1 C

PAGE 63

63 Table 4 1 Household c haracteristics Cacao CCN 51 Cacao n acional Total i ncome per month 1 Mean Median Number of o bse rvations 4.87 5 15 3.44 t 3 34 Farm i ncome per month 1 Mean Median Number of o bservations 3.81 3 16 2.85 t 2 34 Percentag e i ndigenous Number of o bservations 31.25 16 29.41 34 Family s ize Mean Median Number of o bservations 4.75 5 16 4.09 4 34 Fa rm s ize (ha) Mean Median Number of o bservations 28.44 11 16 13.08 tt 8 34 Percentage that own a car Number of o bservations 50.00 16 15.15 ttt 33 Percentage with running w ater Number of o bservations 25.00 16 6.25 tt 32 Rooms in the h ome Mean Median Nu mber of o bservations 4 4 16 3.25 tt 3 32 t S ignificant difference at the 10% level tt S ignificant difference at the 5% level tttt S ignificant difference at the 1% level

PAGE 64

64 Table 4 2. Parcel characteristics Cacao CCN 51 Cacao nacional Parcel area (ha) Mean Median Number of observations 2.57 2.00 18 4.50 tt 3.00 51 Tree density (trees/ha) Mean Median Number of observations 702.78 670 18 502.18 t 600 51 Tree age (years) Mean Median Number of observations 3.94 2.00 18 10.49 ttt 6.00 51 Slope a Mean Median Number of o bservation 1.83 2.00 18 1.82 2.00 51 Number of annual crops per p arcel Mean Median Number of o bservations 0 0 18 0 .25 t 0 51 Number of perennial crops per p arcel Mean Median Number of o bservations 1.33 1.00 18 2.65 ttt 3.00 51 Number of trees varieties per p arcel Mean Median Number of o bservations 0.61 0 18 1.08 t 1.00 51 t S ignificant difference at the 10% level tt S ignificant difference at the 5% level tttt S ignificant difference at the 1% level a 1:Plain, 2:Hilly 3: St eep

PAGE 65

65 Table 4 3 Yield, costs, revenue, and profits per h ectare Cacao CCN 51 Cacao n acional Total l abor (days/ha) Mean Median Number of o bservations 33.62 24.85 18 33.96 28.27 48 Family l abor (days/ha) Mean Median Number of o bservations 18.67 14.5 2 18 22.72 18.00 49 Hired l abor (days/ha) Mean Median Number of o bservations 14.95 11.33 8 10.87 t 8.67 49 Planting costs with family labor (USD/ha) Mean Median Number of o bservations 55.71 28.30 13 43.00 21.00 29 Planting costs without f amil y l a bor (USD/ha) Mean Median Number of o bservations 47.65 21.57 13 18.85 tt 10.12 30 Other input c osts (USD/ha) Mean Median Number of Observations 55.88 48.8 18 21.84 ttt 7 51 Total costs with family l abor (USD/ha) Mean Median Number of Observations 355 .99 295.13 18 296.91 256.11 48 Yield (quintal /ha) Mean Median Number of Observations 16.41 9.41 18 8.97 tt 7.29 49

PAGE 66

66 Table 4 3 Continued Cacao CCN 51 Cacao n acional Revenue in the s tandard market (USD/ha) Mean Median Number of o bservations 141 9.37 652.80 13 648.90 tt 572.71 31 Revenue in the specialty m arket (USD /ha) Mean Median Number of o bservations Na 824.41 600 15 Revenue in the organic m arkets (USD /ha) Mean Median Number of o bservations Na 562.60 165.00 3 Profit in the standard mark et with family l abor (USD /ha) Mean Median Number of o bservations 1066.54 263.06 13 361.31 261.74 30 Profit in the standard market without family labor (USD /ha) Mean Median Number of o bservations 1233.21 495.07 13 543.34 tt 513.00 31 Profit in the s pecialty market with family labor (USD /ha) Mean Median Number of o bservations Na 644.63 511.60 14 Profit in the specialty market without family labor (USD /ha) Mean Median Number of o bservations Na 762.20 600.55 14

PAGE 67

67 Table 4 3 Continued Cacao CCN 51 Cacao n acional Profit in the organic market with f amily labor (USD /ha) Mean Media Number of o bservations Na 745.91 501.67 4 Profit in the organic market without family labor (USD /ha) Mean Median Number of o bservations Na 870.87 614.03 3 t S igni ficant difference at the 10% level tt S ignificant difference at the 5% level tttt S ignificant difference at the 1% level

PAGE 68

68 Table 4 4 Yield, costs, revenue, and profits per hectare e quivalent Cacao CCN 51 Cacao n acional Total l abor (days/hae) a Mean Medi an Number of o bservations 28.75 25.89 18 63.77 tt 41.93 48 Family l abor (days/hae) a Mean Median Number of o bservations 16.16 13.95 18 37.72 tt 22.5 49 Hired l abor (days/hae) a Mean Median Number of o bservations 12.59 11.81 18 25.01 11.19 49 Labor fo r weeding (days/hae) a Mean Median Number of o bservations 9.24 9.17 18 29.14 tt 15 51 Other labor input c osts (USD/hae) a Mean Median Number of o bservations 51.50 36.97 18 28.78 tt 8.77 51 Total costs with family l abor (USD/hae) a Men Median Number of o bservations 304.40 289.60 18 564.61 t 327.21 49 Total cost without family l abor (USD/hae) a Mean Median Numb er of o bservations 165.14 171.70 18 258.40 121.03 49 Revenue in the standard m arket (USD/hae) a Mean Median Number of Observations 1393.97 5 69.92 13 834.88 t 656.25 31 Revenue in the specialty m arket (USD /hae) a Mean Median Number of Observations N a 1060.86 750.00 15

PAGE 69

6 9 Table 4 4 Continued Cacao CCN 51 Cacao n acional Revenue in the organic m arkets (USD /hae) a Mean Median Number of o bservat ions Na 1026.95 1246.09 4 Profit in the standard market with family labor (USD /hae) a Mean Median Number of o bservations 1081.10 382.20 13 285.15 tt 382.20 30 Profit Genera l Market without family labor (USD /hae) a Mean Median Number of o bservations 1223.84 345.45 13 608.65 t 569.92 31 Profit s peci alty market with family labor (USD /hae) a Mean Median Number of o bservations Na 677.90 472.50 15 Profit in the s peci alty market with family labor (USD /hae) a Mean Median Number of Observations Na 677. 90 472.50 15 Profit in the o rgani c market without family labor (USD /hae) a Mean Median Number of Observations Na 182.32 288.39 4 Profit in the organic m arket with out family Labor (USD /hae) a Mean Median Number of o bservations Na 233.57 496.37 4 a h ae = hectare equ ivalent. Each parcel was converted to the amount of hectares it would contain if the trees were plan ted at a density of 625 trees per hectare. t S ignificant difference at the 10% level tt S ignificant difference at the 5% level tttt S ignifi cant difference at the 1% level

PAGE 70

70 Table 4 5 P erceptions of p rofitability Variable Name CCN 51 p arcels Nacional p arcels Entire f arm Profitability 6 years a go r 3.00 3.57 2.90 tt Profitability 3 years a go r 3.25 3.44 2.76 ttt Profitability n ow r 3.28 3.09 2. 67 tt Profitability 3 5 y ears in the f uture r 3.92 4.11 3.46 ttt r Farmers scored the variables on a scale from one to five with one being very bad and five being very good. t Difference between cacao nacional and the entire farm is significant at the 10% level tt Difference between cacao nacional and the entire farm is significant at the 5% level ttt Difference between cacao nacional and the entire farm is significant at the 1% level Figure 4 2 P erceptions of p rofitability 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 6 Yrs Ago 3 Yrs Ago Now Next 3 Yrs Rating on a 1 5 Scale Total Farm Cacao Nacional Cacao CCN-51

PAGE 71

71 Figure 4 3 L ikelihood to plant c acao 44 4 4 Would You Like to Plant More Cacao? Yes No No Response

PAGE 72

72 Figure 4 4 C acao variety planting p references 65.9% 29.5% 4.5% Which Variety Would You Prefer to Plant? Nacional CCN-51 Both

PAGE 73

73 Table 4 6 E nvironmental p erceptions Variable n ame CCN 51 p arcels Nacional p arcels E ntire f arm Soil quality 6 years a go r 3.00 4.36 aaa 3.76 ttt Soil quality 3 years a go r 3.00 4.25 aaa 3 .45 ttt Soil quality n ow r 4.00 4.24 3.10 ttt Soil quality 3 5 y ears in the f uture r 4.00 4.38 a 3.07 ttt Biodiv ersity 6 years a go r 2.00 3.95 3.65 Biodiversity 3 years a go r 1.67 2.95 aa 2.64 Biodiversity n ow r 2.50 2.42 2.04 t r Farmers scored the variab les on a scale from one to five with one being very bad and five being very good. a Difference between cacao nacional and CCN 51 is significant at the 10% level aa Difference between cacao nacional and CCN 51 is significant at the 5% level aaa Difference b etween cacao nacional and CCN 51 is significant at the 1% level t Difference between cacao nacional and the entire farm is significant at the 10% level tt Difference between cacao nacional and the entire farm is significant at the 5% level ttt Difference b etween cacao nacional and the entire farm is significant at the 1% level

PAGE 74

74 Figure 4 5 P erceptions of b iodiversity 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 6 Yrs Ago 3 Yrs Ago Now Next 3 Yrs Ratings on a scale from 1 5 Total Farm Cacao Nacional Cacao CCN-51

PAGE 75

75 Table 4 7 Variables included in production and biodiversity r egressions Variable n ame Description Biodiversity Number of additional varieties of plant species in the parcel Production Additional quintals of cacao p roduced Log of family labor Log of the days of family labor for all cacao production Nacional log of family labor Interaction term of a dummy variable for cacao nacion al with the l og of the days of family labor Log of hired labor Log of the days of hired labor for all cacao production Log of other i nputs Log of cost in US dollars other inputs besides labor for all cacao production Flat p arcel Dummy variable indi cating of the parcel where cacao is planted in a parcel with a flat terrain Hilly p arcel Dummy variable indicating if the parcel where cacao is planted neither a flat nor steep terrain Insecure l and Dummy variable to represent if the househ old has sec ure property rights ( insecure property rights are those defined as communal land, squatted land, or inherited land) Initial l and household began farming Years land owned Number of years that the house hold has the parcel of land Ethnicity (head of household ) Dummy variable if the head of the household is Household s ize Number of members in the household

PAGE 76

76 Table 4 8 Coefficients of the b io diversity e quation Variable n ame Coef ficient Standard Error Log of f am ily l ab or 0.09 0.07 Na cional log of f am ily l ab or 0.15 tt 0.05 Log of hired l ab or 0.05 0.04 Log of other i nputs 0.03 0.03 Flat p arcel 0.64 tt 0.19 Hilly p arcel 0.40 tt 0.19 Insecure l and rights 0.49 tt 0.15 In itial l and size (ha) 0.00 t 0.00 Years land owned 0.01 t 0.00 Ethnicity (head of household ) 0.67 tt 0.19 Household s ize 0.05 t 0.03 Constant 0.52 0.22 F(11, 32) = 20.94 Prob ability > F = 0 R squared = 0.7518 Number of o bservations = 44 t Significanc e at the 10% level tt Significance at the 5% level

PAGE 77

77 Table 4 9 Coefficients in the production e quation Variable Name Coefficient Standard Error Log of f am ily l ab or 0.35 tt 0.16 Na cional log of f am ily l ab or 0.03 0.04 Log of hired l ab or 0.07 0.10 Log of other i nputs 0.06 0.07 Flat p arcel 0.91 tt 0.19 Hilly p arcel 0.73 tt 0.23 Insecure l and rights 0.15 0.27 Initial l and size (ha) 0.00 0.00 Years land owned 0.01 t 0.00 Ethnicity (head of the household ) 0.92 tt 0.28 Household s ize 0.03 0.07 Constant 0.48 0.54 F(11, 32) = 7.99 Prob ability > F = 0 R squared = 0.6302 Number of o bservations = 44 t Si gnificance at the 10% level tt Significance at the 5% level

PAGE 78

78 Table 4 10 Shadow wage components by m arket All Cacao Nacional wQ*(in USD per day of f amily l abor ) in Standard Market 10.35 N.S. t wQ*(in USD per day of f amily Labor ) in Specialty Market 11.93 N.S. t w V *(in varieties per day of family l abor ) N.S. Pv*0.01 a t N.S. means the value is n ot significant a Pv is the value of the additional variety in the parcel Table 4 11 Shadow wage by variety and m arket Price in US dollars per day of l abor Shadow value of l abor for cacao nacional sold in s tandar d m arkets > 10.35 Shadow value for a day of labor for cacao nacional sold in specialty m arkets > 11.93 Shadow value for a day of labor for cacao CCN 51 sold in s tandard m arkets = 10.35

PAGE 79

79 CHAPTER 5 CONCLUSION New Conceptual Framework A need exists to better conceptualize the production decisions of small landholder households. M any models of smallholder farmers have begun to include factors besides profit maximization such as subsistence needs and overall household welfare The envir onmental impact of production methods are also important factors to smallho lder household s as these variables impact the sustaina bility of their farms. In addition, these factors provide the household with aesthetic and other values. A biodiversity measurement is one way to capture this ecological value as the resea rch on agrof orestry cropping systems have shown that the se diversified cropping systems provide many ecological services important to smallholder households. T hese biodiverse systems also provide many monetary benefits such as additional incomes sources and meeting c onsumption needs Th e biodiversity factors have proven to be an important consideration in the production decisions of smallholder cacao producers in Ecuador as these farmers would prefer to raise cacao nacional which provides many ecological and other be nefits that are not obtained in cacao CCN 51 parcels. In order to analyze the importance of biodiversit y in the produc tion of cacao, we created a model that disentangle s the effects of family labor for biodiversity and for the production of cacao by an alyz ing shadow prices. Shadow wage and shadow income value biodiversity which has not been included in past models of shadow wages and labor Our model also takes into account of the effect of segmented markets Se gmented markets exist in the Ecuadorian cas e by certification requirements and limited de mand in specialty cacao nacional markets Our modeling technique is useful

PAGE 80

80 for the study of other crops that participate in segmented markets such as in t he fair trade markets, not just the cacao nacional spec ialty market The shadow wage in this model is the sum of the traditional measure of the shadow wage, MPL for cacao production, and the new variable, the MPL of biodiversity Our analysis of the shadow prices for cacao in northwestern Ecuador reveal s the households receive a higher shadow income for planting c acao nacional than for planting cacao CCN 51 O nly a lower bound for the shadow wage could be developed as the aesthetic and other nonmarket values of biodiversity were not captured in th e survey. The magnitude of the difference between the market and shadow wages depends on the crops planted in the cacao parcels. The shadow prices also show that family and market labor are not perfect substitutes. Family labor receives a very high valu e that would induce family members to work on the farm instead of participating in the labor markets. T he shadow wage for cacao production is at the upper bounds of the labor market and the shadow wage for cacao nacional sold in specialty markets is even higher. Thus, small holder households in Ecuador would be able to maximize the value of their labor by devoting to production of cacao nacional. Impact on Current Research Our analysis of shadow wage extends the theoretical work of Jacoby (1993) and Skou fi as (1994). Our model adds to this work by showing the importance of including externalities, such as the benefits from biodiversity, in computing shadow wage in segmented market s B y examining segmented markets, our analysis of shadow income enhances the work on fair trade of LeClair (2002) and Hayes (2008) by examining the impact of these markets at the household level T he determinants of market access in these specialty and fair trade markets were also examined. Our surveys revealed t hat

PAGE 81

81 many farmers were excluded fro m these markets. Our model could help create a better understanding of the failings of these methods in reach ing smallholder households Our model of shadow prices has shown to be effective in trying to capture the non market benefits t hat farmers realize in their production and take into account in making copping decisions Need for Further Research Our study leaves s everal areas to be examined to better understand smallholder production decisions. First, this research was not able to fully explore the impact of risk on these decisions. Research on agroforestry has shown that these diversified cropping systems are less risky than monoculture production methods as a household has additional income sources to meet its needs ( Reitsma 2001). The household welfare impacts of growing diversified, organic production systems compared to monoculture commercial production systems such as between cacao nacional and cacao CCN 51 need to be examined. The former production system is arg ued to be more ecologically sustainable but the human welfare impact also needs to be examined to truly understand the impact to total welfare of each product ion system. Finally, methods need to b e utilized to give a value to the non market benefits for b iodiver sity Only by placing a value on these factors can shadow wage be crated that includes all the externalities rather than creating a lower bound for t his wage as was included in our study. These additional nonmarket values are especially important i n developing countries where many households are dependent on their farms not only for cash income but also for meeting subsistence needs The Ecuadorian cacao case provides a good example of the need for the inclusion of additional values in the househol d

PAGE 82

82 decision model When comparing the shadow wages, t he traditional production methods for cacao nacional proved to best production decision when the value for ision. T he sha dow wage and income that includes biodiversity as a value in the price provides an additional method for determining the true value of a production method to smallho lder households.

PAGE 83

83 LIST OF REFERENCES ve Valuation of Subsistence Crops: The Case of Traditional Maize in Mexico. American Journal of Agricultural Economics 91:956 972. Asociacion Nacional de Exportadores de Cacao 2007. Manuel del Cultivo de Cacao para Productores. Guayaquil, Ecuador. A uld, G., L. H. Gulbrandsen, and C.L. McDermott. 2008. Certification Schemes and the Impacts on Forests and Forestry. Annual Review of Environmental Resources 33: 187 211. Bardhan, P. A. 1999. Development Microeconomics New York: Oxford University Press Barnum, H. N., and L. Squire 1979. An Econometric Application of the Theory of the Farm Household Journal of Development Economics 6: 79 102. Becchetti, L., and M. Costantino. 2008. The Effects of Trade on Affiliated Producers: An Impact Analysis of Kenyan Farmers. World Development 36: 823 842. Beer, J., R. Muschler, D. Kass, and E. Somarriba. 1998. Shade Management in Coffee and Cacao Plantations. Agroforestry Systems 38: 139 164. Bentley, J., E. Boa, and J. Stonehouse. 2004. Neighbor Trees: Shade, Intercropping, and Cacao in Ecuador. Human Ecology 32: 241 258. Brush, S. B., J. E. Taylor, and M. R. Bellon. 1992. Technology Adoption and BiologicalDiversity in Andean Potato. Agriculture. Journal of Development Economics 39: 365 387. Butler, R A., and W.F. Laurence. 2008. New Strategies for Conserving Tropical Forests. Trends in Ecology and Evolution 2 3: 469 472. Carter, M. R., and Y. Yao. 2002. Local Versus Global Separability in Agricultural Households Models: The Factor Price Equaliz ation Effect of Land Transfer Rights. American Journal of Agricultural Economics 84: 702 715 Central Intelligence Agency. 2009. Factbook. Website https://www.cia.g ov/library/publications/the world factbook/geos/ec.html Accessed 30 November 2009 Conservation International. 2007. Biodiversity Hotspots. Website. http:/ /www.biodiversityhotspots.org/xp/hotspots/tumbes_choco/Pages/default.asp a Accessed 1 October 2009

PAGE 84

84 Coporacin de Promocin de Exportaciones e Inversiones. 2009. Cacao. Ecuador Calidad de Origen Quito, Ecuador. Dahlquist, R. M., M. P. Whelan, L. Winow iecki, B. Polidoro, S. Candela, C. A. Harvey, J. D. Wulfhorst, P. A. McDaniel, and N. A. Bosque Prez. 2007.. Incorporating Livelihoods in Biodiversity Conservation: a Case Study of Cacao Agroforestry Systems in Talamanca, Costa Rica. Biodiversity and Conservation 16: 2311 2333. de Janvry, A., M. Fafchamps, and E. Sadoulet. 1991. Peasant Household Behavior with Missing Markets: Some Paradoxes Explained. The Economic Journal 101: 1400 1417. Duguma, B. J. 2001. Smallholder Cacao (Theobroma cacao Li nn.) Cultivation in Agroforestry Systems of West and Central Africa: Challenges and Opportunities. Agroforestry Systems 51: 177 188. El Universo (Guayaquil, Ecuador) pp. 23. Eswaran M., and A. Kotw al. 1995. Why Poverty Persists in India New York: Oxford University Press Feather, P., and W. D. Shaw. 1999. Estimating the Cost of Leisure Time for Recreation Demand Models. Journal of Environmental Economics and Managemen 38: 49 65. Food and Agr iculture Organization of the United Nations. Statistical Database. Website http://faostat.fao.org/default.aspx Accessed 20 March 2009 Gamma Rodrigues, E.F. 2009. Carbon Storage in Soil Size Fraction s under Cacao School of .Forest Resources and Conservation. Gobierno de la Provincia de Pichincha. 2003. Plan de Desarrollo Participativo 2002 2012 Quito: Gobierno de la Prov incia Pichincha. Gomez Tovar, L., L. Martin, M.G. Cruz, and T. Mutersbaugh. 2005. Certified Agriculture in Mexico: Market Connections and Certification Practices in Large and Small Producers. Journal of Rural Studies 21: 461 474. Hilderbrand, P. E. 2 002. Global Research Challenges: Including Small Holders in Rural Development First Henry A. Wallace Inter American Scientific Conference on Globalization of Agricultural Research. Turrialba, Costa Rica. e Trade Organizations in the Era of A Comment. World Development 36: 2953 2961.

PAGE 85

85 Jacoby, H. G. 1993 Shadow Wage and Peasant Family Labour Supply: An Econometric Application to the Peruvian Sierra, Review of Economic Studies 60: 903 921. Functions. American Journal of Agricultural Economics 91: 685 696. LeClair, M. S. 2002. Fighting the Tide: Alternative Trade Organizations in the Era of Global Fr ee Trade. World Development 30: 949 958. Melo, C. J. 2009. Agroecological Assessment of Ecuadorian Cacao Production Systems. Working Paper, Florida International University, Department of International Relations. Municipalidad de Puerto Quito. 2009. Website http://www.puertoquito.gov.ec/index.php?option=com_content&view=article&id=5 5&Itemid=58 11 March 2010. Municipalidad de Quinind. 2009. W ebsite. http://www.quininde.gov.ec/ 11 March 2010 Neilson, J. 2008. Global Private Regulation and Value Chain Restructuring in Indonesian Small Holder Coffee Systems. World Development 36: 1607 1622. Nelson, V., and Galvez, M. 2000. Social Impact of Ethical and Conventional Cocoa Trading on Forest Dependent People in Ecuador Working Paper. University of Greenwich, Natural Resources and Ethical Trade Programme. Perfecto, I., J. Vandermeer, A. Mas, and L.S Pinto. 2005. Biodiversity, Yield, and Shade Coffee Certification. Ecological Economics ,54: 435 446. Provincia de Santo Domingo de los Tsachilas. 2010. Informacon General. Website http://www.gptsachila.gov.ec/index.php?option=com_content&task=view&id=16& Itemid=33 Accessed on 11 March 2010. Reitsma, R., J. D. Parrish, and W. McLarney. 2001. The Role of Cacao Plantations in Maintaining Forest Avian Diversity in Southeastern Costa Rica. Agroforestry Systems 53: 184 193. Rice, R. A., and R. Greenberg. 2000. Cacao Cultivation and the Conservation of Biological Diversity. Ambio 29: 167 173. Ros Tonen, M. A., T. V. Andel, C. Morsello, K. Otsuki, S. Rosendo, and I. Scholz, 2008. Forest related Partnerships in Brazilian Amazonia: Three is More to Sustainable Forest Management than Reduced Impact Logging. Forest Ecology and Management 1256: 482 1497.

PAGE 86

86 Skoufias, E. 1994. Using Shadow Wages to Estima te to Labor Supply of Agricultural Households. American Journal of Agricultural Economics 76: 215 227. Steffan Dewentera, I., M. Kesslerc, J. Barkmannc, M. M. Bos a, D. Buchorig, Stefan Erasmih, H. Fausth, G. Geroldh, K. Glenke, S. R. Gradsteind, E. Gu hardjai, M. Harteveldd, D. Herteld, P. Hhna, M. Kappash, S. Khlerh, C. Leuschnerd, M. Maertensj, R. Marggrafe, S. Migge Kleiank, J. Mogeai, R. Pitopangl, M. Schaeferk, S. Schwarzem, S. G. Spornd, A. Steingrebek, S. S. Tjitrosoedirdjoi, S. Tjitrosoemitoi, A. Tweleh, R. Weberh, L. Woltmannk, M. Zellerm, and T. Tscharntkea. 2007. Tradeoffs between Income, Biodiversity, and Ecosystem Functioning during Tropical Rainforest Conversion and Agroforestry Intesification Proceedings of the National Academy of Sc iences Vol 104. Available at www.pnas.org/cgi/doi/10.1073/pnas.0608409104 Vasco, J. S. Breve Historia del Cultivo de Cacao en Ecuador. Ministerio de Agricultura, Ganadera, Acuacultura y Pes ca en Ecuador : Website http://www.sica.gov.ec/cadenas/cacao/docs/historia_cacao.htm Accessed 11 March 2010

PAGE 87

87 BIOGRAPHICAL SKETCH Trent Blare was raised on a ranch in western Sou th Dakota where he first gained a passion for agriculture and the rural lifestyle. While in high school Trent was very active in 4 H and FFA. The activities in these organizations enhanced his interest in agribusiness. So, he decided to pursue a degree in Agricultural Economics at the University of Nebraska Lincoln where he was a n Scholar He had the opportunity to participate in two undergraduate research opportunities while at the University of Nebraska. He first wo rked with political science professor Dr. David Rapkin to examine position in trade negotiations in the World Trade Organizations. He continued this work with his adviser Dr. E. Wesley Peterson to build a mode l that explai ned the impact on American small grain producers of a new trade deal in the World Trade Organization. three years. He worked with the community on an eco tourism project as well as helped improve their production and marketing of cacao. This experience spurred Trent to continue his studies in Food and Resource Economics at the University of Florida with a focus on agricultural development. While at the University of Florida, Trent has been active in the Tropical Conservation and Development Program, which provided him with funding to pursue his research on Ecuadorian cacao production in the summer of 20 09. Trent plans to continue his work in agricultural development and especially his research on cacao in Ecuador as he begins a PhD in Food and Resource Economics at the University of Florida in August of 2010.