Determining Potential Demanders of Urea Briquettes in the Cantons of Daule and Santa Lucia in the Ecuadorian Coast

MISSING IMAGE

Material Information

Title:
Determining Potential Demanders of Urea Briquettes in the Cantons of Daule and Santa Lucia in the Ecuadorian Coast Ex-Ante Technology Adoption Analysis
Physical Description:
1 online resource (185 p.)
Language:
english
Creator:
Avila S., Jorge J.
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Food and Resource Economics
Committee Chair:
Useche, Maria Del Pilar
Committee Co-Chair:
Sterns, James A

Subjects

Subjects / Keywords:
adoption -- analysis -- deep -- ex-ante -- innovation -- intensity -- model -- placement -- rice -- technology -- tobit -- urea
Food and Resource Economics -- Dissertations, Academic -- UF
Genre:
Food and Resource Economics thesis, M.S.
Electronic Thesis or Dissertation
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )

Notes

Abstract:
In Ecuador, the main fertilizer for rice cultivation, Urea,may be lost as much as 60% when it is applied with Broadcast technique. UreaDeep Placement (UDP), originally utilized in Asia, has been shown to enablerice farmers to reduce such a loss.  Thisthesis is an ex-ante analysis of the potential for UDP adoption in two major riceproducing cantons in the Ecuadorian Coast, Daule and Santa Lucia. A survey wasimplemented to collect information of rice farmers across 35 villages. A descriptiveanalysis explored variables that may affect adoption decision. For instance, somefarmers obtained negative net incomes, implying the need of more efficient innovationslike UDP. In the Double-bounded exploratory analysis was detected that 93.25% ofthe sample farmers were willing to pay extra for Urea briquette sacks with the introductioneconomic benefits/costs and with the inclusionof economic and environmental impacts. In analyzing the Intensity ofAdoption (IA), potential adopters would dedicate a 49.7% of their total land forUDP production, on average. Finally, the two-limit Tobit model of the technologyadoption decision (in terms of the IA) suggests that the smaller a farmer, thehigher his probability of UDP adoption. However, the subsidy of Urea fertilizermay be an obstacle for UDP acceptance. Social network was also significant inthe model; potential adopters’ behaviors may influence a farmer’s adoptiondecision. Other significant variables were small kids in a household, marketaccess, rented land, credit solicitation, agricultural insurance, riskaversion, UDP knowledge and on-farm hours.
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.
Thesis:
Thesis (M.S.)--University of Florida, 2012.
Local:
Adviser: Useche, Maria Del Pilar.
Local:
Co-adviser: Sterns, James A.
Statement of Responsibility:
by Jorge J. Avila S.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044787:00001


This item has the following downloads:


Full Text

PAGE 1

1 DETERMINING POTENTIAL DEMANDERS OF UREA BRIQUETTE S IN THE CANTO N S OF DAULE AND SANTA LUCIA IN THE ECUADORIAN COAST: EX ANTE TECHNOLOGY ADOPTION ANALYSIS By JORGE J. AVILA S. 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 2012

PAGE 2

2 2012 Jorge J. Avila S.

PAGE 3

3 To Malena S antamaria Alfredo A. Xavier A. and Belen M. son mi vida

PAGE 4

4 ACKNOW LEDGMENTS Firstly, I thank God for giving me everything I will always thank my parents, my family and Belen for their constant love. I am sincerely grateful to Dr. Bowen, Dr. Espinel, Dr. Herrera, Dr. Sterns and Dr. Useche for the trust, support and contr ibution that allowed me to develop this academic resea rch. I would like to thank my great team of enumerators Angie A., Karen R., Robinson M. Victor B., villagers of Daule and Santa Lucia cantons and my friend Samuel whose invaluable help let me obtain th e primary data I have to acknowledge Alfredo A. Belen M and Guido G. for having spent their time on the tabulation I would like to thank Centro de Investigaciones Rurales of Escuela Superior Politecnica del Litoral, Food and Resourc e Economics Department of University of Florida, the PL 480 of USDA and the Secretaria Nacional de Educacion Su perior, Cienc i as, Tecnologias e Innovacion of Ecuadorian g overnment for having funded I also thank Fatima at ESPOL a nd Jessica at UF for having facilitated my life with all the paperwork Last but not least, I also thank all my friends, especially Amanda, Belinda, Clay, Imelda, Isnel, Lara, Lee, Natasha and Olga who help ed me in different ways during this time of my lif e.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 The Problem ................................ ................................ ................................ ........... 14 Research Question ................................ ................................ ................................ 16 Objective ................................ ................................ ................................ ................. 17 Thesis Structure ................................ ................................ ................................ ...... 17 2 LITERATURE REVIEW ................................ ................................ .......................... 20 Technology Adoption Overview ................................ ................................ .............. 20 Determinants of Adoption Decision ................................ ................................ ......... 25 Adoption Phases ................................ ................................ .............................. 25 Credit Market, Risk Av ersion and Insurance ................................ .................... 26 ................................ ................................ ............. 28 ................................ ................................ ................... 30 On farm and Off farm Activities and Non work Income ................................ .... 32 Rice Production Factors ................................ ................................ ................... 33 Social Network and Informat ion Sharing ................................ .......................... 35 3 UREA DEEP PLACEMENT TECHNOLOGY ................................ .......................... 37 Urea Deep Placement Functionality ................................ ................................ ........ 37 Bangladeshi Experiences ................................ ................................ ....................... 39 Ecuadorian Experiences ................................ ................................ ......................... 42 4 ECUADORIAN RICE MARKET AT A GLANCE ................................ ...................... 49 5 METHODOLOGY ................................ ................................ ................................ .... 62 Sampling ................................ ................................ ................................ ................. 62 Unit of Analysis and Target Population ................................ ............................. 62 Sampling Design ................................ ................................ .............................. 63

PAGE 6

6 Questionnaire and Primary Data ................................ ................................ ............. 66 Theoretical and Empirical Model ................................ ................................ ............. 71 Theoretical Model ................................ ................................ ............................. 71 Empirical Model ................................ ................................ ................................ 74 6 EM PIRICAL RESULTS ................................ ................................ ........................... 80 ................................ ................................ ......................... 80 ................................ ................................ ................... 82 Urea Deep Placement Diffusion ................................ ................................ .............. 86 Social Network Analysis ................................ ................................ .......................... 88 Technology Adoption Analysis ................................ ................................ ................ 91 Rice Production System Analysis ................................ ................................ ........... 96 Credit and Insurance Market Analysis ................................ ................................ .. 102 Time Avail ability and Non Work Income Analysis ................................ ................. 104 Econometric Results ................................ ................................ ............................. 105 Descriptive Summary of the Variables ................................ ............................ 106 Tobit Estimation ................................ ................................ .............................. 109 Post Estimation Analysis ................................ ................................ ................ 113 7 CONCLUSION REMARKS AND POLICY IMPL ICATIONS ................................ .. 152 APPENDIX: QUESTIONNAIRE (SPANISH VERSION) ................................ ......... 159 LIST OF REFERENCES ................................ ................................ ............................. 178 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 185

PAGE 7

7 LIST OF TABLES Table page 4 1 Rice production costs in 2011 (US$/ha) ................................ ............................. 57 4 2 World ranking of rice yield, 2000 10 (MT/ha) ................................ ...................... 59 4 3 Average international participation of the three types of rice, 2000 09 ............... 60 5 1 Target zones ................................ ................................ ................................ ....... 78 6 1 Factors matrix (Factor Analysis) ................................ ................................ ....... 120 6 2 Coefficient mat rix (Factor Analysis) ................................ ................................ .. 121 6 3 Statistics of those who knew and did not know about UDP .............................. 124 6 4 Time availability (hrs/day) ................................ ................................ ................. 14 6 6 5 Descriptive summary of dependent and independent variables ....................... 148 6 6 Tobit model estimation of Intensity of Adoption ................................ ................ 149 6 7 Collinearity evaluation of the independent variables ................................ ......... 150

PAGE 8

8 LIST OF FIGURES Figure page 1 1 Fertilized rice land and total rice land in E cuador ................................ .............. 19 3 1 Briquetting machine and Urea briquettes ................................ .......................... 46 3 2 U rea brique ttes placement ................................ ................................ ................. 46 3 3 Replication of briquetting machine and imported briquetting machine ............... 47 4 1 Rice production units and hectares by land size groups ................................ ..... 57 4 2 Rice yield (MT/ha), rice harvested land (ha) and rice production (MT), 2000 10 ................................ ................................ ................................ ....................... 58 4 3 Rice exports, rice imports and rice balance of trade NX, fob (thousands, US$) ................................ ................................ ................................ ................... 60 4 4 Rice credit access (US$) ................................ ................................ .................... 61 5 1 Types of rice sowing ................................ ................................ ........................... 78 5 2 Distributions of applied surveys by sample ................................ ......................... 79 6 1 ................................ ................................ ............................... 116 6 2 ................................ ................................ ................................ .... 116 6 3 Education ................................ ................................ ................................ .......... 117 6 4 A gricultural education ................................ ................................ ....................... 117 6 5 Types of agricultural ................................ ................................ ......................... 118 6 6 Agricultural education providers ................................ ................................ ....... 118 6 7 Land size groups ................................ ................................ .............................. 119 6 8 Rented land ................................ ................................ ................................ ...... 119 6 9 Tot al expenses by land size groups ................................ ................................ 120 6 10 Wealth index ................................ ................................ ................................ ..... 121 6 11 Wealth level ................................ ................................ ................................ ...... 122 6 12 Drought and flood affectation ................................ ................................ ............ 122

PAGE 9

9 6 13 Time to get the main town by sp ent time groups (Market Access) ................... 123 6 14 UDP knowledge ................................ ................................ ................................ 123 6 15 UDP knowle dge sources ................................ ................................ .................. 124 6 16 Observable UDP results ................................ ................................ ................... 125 6 17 UDP knowledge level ................................ ................................ ....................... 125 6 18 Agricultural group affiliation ................................ ................................ .............. 126 6 19 ................................ ................................ ............... 126 6 20 Meeting frequency ................................ ................................ ............................ 127 6 21 Voluntary attendance ................................ ................................ ........................ 127 6 22 ................................ ................................ ......... 128 6 23 Influential groups ................................ ................................ .............................. 128 6 24 Comm unication level ................................ ................................ ........................ 129 6 25 Past technol ogy adoption ................................ ................................ ................. 129 6 26 Adopted innovations ................................ ................................ ......................... 130 6 27 WTP: first question with initial bid by land size groups ................................ ..... 130 6 28 WTP: second question with high er bid ................................ ............................. 131 6 29 WTP: second question with lower bid ................................ ............................... 131 6 30 WTP (US$) by land size groups ................................ ................................ ....... 132 6 31 EWTP: first question with initial bid by land size groups ................................ ... 132 6 32 EWT: second quest ion with higher bid ................................ ............................. 133 6 33 EWTP: second question with lower bid ................................ ............................ 133 6 34 EWTP (US$) by lan d size groups ................................ ................................ ..... 134 6 35 UDP potential area (ha) ................................ ................................ .................... 134 6 36 Intensity of Adoption by land size g roups (%) ................................ ................... 135 6 37 Rice field ................................ ................................ ................................ ........... 135

PAGE 10

10 6 38 Description of rice varieties (in Spanish) ................................ .......................... 136 6 39 Rice v ariety ................................ ................................ ................................ ....... 136 6 40 Soil preparation cos t (US$/ha) by tillage tractor ................................ ............... 137 6 41 Total seed costs (US$/ ha) plotted with land size (ha) ................................ ....... 137 6 42 Urea (50 k g sacks/ha) by land size groups ................................ ....................... 138 6 43 Urea prices (US$/sack) of sub sidized, real and black market s ......................... 138 6 44 Total cost of other fertilize rs (US$/ha) by land size groups .............................. 139 6 45 Total cost of herbicides/pesticid es (US$/ha ) by land size groups ..................... 139 6 46 Total cost of hired labor ................................ ................................ .................... 140 6 47 Total irrigation cost (US$/ha) by land size grou ps ................................ ............ 140 6 48 Total harvest cost (US$) by land size groups ................................ ................... 141 6 49 Rice yield (kg/ha) by land size groups ................................ .............................. 141 6 50 Rice sa ck sold (%) by land size groups ................................ ............................ 142 6 51 Total income (US$/ha) by land size groups ................................ ...................... 142 6 52 Credit Solicitation ................................ ................................ .............................. 143 6 53 Credit providers ................................ ................................ ................................ 143 6 54 C redit (US$) by land size groups ................................ ................................ ...... 144 6 55 Uses of the credit ................................ ................................ .............................. 144 6 56 Rice insurance ................................ ................................ ................................ .. 145 6 57 Main occupat ion ................................ ................................ ............................... 145 6 58 Non work income ................................ ................................ .............................. 146 6 59 Human Development Bonus (US$) ................................ ................................ .. 147 6 60 Intensity of Adopti on R esiduals ................................ ................................ ........ 151 6 61 Jarque B era Normality test of R esiduals ................................ .......................... 151

PAGE 11

11 LIST OF ABBREVIATION S BM Briquetting Machine H A Hec tare HYVs High Yielding Varieties H R Hour IA Intensity of Adoption Kg Kilogram MT Metric Ton N Nitrogen PU S Production Units USDA United States Department of Agriculture UB S Urea Briquettes UDP Urea Deep Placement WTP Willingness to Pay with Econo mic Benefits/Cost EWTP Willingness to Pay with Economic Benefits/ Costs and Environmental Impacts

PAGE 12

12 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 DETERMINING POTENTIAL DEMANDERS OF UREA BRIQUETTES IN THE CANTO N S OF DAULE AND SANTA LUCIA IN THE ECUADORIAN COAST: EX ANTE TECHNOLOGY ADOPTION ANALYSIS By Jorge J. Avila S. December 2012 Chair: Pilar Useche Cochair: James Ster ns Major: Food and Resource Economics In Ecuador, the main fertilizer for rice cultivation, Urea, may be lost as much as 60% when it is applied with Broadcast technique Urea Deep Placement (UDP), originally utilized in Asia, has been shown to enable ric e farmers to reduce such a loss This thesis is an ex ante analysis of the potential for UDP adopti on in two major rice producing cantons in the Ecuadorian C oas t, Daule and Santa Lucia A survey was implemented to collect informa tion of rice farmers acros s 35 villages A descriptive analysis explore d variables that ma y affect adoption decision For instance some farmers obtained negative net incomes implying the need of more efficient innovations like UDP In the D ouble bounded exploratory analysis was d etected that 93.25% of the sample farmers were willing to pay extra for Urea briquette sacks with the introduction economic benefits/costs and with the inclusion of economic and environmental impacts In analyzing the Intensity of Adoption (IA), potential adopters would dedicate a 49.7% of their total land for UDP production on average Finally the two limit Tobit model of the technology adoption decision ( in terms of the IA ) su ggest s that the small er a farmer, the higher his probability of UDP adoption. However, the subsidy of U rea fertilizer may be

PAGE 13

13 an obstacle for UDP acceptance. Social network was also significant in the model; Other significant variables were small kids in a hous ehold, market access, rented land, credit solicitation, agricultural insurance, risk aversion, UDP knowledge and on farm hours.

PAGE 14

14 CHAPTER 1 INTRODUCTION According to the Food and Agriculture Or ganization ( 2011 ) rice consumption has increased rapidly in developing countries where rice intake was expected to raise by 461 million tons (3% more than 2010) ; per capita world rice consumption was 57 kg/year, on average, in 2011 (half a k ilo g ram highe r than in 2010 ). Moreover, around half of the world population has this cereal as a staple food ( International Rice Research Institute 2012 ). In Ecuador, rice is important in term of diet, who se per capita consumption is 112 kg/year. But also, it is the main occupation for 75, 813 production units Additionally, small farme rs are ensuring the access to this grain as 80% of r ice production units are small, less than 20 ha ( Instituto Nacional de Estadstica y Censos M inisterio de Agricultura, Ganaderia, Pesca y Acuacultura and Sistema de Informacion Agraria 2012 ) However, small farmers are struggling against poverty; official st atistics shows that 50 .9% of the rural population was poor in 2011( Instituto Nacional de Estadstica y Censos 2012 ) Also, the most important rice producing zones, Guayas and Los Rios, ar e part of the provinces with the greatest number of poor people in rural zones, around 350000 and 250000 habitants respectively (see Manuel Chiriboga y Brian Wallis 2010). At national level, rice is also relevant for Ecuador. In 2009 rice production cont ributed to the Agricultural G rowth N ational P roduct and the G rowth N ational P roduct in 11.49% and 0.69% respectively ( Instituto Nacional de Estadstica y Censos 2011 ) The Problem The most i mportant fertilizer for rice crops is Urea. According to the Instituto Nacional de Preinversion ( 2011 ) the total Urea demand is determined in about 500,000

PAGE 15

15 MT in Ecuador; mean Urea importation w as 270,000 MT, during 2009 11 Particularly, the 2011 Urea importation was 291,114 MT (in monetary terms, US$146,645,000) ; comparing the first quarter of 2011 and 2012, urea importation decreased in 14. 9% ( Sistema de Informacin Nacional de Agricultura, Ganadera, Acuacultura y Pesca 2012 ) In 2011, farmers could access Urea sacks (50 kg) at a subsidized price of US $10 ; having a market price of around US $25 Thus, the government paid the 60% of the price (Banco Central del Ecuador 2011). This is t he pressure that Ecuador must face as a non producer of Urea given the absence of infrastructure to produce this fertilizer 1 Knowing the actual situation of E cuador with respect to Urea fertilizer, it is time to define the proble m inside the rice sector. As is well known, Urea is the main fertilizer applied on rice crops. Ecuadorian rice farmers principally apply Urea by throwing it on their crops; such techniq ue is called broadcast fertilization (Alvoleo, in Ecuador ). N. K. Savant and P. J. Stangel ( 1990 ) determined that this fertilization technique entails to a loss of Urea up to 60 % which is provoked by NH 3 volatili zation, denitrification, leaching, and/or runoff. To observe the magnitude of this problem, fertilized and total rice land over time in Ecuador ( see F igure 1 1 ) One can observe that the majority of rice land has been fertili zed over time; 97%, on average. The greatest amount of fertilized land was 416 416 ha in 2004, followed by 413 266 ha in 2009 However, between 2005 and 2006, fertilized land dedicated to rice production d ropped 42%; maybe this reduction was caused by the adverse climatic situation, low prices etc. Since 2008, both types of lands were recovered. 1 Instituto Nacional de Preinversion 2011 pointed out that Ecuador has been developing a plan to start the production of urea.

PAGE 16

16 Considering the 2010 fertilized land and assuming that everyone used 20 0 kg/ha of Urea needed to fertilize the rice crop, according to Achim Dobermann ( 2012 ) In total, 80 751.6 MT of Urea could have been us ed for 403 758 ha. As mentioned above given the Urea loss of 60%, 29,070.58 MT could have lost because of the inefficient Broadcast application This value could have represented on average, a 10.78% o f the imported Urea over time (last two years) A potential solution to this problem is the adoption of Urea Deep Placement, which promises more benefits than costs at different levels. For instance, the farm level bene fits and costs are: Urea saving up to 40%; yield increase up to 25% (per se income increase); reduction of weed control cost; l abor increase (hired and/or family) and; harvest cost augmentation. At national level, the benefits from adopting this technology are: employment creation (production of Urea briquettes, briquetting machines and day laborer requirement by farmers); gender income equity; Urea import decline; reduction of Urea subsidy pressure and; reduction of N in the atmosphere and water resources ( International Fertilizer Development Center 2008 ). These impacts motivate the dev elopment of UDP adoption analysis in this thesis. Research Question In Ecuador UDP was also introdu ced to reduce Urea loss. So me rice farmers and students from Escuela Superi or Politecnica del Litoral several UDP trials during 2009 10 There were reduction on Urea fertilizer cost of up to 44. 75%; corroborating the efficiency of this method Additionall y, rice yield ranged was improved, obtaining increases in a range of 8.92% to 56.22% (Escuela Superior Politecnica del Litoral, University of Florida and PL 480 of USDA 2008) These results were diffused through several communication channels: radio, press television, extension work and

PAGE 17

17 agricultural fairs As a result, there is an interest of responding the following research question : What would be the incentives and constraints that may make a n Ecuadorian rice famer be willing to adopt the Urea Deep Plac ement Technology? Objective UDP was brought to Ecuad or by Escuela Superior Politecnica del Litoral, University of Florida and PL 480 of USDA (2008) As UDP is a new technology going into the rice farmers there is still a need of research about this innov ation Moreover, a big restraint of this innovation is the production of the Briquetting Machine whose cost is about US$7500 in Ecuador ( Orlando D. Contreras and Marcelo Espinosa L. 2010 ) ; while, the importa tion cost was estimated in US$2000. Thus, the objective of this thesis is to: Present valuable information about the potential demanders of UDP technology to possible briquette investors in order to create a Briquette Market in Ecuadorian Coast; and finall y, provide general information to policymakers for implementation of agricultural policies. In determining such demand, I could encourage investors to participate in the production of this machine. In order to answer the research question and reach the o bjective a questionnaire was designed to estimate a Tobit model through which I establish what factors may affect UDP adoption and perhaps, agricultural innovation in general. Thesis Structure Without including the Introduction this thesis is finally o rganized as follows: Chapter 2 reviews the agricultural adoption literature in which the main factors affecting adoption decision are described Chapter 3 presents the results of UDP adoption in

PAGE 18

18 n Ecuador. Chapte r 4 describes important variables of the Ecuadorian rice m arket. Chapter 5 introduces the sample design, the survey instrument and, the theoretical and empirical model used in this study. Chapter 6 reports a descriptive analysis and the Tobit estimation ou tcomes with the collected data in Daule and Santa Lucia cantons Finally, Chapter 7 presents the conclusions and policy implications of this thesis.

PAGE 19

19 Figure 1 1 F ertilized rice la nd and total rice land in Ecuador (Source: Ecuador en c ifras 2012 http://www.ecuadorencifras.com/cifras inec/main.html (accessed June 10, 2012). )

PAGE 20

20 CHAPTER 2 LITERATURE REVIEW In countries like Ecuador, small farmers are ensuring food access to an entire nation. Contradictory, those farmers are the poorest and commonly suffer famine, with children and young people being the most stricken ( Organizacin de las Naciones Unidas para la Agricultura y la Alimentacin, Fondo Internacional de Desarrollo Agrcola and Programa Mundial de Alimentos 2002 ). Such a problem can be overcome with the productivity improvement s in farming, ensuring income and food access. One respon se to accomplish this objective is the introduction of improved technologies which would e ; thus, farmers keep producing staple food, such a s rice ( Timothy Besley and Anne Case 1993 ; Cheryl R. Doss 2006 ). As a result, UDP adoption could represent an important imp rovement for Ecuadorian agriculture. Indeed, UDP would accomplish those required objectives to ensure food security: foo d must come from environmental and efficient innovations that take into consideration the biodiversity (see Magdalena Kropiwnicka 2005 ) Given the aforementioned reasons, the potential determinants and conceptual foundations of the technology a doption are explored in this section. Technology Adoption Overview In defining an agricultural inno vation or technology, diverse conceptions are found Mahajan and Perteson (1985) define d any new idea, practice or object imp lemented in the agriculture as a technology (see Lawrence Loh and N. Venkatraman 1992 ) Gershon Feder and Dina L. Umali ( 1993 ) provided a more complete definition: a technolo gy is referred to a factor which modifies the way how a farmer produces, where uncertainties can be undermined with experiences over time.

PAGE 21

21 All these three definition s are consistent with what UDP represent s : a new method/idea/factor going to change rice f a discussion is presented about technologies that were introduced in agriculture in two important periods of times: Green R evolution and Biotechnology Era. These two periods were taken into account because they are breaki ng points for agriculture development T echnological changes or innovations have been proven to have a great influen ce in the economic progress in T he Green Revolution. The introduction of such technologies improved the capability of a farmer to pr oduce a certain crop. High Yielding Varieties ( HYVs ) inorganic fertilizers, other chemical inputs and water control were the most accepted innovations in developed and developing countries, during the 1940s and 1970s. For instance, the widely recognized and name ather of the Green Revolution Norman Borlaug, introduced the high yield wheat to combat starvation worsened by the rapid population growth, first in Mexico and later in Asia. According to the International Food Policy Research Institute ( 2002 ) the problem of starva tion in industrial countries could be avoided by the adoption of improved plant breeding, fertilizers and pesticides during the half of the 20th century. R. E. Evenson and D. Gollin ( 2003 ) carried out a study of the adoption of HYVs over developing countries. They first defined t w o period s of time, Early and Late Green Revolution, being the difference that in the late one the ad option of improved varieties had a greater po sitive impact of yield growth across all developing regions (Asia, Latin American, Middle East North Africa and Sun Saharan Africa). In the early era, Asia and Latin American nations were the regions that most b enefited from HYVs While in the late era, they found that the rest of the regions reached an improvement of yield when suitable HYVs were introduced

PAGE 22

22 Hait ao Wu et al. ( 2010 ) demonstrated that the adoption of HYVs and chemical fertilizers being positively (i.e. increasing income and reducing poverty measurement); they also recognized that newest technologies should be adopted to be better off. Being more specific, HYVs fertilizers and other chemical inputs were widely adopted by rice producers in Asia and Latin America. For instance, R. W. Herdt and C. Capule ( 1983 ) performed an analysis of 11 Asian countries where modern rice varieties were introduced They estimated the increase of rice production monetary value because of HYVs which was $4.5 billion per year since 1965 to 1980 (only Bangladesh, Burma, China, India, Indonesia, Philippines, Sri Lanka and Thailand were considered in this estimation; whe re 85% of rice production comes from) Also, they pointed out to fertilizers, irrigation and other factors as other sources that contributed to the improvement of rice production. Dana Dalrymple (1986) also developed a descriptive analysis of the introduct ion of HYVs in the rice sector of Latin American Countries. HYVs introduction took place in Colombia with the help of the International Center for Tropical Agriculture (CIAT, in Spanish) in 1967. Thus, Latin America was t he second region far behind Asia, in adopting rice HY ( not including Brazil due to its big portion rice land) a 70% of the rice area was cultivated with th is technology in the season 1981 82 when in 1969 70 was around 3 %. Additionally, the increment of rice production in these region s was more by the improvement of yield per hectare rather than land expansion, demonstrating the e fficiency of these HYVs. R ice yield augmented 109% from 19 60 to 2000 (see Prabhu Pingali and Terri Raney 2005 )

PAGE 23

23 Fi nally, the introduction of rice HY Vs were also encouraged by Institutions specialized in this field such as I nternational R ice R esearch I nstitute in Philippines, International Maize and W heat Improvement Center in Mexico, International Center for Tropical Agriculture, Agricultural Research for Development in Africa etc. On the other hand, different researchers had observed that there was still a need of developing innovations in agriculture. As a respond, Biotechnology Era came to replace past technologies in order to continue improving productivity. Biotechnology or Gene Revolution has its start point with the ascertainment of the principle of heredity made by Gregor Mendel, known as the Founder of Genetics F rederick H. Buttel, Martin Kenney and Jr. Jack Kloppenburg ( 1985 ) suggest ed that this Biotechnology Era would be the replacement of Green Revolution. They also made an important differentiation between the Green Revolution and Biorevolution. In the forme r, yield was meant to be mainly improved per hectare; meanwhile, the latter promoted the expansion of crops. According to Matin Qaim (2005) the adoption of biotechnologies may be considered as the most rapid diffused inn ovation not seen before. He notes that the most popular biotechnology is Genetically Modified Organisms (GMO) in agriculture which is a not irreversible technology. He also mentioned to the herbicide to lerance as the most utilized GM event. Biotechnologies were analyzed for rice production, during the 1990s. For instance, IRRI was researching on a type of rice with characteristics such as resistant to biotic and abiotic stress and, blast and drought tolerant ( see J. Ben nett 1995 ). Leonard Gianessi, Sujatha Sankula and Nathan Reigner ( 2003 ) simulate d the benefits for Greece Italy, Spain and Portugal ( countries representing a 97% of rice production in Europe) from adopting a G M rice variety to alleviate the

PAGE 24

24 herbicide resistant weed problem a s this biotechnology had not been introduced at that moment For instance, Greece, Spain and Portugal utilizing the glyphosate resistant rice variety would reduce the cost on weed control b y 50%. Moreover, Kym Anderson, Lee A. Jackson an d Chantal P. Nielsen ( 2005 ) also reproduce d the economic gains of adopting GM crops in Asia. Among those GM crops is golden rice which a type of modified rice providin g vitamin A. According to them, this GM rice was to better feed to those malnourished in Asian countries rather than productivity enhancement T hey found that those unskilled workers would benefit from this biotechnology because the better nourishment woul d cause a productivity improvement. Meanwhile, Bao Rong Lu and Allison A. Snow ( 2005 ) noted that the glufosinate resistant ric e was the unique event deregulated in United States by that time. However, there was a lit tle acceptance of that biotechnology over rice farms. Biotechnology on rice production is still an ongoing research. Most of the papers present potential benefit of this biotechnology for rice farmers. Various authors emphasized that the real adoption of biotechnology has not started yet. F or example Ecuador does not allow any type of transgenic crops ( Ecuadorian Const. of 2008 ): Art. 401. Ecuador is declared a free transgenic crop and seed country. Only in exceptional cases and of National Interest, th e President may demand the introduction of this technology, giving proper fundaments which wi ll be approved by the Congress. R is ky experimental biotechnologies are strictly prohibited. Issues of n egative externalities of those technologies were also parts of T he Green and Gene eras (e.g. adverse environmental impacts, monoculture, soil erosion, labor displacement, biodiversity loss, etc.). Some pointed out that biotechnologies address environmental concerns in some ways that HYVs did not. Yet, uncertainty still exits about full effects on biodiversity. However, those technologies presented a

PAGE 25

25 potential in increasing th e productivity of staple crops, contributing to combat against famine and poverty in developing countries, mainly. For that reason, the develo pment of research on technology adoption is imperative to better comprehend the impact of those technologies on welfare, biodiversity, and other econ omic and non economic development variables. A quick glance was presented of two important phases in agric ulture, Green and Gene Revolution, where a wave of innovations or technological changes arrived to analyzed given its great influence on rice yield increase. Meanwhile, herbicide resistan t rice varieties were the most researched biotechnology for rice production in the still Gene Revolution. UDP is a technology that use efficiently an innovation adopted in T he Green Revolution Urea fertilizer ( widely used to increase rice yield in t he 197 0s ) Determinant s of Adoption Decision After having seen some technologies introduced in agriculture, it is time to observe the main factors affecting the decision of whether a dopt or not To mainta in an order, these factors are grouped i n global topics s uch as: Adoption Phases; Credit Market, Risk Aversion and Insurance ; Characteristics ; Farm Characteristics ; On farm and Off farm activities and Non work Income ; Production Factors; and Social N etwork and Information Sharing Adoption Phase s Some researchers have thought of technology adoption divided into phases. P. Kristjanson et al. ( 2005 ) examined the entire phases of Improved Dual Purpose Cowpea adoption in the dry Savannas of Nigeria. They sep arated such adoption process into 4 phases: 1) Introductory tr aining and demonstration; 2) F arm

PAGE 26

26 participatory trials; and 2) Farmer to farmer seed diffusion. Similarly, David Pannell ( 2007 ) proposed 6 stages th at an innovation introduction should face : 1) Awareness of the problem or opportunity; 2) Non t rial evaluation; 3) Trial evaluation; 4) Adoption; 5) R eview and modification ; and 6) Non adoption or dis adoption. Keeping in mind these possible phase s one ca n think of determin ants taking place at each phase But, this subject is beyond the scope of this thesis and I analyze them as a whole. On the other hand, o ne can ask who is in charge in developing and fostering the modern technology to be utilized in far ming. As this is a national concern, governments through universities or other institutions a re first called to the provision of different agricul tural knowledge to farmers. I nnovations are believed to take place because of inc entives and government polici es (see David Sunding and David Zilberman 2001 ) However, it is important the participation of other actors to develop new ideas in agriculture because of the insufficient government resources. T o measure the e xten t of UDP diffusion, s pecial part of the questionnaire was designed to collect information about the spread of UDP knowledge over the rice zones in Guayas Province. Credit Market, Risk Aversion and Insurance To start a production, a farmer needs to have acc ess to different resources. Similarly, some technologies could require initial investment and access to credit may be a determinant to adopt an innovation. Given the characteristics of UDP, increase of labor, a farmer would probably be more willing to intr oduce it in the rice production having possibilities to borrow money. In Gershon Feder 's and Dina L. Umali 's ( 1993 ) reviews was note d that access to credit market would be affecting the adoption behavior positively For instance, the introduction of MVs into the production system demands a farmer to buy other inputs

PAGE 27

27 such as fertilizers. But, this infl uence of credit availability is stronger in the early phases of adoption. Similarly, Franklin Simtowe and Manfred Zeller ( 2007 ) analyze d the acceptance of hybrid maize in Malawi. They first separated adopters, based on probabilities, in two types: credit constrained and unconstrained. Then, using a Double Hurdle model, the pro bability of hybrid maize was estimated. As a result, access to credit significantly and positively determined adoption of this maize, only in the credit constrained group. According to Rajni Jain, Alka Arora and S.S. R aju ( 2009 ) one of the impediments of technological changes adoption found in India is the lack of rural credit due to the low profitability and no viability of rural finance sector. Risk aversion is also other variable included to determine adoption. O ne can think of small farmers who are the most needed of new improved technologies are constrained by scarce resources that make them more risk averse. For instance, Rajni Jain, Alka Arora and S.S. Raju ( 2009 ) mention ed that poverty would hinder the adoptio n decision, mak ing more risk averse to farmers. And poverty would finally be intensified itself. Conor Keelan et al. ( 2009 ) use d as risk measure a dummy variable (i.e. 1, if one is willing to grow a crop and 0, otherwise); but this variable was not significant. They also mentioned that solvency ratio was utilized as a risk measurement: the higher the ratio, the more risk averse Having adopted technologies in the past may explain the willingness to adopt new ones. This fact may also work as a measure of risk aversion or entrepreneurship. Perhaps, depending on the design of the risk adverse variable, a different effect on adoption could be obtained. Agricultural insurance acquisiti on can also be considered as a determinant of technology embracing. Crop insurance availability make farmer encourage to adopt

PAGE 28

28 improve technologies ( see S. S. Raju and Ramesh Chand 2008 ). Mara Jos Castillo ( 2011 ) made the same affirmation: counting with an insurance market may give better capacities to a farmer to take risks from innovative changes implementation. She also noted that given the complexities of acquiring an insurance (i.e. risk coverage, high prices, etc.) and low demand of insurance have not let the insurance market develop totally, in Ecuador. Household s Characteristics Th ere are several factors that may affect the adoption decision at the farm l evel. F or instance, land tenure, land size, household members, consumption, dista nce of a farm to main town, etc Here, I discuss about these variables. Being a landowner has not been clearly seen as a determinant of adoption choice. D. Joshua Qualls et al. ( 2012 ) indicated that literatures related to the effects of land tenure have not presented consistency on describing the real impact on adoption. On the other hand, Gershon Fed er and Dina L. Umali ( 1993 ) pointed out that tenure is an important determining the adoption speed. However, the affection of land tenure is relied on the type of new technology being introduced (see Conor Keelan et al. 20 09 ). Farm size is also thought of factor to be included in the adoption model. Madhu Khanna ( 2001 ) demonstrated that the farm size had a significant influence on the adoption of variable rate technology (input applicati on through computer controlled device); one of the reasons of the farm size influence is because of return to scale from fixed costs of equipment, the cost of information acquisition and learning. Correspondingly, D. Joshua Qualls et al. ( 2012 ) observed a positive influence of land size on the adoption of switchgrass in United States. Utilizing the intensity of adoption

PAGE 29

29 ratio as a dependent variable, they show that an increase of one acre would lead to a farmer t o dedicate 0.0019 acres to switchgrass. Looking at the family size, if a farmer has a high number of members to take care of, the adoption of any technology may be affected negatively, as this farmer must devote his time to their children or even his elde rly parents. Jorge Fernandez Cornejo, Chad Hendricks and Ashok K. Mishra ( 2005 ) studied the acceptance of herbicide resistant soybean in United States. Through a Multivariate Probit Model, they determined a negative impact of number of children in the household on the adoption of this genetically modified crop. Fidelia N. Nnadi and Chidi Nnadi ( 2009 ) obtained a negative impact of farm size on maize/cassava inter crop adoption. According to their Logistic Regression, an augment of one member in the household would lead to a reduction of 3% on the probability of adoption. Their implication of this result is large family needs more responsibility to ensure the food a ccess and therefore, there would be less resources to finance the production of maize/cassava production. Finally, Gunnar Breustedt, Jrg Mller Scheeel and Uwe Latacz Lohmann ( 2008 ) studying the adoption of GM oilseed rape in Germany, also found that children have negative effect on adoption decision when female farmers are carrying out the production The problem presented here is that female farmers have to take care of their children, dedicating time to othe r activities rather than farming activities. The distance between a household and the marketplace could be a measure of market access Madhu Khanna ( 2001 ) observed s to adopt increase in certain s t ates of USA because of the proximity to professional services, which in the end means facility to access the market. Moreover, S. J. Staal et al. ( 2002 )

PAGE 30

30 show ed the importance of location (embracing market access, demo graphics and agro climate) on technologies improving diary production in Kenya through the utilization of Geographic Information System (GIS) data. Two common problems faced by rice producers are flood and drought. Maybe, these problems can encourage far mers to find technologies to some extent overcome such natural affectations. Progress H. Nyanga et al. ( 2011 ) could observe an insignificant influence of drought and flood perception made by small farmers on th e adoption of conservation agriculture. They conclude that there are other major reasons to adopt this new practice. Farmers Characteristics can also play an important role. Some of the most common ch aracteristics are education, gender and age that may result in different affections on adoption decision. In term of education, perhaps a farmer with higher level of education may be more receptive to new ideas or innovations. However, there have been dif ferent impacts Rajni Jain, Alka Arora and S.S. Raju ( 2009 ) found that literacy was not a key d eterminant of adoption in India; S tates having educated people have a low rate of adoption. They conclude that this low ra tio is because those educated young farmers opt to look for other alternatives. Conor Keelan et al. ( 2009 ) demonstrated that general education has insignificant effect on the adoption of GM crop. But looking at the agricu ltural education, this variable has a great impact on adoption; ideas and the willingness to analyze technological alternatives are sought by those farmers having higher level of agricultural adoption. Instead, Sa nzidur Rahman ( 2008 ) found that the level of household head education has significance in explaining the acceptance of only

PAGE 31

31 diversified cropping systems. His study was carried out to explain the adoption of diversified cropping systems and/or MVs, given t he fact that rice monoculture become popular in Bangladesh. Thus, his results, using a Bivariate Probit model to estimate the adoption, showed that one more year of education obtained by the household head may increase the adoption of this diversified crop ping system by 3%. Gershon Feder and Dina L. Umali ( 1993 ) also mentioned education as those factors affecting the adoption decision in the initial stage of a technology adoption process; in the later phases this fac tor was no longer significant. Finally the education level would have different effects depending on the technology being examined because some technologies would be less complex than others. By history, males have been in a bigger proportion in agricul ture. However, female produ cers, given the needs of welfare improvement, have been integrating in agricultural production. Males and females have different perspectives, objectives and constraints and as a result, innovation adoption may be affected signif icantly. For instance Joseph Bwire ( 2008 ) highlighted in his study of improved meat goat adoption in Uganda that women represent the highest portion of non adopters of this technology. He suggests the empowerment o f women through education and small business encouragement in order to compete fairly with male producers. Cheryl R. Doss and Michael L. Morris ( 200 0 ) did not find a significant effect of gender on HYVs and fertil izer adoption in Ghana. However, they asked themselves if gender may affect adoption indirectly. For instance, they show that land ownership is positive related to the adoption of both technology and land is also related to gender; wom en are less likely to access land. Other fact is that women are not in touch with extension worker frequently;

PAGE 32

32 affecting the transfer of technology. In spite of this gender insignificance, I introduce this variable in the model to really know its effects on UDP case. Age is a lso an explanatory variable for adoption selection. For instance, D. Joshua Qualls et al. ( 2012 ) cite d that because of the long term benefits return of some technologies, older farmer, with life expectancy short less likely to adopt those technologies. Conor Keelan et al. ( 2009 ) also showed that age has a negative effect on adoption choice even though it is insignificant. Haitao Wu et al. ( 2010 ) did not find a significant effect of age on adoption decision. On farm and Off farm A ctivities and Non work Income UDP technology is labor intensive, requiring doubling the workers per hectare with respect to t he traditio nal production syst em. However, having a large family labor with members involved in rice production would produce better gains adopting UDP. According to Samuel Mora and Paul Herrera ( 2010 ) UDP technology would be more accepted in those very small farms w here there is family labor available; this family labor use would not really signify an expense, which may make UDP more attractive. Cheryl R. Doss ( 2006 ) noted an importa nt fact about access to labor market. Given a new technology, if a farmer cannot satisfy the labor required with household members, he/she must go to the labor market. And, if this labor market is not available, labor intensive technologies would not be ad opted. This conclusion emphasizes not only the importance of family labor, but also a correct access to the labor market. Gershon Feder and Dina L. Umali ( 1993 ) mentioned that the decision of how much land must be a llocated with HYVs production is taken simultaneously with the levels of family and hired labor. On the other hand, Gerard E. D'Souza, Douglas Cyphers and Tim T. Phipps ( 1993 ) estimated, through a logit model, the probability of sustainable

PAGE 33

33 agricultural practices in West Virginia. Among the variables with insignificant effects was labor (defined as 1 if a producer hire worker and 0, otherwise). Off farm work must be also considered in the adoption modeling. I f a farmer could make a good income with off farm work, the adoption of a technology may be hampered; he may not have incentives to change their production system while ma king enough money to subsist. Thus, off farm work and income would tend to be negativel y correlated with the adoption of UDP. For instance, Haluk Gedikoglu and Laura M.J. McCann ( 2007 ) found a negative impact of off work income on the adoption of labor intensive technologies. Joseph Bwire ( 2008 ) referred to ambiguous effect of off farm work. Off farm work may reduce the time availability to be dedicated to the production activities of the new technologies. But, the extra income from this off farm work may wor k as a resource of capital for the production. Similarly, government cash transfer to poor people may function as off farm income. Some farmers are beneficiaries of these transfers in Ecuador. Human Development Bonus is given to those most needed families in order for them to ensure access to basic needs such food and education (Ministerio de Inclusion Economica y Social 2012) Thus, farmers would have more resources to be used in the production system and they might take risks from adopting new technolog ies given that their basic needs are being satisfied. In contrast, a farmer may be disincentive to improve his production to get a higher income because of these basic needs satisfaction. Rice Production Factors The main benefit of UDP is the reduction o f Urea applied. Hence, a farmer would liberate resources to other expe nses adopting this technology Moreover, this t echnology makes rice business have economies of scale ; better rice yield by utilizing

PAGE 34

34 less Urea fertilizer and spending less on weeding (a ssuming that labor requirement is satisfied by family labor) Those having greater production costs may be more likely to adopt this technology. Timothy H. Hannan and John M. McDowell ( 1984 ) examined the adopt ion of automatic teller machines (ATMs) in the banking industry, in United States. They found that ATMs adoption may occur in those zones having a high wage rate paid to the employers because of the cost savings; ATMs are labor saving. Biotechnologies are also labor saving, reducing the cost of pesticide/herbicid es application. Similarly, technology adoption such as agricultural precision innovation would be more likely to happen in zones where inputs or labor are scarce which ultimately mean highest total production costs ( see S. M. Swinton and J. Lowenberg deboer 2001 ). On the other hand, current yield may negatively explain adoption of a new improved practice. Keeping a perception that current production practice s are sufficient to get appropriate yield, a farmer could be less willing to innovate. Truong Thi Ngoc Chi ( 2008 ) carried out a study about those factors associated with the adoption decision of three technol ogies: Integrated Pest Management row seeding and three reductions, three gains and among others; qualitative data were collected in the rice region of Mekong Delta, Vietnam. In general, education, lack of suitable extension work and perception of yield l oss made farmer have a low adopt ion ratio of these technologies. Most of the income of Ecuadorian rice farmers comes from on farm activities. This on farm income can be treated as measure of the cash availability ( see Cheryl R. Doss 2006 ). Gershon Feder and Dina L. Umali ( 1993 ) mentioned in his review that income has been a positive determinant of erosion control practices in some re search.

PAGE 35

35 Social Network and Information S haring One of the common assumptions among references is the full information in the introduction of an agricultural adoption This means information about the technology is a public good available to everyone. Also, it is important to study how such infor mation may be spread within farmers; social network and the information sharing would overcome constrained diffusion budget of some Institution promoting this technology. Gershon Feder and Dina L. Umali ( 1993 ) defin e d two types of learning about a certain technology: when a farmer experiment himself the new technology (own learning) and when a farmer could get information from others (learning from others). Also, A. D. Foster a nd Rosenzweig ( 1995 ) demonstrate d the importance of social network and the information sharing through the experiment of HYVs; experiences considerably augment the HYV Conley, Timothy G., and Christopher R. Udry ( 2010 ) found that farmers care not only their own experiences choice model affected by ot Heidi Hogset ( 2005 ) highlighted that adoption decision may be affected positively by participating in a social network which can work as a channel of i nformation. In fact, he concluded that the government of Kenya must better comprehend the functionality of social network to successfully diffuse an innovation to farmers However, Oriana Bandiera and Imran Rasul ( 2006 ) revealed behaviors that restrain th is social learning; some of the farmers will prefer to just effect (being in a group wh conducts and vice versa). Dividing social network of these farmers into three groups:

PAGE 36

36 religious, family and friends/neighbors, they assume that the unique source of endogeneity would come from the latte r group as they select friends/neighbors. If this variable is insignificant in a model, one can say that social network effects may be free of this endogeneity when explaining the adoption decision. They found such results in their adoption model (this app roach is implemented in the Tobit model in this thesis). I have discu ssed about factors that are included in the empirical model of this thesis. There are immense varieties of references with different determinants of adoption choice. But, I limit this an alysis to these aforementioned variables.

PAGE 37

37 CHAPTER 3 UREA DEEP PLACEMENT TECHNOLOGY A summary of UDP experiences in Asian countries, specifically in Bangladesh is presented in this C hapter. Also, a brief story is given of how UDP was introduced in Ecua dor and the results of the experiments carried out by Escuela Superior Poli tecnica del Litoral, University of Florida and PL 480 of USDA (2008) Keeping in mind that UDP is a technology developed to reduce the loss of the main fertilizer used in rice produ ction, Urea. This chemical fert Urea Deep Placement Functionality UDP is a simple technology that has been tested and developed by variou s organizations. A mong these are the International Fertilizer Development Center and Intern ational Rice Research Institute Some of the cou ntries where this technology has been applied are: Afghanistan, Burkina Faso, India, Madagascar, Malawi, Mali, Niger, Nigeria, Rwanda, Senegal and Togo. However, the nation where UDP was most widesp read is Bangladesh where the Bangladeshi government introduced UDP to more than 2 million of farms in 2009 ( International Fertilizer Development Center 2012 ) Two aspect of this innovat ion must be differentiated : a) the production of Urea Briquettes ( UB s ) through the Briquetting Machine ( BM ) and ; b) the new deep application. In this study, the second aspect is examined as the new technology taking the first for granted Before describ ing the real method to be adopted by rice farmer s the briquetting process need s to be explained. T his process has been used commercially since 1840s. But, it was also related to produce comp acted balls or pellets of Urea (or UB s ) ; s in 0.8 g. to 2.2 g. ( M. S. Lupin et al. 1983 ; N. K. Savant et al.

PAGE 38

38 1991 ). N. K. Savant and P. J. Stangel ( 1990 ) pointed that the first developers of a level village BM were the Fujian Academy of Agricultural Sciences and Yongtai Farm Machinery Factory. Then, I nternational F ertilizer D evelopment C enter Metal Industries Development Center and Soil Research modified this machine whose cost was reduced to less than US$1200. F igure 3 1 shows the briquetting machine which is a straightforward device but with a great importance. Thus, farmer only has to load BM with conventional Urea (top of the machine) and then, Urea briquettes are made by p ressuring. The production per hour depends on the type of machine, but considering one with diesel engi ne it can prepare 200 250 kg/h r In relation to the new Urea application, F igure 3 2 i s presented in order to de scribe this innovation visually. Explain ing F igure 3 2 X there are 6 groups of 4 seedlings. Ever y seedling is separate 15 cm. from each other in the same group and each group holds a distance of 25 cm to another one symbolizes a Urea briquette that must be placed by hand (sometimes feet are used) to a depth of around 5 10 cm. Such briquettes are allocated into the soil right in the middle of the four seedlings and this placement is similar to the transplantation of rice plants. UBs must be place d up to 20 days after transplanting. Finally, a farmer only has to fertilize once during the whole rice season. The reason to follow UDP technique is to avoid the loss of Urea When Urea is applied as in its traditional manner ( broadcast/ spreading), fertil izer may only be seized in a 40% by rice plant s because a 60% may be lost through leaching, volatilization, denitrification and/or runoff ( N. K. Savant and P. J. Stangel 1990 ). In applying placed

PAGE 39

39 deep briquettes, Ur ea would not be exposed to such losses and rice plants can catch more of N which is now released slowly. In the end, the benefits of UDP are: Urea saving (not exposed to common losses ) yield increase (N is better caught by plants) and per se better income and positive environment impact (N is part of gasses contributing to the global warming and affects Aquatic resources through runo ff). On the other hand, UDP is labor extensive given the way of how to apply Urea fertilizer. Additionally, International Fertilizer Development Center ( 2008 ) reported that costs of harvest and post harvest are increased as well, but weeding control cost is reduced, instead. At national level, bene fits are: reduction on Urea subsidy and importation, job creation and food security. Bangladeshi Experiences As said before, UDP has been widely adopted in Asia principally. Bangladesh is a representative example with more than 1 million of hectares dedicated to th is innovative fertilization method ( International Fertilizer Development Center 2012 ) In order to document the results of UDP, I base this discussion on two fundame ntal studies : 1) I nt ernational Fertilizer Development Center. 2008. "Expansion of Urea Deep Placement Technology in 80 Upazilas of Bangladesh during Boro 2008: An Assessment of Project Impact." and ; 2) T hompson, Thomas P., and Joaquin Sanabria. 2009. The Division of Labor and Agricultural Innovation in Bangladesh: Dimensions of Gender. Muscle Shoals: International Center for Soil Fertility and Agricultural Development These two investigations were developed in Bangladesh during the period of time known as Boro ( rice cultivati on season from December to February), 2007 08. Moreover, a total of 3,230 UDP production units (or households ) were considered for this analysis. In this part, a discussion is given about the most important benefits/costs of UDP found in

PAGE 40

40 Bangladesh. S uch impacts are: Urea saving, income increase, labor impact, and subsidy and import reduction In 2008 there were 384,550 farmers that adopted UDP across Bangladesh From this number was taken the sample farmers (3,230) whose mean number of hectares dedicate d to UDP was 0.26 ha 1 Meanwhile, traditional paddy production land was 0.54 c onclusion, this result shows that most of the adopters were small farmers. Also, there was a reduct ion on the applied Urea of around 36% when applying UBs. On average, 170 kg/ha of Urea was applied in UDP lands. Meanwhile, Urea use in conventional paddy hectares was 2 67.2 kg/ha, on average. In terms of costs, this reduction signified a 25.3% of saving. A significant reduction experimented by the 97.3% of the interviewed farmers Other objective of UDP was to improve rice y ield and income per se On average, an augmentation of UDP rice yield per hectare of 17.3% let farmers make a 60% of net income per he ctare contrasting to conventional Urea application. In absolute numbers, the mean UDP rice yield was 7,646kg/ha and no n UDP yield was 6,520 kg/ha; the incomes were US$694.56 (UDP) and US$433.62 (non UDP). Finally, the cost benefit ratio was for UDP 1.53 an d for conventional Urea application 1.33, meaning that per each dollar invested in rice production a farmer got US $1.53 in UDP and US $ 1.33 for conventional fertilization. These yield and income increases were reported by the 95.5% and 51.9% of the farmers, respectively In term s of subsidy, Bangladeshi government could save US$6 million given the reduction of 14,000 Urea tons ( or 93 kg/ha). Meanwhile, Urea importations were also diminished, 1 A total of around 280,000 ha dedic ated to UDP

PAGE 41

41 totalizing US$7 million of saving (adapted from I nternational Ferti lizer Development Center. 2008. "Expansion of Urea Deep Placement Technology in 80 Upazilas of Bangladesh during Boro 2008: An Assessment of Project Impact." ) Bangladeshi l abor market was also impacted by UDP adoption According to Thomas P. Thompson and Joaquin Sanabria ( 2009 ) hired labor decreased in an absolute value of 8.6 days/ha, on average. They noted that UDP extra labor requirement was supplied by household labor wh ose increase was 19 days/has. Event ually, the authors concluded that UDP should not influence positively the hired labor cost because household members will cover that demand; of course, this would depend on the household structure in each adopting region. One can think of two facts of UDP adoption : 1) remaining household time that could be dedicated to UDP ; and 2) these farmers were small ones (average UDP land size was 0.26 ha). Perhaps, these two points should be taken into account for other countries to introduce UPD. On the other hand, other costs were also affected by UDP. For instance, hired labor costs for weed control were reduced in 11.3% (weed did not grow because roots did not receive enough Urea because of deep placement and due to UDP plants sooner growth, weed cannot receive su nlight to develop itself ) and post harvest labor, which is an activity done by women, rose in 20 50 days/ha (yield was improved with UDP). At national level, new employments were very significant in B angladesh. There were a total of 1.43 million of d irect agricultural jobs created in 2008. However, this number did not contain those working on briquetting process (production of UB s and BM). An impo rtant consequence of UDP is the wage equality improvement between men and women; only a 3.62% of women received a lower wage than men. This fact is related to the increase

PAGE 42

42 of post ( adapted from T hompson, Thomas P., and Joaquin Sanabria. 2009. The Division of Labor and Agricultural Innovation in Bang ladesh: Dimensions of Gender. Muscle Shoals: International Center for Soil Fertility and Agricultural Development ). Putting benefits and costs in a scale, one can clear ly see the huge positive impact produced in Bangladesh after UDP adoption. Trials of U DP were also performed in Ecuador. A brief summary of the results of these trials are presented in the following part. Ecuadorian Experiences In order to improve agricultural efficiency in Ecuador, UDP was brought through the project Implementation of a Income in the Ecuadorian Coast: Urea Deep Placement and Microcredit ( Escuela Superior Politecnica del Litoral, University of Florida, and USDA PL 480. 2008. "Implementacin de un Programa para Mejoramiento Del Ingreso de Pequeos Productores de Arroz en el Litoral Ecuatoriano: Aplicacin Profunda de Briquetas de Urea y Microcrdito." ) Eventually, the introduction of UDP started with a study called Agro Socio Economic and Ecologic Conditions of Diverse Rice P roduction Systems of Small Farmers in Guayas and Los Rios, Ecuador ( Hildebrand, P. E., L. Andrade, W. Bowen, R. Espinel, P. Herrera, P. Jaramillo, I. Medina, S. Mora, A. Santos, C. Toledo, e t al. 2008. "Condiciones Agro Socio Econmicas y Ecolgicas de los Diversos Sistemas de Produccin de Arroz de Pequeos Productores en Guayas y Los Ros, Ecuador." ) main objective of study was to observe the feasibility of this innovation in Ecuadorian rice zones Subsequently, UDP promotion too k place through exte nsion

PAGE 43

43 work and experiments 2 ; its final productio t hesis is one of them). In this part, I describe the work done since 2009 Starting with the first study mentioned implement collect relevant information through, more than interview, an informal conversation without a determined list of topics to be discussed. The idea behind this method is that far mers feel more comfortable and relax and without a frame of questions, the quality and the quantity of information are improved. As mentioned earlier, the main goal of this study was to better understand the liveli hood system of Ecuadorian rice farmers in Guayas and Los Rios P rovinces mainly 3 The following is a summary of the most important findings of this study (adapted from Hildebrand, P. E., L. Andrade, W. Bowen, R. Espinel, P. Herrera, P. Jaramillo, I. Medina, S. Mora, A. Santos, C. Toledo, et al. 200 8. "Condiciones Agro Socio Econmicas y Ecolgicas de los Diversos Sistemas de Produccin de Arroz de Pequeos Productores en Guayas y Los Ros, Ecuador." ) : Farmers are very organized in term of rice production but not in commercialization. There are dive rse types of rice production systems; some farmers differ in their rice practices even in the same Cooperative or sector. Urea price growth resulted to a great incentive for UDP adoption. The majority of farmers only have elementary education Despite the risk aversion level, farmers showed interest for innovations. through the aforementioned study the following phases were organized to effectively introduce UDP 2 Some experiments were conducted by students of Escuela Superior Politecnica del Litoral: J. Aguiar, D. Aguirre, L. Barzola, O. Calle, J. Mayorga, R. Romero, C. Saenz and T. Villalva. 3 A total of 39 visits were set in 18 villages of Guayas and Los Rios

PAGE 44

44 in Ecuador A basic repl icat ion of BM was made while the original machine was being imported from Bangladesh ( see F igure 3 3 ). This let controlled and uncontrolled UDP experiments be developed 4 In general information is provided below (adapted from Escuela Superior Politecnica del L itoral, University of Florida, and USDA PL 480. 2008. "Implementacin de un Programa para Mejoramiento Del Ingreso de Pequeos Productores de Arroz en el Litoral Ecuatoriano: Aplicacin Profunda de Briquetas de Urea y Microcrdito." ) : There were 5 controll ed experiments were analyzed. The land size us ed in such experiments ranged between 0.01 0.41 ha. There were 11 uncontrolled experiments with farmers who were willing to try UDP. Their areas were between 0.18 0.52 ha. UDP project gave incentives in order to make farmers try UDP fertilization: 50% of the Urea fertilization cost of a land size of 0 .17 ha would be covered by the project. The results of such experiments are now listed as follows (see T able 3 1) : The mean amount of Urea applied was 154.99 kg/ ha for UDP land and 222.31 kg/ha for broadcast land; meaning a reduction of 30.28% of Urea used in a hectare. In almost all of the experiments, UDP yield (Kg/ha) were superior to the broadcast yield. For instance, the best UDP yield of 10,017.43 kg/ha exce eds the conventional production yield in 2,636.17 kg/ha. However, there were 3 farmers with UDP yield inferior ; the lowest UDP yield was 3,197.80 kg/ha 2,070.81 kg/ha less than the broadcast yield. There were UDP yields superior to broadca st yields even with a lower amount of Urea applied in a hectare. For instance, the best UDP yield used 170 kg/ha of Urea less than the traditional fertilization. This explains the inefficiency of this broadcast Urea application. On average, a farmer could make US$1487.92/ha of net income when producing with UDP. In broadcast fertilization, the mean net income was US$1222.03/ha. In this sense, rice net income was in creased by 21.75%, on average. T he average 4 Controlled experiments (C.E.) are referred when students worked themselves with UDP production. Uncontrolled experiments (U.E.) are for those students who based their analysis on result s of farmers experimenting with UDP production.

PAGE 45

45 cost of UDP was almost equal to broadcast cost US $1149.81/ha and US$1149.76/ha respectively Finally, BM replication could produce UBs with different weights. Meanwhile the imported BM only created 2.7 g; the best yield was obtained with 2.7 g. These are those very encouraging results of UDP experiment ation. In addition to these experiments, there were various ways that UDP was promoted in Ecuador. For air, it is the most important one happen ing every year in Guayas Province Moreover, several meetings were set in different rice agricultural associations in Guayas mainly. One of the most representative meetings occurred in Higueron Irrigation Board where around 150 farmers were present for UDP hearing Finally, UDP was also fostered through television, radio, printing press and internet. At this moment, the biggest constraint of this technology in Ecuador is the elaboration of the BM which is very costly given tha t Ecuador is not a steel producer. A thesis c arried by Orlando D. Contreras and Marcelo Espinosa L. ( 2010 ) showed that an investment of US$7947.45 is needed to produce a BM having the characteristics of the imported one whose final cost was around US$2000.

PAGE 46

46 Figure 3 1. Briquetting machine and Urea briquettes (Sources: Internet) Figure 3 2. Urea briquettes placement (Sources: Author)

PAGE 47

47 Figure 3 3 Replication of briquetting machine and imported briquetti ng machine (Sources: Escuela Superior Politecnica del Litoral, University of Florida, and USDA PL 480. 2008. "Implementacin de un Programa para Mejoramiento Del Ingreso de Pequeos Productores de Arroz en el Litoral Ecuatoriano: Aplicacin Profunda de Bri quetas de Urea y Microcrdito." )

PAGE 48

48 Table 3 1. Results of UDP experiments in Ecuador Type of Experiment Ha UB s weight (gr) UDP Urea (kg /ha) Broadcast Urea (kg/ha) UDP yield (kg /ha) C. E 0.41 4.33 214.50 260.00 8386.45 C. E 0.01 3.60 80.00 120.00 6221 .21 C. E 0.01 3.60 80.20 120.00 7146.94 C. E 0.15 3.60 80.20 120.00 7397.07 C. E 0.01 2.70 120.00 120.00 3197.80 U. E. 0.23 2.70 180.00 250.00 7552.36 U. E. 0.52 2.70 180.00 350.00 6281.23 U. E. 0.11 2.70 180.00 200.00 10544.66 U. E. 0.18 2.70 1 80.00 300.00 10017.43 U. E. 0.18 2.70 180.00 250.00 5272.33 U. E. 0.18 2.70 180.00 200.00 10170.86 U. E. 0.18 2.70 180.00 250.00 9298.64 U. E. 0.18 2.70 180.00 350.00 9879.81 U. E. 0.13 2.70 5128.65 U. E. 0.13 2.70 4485.15 U. E. 0.13 2.70 4434.30 Mean 0.17 2.97 154.99 222.31 7213.43 Type of Experiments Broadcast Yield (kg/ha) UDP total costs (US$/ha) Broadcast total cost (US$/ha) UDP Income (US$/ha) Broadcast Income (US$/ha) C. E 7728.10 969.49 974.77 2525.45 2326.99 C. E 6810.44 C. E 6613.20 1193.25 1170.76 3012.91 2588.77 C. E 6840.08 1219.36 1053.48 3088.96 2206.40 C. E 5268.61 558.50 558.51 790.97 1303.18 U. E. 6934.00 1085.05 1128.42 2274.16 2087.96 U. E. 4020.73 1050.87 1050.10 1891.40 1210.72 U. E. 8582.65 1128.50 1135.75 3175.20 2584.40 U. E. 7381.26 1151.32 1140.45 3016.44 2222.64 U. E. 5281.63 1023.75 1084.00 1587.60 1590.40 U. E. 10080.66 1120.37 1129.87 3062.64 3035.48 U. E. 8585.44 1022.37 1067.37 2800.00 2585.24 U. E. 9824.02 10 22.37 1137.87 2975.00 2958.20 U. E. 4578.75 1575.57 1534.93 3419.00 3052.40 U. E. 4422.60 1565.15 1529.36 2990.00 2948.40 U. E. 4313.25 1561.18 1550.69 2956.20 2875.60 Mean 6704.09 1149.81 1149.76 2637.73 2371.79 Source: Escuela Superior Politecni ca del Litoral, University of Florida, and USDA PL 480. 2008. "Implementacin de un Programa para Mejoramiento Del Ingreso de Pequeos Productores de Arroz en el Litoral Ecuatoriano: Aplicacin Profunda de Briquetas de Urea y Microcrdito."

PAGE 49

49 CHAPTER 4 ECUAD ORIAN RICE MARKET AT A GLANCE A n examination of Ecuadorian official data of rice market is carried out in this section In doing so, data from this thesis can be compared with the n ational statistics. The analysis is focused on the main producing zones, f arm size, production costs, rice production over time, exports and imports, international rice partners, and credit and labor markets. It is important to clarify that III Agricultural Census of 2000 is the last national census; the sample was 162.818 produ ction units (PUs) in total ( Sistema de Informacin Nacional de Agricultura, Ganadera, Acuacultura y Pesca. 2012 ) However, there exists the Continuous Survey of Agricultural Land an d Production which has been carried out every year after the 2000 Census to permanently monitor Ecuadorian agriculture ; its sample consists of 6,000 PUs in general. In Ecuador, the main rice zones have always been placed in two provinces, Guayas and Los Rios. These provinces represent th e 94 % of the rice cro pland in 2010, acco rding to the Instituto Nacional de Estadstica y Censos (2012) However, there are more provinces producing rice: Provinces in Coast region: Esmeraldas, Guayas, El Oro, Los Rios, Manabi and Sa nta Elena Provinces in Sierra or Andean: Bolivar, Caar, Cotopaxi, Loja and Pichincha. Provinces in Middle Southeast and Northeast (Amazon region): Napo, Orellana, Sucumbios, and among others. In term s of yield, the best provinc es of those mentioned abov e are : Caar (4.56 MT /ha), Los Rios (4.64 MT /ha), Guayas (4.25 MT /ha) and El Oro (4.15 MT /ha); the lowest yield was seen in Middle Southeast, 0.89 MT /ha At national level, rice yield is 4.34 MT /ha in 2010, while Colombia and Peru (bordering countries) hav e 5.19 and 7.5 MT /ha

PAGE 50

50 respectively; United States presents 8 MT /ha approximately ( Food and Agriculture Organization 2012 ). Farm size is analyzed but only for single crop rice PUs in Ecuador The 2000 Census data are considered for this analysis. Figure 4 1 shows the number of RPUs and hectares categorized into farm size groups. Almost all rice PUs (15,165 ) hold a farm size between 5 ha and 10 ha (>=5 ha /<10 ha ). The smallest (<1 ha) and largest fa rm (>=200ha ) groups contain 6,797 and 498 rice PUs respectively. According to the Ministry of Agriculture ( MA GAP ) farms with less than 20 ha are acknowledged as small ones ( Sistema de Informacin Nacional de Agricultura, Ganadera, Acuacultura y Pesca 201 2). Thus, around 80% of the rice PUs are falling below this threshold. In looking at the total rice hectares, g roup of >=20ha />50ha presents the bigge st amount 63107 ha Meanw hile, the smallest and largest farm size groups possess 3,473 ha and 43,872 ha r espectively. In combining these RPUs and hectares, on one hand, the 80.52% of rice PUs holds 49.68% of the rice hectares appro ximately and on the other hand a 19.48% of rice PUs owns a 50.32% of rice hectares. This disparity of land holding is one of curr ent hotspot issues being discussed in agriculture at this moment in Ecuador As a result, Ecuadorian g in order to distribute the land that is not fulfilling the environmental and social condition (n ot incl 1 Also, average rice production costs are described; costs of 2011. In T able 4 1, one can see three types of mean p er hectare production costs: traditional, semi technified 1 See Ministerio de Agricultura, Ganaderia, Acuacultura and Pesca 2012

PAGE 51

51 and technified 2 Thus, traditional production has its m ai n direct cost per hectare on l abor (soil cleaning, sowing and fertilizer/herbicide/insecticide app lication); US$518/ha or 59% of total cost. Another import cost is related to the expenses on fertilizers (the main is Urea ), US$75.50/ha (9%). While, in the i ndirect costs administration and technical assistance, financial cost and rent cost add up to US$144.48/ha. Hence, a person would go to invest on this rice production US$875.97/ha in a season. Meanwhile, the semi technified production demands more use of m achine/equipment/material; such cost represents the 48% of the total cost (direct and indirect) Fertilizer cost is also significant for a semi technified farmer, 13% (US$151/ha). Here, indirect costs are equal to US$185.39/ha. Consequently, a farmer with this specific production is making an investment of US$1124.03/ha. Finally, technified rice cultivation also shows its strongest expense on machine/equipment/material, US$517.25ha (38% of total cost). The second relevant expense is the Labor with US$252/ha US$217/ha (16%). A farmer dedicates US$221.58/ha to indirect costs. Total investment in this production is up to US$1343.52/ha. To sum up, labor cost decrease for semi technified and technified productions becaus e of the machine/equipment/material increase. Seed costs are the same for three types, fertilizer costs increase importan tly and P hytosanitary cost is not so different between semi technified and technified but with respect to traditional, this cost increa ses in around US$57/ha, on average. In conclusion, if a traditional farmer wants to change to semi technified, his cost would 2 The degree of technification is related to the n ew processes being introduced to improve production (e.g. type of irrigation systems, machines, control of herbicides, and among others). Such degree makes traditional, semi technified and technified b e different between each other.

PAGE 52

52 increase in absolute term around US$248/ha (or 22% more); and from semi technified to technifie d the cost augment would be US$ 219/ ha (or 16% more). Rice yield, rice harvested land and rice produ ction are examined over time ( 2000 10 ) I n F igure 4 2 is seen that rice harvested lands look relatively constant over the ten years in analysis (blue line). In 2003, these lands was the lowes t number put into rice production, 332,837 has. However, this number peaked up to 433,377 has in the next year. Since then, rice hectares have not changed dramatically over time. With reference to the rice production ( MT ), it presents a volatile behavior c ompared to th e harvested land. During 2000 03, rice production was constant. A fterward, there is a drastic change in 2004, when the rice production soared to 1,778,380 MT According to the Inst ituto Nacional de Estadstica y Censos (2011), the latter fact was because of external growth for 2005. Regarding to the yield ( MT /ha), it exhibits an invariable trend during 2000 03. As harvested land and production, yield also raised in 2004. The best yield was of 4.36 MT /ha, in 200 7; and the worse was 2.73 MT /ha, in 2003. In T able 4 2 is showed a ranking of rice yield (MT/ha) of different countries during 2000 10. Egypt wa s selected as the best rice yielding country, having 10.62 MT /ha on average, over time. It demonstrates a stable yield range of 10.03 MT /ha (2000) to 11.11 MT /ha (2011). Australia was the second in this ranking, showing a mean yield of 9.58 MT /ha during this period; its worse yield was 7.29 MT /ha, in 2005 and its best is 11.95 MT /ha, in 2010. The third country was USA with 8.36 MT /ha on average, over time. It shows its lowest and highest yields of 7.76 MT /ha (2000) and 8.92 MT /ha (2007), respectively. Focusin g on two neighbors, Colombia (24 place ) and Peru (7

PAGE 53

53 place), they show an average yields of 6.39 MT /ha and 7.74 MT /ha, respectively. Ecuador is placed in the 49 position of this ranking with 4.42 MT /ha. In conclusion, Ecuador presents a big gap between bord ering countries and other countries in term of rice yield requiring improving production through new technology such as UDP A brief analysis of Ecuadorian rice production in the international market is presented looking at the importations, exportation s, competing countries and main commercial partners in the world. F igure 4 3 shows that Ecuador has unstable exports over time (blue line ). During 2002 03, the exports held constant; right after, they plummeted to the lowest level, US$566,160. Exports reco vered the ir rising trend during 2005 07, having their maximum of US$62,014,430 in 2006. However, they again sank to low levels in 2008 09. On the other hand, imports have a steady climbing trend over these years; the highest values is f ound in 2009, US$167 ,860. Though volatile exports and constant increasing imports, Ecuador presents a positive balance of rice trade (NX) over t ime; meaning that Ecuador is, in the end, a rice seller. For instance, the best NX was of US$61,942,440 in 2006 In T able 4 3 is ob served the types of rice and commercial partners with which Ecuador participates internationally. Ecuador has mainly 6 countries to import and/or export rice production: Chile, Colombia, France, Italy, Spain and USA. For instance, the main rice buyers of E cuadorian broken, white or semi white and huller rice production are Italy (96.59%), Colombia (99.76%) and Chile (81.51%) respectively On the other hand, Ecuador only imported white or semi white rice primarily from USA, 86.75%. Recently, there was a red uction on rice exports to Colombia because of a prohibition to stabilize the internal price and also, Ecuador and Colombia broke up commercial

PAGE 54

54 relationship due to a military issue. However, that portion (FOB: US$8,170,930) was taken by Venezuela that signe d up agricultural agreements with Ecuador. Maybe, Ecuador may lose Colombian rice market participation given the Free Trade Agreement between this country and United States (USA has a rice yield of 8.31 MT /ha, around 4 MT /ha more than Ecuador). Then, new p roposals must appear to be better rice competitors in the international market; UDP might improve that competitiveness Focusing on the credit market there are so me credit programs being given to farmers currently 3 Such programs are: Credito 5 5 5 (up to US$5000, 5 years, 5%, and no collateral), Credito de Desarrollo Humano (up to US$420, 1 year, 5% and no collateral) and microcredits (up to US$20,00 0, 5 years, 11% and collateral) As is observed in F igure 4 4, loans acquired by rice farmers have an incre asing tendency over time. For instance, a drastic growth of 155.45% was observed in 2001. However, this curve presented negative variations in 2002 ( 6.78%), 2003 ( 5.45%) and 2006 ( 11.90%). While, credits dramatically increased in 100.85% in 2 008. Final ly, BNF started lending money at a level of US$2,580,032.00 in 2000 and ended up at US$22,513,164 .00 in 2009. To sum up, it seems tha t farmers are able to access credit s especially to those credit program s fostered by Ecuadorian g overnment. In relation to de offering insurance to mainly four crops: rice, corn, potato and wheat. Also, d isasters to be covered by this insurance are: drought, flood, cold season, humidity excess, fire, uncontrolled plagues and diseases, h ailstorm and hurricane winds. Insurances are subsi dized in 60% by the government; around 1,939.78 hectares 3 See Banco Nacional de Fomento 2012.

PAGE 55

55 have been assured in provinces such as Guayas Los Ros, Manab, Bolvar, Canar and Loja s o far; totalizing US$1,918,857.94. In describing the labor market in the rice sector, the Central Bank provides an updated description of last rice production season. R ice producers hire day laborer for activities like sowing, fertilizer/herbicide/insecti cide application and harvest. Thus, a day laborer may cost, on average, around US$9 when the employer affords food; otherwise, the day laborer cost is US$10. Additionally, Daule and Santa Lucia present ed a low cost of hiring day laborer US $7 (employer do es not pay food). However, there rice producers also paid US$15 per day laborer without food ( Banco Centra del Ecuador 2011 ) The III Agriculture Census of 2000 shows that in general there were 493,003 hired day laborers shared out by 127,834 production un its; around 3 laborers a production unit. In conclusion, having observed the lowest laborer cost in the analyzed canto n s may indicate good labor market access; many people participating makes prices go down. To sum up, Guayas, where this study took place is an important producing province. Also, official data demonstrated that small farmers are the majority in the rice sector, around 80%. But, they are holding not more than 50% of the land for rice. Similarly, this study surveyed very small farmers. On t he other hand, m ean cost a hectare changes importantly from type of production. For those traditional producers, UDP could significantly reduce their costs given that they mostly base their production on the use of fertilizer (knowing that Urea is the main ). In order to be a competitive rice producer internationally, Ecuador needs to improve its rice yield. Countries like Colombia and Peru exceed Ecuadorian yield in 2 and 3 TM/ha. In terms of credit and

PAGE 56

56 labor, rice farmers have good access to these markets. Given UDP characteristics, national statistics presents a scenario where UDP may occur with high probabilities. However, this adoption would be analyzed in the empirical examination.

PAGE 57

57 Figure 4 1. R ice production units and h e ctares by land size group s (Source: Instituto Nacional de Estadistica y Census, Ministerio de Agricultura y Ganaderia and Sistema de Informacion Agropecuaria 2012 ) Table 4 1. Rice product ion c osts in 2011 (US$/ha) US$/h a Traditional Semi technified Technified Direct costs Labor 518.00 120.00 252.00 Seed 47.00 47.00 47.00 Fertilizers 75.50 151.00 217.00 Phytosanitary 30.00 86.65 88.69 Machine/equipment/materials 61.00 534.00 517.25 Total Direct Costs (US$/h a ) 731.50 938.65 1121.94 Indirect Costs Technical assistance and administration (10% direct costs) 73.15 93.87 112.19 Financial costs (yearly interest rate 9.50%/6 months) 34.75 44.59 53.29 Rent costs (5% of total cos ts) 36.58 46.93 56.10 Total Indirect Costs (US$/h a ) 144.47 185.38 221.58 Total Cost (US$/h a ) 875.97 1124.03 1343.52 Source: Instituto Nacional de Estadstica y Censos (2011).

PAGE 58

58 Figure 4 2 Rice y ield ( MT /ha), rice harves ted land (ha) and rice production ( MT ), 2000 10 (Source: Instituto Nacional de Estadstica y Censos 2012 )

PAGE 59

59 Table 4 2. World ran king of rice yield 200 0 10 ( MT /ha ) Ranking Country 2000 2001 2002 2003 2004 2005 1 Egypt 10.03 10.23 10.35 10.75 10.84 11.01 2 Australia 9.1 10.23 9.12 10.54 9.18 7.29 3 USA 7.76 8.03 8.13 8.24 8.63 8.18 4 Greece 7.72 7.85 8.13 7.71 8.77 7.98 5 Spa in 7.79 8.35 7.96 8.02 7.94 7.62 6 Uruguay 7.04 7.39 6.46 6.51 7.46 7.28 7 Peru 7.26 7.38 7.37 7.49 7.1 7.6 8 Turkey 6.65 6.73 6.61 6.31 7.72 7.78 9 El Salvador 6.38 6.65 6.55 7.48 7.32 7.95 10 Korea Republic 7.4 7.54 7 6.53 7.42 7.24 24 Colombia 6. 32 6.23 6.47 6.46 6.54 6.39 49 Ecuador 4.06 3.97 4.27 4.27 4.65 4.3 World 4.29 4.35 4.27 4.36 4.45 4.51 Ranking Country 2006 2007 2008 2009 2010 Average (2000 10) 1 Egypt 11.11 10.77 10.73 10.57 10.39 10.62 2 Australia 11.17 8.98 8.82 8.98 11.95 9.58 3 USA 8.52 8.92 8.46 8.75 8.31 8.36 4 Greece 8.5 8.41 7.42 7.79 7.44 7.98 5 Spain 7.49 7.85 7.63 8.31 8.34 7.94 6 Uruguay 8.04 8.69 8.71 8.83 7.82 7.66 7 Peru 7.58 7.95 8.11 8.15 8.03 7.64 8 Turkey 7.74 7.62 8.35 8.57 9.58 7.6 9 El Salvador 8.15 8.23 8.75 7.61 6.75 7.44 10 Korea Republic 7.4 7 8.15 8.37 7.17 7.38 24 Colombia 6.51 6.65 6.94 6.06 5.72 6.39 49 Ecuador 4.63 4.8 4.48 4.41 4.78 4.42 World 4.55 4.67 4.82 4.77 4.82 4.53 Source: Food and Agriculture Organization 2012

PAGE 60

60 Figure 4 3. Rice e xports rice imports and rice balance of trade NX, fob (thousands, US $) (Source: Instituto Nacional de Estadstica y Censos 2011 ) Table 4 3. Average international p articipation of t he three types of rice, 2000 0 9 Broken White or Semi white Hulled Exports Imports Exports Imports Exports Imports Chile 81.51% Colombia 99.76% France 3.41% 12.68% Italy 96.59% 0.24% 13.25% 3.38% Spain 2.44% USA 86.75% Source: Instituto Nacional de Estadstica y Censos ( 2011).

PAGE 61

61 Figure 4 4. R ice credit access (US$) (Source: Instituto Nacional de Estadstica y Censos 2011 )

PAGE 62

62 CHAPTER 5 METHODOLOGY This section focuses on the justifications and expla nations of the instruments that are employed to respond the rese arch question and to achieve the objective. Thus, the following parts are discussed: 1 ) The sampling method; 2) questionnaire and primary da ta; 3) introduction of the theoretical model and ; 4) empirical model of willingness to adopt (or pay). Sampling Unit of Analysis and Target Population In Ecuador, a n 80% o f the rice production units are small farmers having less than 20 ha. Thus, they are key actors allowing the food security in Ecuador Furthermore, those small farmers are part of a family (or household) who makes decisions about their own production systems. In this sense, the units of analysis are rice producers (or households) in this stud y. Thu s, all the observations are rice farmers. In Ecuador, t here are different typ es of rice sowing: transplantation, broadcast and population would be those rice fa rme rs utilizing transplantation t echnique because plants are placed in rows or lines and the placement of Urea briquettes is easier and more precise However, conversion from broadcast sowing to transplanting may h appen without any inconvenient (both techniqu es incur in similar costs). In fact, sometimes some farmers use transplantation in one season and broadcast in others. Given this chance of conversion and with the purpose of having a greater number of observations the target population considers eith er t ype of farmers (see F igure 5 1) Those producing

PAGE 63

63 with mechanized sowi ng are not taken into account because UDP would not be suitable for them at this moment. Current ly, most of the target population is l ocated in the Ecuadorian Coast. This study examined t he acceptance of the UDP in one of the two main rice Province s of Ecuador, Guayas The fieldwork was developed during the 2011 summer (June July August). T able 5 1 lists all the villages where the survey instrument was applied, 35 villages in total Figure 5 2 shows the distribution ( percentage s ) of the surveys per villages. There was no a visit plan: s urvey application started out in Daule, followed by Santa Lucia and ended up in Daule. In this sense, Peninsula de Animas, El Mate and Bermejo with similar p ercentages, and Naupe were the zones with the majority of applied questionnaires, 8.31%, 7.53% and 6.75% of the surveyed rice farmers respectively. In total, 401 rice farmers were surveyed but after data cleaning, 385 farmers were included in the analysis. This latter number represents a 3.82% of the total rice production units across Daule and Santa Lucia cantons. Sampling Design A n on random sampling method was used to collect the primary data given that there are problems regarding to the information ga thering for agricultural research: incomplete sample frame, interview rejecti on, observation absence, among others. Unfortunately, there was not any l eader or representative of the c ooperatives who can provide W hile doing the pilot survey application, some farmers said no to the questionnaire and others were not at home ; these problems turn more complicated because of the location of units of analysis and the difficulties to reach them to apply the survey in rural zones. Coming back to the same household was not

PAGE 64

64 su itable due to budget constraint In this sense, disadvantages and reasons of having chosen this process of case selection are better explained here There are some disadvantages weakening a non random sampling design. Jr. Royce A. Singleton and Bruce C. Straits ( 2010 ) refer to two weaknesses: 1) the investigator bias could be presented in the selection of the units and 2) Sampling errors and sample precision cannot be estimated be cause the cases were not taken randomly or probability theory would no t be applicable. Consequently, any generalizations cannot be made as a product of an uncertain representativeness of this non random selection. However, there are also strong reasons to consider the application of a n on probability sampling. Even if I consider a list of all f armers participating in all agricultural cooperative s as sample frame, which I could not obtain, would be incomplete because a significant number of rice farmers are not affiliated to any c o operative. As a consequence, the entire population could not be identified and then assign a probability to the cases is not possible Also, difficulties come availability. A common situation is that a farmer is not available or simply refuses to participate in a study and the problem is to reach units of analysis given their locations in rural zones and the dedic ation of more time and money is needed Because of the limited resources, selecting cases not randomly was very feasible. According to Jr. Royce A. Singleton and Bruce C. Straits ( 2010 ) given that one is more interest in know ing more about the problem or the topic being examined that r andom selection may be needle ss at the very early phase. In this case, U DP technology has hardly been introduced to rice farmers in Ecuador; being this technology

PAGE 65

65 unknown by farmers and this study is the very first work on the adoption of this innovation. This is also a justification of a non random sampling. Thus, the non random sampli ng considered was the Snowball or Referral s ampling. However, there are some disadvantages utilizing this sampling: 1) statistical inferences must be done on the initial sam ple, as the inclusion of new individuals in the sample is not random ; 2) sample bias could be generated by the participation of the most willing people to be surveyed because they would be considered outlier s based on cooperation; 3) sample selection bias because some individuals would no t refer to their friends with privacy concerns; protecting their friends from possible sensitive questions in the survey and ; 4) those who have a larger social network would have more chances to be referred and those with s maller social network would be more likely to be excluded from the survey, creating the bias in the sample selection (see Douglas D. Heckathorn 1997) The lack of a sample frame first moved to implement this method. As said above, UDP introduction is a ver y beginning phase where awareness about random selection could be forgotten at this moment. Also, the problem of privacy concerns could be avoided or at least reduced in this study because the survey did not ask for very sensitive information. The social n etwork bias may also be reduced given that these farmers live in relative small villages and the probability that farmers know each other and refer to any other farmer is high T he problem would come from the willingness to participate in the survey or the cooperation bias. However, a researcher could find in any sampl ing very enthusiastic individuals and very reluctant people that could provide true or fake information Still, the aforementioned disadvantages are present in this thesis but there are reason s to think of a reduced affectation of these problems.

PAGE 66

66 Thus, e numerators went to households asking for the application of a questionnaire. If they got a positive respond from the famer, this farmer was asked to name other rice producers. Perhaps, what can make random this sampling method is the probability of finding a farmer willing to participate (or finding a farmer at home). In all villages visited, nobody knew beforehand about this fieldwork. On the other hand, a basic assumption of this sampling desi gn is the desirable cases (households represented by the head) know each other. One reason to hold this assumption is because these rice farmers live in villages composed of 20 up to 200 households. I cannot access any available data to certainty know how many households are in each village but I based these numbers on the fieldwork. In addition, farmers might also be related through cooperatives, irrigation b oards other groups Questionnaire a nd Primary Data In this part a description of the questionnair e is given. The primary data were collected through 11 sections of the survey instrument. The most important sections for this study were: UDP diffusion; social network ; past adoption, perceptions and willingness to pay; production system; credit market; These sections are described briefly. Filter s ection: This section works to identify the units of analysis correctly. The first question asks if the person is a rice producer; a second identifi es the type of sowing (broadcast or transplantation ) ; the last question is to see if a farmer is in charge of the rice. If it was found a person not accompli shing one of these questions, enumerators just moved to other household. UDP diffusion s ection: Du e to the extension work, I use this section to identify users of this UDP, information channels, UDP results, level of knowledge and current

PAGE 67

67 users. To know the results, a matrix asking for amount of inputs and their co st was set in this section. A p articu lar objective of this section is to evaluate the performance of the diffusion work and how this information was shared or transmitted among the households. Social network s ection: Questions about affiliation in an agricultural association or other groups, members, relatives, frequency of meetings, and behavior in each meeting were asked in this section. The goal is to collect valuable data to carry out a social network analysis. Past adoptions, perceptions and willingness to pay s ection: This section has q uestions asking for past technology adoption, time of the adoption and the overall performance of those technologies. Additionally, perceptions on communication among farmers and the possible use of a new fertilization method also were measured through sta tements. Q uestions of w illingness to pay were included in this sect ion; D ouble bounded dichotomous choice was used. In order to measure the environmental impact of UDP on the willingness, two rounds of questions were performed : o ne round consideri ng all the tangible benefits/costs of the UDP and; the other describing both tangible benefits/costs and environmental impacts Then, farmers were told to respond whether accept or not to pay extra dollar for one Urea briquette sack (50 kg) There were thr ee version of the extra payment for the initial and second question of the D ouble bounded format in both rounds. The first bid versions were US$1, US$2 and US$3 Then depending on t he answer of the first question, farmers faced a lower or higher bid: US $0. 50, US $1.50, US $2.50 and US$3.50 For example, if farmer said yes to the first bid of US$2, he would have faced a second bid of US$2.50; otherwise, the

PAGE 68

68 second would have been US$1.50 (see Appendix : Questionnaire). T hese initial bids were established based on: 1) previous conversations with rice farmers when recognizi ng all villages to be surveyed, f armers pointed out that Urea price changed very drastically from place to place, between US$1 to US$3 and; 2) there is the Urea to briquettes whose cost is two dollar more than the conventional Urea Such bids were accepted by rice farmers (see Chapter 6: Empirical Results). However, a D ouble bounded estimation is not utilized in this study; only an exploratory analysis is held lat er. In addition, an open ended question of willingness to pay was asked in both rounds. The decision of asking first the binary questions and then, the open ended question was based on the pilot survey application, where farmers spent too much time idealiz ing what extra payment should be the correct. The risk is to have bias ed farmers to respond the open ended question with values below or as much as the bids of the D ouble bounded questions. But, these bids are consistent with the reality and overestimated extra payments could be avoided from this analysis. Besides, the resource time could be saved and used in other survey applications. Finally, i f farmers responded positively to one of questions in any aforementioned round they were able to respond how man y hectare s they would be willing to dedicate to UDP producti on This question is to construct the key variable Intensity of adoption (UDP potential land divided by the total operated land). This ratio is the dependent variable for the Tobit model because i t is more informative, giving the interest in adopting UDP but also the intensity of that adoption in terms of the hectares dedicated for UDP production.

PAGE 69

69 Production system s ection: A matrix with specific questions a bout the rice production was constructed in this section. As farmers do not have any accountability of their productions, it is understandable that such data obtained from this matrix might be exposed to errors. Latterly the data related to the following information : soil preparation cost, seed cost, irrigation system cost, labor cost, harvest info rmation, etc. These questions are associated with the last period of production. A more summarized matrix was created to ask for other types of cultivation. Credit market s ection: The questionnaire a l so contains a section incorporating data of the credit availability The objective is to know from what sources families can get a credit, the amount of money, the uses of that money and etc. Actually, the Ecuadorian government is providing subsidies to ag ricultural insurances; 60% of the cost. Those subsidies are for the short cycle cultivations (e.g. rice or corn ) mainly. A question was also set to observe if farmers own an agricultural insurance. Labor participation s ection: There are questions asking fo r both on farm and off farm activities. Thus the household head was asked about his time availability and other household members (i.e. mother and one son/daughter). The intention is to estimate the available time of each household and its uses. In additi on, a set of questions was determined to obtain information of other incomes from off farm jobs and non worked money (e.g. remittances, donations, etc. ); also, there are questions regarding savings Household s health s ection: One of the environmental ben efits of this technology UDP is that there could be a reduction of N going into the atmosphere and the Daule River (N is harmf ul in high levels in the water for the aquatic lives but also for

PAGE 70

70 those living close to the river who drink that water ) This sect ion asks for the water consumption, stomach illness suffering, and exp enses on that illness, etc. The objective is to know what people could be affected by in the river and calculate approximately the money saved by implementing the UDP. However, this topic was not considered in this thesis because it is beyond the scope of objective. ection: The instrument also contains a section dedicated to Through a set of questions, household head provide d informat ion of his assets types of utilities, consumption expenses (e.g. food, clothes, and school items), etc. Also, farmers were asked if their cropland faced natural disasters such as flood and drought in the last ye ar One complicated question for farmers was related to the land size. The g overnment has been discussing a new land law which promotes the redistribution of lands. There is article stipulating the basis to expropriate lands and it remarks that small famer s will not be affected by this law. However, those farmers could have been afraid of telling the true. As a result, this question was placed at the end of the questionnaire to not bias other answers. ection: In this part farmers r espond questions about their gender, age, education and marital status. There was a question asking particularly for the a gricultural education received; I could identify what agricultural knowledge was learned by these farmers and who facilitated that edu cation (e.g. government, cooperative, universities, etc. ).

PAGE 71

71 Geographic information s ys tem s ection: S patial data were also collect ed in this study. Farmers were ask ed for the ir location of their household s and location codes were taken through a Global Posi tioning System device. To sum up, t his questionnaire included nine sections, or 57 questi ons. To survey a person took around 30 minutes, o n average. The most time consuming question s were the willingness to pay, which should be explained to the farmers car efully to have a better understanding of it and the production system section that contains several questions about rice cultivation Finally, 5 enumerators were hired; three men and two women. As three of the enumerators live d in one of the target zone s they helped to this study recommending sectors where rice produ cers could be found. This field work started in June and ended in August of 2011. Theoretical and Empirical Model The setting of the theoretical and empir ical model is discussed in thi s part. I n order to estimate the Intensity of Adoption (the percentage of total land that would be dedicated for UDP production) model, is first needed In doing so, a typical producer consumer household model was determined, follow ing Howard N. Barnum and Lyn Squire ( 1979 ) Then, this conceptual model with the empirical one is associated Finally, the Tobit model is explained as well as its maximum likelihood estimation in this section T heoretical Model In this section, I developed a theoretical model through which I expect to see the I first need to set out the theoretical agricultural household model ( Howard N. Barnum and Lyn Squire 1979 ). Suppose that a house R on a given

PAGE 72

72 production unit demands inputs, vector f j to Urea insectic j labor market, household has to decide how to use its time farm activity (Z>0) or when it has to hire labor (Z<0) in a marketplace where everyone offers an own rice consumpt a competi tive market; also it consumes other M problem to be faced by this household is: Max U (R, M, L; E, o i ) R, M, L subject to: Y=Y (H, f j ; A, C) T=H+L+Z p R R+ p M M = p R Y+B+ wZ j *f j Where U(* which is also explained by environmental factors the vector of o i can be number of children, agri cultural affiliation, etc.); Y(* ) is the production f unction af fected by those aforementioned factors and for current broadcast Urea Thus, the first restriction is associated with the level of rice production, the second is the time availability and the third is the expenditure constraint. This is a partial representation of Replacing the optimal amounts in the utility function I may rewrite this function as follows:

PAGE 73

73 V [ p R p M L*, B, w, Y*(H*, f j *; A, C); E, o i ] (5 1) Where V [*] is the ind irect utility and L *, Y*, H and f j are the optimal amount of leisure, hired or sold labor, rice production, on say UDP is introduced in the rice production livelihood of this household ; all the land is for UDP production From E quation 5 1, I set out the new welfare level at: V UDP [ p R p M L UDP B, w, Y UDP (H UDP f UDP j ; A UDP UDP ) G; E UDP o UDP i ] As is known, UDP is technology that let farmers save Urea as well as improve rice yield and per se, in come. Also, it has a positive environmental impact given the lower amount I assume that environmental enhancement takes place in a short term as farmers breath less contaminated air and use cleaner w ater from rivers. However, one requirement is to increase the labor for rice production. Finally, this technology has a direct monetary cost as well, G 1 Thus, UDP adoption occurs if a farmer assuming the cost of UDP, would at least obtain an equal or bet ter utility ( Bryan J. Hubbell, Michele C. Marra and Gerald A. Carlson 2000 ): V UDP [ ] V [ ] Thus, factors impacting directly and indirectly the new level of welfare given the adoption of a new improved technol ogy can be identified Following Bryan J. Hubbell, Michele C. Marra and Gerald A. Carlson ( 2000 ) I assume a stochastic utility function: VUDP [IUDP (5 2) Where matrix of all those factors affecting the utility (not including environment), I is the net income of this family (most of the income would come from rice production) 1 The cost may be for the acquisition of BM in order to produce UB s or only extra cost for UB s sacks

PAGE 74

74 and Urea The deterministic part of the utility is assumed to be a particular linear form: V = X* i + i f each factor in X environment, etc.) subscript C. Thus, E quation 5 2 could be rewritten as follows : UDP I C G) X* C X* UDP C UDP ) By assuming any distribution of the stochastic terms, the vector of parameter s and can be estimated by employing a Likelihood Estim ation (see Bryan J. Hubbell, Michele C. Marra and Gerald A. Carlson 2000 ). Hence, the probability that a farmer is willing to adopt UDP can be estimated, but with a binary model Instead, this study utilizes a m ore informative model, the Intensity of Adoption, that not jus t shows the interest in adopting UDP but also the level of that interest in terms of the land dedicated for such production. Now, t he econometric process is shown in the next section. Empirical Model The IA ratio was designed In defining IA farmer s first took the decision whether pay or not extra for acquiring Urea briquettes sacks, through the D ouble bounded and open ended questions set in t he questionnaire (see Appendix : Questionnaire). Then, if a farmer responded positively to any of these questions, they select the hectares to work with UDP. In calculating IA, I divided the possible area dedicated to UDP production by the total cu rrent operat ed area Again, IA is a more informative variable given that it hectares given to UDP production. On the other hand some values are missing

PAGE 75

75 because some farmers showed n o interest in UDP production (they were not asked for potential UDP land, skip the question because of not interest) or even pre senting awareness they do not want to experiment on their operated land (e.g. they could experiment on donated land). In this s ense, this variable was censored at 0. On the other hand there were values that exceeded the limit of 100%, meaning, farmers want to acquire more than their current hectares (buying or renting) to work with UDP. However, to base this study on the actual r esources of these farmers or be more realistic ( they can just give what they have currently ), IA is also censored at 100%. In sight of such results, IA must be treated as a two limit Tobit. In this section is developed the presentation of the empirical mod e l and how it is estimated 2 Tobit model : Defining the latent Intensity of Adoption as IA* which is explained by a set of independent variables. The empirical model is defined as follows: IA i = X + i Where X is a matrix of exogenous variables (e.g social network factors, production environmental concerns and among others), is vector of the parameters belonging to each regressand (including a constant) i the error term followin g a normally identically and independently distributed with mean 0 and variance 2 ; As a result, the distribution of the conditional latent variable IA* given X is also Normally Distributed ( X 2 ). How ever, this variable can only be observe d in a certain range. rvable intensity of adoption is defined as follows: 2 As first thought, a Heckman Model was estimated to solve the problem of the self selective bias. However, there was no significant resu lt of the correlation between the error term of the selection rule and the IA mo del. Therefore, the analysis is based on a two limit Tobit model.

PAGE 76

76 IA i = 0 if IA i 0 or X + i IA i = X + i if 0 < IA i < 1 00 or 0 < X + i < 100 IA i = 100 if IA i 00 or X + i IA is censored at 0 and 100 percent. E stimating this mode l with Ordinary Least Square, bias estimators are produced because this method does not take into consideration the censoring created in the observations (see Peter Kennedy 2008 ; Damodar N. Gujarati and Dawn C. Porter 2009 ). Ma ximum Likelihood Estimation will allow estimating the three parts of this model as follows: / IA i X) = IAi 0 [1 )]* 0
PAGE 77

77 being censo red at 100 and the probability of being censored at 0 (in parenthesis) times the coefficient of the variable that is being analyzed.

PAGE 78

78 F igure 5 1. Types of rice sowing (Source: Author) Table 5 1. Target z ones Province Canton Villages Guayas Daule Brisas de Daule, Bella Esperanz a Clarisa, Coloradel, El Limonal/Los Almendros, Flor de Maria, Huanchichal, Jesus del Gran Poder, Jigual, La Aurora, La Elvira, Las Maravillas, Loma de Papayo, Los Moranillos, Los Que mados, Pajonal/Arriba/Abajo, Patrio Nuevo, Pennsula de Animas, Pinal, Porvenir, Rebeldia, Rio Perdido, San Gabriel, San Vicente, Villa Filadelfia and Yurima. Santa Luci a Barbasco/Central, Barranquilla, Bermejo/Abajo/del Frente, Cooperativa 14 de Octu bre, El Mate/El Encanto, El Limon, El Porvenir Higuern, La Fortuna, La Carmel, Marcela, San Jacinto, San Pablo and Playones, Source: Author

PAGE 79

79 Figure 5 2. Distributions of applied surveys by sample (Source: Author)

PAGE 80

80 CHAPTER 6 EMPIRICAL RESULTS In th is C hapter, a descriptive analysis of the primary data is developed. Af terwards the results of the econometric model are introduced As said before, these analyses were based on information from two cantons, Daul e and Santa Lucia. Finally, MICROSOFT EXC EL EVIEWS a nd STATA were utilized to develop the empirical analyses. Two types of data were found : missing and extreme values For the first case, those observations were replaced with the mean (or mode depending on the variable). Regardin g the second t ype of data, I decide d to leave them in the analysis to avoid further manipulation of the real data and to show what is happing in these zones in reality (these values would be totally justified because of the different agricultural practices or beliefs of these farmers ; other source of these outliers is the extrapolation to mainly rice production data to a hectare ) In addition, I estimated the Tobit model with and without extreme values, getting similar results. Therefore, I did not have any motive to man ipulate these extreme values. In total, 385 farmers are taken into consideration in these analyses. This section is to understand the demographic struc ture of the rice farmers in the sample. I examine gender and age of household heads, education and agricultural instruction. M ost of the household heads are male, 92.2 1%; being a 7.79% female ( see F igure 6 1). However, women are currently participating in the agricultural activities directly or indirectly. I could observe female hav e rice production knowledge when they answered

PAGE 81

81 some questions related to production costs on behalf of males. Moreover, some females are in charge of buying inputs such as fertilizers, pesticides and so on, and others are directly involved in sowing and fe rtilizers/pesticides application. Such helps liberate time needed by the farmer working on the rice crop. On the other hand, F igure 6 2 shows that farmers are in a range of 40 and 60 years old mainly. On average, a farmer is 51 years old over all the vill ages analyzed. lands. Perhaps, the average age could be increased if I would have included the formal owners of the lands 1 Farmers also mentioned their maximum level of ed ucation As F igure 6 3 shows, the majority, 233 farmers got element ary school level, 63 finished high school, 3 an important number of farmers said they did not attend an y forma l education, 85. Subsequently farmers were asked if they have received any agricultural education provided by any organization/institute/government in any place, formally or informally Figure 6 4 demonstrates that o nly 20% of these producers have received such instruction. W hat ha ppened to the other farmers why they said to not have received such education are interesting questions to be responded Maybe, they thought of formal education. However, s ome questions arise due to this fact: do extens ion workers not visit all farmers? Do farmers not accept any help from others? Would this lack of agricultural education and general education be a problem with UDP adoption? 1 This study d ecided to include who are actually producing rice, not matter if farmers have not inherited land s formally. In the end, these farmers are in charge of the rice business and who make the decisions.

PAGE 82

82 According to F igure 6 5, the types of agricultural education received were as f ollows: pesticides/herbicides (32), fertilization (30), soil preparation (9), harvest (6), sowing (7) and others (3). Again, fertilization was important in term of education. Figure 6 6 shows that cooperative organizations were the main sources of this edu cation, 48 farmers mentioned that fact. Government was also a provider of agricultural knowledge according to 16 farmers. Elementary school/high school/universities and others such as agricultural fair, extension workers and specialized companies were also providers of this education, 6 and 10 farmers respectively. Households In this section, an analysis at farm household level is developed with the collected data in these two cantons. The variables to be examined are: l and size, land owner ship drought/flood affection, consumption (i.e. food, clothes, schooling supplies, etc.), and distance to the main town. Additionally, wealth index was constructed with a Factor r, water pump, household ownership, etc. First, I describe hectares given for rice production. This number is very high because these zones are plenty for rice production. As a result of the data analysis, I found that 382 farmers dedicated the whole ope rated land size to rice cultivation, and just three cases did not fully utilized their land. After all their main occu pation for these farmers is the rice production. In this thesis, a ll the studied farmers are s mall because land size ranges between 0.04 and 16.33 hectares (less than 20 h a is a small farm). Looking at F igure 6 7, most of these farmers hold less than 1 hectare, 154 cases. Then, landholders with land size between 1 and 2.99 hectares also represent a significant portion of this sample, 14 0 ob servations. Finally, the leas t landholders are those

PAGE 83

83 maintaining an area between 3 and 4.99 hectares and those having 5 or more hectares, 60 and 31 cases respect ively. On average, a farmer holds 2 hectares for rice production. On the other hand, F igure 6 8 demonstrates that a vast number of producers own their land, 331 observations (85.97%) Meanwhile, 54 farmers rent their land partially or totally. In detail, a person rented 1.6 hectares, on average; while the smallest and largest renter got 0.04 and 9 h ectares, respectively. In addition, the mean cost of renting a hectare w as US$193 for these farmers Also, information of the recent expenses on food, clothes, school supply, household arrangements and other items was captured. Figure 6 9 presents such exp enses by land size groups A slightly positive relationship between these two variables is perceived In fact, the largest landholders are those who most spent on these items on average, US$175.26. However, some farmers of the two smallest land size groups spent more than those on the largest land size groups. These results may have taken place because a greater number of members existed in some small farms. For instance, controlling the sample by family size and land size, the smallest farmers had a per ca pita food consumption of US$13, while, the largest group had US$25. On average, the per capita food consumption was US$16. E xpending more on household consumption does not completely define how much poor or wealthy a farmer can be. According to M. Dekker ( 2006 ) income and expenditure are subjected to measure problems (e.g. fake information) and they can be seen as short term wealth index (WI) because they represent information of a certain point in time (time when the survey was applied) As a result I utilize Factor Analysis to

PAGE 84

84 estimate the WI considering (with relatively long term duration) : Utilities: electricity, drinkable water, telephone service, etc. Household assets: motorcycle/car, tractors, water pump, electric generator, etc. Household construction materials: wall, roof, floor, etc. Household ownership: own, rented, shared and borrowed. The computation of t he Factor Analysis is shown in T a ble 6 1 and T able 6 2. In T able 6 1 is presented the eigenvalue of each factor that are used to decide how many factors should be considered in the analysis; the rule of thumb is that a factor with eigenvalue greater than 1 should be taken into consideration. How ever, one factor is only required to represent the WI. Such a factor explains the 12.45% of the total variance of the linear combination of these variables. Table 6 2 shows the coefficient of each variable in the linear model of the factor 1 (or WI). For i nstance, having storeroom, irrigation pump, electric plant and harvester impact the WI positively; meaning that those with these assets enjoy a better wealth. Also, having a car as well as phone service increases the wealth of a farmer; these items may hel p to the commercialization of their rice production. In term s of health, drinkable water is an important factor for better life Meanwhile, those with a rented or borrowed household would have a lower wealth as well as those with household built with cane; cane is a construction material very cheap and found in poor household usually. However, the unique variable that would not make sense in its coefficient is the Electricity Service whose impact is negative; having this service lets to enjoy the benefits o f assets that enhance wellbeing such as refrigerators, television sets and among others. One assumption about the negative impact of this varia ble o n WI may be that f armers may not have many of these appliances and the Electricity Service is not so indispe nsable and farmers must pay for that service

PAGE 85

85 anyway. T he variable shared household (more than one family living in the same household) would almost have a null effect on WI ; its coefficient tends to zero In general, 17 variables were included in the Facto r Analysis, which determine the wealth index. In Figure 6 10 is shown the WI scores, which can be interpreted as follows: the higher the index, the wealthier a farmer is in terms of assets In Figure 6 11 the wealth level v ariable constructed with the co ntinuous WI variable, and between the aforementioned percentiles are wealthy farmers. Thus, those having a low, average and high we alth level sum 100, 189 and 96 farmers respectively. Also, information about natural affections such as drought and flood suffered by farmers was captured in the questionnaire. Figure 6 12 shows that 51 farmers said to have been affected by droug ht during the last year (2010 11). Meanwhile, 49 producers suffered flood during the same time. These results demonstrate that the majority of farmers did n ot face these natural threats. Also, drought and flood were detrimental to the rice production in the season J anuary April of 2011 Also, Daule River passes through some of the villages visited for this study. Therefore, they are most exposed to suffer flood; meanwhile, those far from the river would be more likely to have drought. Market access was measured throu Time to get the main town Over these villages, transportation (busses and trucks) was a service very limited; the fare was between 25 50 cents. As is shown in Figure 6 13 three groups of time spent to get the market were set: less than 15 minutes, 20 30 minutes and more than 30 minutes. Most of the farmers (171) are placed in the group of 20 30 minutes to

PAGE 86

86 get the main town in both Daule and Santa Lucia cantons ( average spent time : 25 minutes). There are 162 farmers in group less th an 15 minutes the closest to the market (average spent time : 11 minutes). Also, there are 52 farmers in group of more than 30 minutes, the further to the market (average spent time : 54 minutes). This measurement reveals, in some way, the difficulties of s ome farmers to go to the market in terms of time Based on the fieldwork one big obstacle was identified to access the market from some villages: bad condition of the roads. Urea Deep Placement Diffusion To know how UDP knowledge was spread within farme rs in the Rice Zones in Guayas Province, a section was designed to collect this information in the questionnaire. It is important to highlight UDP was first introduced to the rice farmers with meetings in Cooperatives such as Bermejo, Alianza Definitiva an d Agricultores de Babahoyo. Other diffusion tools were: flyers, radio, television, agricultural fair and informal meetings with farmer over the Rice Zones of Daule and Santa Lucia. In this part, a discussion is presented of how many farmers were familiar w ith UDP, how they obtained that knowledge, if they used this innovation and so on. At the midterm of July 2009, the diffusion of UDP started with the visits of extension collaborators to Daule and Santa Lucia cantos mainly. According to Figure 6 1 4 such diffusion was not captured by the majority of these sample farmers; 90.65% (349) said not to have knowledge about UDP technology 2 Perhaps, such a result is a consequence of forgetfulness given the time of introduction of UDP (2009) and the 2 I was not ask ed for UDP per se, but for the urea briquettes or better recognized by them as small urea balls which are put into the soil. Besides, we pro vided farmers flyers to let them identify the technology at the moment of the interview.

PAGE 87

87 questionnaire a pplication (2011). However, a 9.35 % (36) of farmers knew about this new fertilization method. In T able 6 3 is given some statistics of these fa rmers who k n ew and did not know about UDP. For instance, farmers who had knowledge about UDP were, on average, 53 years old, had elementary school studies, lived in the El Mate/El Encanto village in Santa Lucia in their majority affiliated to an agricultural group and held less than 1 ha of rice land size. Meanwhile, farmers who said to not have any UDP knowledge pr esented an mean age of 51 year old, with elementary school instruction mainly as well, with location in the village of Peninsula de Animas in Daule (majority of these farmers) with agricultural group affiliation and with land size group less than 1 ha. Ho wever, farmers who knew and did not know about UDP presented other characteristics of the same variables such as high school level, location in other villages (across all the 35 villages found here) with and without affiliation in an agricultural cooperat ive and with different land size groups. A systematic pattern was not seen of having UDP knowledge or not. On th e other hand, those 36 farmers with knowledge of UDP were asked how they accessed such information. As is showed in Figure 6 15 the ways to ac cess UDP knowledge were: neighbors (8 farmers), Higueron meeting (6 farmers), friends ( 5 farmers) and flyer announcements, agricultural fairs and Alianza Definitiva meeting (3 farmers for the three categories ). For those who said others (10 farmers), the main sources of information were family members, other relative, input suppliers and informal meetings with the extension workers. Moreover, s ome experiments were developed in some villages in the Guayas rice zo nes. Therefore, farmers who knew about UDP we re asked if they also observed the

PAGE 88

88 results of the UDP application (i.e. yield increase, Urea reduction, etc.). Figure 6 16 demonstrates that only 13 out of those 36 farmers were familiar with the final results of UDP experiments. They observed these r esult U nfortunately, these observers could not identify the location of the experiments or enumerators did not find experimenters at home. In the end, only one of these farmers was an UDP experimenter, seeing the results of UDP in his own rice crop (See results in Chapter 3). Also, seven of these farmers affirmed to have heard about UDP technologies again. Currently the imported BM is broken, stopping the promotion and adoption process (BM was transferred to a Cooperative in Daule Can ton and there it was broken ). Finally, a Likert scale was employed to measure the UDP knowledge of these farmers (see Figure 6 17 ). Having three categories for the scale, the results show that 29 farmers know a little about UDP, 4 somewhat and only 3 much T hose who said to have a better understanding of UDP are the farmer s who had experience with it. As a conclusion of this part the diffusion an d promotion the UDP technology must continue for the better understanding of it 80% of the Ecuadorian rice fa rms are small and on which this innovation will have high probabilities of being successful ly adopted Social Network Analysis In the surveyed zones are some well organized groups whose main purpose is to discuss any affair connected with their likelihood production system. Such groups may facilitate the transfer of new technology knowledge within farmers in the group and maybe, between groups. Moreover, social network as part of the so called Social Capital has played an important character to the developm ent o f some rural zones in Ecuador (s ee Pal Herrera et al. 2010 ).This section is intended to confer a modest

PAGE 89

89 description of the actual relationship among farmers in the Daule and Santa Lucia Cantons mainly. The su rvey instrument contains an elaborated social network section behavior in meetings, identi and the estimated number of the social network of each farmer. As said abov e, a majority is involved in agricultural group s 64.42% of farmers ( see F igure 6 18 ). Also, two types of groups were Cooperative Organizations Now t aking a look at Figure 6 19 one can see that the sample is concentrated in the following agricultural groups: Alianza Definitiva Cooperative (56), Mate Irrigation Board (49) and Higuero n Irrigation Board (41). In important, 46 farmers. Additionally, these agricultural associations can be deem ed as small and large, relying on the number of members. On aver age, a farmer said to be part of organization of 515 members; one pointed out to be in one with only five farmers and other alleged to participate in a group of 9000 members. Unfortunately, the verification of these numbers w ith official documents from tho se c oope ratives or irrigations b oards was not possible because there was nobody at the offices at the time of the visits ( coming back was not suitable due to budget constraints). In general, 39 agricultural associations were found in this study. On the o ther hand, I could also see how often farmers meet with their group fellows. Figure 6 20 demonstrates that groups may principally hold their meetings

PAGE 90

90 monthly; 214 farmers confirmed such statistic. Also, weekly and even yearly meetings are attended by these farmers. In addition, these affiliated farmers were required to declare if they are enforced to attend such meetings. As is beheld in Figure 6 21 a 68.95% participated voluntarily without any enforcement. However, 31.05% argued that they have to go in o rder for them to avoid any pecuniary punishment. Once in the meetings, farmers said not to be very participative: non participative and somewhat participative, 21 and 159 respectively (see Figure 6 22 ) As personal experience, I witnessed the development o f a meeting in Higueron where around 200 or 300 farmers were present to hear about UDP technology, of course, after finishing their own discussion. Based on fieldwork, t heir participation was more active in subgroups inside the association; as there were m ore than 200 members; farmers bu ilt their own small cluster. However, no constraints were observed for these farmers to convey their doubts or ideas. In Figure 6 23 is revealed those influential groups which, to certain extent, sway sion s. According to their responses, neighbors/friends, input suppliers and family might be deemed as the most relevant groups; 243, 241 and 210 responses respectively 3 Neighbors/friends and family are obvious important groups. Also, farmers are usually in touch with input suppliers while the rice production season is taking place Interestingly, these sample farmers did not perceive agricultural group 3 It is not an attempt that farmers separate people who are neighbors or friends or agricultural group members b ecause of the difficulty. What was want ed from them is to affirm where most significant or influential people come from.

PAGE 91

91 nor extension workers as influential in their majority; only 54 and 6 cases see them as important, respe ctively. Finally, Figure 6 24 shows a Likert scaled measure of communication level A total of 271 farmers rated communication in their villages as good or very good; while 105 declared the opposite; and 9 farmers were indifferent. Pal Herrera et al. ( 2010 ) described reliable and fellowship. Still, there is a need of further research on social network as c ooperatives or irrigation boards and extension w orkers are source s of new technology knowledge. Technology Adoption Analysis In this section, there is a review of the past technologies that have been introduced in the production systems by these surveyed rice farmers. However, there is a part icular interest to analyze the willingness to adopt UDP technology as well as the key vari able, Intensity of Adoption F armers were asked if they have introduced a new technique (or technology/innovation) into their p roduction system before Figure 6 25 indicates that only 18% of these producers responded positively. In contrast, 79% declared not to have included any innovation; the rest said they did not remember (3%). Perhaps, some farmers may have been using the sam e techniques the whole life; or others may have said no to this question because they did not perceive a past innova tion as a big change. However, A griculture is always in constant adjustment and its methods as well ; implying the persistent introduction of new techniques. Additionally, those past adopters pointed out the innovations introduced in th eir rice activities ( see F igure 6 26 ). For instance, the most common innovation accepted

PAGE 92

92 was linked to the fertilization procedure (39 farmers); one of fertiliz ation techniques mentioned was organic method In this sense, UDP might be an option that can be taken by these farmers given its simple, innovative and efficient fertilization technique. Also, 18 farmers affirmed to have included new sowing methods While pesticides to ha ve utilized new other methods on harvest, soil preparation, irrigation syst em, and plant disease treatment. Consequently, farmers were also asked for th eir experience by adopting the aforementi oned innovations. Thus, they could see an experience as bad, equal or good 4 For example, those adopters of past fertilization techniques had a good (37) and equal (2 ) experiences. Again, results of new fertilizatio n methods are seeing as positive encouraging UDP introduction in agriculture. The rest of the innovations were also considered as good in its majority. In addition, a statement was included to observe how disagree or agree farmers are to adopt a new fer tilization method. A positive response predominated, 350 farmers were agreed. At a glance, one can say UDP adoption could occur without any problem. To clearly observe so, I prepared a section addressing the willingness to adop t Urea Deep Placement. Willi ngness to Pay : A D ouble willingness to adopt UDP (see Chapter 5 Methodology). Here, I name the willingness to pay considering the first round, where the economic benefits and costs of UDP were 4 In term of production, maybe a bad experience is for a decline; equal is for no improvement at all; and a go od is for improvement. However, these experiences were not specified to farmers and then, they decided what to select.

PAGE 93

93 impacts (economic and environmental For this analysis, I do not divide the observations into the three bid versions; I treat the data as a whole. Firstly I analyze the WTP whose benefits and costs are: reduction of applied Urea yield im provement, income increase and labor increase. According to Figure 6 27 a high percentage of this studied farmers are willing to pay extra for Urea briquette sack s independently of the land size: 91.56% (less than 1 ha), 95.71% (1 2.99 ha), 93.33% (3 4.99 ha) and 90.32% (equal/more 5 ha), respectively In general, 359 producers are willing and 26 are hesitant for UDP adoption. Subsequently, the second question was asked with a higher or lower bid based on the negative or positive first answer (see Figure 6 28 ). From 359 positive responses in the first question, 321 farmers ratified their response at higher extra payment; only 38 cases rectified their decision facing a higher bid. While those farme rs who answered no to UDP in the first question, they also held their negative response in the second questi on, even with a lower bid ( see F igure 6 29 ). Also these farmers decided what extra payment for the briquette sack would be the best, according to t heir thought s Figure 6 30 shows that these producers might pay extra US $2.45 on average. The majority of values range in $1 and $3.50 normally at all land size category. However, observations with highest monetary WTP wer e found in less than 1 ha category Then, the above questions were repeated to these farmers but now introducing the environmental im pacts of UDP; capturing how important the environment is for these sampled farmers (EWTP) Once again, Figure 6 31 presents that high portion of farmers said yes to UDP, at all land size categories: 92.21% (less than 1 ha), 95.71% (1 2.99

PAGE 94

94 ha), 93.33% (3 4.99 ha) and 83.87% (eq ual/more than 5 ha). Only one farmer in the largest category changed his mind about WTP and become unwilling; the other categories maint ained similar responses like in the first round. What made this farmer to rectify his decision would be answered with the econometric model where I identify the most relevant factors affecting adoption decision. In general, 358 farmers said yes and 27 said no in the first question of the second round. Now, 323 out of those 358 positive responses ratified their interest in possibly adopting UDP; only 3 5 changed their decision ( see F igure 6 32 ). A point to be pointed out is that the positive responses of the high bid question augmented by two farmers compared to the first round; this result implies that environmental may have made these two individuals reconsider the negative response to the adoption of UDP even with higher extra payment. Going through the se cond qu estion with a lower bid ( see F igure 6 33 ), I found that only 1 out 27 people is willing to adopt at this lower extra payment; the rest is still reluctant to pay for Urea briquettes Meanwhile, in the open ended question of the second round, the mon etary EWTP has a similar distribution as that in the first round. Looking at Figure 6 34 one can see that EWTP ranges in $0 to $3.50. On average, a farmer would pay US$2.57 extra for a Urea briquette sack, 12 cents more than in the first round. On the o ther hand, the potential hectares that can be dedicated for UDP production are seen in Figure 6 35. O ne person was decided to use 6 hectares with this technology; while one, even presenting interest in UDP, would not give any hectare (he might try UDP on d onated land) On average, a farmer would utilize 0.80 ha for UDP

PAGE 95

95 production ; the data are concentrated on 0.04 an d 0.71 ha hypothetical UDP land sizes Now, the key variable is the intensity of adoption ; this variable is more informative because it does no t just present the willingness of adoption, but also a level of adoption in term of the percentage of total land given to UDP production. Figure 6 36 shows that those in the smallest landholders (less than 1 ha) would dedicate more land than those largest ones (equal/more than 5 hectares). Particularly, I identified a vast number of farmers would expect to use UDP on 100% of their land, 131 cases. This result i s based on the decision to censo r the data at 100% (all values greater than 100% would be converte d at 100%). Considering the uncensored IA, there were 22 farmers with the interest in utilizing more than their total land with UDP; valid answers because they can rent more land. On average, 49.70% of the total land may be dedicated to the UDP technology by these potential adopters. In conc lusion, this analysis allowed seeing a surprising h igh rate of adoption with this D ouble bounded format. However, there was not a significant change of UDP adoption rate when the enviro nmental impacts were introduced; W TP and EWTP presented the same adoption rate; when introducing environmental impacts 2 farmers were willing to pay extra with a higher bid and 1 farmers, who said no to the first question, said yes to pay extra for urea briquettes. The rest of the reluct ant farmers maintained their negative responses. In the end, one can assume that environment benefits might not be a significant factor when deciding the adoption of a certain tech nology. Additionally, the acceptance of UDP co uld also be seen through the I A, which shows some farmers willing to dedicate more than the current rice hectares for UDP production.

PAGE 96

96 Rice Production System Analysis One of the most important a nd time consuming section in the questionnaire was Rice Production System Section. Fortuna tely, most of the farmers were willing to dedicate a s ignificant time in order to collect valuable data such as: rice field, rice Urea foliar fertilizers, herbicides, etc.), hired labor, and yield, among others; this informati on corresponds to the last cultivation season (production of 2011 spring). Finally, for comparison reasons, I divided all variable related to production by the total rice hectares. I start out with the type of rice field of these rice producers. As is sh own in Figure 6 37 almost every farm er employed the Paddy field in the data, 98.44%; the rest, 1.56 %, implemented Poza field 5 Farmers were also asked for rice varieties being used. The characteristics of these varieties are differentiated by: yielding, vegetative cycle, and grain size, among others (see Figure 6 38) Taking a look at F igure 6 39 variety 11 is t he most utilized by the sample farmers, 288; varieties 14 and 15 were sown in 84 and 39 crops respectively. Finally, there are other varieties (e .g. 9, 10, 12, 16, 18, 21, Carvajal, Mancha Costa and Arroz el Pavo ) being employed by 30 farmers 6 In Figure 6 40 I have the total soil preparation cost per hectare given tha t a famer owned or did not own a tillage t racto r (there are small tractor that c an be afforded by farmers) I observe that those having tractor co nsiderably reduced this cost: these 5 Poza field are natural hollow which accrues water from rain. When the raining season is over, farmers cultivate rice in this land, only sowing on the top of this hollow at the beginning. As water goes down, farmers sow more rice seedlings. This traditional production is principally found in Los Rios Province, not visited by this study. 6 A confirmation that Carvajal, Mancha Costa and Arroz el Pavo are native ones or only commercial variet ies recognized with these n ames could not be reached.

PAGE 97

97 farmer s paid for this soil preparation US $35 less than producers without having tractor, on average. On the other hand, analyzing this cost per hectare b y land size categories, those farmers with the less than one hectare spent more on average than other landholders: US$218.03 (less than 1 ha); US$139.22 (1 2.99ha); US$115.81 (3 4.99ha); and US$136.60 (equal/more than 5 ha). In Figure 6 4 1 is presented se ed cost s (US$/ha). H ow the cost decreases while the land size increases is observed A gain the smallest landholders, on average, invested more than those largest ones on seed cost per hectare: US$89.60 (less than 1ha); US$52.10 (1 2.99 ha); US$53.80 (3 4.9 9 ha); and US$60.96 (equal/more than 5 ha). Acquisition of the m ain rice fertilizer, Urea is also analyzed in this part In Ecuador, there are different ways to obtain Urea sack (50 kg): government provision, market and others. Ecuadorian government is pr oviding subsidized Urea sacks for those small farmers, but they can also buy at the marketplace. In a ddition t o these alternatives, there is a black market (or other providers) where the Urea sack s are sold informally (such black market is composed of tho se farmers or other people who buy Urea sacks to resell them at higher prices; perhaps, these people buy and resell the subsidize d Urea ). There is a question asking farmer s for the Urea s acks bought in the last season. Displayed in Figure 6 42 is the total Urea (sacks/h a) bought by landholder groups. In t he smallest land size group, there is a farmer acquiring up to 100 sacks/ha. However, the other groups present a less wide range, between 2 up to 40 sacks/ha. On average, the group of less than 1 ha bought 12 sacks/ha, 1 2.99 ha and 3 4.99 ha purchased 6 sacks/ha, and equal/more 5 ha obtained 5 sacks/ha. This result may show

PAGE 98

98 that some farmers may be utilizing more than the necessary, keeping Urea sacks for future seasons or res elling them. In the first case, 4 Urea sacks are appropriate to fertilize one hectare. For instance, a producer said to have had 0 .04 ha and bought one Urea sack; if he used the entire sack, this represents 25 Urea sacks in a hectare utilizing 19 sacks in exceed. In the second case, Ur ea is not produced in Ecuador and this fertilizer is imported. As a result, a farmer might have gotten more Urea sacks to prevent production from possible shortages. The last case, based on th e fieldwork farmer informed that buy ing Urea sacks a nd resell t hem at higher prices is a common activity there A problem can occur when subsidized Urea is resold at a better price; if so, this subsidy would not be reaching its objective of reducing costs as low as possible. An example of these two latter cases is a f armer with 10 ha bought 100 sacks, presenting exceed of around 40 sacks. In detail, subsidized, marketplace and black market acquirers summed 174, 248 and 10 cases respectively. Similarly, data about Urea sack prices were also collected for these different markets. Figure 6 43 presents these prices. An interesting point is associated with the government price. As farmer were asked for Urea sack price of the last season (December, 2010 Marzo, 2011) The government prices that should have received as an answe rs are: US$10/sack or US$15/sack 7 However, demonstrate that the subsidized Urea sack price fluctuates between US$10/sack and US$29/sack (blue line). On average, subsidized price, market price and black market price were US$14.88/sack, U S$28.92/sack and US$24.11/sack. Different Urea markets 7 The price increased in January of 2011

PAGE 99

99 with several prices. Such results demonstrate how a farmer faces uncertainties not only in the end of the production with vari ation of rice prices but also in the beginning. Figure 6 44 shows costs of other fertilizers a hectare. Once again, the smallest landholder group presents observations spending more than those in the largest group. On average, one farmer in group less than 1 ha invested US$191/ha on these additional fertilizers, followed by 1 2.9 9 ha cluster with US$129/ha, 3 4.99 ha with US$127/ha and equal/more than 5 ha with only US$66/ha. Furthermore, herbicides/insecticides costs per hectare were displayed in Figure 6 45 Only four farmers mentioned not to have acquired these inputs in the last season. Most of the farmers spent less than US$200/ha in these inputs. Still, the smallest farmers spent, on average, more than the other others categories. Thus, less than 1 ha groups exceeds to the other groups in US$45/ha (1 2.99 ha), US$54 (3 4.99 ha) and U S$38/ha (equal/more than 5 ha). Day laborers are hired to mainly work on sowing and application of inputs (i.e. Urea other fertilizers, herbicide and pesticides) Therefore, i n Figure 6 46 are plotted day laborer cost per hectare by numbers of household members. As a first thought, one can assume that if there are more members in a family the cost of the hired day laborer should be lower. However, the aforementioned intuition is not seen clearly. Having 1 or 8 members in a household does not si gnificantly the mean hi red labor cost according to the sample. Actually, comparing the average hired labor cost of one member households (US$192/ha) with five member ones (US$194/ha), I found that the latter spent even more, US$2. In general, a farmer inve sted US$188/ha for this resource on average Finally, analyzing the mean cost by land size group, still those smallest

PAGE 100

100 farmers, less than 1ha, paid US$13/ha, US$14/ha and US$17/ha in exceed compared to 1 2.99 ha, 3 4.99 ha and equal/more 5 ha groups respe ctively Furthermore, water provision cost was also asked. There are two different sources of water provision: by the Agricultural Organization or/and by taking from the river or raining (there could be no cost or the cost comes from buying fuel to transf ers water from the river to the crop with water pump). To simplify, I added these source of water provision as unique cost per cultivation. Then this cost was divided by the total hectares. Thus, Figure 6 47 shows that the average cost of each group is as follows: less than 1 ha with US$75/ha, 1 2.99ha with US$41/ha, 3 4.99ha with US$36/ha and equal/more than 5ha with US$26/ha. In the harvest time, farmers have several costs. For example, they usu ally rent a combine from their agricultural o rganization, hi re day laborer to pack rice grains into the sacks, pay for transportation of these rice sacks, etc. How ever, some farmers do not rent any c ombine or hire labor and this is t he reason for what costs differ within farmers. To simplify the information gatheri n g, I estimated the total cost a hectare. As is seen in Figure 6 48 categories. However, on average, a farmer spent in the groups as follows: US$368 (less than 1 ha), US$282 (1 2.9 9 ha), U$354 (3 4.99 ha) and US$280 (equal/more than 5 ha). In Figure 6 49 is presented the rice yield (kg/ha) by land size group. The range of this variable is between 1,000 kg/ha t o 19,000 k g/ha. Surprisingly, a farmer go t 743.89 kg (8 sack of 205 lb.), having 0.04 ha which means 18,597.28 kg/ha. This is why there are some observations with extreme values, but those are, in fact, real values. On

PAGE 101

101 average, a farmer could get 5,892.15 kg/ha (the national mean is 4360 kg/ha). Taking a look at land size group s, the mean yields are as follows: 6,427.95 kg/ha (less than 1 ha); 5,590.57 kg/ha (1 2.99 ha); 5,475.65 kg/ha (3 4.99 ha) and ; 5,398.51 kg/ha (equal/more than 5 ha). A surprising fact was found those smallest farmers have better rice yield per hectare tha n the largest landholders. Once harvested and know the amount of rice sacks yielded, farmers decided how many sacks would be for own consumption and for sales. Figure 6 50 presents the percentage of rice sacks sold by land size category. As is seen, much o f the production is sold, 87.73%, on average. Few farmers just produced for own consumption 5 in total Furthermore, those in 3 4.99 ha and equal/more than 5 ha groups may be categorized as complete rice sellers because the majority of their production wa s put into the marketplace. On average, the percentages of rice sacks sold by land size category are as follows: 81% (less than 1 ha), 91% (1 2.99 ha) and 94% (3 4.99 ha and equal/more than 5 ha). Incomes per hectare are shown in Figure 6 51 On average, a farmer got US $1556.60/ha. In contrast, three farmers did not gain any income because they produce for own consumption. The mean incomes per hectare by land size group were: US$1648/ha (less than 1 ha), US$1508 (1 2.99 ha), US$1539/ha (3 4.99 ha) and US$13 54 (equal/more than 5 ha). On the other hand, calculating the rice net income is a special task because these production costs may not only be representing to the last season but also future ones (e.g. Urea sacks bought for futures fertilizations or resale ). As suming that those costs were incurred only for the last rice production, a farmer would make, on average, US$305.51/ha. However, an important portion of farmers

PAGE 102

102 would make a negative or zero rice net income, 118 cases. Comparing the land size categori es, I got as mean net rice income: US$71/ha (less than 1 ha), US$464 (1 2.99 ha), US$480 (3 4.99 ha) and U S$413/ha (equal/more than 5 ha) The smallest landholders made the lowest net income per hectare because of their higher costs. Credit and Insurance M arket Analysis A credit section was also prepared in the survey instrument in order to know about the credit access of these sampled farmers. The collected information attempts to identify those credits solicitants, moneylenders, sum of money, etc. The ana lysis of these data is l imited to the last year (2010 11). The question to observe credit market participation of these farmers was: have you ever asked any credit during the last year? Figure 6 52 shows that a 47% (181) of thes e farmers applied for a cred it. However, the majority did not ask for a credit, 52.88% (204). In the end, those credit solicitants actually obtained the credit without any problem. The last result may demonstrate that there is credit availability over the rice zones visited. Addition ally, farmers mentioned to those institutions, people or others that provided credits in their zone. As shown in Figure 6 53 the most mentioned provider is hulquero is a person offering loans usually at usurious interest rate. This is an illegal practice that is very common not just in the rural areas, but also in big cities. However, many farmers opt for credit s from these provider s because there are no requirements to get the credit; just to pay off the loan at a higher price. Other mentioned sources of credit were banks (46), Rice Mill (31), Cooperative (12), Friends/relatives (10) and others (1).

PAGE 103

103 Farmers were also asked for the amount of money they required as a credit. Taking a look at Fi gure 6 54 one can see that being a larger landholder requested more money. For instance, a farmer with equal/more than 5 ha asked, on average, US$3964.706. Meanwhile, those have less than 1 ha solicited US$700.625, on average. This is a logic result given that the majority of the credit was spent on inputs. Figure 6 55 demonstrates that most of the farmers (133) utilized the money borrowed on inputs such as seeds, fertilizers, herbicides, etc. In addition, 130 rice producers employed the credit on machine and equipment for the production; machine stands for the tillage tractor that some farmers must rent or sometimes acquire to prepare the soil before so wing. Credit was also used for f amily expenses such as health and education and the payment of previous debts, 79 farmers. Other items on what the credit was spent are: vehicle parts (54), house arrangements (14) and other such as entertainment and household consumption (8). Finally, there was also question identifying if these farmers were crop in surance holders in these zones In F igure 6 56 is seen that only a 4.67% (18 observations) of the sample was insurance holders. Such results can be justified by Mara Jos Castillo ( 2011 ) about the i nsurance market s ituation in Ecuador: lack of insurance culture and the complexities of insurance agricultural market (i.e. risk coverage, high prices, etc.). According to MAGAP there were a total of 3,885 insurance policies in 2011 (January August) Ecuadorian government is subsidizing a 60% of the crop insurance cost mainly for rice, wheat, maize and potato 8 None of these insurance holders mentioned to have a subsidized insurance. 8 See Ministerio de Agricultura, Ganaderia, Acuacultura y Pesca 2011.

PAGE 104

104 Time Availability and Non Work Income Analysis Another interest of this thesis was to gat her information about the family workforce over these households and the main source of income. Consequently, this section is to analyze labor data such as main occupation and the hours spe nt on on farm and off farm work, and education T his information is available for the father, mother and the fir st son/daughter of a household. Additionally, a question about non work income (e.g. donation or government monetary transfers) was introduced. Starting with the main sources of money, Figure 6 57 demonstrates that the main occupation of these farmers is the on farm activities (or rice production). For instance, 358 out of 375 fathers work in rice production and 15 in off farm activities. Meanwhile, 5 fathers do not generate any income; these fathers were mentio ned as household heads but their sons/daughters were in charge with the rice production. On the other hand, mothers are also participating in farming and in other activities. In total, 20 mothers were related to the rice family business directly 38 in off farm work and 304 in housework. Additionally, main occupation of a son/daughter was also collected. Thus, 126 son/daughters were included in the rice activities; but there were 36 sons/daughters working outside of the rice production. Finally, some of t hese sons/daughters were not making any income, 195. Similarly, T able 6 4 presents hour availability of a household in a day. On average, a household dedicated 6.05 hrs/day to the rice business; o f f farm jobs demanded 1.75 hrs/day ; and those studying utili zed 1.99 hrs/day in this activity. Taking into consideration to only those with off farm and on farm work, the mean hours spent on these activities are 6.65 hrs/day and 6.76 hrs/day respectively. Two points should be highlighted here: 1) there are job alte rnatives which can be produced income that could

PAGE 105

105 function as extra capital for family rice production; and 2) this in significant difference between hours of off farm and on farm activities may tell that job alternatives could be being becoming in more prof itable over time 9 However, if farmers are changing their work on these substitute activities, such a fact may be a threat for the rice prod uction in the Ecuadorian Coasts in the long time. Further research may be clarified this fact. Figure 6 58 shows th at a big part of these rice farmers is receiving non worked income, 60% (154 farmers). The unique source of a non worked mentioned by these farmers was the government transfers. In Ecuador Human Development Bonus (US$35 monthly trans fers), the Solidarity Credit and Human Development Credit ( credits up to US$840 for two years) are p rograms mainly for poor people. The last two programs are directed for those who want to develop a business (e.g. rice production). The final objective is to allow these poor to alleviate poverty. To simply the analysis, all these transfers and credits were converted to monthly. Figure 6 59 shows thes e monthly transfers. On average, a farmer received US$34.95 10 Econometric Results At this moment, a descriptive picture of the samp examined with a descriptive analysis. Now, results of the dependent analysis are presented in this section. The econometric model that better fits UDP adoption was a Tobit. Intensity of Adoption is a right and left censore d variable through which the research question will be answer ed : Who are the potential adopters of UDP? 9 O ff education, c occupations 10 Those cred its are paid off with the same Human Development B onus, so, farmers would not receive any transfers until cancel all the debts. These credits assumed to be transfers in advance.

PAGE 106

106 Determinants of adoption were established based on the extensive literature consulted in this the sis. Therefore, this section is mana ged in the followin g order: 1) e xplanation of the independent variable; 2) Tobit model outcomes an d the marginal effects; and 3) P ost estimation analysis. Descriptive Summary of the Variables T able 6 5 prov ides a statistic summary of the data that is employed to estimate th e IA model : mean, standard deviation, minimum and maximum values. Each variable was labeled with the real name to have a better understanding. The Tobit model is finally composed of variables related to subjects such as farm and farm er s characteristics, production var iables, credit participation, agricultural insurance and risk aversion, UDP diffusion, income sources and social network. Starting with the dependent variable, it is a measure that indicates the percent of the hectares that will be given for UDP production ( D. Joshua Qualls et al. 2012 ). This variable does not just provide a far the level of the adoption of such technology. The values are in percentages. Referring to f arm ers used to explain adoption decision ( Cheryl R. Doss and Michael L. Morris 2000 ; Rajni Jain, Alka Arora a nd S.S. Raju 2009 ; Conor Keelan et al. 2009 ; D. Joshua Qualls et al. 2012 ). Gender is a binary variable taking values of 1 for male and 0 for female. The age of a household is represented by the variable Age which is in years. Also, there are two types of education variable, one containing the years of general education (continuous variable) and a binary variable specifying if a farmer attained any agricultural education (i.e. fertilizer, harvest, tillage and other techniques).

PAGE 107

107 Khanna and Madhu 2001 ; Gunnar Breustedt, Jrg Mller Sche eel and Uwe Latacz Lohmann 2008 ). As a result, land size, number of small kids, wealth index, drought or/and flood affections, and distance to the main town were included. Particular interest is focused on the land size because the UDP technology seems t o better work in small farms. On the other hand, distance to the main town may be seen as a market access variable; some household are really far away to market which is placed in the downtown of Daule and Santa Lucia. Similarly, variables related to the p roduction system were also included in the model: binary rented land variable; per hectare subsidized Urea sacks (50 kg) and non subsidized Urea sacks (50 kg); labor cost per hectare (it contains costs of sowing and fertilizer/herbicide/insecticide applica tion); total other cost per hectare (it includes costs of tillage, seeds, herbicides/insecticides, non urea fertilizers, irrigation and harvest); and yield variable ( sacks of 205 lb /ha). Moreover, Urea sacks and labor cost require special attention given the UDP char acteristics: reduction of Urea sacks applied and increase of labor 11 Also, past adoption variables was taken into consideration in this analysis. This binary variable refers if a farmer has introduced a technique, idea or technology into his r ice production system (yes is 1 and no is 0). In this sense, one could see how likely a farmers was in the past to adopt a technology. The effect of this factor must be positive as farmers who adopted agricultural innovations before would be more willing t o utilize new ones, like UDP, in the present. 11 The consideration of urea sacks are discussed in the Rice Production System Analysis in Chapter 6.

PAGE 108

108 Credit participation, Insurance solicitation and risk aversion were also explanatory factors taken into account ( Gershon Feder and Dina L. Umali 1993 ; Franklin Simtowe and Manfred Zeller 2007 ; Conor Keelan et al. 2009 ; Mara Jos Castillo 2011 ). Credit and insurance variables are binary response. Meanwhile, the ris k aversion is measured through the following statement: I will use a new Urea fertilization method without caring if others used it firstly; the Likert scale is 1=disagree, 2=Indifferent and 3=agree. Other determinants of adoption are rel ated to: hours spent on on farm and off farm works (hours on education is considered as well) and if a farmer obtained non work income ( Cheryl R. Doss 2006 ; Haluk Gedikog lu and Laura M.J. McCann 2007 ). Total on farm hours stand for the hours dedicated to the own rice activity by a household. In contrast, total off farm hours represent the time used in other works by a household (e.g. daily laborer, teacher, janitor, etc.) Finally, there is the binary variable non work income which 1 if a farmer received such income and 0 otherwise. The last variables are linked to the peers effects or social network in general ( A. D. Foster and Rose nzweig 1995 ; Conley, Timothy G., and Christopher R. Udry 2010 ; Oriana Bandiera and Imran Rasul 2006 ). In order to avoid the endogenous probl em of these variables, I followed Oriana Bandiera and Imran Rasul ( 2006 ) Therefore, three variables were designed: numbers of people in their religious group, number of people in the extended family group and nu mber of people in the agricultural/other groups. In this sense, one can know by intuition the social network size of a farmer. Thus, 27 regressands were placed in the model, trying to take into considerations the widely utilized variables in different tec hnology adoption researches and also, attempting to maintain the parsimony of the model.

PAGE 109

109 Tobit Estimation The final Tobit model was grounded by the evaluation of measurements of goodness of fit 12 Also, Intensity of Adoption was censored at 0 (19 observati ons) and 100% (131 observations). On the other hand, regarding the Heteroscedasticity, there are data with values that behave very different to the rest of other observations (outliers) but they were left because of the total logical sense with reality. Mo reover, this study is dealing with microeconomic data. Those two aforesaid facts may be sources of Heteroscedasticity ( Damodar N. Gujarati and Dawn C. Porter 2009 ). In ord er to obtain robust estimators, the v ariant estimation of Bootstrapped Tobit Model was applied ( A. Colin Cameron and Pravin K. Trivedi 2010 ) Also, the assumption of well specified model is held. Fin ally, in the Post estimation analysis is discuss ed the possible Multicollinearity problem and the analysis of the R esiduals. Table 6 6 contains the outcomes of the Tobit estimation. Utilizing a Maximum Likelihood a pproach, a total of 11 significant variables were found: rented/owned land size (ha), numb ers of small kids in a household, time to get the main town, binary rented land variable, subsidized Urea sacks per hectare binary credit solicitation, binary agricultural insurance, Likert scaled risk aversion, binary UDP knowledge, on farm hours/ha and social networking variable number of people in the family group. These variables were significantly important at confidence levels of 1%, 5% and 10%, according to the t test. As additional information, the marginal effects on the continuous part of IA Tobi t model are also di splayed in T able 6 6 T hese significant regressands 12 Several Tobit models were es timated which did not differ significantly in any of measurements of goodness of fit and producing stability in estimators (similarity in signs and magnitude).

PAGE 110

110 have the hypothesized effect (cor rect sign) on adoption decision based on the literature review. Urea Deep Placement would better function in small parcels because the cost of placing the Urea briquette in soil can be absorbed by the family labor without hiring people ( Thomas P. Thompson and Joaquin Sanabria 2009 ) On the other hand, large farmers would have to hire more people given the ha rder work to apply the briquettes. As a result, the total rented/owned operated land (ha) coefficient is negative: the smaller cropland, the higher intention to adopt UDP. In order words, if a farmer acquires one more hectare, the mean value of IA would de cr ease in 1.74%, ceteris paribus. In this study, all these farmers are considered small (less than 20 hectares). A lso, a possible endogenous problem of farm size can be faced because land size can be related to wealth or capacity to be ar risks, and so on ( see D. Joshua Qualls et al. 2012 ). However, as these farmers present similar characteristi cs (e.g. they are holding small farm s, facing the same credit market, hving slightly different level of wealth, etc.), no endogeneity is assumed in this thesis Regarding the children in a household, this could restrain the experimentation/adoption of an innovation given that farmers must dedicate enough time to the caring of their children. Gunnar Breustedt, Jrg Mller Scheeel and Uwe Latacz Lohmann ( 2008 ) corroborated this result in the adoption of Oilseed Rape where the presence of small kids may influence such adoption negatively. Following this intuition, this variable produ ces the same effect in this Tobit model signifying a reduction of 1.30 % of the expected average hectares dedicated to UDP when a child is added to the household, holding other factors constant. Not just mothers have to take

PAGE 111

111 care of their children but also the fathers. Perhaps, further research, having a comparable sample size between female and male farmers, may conclude that females would be less likely to adopt because they are more related to the housework and kids caring than males. Unsurprisingly, t he time to get to the main town seen as the market access measure is negatively associated with the Adoption of UDP. This coefficient asserts that being close to the market (or less time needed to get the main town) so that reduction of transaction cost o f reaching agricultur al services, input suppliers trading house or in this case, Urea briquettes increases the probability of UDP adoption ( Madhu Khanna 2001 ; S. J. Staal et al. 2002 ) According to the marginal effect of the market access, on average, if a farmer increases one minute to get the main town, he would dedicate 0.18% less of their total land to the UDP production, holding other things constant. According to the Literature Review, it is difficult to clearly know what effects land tenure (owned or rented) could have on adoption ( Gershon Feder and Dina L. Umali 1993 ; Conor Keelan et al. 2009 ; D. Joshua Qualls et al. 2012 ). In this Tobit model, if a farmer rented his cropland, he is more e nthusiastic for UDP innovation. Thus, a farmer will on average put into UDP production 5.03% more of his rented/owne d land when he is a renter, ceteris paribus. As a first ideation, more Urea sacks applied per hectare would imply a higher willingness to adopt UDP because of Urea reduction benefit. However, the subsidized Urea sacks in a hectare yielded a negative affect ion on Adoption: one more subsidized Urea sack acquired reduces the expected hectares for UDP in about 0.37%, ceteris

PAGE 112

112 paribus. As said earlier, this variable encloses different effects in reality: the subsidy, resale and future employment. Analyzing th e fi nal result carefully, this study assumes t hat a peasant purchasing subsidized Urea may perceive UDP as a threat: they are bearing a relatively low Urea cost with the subsidy (or profiting with the resale at higher prices) and the coming of UDP technology m ay make government reconsider the subsidy as farmers would utilize Urea more efficiently. As practical exercise, more than the 50% of those subsidized farmers, who bought more than the required amount of Urea sacks, were willing to give less than 50% of th eir hectares to UDP technology. Credit solicitation would impact positive ly on UDP adoption. The intuition of this result may be that cash resource is available to invest on new and improved ideas. These sources of cash let farmers have the capacity to fac e the costs of UDP or any innovation inside their production system (e.g. more labor). Thus, they would be the first that will take advantages of the benefits of UDP (e.g. yield and income increase and Urea reduction ). According to the estimated Intensity of Adoption model, farmers that asked for a credit would probably give, on average, 3.04% more of their total hectares for UDP production ceteris paribus. Additionally, having rice insurance may positively explain the occ urrence of UDP in these surveyed f arms. Being covered from different risks (e.g. drought, flood, uncontrolled plagues, etc.) ma kes a farmer allocate 8.37 % of his total land in UDP production, ceteris paribus. Also, the risk aversion gauged by a statement exhibited a positive association with the adoption behavior. This means that those more risky farmers may on average

PAGE 113

113 increase the UDP hectare by 2.79 %. These types of farmers can also be properly Possessing information of UDP would intensify the adoption of UDP. Although these farmers affirmed to need more knowledge of UDP, some of them could observe the encouraging outcomes of UDP experiments in Daule which stimulates its adoption in the future. The marginal effect is of 5.17 % on IA. UDP technique requires labor for its development. Therefore, one solution is to hire more workers. However, the ideal situation is to have more household members or in its defect more time available for on farm work. Consequently, those households with more on farm hours used in a he ctare are more likely to insert UDP method in their production; an additional hour spent on on farm work may make farmers add 0.23% of their total hectares to UDP production, ceteris paribus. These households have enough family workforces to avoid the main Finally, the social network effect was positive on I ntensity of Adoption. According to the results, endogenous problem would be avoided given the insignificancy of the number of people on agricultural/other group In the end, this coefficient states that a farmer having a numerous social network with potential UDP adopters may dedicate 0.46% more of his total rice land for UDP production holding the rest constant. Post Estimation Analysis There some issues that t he any estimation can arise. However, those affecting this econometric analysis are : multicollinearity, Heteroscedasticity and Normality. Through the section, they will be explained and tested. The methods to analyze these problems are through tests, graph s and indexes.

PAGE 114

114 Multicollinearity may always be present in all models even more when 26 regressands have been incorporated in this Tobit setting. In the presence of collinear variables, error standards and variances tend to be infinite and estimators are n ot converging to a unique solution (see Damodar N. Gujarati and Dawn C. Porter 2009 ). In William H. Greene ( 2012 ) are pointed out some of the consequences of the collinearity: enormous fluctuation in estimators with small data changes; jointly significance of the estimators and elevated R square but high standard errors and insignificance t ratios of these estimators; and mistaken signs and magnitudes of the estima tors. None of these symptoms are observed in the T obit model. However, using the tolerance i ndex and th e variance inflation factor one is able to see to what extent is found such problem ( Damodar N. Gujarati and Dawn C. Porter 2009 ) Tolerance index and the VIF are reciprocal. T herefore, the interpretation of the variance inflation factor is that the higher the value, the more eminent the collin earity; meanwhile, for t olerance is the opposite. As is shown in T able 6 7 the vari ance inflation factor of each 27 variables is not greater than 10 and the t olerance is not so close to zero. As a conclusion, one can make the assumption of relatively low presence of multicollinearity in the Tobit model proposed in this thesis. In examining the R esiduals, Figure 6 60 demonstrates that they have v olatile behaviors. As said above this econometric examination is facing with some atypical observations and microeconomic data which are important Heteroscedasticity sources. In the end, I may conclude that the model is suffering this problem. However, bootst rapping estimation would let estimators be robust ( A. Colin Cameron and Pravin K. Trivedi 2010 ).

PAGE 115

115 On the other hand, once I compu ted the R esiduals of the Tobit model, Jarque Bera test was used to analysis the Normality Distribution of the model. The null hypothesis under this test is Normality. Figure 6 61 shows R slightly looks as a Normal. However, the t est produc es a p value of 0.05% telling that Normality is not reached by the R The purpose is to circumvent any violation of the regression assumption. As a result, I went over these tests and used the b ootstrapping method. A more p rofound analysis is left for further research.

PAGE 116

116 Figure 6 1. (Source: Author) Figure 6 2. (Source: Author)

PAGE 117

117 Figure 6 3. Education (Source: Author) Figure 6 4 Agricultural education (Source: Author)

PAGE 118

118 Figure 6 5. Type s of agricultural (Source: Author) Figure 6 6 Agricultural education providers (Source: Author)

PAGE 119

119 Figure 6 7. Land size g roups (Source: Author) Figure 6 8. Rented land (Source: Author)

PAGE 120

120 Figure 6 9. Total expenses by land size group s (Source: Author) Table 6 1. Factors matrix (Factor Analysis) Factor analysis/correlation Number of obs = 385 Method: principal component Retained factors = 1 Rotation: (unrotated) Number of params = 17 Factor Eigenvalue Difference Proportion Cumulative Factor1 2.1171 0 .155 0.1245 0.1245 Factor2 1.9621 0.2662 0.1154 0.24 Factor3 1.6959 0.5262 0.0998 0.3397 Factor4 1.1697 0.015 0.0688 0.4085 Factor5 1.1548 0.0292 0.0679 0.4764 Factor6 1.1256 0.0359 0.0662 0.5427 Factor7 1.0896 0.062 0.0641 0.6068 LR test: independe nt vs. saturated: chi2(136) = 1.4e+04 Prob>chi2 = 0.0000 Source: Author

PAGE 121

121 Table 6 2. Coefficient matrix (Factor Analysis) Scoring coefficients (method = regression) Variable Factor Tractors and other major equipment 0.23339 Storeroom 0.25414 Irriga tion Pump 0.18732 Harvester 0.2138 Fumigation machine 0.29288 Car 0.27204 Electric Plant 0.10194 Drinkable water service 0.06549 Electricity Service 0.02403 Phone Service 0.23968 House made up of cement 0.05945 House made up of wood 0.0854 House made up of cane 0.13946 Own household 0.0878 Rented h ousehold 0.03919 Borrowed household 0.03865 Shared household 0 Source: Author Figure 6 10 Wealth index (Source: Author)

PAGE 122

122 Figure 6 11 Wealth level (Source: Author) Figure 6 12 Drought and flood affectation (Source: Author)

PAGE 123

123 Figure 6 13. Time to get the main town by spent time groups (Market Access) (Source: Author) Figure 6 14. UDP knowledge (Source: Author)

PAGE 124

124 Table 6 3. Statistics of those w ho knew and did not know about UDP Variable Who knew UDP Who did not know UDP Age (years) 53.88 51.28 Education (majority) Elementary School Elementary School Village (majority) El Mate/El Encanto Peninsula de Animas Agricultural Group Affiliation (majority) Yes Yes Land Size Groups (majority) Less than 1 ha Less than 1 ha Source: Author Figure 6 15. UDP knowledge sources (Source: Author)

PAGE 125

125 Figure 6 16. Observable UDP results (Source: Author) Figure 6 17. UDP knowledge level (Sou rce: Author)

PAGE 126

126 Figure 6 18 Agricultural group affiliation (Source: Author) Figure 6 19. (Source: Author)

PAGE 127

127 Figure 6 20. Meeting frequency (Source: Author) Figure 6 21. Voluntary attendance (Source: Author)

PAGE 128

128 Figure 6 22. F (Source: Author) Figure 6 23. Influential groups (Source: Author)

PAGE 129

129 Figure 6 24. Commu nication level (Source: Author) Figure 6 25. Past technology adoption (Source: Author)

PAGE 130

130 Figure 6 26. Adopted innovations (Source: Author) Figure 6 2 7 WTP: first question with initial bid by land size groups (Source: Author)

PAGE 131

131 Figure 6 28. WTP: second question with higher bid (Source: Author) Figure 6 29 WTP: second question wi th lower bid (Source: Author)

PAGE 132

132 Figure 6 30 WTP (US$) by land size groups (Source: Author) Figure 6 31. EWTP: first question with initial bid by land size group s (Source: Author)

PAGE 133

133 Figure 6 32. EWT: second question with higher bid (Source: Author) Fi gure 6 33. EWTP: second question with lower bid (Source: Author)

PAGE 134

134 Figure 6 34. EWTP (US$) by land size groups (Source: Author) Figure 6 35. UDP potential area (ha) (Source: Author)

PAGE 135

135 Figure 6 36. Intensity of Adoption by land size groups (%) (Source: Au thor) Figure 6 37. Rice field (Source: Author)

PAGE 136

136 Figure 6 38. Description of rice varieties (in Spanish) (Source: In stituto Nacional Autonomo de In ve s tigacion Agropecuaria) Figure 6 39. Rice variety (Source: Author)

PAGE 137

137 Figure 6 40 Soil preparation cost (US$/ha) by tillage tractor (Source: Author) Figure 6 41. Total seed costs (US$/ha) plotted with land size (ha) (Source: Author)

PAGE 138

138 Figure 6 42. Urea (50 kg sacks/ha) by land size groups (Source: Author) Figure 6 43. Urea pr ices (US$/sack) of subsidized, real and black markets (Source: Author)

PAGE 139

139 Figure 6 44. Total cost of other fertilizers (US$/ha) by land size groups (Source: Author) Figure 6 45. Total cost of herbicides/pesticides (US$/ha) by land size groups (Source: Aut hor)

PAGE 140

140 Figure 6 46. Total cost of hired labor (Source: Author) Figur e 6 47. Total irrigation cost (US$/ha) by land size groups (Source: Author)

PAGE 141

141 Figure 6 48. Total harvest cost (US$) by land size groups (Source: Author) Figure 6 49. Rice yield (kg/ha) by land size group s (Source: Author)

PAGE 142

142 Figure 6 50. Rice sack sold (%) by land size groups (Source: Author) Figure 6 51. Total income (US$/ha) by land size groups (Source: Author)

PAGE 143

143 Figure 6 52. Credit Solicitation (Source: Author) Figure 6 53. Credit providers (Source: Author)

PAGE 144

144 Figure 6 54. Credit (US$) by land size groups (Source: Author) Figure 6 55. Uses of the credit (Source: Author)

PAGE 145

145 Figure 6 56. Rice insurance (Source: Author) Figure 6 57. Main o ccupation (Source: Author)

PAGE 146

146 Table 6 4 Time a vailability (hrs/day) Variable Obs Mean Std. Err. Min Max On farm 385 6.05 2.45 0.50 18.00 Off farm 385 1.75 3.12 0.00 14.00 Education 385 1.99 3.05 0.00 14.00 Sou rce: Author Figure 6 58. Non work income (Source: Author)

PAGE 147

147 Figure 6 59. Human Development Bonus (US$) (Source: Author)

PAGE 148

148 Table 6 5 Descriptive summary of dependent and independent variables Variables Mean Std. Dev. Min Max Intensity of A doption (%) 49.70 40.69 0 100 Gender (1=male/0=female) 0.92 0.27 0 1 Age (years) 51.52 14.26 18 95 Education (years) 5.89 3.49 0 17 Agricultural education (1=yes/0= no) 0.2 0.4 0 1 Total operated land size (Ha) 2.01 2.18 0.04 16.33 Number of small ki ds in a household 1.3 1.39 0 8 Wealth index 0 1 1.46 8.86 Flood or/and Drought affection (1=yes/0= no) 0.23 0.42 0 1 Tim e to get the main town (minutes ) 23.5 15.34 1 120 Rented land (1=yes/0= no) 0.14 0.35 0 1 Past adoption (1=yes/0=no) 0 .1 8 0 .38 0 1 Subsidized Urea sacks (50 kg /ha) 3.19 4.97 0 42.25 Non s ubsidized Urea sacks (50 kg/ha) 7.14 30.94 0 583.33 Labor c ost (US$)/ha 188.06 43.73 0 395.77 Total other cost (US$/ha) 845.48 468.57 215.21 2674.33 Total yield (sacks 205 lb / ha) 63.37 21.51 14.08 200 Credit s olicitation (1=yes/0= no) 0.47 0.5 0 1 Agricultural i nsurance (1=yes/0= no) 0.05 0.21 0 1 Risk a version 2.8 0.52 1 3 UDP k nowledge (1=yes/0= no) 0.09 0.29 0 1 Total o n farm hours/Ha 8.15 11.69 0.36 100 Total o ff farm hours 1.75 3.1 2 0 14 Total e ducation hours 1.99 3.05 0 14 Non worked income (1=yes/0= no) 0.4 0.49 0 1 # of people of a religious group 4.81 43.63 0 650 # of people of a family group 2.11 4.29 0 65 # of people of an agricultural group/other group 106.08 296.84 0 3000 Source: Author

PAGE 149

149 Table 6 6 Tobit model estimation of Intensity of Adoption Regressor s Observed Coef. Boostrap Std. Err Z value Marginal Effects Gender (1=male/0=female) 1.08 14.88 0.07 0.28 Age (years) 0.14 0.30 0.46 0.04 Education (ye ars) 0.51 1.17 0.43 0.13 Agricultural education (1=yes/0= no) 1.78 8.50 0.21 0.46 Total operated land size (Ha) 6.77 1.48 4.58 1.74 Wealth index 3.77 8.06 0.47 0.97 Number of small kids in a household 5.05 ** 2.39 2.11 1.30 Flood or/and D rought affection (1=yes/0= no) 3.42 3.25 1.05 0.88 Time to get the main town (minutes) 0.71 0.19 3.69 0.18 Rented land (1=yes/0= no) 19.72 ** 7.99 2.47 5.02 Past adoption (1=yes/0=no) 1.24 8.80 0.14 0.32 Subsidized Urea sacks (50 kg/ha) 1.43 ** 0.59 2.43 0.37 Non subsidized Urea sacks (50 K g/ha) 0.13 0.36 0.37 0.03 Labor cost (US$)/ha 0.04 0.08 0.51 0.01 Total other cost (US$/ha) 0.01 0.01 1.12 0.003 Total yield (sacks 205 lb/ha) 0.09 0.17 0.52 0.02 Credit solicitation (1=yes/0= no) 11.86 ** 5.89 2.01 3.05 Agricultural insurance (1=yes/0= no) 33.55 *** 18.18 1.85 8.37 Risk aversion 10.85 *** 5.60 1.94 2.79 UDP knowledge (1=yes/0= no) 20.35 *** 12.12 1.68 5.17 Total on farm hours/Ha 0.89 *** 0.48 1.86 0.23 Total off farm hours 1.56 0.9 8 1.58 0.40 Total education hours 1.38 1.00 1.37 0.35 Non worked income (1=yes/0= no) 7.87 6.79 1.16 2.02 # of people of a religious group 0.001 0.77 0.00 0.0003 # of people of a family group 1.77 *** 0.97 1.83 0.46 # of people of an agricultural group /other group 0.0032 0.01 0.23 0.001 Constant 40.70 32.16 1.27 Variance of the error term 53.47 2.75 Log likelihood 1397. 80 Akaike's Information Criterion 7.41 Pseudo R2 0.035 Obs. Summary: 19 left censored observations at SIA<=0 235 unc ensored observations 131 right censored observations at SIA>=100 Significant at *1%, **5% and ***10%. (Source: Author)

PAGE 150

150 Table 6 7 Collinearity e valuation of the independent variables VIF Tolerance R Squared Gender (1=male/0=female) 1.04 0.92 0.08 Age (years) 1.17 0.73 0.27 Education (years) 1.17 0.73 0.27 Agricultural education (1=yes/0= no) 1.04 0.92 0.08 Total rented/owned operated land size (Ha) 1.19 0.71 0.29 Number of small kids in a household 1.09 0.84 0.16 Wealth index 1.08 0.85 0.15 Flood or/and Drought affection (1=yes/0= no) 1.11 0.82 0.18 Time to get the main town (minutes ) 1.03 0.95 0.05 Rented land (1=yes/0= no) 1.04 0.92 0.08 Past adoption (1=yes/0=no) 1.08 0.85 0.15 Subsidized Urea sacks (50 kg/ha) 1.13 0.79 0.21 Non su bsidized Urea sacks (50 kg/ha) 1.09 0.84 0.16 Labor cost (US$)/ha 1.09 0.84 0.16 Total other cost (US$/ha) 1.2 0 0.69 0.31 Total yield (sacks 205 lb/ha) 1.14 0.77 0.23 Credit solicitation (1=yes/0= no) 1.12 0.80 0.20 Agricultural insurance (1=yes/0= no ) 1.07 0.87 0.13 Risk aversion 1.02 0.96 0.04 UDP knowledge (1=yes/0= no) 1.08 0.86 0.14 Total on farm hours/Ha 1.33 0.56 0.44 Total off farm hours 1.06 0.88 0.12 Total education hours 1.07 0.87 0.13 Non worked income (1=yes/0= no) 1.07 0.87 0.13 # of people of a religious group 1.05 0.91 0.09 # of people of a family group 1.01 0.97 0.03 # of people of an agricultural group/other group 1.13 0.78 0.22 Source: Author

PAGE 151

151 Figure 6 60 Intensity of Adoption R esiduals (Source: Author) Figure 6 61. Jarque Bera Normality test of R esiduals (Source: Author)

PAGE 152

152 CHAPTER 7 CONCLUSION REMARKS A ND POLICY IMPLICATIONS A n ex ante analysis was carried out to ascertain which factors may affect the adoption process of Urea Deep Placement in the Ecuadorian Coast. This study only takes into consideration s mall rice fa rmers in Daule and Santa Lucia c antons. The use of primary data, which were collected through a survey instrument, made possible the development of the descriptive and econometric analyses. Such analys es allow respond ing the research question of this thesis. Consequently, I could obtain in teresting findings that are summarized in this sect ion. Such findings would let to better comprehend the decision making of rice farmers (and perhaps, farmers in gene ral) on the adoption of new technologies. Moreover, I provide recommendations for a suitable introduction of UDP. As a relevant observation, I noted that farmers were principally males (92.2%), but behind these males there are wives also contributing with the rice activities directly or indirectly. Some wives have the responsibility to purchase inputs, interacting with input dealers and learning about rice business; a participation more intellectual tha n physical on rice production. I observed that few fa rmers mentioned to have adopted an innovation/technology/technique, 18%. Additionally, only a 20% of the sample farmers declared to have received agricultural instruction. This result would imply that farmers need to be updated about new knowledge and new improved technology pr esent in agriculture currently. A further research is suggested on technology adoption and agr icultural education over time.

PAGE 153

153 In spite of the fact that the smallest farmers obtained a better yield (kg/ha), their production costs were the highest. Some of these farmers presented a negative net income. There is a need of innovations that make rice business more profitable. For instance, UDP would let these farmers not just increase their yield (income increase per se) but also reduce the ir cost through Urea s aving and weeding cost reduction; when the labor requirement is supplied by family labor (remaining time not used by these farmers). Also, UDP requires more labor that may augment cost s. But, the labor required might b e satisfied by f amily labor especially in small farm, without incurring in any cost ( Thomas P. Thompson and Joaquin Sanabria 2009 ). However, the opportunity cost of working in rice production must be analyzed. As seen before some households are using considerable amount of hours on o ff farm activities; being this c ost high if there exits more profitable jobs. Since farmers in these two rice producing zones have rice production as main activity (there are not many jobs being offered in these zones and Guayaquil city is the best place with more profitable alternatives but it is far away from one may think of a low opportunity cost and farmers would dedicate the required hours for UDP production if adoption occurs. Thus, economy of scale is produced by UDP having a better yield but reducing mean cost of rice production as there are urea saving an d weeding cost reduction Moreover, land size is significantly and negatively associated with Intensity of Adoptio n. This effect was expected because UDP becomes more labor intensive as rice cropland size increases. Moreover, UDP adopters were mainly small farmers in Bangladesh. In Ecuador, 80% of the farmers are consider ed as small ( less than 20 ha). Therefore, the m t would be those small farmers.

PAGE 154

154 Another substantial variable was number of children in the household. This factor was found inversely related to IA. As farmers have to take care of their small kids, they would be constrained to dedicate time to experiment new technologies. Also, while a farmer is closer to the market, he would be more willing to giv e more land for UDP production. The two main constraints for market access were bad conditions of the roads and lack of transportation. Con struction of paved roads and transportation service would not just work for UDP and other technology adoption but also for the ease of rice trade, in general. Being a land renter would apparently affect IA positively. These farmers are willing to rent or possibly buy land to implement UDP as this technology produces benefits such as urea saving and income increase, making rice business more attractive. Thus, a land market is important to integrate more people to UDP production. In fact, with the new land l aw being discussed at this moment, new rice farmers may be interest in starting their rice production with this technology. Additionally, the effects of three different markets of Urea fertilizer were able to be differentiated : a subsidized market, a rea l market and a black market. In the end, only the total number of subsidized Urea sacks had a significant and negative impact on IA. I conclude that the effects of the subsidy and the black may be being reflected in this variable. UDP might be perceived as a threat for those benefiting from the subsidy (including those Urea resellers) because Urea saving with UDP may make government reconsider about the subsidy. Hence, subsidies may be an important restraint for UDP or other innovation adoption. However, th e introduction of these types of technologies

PAGE 155

155 that make farmers more competitive, requiring less help from government, are required to reduce this fiscal pressure. It seems that farmers have no problem with credit access; those who applied for credit rec eived the loan. These credits were important for rice production because this money was spent on inputs mainly. On the other hand, special credit providers were the friends; playing a relevant role in t he development of rice bus iness ( an important research to be developed is the impact of the aforementioned credit sources on technology adopt ion) According to Tobit model, c redit participation determines IA positively; cash availability let s farmers try new techniques in their pr oduction system. Thus, a credit market, with acces sible agricultural credits, is strongly recommended for the UDP adoption in rice production; the availability of these types of credits could also work for adoption of other innovations related to the effic ient uses of rice production inputs as credits were used for inputs acquisition mainly. I nsurance s were also significant and positive influence in the determination of UDP adoption ; being protected from natural disasters makes a farmer willing to try new ideas as agricultural resources are ensured As said before, Ecuador does not have a developed insurance market because of different reasons For instance, in agriculture everyone is exposed to many natural disasters every year such as flood and drought ; this makes agricultural insurance market too risky and companies decide not to participate in it. It is very important the current participation of the Ecuadorian government in this market subsidizing such agricultural insurances. However, it seems that th ere still a need of a development of an insurance market (e.g. only 18 farmers mentioned to be a crop insurance holder in this study). With a better accessibility to

PAGE 156

156 these insurances and with an insurance culture, adoption of UDP and other improved technol ogy can take place in the Ecuadorian agriculture. Moreover the risk aversion measurement (the hig her the value, the riskier) w as directly and significantly related to the IA. One can think of these risky farmers as entrepreneurs. They are more willing t o adopt new ideas, of course, with the availability of resources such as money. Again, the importance of accessible credit market is relevant for those more enthusiastic for adoption of UDP or any agricultural technique. In estimating household time avail ability, I could note that those working on farm and off farm dedicated almost the same hours to both activities, on average. I think of two possible scenarios: 1) off farm work would mean extra income that would be taking as an extra capital for family ri ce production; or 2) other alternatives could be more profitable than rice production and household time is being allocated to them over time; this might mean a threat for rice production in the future In the Tobit model on farm hours are positively sign ificant in explaining the Intensity of Adoption; those with more hours spent on on farm activities are more willing to give hectares for UDP production. Farmers were also told that UDP is labor intensive and for that reason those spending few hours on rice production hesitate to pr ovide land for UDP production. Thus, improved technolog ies are needed to make rice production more profitable UDP is an incentive that coul d maintain these people wo rking in agriculture given the rice income improvement and cost reduction. Also, the majority of farmers belong to a formally built agricultural organization. However, in asking famers for groups that may influence on their rice production decisions, they named neighbors/friends, input suppliers and family as the most

PAGE 157

1 57 influential. Additionally with a simple methodology proposed by Bandiera and Rasul (2002), social network effect may be caught without falling in the endogeneity problem in the econometric model Social network has a positive impact on IA which means th at a farmer would be more willing/reluctant to adopt UDP if there are potential/non potential adopters in his social network. In the end, the purpose was to identify the most important social channels of these rice farmers. Working on the establishment of these agricultural groups would be very decisive for the UDP adoption as these channels can be used to spread UDP knowledge across all villages. Despite the fact that the extension work carried out in Ecuador to diffuse UDP, the promotion of this innovation is still needed: a 9.36% (36) of farmers knew about this new fertilization method. Briquetting machine must be reproduced in order for farmers to have access to briquettes and try this technology. Given the budget constraint, only one briquetting machine was imported to experiment with UDP, obtaining very promising results in term of yield and Urea saving. However, i n the descriptive analysis of t he Double bounded Q uestions, one could observe a surprising number of the studied farmers willing to buy Urea b riquette at a price higher than the normal Urea sack 93.25 % when introducing economic benefits/cost (WTP) and adding environmental impacts (EWTP). Moreover, they showed their willingness to adopt UDP when they adduced to give, on average, 49.70% of thei r land to produce with this technology This willingness could have been even greater if the diffusion of UDP would have continued. In the Tobit estimation having knowledge about UDP makes a farmer more interesting in dedicating hectares for UDP productio n. Consequently, investment s are needed to start out this market I encourage public and private investors to ta ke part of this project

PAGE 158

158 that have shown benefits for both firms and government (see International Fertilizer Development Center 2008 ). Finally, one point that must be highlighted is the possible lack of interest in environmental benefit s of this technology Education about environment benefits must be considered in order to mak e farmers more aware of this relevant resource (e.g. Daule River is a source of water for agricultural production and for family consumption). Considering the most significant variables the potential adopter is defined as a farmer having few hectares and small kids, being close to the market, renting land, buying few subsidized Urea sack s participating in the credit and insurance market, taking risks, having UDP knowledge, spending most of his time on on farm activities and keeping potential UDP adopters in his social network. There are two investigations that should be taken into consideration for UDP development: 1) elaboration of a device to facilitate the Urea briquette placement ( N. K. Savant et al. 1991 ) and ; 2 ) examination of UDP impact on the environment (reduction of N going into the atmosphere and aquatic resources). To sum up, important factors were identified, which would be determining adoption of UDP, but also of any agr icultural innovation in gener al. For instance, important constrains for UDP adoptions would be a lack of credit and insurance markets. Hence i nformation was provided to investors and policymakers of potential demanders of a market with economic and social benefits. Finally, t his the sis is a contribution to the Ecuadorian agricultural technology adoption literature

PAGE 159

159 APPENDIX QUESTIONNAIRE (SPANISH VERSION) Preguntas de Filtro Usted maneja un cultivo de arroz? ____Si ____No(Pase a Seccin 7: Salud, hoja 6) Qu clase de manejo usted ha llevado en su cultivo de arroz? ____Arroz al trasplante ____Arroz Al voleo ____Otro: Cul?_________ Usted toma las decisiones sobre su cultivo de arroz? ____ Si ____No (siga a otra casa) SECCI"N 1: DIFUSI"N DE LA APLICAC I"N PROFUNDA DE BRIQUETAS DE UREA 1. Usted ha escuchado, visto o utilizado la Aplicacin Profunda de Briquetas de Urea (las bolitas de Urea que se enterraban en el suelo)? ___Si ___No(pase seccin 2: Redes sociales ) 2. De qu forma conoci la Aplic acin Profunda de Briquetas de Urea (las bolitas de Urea que se enterraban en el suelo)? (Ponga s o no, o escoja el nmero de acuerdo a las claves) 3. De qu forma por primera vez conoci personalmente los resultados de la Aplicacin Profunda de Briq uetas de Urea (las bolitas de Urea que se enterraban en el suelo)?(Marque X o escoja el nmero de acuerdo a la clave4) No he observado Experimento en mi finca Experimento en finca ajena Observo a (clave4) Otras (escriba cuales) NOTA: SI OBSERVO (CLAVE4) A OTRA PERSONA LOS RESULTADOS DEL APBU CONTESTE PREGUNTA 4 EN OTRO CASO PASE A LA PREGUNTA 5.

PAGE 160

160 4. Cul es e l nombre del dueo y la distancia aproximada desde su finca hasta la finca en donde usted observ los resultados de la Aplicacin Profunda de Briquetas de Urea (las bolitas de Urea que se enterraban en el suelo)? (Distancia en tiempo que el encuestado le d iga) Unidad Cantidad Nombre 5. Si usted volvi a saber sobre la Aplicacin Profunda de Briquetas de Urea (las bolitas de Urea que se enterraban en el suelo), hace cunto tiempo y como volvi a saber? (Poner cero en cantidad si no supo nunc a ms y pasar a la pregunta 6 o 7) 6. NOTA: CONTESTE ESTA PREGUNTA SI EL ENCUESTADO EXPERIMENTO CON LAS BRIQUETAS DE UREA EN LA PREGUNTA 3 O 5. Por favor, dganos la informacin aproximad a sobre la produccin con la Aplicacin de briquetas de Urea de la ltima vez que la utiliz (VER HOJA PRODUCCI"N CON APBU O PGINA 2). 7. De lo que ha escuchado, observado o experimentado, Qu conocimiento tiene sobre la Aplicacin Profunda de Briquetas d e Urea(las bolitas de Urea que se enterraban en el suelo)? (marque con una X el casillero correspondiente) Poco Regular Bastante Hace cunto? Cmo supo de nuevo sobre la Aplicacin Profunda de briquetas de Urea (las bolitas de Urea que se enterraban en el suelo)? Unidad Cantidad Reuniones en Asociaciones Experimento en finca propia Experiment o en finca ajena Conversando u Observando a (clave4) Por esta encuesta Otros (escriba cuales)

PAGE 161

161 SECCI"N 2: REDES SOCIALES 8. Indique todos los Grupo u Organizacin que usted u otro miembro de su hogar p ertenece. Miembro de su hogar (Jefe del hogar, esposa, hijo, etc.) Tipo de Organizacin / Grupo (clave5) Nombre del Grupo u Organizacin Nmero de personas dentro de su Grupo u Organizacin Cuntas de estas personas son familiares Participa en su grupo voluntari amente (si/no) Frecuencia de participacin Seale como es su participacin en cada organizacin Semanal Mensual Anual No participa Algo Activo Muy Activo Lder SOLO SI HA USADO LA APLICACI"N PROFUNDA DE BRIQUETAS DE UREAS: PRODUCCI"N CON APBU 6. BASADA EN LA LTIMA PRODUCCI"N DE ARROZ CON LAS BRIQUETAS DE UREA (LA BOLITAS DE UREA QUE SE ENTERRABAN EN EL SUELO) Ciclo Corto Produccin de Arroz Cultiv o Rubros Unidad Cantidad Valor Unitario Observaciones Arroz Gastos en Semilla $ Nombre de la semilla: Tipo: R V_ H Gastos en Fertil izantes Qu Cantidad de Urea us y a quien se la compr? $ Casas ... $ $ Quien: Qu cantidad de NPK (Completo) us? $ Qu cantidad de DA P (el de la pata, inicio) us? $ Gastos en Foliares (escriba el costo total de todo el ciclo del cultivo de arroz) $

PAGE 162

162 Gastos en Herbicidas/Malezas (escriba el costo total de todo el ciclo del cultivo de arroz) $ Gastos en Insecticida /Bichos y otros /lquidos (escriba el costo total de todo el ciclo del cultivo de arroz) Gastos en pago a Jornales Qu cantidad de jornales/personas uso para el Trasplante? $ Qu cantidad de jornales/personas uso para Aplicaciones de Productos? $ Qu cantidad de jornales/personas uso para cosechar? $ Costos relacionados a Cosecha y Pilado: Cosechadora, transporte, cargada, sacos y pilada (escriba el costo total en el cultivo de arroz) $ Arroz Cosecha En el ltimo cultivo, Qu cantidad cosech? En el ltimo cultivo, Qu cantidad dej para consumo propio? Qu cantidad y a qu precio vendi en Sacas (arroz en cascara)? $ Qu cantidad y a qu precio vendi en Arroz Pilado? $ De esta cosecha Qu cantidad de dinero destin para ahorro (escribe solo la cantidad de dinero ) $ a) Cunta rea uti liz para pro ducir con las briquetas de Urea ? _________Cuadras ____________Tareas C) Si alquilo las tierras, cunto pago en total? $_________ b) De esta rea Cunto alquilo? ( PONGA CERO SI NO ALQUILO Y PASE A LA PREGUNTA 9 ) ______Cuadr a _______Tareas 9. Por favor, conteste el siguiente conjunto de preguntas. (Para el encuestador: mencionar los grupos al encuestado en cada preg unta) Grupos Preguntas Vecinos Miembros de Asociacin/Gru po Amigos no incluidos en las anteriores c ategoras Familiare s Dirigentes President es o lderes Trabajadore s de extensin Comerciant e Otros (detalle nombre y que es para el encuestad o) De quin aprendi el uso de fertilizantes? (puede escoger ms de una opcin) Con qu frecuencia y c on qu grupo usted habla sobre tcnicas agrcolas? (Clave7) Si tiene un problema a quin pedira

PAGE 163

163 un consejo o ayuda? Elija los 3 grupos que influye ms en sus decisiones sobre el manejo de su cultivo de arroz. (Siendo 1 el ms important e y 3 el menos importante) Cul es el nombre de la persona a la que gente le pide consejos sobre el manejo de cultivos? (Anote en el nombre incluso cua ndo el encuestado mismo sea esa persona y cualquier otra identificacin que sea posible encont rarlo) Nombre: _____________________ Qu es para usted?:_______________ A cuntos minutos de distancia esta la casa de esta persona?:_____________________ 10. Indique con cuntos hombres y mujeres aproximadamente usted conversa sobre tcnicas agrcolas y a que grupo pertenecen. (escriba el nmero de personas en todos los grupos que diga el encuestado) Grupo Religioso Familiares Asociacin/otro grupo/vecinos/amigos Hombres Mujeres 11. En general, Cmo usted considera la comunicacin entre los agricultores de su sector? Marque con una X Baja Regular Indiferente Buena Muy Buena

PAGE 164

164 SECCI"N 3: ANTERIORES ADOPCIONES DE TCNICAS AGRCOLAS, PERCEPCIONES Y DISPONIBILIDAD A PAGAR 12. Usted ha usado nuevas tcn icas agrcolas en sus cultivos? ___Si ___No (pase 14) ____ No recuerdo(pase 14) 13. Si usted ha usado nuevas tcnicas agrcolas en sus cultivos, Cules fueron, hace cunto tiempo las us y cmo calificara sus experiencias con los mismas? Cules fuero n los Cambios? (Clave 6) Hace cunto tiempo? Cmo calificara sus experiencias con esas formas de producir? Unidad Cantidad Psima Me dio Igual Buena 14. Por favor, responda si esta desacuerdo, indiferente o acuerdo sobre las siguientes oraciones (Marque con X) Desacuerdo Indiferente Acuerdo a) Usted comparte cualquier conocimiento agrcola con los agricultores de su sector. b) TODOS los agricultores de su sector comparten experiencias del manejo de cultivos. c) Usted utiliz ara un nuevo mtodo de aplicar la Urea en su cultivo de arroz, aunque otros no hayan usado este mtodo.

PAGE 165

165 15. Considerando los beneficios y costos de las Briquetas de Urea : Version 1: a. Estara usted dispuesto a pagar por c ada quintal de briquetas de Urea UN D"LAR DE MS de lo que usted paga por la Urea normalmente? Si (siga b) b. Ya que est dispuesto a pagar un dlar, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea UN D"LAR y CINCUENTA CENT AVOS DE MS de lo que usted paga por la Urea normalmente? S No d. Sin importar las respuestas anteriores, En Su Mente Cunto usted estara dispuesto a pagar por cada quintal de briquetas de Urea adicionalmente a lo que usted paga por la Urea norm almente? $______ No (siga c) c. Ya que no est dispuesto a pagar un dlar, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea CINCUENTA CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No NOTA: PARA EL ENCUESTADOR: Debe de leerle este texto al encuestado antes se preguntarle la Pregunta 15: Explicacin de la Aplicacin Profunda de Briquetas de Urea y de sus Beneficios y Costos. La Aplicacin Profunda de Briquetas de Urea (APBU) es una tecnologa que convierte la Urea normal en briquetas (o pequeas bolas de Urea ) mediante una mquina llamada Briquetadora. Estas briquetas, que pe san alrededor de 2.7 a 4 gramos, son enterradas manualmente en medio de cuatro plantas en el suelo fangoso por debajo de la lmina de agua, despus del trasplante y se lo hace por una sola vez en el ciclo de cultivo. Mediante las briquetas de Urea se puede evitar alrededor del 50% de la prdida de la Urea por evaporacin y escorrentas. De esta forma, las plantas podrn captar ms el Nitrgeno contenido en la Urea Los beneficios y costos de la Aplicacin de las briquetas de Urea seran: Beneficios: Con las briquetas de Urea se aplicara de 4 a 4,5 quintales de Urea por hectreas. Una reduccin de Urea usada del 40%. Incremento de la Produccin de arroz del 25%, y por ende mayores ingresos. Posible creacin de pequeas empresas encargadas de vender las briqu etas de Urea y por ende generacin de empleos. Costos: Se necesitara dos jornales ms por hectrea para aplicar la Urea Se debe tener comprados los 4 o 4,5 quintales de Urea para la aplicacin entre los 15 a 20 das despus del trasplante, ya que se fert iliza una solo vez durante todo el ciclo del cultivo. Po cual se debe hacer un solo gasto en todos los quintales de Urea al inicio

PAGE 166

166 Version 2: a. Estara usted dispuesto a pagar por cada quintal de briquetas de Urea TRES D"LARES DE MS de lo que usted paga por la Urea normalmente? Si (siga b) b. Ya que est dispuesto a pagar tres dlares, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea TRES D"LARES Y CINCUENTA CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No d. Sin importar las respuestas anteriores, En Su Mente Cunto usted estara dispuesto a pagar por cada quinta l de briquetas de Urea adicionalmente a lo que usted paga por la Urea normalmente? $______ No (siga c) c. Ya que no est dispuesto a pagar tres dlares, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea DOS D"LARES Y CINCUENTA CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No Version 3: a. Estara usted dispuesto a pagar por cada quintal de briquetas de Urea DOS D"LARES DE MS de lo que usted paga por la Urea normalmente? Si (siga b ) b. Ya que est dispuesto a pagar dos dlares, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea DOS D"LARES Y CINCUENTA CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No d. Sin importar las respuestas anter iores, En Su Mente Cunto usted estara dispuesto a pagar por cada quintal de briquetas de Urea adicionalmente a lo que usted paga por la Urea normalmente? $______ No (siga c) c. Ya que no est dispuesto a pagar dos dlares, Estara usted dispues to a pagar por cada quintal de briquetas de Urea UN D"LAR Y CINCUENTA CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No

PAGE 167

167 16. Considerando los beneficios y costos de las Briquetas de Urea pero ahora considerando el b eneficio ambiental: Version 1: a. Estara usted dispuesto a pagar por cada quintal de briquetas de Urea UN D"LAR DE MS de lo que usted paga por la Urea normalmente? Si (siga b) b. Ya que est dispuesto a pagar un dlar, Estara usted di spuesto a pagar por cada quintal de briquetas de Urea UN D"LAR y CINCUENTA CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No d. Sin importar las respuestas anteriores, En Su Mente Cunto usted estara dispuesto a pagar por cada q uintal de briquetas de Urea adicionalmente a lo que usted paga por la Urea normalmente? $______ No (siga c) c. Ya que no est dispuesto a pagar un dlar, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea CINCUENTA CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No NOTA: PARA EL ENCUESTADOR: Debe de leerle este texto al encuestado antes se preguntarle la Pregunta 16: Beneficios ambientales de la Aplicacin Profunda de Briquetas de Urea Sabiendo que los primeros beneficios y costos de las briquetas de Urea son: Beneficios: Reduccin del uso de Urea en un 40%, Incremento de la produccin de un 25% y una Po sible generacin de empleo. Costos: Aumento de los jornales utilizado y Compra al inicio de todos los quintales de Urea que van a ser aplicados. Urea que se puede es tar perdiendo por escorrentas o por evaporacin sera alrededor del 50%, lo cual afecta negativamente al medio ambiente. Mediante la aplicacin con las briquetas de Urea se reduce la perdida de la Urea consiguiendo los siguientes impactos ambientales: S e reducira los gases, en este caso el Nitrgeno contenido en la Urea que generan el Calentamiento global. Se reducira el nitrgeno en los ros, conservando el oxgeno en el agua y la vida acutica. Se evitaran enfermedades de las personas que beben a guas de los ros, ya que se reduce el nitrgeno en esas aguas y el cual es malo para la salud cuando existe en grandes cantidades en el agua que es bebida.

PAGE 168

168 Version 2: Version 3: a. Estara usted dispuesto a pagar por cada quintal de briquetas de Urea DOS D"LARES DE MS de lo que usted paga por la Urea normalmente? Si (siga b) b. Ya que est dispuesto a pagar dos dlares, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea DOS D"LARES Y CINCUENTA CENTAVOS DE MS de lo que ust ed paga por la Urea normalmente? S No d. Sin importar las respuestas anteriores, En Su Mente Cunto usted estar dispuesto a pagar por cada quintal de briquetas de Urea adicionalmente a lo que usted paga por la Urea normalmente? $______ N o (siga c) c. Ya que no est dispuesto a pagar dos dlares, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea UN D"LAR Y CINCUENTA CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No 17. NOTA: SE RESPONDE ESTA PREGUNTA SI ES QUE HUBO UNA DISPONIBILIDAD A PAGAR POR LAS BRIQUETAS DE UREA EN LAS PREGUNTAS 15 0 16 Dado que usted est dispuesto a pagar por las briquetas de Urea En cuntas cuadras tareas usted estara dispuesto a utilizar la Aplicaci n Profun da de Briquetas de Urea ? ______Cuadra(s) ________ Tarea(s) a. Estara usted dispuesto a pagar por cada quintal de briquetas de Urea TRES D"LARES DE MS de lo que usted paga por la Urea normalmente? Si (siga b) b. Ya que est dispuesto a pagar tres dlares, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea TRES D"LARES Y CINCUENTAS CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No d. Sin importar las respuestas anteriores, En Su Mente Cunto usted estara dispuesto a pagar por cada quintal de briquetas de Urea adicionalmente a lo que usted paga por la Urea normalmente? $______ No (siga c) c. Ya que no est dispuesto a pagar tres dlares, Estara usted dispuesto a pagar por cada quintal de briquetas de Urea DOS D"LARES Y CINCUENTAS CENTAVOS DE MS de lo que usted paga por la Urea normalmente? S No

PAGE 169

169 SECCI"N 4: SISTEMA DE PRODUCCI"N 17. BASADA EN LA LTIMA PRODUCCI"N DE ARROZ, INDIQUE LOS COSTOS E INGRESOS QUE TUVO a) C mo lleva su cultivo? Piscina______ Poza__________ Ciclo Corto Produccin de Arroz Cultivo Rubros Unidad Cantidad Valor Unitario Observaciones Arroz Gasto en alquiler de cada unidad de terreno (Para el encuestador: el pago depende de la can tidad de tierra que se alquila para cultivar) $ Gasto en preparacin de cada unidad de Terreno: arado y fangueado $ Gastos en Semilla $ Nombre de la semilla: Tipo: R V_ H Gastos en Fertilizantes Qu Cantidad de Urea us y a quien se la compr? Gobierno: $ Casas $ $ Quien: Qu cantidad de NPK (Completo) us? $ Qu canti dad de DAP (el de la pata, inicio) us? $ Gastos en Foliares (escriba el costo total de todo el ciclo del cultivo de arroz) $ Gastos en Herbicidas/Malezas (escriba el costo total de todo el ciclo del cultivo d e arroz) $ Gastos en Insecticida /Bichos y otros /lquidos (escriba el costo total de todo el ciclo del cultivo de arroz) Gastos en pago a Jornales Qu cantidad de jornales/personas uso para el Trasplante? $ Qu can tidad de jornales/personas uso para Aplicaciones de Productos? $ Qu cantidad de jornales/personas uso para cosechar? $ Gastos en acceso al agua de riego Paga a Junta de Riego_____ (cada cuanto tiem po y cunto paga) $ Paga por Riego por bomba_____ (gasto por cada galn de diesel para la bomba o pago con sacas de arroz) $ Costos relacionados a Cosecha y Pilado: Cosechadora, transporte, $

PAGE 170

170 cargada, sacos y pilada (escriba el costo total en el cultivo de arroz) Arroz Cosecha En el ltimo cultivo, Qu cantidad cosech? En el ltimo cultivo, Qu cantidad dej para consumo propio? Qu cantidad y a qu pre cio vendi en Sacas (arroz en cascara)? $ Qu cantidad y a qu precio vendi en Arroz Pilado? $ De esta cosecha Qu cantidad de dinero destin para ahorro (escribe solo la cantidad de dinero ) $ 18. LLENE LA TABLA SI ES QUE TIENE OTROS CULTIVOS (PONGA EL NOMBRE DEL CULTIVO A LA IZQUIERDA) DE CICLO CORTO O CULTIVOS DE CICL O PERMANENTE CICLO CORTO Cultivo Rubros Unidad Cantidad Valor Unitario Observaciones De ciclo corto: ______ ___ Costo Total Estimado (escriba el costo total solo) $ COSECHA Qu cantidad cosech? Qu cantida d dej para consumo propio? Qu cantidad vendi y a qu precio? $ De esta cosecha Qu cantidad de dinero destin para ahorro (escribe solo la cantidad de dinero ) $ CICLO PERMANENTE De ciclo perm an ente: ______ ____ Costo Total Estimado (escriba el costo total solo) $ COSECHA Qu cantidad cosech? Qu cantidad dej para consumo propio? Qu cantidad vendi y a qu precio? $ De esta cosecha Qu cantidad de dinero destin para ahorro (escribe solo la cantidad de dinero ) $ De ciclo perman ente: ______ ____ Costo Total Estimado (escriba el costo total solo) $ COSECHA Qu cantidad cosech? Qu cantidad dej para consumo propio? Qu cantidad vendi y a qu precio? $ De esta cosecha Qu cantidad de dinero destin para ahorro (escribe solo la cantidad de dinero ) $ Cunta rea dedico a este cultivo? _____ cuadras _______tareas Cuntos ciclos de cultivos de arroz usted tiene al ao? __________ciclo(s)

PAGE 171

171 19. Cunto tiempo usualmente se demora en vender su produccin de: Verano Invierno Arroz Unidad: Cantidad: Unidad: Cantidad: Otro cultivo Cul?: Unidad: Cantidad: Unidad: Cantidad: 20. A cuntas piladoras usualmente usted puede ir a vender su produccin de arroz? (si no va a las piladoras poner 0 y pasar 22) _________piladora(s) _________piladora(s) 21. ( No conteste si no va a vender su arroz a las piladoras ). En las piladoras que usualmente usted vende su produccin de arroz, dganos un nmero aproximado de cuanto otros productores de arroz venden su produccin en esa misma pila dora ________ productor(es) ________ productor(es)

PAGE 172

172 SECCI"N 5: MOVIMIENTOS DE DINERO 22. En el ltimo ao, Usted pedio un prstamo? ____Si ____No(pase a 28A) 25. En que gast el prstamo que hizo de mayor monto? Elija los 3 gastos ms importantes, siendo 1 el ms importante y 3 menos importante. Gastos (mencione los gastos al encuestado) semillas, alquilar maquinaria, equipos, repuestos para equipos vehculos, Educacin, .... Gastos en entretenimiento, viajes, .... Otros gastos: Cules?__________ 26A. Usted vendi su arroz a algn pre stamista? 27) 26B. El precio que vendi al prestamista fue precio al del 26C. Cul fue el precio y como lo vendi su arroz? Cmo? Por: ..... 22A A quin ha acudid o? (marque con una X y puede elegir ms de una opcin) 22B Si le aprobaron el crdito, dganos un aproximado del monto que le prestaron y quien se lo presto. (Marque con una X en el casillero si no le aprobaron el prstamo) Bancos Cooperativas Cajas de ahorro Banco Comunal Fundaciones Chulquero Piladoras Familiar o amigo Almacenes Otro: __________ No me aprobaron el prstamo (pase a 23) $______Bancos $______Cooperativas $______Cajas de ahorro $______Banco Comunal $______Fundaciones $______Chulquero $______Piladoras $______Familiar o amigo $______Almacen es $______Otro:___________ 27. En el siguiente ciclo va a intentar pedir un prstamo a la misma persona/institucin: 23. En caso de NO haberse concretado el prstamo, Cules fueron los motivos? (Conteste y luego contine desde la 27 ) ___Motivos del prestamista ____Motivos relacionados a usted ____Otros: cules?_________ 28B. Para qu cult 28E. Tiene algn seguro subsidiado por el gobierno? Para qu cultivo? (escriba C1, C2, etc, seg n el 24A. Cul fue el plazo de pago para el prstamo que hizo de mayor monto?(Marque con X, puede escoger solo 1 opcin) __Pagos diarios __ Pagos semanales __ Pagos quincenales __ Pagos mensuales __ Pagos trimes trales __Otros: Cules?____________ 24B. De acuerdo al plazo de pago, Cunto usted pago por este prstamo? Pago:$___________ Nota: Si logro algn prstamo, conteste desde la 2 4 en adelante.

PAGE 173

173 SECCI"N 6: MERCADO LABORAL 29. TRABAJO ( Nota: Las siguientes preguntas estn basadas en un da normal de trabajo (n mero de horas)) Integrante Padre Cul es la principal fuente de ingreso? (Puede trabajar en agricultura o en otra cosa) Agricultura ___ ; No agrcola __ No ingresos __ Cuntas horas usted trabaja en actividades agrcolas? hora(s) Cuntas hor as usted trabaja en actividades NO AGRCOLAS? ( Si contesta esta pregunta, ver a qu actividad se dedica en pregunta 33) hora(s) Cuntas horas le ocupa los estudios? hora(s) Cuntas horas le ocupa en las actividades de casa? Madre Cul es la pri ncipal fuente de ingreso?(Puede trabajar en agricultura o en otra cosa) Agricultura ___ ; No agrcola __ No ingresos __ Cuntas horas su pareja trabaja en actividades agrcolas? Cuntas horas su pareja trabaja en actividades NO AGRCOLAS? ( Si c ontesta esta pregunta, ver a qu actividad se dedica en pregunta 33) Cuntas horas su pareja le ocupa los estudios? Cuntas horas su pareja le ocupa en las actividades de casa? Hijo Cul es la principal fuente de ingreso?(Puede trabajar en ag ricultura o en otra cosa) Agricultura ___ ; No agrcola __ No ingresos __ Cuntas horas su hijo(a) trabaja en actividades agrcolas? Cuntas horas usted trabaja en actividades NO AGRCOLAS? ( Si contesta esta pregunta, ver a qu actividad se dedi ca en pregunta 33) Cuntas horas le ocupa los estudios? Cuntas horas le ocupa en las actividades de casa? 30. Cuntas personas en su hogar trabajan en la agricultura, incluyendo a usted? __________persona(s) 31. Cuntas personas en tu hoga r trabajan en actividades NO AGRICOLAS, incluyen a usted? __________persona(s) 32. Cuntas personas en su hogar estudian, incluyendo a usted? __________persona(s)

PAGE 174

174 34. EN EL LTIMO AO, indique si recibi algn tipo de dinero que no es resultado del trabajo como por ejemplo de: Monto en $ Frecuencia en la recibi este dinero Del dinero que recibi, cunto ahorr? Familiares en otras pa rtes del pas Unidad: Cantidad: Familiares en el extranjero Unidad: Cantidad: Bono solidario Unidad: Cantidad: Arriendos Unidad: Cantidad: Ganancias financieras Unidad: Cantidad: Donaciones Unidad: Cantidad: Otros: _______________ Unidad: Cantidad: 33. Trabaja Usted o alguien ms de su familia en alguna otra actividad adems de la agrcola que les de dinero? ___ Si(llene tabla de abajo) ___No(pase 34) Nota: Responda las siguientes preguntas de acuerdo a las actividades que usted trabaj en el ltimo ao. Preguntas Ac tividades Cul fue su Ingreso Aproximado por esta actividad? Cada cunto tiempo reciba este ingreso? De este monto cuanto ahorr? Pesca $ ____Diario ____Semanal ___Mensual $ Proyectos forestales $ ____Diario ____Semanal ___Mensual $ Cri anza de animales $ ____Diario ____Semanal ___Mensual $ Apicultura (Abejas) $ ____Diario ____Semanal ___Mensual $ Algn proceso Agroindustrial $ ____Diario ____Semanal ___Mensual $ Turismo $ ____Diario ____Semanal ___Mensual $ Ar tesana $ ____Diario ____Semanal ___Mensual $ Algn tipo de Comercio $ ____Diario ____Semanal ___Mensual $ Alquiler de equipos $ ____Diario ____Semanal ___Mensual $ Otras: __________________ $ ____Diario ____Semanal ___Mensual $

PAGE 175

175 SECCI"N 7: SALUD 35. Responda el siguiente conjunto de preguntas: a) De dnde obtiene el agua para el uso diario? b) Usted usa el agua del rio o pozo para beber y cocinar? c) Qu mtodos usa para purificar el agua? d) Cunto gasta por purificar esa agua? dlares POR CADA CADA e) Se ha enfermado del estmago en el ltimo ao? Si responde NO; Pasar a pregunta m ) f) Ust ed cree que esa enfermedad estuvo relacionada al tipo de agua que bebe? Si responde NO; Pasar a pregunta m ) g) Dnde se hizo ver? Subcentro de salud de la El pueblo ms cercano ..... Cul Cu l?................... h) Cun lejos queda el dispensario mdico desde su casa? i) Cunto gast por transportarse hasta ese lugar? j) Cunto gast en la consulta? k) Cun to gast en la medicina? l) En total, Cuntas veces se enferm del estmago este ltimo ao? m) Otras personas dentro de su hogar se han enfermado del estmago en el ltimo ao? Si responde NO; Pasar a preg unta 36 ) n) Cuntos? o) Usted cree que esas enfermedades estuvieron relacionadas al agua que ellos beben?

PAGE 176

176 SECCI"N 8: CARACTERSTICAS DEL HOGAR 36. Indique cuales de los siguientes bienes posee y en cuanto us ted cree que estn. tem Valor tem Valor __Casa __ Cosechadora __ Tractor u otros equipos mayores __ Fumigadora u otras equipos __ Bodega __ Carro __ Piscina o Reservorio __ Planta elctrica __ Bombas de riego __ Empacadora 37.S u casa es: __ Propia; __ Alquilada; __ Prestada; __ Compartida 38.Servicios que posee: __ Agua Potable; __ Luz; __ Alcantarillado; __ Telfono; __ Alumbrado Pblico; ____Internet 39. Su casa est hecha de : __ Madera; __ Caa; __ Cemento; __ Mixt a 40. Cul es la distancia en tiempo del pueblo ms cercano y cunto gasta en ir y regresar de all? Unidad: Cantidad: ____ dlares 41. Cuntos trasbordos necesita hacer hasta llegar all? __ 1; __ 2; __3; __4; __5 42. Cuntas personas estn viviendo en su hogar? _______ adulto(s) ______nio(s) ______joven(es) 43. Cul fue el gasto que hizo la ltima vez en Alimentos Ropa tiles Escolares Arreglos en la casa Otros: Cules? $ $ $ $ $ 44A. Durante el ltim 45. De la ltima vez que sufri estas INUNDACIONES Y/O SEQUIAS, qu porcentaje de tierra perdi en donde tena cultivos de: Arro z Mango Maz Otros: 1) 2) 3) 4) % (Sequia) 46. De la ltima vez que sufri estos INUNDACIONES Y/O SEQUIAS, cunto gast para recuperar la normalidad en su actividad agropecuaria?

PAGE 177

177 IDENTIFICACI"N DEL HOGAR Canton:___________ Sector:___________ Recinto/Barrio:____________ Tel.: _______ A2 9.Otras Indicaciones exactas para llegar a la casa: GPS Cdigo: 47. Actualmente Cunta rea en total usted tiene para cultivar? 48. Seale si alquila o da en alquiler esta rea para cultivo(marque con una X): ____doy en alquiler (1) ____ alquilo(2) _____Ninguna de las anteriores 1: 2: 49. Actualmente Cunta rea es destinada al cultivo de Arroz? Tareas 50. Actualmente Cunta rea es destinada a otros cultivos? Tareas 51. Cunta rea tiene sin usar? S ECCI"N 9: CARACTERSTICAS DEL INDIVIDUO 52. Sexo M F 53. Edad_______ aos 54. Cuntos aos en total ha estudiado Usted: ____ aos 55. Mximo nivel de educacin terminado Analfabeto Secundaria Otros:Cule s? Primaria Universitaria Ninguno 56A Usted ha recibido estudios agrcolas? S No(pase 57) 56B Qu temas estudio en esos cursos? (Clave6) 56C Mediante quin o como recibi esos estudios agrcola? ____Gobierno ____ Organizacin ____ Ferias Agropecuarias ____ Escuelas ____ Colegios ____Universidades ____ Otros(cuales): __________ 57. Estado civil ____Soltera(o) ____Viuda(o) ____Casada(o) ____Unin Libre. ____Divorciada(o)

PAGE 178

178 LIST OF REFERENCES Anderson, Kym, Lee A. Jackson, and Chantal P. Nielsen. 2005. "Genetically Modified Rice Adoption: Implications for Welfare and Poverty Alle viation Journal of Economic Integration 20(4): 771 788. Banco Central del Ecuado r. 2011 Encuesta de Coyuntura Sector Agropecuario. http://www.bce.fin.ec/documentos/PublicacionesNotas/Catalogo/Encuestas/Coyun tura/Integradas/etc201004.pdf. Banco Nacional de Fomento. 2012. Microcreditos. https://www.bnf.fin.ec/index.php?option=com_content&view=category&id=5&Itemi d=4&lang=es (accessed June 18, 2012) Bandiera, Oriana and Imran Rasul. 2006. "Social Networks and Technology Adoption in Northern Mozambique The Economic Journal 116(51 4): 869 902. Barnum, Howard N. and Lyn Squire. 1979. "An Econometric Application of th e Theory of the Farm Household. Journal of Development Economics 6: 79 102. Bennett, J. 1995. "Biotechnology and the Future of Rice Production GeoJournal 35(3): 3 33 335. Besley, Timothy and Anne Case. 1993. "Modeling Technology Adoption in Developing Countries The American Economic Review, Papers and Proceedings of the Hundred and Fifth Annual Meeting of American Economic Association 83(2): 396 402. Breustedt Gunnar, Jrg Mller Scheeel, and Uwe Latacz Lohmann. 2008. "Forecasting the Adoption of GM Oilseed Rape: Evidence from a Discrete Choice Experiment in Germany Journal of Agricultural Economics 59(2): 237 256. Buttel, Frederick H., Martin Kenney, and Jr. Jack Kloppenburg. 1985. "From Green Revolution to Biorevolution: Some Observations on the Changing Technological Bases of Economic Transformation in the Third World Economic Development and Cultural Change 34(1): 31 55. Bwire, Joseph. 2008. "Facto rs Affecting Adoption of Improved Meat Goat (Boer) Production in Rangelands of Sembabule District http://news.mak.ac.ug/documents/Makfiles/theses/Bwire_Joseph.pdf Cameron A. C. and Pravin K. Trivedi. 2010. Microeconometrics U sing Stata Revis ed E d. Texas : Stata Press. Castillo, Mara J. 2011. "Seguro Agrcola en Ecuador: Un Servicio a la Comunidad o u n Negocio Rentable?" Paper presented at Coffe B reak Opinion desde la Academia, Graduate School of Management ESPAE Guayaquil Ecuador.

PAGE 179

179 Chiriboga Manuel and Brian Wallis 2009. Diagnostico de la Pobreza Rural en Ecuador y Respuestas de Politica Publica. Quito: Centro Latinoamericano para el Desarrollo Rural. Conley, Tim othy G., and Christopher R. Udry. 2010. Learning about a New Technology: Pineapple in Ghana. American Economic Review 100(1): 35 69. Contreras, Orlando D. and Marcelo Espinosa 2010. "D iseo y Clculo de una Mquina para Producir Briquetas d e Urea http://www.dspace.espol.edu.ec/bitstream/123456789/8027/1/Dise%C3%B1o%20 y%20C %C3%A1lculo%20%20de%20una%20m%C3%A1quina%20para%20produ cir%20Briquetas%20de%20Urea.pdf Dalrymple, Dana. 1986. Development and Spread of High Yielding Rice Varieties in Developing Countries Washington, D.C.: Agency for International Development. Dekker, M. 2006. "Estimating Wealth Effects without Expenditure Data: Evidence from Rural Ethiopia Ethiopian Journal of Economics 15(1): 35 54. Dobermann Achim 2012. IRRI Agronomy Challenge : How Much Fertilizer http://www.irri.org/index.php?option=com_k2&view=item&id=11712%3Airri agronomy challenge how much fertilizer&lang=en. Doss, Cheryl R 2006. "Analyzing Technology Adoption U sing Microstudies: Limitations, Challenges, and Opportunities for Improvement Agricultural Economics 34: 207 219. Doss, Cheryl R. and Michael L. Morris. 2000. "How does Gender Affect the Adoption of Agricultural Innovations? The Case of Improved Maize Technology in Ghana Agricultural Economics 25(1): 27 39. D'Souza, Gerard E., Douglas Cyphers, and Tim T. Phipps. 1993. "Factors Affecting the Adoption of Sustainable Agricultural Practices Agricultural and Res ource Economics Review 22(2). Ecuadorian Const. of 2008, art. CDI. Escuela Superior Politecnica del Litoral, University of Florida, and USDA PL 480. 2008. "Implementacin de un Programa para Mejoramiento Del Ingreso de Pequeos Productores de Arroz en e l L itoral Ecuatoriano: Aplicacin P rofunda d e Br iquetas d e Urea y Microcrdito." Unpublished. Evenson, R. E. and D. Gollin. 2003. "Assessing the Impact of the Green Revolution, 1960 to 2000 American Association for the Advancement of Science 300(5620) : 758 762. Feder, Gershon and Dina L. Umali. 1993. "The Adoption of Agricultural Innovations A Review." Technological Forecasting an d Social Change 43: 215 239.

PAGE 180

180 Fernandez Cornejo, Jorge, Chad Hendricks, and Ashok K. Mishra. 2005. "Technology Adoption a nd Off Farm Household Income: The Case of Herbicide Tolerant Soybeans Journal of Agricultural and Applied Economics 37(3): 549 563. Food and Agriculture Organization of the United Nations 2012. FAOSTAT http://faostat.fao.org/?lang=en (accessed May 15, 2012). Food and Agriculture Organization of the United Nations 2011. "Rice Market Monitor http://www.fao.org/docrep/013/am1 56e/am156e00.pdf Foster, A. D. and M. R. Rosenzweig. 1995. "Learning by Doing and Learning from Others: Human Capital and Technical Change in Agriculture The Journal of Political Economics 103(6): 1176 1209. Gedikoglu, Haluk and Laura M.J. McCann. 2007. "Impact of Off Farm Income on Adoption of Conservation Practices Paper presented at the Annual Meeting from American Agricultural Economics Association, Oregon, TN. Gianessi, Leonard, Sujat ha Sankula, and Nathan Reigner, ed. 2003. Plant Biotechno logy: Potential Impact for Improving Pest Management in European Agriculture: A Summary of Nine Case Studies Washington, D C : The National Center for Food and Agricultural Policy Greene, William H. 2012. Econometric Analysis 7th E d. Boston : Prentice Hall. Gujarati, Damodar N. and Dawn C. Porter. 2009. Basic Econometrics 5th E d. New York: McGraw Hill Irwin. Hannan, Timothy H. and John M. McDowell. 1984. "The Determinants of Technology Adoption: The Case of the Banking Firm The RAND Journal of Economics 15(3): 328 335. Heckathorn Douglas D. 1997 Respondent Driven Sampling: A New Approach to the Study of Hidden Populations S ocial Problems 44(2): 174 199 Herdt, R. W. and C. Capule. 1983. Adoption, Spread, and Production Impact of Modern Rice Varieties in Asia Los Banos: International Rice Research Institute Herrera, Pal, Katherine Jimnez, Graciela Prado, and Ramn Espinel. 2010. "Capital Social y Desarrollo Comu nitario: El Caso de las Juntas de Usuarios del Sistema de Riego del Valle d el Daule." In Tierra y Agua: Interrelaciones de u n Acceso Inequitativo ed. Edgar Isch and Alex Zapatta, 111 130. Quito: Sistema de Inve stigacin sobre la Problemtica Agraria en e l Ecuador. Hildebrand, P. E., L. Andrade, W. Bowen, R. Espinel, P. Herrera, P. Jaramillo, I. Medina, S. Mora, A. Santos, C. Toledo, et al. 2008. "Condiciones Agro Socio Econmicas y Ecolgicas de los Diversos Sistemas de Produccin de Arroz d e Pequeo s Pr oductores e n Guayas y Los Ros, Ecuador Unpublished.

PAGE 181

181 Hogset, Heidi. 2005. "Social Network and Technology Adoption Paper presented at the American Agricultural Economics Association Annual Meeting, Providence, Rhode Island Hubbell, Bryan J., Michel e C. Marra, and Gerald A. Carlson. 2000. "Estimating the Demand for a New Technology: Bt Cotton and Insecticide Policies American Journal of Agricultural Economics 82: 118 132. Instituto Nacional de Estadstica y Censos. 2011 "Sistema Agroalimenta rio d el Ecuador http://www.ecuadorencifras.com/sistagroalim/pdf/Arroz.pdf Instituto Nacional de Estadstica y Censos 2012. Ecuador en Cifras http://www.ecuadorencifras.com/cifras inec/main.html (accessed June 10, 2012). Instituto Nacional de Preinversion. 2011. "Enfoques De Preinversion htt p://www.preinversion.gob.ec/pmo/enfoquespreinversion.pdf I nternational Fertilizer Development Center. 2008. Expansion of Urea Deep Plac ement Technology in 80 Upazilas of Bangladesh during Boro 2008: A n Assessment of Project Impact Unpublished Inter national Fertilizer Development Center 2012. Fertilizer Deep Placement (FDP). http://www.ifdc.org/Expertise/Fertilizer/Fertilizer_Deep_Placement_(UDP) (a ccessed March 20, 2012) International Food Policy Research Institute. 2002. "Green Revolution: Curse Or Blessing? http://www.ifpri.org/sites/default/files/pubs/pubs/ib/ib11.pdf International Rice Research Institute. 2012. http://www.irri.org (accessed June 6, 2012 ). Jain, Rajni, Alka Arora, and S.S. Raju. 2009. "A Novel Adoption Index of Selected Agricultural Technologies: Lin kages with Infrastructure and Productivity Agricultural Economics Research Review 22: 109 120. Keelan, Conor, Fiona S. Thorne, Paul Flanagan, Carol Newman, and Ewen Mullins. 2009. "Predicted Willingness of Irish Farmers to Adopt GM Technology The Jou rnal of Biotechnology Management and Economics 12(3&4): 394 403. Kennedy, Peter. 2008. A Guide to Econometrics 6th E d. Malden: Wiley Blackwell. Khanna, Madhu. 2001. "Sequential Adoption of Site Specific Technologies and its Implications for Nitrogen Pr oductivity: A Double Selectivity Model American Journal of Agricultural Economics 83(1): 35 51.

PAGE 182

182 Kristjanson, P., I. Okike, S. Tarawali, B. B. Singh, and V. M. Manyong. 2005. "Farmers' Perceptions of Benefits and Factors Affecting the Adoption of Improv ed Dual Purpose Cowpea in the Dry Savannas of Nigeria Agricultural Economics 32: 195 210. Kropiwnicka, Magdalena. 2005. "Biotechnology a nd Food Security in Developing C ountries : T he Case for Strengthening International Environmental Regimes Journal o n Science and World Affairs 1(1): 45 60. Loh, Lawrence and N. Venkatraman. 1992. "Diffusion of Information Technology Outsourcing: Influence Sources and the Kodak Effect Information Systems Research 3(4): 334 358. Lu, Bao Rong and Allison A. Snow. 2005. "Gene Flow from Genetically Modified Rice and its Environmental Consequences BioScience 55(8): 669 678. Lupin, M S., J. R. Lazo, N. D. Le, and A. F. Little. 1983. Alternative Process for Urea Supergranules Muscle Shoals: International Fertilize r Development Center Ministerio de Agricultura, Ganaderia, Acuacultura and Pesca 2011. El MAGAP, a travs de la Unidad d e Seguro Agrcola se rene con P roductores Arroceros del C antn Daule http://www.magap.gob.ec/mag01/index.php/prensa boletinesprensa/1783 seguro agricola se reune con productores arroceros del canton daule (accessed March 12 2012). Mi nisterio de Agricultura, Ganaderia, Acuacultura and Pesca 2012. Plan Tierras. http://www.magap.gob.ec/mag01/index.php/boletines viceministerio desarrollo rural/122 (accessed June 1, 2012). Ministerio de Inclusion Economica y Social 2012. Programa de Proteccion Social. http://www.pps.gob.ec/PPS/PPS/QuienesSomos.aspx (accessed June 01, 2012) Mora, Samuel and Paul Herrera. 2010. "Comparacin de d os Tecnologas de Aplicacin de Nitrgeno (Urea) en Diferentes Niveles en el Cultivo d e Arroz: A plicacin Profunda de Briquetas de Ur ea y la Aplicacin Tradicional Alv oleo http://www.dspace.espol.edu.ec/bitstream/123456789/11362/10/Resumen%20CIC YT.pdf Ngo c Chi, Truong T. 2008. "Factors Affecting Technology Adoption among Rice Farmers in the Mekong Delta through the Lens of the Local Authorial Managers: An Analysis of Qualitative Data Omonrice 16: 107 112. Nnadi, Fidelia N. and Chidi Nnadi. 2009. "Farm Behaviors of Maize/Cassava Intercrop Technology in Imo State: Lessons for Extension Policy Development World Rural Observations 1(2): 87 92.

PAGE 183

183 Nyanga, Progress H., Fred H Johnsen, Jens B Aune, and Thomson H Kalinda. 2011. Agriculture: Evidence from Zambia Journal of Sustainable Development 4(4): 73 85 Organizacin de las Naciones Unidas para la Agricultura y la Alimentacin, Fondo Internacional de Des arrollo Agrcola, and Programa Mundial de Alimentos. 2002. "La Reduccin de La Pobreza y e l Hambre: La Funcin Fundamental de l a Financiacin de la Alimentacin, l a Agricultura y e l Desarrollo Rural." Paper presented at Conferencia Internacional sobre la F inanciacin para el Desarrollo, Monterrey, Mexico. Pannell David J. 2007. Influences on Technology Adoption in Different Phases http://www.pannelldiscussions.net/2007/06/ Pingali Prabhu and Terri Raney. 2005. "From the Green Revolution to the Gene Revolution: How Will the Poor Fare?" Agricultural and Development Econo mics Division of Food and Agriculture Organization of the United Nations Working Paper 05 09. Qaim, Matin. 2005 "Agricultural Biotechnology A doption in Developing Countries. American Journal of Agricultural Economics 87(5): 1317 1324. Qualls, D. J., Kimberly L. Jensen, Christopher D. Clark, Burton C. English, James A. Larson, and Steven T. Yen. 2012. "Analysis of Factors Affecting Willingness to Produce Switchgrass in the Southeastern United States Biomass & Bioenergy 39: 159 167. Rahman, Sanzidur. 2008. "Determinants of Crop Choices by Bangladeshi Farmers: A Bivariate Probit Analysis Asian Journal of Agri culture and Development 5(1): 29 42. Raju, S. S. and Ramesh Chand. 2008. "Agricultural Insurance in India Problems and Prospects National Centre for Agricultural Economics and Policy Research 8: 1 80. Savant, N. K. and P. J. Stangel. 1990. "Deep Pl acement of Urea Supergranules in Transplanted Rice: Principles and Practices Fertilizer Research 25: 1 83. Savant, N. K., P.S. Ongkingco, I.V. Zarate, F.M. Torrizo, and P.J. Stangel. 1991. "Urea Briquette Applicator for Transplanted Rice Fertilizer R esearch 28: 323 331. Simtowe, Franklin and Manfred Zeller. 2007. "The Impact of Access to Credit on the Adoption of Hybrid Maize in Malawi: An Empirical Test of an Agricultural Household Model Under Credit Market Failure Munich Personal RePEc Archive 45: 1 28.

PAGE 184

184 Singleton, Royce A., Jr., and Bruce C. Straits. 2010. Approaches to Social Research 5th E d. New York: Oxford University Press. Sistema de Informacin Nacional de Agricultura, Ganadera, Acuacultura y Pesca. 2012. III Censo Nacional Agropecuar io. 2012. http://www.magap.gob.ec/sinagap/index.php?option=com_wrapper&view=wrapper &Itemid=400 (accessed June 3, 2012 ). Sistema de Informacin Nacional de Agricultura, Ganadera, Acuacultura y Pesca. 2012. Principales Productos d e Importacion. http://www.magap.gob.ec/sinagap/index.php? option=com_wrapper&view=wrapper &Itemid=100 (accessed May 29, 2012 ). Staal, S. J., I. Baltenweck, M.M. Waithaka, T. DeWolff, and L. Njoroge. 2002. "Location and Uptake: Integrated Household and GIS Analysis of Technology Adoption and Land use, with Applic ation to Smallholder Dairy Farms in Kenya Agricultural Economics 27(3): 295 315. Sunding, David and David Zilberman. 2001. "Chapter 4 T he Agricultura l Innovation Process: Research a nd Technology Adoption in a Changing Ag ricultural Secto r ." In Handbook of Agricultural Economics Vol. 1A, ed. Bruce L. Gardner and Gordon C. Rausser, 207 261. North Holland: Elsevier science Swinton, S. M. and J. Lowenberg deboer. 2001. "Global Adoption of Precision Agriculture T echnologies: Who, when and Why? Paper pr esented at the Third European Conference on Precision Agriculture, Montpellier, France Thompson, Thomas P., and Joaquin Sanabria. 2009. The Division of Labor and Agricultural Innovation in Ba ngladesh: Dimensions of Gender. Muscle Shoals: International Ce nter for Soil Fertility and Agricultural Development Wu, Haitao, Shijun Ding, Sushil Pandey, and Dayun Tao. 2010. "Assessing the Impact being using Propensity Score Matching Analysis in Rural China A sian Economic Journal 24(2): 141 160.

PAGE 185

185 BIOGRAPHICAL SKETCH Jorge Avila was born in the city of Guayaquil, Ecuador. He went to the university Politics and Theory, he received fundamental classes such as Econometrics, Microeconomics, Macroeconomics, Public Economics and Development and Economic d started his professional life as a research assistan t at Rural Research Center and Graduate School of Management of ESPOL. He was engaged in works such as Social Corporate Responsibility and Socioeconomic Analysis. With Dr. Espinel and Dr. Herrera, he Determinants of the Current Trends of Agricultural vent held by the Latin American Council of School Administration (CLADEA, in Spanish). Afterwards, he was included in the Urea Deep Placement Project carried out by ESPOL and U F His function was to promote and understand adop tion process program which was funded by ESPOL, UF, PL 480 of USDA and the Secretaria Nacional de Educacion Superior, Ciencias, Tecnologias e Innovacion While at UF, he complemented his economic studies with two practical and innovative Currently h is research interests are associated with agricultural technology adoption, social networks functionality and biodiversity. Taking advantage from his econometric and GIS knowledge, he expects to combine his research interests during his PhD program which w ould begin in August of 2012.