Cost-Benefit Analysis for the Adoption of the Urea Deep Placement Technology by the Rice Farmers of Daule and Santa Luci...

MISSING IMAGE

Material Information

Title:
Cost-Benefit Analysis for the Adoption of the Urea Deep Placement Technology by the Rice Farmers of Daule and Santa Lucia, Ecuador
Physical Description:
1 online resource (128 p.)
Language:
english
Creator:
Mora Vargas, Samuel Alejandro
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:
Sterns, James A
Committee Co-Chair:
Useche, Maria Del Pilar

Subjects

Subjects / Keywords:
briquettes -- broadcasted -- ecuador -- farmers -- rice -- technology -- udp -- 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:
Ecuador is a rice producing country that has not increased significantly its per hectare production in recent decades. Because of limited development of both technologies and high-yielding varieties, average Ecuadorian production is under the regional average for rice production in Latin America. The need to increase rice yields led some Asian countries to develop and test new technologies. One such technology is the UDP technology; which only recently has been field tested in Ecuador. The technology consists of using urea briquettes, inserted directly into the soil, only once per production cycle. The benefits reported are related to higher rice yields, reduction of urea applications, and generation of new employment. Therefore, this study assessed the costs and benefits of introducing the UDP technology in the rice sector of Ecuador’s Daule and Santa Lucia cantons. This study contrasted different scenarios that were projected to the future to evaluate the benefits/costs over time. The results show that small farmers would benefit the most from the adoption of the UDP technology because survey results suggest that they would adopt the technology on larger proportions of their land and because they currently are the most inefficient in urea fertilization. In addition, the macroeconomic analysis determined that the UDP technology would 1) help rice farmers’ both save money by reducing total quantity of urea used and increase income with higher per hectare yields; 2) help the environment by reducing the amount of urea that would be lost through physical and chemical processes that occur in rice fields; and, 3) help the government by reducing cash outflows spent on imported urea, and possibly also through a reduction/restructurating of urea subsides for rice farmers.
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: Sterns, James A.
Local:
Co-adviser: Useche, Maria Del Pilar.
Statement of Responsibility:
by Samuel Alejandro Mora Vargas.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044808:00001


This item has the following downloads:


Full Text

PAGE 1

1 COST BENEFIT ANALYSIS FOR THE ADOPTION OF THE UREA DEEP PLACEMENT TECHNOLOGY BY THE RICE FARMERS OF DAULE AND SANTA LUCIA, ECUADOR By SAMUEL ALEJANDRO MORA VARGAS 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 Samuel Alejandro Mora Vargas

PAGE 3

3 To my mom and Ecuador

PAGE 4

4 ACKNOWLEDGMENTS First of all I want to thank my mother because sh e has always believed in me and formed my character and perseverance that encouraged me to stay here in this foreign country until my return. I would also like to thank Dr. Sterns because he help ed me to achieved my admission at the Un iversity of Florida, guided me through this entire process and let me know his family while he was working in Ecuador He and Dr. Useche were the main source of knowledge for my thesis research and completion of my project In addition, I want to thank my friends Imelda, Natasha, Olga, and Jorge because I cannot leave out of my thanks Dr. Herrera and Dr. Espinel because they gave me the opportunity to follow my master degree in UF. Finally I want to thank the Government of United States, which t h 480 program supported the first year of my studies; as well as the Government of Ecuado r, which through the SENESCYT supported the last year of my studies.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 16 Problem St atement ................................ ................................ ................................ 16 Objectives ................................ ................................ ................................ ............... 17 Hypothesis ................................ ................................ ................................ .............. 17 Thesis Outline ................................ ................................ ................................ ......... 17 2 LITERATURE REVIEW ................................ ................................ .......................... 19 Rice Production in Ecuador ................................ ................................ .................... 19 The Urea Deep Placement Technology ................................ ................................ .. 21 Diffusion of the UDP Technology ................................ ................................ ..... 25 Urea Deep Placement Technology for Ecuadorian Rice Conditions ................ 26 3 DATA COLLECTION AND METHODS ................................ ................................ ... 33 Methodology ................................ ................................ ................................ ........... 33 Location ................................ ................................ ................................ ............ 33 Data Collection ................................ ................................ ................................ 33 Primary information collection ................................ ................................ .... 34 Secondary information ................................ ................................ ............... 37 Data Analysis ................................ ................................ ................................ ... 38 Baseline scenario ................................ ................................ ....................... 39 Hypothetical s cenario based upon survey data ................................ .......... 40 Future scenarios ................................ ................................ ........................ 40 Macroeconomic analysis ................................ ................................ ............ 41 4 ANALYSIS ................................ ................................ ................................ .............. 44 Construction of the Baseline Scenario ................................ ................................ .... 44 Construction of the Categories ................................ ................................ ......... 44 Analysis of the Production Cost Variables ................................ ........................ 45

PAGE 6

6 Rent cost/ha variable ................................ ................................ ................. 45 Cost of soil preparation/ha variable ................................ ............................ 46 Seed cost/ha variable ................................ ................................ ................ 47 Urea cost/ha variable ................................ ................................ ................. 47 Cost of other fertilizers/ha variable ................................ ............................. 48 Cost of foliar fertilizer/ha variable ................................ ............................... 49 Herbicide cost/ha variable ................................ ................................ .......... 50 Insect icide cost/ha variable ................................ ................................ ........ 51 Seeding cost/ha variable ................................ ................................ ............ 52 Cost of products application/ha variable ................................ .................... 52 Cost of fertilizers application/ha variable ................................ .................... 53 Irrigation cost/ha variable ................................ ................................ ........... 5 4 Harvesting cost/ha variable ................................ ................................ ........ 55 Post Harvest Variables ................................ ................................ ..................... 56 Yield/ha variable ................................ ................................ ........................ 56 Own consumption/ID variable ................................ ................................ .... 57 Income by paddy rice/ha variable ................................ .............................. 58 Income by pulled rice/ha ................................ ................................ ............ 58 Economic Variables ................................ ................................ .......................... 59 Total cost/ha variable ................................ ................................ ................. 59 Total income/ha variable ................................ ................................ ............ 59 Total in come/ha* variable ................................ ................................ ........... 60 Hypothetical Scenario from Survey ................................ ................................ ......... 62 Adoption Rate of the UDP Technology ................................ ............................. 62 Estimating the Money Saved from Reduced Urea/ha Usage by UDP Technology ................................ ................................ ................................ .... 63 Estimating the Cost of Briquette Application per hectare ................................ 65 Estimating Incremental changes in Rice Yield/ha with UDP technology adoption ................................ ................................ ................................ ........ 67 Estimating the Incremental Change in Harvesting Cost with UDP Technology ................................ ................................ ................................ .... 68 Estimating the Extra Income Generated by Adopting the UDP Technology ..... 69 Net Income Generated by the UDP Technology ................................ .............. 70 Adoption Rate for Ex Ante Benefit Cost Analysis ................................ .................... 71 Construction of the Logistic Function ................................ ............................... 71 Spill over Effects of UDP Adoption ................................ ................................ ......... 73 Employment Generation ................................ ................................ ................... 73 Impacts of Reduced Urea Usage ................................ ................................ ..... 74 In crement of Rice Production and Income ................................ ........................ 75 5 CONCLUSIONS ................................ ................................ ................................ ... 121 Summary ................................ ................................ ................................ .............. 121 Politics Implications ................................ ................................ .............................. 123 REFERENCES ................................ ................................ ................................ ............ 125

PAGE 7

7 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 128

PAGE 8

8 LIST OF TABLES Table page 2 1 Planted Area and urea tablet consumption in Java, 1992 1995 ......................... 31 2 2 Results of Urea Technology Research obtained from experimental and demonstration plots conducted in Ecuador, 2010 ................................ ............... 31 2 3 Results of the rice yields from the experimental and demonstration plots with UDP technology conducted in Ecuador, 2010 ................................ .................... 32 3 1 Villages visited by the enumerators in Daule and Santa Lucia ........................... 43 4 1 Categories of the Farmers Land Size Variable ................................ ................... 95 4 2 Production Cost Variables ................................ ................................ .................. 95 4 3 Post Harvest Variables ................................ ................................ ....................... 95 4 4 Economic Variables ................................ ................................ ............................ 96 4 5 Chi .......... 96 4 6 Variable ................................ ................................ ................................ .............. 97 4 7 Chi ........... 97 4 8 Crosstabulation of Farmers' Land Size Variable Seed Cost/ha in 3 Categories ................................ ................................ ................................ .......... 98 4 9 Chi Square ... 98 4 10 Crosstabulation of Farmers' Land Size variable Urea Cost/ ha variable ........... 99 4 11 Chi ............... 99 4 12 Crosstabulation of Farmers' Land Size Variable Cost of other Fertilizers Variable ................................ ................................ ................................ ............ 100 4 13 Chi .................. 100 4 14 Crosstabulation of Farmers' Land Size Variable Cost of Foliar Fertilizers Variable ................................ ................................ ................................ ............ 101 4 15 Chi ............ 101 4 16 Crosstabulation of Farmers' Land Size Variable Herbicide Cost/ha Variable 102

PAGE 9

9 4 17 Chi Square Tests of Insecticide Cost/ha Variable Farmers land Size Variable ................................ ................................ ................................ ............ 102 4 18 Crosstabulation of Farmers' Land Size Variable Insecticide Cost/ha Variable ................................ ................................ ................................ ............ 103 4 19 Chi 103 4 20 Crosstabulation of Farmers' Land Size Variable Seeding Cost/ha Variable ... 104 4 21 Chi Square Test Variable ................................ ................................ ................................ ............ 104 4 22 Crosstabulation of Farmers' Land Size Variable Cost of Products Application Variable ................................ ................................ .......................... 105 4 23 Chi Size Variable ................................ ................................ ................................ .... 105 4 24 Crosstabulation of Farmers' Land Size Variable Cost of Fertilizers Application/ha Variable ................................ ................................ ..................... 106 4 25 Chi .... 106 4 26 Crosstabulation Farmers' Land Size Variable Irrigation Cost/ha Variable ...... 107 4 27 Chi Variable ................................ ................................ ................................ ............ 107 4 28 Crosstabu lation of Farmers' Land Size Variable Harvesting Cost/ha Variable ................................ ................................ ................................ ............ 108 4 29 Chi ize Variable ............. 108 4 30 Crosstabulation of Farmers' Land Size Variable Yield/ha Variable ................ 109 4 31 Chi Square Tests of Own Consumption/ID Variable Farmers Land Size Variable ................................ ................................ ................................ ............ 109 4 32 Crosstabulation of Farmers' Land Size Variable Own Consumption/ha Variable ................................ ................................ ................................ ............ 110 4 33 Chi Square Tests of Income Paddy Rice/ha Variable Farmers Land Size Variable ................................ ................................ ................................ ............ 110 4 34 Crosstabulation of Farmers' Land Size Variable Income by Paddy Rice/ha Variable ................................ ................................ ................................ ............ 111

PAGE 10

10 4 35 Chi Variable ................................ ................................ ................................ ............ 111 4 36 Comparison of Total Income/ha Variable through Farmers Land Size Variable ................................ ................................ ................................ ............ 112 4 37 Chi 112 4 38 Crosstabulation of Farmers' Land Size Variable Total Income Variable ......... 113 4 39 Variable ................................ ................................ ................................ ............ 113 4 40 Chi Variable ................................ ................................ ................................ ............ 114 4 41 Crosstabulation of Farmers' Land Size Variable Total Income/ha* Variable .. 114 4 42 Comparison of UDP AR Variable through U DP Area Variable ......................... 115 4 43 Comparison of MSUR Variable through UDP Area Variable ............................ 115 4 44 Comparison of CBA/ha Variable through UDP Area Adopted .......................... 116 4 45 Comparison of IRY UDP Variable thro ugh UDP Area Variable ........................ 116 4 46 Size Categories ................................ ................................ ................................ 117 4 47 Land Size Categories ................................ ................................ ....................... 117 4 48 Land Size Categories ................................ ................................ ....................... 118 4 49 ............ 118 4 50 Projection of the UDP technology adoption in Daule and Santa Lucia, Ecuador ................................ ................................ ................................ ............ 119 4 51 Total Cost and Benefits of producing rice with the UDP technology in Daule and Santa Lucia, Ecuador ................................ ................................ ................ 119 4 52 Sensitive Analysis of Pessimistic and Optimistic Scenarios ............................. 120

PAGE 11

11 LIST OF FIGURES Figure page 2 1 Monetary differences between the UDP technology and the broadcasted urea technology in Bang l adesh ................................ ................................ .......... 29 2 2 Annual Records of the Ecuadorian rice production represented in th ousand tons and million dollars ................................ ................................ ....................... 29 2 3 Annual records of the Ecuadorian rice production and the annual yields per hectare since 2 000 to 2009 ................................ ................................ ................ 30 2 4 Rice yields (paddy) obtained in Bangladesh using deep point placement of urea briquettes ................................ ................................ ................................ .... 30 3 1 Target Population of the study ................................ ................................ ............ 42 4 1 Histogram of the Rent Cost/ha Variable ................................ ............................. 77 4 2 Histogram of Cost of Soil Preparation/ha Variable ................................ ............. 77 4 3 Histogram of Seed Cost/ha Variable ................................ ................................ .. 78 4 4 Histogram of the Urea Cost/ha Variable ................................ ............................. 78 4 5 Histogram of Cost of Other Fertilizers/ha Variable ................................ ............. 79 4 6 Histogram of Cost of Foliar Fertilizers/ha Variable ................................ ............. 79 4 7 Histogram of Herbicide Cost/ha Variable ................................ ............................ 80 4 8 Histogram of the Insecticide Cost/ha Variable ................................ .................... 80 4 9 Histogram of Seeding Cost /ha Variable ................................ ............................. 81 4 10 Histogram of Cost of Products Application/ha Variable ................................ ...... 81 4 11 Histogram of Cost of Fertilizers Application/ha Variable ................................ ..... 82 4 12 Histogram of Irrigation Cost/ha Variable ................................ ............................. 82 4 13 Histogram of Harvesting Cost/ha Variable ................................ .......................... 83 4 14 Histogram of Yield/ha Variable ................................ ................................ ........... 83 4 15 Histogram of Own Consumption Variable ................................ ........................... 8 4 4 16 Histogram of Income by Rice Paddy/ha Variable ................................ ............... 84

PAGE 12

12 4 17 Histogram of Income by Pulled Rice/ha Variable ................................ ................ 85 4 18 gory ............................. 85 4 19 Histogram of Total Income/ha Variable ................................ .............................. 86 4 20 Histogram of Total Income/ha* Variable ................................ ............................. 86 4 21 Histogram of UDP Adoption Rate (Base Line) Variable ................................ ...... 87 4 22 Histogram of Unit Cost of Urea from Market Variable ................................ ......... 87 4 23 Histogram of MSUR Variable ................................ ................................ .............. 88 4 24 Histogram of Unitary Cost of Hiring a Worker Variable ................................ ....... 88 4 25 Histogram of Cost of Briquette Application Variable ................................ ........... 89 4 26 Histogram of Increment of Rice Yield by U DP Variable ................................ ...... 89 4 27 Histogram of Unitary Cost of Harvesting One Sack Variable .............................. 90 4 28 Histogram of Unitary Price of Paddy Rice Variable ................................ ............ 90 4 29 Histogram of Income that would Increase by UDP Variable ............................... 91 4 30 Histogram of Net Income by UDP Variable ................................ ........................ 91 4 31 Logistic Function of the UDP technology for Ecuador ................................ ........ 92 4 32 Employment Generation in Daule and Santa Lucia by UDP Technology ........... 92 4 33 Impacts of Reduced Urea Usage on the Farmers of Daule and Santa Lucia ..... 93 4 34 Historical Urea Imports of Ecuador ................................ ................................ ..... 93 4 35 Increment of Rice Production and Income of Daule and Santa Lucia ................. 94

PAGE 13

13 LIST OF ABBREVIATION S ha It is a metric unit of area that represents 10,000 square meters IHC It represents the increment al change of harvesting costs caused by the UDP technology mea sured in US dollars INEC Spanish acronym for the National Institute of Survey and Census of Ecuador IRY It represents the increment al change of rice yield obtained with the use of the UDP technology measured in pounds MAGAP Spanish Acronym for the Ministr y of Agriculture of Ecuador MSUR It represents the money saved by urea reduced use of urea associated with the adoption of UDP technology measured in US dollars SACK It represents a unit of 205 pounds for the rice sacks and 50 kilograms for the urea sacks UDP Urea Deep Placement [ Technology ] UDP AR It represents the adoption rate of the UDP technology USD United States Dollar UT Urea Tablets Technology

PAGE 14

14 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 COST BENEFIT ANALYSIS FOR THE ADOPTION OF THE UREA DEEP PLACEMENT TECHNOLOGY BY THE RICE FARMERS OF DAULE AND SANTA LUCIA, ECUADOR By Samuel Alejandro Mora Vargas December 2012 Chair: James Sterns Cochair: Pilar Useche Major: Food and Resource Economics Ecuador is a rice produc ing country that has not increased significantly its per hectare production in recent decades. Because of limited development of both technologies and high yielding varieties average Ecuadorian production is under the region al average for rice production in Latin America The need to incr ease rice yields le d some Asian countries to develop and test new technologies. One such technology is the UDP technology; which only recently has been field tested in Ecuador. The technology consist s of using urea briquettes inserted directly into the so il, only once per production cycle. The benefits reported are related to higher rice yields, reduction of urea applications, and generation of new employment. Therefore, this study assessed the costs and benefits of introducing the UDP technology in the ri ce sector of Daule and Santa Lucia cantons This study contrasted different scenarios that were projected to the future to evaluate the benefits/costs over time. The results show that small farmers would be nefit the most from t he adoption of the UDP technology because survey results suggest that they would adopt the technology on larger proportions of their land and be cause they currently are the most inefficient in urea fertilization. In

PAGE 15

15 addition, the macroeconomic analysis determined that t he UD P technology would 1) help rice farmers both save money by reducing total quantity of urea used and increase income with higher per hectare yields; 2) help the environment by r educ ing the amount of urea that would be lost through physical and chemical pro cesses that occur in rice fields ; and, 3) help the government by reducing cash outflow s spent on imported urea and possibly also through a reduction/res tructurati ng of urea subside s for rice farmers.

PAGE 16

16 CHAPTER 1 INTRODUCTION Problem Statement Ecuadorian ric e production has remained relatively stagnant in recent decades and per hectare productivity remains under the mean for Latin American countries Fertilizer is a critical input in Ecuadorian rice production and most rice farmers broadcast pearled urea to fertilize their rice paddies, even though this method for supplying nitrogen to rice plants leads to losses in effective nitrog en uptake in the rice paddies. Hence, rice farmers are paying in vain for fertilizers that are lost due to physical and chemical processes in the flooded soil of the rice fields. P rilled urea ( small granular urea with a minimum of 46% Nitrogen) is commonly used even though about 60% of the nitrogen content ends up in rivers and the atmosphere as run off and volatilized nitrogen Con sequently, the U rea D eep P lacement (UDP) technology which was originally developed in south Asia, represents a possible alternative to increase Ecuadorian rice yields, and also, to reduce the nitrogen loss associated with broadcasted urea. However, Ecuado r has not yet adopted the UDP technology and there has not been any analysis of the possible benefits and costs to be realized if adoption were to happen. Hence, this ex ante study estimates the potential benefits and costs associated with several scenari os for the introduction and adoption of UDP technology by Ecuadorian rice farmers in one of the primary rice producing regions of the country.

PAGE 17

17 Objectives The UDP technolog y has a lready been tested in field trial plots in Ecuador; but before planning its diffusion it is necessary to assess the benefits and costs that it would generate for small and medium sized rice farmers who represent a large portion of total land holdings in rice production. Therefore, this study proposes an economic analysis of the introduction of this technology in the rice producer cantons of Daule and Santa Lucia. The specific objectives for the microeconomic analysis are t o a sses s the total cost of producing rice with the UDP technology t o a ssess the total income of produci ng ri ce with the UDP technology, and t o d etermine the net income of the UDP technology for each farmers land size category. Additional objectives for a more macroeconomic analysis are t o a ssess the impacts on the demand for rural labor associated with UDP adop tion, t o a ssess the impacts of reduced demand and use of urea that would result from the adoption of the UDP technology, and t o a ssess the impacts of i ncrement al increases in rice production that would be generated by the introduction of the UDP technology Hypothesi s The urea briquettes represent a feasible technology that will have positive net benefits on the economy and environment of the rice farmers from Daule and Santa Lucia, Ecuador. Thesis Outline Chapter 2 provides a literature review of the research context (i.e., rice production and previous research related to the introduction and field testing of the UDP techn ology in Ecuador. In Chapter 3, the data colle ction and research methods used to

PAGE 18

18 conduct a farm level questionnaire of over 400 rice farmers is documented. Chapter 4 is a detailed description and discussion of the analysis completed for this research. Chapter 5 provides a summary of these findings.

PAGE 19

19 CHAPTER 2 LITERATURE REVIEW Rice Production in Ecuador In 2002 Ecuador had 78,814 Agricultural Production Units (APU) of which 343,936 hectares were cultivated with rice Al most 50 % of those lands were owned by small farmers of fewer than 20 hectares. On the other hand, larger farms represented 19.47% of the APUs and farmed half of all Ecuadorian in rice production. Guayas and Los Rios are the provinces most cultivated with rice and combined, these two represent 91.6% of the land cultivated with rice (SINAGAP 2012 ) Even though these two provinces are located in the coastal region they have different production systems M ost of the rice land planted in Los Rios Province is known as low to semi tech dry season rice production since those areas are not flat and there are few irrigation systems in use. These conditions limit rice production to only t he rain y season though the land is then multi cropped with soybean s during in the dry season. In Guayas Province, where the towns of Nobol Daule, Santa Lucia, Palestina, and Yaguachi are where the most rice is planted which takes advantage of the flatter topography that has well developed irrigation systems In these areas, rice is grown year round, both in the winter dry season s (Vitery 2007, p. 145) Based on the Statist ic Division of the FAO rice represents one of t he most produced agricultural commodities of Ecuador S ince 1992, rice was ranked as the fourth most prod uced commodity (F AO 2012 ) Figure 2 1 shows that 2003 was the year when the Ecuadorian rice prod uction achieved its highest level of rice production in the last 20 years. However, the increment al increase in production has not been significant rising only about 21% during the last 10 years. Rice generated 4 28.42 million dollars in

PAGE 20

20 2009. The records of rice production in tons and dollar s are shown in the Figure 2 1 Similarly, the Department of Agriculture of Ecuador ( MAGAP ) estimates that the national produc tion of the last five years has been a little above the 1 million tons per year Even though the Ecuadorian average yield has incre ased since 2003 it still continue s to be under the 4 tons /ha (Figure 2 2) In general, rice production requires a high level of inputs. However, Ecuadorian production costs depend on the production system For instance, the submerged rice system is the mo st expensive because of the extra labor needed to trans plant the rice plants at the beginning of the cycle, as well as the higher amount of fertilizers and pesticides that this system requires (Vitery 2007, p. 145) Th e Ministry of Agriculture of Ecuador in 2011 reported that the direct cost of the submerged system was 1,345 USD per hectare, where fertilization cost ranked as the third highest cost after labor and mechanization. In fact, fertilization costs represent 22.37% of the total cost The fertilization consists of macro nutrients such as Nitrogen, Phosphorus, Potassium Magnesium, Calcium and Sulfur, and also micro nutrients such as Chlorine, Copper, Manganese, Zinc, Boron, Iron, and Molybdenum. Farmers usuall y apply the macro nutrients as granular fertilizers and the micro nutrients as liquid that are applied jointly with the pesticides. Macronutrients are applied at least three times during the crop cycle and they are broadcasted by hand throughout the flood ed field A large number of research reports have demonstrated that the broadcasted application is an inefficient technique because of its high loss rate of nitrogen due to run off, leaching and volatilization

PAGE 21

21 With granular fertilizers, nitrogen is the mo st important nutrient for rice production and the urea is the most used fertilizer because it is composed of 46% of nitrogen. However, the portion of urea actually used by rice plants is very low. Numerous studies have determined that around third fourth s of the nitrogen is lost (Cao, Z.H., S.K. De Datta, and I.R. Fillery 1984, p. 196 203 ; Choudhury, T. M. A., and Y. M. Khanif 20 01, p. 855 871 ; Choudhury, A.T.M.A., and Y.M. Khanif 2011, p. 201 206) Urea may be lost in three different ways The first is ammonia volatilization that occurs when the broadcasted urea is hydrolyzed to ammonium + aqueous ammonia in the floodwater. The leaching urea occurs when the water percolation rates are high; in the same way, urea may be lost by runoff (Cho 2003, p. 43 52) The nitrogen inefficienci es potentially have both environmental and health impacts. Product s of the volatilization and denitrifrication of urea include the gases nitrous oxide, nitric oxide, and ammonia causing atmospheric pollution (Azam, F., C. Mller, A. Weiske, G. Benckiser, and J. Ottow 2002, p. 54 61) Also, Reeves et al. ( 2002 ) determined that deposition of nitric oxide and ammonia in terrestrial and aquatic ecosystems can lead to eutrophication, shifts in species diversity, and effects on predators and parasite systems (Reeves, T.G., S.R. Waddington, I. Ortiz Monast erio, M. Banziger, and K. Cassaday 2002, p. 11) On the other hand, leaching of nitrate results in drinking water and food with elevated levels of nitrates, which may cause methemoglobinemia in infants, respiratory illness, and decreased content of vita min A in the liver (Phupaibul, P., N. Chinoim, and M. Toru 2002, p. 295 302) The Urea Deep Placement Technology Facing high fertilization inefficiencies, the Urea Deep Placement (UDP) techn ology was developed in response to the high rates of nitrogen loss associated with

PAGE 22

22 broadcasted application. Many researchers began to study this technology for rice production and concluded that the deep placement of super granules is an effective technolo gy to lower nitrogen losses as well as increase rice yields. Th ose conclusions were supported by the International Network on Soil Fertility and Fertilizer Evaluation for Rice (INSFFER) which evaluated the UDP technology during 1975 1978. (De Datta, S.K., and E T. Craswell 1980, p. 283 316) Similarly, field experiment results by the was higher when urea was pla ced at 10 cm soil depth (De Datta 1981 ) That was also reported by the Bangladesh Rice Research Institute (BRRI) which applied urea briquettes with an instrument called a (Choudhury, ATMA, and N. Bhuiyan 1994, p. 104 107) In spite of the se early positive results the UDP technology was not significantly diffused in the Asian countries where it was prove n effective because of (a) n onavailability of properly sized (weight) Urea Briquette fertilizer material at an affordable price in local markets, and (b) extra labor required for its hand deep placement after transplanting using regular hill spacings (Savant, N. and P. Stangel 1998, p. 85 94) Over time, with modifications and advances in the adaptation of the general concept of placing urea directly into the soil, UDP technolog ies have been widely adopted by small scale rice farmer s in Bangladesh, Vietnam, Cambodia, India and other Asian countries. The urea briquettes are made using a roller mill machine with indented pocket rollers that compress the granular urea to create the briquettes. On farms, once the rice plants are transpl anted, the briquettes are inserted into the puddled soil by hand in the

PAGE 23

23 middle of four plants These applications are only needed once during the crop cycle shortly after transplanting (De Datta, S.K ., and R.J. Buresh 1989, p. 143 169) In seven districts of Bangladesh Bogra, Chandpur, Jessore, Kishoreganj, Mymensingh, Pabna, and Tangail 531 on farm trials were conducted to evaluate the UDP technology compared to the traditional way of broadcast ing urea fertiliz er, Practice (Bowen, W., R.B. Diamond, U. Singh, and T.P. Thompson 2005, p. 369 372) The se trials measured rice yields through side by sid e comparisons between the two technologies. The results were evaluated for both the Boro season (raining season) and the Aman season (dry season). That research was able to prove that UDP technology improved rice yields on the 531 o n farm trials along 2000 2004 (Figure 2 3). On 1 (SE = 32.4) during the Boro season and 890 kg grain ha 1 (SE = 32.5) during the n saved by the new technology, which was 70 kg N ha 1 (SE = 2.4) during the Boro season and 35 kg N ha 1 (SE =9.1) during the Aman. In 2008 Bangladesh researchers reported estimates of economic and social impacts of the UDP technology. The main benefits are related to Net Incremental Farm Income, National Benefits and Government savings, and Employment Generation from Urea Super Granules (USG) Business (IFDC 2008 ) All of these assessments were measured through a sample de sign that surveyed 3,230 farmers who represented 3% of the new UDP users in Boro in 2008. The results show that there is a positive increment of farm income. However, the UDP technology generated higher production costs that are mainly caused by increased need for human labor On the other hand, the rice

PAGE 24

24 production with briquettes required fewer amounts of the rest of fertilizers than the broadcasted urea technique; only Zinc fertilizers were used in more proportions in the UDP production. Figure 2 4 shows the difference of cost between the UDP technology and broadcasted urea. There, the negative values of each component mean that those resources were less expensive for the rice production with UDP technology. The analysis shows that at the end of the cycle, the UDP technology is, on average, 20.2 USD more expensive than the broadcasted urea. However, there is a significant difference in the net return between the two technologies because the farmers surveyed gained, on average, 265 USD extra by usi ng the UDP technology. The Cost/ Benefit Ratio of the UDP technology was 1.53 and of the broadcasted urea was 1.33 (IFDC 20 08 ) The national benefits that the UDP technology had in the 80 Upazilas of Bangladesh w ere given by: 1) the increment of rice supply, 2) the reduction in the quantity of urea imports, 3) the reducti on of the foreign exchange spent on urea imports and 3) generation of new employment. The results show that farmers saved, on average, 93.0 kg of urea per hectare. C onsequently, there was 14,000 tons of urea saved by using the UDP technology which reduced the budgetary burden of the Government of Bangladesh to USD 6.0 million, and, it also made foreign exchange decrease to USD 7.0 million due to reduction of urea imp orts On the other hand, the increment al increase in rice production generated 42.3 million dollars in the rural net T he rice yields increased on average, 0.74 tons per hectare. Consequently, benefited farme increased by

PAGE 25

25 113,310 tons in Boro 2008 The employment of the rural sector was incentivized because, on average, one hectare of rice that was fertilized with urea briquettes needed an additional 9.5 person days of labor for transplanting and deep placement of the briquettes (IFDC 2008 ) Consequently, 1.43 million person days of rural employment w ere generated in the Boro in 2008. Additionally, the increasing demand of urea briquettes encouraged the creation of new business es that constructed the roller mill machines used to make urea briquettes In addition, greater rural employment was observed because new businesses were created for retail stores that manufactured and sold the urea briquettes. Two hundred twe nty nine briquette machine owners were surveyed in order to determine the production and profitability of the briquette machines. The results show that, on average, one diesel operated machine produced 58.95 tons of briquettes and an electric driven machin e produced 80.95 tons of briquettes during the Boro 2008 rice growing season The net profits for briquette machine owners were USD 585.00 for diesel machines and USD 1,040 for electric machines. The difference is due to the diesel costs. Jointly, the briq uettes manufacturing also generated the employment of 754 persons. Diffusion of the UDP T echnology The Urea tablets technology was introduced in the Java Island of Indonesia in 1992 where researchers determined that it saved, on average, 25% of N fertilizer rate and also increased, on average, 400 Kg ha in rice yield (Pasandaran, E., B. Gulton, J. Sri Adimingsih, H. Apasari, and S ri Rochayati 1999, p. 113 119) This study measured how the urea tablets technology was covering the areas de signated for rice production from 1992 to 1995; also, these researchers measured the percentage of urea used to produce urea tablets from the total urea consumed. The results show that area planted

PAGE 26

26 with urea tablets was less than 0.3% at the beginning of the diffusion and then increased by 11%, 7%, and 14% in West Java, Central Java, and East Java; respectively for the period 1992/1993. In subsequent years, the ado ption rates increased significantly until 1995 when it was reported that West Java adopted 77% of their planted area. However, Central Java and East Java decreased their planted area with urea tablets since 1993/1994 period to 1995 period. The same happene d for the percentage of urea used to made urea tablets. The adoption rates and percentage of urea used are expressed in Table 2 1 Urea Deep P lacement Technology for Ecuadorian R ice C onditions The Escuela Superior Polit cnica del Litoral (ESPOL), an Ecuado rian University, jointly with The University of Florida carried out a project to introduce the UDP technology in Ecuador. That project was supported by the PL 480 of the U .S. Department of Agriculture. The first experiment was done on experimental plots on the main campus of ESPOL located on the outskirts of Guayaquil. It was a high ly controlled experiment where dif ferent briquette sizes were tested. Results included a key finding that the 3.6 gr am briquettes which provided 179 Kg of urea per hectare wer e the best treatment. Also it demonstrated that the UDP t echnology increases rice yields despite researchers applying less urea to the UDP rice paddy than the broadcasted urea treatment paddies In addition, soil analysis indic a ted that the urea briquette kept nitrogen for a ( Calle, O., and I. Medina 2010 ) The profitability analysis showed that the UDP technology was more profitable (generatin g net profits of USD 401.65 more than the broadcasted urea ), mainly because it saved 33% of the urea applied, and, increased the quality of the rice grain, which made it more valuable (Mora. S., and P. Herre ra 2010 )

PAGE 27

27 Based on that finding, eight demonstration plots were implemented to test the feasibility of the UDP technology in farm conditions The demonstration plots were conducted in the villages of Alianza Definitiva, Brisas de Daule, Huachichal, La R inconada, and Cooperativa 25 de Abril which are villages in the Daule Canton of the Guayas Province. There, o nly the fertilization techniques were modified in order to keep constant all other inputs and production practices, plus allowing for the technolo gy to be tested under the management practices of actual rice farmers. The results determined that UDP technology saved, on avera ge, 30.14% of the urea applied (Table 2 2); also, expressed very high levels of interest and a strong desire to co ntinu e using the UDP technology in their rice paddies (Mayorga, J., and P. Herrera 2010 ; Saenz, C., and P. Herrera 2010 ; Ba rzola, L., and P. Herrera 2010 ) Th e re was a last experimental plot that was conducted in Los Rios Province where the UDP technology saved 31.54 % of the urea applied in cont rast to the broadcasted urea technology (A guirre, D., and I. Medina 2010 ) The experimental plots and demonstration plots allow ed this study t o determine that the UDP technology increased, on average, 14.99% of the rice yields (T able 2 3). On the other hand, the demonstration plots done in Daule determined that 20 worker hours were necessary to insert the urea briquettes in one hectare. The rice workers usually begin their labors at 4 hours per day, beginning at 6:00 am working until 10:00 am. T he benefit of the UDP technology in terms of reduced environmental impacts of fertilizers use was also addressed The rice production of Daule and surrounding zones are connected by t he Daule River which jointly with The Babahoyo River creates The Guayas basin that has 32,024 Km 2 and represents the most important sou rce of fresh

PAGE 28

28 water in the zone. A tmospheric deposition, fertilizers and disposal of waste water are the principal sources for increased carrying loads of in puts of nitrogen and phosphorus in these rivers (Valiela, I., and J. Bowen 2002, p. 239 248) Specifically for the Guayas Estuary, a griculture and natural processes are the most significant sources for the amount of dissolved inorganic nitrogen (DIN) that are caused by inefficient fertilize rs use and poor runoff control that represented 96% of the nitrogen input and 65% of the phosphorus input, followed by city discharges and aquaculture (Borbor 2004, ) Therefore, the UDP technology would help to decrease the amount of nitrogen loading of the rice zones where it is adopted.

PAGE 29

29 Figure 2 1. Monetary differences between the UDP technology and the broadcasted urea technology in Bang l adesh Placement (UDP) Technology in 80 Upazilas of Bangladesh during Boro ; Units are in US Dollars. IFDC, 2008 Figure 2 2 Annual Records of the Ecuadorian rice production represented in thousand tons and million dol lars. FAOSTAT, Top 20 commodities by country

PAGE 30

30 F igure 2 3 Annual records of the Ecuadorian rice production and the annual yields per hectare since 2000 to 2009. Third Agricultural Census of Ecuador, 2000; and MAGAP, 2011 Figure 2 4 Rice yields (paddy) obtained in Bangladesh using deep point placement of urea briquettes (UDP) plotted against yields obtained using broadcast urea by 04. Source: Bowen, W. et al 2005)

PAGE 31

31 Table 2 1. Planted Area and urea tablet consumption in Java, 1992 1995 (Source: Pasandaran et al 1999) Year West Java Central Java East Java PA UT PA UT PA UT 1992 0.3% 0.4% 0.2% 0.6% 0.2% 0.5% 1992/1993 11.0% 9.0% 7.0% 7.0% 10.0% 15.0% 1993 27.0% 24.0% 24.0% 24.0% 14.0% 16.0% 1993/1994 58.0% 47.0% 50.0% 50.0% 62.0% 67.0% 1995 77.0% 69.0% 49.0% 49.0% 50.0% 54.0% PA: planted area covered by urea tablets (% of the total planted area), UT=urea tablet (%of total urea consumed) Table 2 2 Results of Urea Technology Research obtained from experimental an d demonstration plots conducted in Ecuador, 2010 Experiment carried out by: Location UDP* (Kg) BU ** (Kg) Diff*** Samuel Mora 2010 Guayaquil, Guayas, Ecuador 174.75 260.87 33.01% Cesar Saenz, 2010 Daule, Guayas, Ecuador 180 250 28.00% Cesar Saenz, 2010 Daule, Guayas, Ecuador 180 200 10.00% Cesar Saenz, 2010 Daule, Guayas, Ecuador 180 300 40.00% Joy Mayorga 2010 Daule, Guayas, Ecuador 180 250 28.00% Joy Mayorga, 2010 Daule, Guayas, Ecuador 180 350 48.57% Luis Barzola, 2010 Daule, Guayas, Ecuador 180 200 10.00% Luis Barzola, 2010 Daule, Guayas, Ecuador 180 250 28.00% Luis Barzola, 2010 Daule, Guayas, Ecuador 180 350 48.57% David Aguirre, 2010 San Juan, Los Ros, Ecuador 178 260 31.54% Mean 179.28 267.09 30.57% Standard Deviation 1.71 52.49 13.37%

PAGE 32

32 Table 2 3 Results of the rice yields from the experimental and demonstration plots with UDP technology conducted in Ecuador, 2010 Experiment carried Location UDP* BU** IF** out by: (Kg) (Kg) (%) Samuel Mora 2010 Guayaquil, Guayas, Ecuador 7161.95 6720.27 6.57% Cesar Saenz, 2010 Daule, Guayas, Ecuador 5283.31 5292.62 0.18% Cesar Saenz, 2010 Daule, Guayas, Ecuador 10566.61 8600.51 22.86% Cesar Saenz, 2010 Daule, Guayas, Ecuador 10038.28 7396.63 35.71% Joy Mayorga 2010 Daule, Guayas, Ecuador 7568.08 6948.43 8.92% Joy Mayorga, 2010 Daule, Guayas, Ecuador 6294.31 4029.10 56.22% Luis Barzola, 2010 Daule, Guayas, Ecuador 10192.23 10101.84 0.89% Luis Barzola, 2010 Daule, Guayas, Ecuador 9318.18 8603.48 8.31% Luis Barzola, 2010 Daule, Guayas, Ecuador 9900.57 9816.70 0.85% David Aguirre, 2010 San Juan, Los Ros, Ecuador 7744.34 7059.45 9.70% Mean 8406.79 7456.90 14.99% Standard Deviation 1746.05 1805.09 13.37% *UDP represents the yields in kilograms (Kg) obtained by the UDP technology. **Bu represents the yields in kilograms (Kg) obtained by the Broadcasted Urea technology. relation

PAGE 33

33 CHAPTER 3 DATA COLLECTION AND METHODS This study is based on the comparison between the Ecuadorian rice producers current situation against hypo thetical scenarios; where the UDP technology is adopted by rice farmers. Therefore, the first part of the methodology is about collecting all information through a sampling design that allowed us to set a baseline scenario that captures the current reality of the farmers that are producing rice in the zones located along Daule River. The hypothetical scenarios will project the baseline scenario in order to assess the impacts of the UDP tech nology on s S econdary information about the UDP technology which was generated by previous research in Ecuador and some Asian countries also was used to supplement the analysis Methodology Location This study is focused on the rice farmers of the Daule Canton and Santa Lucia Canton. The sampling design, detailed in the next section, led the research team to identify 28 villages in Daule and 15 villages in Santa Lucia (Figure 3 1 ) Data Collection The data collection is divided in to two phases. The firs t consist ed of the collection of the primary information through a sampling design that used questionnaires that were implemented in the identified villages. The second is the compilation of research findings from earlier work, particularly research focuse d on the adoption and diffusion of the UDP technology in Indonesia and t he preliminary introduction and evaluation of the UDP technology in Ecuador.

PAGE 34

34 Primary information c ollection The primary information collection and analysis consisted of the selection of the sampling design, construction of the instrument, surveyors training, and implementation of the surveys. Sampling Design The sampling design selected is a non random sampling technique called snowball sampling. The reason why this method was select ed instead of randomized sampling is because the target population the rice farme rs of Daule and Santa Lucia is not registered in a sample frame; consequently, there was no means for creating a randomized selection of the units of study (i.e., persons w ho m ake production decision s for rice farms in the targeted region) T he characteristics that made the snowball sampling the best for this study were: It allow ed us to pre select the villages where the re are large numbers of rice farmers The farmers surveyed can suggest the villages or houses where other rice farmers live. At the moment when survey enumerators go to the selected villages they can rapidly determine if farmers are willing to participate in the study, and if not, enumerators can move to other villages without losing time. Instrument. The instrument used for the data collection is a questionnaire which was 9 page s long and divided into nine topical sections. Most of the questions were close ended with premeditated answer options that fac However, there were also open ended questions that provided respondents to follow up their quantitative responses with additional commentary The analysis for this thesis focuses on the data from S ection s 3 and 4 of the que stionnaire plus question 49 of S ection 8. All sections are described in the following paragraphs.

PAGE 35

35 S ection 1 was entitled purpose of this section was to measure the degree of diffusion that UDP technol ogy had in the last two years. It had seven questions designed to obtain information about how many different ways farmers knew (if that is the case) of the UDP technology and also about their level of knowledge about it. S ection 2 was en titled and contained questions designed to understand the social connections between farmers. The questions were related to their participation in farmer associations, cooperatives and any other group that they or any other family member may belong. This section also collect ed information about the level of communication between farmer neighbors, friends, and members of the ir groups. S ection 3 was en titled This section measured the ways farmers share their agricultural knowledge. Also, this section measured the desire to adopt the UDP technology. There were two questions that asked farmers if they would use the UDP technology based on the advantage s and disadvantages th at were read to the respondents by the enumerators. F armers also were asked if they would pay a pre determined extra amount of money for the urea briquettes since the manufacturing of urea briquettes will likely add a slight per kilo price premium over th e price of prilled urea traditionally used when broadcasting fertilizer in rice paddies There were three different prices that were proposed pegged to current prilled urea price s plus a range of price premiums for the briquettes Also, there is a questio willingness to pay more for urea briquettes due to positive environmental benefits that the UDP technology generates.

PAGE 36

36 S ection 4 was en and it was designed to collect information related to the producti on cost s of cultivating rice and the income that it generates. T his section had additional questions for the other crops (if the respondents indicated that they grew other crops) that were classified by short term crops and long term crops. S ection 5 was e n and had questions to determine how the farmers usually obtain the money and/or loans for the ir farm and non farm activities. This section also collected information to determine the cash flows associated with informal credit markets and illegal loans that have to be paid in short time periods with high interest rates. It is known that this loan type is frequently used in the Ecuadorian rural sector by small scale farmers and their families. S ection 6 was en and was focused upon farm and non farm activities that family members do. It also asked about the time that each activity required and the income it generated. There also were questions that make reference to the external money that the household may rece ive from donations, government help, lease s economic remittances, or other sources of outside money. S ection 7 was en title d was about how farmers obtain water to drink as well as the cost that this involves. It also ha d questions about the p erception of the causes of their illness and the ir costs. S ection 8 was en was about the material goods and basic services in the s It also had questions about the household expenses such as water electricity, food, and clothing At the end of this

PAGE 37

37 section there were the El Nio Southern Oscillation (ENSO) or any other flood ing in recent years. Section 9, t he last section was e n questions about personal information of the respondent, such as gender marital status, education and age. At the end of the questionnaire, there was a chart that the enumerator had to fill out with information about the interview location including a GPS code to identify the exact location of the surveyed household. Enumerator training. The enumerators selected for the research project were five undergraduate students enrolled at the State University of Guayaquil; three of the se also studied at which is located in Daule. The training was conducted by two graduate students from the University of Florida as part of their thesis research. The training consiste d of a review of all sections of the questionnaire completion of practice surveys in order to clarify misunderstandings as well as to determine that fill ing out a questionnaire required between 45 to 60 minutes. Implementation of the survey. T he goal of th e study was to obtain at least four hundred completed questionnaires. However, the project team could not pre determine the number of farmers per village that would be surveyed due to the characteristics of the snowball sampling design. Table 3 1 shows t he villages of Daule and Santa Lucia that were visited by the su rveyors. Secondary i nformation S econdary data for this study consist ed of information drawn from previous UDP studies These included,

PAGE 38

38 The National Institute of Statistics and Census of Ecuador (INEC by its Spanish acronym). Reports of Annual Imports from the Central Bank of Ecuador (BCE by its Spanish acronym) Survey of Agricultural Land and Continuous Agricultural Produ ction of Ecuador (ESPAC by its Spanish acronym) Effects of Land Use Change and Agriculture on the Discharges of Nutrients from the Guayas Basin, Borbor 2002 All research done by the ESPOL University through the Center of Rural Research (CIR by its Spanish acronym). There were 3 UDP experimental plots and 7 UDP demonstratio n plots done in 2010. The adoption rates that the technology of urea tablet ( a kind of UDP) had in Ind onesia (E. Pasandaran et al 1999). Data Analysis The data analysis begins with the construction of a baseline scenario of likely adoption rates and then a ssesses the potential diffusion of the U DP technology into the future. Thus, the baseline scenario is contrasted to a range of hypothetical scenarios that consider a range of projected adoption rates for the UDP technology by the target populat ion of Ecuad orian rice farmers. The se scenarios are analyzed through time ; specifically, all scenarios are projecte d from the baseline scenario for the next 10 years in order to estimate, a priori, the potential long term benefits/costs of the introduction, adoption a region Different adoption rates are considered. For e xample, the baseline scenario that technology with the adoption rates that reflect their intentions to adopt as indicated by their answer s in the survey of Ecuadorian rice farmers conducted for this thesis But, because of the extremely high levels of adoption rates that are expressed in these

PAGE 39

39 survey responses, this baseline scenario was not projected into the future for the next ten years A more conservative (and more likely) estimated adoption rate was assumed for the bene fit/cost analysis projections. These future scenarios used a standard model determined by a logistic function fitted to the empirically observed adoption rates of a similar technology and farming system that researchers had evaluated i n Indonesia (E. Pasandaran et al 1999). Baseline s cenario The baseline scenario shows the cur rent situation of the rice farmers surveyed. To do that, all data were standardized to U.S. dollars per hectare (USD/ha) in order to make the analysis comparable between farmers Also, with the objective of identifying the heterogeneity between the farmers of different land size the data were divided in to five farm land size categories and then the analysis compared findings across these five categories of respondents. For the construction of the baseline scenario the following variables were compiled fr om survey results : P roduction cost variables: Rent Cost, Cost of Soil Preparation, Seed Cost, Urea Cost, Cost of other Fertilizers, Cost of Foliar Fertilizers, Herbicide Cost, Insecticide Cost, Seeding Cost, Cost of Chemical Applications, Cost of Fertiliz er Applications, Irrigation Cost, and Harvesting Cost. P ost harvest variables : Yield/ha, Own Household Consumption of harvested rice Income generated from the production of Padd y Rice and Income generated from the production of Pulled Rice 1 E conomic variables: Total Cost, Tot al Income, Total Income* 2 1 Pulled rice means the rice grain that was removed the rice husk.

PAGE 40

40 Hypothetical scenario based upon s urvey data This hypothetical scenario measures the on farm financial costs and benefits of the UDP technology for the surveyed rice farm ers for a single growing cycle. T he analysis simply estimates the financial returns associated with partial adoption of the UDP technology, benchmarked to the financial returns realized using the traditional practices of broadcasting urea. To construct this scenario, the UDP adoption rate of each farmer is estimated such that farmers adopt the technology incrementally by using the UDP technology on only a portion of their total land in rice cultivation Also, this scenario only considers the financial variables that are di rectly affected by the UDP technology and how their values change incrementally with the adoption of the new technology. Specifically, t he variables are : Money Saved by Reducing the total amount of Urea applied to the field Added labor Cost of Briquette A pplications Increment al increase of Rice Yield when using UDP Increment al increase of Harvesting Cost when using UDP (higher yields have higher harvesting costs) Income generated by UDP Net Income generated by UDP Future s cenarios For the construct ion of the future scenarios which will be the basis for an a priori economic analysis of the economic costs and benefits of the UDP technology, a logistic function is assumed as the appropriate expression of the diffusion path for UDP technology in Ec e producing region. The logistic function is fitted using adoption rate data from the Urea Tablets (UT) Technology introduced and adopted in Indonesia (Pas andaran, E., B. Gulton, J. Sri Adimingsih, H. Apasari, and S ri Rochayati 2 Total Income* represents the income that would be generated if rice farmers had sold the whole production as paddy rice.

PAGE 41

41 1999, p. 113 119) Using the logistic function to model the adoption rate for the UDP technology, net benefits associated with this technology are projected for the next 10 years. Macroeconomic a nalysis The objective of the macroeconomic analysis is to extrapolating the findings from the survey and baseline scenario for all rice farmers in the cantons of Daule and Santa Lucia Additional variables considered with this macroeconomic analysis are : The impacts of the i ncrement al change in rice production that would be generated by the introduction of the UDP technology in Daule and Santa Lucia. The g eneration of new employment for farm laborers needed for briquette a pplications in the rice fields where the UDP technology would be used. The impacts of the urea saved by the UDP technology

PAGE 42

42 Figure 3 1 Target Population of the study

PAGE 43

43 Table 3 1 Villages visited by the enumerators in Daule and Santa Lucia Villages Names of Daule Canton Bella Esperanza (14) Jigual Los Quemados Porvenir Brisas de Daule La Aurora Naupe Rebeldia Clarisa La Elvira Pajonal Rio Perdido El Limonal Las Maravillas Patria Nueva San Gabriel Flor de Maria Loma de Papayo Pajonal San Vicente Huanchichal Los Almendros Peninsula de Animas Villa Filadelfia Jesus del Gran Poder Los Moranillos Pial Yurima Villages of Santa Lucia Canton Barbasco El Encanto La Carmela Playones Barranquilla El Limn La Fortuna San Jacinto Bermejo El Mate Marcela San Pablo Coop 14 de Octubre El Porvenir Pajonal

PAGE 44

44 CHAPTER 4 ANALYSIS Construction of the Base line Scenario The baseline scenario represents the reality of the rice farmers surveye d in terms of the financial costs and benefits associate d with the cultivation of rice. It in cludes the production costs generated during rice production as well as, the income generated by selling the harvest ed crop In order to test for interaction s size and farm level cash flows the data were divided in to farm size categ orie s which will allow for a disaggregated analysis using Cross Tables and the Chi Square Test. These analytical tools are used to test for correlation s between the target variables, which will be shown below and the categories which are obtai ned for a percentiles analysis. The results obtained in this section are the base line for the analysis of the future scenarios. Construction of the C ategories ize V ariable was recoded and arranged in to five ca tegories. Each category has the same number of valid values. In other words, the whole population of respondents was divided in to 5 percentiles of 20% of respondents per category, grouped by the size of their farm land holdings Based on the percentile ana ly sis in Table 4 1, which shows that the Categor y 1 i s composed of farmers with very small land plots of 0.04 0.52 ha; C ategory 2 farmers cultivate lands of 0.52 1 ha; Category 3 farmers cultivate lands of 1 1.5 ha; Category 4 farmers cultivate lands of 1.5 3.19 ha., and Category 5 farmers cultivate lands of 3.19 16.33 ha A similar procedure was done for the rest of the variables from the questionnaire, although all other variables are

PAGE 45

45 recoded and arranged in to sets of 3 categories which represent the low, medium, and high values for each variable (and thus these are grouped by defined population percentiles of 33.3 percent per c ategory) Table 4 2, Table 4 3, and Table 4 4 show the values that were used for the construction of the new categorized variables. For all variables the minimum and 33.33 percentile value s were used to construct the first category. Then 33.33 percentile value s up to the 66.66 percentile value s were used to construct the second category, and finally the 66.6 7 percentile value s and maximum value s were used to construct the third category. Analysis of the Production Cost Variables The production cost varia bles are composed of inputs related to renting, soil preparation, herbicides, insecticides, fertilizers, labor, and harvesting costs. Each variable is order to determine if there exists a correlation between them. Rent c ost/ha v ariable The Rent Cost/ha variable had 384 valid observations out of a raw count of 401 interviews conducted by the research team In m ost cases farmers did not pay any monetary value for renting land. In fact, there were 332 far mers who said that they paid zero dollars for land rental Only 52 farmers had to pay from 59.15 USD/ha up 528.16 USD/ha as is illustrated in the Figure 4 1 histogram T he mean value of rent paid by farmers is $ 248.42 USD with a st andard deviation of $ 127.26 USD/ha. Due to the high number of observations with zero values, this vari able was not considered for cross tab analysis

PAGE 46

46 Cost of s oil p reparation /ha v ariable The Cost of Soil Preparation/ha variable ha d 370 valid observ ations with a mean value of 199.49 USD/ha and standard deviation of 195.03 USD/ha. Its minimal value was 8 USD /ha and its maximum was 1250 USD/ha However, t he distribution analysis (Figure 4 2) shows that most of the farmers (90%) paid up to 400 USD/ha. T he Chi Square Test (Table 4 5) indicates that the C ost of Soil Preparation /ha variable is correlated to the categorized land size variable (at the level of significance of 0.05). Therefore, t he changes in the costs of soil preparation and changes of land size are related to each other. T he crosstab analysis (Table 4 6 ) indicates that most of the smallest farmers (row 1) are arranged in column 5 which represents the highest Cost of Soil Preparation/ha. In fact, the ir observed value (26) is higher than the expe cted value (15.8), which suggests that those smallest farmers ha ve a greater incidence o f high cost s of soil preparation/ha than would be expected by chan ce alone. This is supported statistically, as their Stan dard Residual value (2.6) was higher than the Z critical value (1.96) On the other hand, m any of the largest farmers (listed on row 5 ) had the lowest cost of soil preparation because most of them were arranged in the first and second column s of the crosstab table These findings suggest that farmers had who had larger land size also reported lower per hectare soil preparation costs. That hypothesis is demonstrated through the highest residual values of the farmers of the row 1 row 2 row 3, row 4 and row 5 ; that were arranged in the colu mn 5, 4 2, 1, and 1; respectively. A nd jointly with their Standard Residual values that are greater than the Z critical value (1.96).

PAGE 47

47 Seed c ost/ha v ariable The variable Seed Cost/ha had 132 missing cases which limit ed the analysis to only 269 valid cases The mean value of this v ariable was 61.24 USD/ha. The distribution analysis s hows that most of the farmers surveyed (88.5%) paid less than 100 USD/ha (Figure 4 3) The statistic of the Chi Square Test (Table 4 7 ) supports the hypothesis that the categorized Seed Cost/ha variable is cor related with the categorized F Land S ize variable bec ause it (0.000) was smalle r than the level of significance of 0.05 indicating a need for crosstab analysis to further test for the gories and The crosstab analysis (Table 4 8 ) shows that most of the smallest farmers (row 1) had the highest seed cost values and that per hectare seed costs decrease d as land size increase d This finding is supp orted by the highest residual value (11.8) in row 1 ( the smallest farmers) paired with the column of the highest co s ts (column 3). Also, the Standard Residual value support s the assertion becaus e it (3.5) is greater than the Z critical value (1.96). And, f or the farmers grouped in the row s 2, 3, 4 and 5 their highest residual values are in the columns 2, 1, 1, and 1 respectively. However, the assumption for the farmers of the row 3 is not well supported becaus e their Standard R esi dual value (0.9 ) was smal ler than the Z critical value (1.96). Urea c ost/ha v ariable The Urea Cost/ha variable had 347 valid values and 54 missing values. The descriptive analysis indicates that farmers spent, on average, 159.62 USD/ha to buy urea and they had a standard deviation of 101.31 USD/ha. The distribution analysis (Figure 4 4) of this variable shows that m ost of the farmers (80.1%) had to pay from 7.68 USD/ha up to 234.74 USD/ha. The highest amount paid was 621.42 USD/ha.

PAGE 48

48 The results of the Chi Square Test demo nstrated that the Urea Cost/ha v ariable is related to the categorized ize variable because it s statistic (0.000) is lower than the p value (0.05) as is illustrated in the Table 4 9 Therefore, a crosstab analysis is justified to further test the relation between the urea cost/ha variable and the The crosstab analysis (Table 4 10 ) shows that most of the smaller farmers (the first two categories i.e., row s 1 and 2) ha ve the highest cost by purchasing urea on a per hectare basis Th eir Standard Residual values, 3.3 and 1 .7 respectively, are higher than the Z critical value (1.96), supporting the assumption that they have a higher incidence tha n would be expected by chance alone. On the other hand, the farme rs arranged in the row s 3 and 4 had lower urea cost than the smaller farmers in the first two categories The highest residual va lues of these two categories were arranged in the column 2 and column 1 respectively B ut only the farmers in row 4 are well supported to say that there are more farmers in that column than would be expected because their standard residual value (2.0) is greater than the z critical value This last assumption is supported by its Standard Residual value (2.3) that was g reater than the Z critical value (1.96 ) Again, the farmers with the highest land size (i.e., farmers grouped in row 5 ) have the lowest cost of urea with most of them sorted into C olumn 1. Their Residual value ( 9.1 ) and Standard Residual (1.9 ) support the assumption tha t there are more farmers in those categories than would be expected. Cost of o ther f ertilizers /ha v ariable The variable c ost of o ther fertilizers/ha had 386 valid cases and only 13 cases were missing. The descriptive analysis s hows that farm ers spent, on average, 92.94 USD/ha with a standard deviation of 101.97 USD/ha. An important portion (14.5 %) of

PAGE 49

49 them did not use any micro fertilizers and as a consequence, the value for these respondents is zero. On the other hand, 89.4 % of the farmers paid less than 200 USD/ha as i s illustrated in Figure 4 5 The C hi Square Test supports an association between this variable and the c ategorized F Land S ize variable because its statistic (0.002 ) is smaller than the p value (0.05) as i s illustrated in Table 4 11 T he c rosstab analysis (Table 4 12 ) indicates that size of land holdings and costs of other fertilizers are inversely correlated (i.e., as land size increases, cost of other fertilizers/ha goes down) However, only the farmers of the first category can support the assumptions that they have more farmers in that cost column (3 ) than would be expected because their Standard Residuals ( 2.6) was greater than the Z critical value (1.96). On the othe r hand, despite that row s 4 and row 5 ; which belong to the farmers with the largest land area have positive residual values in the column 1 and column 2 ; none of them have Standard Residuals higher than the Z criti cal value (1.69). Therefore, for them, the assumption that these last two categ ories have more farmers than would be expected cannot be supported Cost of f oliar f ertilizer/ha v ariable This variable had 22 missing values which represents 5.5 % of the sample data set O n average, farmers spent 41.37 USD/ha when purchasing foliar fer tilizers. This input did not represent a significant outflow of money. In fact, 35.3 % of the farmers surveyed said that spent less than 15 USD/ha. However, there were 7 cases when respondents stated that they spent more than 200 U SD/ha; but they only represent 2 % of the data set The statistics mentioned above are illustrated graphically in Figure 4 6 The Chi Square Test supports the assertion that there is a relationship between the Cost of Foliar Fertilizers/ha variable and the

PAGE 50

50 because its statistic (0.000) was smaller than the p value (0.05) as it is expressed in the Table 4 13 The crosstab analysis ( Table 4 14) shows that most of the farmers with the smallest land size also had the high est cost of foliar fertilizers because most of them were arranged in the column 3 ( with a Standard Residual value of 17.7 ) Further, the Standard Residual value (3.5) is greater than the z critical value (1.96) Hence, the hypothesis that there are more fa rmers in those tw o categories (row 1 and column 3 ) than would be expected is supported The same happened for most of the farmers grouped in row 2 column 2 as well as farmers with the largest land size who had the smallest cost of foliar fertilizers/ha ( with a Standard Residual Value of 2.5) Herbicide c ost/ha v ariable The costs related to weed control, the Herbicide Cost/ha variable had 39 0 valid cases with only 19 missing cases. Farmers spent, on average, 44.44 USD/ha to buy herbi cides; the standard deviation was 53.83 USD/ha. T he distribution analysis s hows that 90.5 % of the farmers surveyed spent less than 100 USD/ha and as consequence the normal c urve trends to the left side of the histogram ( Figure 4 7 ) The Chi Square T est indicates a correlation between this variabl S ize variable because its statistic (0.000) was smaller than the p value (0.05) as it is observed in the Table 4 15 The crosstab analysis (Table 4 16) shows that t he smallest farmers (ro w 1) also had the highest herbicide cost/ha. However, despite farmers in the row 2 ha ving positive values in the three columns ; none of the standard residual values were greater than the Z crit ical value (1.96). Therefore, t he hypothesis that there are more farmers in those categories than would be expected by chance alone is not supported The same

PAGE 51

51 happened for the farmers in row 4. On the other hand, most of the farmers with the largest land size (row 5 ) had their highest residual value in the column 1, and their Standard Residual value ( 2.2 ) support s the hypothesis that there are more farmers in the row 5 with the lowest herbicide costs than would be expected. Insecticide c ost/ha v ariable The Insecticide Cost/ha v ariable had 38 6 valid cases, and 1 1 cases w ith missing values. F armers spent on average, 40.05 USD/ha with a standard deviatio n of 48.72 USD/ha. There were respondents (3.6 %) that answered that did not have any outflow of money in this input On the othe r hand, 74.1 % of them spent less than 50 USD/ha. The descriptive and distribution analysis can be observed in Figure 4 8 The Chi Square Test s hows that there is correlation between the categorized F Land S ize variable and the Insecticide Cost/ ha variable because its statistic (0.000) was smaller than the level of confidence of (0.05) as i s illustrated in Table 4 17 The Crosstab analysis s hows that most of the farmers in row 1; which represent s the farmers with the smallest land size also had the highest per hectare insecticide costs. This assertion is supported by their highest Residual V alue (16.3) being in column 3 of Table 4 18 In addition, since their Standard R esidual value (3. 3 ) was greater t han the Z critical value (1.96) the ass ertio n that the re are more farmers in that column than would be expected is supported T he table also shows that famers who had larger land size also had lower herbicide costs. This ass ertion is supported since the highest residual values of row s 2, 3, 4, and 5 trended to column 1. Yet, only the largest land sized category of farmers ( row 5 ) had a standard residual value great er than the Z critical val ue (1.96).

PAGE 52

52 Seeding c ost/ha v ariable The variable Seeding Cost/ha ha s 350 valid cases and 51 cases with missing values The values of this variable are given by multipl ying the unit cost for seeding one a local land area unit equal to 441 m 2 and the This variable did show significant variance as indicated by its Standard De viation of 18.66 USD/ha and mean of 197.09 M ost of the farmers surveyed (54.9%) stated that they have paid up to 159 USD/ha, while others (36%) that they have paid up to 181 USD/ha. In addition, there were 6 farmers who indicated that they seed their land by themselves. The histogram s hows the distribution of this variable (Figure 4 9) T he Chi Square Test supports the correlation ze variable and the Seeding Cost/ha variable because its statistic ( 0.007 ) was smaller than the p value (0.05) as is ill ustrated in Table 4 19 The Crosstab analysis (Table 4 20 ) shows that most of the farmers (i.e., those in categories 1, 2, 3, and 4 ) did not demonstrate a correlation with land size categories, as results show that all of the Standard Residual values were smaller than the Z critical value (1. 96). However, row 5 results indicate that most of the largest farmers had the highest seeding cost becau se their highest Residual value was arranged in column 3 Further, their Standard Residual value (3.4) support s the hypothesis that there were more farmers in that column than would be expected by chance alone. Cost of p roducts a pplication /ha v ariable The variable Cost of Product s Application/ha ha s 277 valid cases and 120 cases with missing values The descriptive analysis shows that f armers spent, on average, 37.14 USD/ha to apply the insecticides, herbicides and foliar fertilizers per crop cycle. This v ariable also had values with more frequencies than others. Respondents gave the

PAGE 53

53 answers, 30 USD/ha, 34 USD/ha, and 42 USD/ha at the following frequencies: 19.1 %, 1 6.2 %, and 36.5 % respectively. T here also are 23 farmers that stated they apply the ir products by themselves. Figure 4 10 summarizes these statistics. T he Chi Square Test supports the assertion that there is a correlation Land Si ze variable and the Cost of Products Application /ha variable because its stat istic (0.003 ) wa s smaller than the p value (0.0 5 ), as is illustrated in the Table 4 21 The Crosstab analysis (Table 4 22 ) shows th at most of the smallest farmers (row 1) report the lowest cost of products application/ha because their highest Residual valu e is in column 1 and their Standard Residual value support s the hypothesis that there are more farmers in that category than would be expected by chance alone. The f armers in row s 2, 3, and 4 did not show significant incidence for any specific column beca use their Standard Resi duals are smaller than the Z critical value (1.96). On the other hand, most of the farmers in the largest land size category (i.e., those in row 5 ) ha ve medi um costs because their highest Residual value is located in column 2. Their Standard Residual value (1.9) supports the hypothesis that there are more farmers in those categories than would be expected by chance alone. Cost of f ertilizers a pplication/ha v ariable The Cost of Fertilizer Application s /ha variable ha d 366 valid cases and 35 cases with missing values The values of this variable depend ed on the number of sacks that were applied and by the unit cost of applying each sac k to the field s Farmers spent, on average, 18.48 USD/ha with a Stand ard Deviation of 18.43 USD/ha. The distribution of this variable can be observed in Figure 4 11 T he Chi Square Test shows that there is correlation between the cost of fertilizer application s because

PAGE 54

54 its statistic (0.000) was smaller than the le vel of confidence (0.05) a s reported in Table 4 23 Th e Crosstab analysis (Table 4 24 ) shows that the smallest farmers (those reported in rows 1 and 2 ) have the highest costs of applying fertil izers This ass ertion is supported by their highest Residual V a lues being in column 3. T heir Standard Residuals ( 4.6 and 3.4; respectively) were high er than Z critical value (1.96) which support s the hypothesis that there are more farmers in those categories than would be expected by chance alone. On the ot her hand, the large st land size category correlates with the lowest per hectare cost of fertilizer application s because their Residual values are in column 1 Their Standard Residual values (3.5 and 1.8; respectively ) support the hypothesis that there are more farme rs in those categories than would be expected by chance alone. Irrigation c ost/ha v ariable The variable Irrigation Cost/ha is constructed by summing the annual payments to the irrigation association s and other general expenses like payment for fuel, oil a nd controller of the pump. However, the annual cost was converted/modified to be a production cycle cost in order to make it consistent with other production costs, allowing it to be added to the calculations for total production costs by production cycle (Note, in the Daule region, rice farmers may grow up to three crops of rice in the same calendar year) This variable had 376 valid cases and 25 cases with missing values Descriptive analysis shows that farmers spent, on average, 48. 94 USD/ha wi th a Stand ard Deviation of 70.07 USD/ha. The dist ribution analysis (Figure 4 12) s hows that 66.1 % of the farmers surveyed spent less than 50 USD/ha. T he Chi Square Te st statistic supports the hypothesis that the Irrigation C ost/ha ize

PAGE 55

55 variable are correlated because it (0.000) is smaller than the level of significance (0.05). The results of this analysis are reported in Table 4 25 The crosstab analysis (Table 4 26 ) shows that most of the smallest farmers (i.e., those in row s 1 and 2 ) report the highest per hectare irrigation cost with their highest Residual V alues in column 5. However, these results are mixed as only the farmers in row 2 have a statistically significant correlation between land size category and irrigation costs (i.e. for this category of land size, the Standard Residual value of 2.6 was greater than the z critical value of 1.96). None of the other four categories of land size had Standard Residual values greater than the Z critical value (1.96). Therefore, crosstab a nalysis does not support the hypothesis that there is a correlation between land size category and irrigation costs. Harvesting c ost/ha variable The variable Harvesting Cost ha d 359 valid cases and 42 cases with missing values. F armers spent, on average, 310 USD/ha on harvesting costs, with a Standard Deviation of 303.66 USD/ha. The distribution of the data shows that 54.3% of the farmers surv eyed spent less than 200 USD/ha as reported in Figure 4 13 On the other hand, t he Chi S quare Test suggests that there is a correlation between harvesting costs and the f size because its statistic (0.014 ) was smaller than the p valu e (0.05) as is shown in T able 4 27 The c rosstab analysis (T able 4 28 ) shows that most of the smallest farmers (those in rows 1, and 2 ) had their greatest residual values in the column 3 and 2; respectively. However, the hypothesis that there is a correlation present in the data is only supported for farmers in th e first row, because the Standard Residual value (2.1) for this category of land size is greater than the Z critical value (1.96) and Similarly, farmers

PAGE 56

56 grouped in rows 3, 4, and 5 all have the highest Residual values in the column 1 ; yet, only the fa rmers grouped in r ow 3 ha ve a Standard Residual value higher than the Z critical value. Post Harvest Variables Yield/ha variable The variable Yield /ha is measured by the number of sacks each weighing 205 pounds that were produced per hectare of land farmed This variable has 38 7 valid cases and 1 0 cases w ith missing values. Farmers indicated that, on average, they harvested 61.64 sac ks /ha with a Standard Deviation of 17.312 s ac ks /ha. The distribution analysis shows that 53.2 % of the farme rs surveyed had per hectare yields less t han the mean value Figure 4 14 shows the distribution of this variable. T he Chi Square Test determined th at there is a statistically significant correlation between the Yield/ha variable L and Size variable because its statisti c (0.006 ) was smalle r than the p valu e (0.05) as it is shown in the Table 4 29 The Crosstab analysis (Table 4 30 ) shows that most of the smallest farmers are arranged in row 1 and had the highest rice yields because their highest Residual V alue was loc at ed in column 3. Also, their Standard Residual value (2.4) supports the hypothesis that there are more farmers in that category than would be expected by chance alone. On the other hand, most of the fa rmers in rows 3 and 4 had the lowest rice yield because their Residual values are located in the column 1. However, as their Standard Residual values are lower than the z critical value (1.96) the data does not support th ose hypotheses. The same happened with the farmers in row 5 that are arranged in the medium cost column.

PAGE 57

57 Own c onsumption /ID variable The Own Consumption variable is given by the number of sacks that farmers saved for the own consumption by their families. Therefore, this variable represents e ach farmer ( i.e., by ID) unlike to the previous variables which are evaluated by an ar ea unit ( i.e., hectares ). Th e own consumption per household variable had 396 valid cases and only 1 case w ith a missing value The descriptive analysis show s that farmers saved, on average, 8.21 Sacks/ID with a Standard Deviation of 13.44 Sacks/ID. The dist ribution analysis says that 84.1 % of the farmers surveyed saved less than 10 sack s as illustrated in Figure 4 15. The Chi Square Test supports the hypothesis of correlation between the Ow n Consumption/ID variable Land Size variable because it (0.000 ) was smaller than the p value (0.05) as it is illustrated in T able 4 31 The Crosstab analysis (Table 4 32 ) shows that most of the smallest farmers (i.e., those arranged in row 1 ) saved the smallest amount of rice for their own consumption. Their Standard Residual value (3.3) supports the hypothesis that there are more farmers in that category than would be expected by chance alone. On the other hand, most of the farmers in rows 3 and 4 saved medium amounts of rice because all of them had their highest Residual value in C olumn 2 However, none of th eir Standard Residual values were greater than Z critical value (1.96). Finally, the largest f armers (i.e., those in row 5) saved the highest amount of rice for their own consumption. Their Standard Residual (4.4) for column 3 supports the hypothesis that there are more farmers in those categories than would be expected by chance alone.

PAGE 58

58 Income by paddy rice /ha variable The Income by Paddy Rice/ha variable represents the money that farmers earned by selling their rice harvested as paddy rice. This variable had 362 valid cases and 39 cases with missing values The farmers surveyed gained, on average, 1,551.24 USD/ha with a Stand ard Deviation of 620.19 USD/ha. The distribution analysis shows that 79.8% of the farmers surveye d gained less than 2,000 USD/ha as illustrated in F igure 4 16 T he statistic of the Chi Square Test supports the hypothesis of correlation between the I because it (0.00) was sm aller than the p value (0.0 5) as illustrated in T able 4 33 The crosstab analysis (Table 4 34 ) shows that the smallest farmers grouped in the first two categories (row 1 and row 2) had the lowest income be cause their Residual values were arranged in C olumn 1. T he hypothesis that there are more farmers arranged in a column than would be expected only is the case for farmers in row 2, with a Standard Residual value (5.7) that is greater than the Z critical value (1.96) On the other hand, most of the farmers in row 4 had the highest income because their highest Residual values are arranged in column 3, and thei r Standard Residual value (2.8) s upport s the hypot hesis that there are more farmers in that column than wou ld be expected by chance alone. M ost of the farmers that are arra nged in the row 5 had medium incomes (column 2). Income by p ulled r ice /ha The Income by Pulled Rice per hectare variable was construct ed only for the farmers who sold their production as pulled r ice. As a consequence, it only had 40 valid values and 361 cases had missing values. For the 40 surveyed farmers who did sell pulled rice, their income was, on average, 1378.68 USD/ha with a sta ndard d eviation of

PAGE 59

59 1050.20 USD/ha. The distribution analysis shows that 80% of the farmers surveyed gained less than 2025 USD/ha ( Figure 4 17 ). T he statistic of the Chi Square Tests shows that there is not a statistically significant relationship between t he income by pulled r variable Therefore, for this variable is not necessary to present the crosstab analysis that would have helped to determine categories over the categories of the income by pulled rice /ha variable Details of this analysis are presented in Table 4 35 Economic Variables The economic variables include the analysis of production cost variables and income variables in order to determine the n et income of the rice farmers surveyed. sold the whole rice yield as paddy rice. In this section, the mean values of economic variables are a s ize category. Total c ost/ha variable The Total Cost per hectare variable is constructed by the sum of the mean val ues o f the production cost v ariables analyzed above. The analysis shows that the smallest farmers (Category 1) are the farmers who had the h ighest production cost/ha. It was around 1,6 70 USD/ha. They are f ollowed by category two farmers who had a production cost of 1,2 90 USD/ha. The production costs of the farmers in categor ies 3, 4, and 5 had similar values of approximately 1,010 USD/ha. Figure 4 18 graphically summarized these findings. Total i ncome/ha variable The Total Income per hectare variable is constructed by summing the incomes obtained from selling the harvest as paddy rice and as pulled rice. This variable had 389

PAGE 60

60 valid cases and 12 cases with missing values. Farmers gai ned, on average, 1585 USD/ha with a Standard Deviation of 67 1 USD/ha. The distribution analysis shows that 77.4% of the farmers surveyed gained less than 2000 USD/ha. The se results are summarized in Figure 4 19 When this variable is Size variable it becomes apparent that the smallest farmers ( those grouped in row s 1 and 2) had on average, the lowest t otal income. On the other hand, the farmers in row 4 had, on average, the hi ghest total income followed by the farmers in row 3 and the n farmers in row 5 T able 4 36 shows in detail the se statistics When testing for correlation between these two variables, Chi Square Test statistic was significant (Table 4 37 ) Hence, the hypothesis that said that Total Income/ha are related could not be rejected The estimated statistic was 0. 000 ; which is smaller than the p value (0.05). Consequently the crosstab analysis is presented in Table 4 38 which indicates that despite most of the smallest farmers (row 1) with positive Residual values in column 1 and column 3 none of them had Standard Residual values greater than the Z critical value (1.96) Hence, the data does not support the hypoth esis that there are more farmers in those categories than would be expected by chance alone. In contrast as noted in Table 4 38 farmers grouped in rows 2 to 5 did have statistically significant Standard Residual Values that were greater than the Z critic al value, thus suggesting that these farmers group in one of the three columns more than would be expected by chance alone. Total i ncome/ha* variable The Total Income per hectare variable is constructed to estimate the effects of selling all rice production as paddy rice. This variable also tries to measure the opportunity cost of the saved rice for own consumption. The descriptive analysis shows

PAGE 61

61 that farmers would have gained, on average, 1 77 0 USD/ha with a Standard Deviation of 588 USD/ha. Figure 4 20 shows the statistics and distribution of this variable. When the Total Income/ha* variable is analyzed within the context of the five categories of the smallest farmers (row 1) have the second highest total income*. T heir m ean value is 1 955 USD/ha. F armers in row 4 ha ve the highest total income/ha* with a mean value of 1 98 0 USD/ha. On the other hand, the farmers in row 2 have on average, th e lowest total income/ha*. T able 4 39 details the se findings As noted in Table 4 4 0 t he stati stic of the Chi Square T est is significant, hence the hypothesis that are r elated cannot be rejected because the test statistic (0.000) is sm aller than the p value (0.05) The Crosstab analysis (Table 4 41 ) shows that most of the smallest farmers gained the highest total income/ha* because their Residual Valu es are arranged in column 3 and their Standa rd Residual value (2.0) support s the hypothesis that there are more farmers i n that category than would be expected by chance alone. However, the analysis indicates that the same is true for most of the farmers in row 2 that had their highest Residual Value (4.8) in column 1 which represents the lowest total income/ha*. The larges t f armers (those in rows 4 and 5 ) had the highest and medium total income/ha*; respectively. Their Residual values (11.2 and 8.2 ) were arranged in the third and second column; respectively. As their residual values were greater than Z critical value, the h ypothesis that there are more farmers than would be expected by chance alone in those categories is supported by the data

PAGE 62

62 Hypothetical Scenario from Survey The following analysis presents a hypothetical scenario based upon survey results, in order to determine ex ante an estimated set of benefit and cost streams associated with the introduction of the UDP technology in the zones where the survey was conducted. Therefore, the following analysis contrasts the benefits and costs that the UDP technology would generate for farmers who expressed an interest in adopting the UDP technology in their rice fields Adoption Rate of the UDP T echnology As supported by the survey results, and by previous published research, there will not be 100 % a doption of the UDP technology. Hence an adoption rate is estimated for the purposes of creat ing the hypothetical scenario. This annual adoption rate is estimated with the following equation : Where, = It is the adop tion rate of the UDP technology expressed as a percentage of farmland dedicated to the use of the UDP technology = It is the desire of adopt the UDP technology. It is a dummy variable with only the values of 0 or 1, the value of which is t aken directly fr om S ection 3, question 15 of the survey. = It is amount of land a farmer would assign to test/adopt the UDP technology. It is given in hectares and it value is taken directly from S ection 3, question 17 of the survey.

PAGE 63

63 = It is the total area for rice pr oduction of each farmer. It is given in hectares and its value is taken directly from the section 9, question s 49 50 of the survey. willingness to test/trial/adopt the UDP Technology of each farmer surveyed. The Descriptive analysis of the UDP AR variable shows that the farmers wanted to adopt the UDP technology, on average, in 52% of their lands designated to rice production 19.2 % of the farmers surveyed said that they would use the technology in less than 10% of their land. On the other hand, there were 130 farmers (35.7%) who said that they would use the UDP technology in all of their rice fields The se statistics and distribution s are summarized in the Figure 4 21 The descriptive analysis of this variable, in terms of the smallest farmers (Categor ies 1 and 2) have on average, the biggest adoption rates. Their expressed percentages of land to be dedicated to t he UDP technology are 68.7% and 59%; respectively. However, the total area of thos e farmers that would be produced with UDP technology is 61.41 ha; which is significant ly fewer than the area of the largest farmers (98.73 ha) who have, on average, a 45.4 % adoption rate. Also, it is noticed that in general, increases in farm size was inversely related with the mean adoption rates i.e., expressed willingness to adopt the UDP technology decrease s as total farm size increases. Table 4 42 shows these results in detail. Estimating the Money Saved from Reduced Urea /ha Usage by UDP Technology As noted in the literature review chapter, preliminary results from test plots at ESPOL as well as published results obtained in other countries where the UDP technology h as been adopted, farmers will need less urea for their rice production. To con struct an estimate for this savings, the following equation is used,

PAGE 64

64 Where; : This represents the monetary differences between the two technologies in terms of the costs associated with the amount of urea applied. This estimate of cost differences only make includes the percentage of rice land assigned by each farmer to test/trial/adopt the UDP technology. UDP AR : This represents the Adoption Rates of the UDP technology as defined in the previous section Urea Cost /ha : This represents the cost of purchasing urea for rice production. The UDPec: This represent s the amount of urea briquettes needed to produce one hectare of rice in Ecuadorian conditions Its value is 3.58 sacks (where each sack weighs 50 Kg ) This fertilizer rate is obtained from previous test plot research about the evaluation an d introduction of the UDP technology in Ecuador that was conducted by the author and his colleagues at ESPOL University in collaboration with researchers at the University of Florida. This represents the market price paid by farmers for urea This p rice is a weighted price ( 28.20 USD) that is estimated Figure 4 22 shows its distribution a S UR After running the equation for all applicable farmers, it was determined that farmers willing to adopt and test the UDP technology would save on average, 3 5 USD per growing season. This savings is realized since some farmers would purchase less urea than if they were to continue to broadcast urea by hand (i.e., continue with the

PAGE 65

65 existing t echnology ) Th e equation is estimated for the 320 farmers who indicated an interest in adopting the UDP technology on at least a portion of their rice land However, an estimated 33.8% of the se farmers do not realize any urea cost savings from using the UD P technology because their urea cost s are the same with or without the adoption of the UDP technology. This result, h owever, is somewhat influenced by the methods used to estimate the weighted cost of urea. First, the survey results of farmers costs include the subsidized urea prices from the Ecuadorian government and other sources which lowers the value of the average weighted market price used in the analysis. Second, some of the surveyed farmers use fewer sacks of urea than what is recommended for rice production with urea briquettes. As a consequence, this analysis assumed for the se farmers (n = 59) that their use of urea would remain constant, i.e., they would use the same number of sacks of urea, though some portion of this urea would be made in to briquettes (thus, the production costs for these farmers constant in terms of urea costs) Figure 4 23 shows the statistics associated with this estimation and analysis of the MSUR values The results of the analysis of the MSUR in terms of the categories of size are summarized in Table 4 43. As noted there, the smallest farm size ( c ategor ies 1 and 2 ) would have, on average, the highest savings per hectare Their mean values are 80.28 USD and 86.67 USD; respectively. Also it is noticed that the average of money saved decrease s s. Estimating the Cost of Briquette Application per hectare The estimation of the costs of applying the UDP briquette technology includes the additional labor costs of placing the briquette into the ground, as compared to broadcasting the urea. Th e briquette application is more labor intensive at the time of

PAGE 66

66 application, although the UDP technology only requires one application per growing cycle, whereas the typ ical practice with broadcasting involves three applications in a growing season. Still, in general, the UDP technology is more labor intensive. The estimation of c ost s of applying briquettes is based on the following equation Where; CBA: This represents the cost of inserting the urea briquettes in to the ground in the portion of land each farmer is dedicating to the adoption of the UDP technology. UDP AR: This represents the Adoption Rate of the UDP technology of each farmer. Land Size: This represents the total area of each farmer surveyed that will be used to produce rice. L: This represents the number of farm laborers that are required to insert briquettes in one hectare. The experimental plots by ESPOL estimated that labor needed to place the urea briquettes in the field is, on average, five workers to insert the urea briquettes in one hectare. Pl: This represents the unitary cost of hiring workers to plant rice. This variable ha s a mean value of 7 .24 USD/worker as detailed in Figure 4 24 The descriptive analysis of the Cost of Briquette Application variable shows that famers would have paid, on average, 23.74 USD to insert the urea briquettes in the lands assigned to produce with the UDP technolo gy. The distribution analysis shows that 60.7% of the farmers surveyed would have paid less than the mean value. Figure 4 25 s ummarizes these statistic s Further, w hen the estimated costs of b riquette

PAGE 67

67 a pplication is ana lyzed in terms of ize larger farms tended to have lower costs for urea briquette applications in the areas assigned to prove the UDP technology The smallest farmers would have pa id on average, 35.47 USD/ha and the largest farmers would have paid on average, 26.59 USD/ha. T a ble 4 44 summarizes these findings. Estimating Increment al changes in Rice Yield /ha with UDP technology adoption T he increment al change in rice yield resulting in the adoption of the UDP technology is given by the following equation: Where; : This represents the difference between the rice yield that would be generated with the UDP technology and the rice yield th at is generated with the broadcasted urea technology for the portion of f armland used to adopt/trial the new technology. : This represents the Adoption Rate of the UDP technology. : This represents, in percentage, the mean of the incremental rice yield increase that was found with the use of the UDP technology i n 10 experimental and demonstration plots carried out by ESPOL in Ecuador during the 2010 growing season The IF used for the equation was 1.1499 which came from the 14.99% increment al increase in rice yields obtained experimentally from the ESPOL plots d one in Guayaquil, Daule and San Juan. After running the e quation for the 354 applicable farmers, it was determined that they would increase their rice yield, on average, by 5.72 sacks (where each sack

PAGE 68

68 weighs 50 kilos) by using the UDP technology However, the distribution analysis (Figure 4 26 ) shows that 42.4% of the farmers would increase their yields by less than one sack. When the Increment of Rice Yield by UDP variable is analyzed considering variable the sm allest farmers (Category 1) are on average, the highest increment al improvement in rice yield by using the UDP technology. Also, it is noticed that the largest farmers would have, on average, the lowest increment al increase in rice yield; they are followe d by the farmers in row s 2 3 and then 4 On the other hand, the aggregate area assigned to prove the UDP technology varies in total produce rice with the UDP technology on 16 .21 ha; the farmers in category 2 would total 45.20 ha and the farmers of the c ategory 3 would total 37.88 ha; the farmers in category 4 would total 77.91 ha, and fa rmers in category 5 would total 98.73 ha. T able 4 45 shows in details the se data and statis tics Estimating the Increment al Change in Harvesting Cost with UDP Technology The Increment of Harvesting Cost by UDP variable takes into account the higher harvesting costs associated with higher yields. To the degree that UDP technology increases per hectare yields, the harvesting costs will also increase incrementally. This increase is estimated by the following equation: Where; : This represents the cost s related to harvest, transportation and pullin g service directly linked to the extra rice sacks that would be generated by the UDP technology.

PAGE 69

69 : This represents the Increment al change in Rice Yield from UDP technology adoption, which is estimated in the previous section : This represents the unitary cost of harvesting a rice sack. The mean value is used as a weighted cost for all cases and for the 333 valid cases the mean value for is 4.64 USD/Sack (Figure 4 2 7 ). The descriptive analysis shows that farmers would pay, on average, 26.54 USD for harvesting cost associated with increased rice sacks. There are 354 valid cases for this analysis for which the standard deviation is 45.83 USD. When reviewed within the mean valu es have the same trend tha t the variable s did, because expresses results in monetary terms, of th is yield variable The details of this analy sis are shown in the Table 4 46 Estimating the Extra Income Generated by Adopting the UDP Technology The extra income generated by adopting the UDP technology is estimated as follows, Where; : It represents the extra income that the UDP technology would gene rate as a result of farmers increm entally adopting the technology on a portion of their rice land. : this is given by the Increment al change in Rice Yield by UDP variable. : This represents the unitary price of one rice sack. It is estimated from the e represents a weighted price obtained from the mean value

PAGE 70

70 (29.78 USD/sack) of the Unitary Price of Paddy Rice. Figure 4 28 shows the distribution and descriptive statistics of this variable. After estimating the equation the descriptive analysis (Figure 4 29 ) indicates that farmers would gain, on average, 170.34 USD by using the UDP technology When the Inc UDP variable is analyzed in terms of the categories, the smallest farmers have, on average, the hi ghest income/ha by using the UDP technology. On the other hand, the largest farmers (i.e., those in the category 5 ) would have the lowest benefit s from adopting the UDP technology T able 4 47 shows in detail these results Net Income Generated by the UDP Technology The Net Income of UDP variable is given by the following equation: Where; : This represents the net income that the UDP technology would generate if it is adopted in a portion the survey questions. : This represents the i ncome that the UDP would generate. : This represents the money saved by using the UDP technology. : This represents the cost of briquettes applications. : This represents the increment al increase in harvesting costs resulting from the higher yields associated with using the UDP technology. The findings of this analysis indicates that farmers would gain, on average, 149.54 USD in the areas assigned to prove the UDP technology. The distribution analysis

PAGE 71

71 (Figure 4 30 ) indicates that 63.5% of the farmers surveyed would gain less than 100 USD On the other hand, 8.8% of them would gain more than 500 USD. in relation to net changes in farm income shows that the while farmers with larger size would have lower net income/ha by using the UDP technology. As a consequence, the s mallest farmers obtain the highest net income. The details of this analy sis are expressed in the T able 4 48 Adoption Rate for Ex Ante Benefit Cost Analysis Construction of the Logistic Function For the construction of the benefit and cost streams in the c oming years, it was necessary to determine an ex ante adoption rate that was reasonable and plausible. Farmers expressed exceedingly ambitious and perhaps overly optimistic interest i n adopting the UDP technology. To be more consistent with published benef it cost studies, a more conservative adoption rate was estimated using a logistic function that expresses the adoption rates for the UDP technology over time in Daule and Santa Lucia. Given general similarities between the two production systems, the diffu sion path that was observed for urea tablet (UT) technology in Java Indonesia from 1992 until 1995 is used to parameterize the logistic function ( E. Pasandaran et al 1999). Table 4 49 in West Java, Central Java, and East Java; where, PA represents, in percentage s the planted area covered by the urea tablets. In addition, UT represents the percentage of total ur ea consumed. To construct the estimated using the logistic function, the following equatio n was used,

PAGE 72

72 The values used for the logistic functi on were obtained from Table 4 49 Where: The adoption rate values represent : for the first year ; for the third year ; for the fifth year. And the period values: Th e construction of the logistic function comes from the determination of the following parameters : Once estimated the parameters a re replaced in the equation and the logistic function looks like: Where; t represents the value of each period. The values estimated are expressed in T able 4 50 The estimation of the logistic function shows that the UDP technology would have their maxi mum increment al growth in adoption in the first four years T he area that woul d be cultivated with the UDP technology would pass from 0.33% in the first year to 3.76% in the second ye ar that t hen would increase to 26.70% in the third year and doubled to 53.9% in year four. T he years after year 5 would have annual adoption rate s around 59% and after that period, the increment al adoption rate would

PAGE 73

73 be almost constant Therefore, this stu dy limits this ex ante analysis to only for the next ten years because it was noticed that the estimated adoption rates would not increase significantly after year 6 as indicated in Figure 4 31 Spill over Effects of UDP Adoption Several macroeconomic factors contribute to the overall cost s /benefits of the introduction of the UDP technology in the Cantons of Daule and Santa Lucia This area jointly total s 46,704 hectares in 2000 (SINAGAP 2012 ) much of which is dedicate d to rice production, those there are also high concentrations of urban and rural community populations. Spill over effects analyzed here in this section include the employment generated by the briquette applications, the impacts of urea reduction, and the impacts of increment al increases in rice yields. All these points are an alyzed annually for the next 10 years. Employment G eneration The experimental and demonstration plots determined that to insert the urea briquettes in one hectare of rice land requir ed 5 person days. Therefore, this analysis (Figure 4 32 ) estimate s that the UDP technology would require in Daule and Santa Lucia 1,726 person days of labor that would generate 7,568 USD in additional income in the first year Then, f or the second year, the employment would increase by 19.63 thousand person days with a money generation of 86.05 thousand dollars. In the third year the employment generation would be around 139.28 thousand person days and 61.45 thousand dollars. After year 4, the annu al employment generation would be more than 280 thousand person days and 1 million dollars T he estimat ed values are almost constant after year 5 when the annual employment generation would be around 309.7 thousand person days generating around 1.35 mill ion dollars per year.

PAGE 74

74 Impacts of Reduced Urea Usage The impacts of urea reduced by the UDP Technology were analyzed from the perspectives of money that would be saved by the rice farmers by applying less urea with the UDP technology. Also, the amounts of u rea that would not be applied to the rice fields and that would ultimately pollute the atmosphere and Daule River. A further impact would be the money saved as a result of reduced quantities of urea imports. The analysis for the urea reduction of the rice farmers of Daule and Santa Lucia (Figure 4 33 ) shows that the urea that would not be applied in the first year would be around 75.73 tons which would increase to 857 tons in the second year. The values for the third year would be around 6 thousand tons t hat then would be doubled in the next year. After year 4, the annual values would be around 13.5 thousand tons of urea. Th ose annual urea amounts would generate saving s for rice farmers of Daule and Santa Lucia of 24 thousand dollars in the first year whi ch would increase by almost 273 thousand dollars for the second year. In year 3 the savings are estimated to be 1.9 million dollars which would be doubled in year 4. The annual savings after year 4 would be more than 4 .3 million dollars per year Another benefit from reduced urea usage is the reduction of urea imports. Based on an analysis of current urea imports (Figure 4 34 ) it was determined that national urea imports have increased significantly in the last 20 years. Since 2000 urea imports have bee n over 1 50 thousand of tons annually. However, in the last three years, they increased sharply to be more than 200 thousand tons annually, and the 2011 reports state t hat they reached 235,950 tons of urea with a CIF cost of 113.19 million dollars. This ana lysis used the CIF cost per ton of urea imported that was reported in 2011 for the ex ante analysis of the next 10 years. Despite that the CIF costs have had a positive tendency to increase annually, especially

PAGE 75

75 in the next few years (Figure 4 34 ); it s annu al value is assumed to be constant at a CIF Cost of 479.720 USD The UDP technology would save 36.1 thousand dollars in the first year by the urea that would not be imported in the first year. The saving would increase to 411.1 thousand dollars for the sec ond year. The saving would be alm ost 3 million dollars in year 3 that then it would increase to 5.8 million dollars in year 4. The annual savings in urea imported after year 5 would be around 6.4 million dollars Increment of Rice Production and Income Th e UDP technology would add to the total production of rice farmers of Daule and Santa Lucia aproximately 190 t ons of paddy rice in year 1 and that then would increase to 2.1 thousand tons in year 2. In year 3, the values would increase to 15.3 5 thousand tons After year 4, the values would be more than 30 thousand tons and after year 5 the annual values would be arround 34.1 thousand tons of paddy rice that would be added to the rice farmers overall production ( Figure 4 35 ) Those extra amounts of rice pl us the amount of urea saved would generate a net income of 62.44 thousand dollars in the year 1 that then would increase to 0.71 million dollars in the year 2. In the year 3 it would be aproximately 5 million dollars which then would de doulbled in the ne xt year. After year 5, the annual net income would be around 11 million dollars Table 4 51 summaries the costs and benefits of producing rice with the UDP technology. It shows the monetary values that create the base scenario; which are given by the cost s of briquette applications (CBA) and the costs of harvesting the surplus of rice caused by the UDP technology (IHC UDP). The positive values that would increase the net income are given by the savings (MSUR) caused by the reduced urea usage by the UDP tec hnology and the income (Inc UDP) by selling the extra rice

PAGE 76

76 generated by the UDP technology. The values of the net income (NI UDP) came from the calculation of the whole data, not by the sum of all components mentioned above The extra net income generated by the adoption of the UDP technology by the rice farmers of Daule and Santa Lucia is also analyzed in terms of a sensitive analysis (Table 4 52 ) to determine if the baseline results are r obust. This includes additional estimates for which the estimated ad option rates are increased or decreased by 10% and 25% in order to compare optimistic and pessimistic scenarios The values to compare are the sum of the extra net income of each year. Theref ore, this analysis contrasts 5 different scenarios, one is a base scenario (BS) that was estimated before T here are two optimistic scenarios that increase the UDP adoption rate ( AR ) by 10% and 25%. On the other hand, there are two pessimistic scenarios that lower the adoption rate by 10% and 25% of the the UDP AR. The resuts show that if the UDP AR increases by 10 % more than the baseline estimate, the extra net income would increase from 83.08 million dollars up 91.31 million dollars. And, if the UDP AR incr eases by 25%; the extra net income would increase to 103.85 mi llion dollars However, if the UDP AR descease s by 10% from the values estimated; the extra net income would decline to 74.77 million dollars. And, if the UPD AR decreases 25%, the extra net income would decrease to 62.31 million dollars.

PAGE 77

77 Figure 4 1 H istogram of the Rent Cost/ha V ariable Figure 4 2 Histogram of Cost of Soil Preparation/ha V ariable

PAGE 78

78 Figure 4 3 Histogram of Seed Cost/ha V ariable Figure 4 4 Histogram of the Urea Cost/ha V ariable

PAGE 79

79 Figure 4 5 Histogram of Cost of Other Fertil izers/ha Variable Figure 4 6 Histogram of Cost of Foliar Fertilizers/ha Variable

PAGE 80

80 Figure 4 7 Histogram of Herbicide Cost/ha V ariable Figure 4 8 Histogram of the Insecticide Cost/ha Variable

PAGE 81

81 Figure 4 9 Histogram of Seeding Cost /ha V ariable Figure 4 10 Histogram of C ost of Products Application/ha V ariable

PAGE 82

82 Figure 4 11. Histogram of Cost of Fertilizers Application/ha Variable Figure 4 12 Histogram of Irrigation Cost/ha Variable

PAGE 83

83 Figure 4 13. Histogram of Harvesting Cost/ha Variable Figure 4 14 Histogram of Yield/ha Variable

PAGE 84

84 Figure 4 15 Histogram of Own Consumption Variable Figure 4 16 Histogram of Income by Rice Paddy/ha Variable

PAGE 85

85 Figure 4 17 Histogram of Income by Pulled Rice/ha Variable Figure 4 18 $1.6690 $1.2883 $1.0581 $1.0874 $1.1312 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Category 1 Category 2 Category 3 Category 4 Category 5 Thousand dollars/ha Total Production Cost for each Farmers' Land Size Category

PAGE 86

86 Figure 4 19. Histogram of Total Income/ha Variable Figure 4 20. Histogram of Total Income/ha* Variable

PAGE 87

87 Figure 4 21. Histogram of UDP Adoption Rate (Base Line) Variable Figure 4 22. H istogram of Unit Cost of Urea from Market Variable

PAGE 88

88 Figure 4 23. Histogram of MSUR Variable Figure 4 24. Histogram of Unitary Cost of Hiring a Worker Variable

PAGE 89

89 Figure 4 25. Histogram of Cost of Briquette Application Variable Figure 4 26. Histogram of Increment of Rice Yield by UDP Variable

PAGE 90

90 Figure 4 27. Histogram of Unitary Cost of Harvesting One Sack Variable Figure 4 28. Histogram of Unitary Price of Paddy Rice Variable

PAGE 91

91 Figure 4 29. Histogram of Income that w ould Increase by UDP Variable Figure 4 30. Histogram of Net Income by UDP Variable

PAGE 92

92 Figure 4 31. Logistic Function of the UDP technology for Ecuador Figure 4 32. Employment Generation in Daule and Santa Lucia by UDP Technology 0.33% 3.76% 26.70% 53.90% 58.87% 59.33% 59.36% 59.37% 59.37% 59.37% 0% 10% 20% 30% 40% 50% 60% 70% 0 1 2 3 4 5 6 7 8 9 10 UDP AR Years Logistic Funtion for the Adoption of the UDP technology in Daule and Santa Lucia 0 50 100 150 200 250 300 350 $0 $200 $400 $600 $800 $1,000 $1,200 $1,400 $1,600 1 2 3 4 5 6 7 8 9 10 Thousand of person days Thousand dollars Years Impacts on Employment Generation Employment in thousand of person days Employment in Thousand Dollars

PAGE 93

93 Figure 4 33. Impacts of Reduced Urea Usage on the Farmers of Daule and Santa Lucia Figure 4 34. Historical Urea Imports of Ecuador 0 2000 4000 6000 8000 10000 12000 0 1 1 2 2 3 3 4 4 5 5 1 2 3 4 5 6 7 8 9 10 Thousand tons Million dollars Years Impacts of Urea Reduced by UDP Urea Saved in Thousand Tons Urea Saved in Million Dollars Import Reductions in Million Dollars 0 20 40 60 80 100 120 0 50 100 150 200 250 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Millions Dollars Thousand Tons CIF of Urea Imports Urea Imports

PAGE 94

94 Figure 4 35. Increment of Rice Production and Income of Daule and Santa Lucia 0 0 0 1 1 1 1 1 2 2 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 Million dollars Axis Title Thousand tons Years Impacts of the Increments of Rice Yields Net Income of UDP in Million Dollars Increment of Rice Yield in Thousand Tons

PAGE 95

95 Table 4 1 Categories of the Farmers Land Size Variable Category Label Area Range s 1 (0.04 0.52 ha) 2 [0.52 1 .00 ha) 3 [1 .00 1.50 ha) 4 [1.50 3.19 ha) 5 [3.19 16.33 ha) Table 4 2 Production Cost Variables Cost of Soil Prep. Rent Cost Seed Cost Urea Cost Cost of Other Fertilizers Cost of Foliar Fertilizers Minimum 8.00 0.00 9.39 7.68 0.00 0.00 Maximum 1250.00 528.17 281.25 621.43 950.00 421.35 Percentiles 33.33 100.00 0.00 42.25 100.00 42.25 14.70 66.66 198.12 0.00 63.38 177.46 102.82 40.00 Table 4 2. Continued Herbicide Cost Insec ticide Cost Seeding Cost Cost of Product Application Cost of Fert ilizer App lications Irrigation Cost Harvesting Cost 8 0 0 56.34 0 0 0 1250 400 421.35 226.75 85 62.5 700 17.8 14.08 158.73 33.84 13.21 14.22 161.97 42.25 42.25 181.4 42.3 22 51.56 246.48 Table 4 3 Post Harvest Variables Yield Own Consumption Income by Paddy Rice Income by Pulled Rice Minimum 14.08 0.00 394.37 22.73 Maximum 125.00 210.00 4025.70 4860.00 Percentiles 33.33 56.34 4.00 1500.00 700.00 66.6 7 69.81 10.00 1971.83 1677.00

PAGE 96

96 Table 4 4. Economic Variables Total Income 1 Total Income 2 Net Income 1 Net Income 2 Minimum 394.37 0.00 2370.03 2426.03 Maximum 4025.70 4860.00 3442.62 3245.44 Percentiles 33.33 1500.00 1300.00 335.47 123.73 66.67 1971.83 1819.25 919.03 755.16 Table 4 5 Chi Value df Asymp. Sig. (2 sided) Pearson Chi Square 54.923 a 16 .000 Likelihood Ratio 53.253 16 .000 Linear by Linear Association 28.885 1 .000 N of Valid Cases 370 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 9.22.

PAGE 97

97 Table 4 6 Variable Cost of Soil Preparation/ha in 5 Categories T ot al 1.00 2.00 3.00 4.00 5.00 Size in 5 Categories 1 Count 4 11 12 19 26 72 Expected Count 12.1 15.6 13.0 15.6 15.8 72.0 Residual 8.1 4.6 1.0 3.4 10.2 Std. Residual 2.3 1.2 .3 .9 2.6 2 Count 8 18 21 30 19 96 Expected Count 16.1 20.8 17.4 20.8 21.0 96.0 Residual 8.1 2.8 3.6 9.2 2.0 Std. Residual 2.0 .6 .9 2.0 .4 3 Count 6 18 12 11 8 55 Expected Count 9.2 11.9 10.0 11.9 12.0 55.0 Residual 3.2 6.1 2.0 .9 4.0 Std. Residual 1.1 1.8 .6 .3 1.2 4 Count 19 19 14 11 13 76 Expected Count 12.7 16.4 13.8 16.4 16.6 76.0 Residual 6.3 2.6 .2 5.4 3.6 Std. Residual 1.8 .6 .1 1.3 .9 5 Count 25 14 8 9 15 71 Expected Count 11.9 15.4 12.9 15.4 15.5 71.0 Residual 13.1 1.4 4.9 6.4 .5 Std. Residual 3.8 .3 1.4 1.6 .1 Total Count 62 80 67 80 81 370 Expected Count 62.0 80.0 67.0 80.0 81.0 370.0 Table 4 7 Chi Land Size variable Value df Asymp. Sig. (2 sided) Pearson Chi Square 42.988 a 8 .000 Likelihood Ratio 42.814 8 .000 Linear by Linear Association 26.726 1 .000 N of Valid Cases 269 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 12.65.

PAGE 98

98 Table 4 8 Crosstabulation of Farmers' Land Size Variable Seed Cost/ha in 3 Categories Seed Cost/ha in 3 Cat egories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 9 7 25 41 Expected Count 14.9 13.4 12.7 41.0 Residual 5.9 6.4 12.3 Std. Residual 1.5 1.8 3.5 2.00 Count 16 36 28 80 Expected Count 29.1 26.2 24.7 80.0 Residual 13.1 9.8 3.3 Std. Residual 2.4 1.9 .7 3.00 Count 20 16 9 45 Expected Count 16.4 14.7 13.9 45.0 Residual 3.6 1.3 4.9 Std. Residual .9 .3 1.3 4.00 Count 26 11 13 50 Expected Count 18.2 16.4 15.4 50.0 Residual 7.8 5.4 2.4 Std. Residual 1.8 1.3 .6 5.00 Count 27 18 8 53 Expected Count 19.3 17.3 16.4 53.0 Residual 7.7 .7 8.4 Std. Residual 1.8 .2 2.1 Total Count 98 88 83 269 Expected Count 98.0 88.0 83.0 269.0 Table 4 9 Chi Value df Asymp. Sig. (2 sided) Pearson Chi Square 41.364 a 8 .000 Likelihood Ratio 41.795 8 .000 Linear by Linear Association 34.202 1 .000 N of Valid Cases 347 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 16.43.

PAGE 99

99 Table 4 10 Crosstabulation of Farmers' Land Size variable Urea Cost/ha variable Urea Cost /ha in 3 Categories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 11 18 37 66 Expected Count 22.3 22.1 21.7 66.0 Residual 11.3 4.1 15.3 Std. Residual 2.4 .9 3.3 2.00 Count 21 31 39 91 Expected Count 30.7 30.4 29.9 91.0 Residual 9.7 .6 9.1 Std. Residual 1.7 .1 1.7 3.00 Count 19 20 11 50 Expected Count 16.9 16.7 16.4 50.0 Residual 2.1 3.3 5.4 Std. Residual .5 .8 1.3 4.00 Count 33 24 12 69 Expected Count 23.3 23.1 22.7 69.0 Residual 9.7 .9 10.7 Std. Residual 2.0 .2 2.2 5.00 Count 33 23 15 71 Expected Count 23.9 23.7 23.3 71.0 Residual 9.1 .7 8.3 Std. Residual 1.9 .2 1.7 Total Count 117 116 114 347 Expected Count 117.0 116.0 114.0 347.0 Table 4 11 Chi Value Df Asymp. Sig. (2 sided) Pearson Chi Square 24.340 a 8 .002 Likelihood Ratio 24.137 8 .002 Linear by Linear Association 16.366 1 .000 N of Valid Cases 386 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 18.75.

PAGE 100

100 Table 4 12 Crosstabulation of Farmers' Land Size Variable Cost of other Fertilizers Variable Cost of other Fertilizers/ha in 3 Categories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 20 17 38 75 Expected Count 25.5 24.7 24.9 75.0 Residual 5.5 7.7 13.1 Std. Residual 1.1 1.5 2.6 2.00 Count 24 36 38 98 Expected Count 33.3 32.2 32.5 98.0 Residual 9.3 3.8 5.5 Std. Residual 1.6 .7 1.0 3.00 Count 20 23 14 57 Expected Count 19.3 18.8 18.9 57.0 Residual .7 4.2 4.9 Std. Residual .1 1.0 1.1 4.00 Count 33 27 18 78 Expected Count 26.5 25.7 25.9 78.0 Residual 6.5 1.3 7.9 Std. Residual 1.3 .3 1.5 5.00 Count 34 24 20 78 Expected Count 26.5 25.7 25.9 78.0 Residual 7.5 1.7 5.9 Std. Residual 1.5 .3 1.2 Total Count 131 127 128 386 Expected Count 131.0 127.0 128.0 386.0 Table 4 13 Chi Square Tests of Foliar Fertilizer Value df Asymp. Sig. (2 sided) Pearson Chi Square 37.458 a 8 .000 Likelihood Ratio 36.738 8 .000 Linear by Linear Association 19.244 1 .000 N of Valid Cases 385 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 18.51.

PAGE 101

101 Table 4 14 Crosstabulation of Farmers' Land Size Variable Cost of Foliar Fertilizers Variable Cost of Foliar Fertilizers/ha in 3 Categories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 13 22 43 78 Expected Count 25.9 26.7 25.3 78.0 Residual 12.9 4.7 17.7 Std. Residual 2.5 .9 3.5 2.00 Count 29 41 27 97 Expected Count 32.2 33.3 31.5 97.0 Residual 3.2 7.7 4.5 Std. Residual .6 1.3 .8 3.00 Count 21 25 11 57 Expected Count 19.0 19.5 18.5 57.0 Residual 2.0 5.5 7.5 Std. Residual .5 1.2 1.7 4.00 Count 27 22 27 76 Expected Count 25.3 26.1 24.7 76.0 Residual 1.7 4.1 2.3 Std. Residual .3 .8 .5 5.00 Count 38 22 17 77 Expected Count 25.6 26.4 25.0 77.0 Residual 12.4 4.4 8.0 Std. Residual 2.5 .9 1.6 Total Count 128 132 125 385 Expected Count 128.0 132.0 125.0 385.0 Table 4 15 Chi Value df Asymp. Sig. (2 sided) Pearson Chi Square 56.970 a 8 .000 Likelihood Ratio 56.868 8 .000 Linear by Linear Association 28.332 1 .000 N of Valid Cases 390 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 18.85.

PAGE 102

102 Table 4 16 Crosstabulation of Farmers' Land Size Variable Herbicide Cost/ha Variable Herbicide Cost/ha in 3 Categories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 10 19 50 79 Expected Count 26.3 26.5 26.1 79.0 Residual 16.3 7.5 23.9 Std. Residual 3.2 1.5 4.7 2.00 Count 29 41 28 98 Expected Count 32.7 32.9 32.4 98.0 Residual 3.7 8.1 4.4 Std. Residual .6 1.4 .8 3.00 Count 28 20 9 57 Expected Count 19.0 19.1 18.9 57.0 Residual 9.0 .9 9.9 Std. Residual 2.1 .2 2.3 4.00 Count 26 24 28 78 Expected Count 26.0 26.2 25.8 78.0 Residual .0 2.2 2.2 Std. Residual .0 .4 .4 5.00 Count 37 27 14 78 Expected Count 26.0 26.2 25.8 78.0 Residual 11.0 .8 11.8 Std. Residual 2.2 .2 2.3 Total Count 130 131 129 390 Expected Count 130.0 131.0 129.0 390.0 Table 4 17 Chi Square Tests of Insecticide Cost/ha Variable Farmers land Size Variable Value Df Asymp. Sig. (2 sided) Pearson Chi Square 41.613 a 8 .000 Likelihood Ratio 43.709 8 .000 Linear by Linear Association 20.288 1 .000 N of Valid Cases 386 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 18.33.

PAGE 103

103 Table 4 18 Crosstabulation of Farmers' Land Size Variable Insecticide Cost/ha Variable Insec ticide Cost/ha in 3 Cat egories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 10 27 41 78 Expected Count 26.3 27.1 24.7 78.0 Residual 16.3 .0 16.3 Std. Residual 3.2 .0 3.3 2.00 Count 27 43 29 99 Expected Count 33.3 34.4 31.3 99.0 Residual 6.3 8.6 2.3 Std. Residual 1.1 1.5 .4 3.00 Count 26 23 9 58 Expected Count 19.5 20.1 18.3 58.0 Residual 6.5 2.9 9.3 Std. Residual 1.5 .6 2.2 4.00 Count 32 25 21 78 Expected Count 26.3 27.1 24.7 78.0 Residual 5.7 2.1 3.7 Std. Residual 1.1 .4 .7 5.00 Count 35 16 22 73 Expected Count 24.6 25.3 23.1 73.0 Residual 10.4 9.3 1.1 Std. Residual 2.1 1.9 .2 Total Count 130 134 122 386 Expected Count 130.0 134.0 122.0 386.0 Table 4 19 Chi Variable Value Df Asymp. Sig. (2 sided) Pearson Chi Square 21.158 a 8 .007 Likelihood Ratio 23.193 8 .003 Linear by Linear Association 4.131 1 .042 N of Valid Cases 350 a. 5 cells (33.3%) have expected count less than 5. The minimum expected count is 1.82.

PAGE 104

104 Table 4 20 Crosstabulation of Farmers' Land Size Variable Seeding Cost/ha Variable Seeding Cost /ha in 3 Cat egories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 36 20 0 56 Expected Count 33.6 20.5 1.9 56.0 Residual 2.4 .5 1.9 Std. Residual .4 .1 1.4 2.00 Count 58 29 4 91 Expected Count 54.6 33.3 3.1 91.0 Residual 3.4 4.3 .9 Std. Residual .5 .7 .5 3.00 Count 30 23 0 53 Expected Count 31.8 19.4 1.8 53.0 Residual 1.8 3.6 1.8 Std. Residual .3 .8 1.3 4.00 Count 48 28 0 76 Expected Count 45.6 27.8 2.6 76.0 Residual 2.4 .2 2.6 Std. Residual .4 .0 1.6 5.00 Count 38 28 8 74 Expected Count 44.4 27.1 2.5 74.0 Residual 6.4 .9 5.5 Std. Residual 1.0 .2 3.4 Total Count 210 128 12 350 Expected Count 210.0 128.0 12.0 350.0 Table 4 21 Chi Size Variable Value Df Asymp. Sig. (2 sided) Pearson Chi Square 23.243 a 8 .003 Likelihood Ratio 24.323 8 .002 Linear by Linear Association 15.912 1 .000 N of Valid Cases 277 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 6.96.

PAGE 105

105 Table 4 22 Crosstabulation of Farmers' Land Size Variable Cost of Products Application Variable Cost of Prod uct Appli cation in 3 Cat egories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categ 3 ories 1.00 Count 28 9 6 43 Expected Count 20.0 15.7 7.3 43.0 Residual 8.0 6.7 1.3 Std. Residual 1.8 1.7 .5 2.00 Count 40 22 10 72 Expected Count 33.5 26.3 12.2 72.0 Residual 6.5 4.3 2.2 Std. Residual 1.1 .8 .6 3.00 Count 22 15 4 41 Expected Count 19.1 14.9 7.0 41.0 Residual 2.9 .1 3.0 Std. Residual .7 .0 1.1 4.00 Count 24 24 12 60 Expected Count 27.9 21.9 10.2 60.0 Residual 3.9 2.1 1.8 Std. Residual .7 .5 .6 5.00 Count 15 31 15 61 Expected Count 28.4 22.2 10.4 61.0 Residual 13.4 8.8 4.6 Std. Residual 2.5 1.9 1.4 Total Count 129 101 47 277 Expected Count 129.0 101.0 47.0 277.0 Table 4 23 Chi Square Tests of Cost of Fertilizers Land Size Variable Value Df Asymp. Sig. (2 sided) Pearson Chi Square 96.800 a 8 .000 Likelihood Ratio 100.104 8 .000 Linear by Linear Association 69.874 1 .000 N of Valid Cases 366 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 17.23.

PAGE 106

106 Table 4 24 Crosstabulation of Farmers' Land Size Variable Cost of Fertilizers Application/ha Variable Cost of Fertilizer Appli cations in 3 C at egories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 10 16 45 71 Expected Count 23.7 24.2 23.1 71.0 Residual 13.7 8.2 21.9 Std. Residual 2.8 1.7 4.6 2.00 Count 15 29 49 93 Expected Count 31.0 31.8 30.2 93.0 Residual 16.0 2.8 18.8 Std. Residual 2.9 .5 3.4 3.00 Count 21 25 7 53 Expected Count 17.7 18.1 17.2 53.0 Residual 3.3 6.9 10.2 Std. Residual .8 1.6 2.5 4.00 Count 42 24 8 74 Expected Count 24.7 25.3 24.1 74.0 Residual 17.3 1.3 16.1 Std. Residual 3.5 .3 3.3 5.00 Count 34 31 10 75 Expected Count 25.0 25.6 24.4 75.0 Residual 9.0 5.4 14.4 Std. Residual 1.8 1.1 2.9 Total Count 122 125 119 366 Expected Count 122.0 125.0 119.0 366.0 Table 4 25 Chi Square Tests of Irrigation Cost Variable Value df Asymp. Sig. (2 sided) Pearson Chi Square 28.208 a 8 .000 Likelihood Ratio 28.406 8 .000 Linear by Linear Association 14.952 1 .000 N of Valid Cases 375 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 18.33.

PAGE 107

107 Table 4 26 Crosstabulation Farmers' Land Size Variable Irrigation Cost/ha Variable Irrigation Cost/ha in 3 Categories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 21 19 32 72 Expected Count 24.0 24.0 24.0 72.0 Residual 3.0 5.0 8.0 Std. Residual .6 1.0 1.6 2.00 Count 27 23 47 97 Expected Count 32.3 32.3 32.3 97.0 Residual 5.3 9.3 14.7 Std. Residual .9 1.6 2.6 3.00 Count 19 21 15 55 Expected Count 18.3 18.3 18.3 55.0 Residual .7 2.7 3.3 Std. Residual .2 .6 .8 4.00 Count 26 34 17 77 Expected Count 25.7 25.7 25.7 77.0 Residual .3 8.3 8.7 Std. Residual .1 1.6 1.7 5.00 Count 32 28 14 74 Expected Count 24.7 24.7 24.7 74.0 Residual 7.3 3.3 10.7 Std. Residual 1.5 .7 2.1 Total Count 125 125 125 375 Expected Count 125.0 125.0 125.0 375.0 Table 4 27. Chi Variable Value df Asymp. Sig. (2 sided) Pearson Chi Square 19.075 a 8 .014 Likelihood Ratio 18.621 8 .017 Linear by Linear Association 8.189 1 .004 N of Valid Cases 359 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 18.58.

PAGE 108

108 Table 4 28 Crosstabulation of Farmers' Land Size Variable Harvesting Cost/ha Variable Harve sting Cost /ha in 3 Cat egories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 17 20 33 70 Expected Count 23.4 23.8 22.8 70.0 Residual 6.4 3.8 10.2 Std. Residual 1.3 .8 2.1 2.00 Count 22 40 31 93 Expected Count 31.1 31.6 30.3 93.0 Residual 9.1 8.4 .7 Std. Residual 1.6 1.5 .1 3.00 Count 27 16 14 57 Expected Count 19.1 19.4 18.6 57.0 Residual 7.9 3.4 4.6 Std. Residual 1.8 .8 1.1 4.00 Count 28 24 19 71 Expected Count 23.7 24.1 23.1 71.0 Residual 4.3 .1 4.1 Std. Residual .9 .0 .9 5.00 Count 26 22 20 68 Expected Count 22.7 23.1 22.2 68.0 Residual 3.3 1.1 2.2 Std. Residual .7 .2 .5 Total Count 120 122 117 359 Expected Count 120.0 122.0 117.0 359.0 Table 4 29 Chi Value df Asymp. Sig. (2 sided) Pearson Chi Square 21.424 a 8 .006 Likelihood Ratio 21.960 8 .005 Linear by Linear Association 6.782 1 .009 N of Valid Cases 387 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 12.89.

PAGE 109

109 Table 4 30 Crosstabulation of Farmers' Land Size Variable Yield/ha Variable Yield/ha in 3 Categories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 19 19 37 75 Expected Count 33.3 16.7 25.0 75.0 Residual 14.3 2.3 12.0 Std. Residual 2.5 .6 2.4 2.00 Count 44 26 29 99 Expected Count 44.0 22.0 33.0 99.0 Residual .0 4.0 4.0 Std. Residual .0 .9 .7 3.00 Count 32 10 16 58 Expected Count 25.8 12.9 19.3 58.0 Residual 6.2 2.9 3.3 Std. Residual 1.2 .8 .8 4.00 Count 43 11 24 78 Expected Count 34.7 17.3 26.0 78.0 Residual 8.3 6.3 2.0 Std. Residual 1.4 1.5 .4 5.00 Count 34 20 23 77 Expected Count 34.2 17.1 25.7 77.0 Residual .2 2.9 2.7 Std. Residual .0 .7 .5 Total Count 172 86 129 387 Expected Count 172.0 86.0 129.0 387.0 Table 4 31 Chi Square Tests of Own Consumption/ID Variable Farmers Land Size Variable Value df Asymp. Sig. (2 sided) Pearson Chi Square 55.385 a 8 .000 Likelihood Ratio 51.745 8 .000 Linear by Linear Association 24.826 1 .000 N of Valid Cases 396 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 9.23.

PAGE 110

110 Table 4 32 Crosstabulation of Farmers' Land Size Variable Own Consumption/ha Variable Own Consumption/ha in 3 Categories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 45 31 4 80 Expected Count 27.5 39.8 12.7 80.0 Residual 17.5 8.8 8.7 Std. Residual 3.3 1.4 2.4 2.00 Count 34 54 13 101 Expected Count 34.7 50.2 16.1 101.0 Residual .7 3.8 3.1 Std. Residual .1 .5 .8 3.00 Count 14 37 7 58 Expected Count 19.9 28.9 9.2 58.0 Residual 5.9 8.1 2.2 Std. Residual 1.3 1.5 .7 4.00 Count 18 50 11 79 Expected Count 27.1 39.3 12.6 79.0 Residual 9.1 10.7 1.6 Std. Residual 1.8 1.7 .4 5.00 Count 25 25 28 78 Expected Count 26.8 38.8 12.4 78.0 Residual 1.8 13.8 15.6 Std. Residual .3 2.2 4.4 Total Count 136 197 63 396 Expected Count 136.0 197.0 63.0 396.0 Table 4 33 Chi Square Tests of Income Paddy Rice /ha Variable Farmers Land Size Variable Value df Asymp. Sig. (2 sided) Pearson Chi Square 86.440 a 8 .000 Likelihood Ratio 85.606 8 .000 Linear by Linear Association 22.721 1 .000 N of Valid Cases 362 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 19.23.

PAGE 111

111 Table 4 34 Crosstabulation of Farmers' Land Size Variable Income by Paddy Rice/ha Variable Incom e by Paddy Rice /ha in 3 Categories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 28 18 26 72 Expected Count 24.1 24.1 23.9 72.0 Residual 3.9 6.1 2.1 Std. Residual .8 1.2 .4 2.00 Count 60 16 11 87 Expected Count 29.1 29.1 28.8 87.0 Residual 30.9 13.1 17.8 Std. Residual 5.7 2.4 3.3 3.00 Count 11 26 21 58 Expected Count 19.4 19.4 19.2 58.0 Residual 8.4 6.6 1.8 Std. Residual 1.9 1.5 .4 4.00 Count 11 26 39 76 Expected Count 25.4 25.4 25.2 76.0 Residual 14.4 .6 13.8 Std. Residual 2.9 .1 2.8 5.00 Count 11 35 23 69 Expected Count 23.1 23.1 22.9 69.0 Residual 12.1 11.9 .1 Std. Residual 2.5 2.5 .0 Total Count 121 121 120 362 Expected Count 121.0 121.0 120.0 362.0 Table 4 35. Chi Size Variable Value Df Asymp. Sig. (2 sided) Pearson Chi Square 5.926 a 8 .655 Likelihood Ratio 8.374 8 .398 Linear by Linear Association .014 1 .907 N of Valid Cases 40 a. 12 cells (80.0%) have expected count less than 5. The minimum expected count is .65.

PAGE 112

112 Table 4 36. Comparison of Total Income/ha Variable through Farmers Land Size Variable Farmers' Land Size in 5 Categories N Minimum Maximum Mean Std. Deviation 1 Total Income (USD/ha) 76 0.00 3625.00 1541.01 763.14 2 Total Income (USD/ha) 100 0.00 4860.00 1322.24 687.83 3 Total Income (USD/ha) 58 566.14 2783.80 1640.35 420.55 4 Total Income (USD/ha) 78 700.00 3828.52 1895.76 656.50 5 Total Income (USD/ha) 77 0.00 3038.00 1614.90 584.04 Table 4 37. Chi Size Variable Value Df Asymp. Sig. (2 sided) Pearson Chi Square 69.534 a 8 .000 Likelihood Ratio 68.565 8 .000 Linear by Linear Association 16.053 1 .000 N of Valid Cases 389 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 19.23.

PAGE 113

113 Table 4 38. Crosstabulation of Farmers' Land Size Variable Total Income Variable Total Inc ome in 3 Cat egories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 28 20 28 76 Expected Count 25.4 25.4 25.2 76.0 Residual 2.6 5.4 2.8 Std. Residual .5 1.1 .6 2.00 Count 63 21 16 100 Expected Count 33.4 33.4 33.2 100.0 Residual 29.6 12.4 17.2 Std. Residual 5.1 2.1 3.0 3.00 Count 11 26 21 58 Expected Count 19.4 19.4 19.2 58.0 Residual 8.4 6.6 1.8 Std. Residual 1.9 1.5 .4 4.00 Count 12 27 39 78 Expected Count 26.1 26.1 25.9 78.0 Residual 14.1 .9 13.1 Std. Residual 2.8 .2 2.6 5.00 Count 16 36 25 77 Expected Count 25.7 25.7 25.5 77.0 Residual 9.7 10.3 .5 Std. Residual 1.9 2.0 .1 Total Count 130 130 129 389 Expected Count 130.0 130.0 129.0 389.0 Table 4 39. Size Variable Farmers' Land Size in 5 Categories N Minimum Maximum Mean Std. Deviation 1 Total Income/ha* 69 900.00 3662.50 1955.28 677.23 2 Total Income/ha* 87 394.37 2957.75 1482.54 499.93 3 Total Income/ha* 58 676.00 2838.03 1771.19 414.99 4 Total Income/ha* 76 840.00 4025.70 1981.57 624.78 5 Total Income/ha* 68 420.00 3100.00 1731.65 526.91

PAGE 114

114 Table 4 40. Chi Variable Value df Asymp. Sig. (2 sided) Pearson Chi Square 61.385 a 8 .000 Likelihood Ratio 59.653 8 .000 Linear by Linear Association 1.472 1 .225 N of Valid Cases 358 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 18.63. Table 4 41. Crosstabulation of Farmers' Land Size Variable Total Income/ha* Variable Total Income 2 in 3 Cat egories Total 1.00 2.00 3.00 Farmers' Land Size in 5 Categories 1.00 Count 19 18 32 69 Expected Count 24.3 22.2 22.6 69.0 Residual 5.3 4.2 9.4 Std. Residual 1.1 .9 2.0 2.00 Count 57 16 14 87 Expected Count 30.6 27.9 28.4 87.0 Residual 26.4 11.9 14.4 Std. Residual 4.8 2.3 2.7 3.00 Count 14 25 19 58 Expected Count 20.4 18.6 19.0 58.0 Residual 6.4 6.4 .0 Std. Residual 1.4 1.5 .0 4.00 Count 14 26 36 76 Expected Count 26.7 24.4 24.8 76.0 Residual 12.7 1.6 11.2 Std. Residual 2.5 .3 2.2 5.00 Count 22 30 16 68 Expected Count 23.9 21.8 22.2 68.0 Residual 1.9 8.2 6.2 Std. Residual .4 1.7 1.3 Total Count 126 115 117 358 Expected Count 126.0 115.0 117.0 358.0

PAGE 115

115 Table 4 42. Comparison of UDP AR Variable through UDP Area Variable Farmers' Land Size in 5 Categories N Minimum Maximum Sum Mean Std. Deviation 1 UDP AR 75 11.27% 100.00% 68.73% 34.03% UDP Area (h a) 75 0.04 0.5 16.21 0.2161 0.14521 2 UDP AR 95 4.00% 100.00% 58.99% 41.08% UDP Area (h a) 95 0.04 1 45.2 0.4758 0.34449 3 UDP AR 51 2.82% 100.00% 53.68% 38.95% UDP Area (h a) 51 0.04 1.42 37.88 0.7427 0.53599 4 UDP AR 73 1.41% 100.00% 45.41% 39.02% UDP Area (h a) 73 0.04 3 77.91 1.0673 0.93373 5 UDP AR 70 0.00% 100.00% 30.76% 36.36% UDP Area (h a) 70 0 6 98.73 1.4104 1.63293 Table 4 43. Comparison of MSUR Variable th r ough UDP Area Variable Farmers' Land Size in 5 Categories N Minimum Maximum Sum Mean Std. Deviation 1 MSUR/ha (USD) 62 0.00 520.33 8039.69 129.67 120.24 UDP Area (ha) 75 0.04 0.50 16.21 0.22 0.15 2 MSUR/ha (USD) 87 0.00 291.86 8210.44 94.37 85.65 UDP Area (ha) 95 0.04 1.00 45.20 0.48 0.34 3 MSUR/ha (USD) 44 0.00 253.83 1549.60 35.22 57.30 UDP Area (ha) 51 0.04 1.42 37.88 0.74 0.54 4 MSUR/ha (USD) 64 0.00 170.90 2010.64 31.42 47.53 UDP Area (ha) 73 0.04 3.00 77.91 1.07 0.93 5 MSUR/ha (USD) 62 0.00 236.40 2365.36 38.15 55.93 UDP Area (ha) 70 0.00 6.00 98.73 1.41 1.63

PAGE 116

116 Table 4 44. Comparison of CBA/ha Variable t h rough UDP Area Adopted Farmers' Land Size in 5 Categories N Minimum Maximum Sum Mean Std. Deviation 1 CBA/ha (USD) 55 4.37 465.94 2339.92 42.54 69.26 UDP Area (ha) 75 0.04 0.50 16.21 0.22 0.15 2 CBA/ha (USD) 86 1.06 56.80 1885.75 21.93 16.10 UDP Area (ha) 95 0.04 1.00 45.20 0.48 0.34 3 CBA/ha (USD) 47 0.85 87.50 970.99 20.66 17.87 UDP Area (ha) 51 0.04 1.42 37.88 0.74 0.54 4 CBA/ha (USD) 71 0.49 45.18 1140.58 16.06 14.12 UDP Area (ha) 73 0.04 3.00 77.91 1.07 0.93 5 CBA/ha (USD) 67 0.20 50.00 806.68 12.04 14.58 UDP Area (ha) 70 0.00 6.00 98.73 1.41 1.63 Table 4 45. Comparison of IRY UDP Variable t h rough UDP Area Variable Farmers' Land Size in 5 Categories N Minimum Maximum Sum Mean Std. Deviation 1 IRY UDP/ha (s ack) 71 0.98 155.90 893.13 12.58 23.02 UDP Area (ha) 75 0.04 0.50 16.21 0.22 0.15 2 IRY UDP/ha (s ack) 93 0.17 14.15 425.25 4.57 3.62 UDP Area (ha) 95 0.04 1.00 45.20 0.48 0.34 3 IRY UDP/ha (s ack) 51 0.18 26.39 257.31 5.05 4.71 UDP Area (ha) 51 0.04 1.42 37.88 0.74 0.54 4 IRY UDP/ha (sack ) 72 0.11 14.99 312.97 4.35 4.01 UDP Area (ha) 73 0.04 3.00 77.91 1.07 0.93 5 IRY UDP/ha (sack ) 66 0.02 13.87 199.14 3.02 3.75 UDP Area (ha) 70 0.00 6.00 98.73 1.41 1.63

PAGE 117

117 Table 4 46. Size Categories Farmers' Land Size in 5 Categories N Minimum Maximum Sum Mean Std. Deviation 1 IHC UDP/ha (USD) 71 4.55 723.38 4144.13 58.37 106.83 UDP Area (ha) 75 0.04 0.50 16.21 0.22 0.15 2 IHC UDP/ha (USD) 93 0.77 65.64 1973.18 21.22 16.78 UDP Area (ha) 95 0.04 1.00 45.20 0.48 0.34 3 IHC UDP/ha (USD) 51 0.83 122.45 1193.92 23.41 21.88 UDP Area (ha) 51 0.04 1.42 37.88 0.74 0.54 4 IHC UDP/ha (USD) 72 0.52 69.55 1452.19 20.17 18.59 UDP Area (ha) 73 0.04 3.00 77.91 1.07 0.93 5 IHC UDP/ha (USD) 66 0.11 64.34 923.99 14.00 17.40 UDP Area (ha) 70 0.00 6.00 98.73 1.41 1.63 Table 4 47. Land Size Categories Farmers' Land Size in 5 Categories N Minimum Maximum Sum Mean Std. Deviation 1 Inc UDP/ha (USD) 71 29.22 4643.30 26600.59 374.66 685.74 UDP Area (ha) 75 0.04 0.50 16.21 0.22 0.15 2 Inc UDP/ha (USD) 93 4.96 421.30 12665.53 136.19 107.70 UDP Area (ha) 95 0.04 1.00 45.20 0.48 0.34 3 Inc UDP/ha (USD) 51 5.31 786.01 7663.63 150.27 140.42 UDP Area (ha) 51 0.04 1.42 37.88 0.74 0.54 4 Inc UDP/ha (USD) 72 3.32 446.45 9321.43 129.46 119.31 UDP Area (ha) 73 0.04 3.00 77.91 1.07 0.93 5 Inc UDP/ha (USD) 66 0.69 412.97 5930.98 89.86 111.71 UDP Area (ha) 70 0.00 6.00 98.73 1.41 1.63

PAGE 118

118 Table 4 48. Comparison of NI UDP/ha Variable and UDP Area Variable through Farmers' Land Size in 5 Categories N Minimum Maximum Sum Mean Std. Deviation 1 NI UDP/ha (USD) 45 46.69 3615.38 18703.13 415.63 657.85 UDP Area (ha) 75 0.04 0.50 16.21 0.22 0.15 2 NI UDP/ha (USD) 76 2.22 513.42 13888.96 182.75 110.39 UDP Area (ha) 95 0.04 1.00 45.20 0.48 0.34 3 NI UDP/ha (USD) 41 4.12 589.05 5384.62 131.33 128.55 UDP Area (ha) 51 0.04 1.42 37.88 0.74 0.54 4 NI UDP/ha (USD) 62 4.33 407.27 7325.67 118.16 96.94 UDP Area (ha) 73 0.04 3.00 77.91 1.07 0.93 5 NI UDP/ha (USD) 57 0.37 319.38 5504.18 96.56 83.69 UDP Area (ha) 70 0.00 6.00 98.73 1.41 1.63 Table 4 49. Year Average Adoption Rates of Java PA UT ** 1992 0.23% 0.50% 1992/1993 9.33% 10.33% 1993 21.67% 21.33% 1993/1994 56.67% 54.67% 1995 58.67% 57.33% *PA represents the mean percentage value of the planted area cultivated with urea tablets technology from the total area of Java. **UT represents the mean percentage value of the urea consumed by urea tablets technology in Java.

PAGE 119

119 Table 4 50. Projection of the UDP technology adoption in Daule and Santa Lucia, Ecuador Years ** 1 0.33% 2 3.76% 3 26.70% 4 53.90% 5 58.87% 6 59.33% 7 59.36% 8 59.37% 9 59.37% 10 59.37% The years are proje cted for the next 10 years from the present time, 2012. ** The adoption rates ( ) are expressed by the percentage of land planted with urea briquettes from the total land cultivated in Daule and Santa Lucia, Ecuador Table 4 51. Total Cost and Benefits of producing rice with the UDP technology in Daule and Santa Lucia, Ecuador Years UDPe CBA IHC UDP MSUR ** Inc UDP ** NI UDP 1 0.3310% $7,568 $9,478 $24,009 $60,838 $62,445 2 3.7634% $86,058 $107,772 $272,997 $691,770 $710,042 3 26.6956% $610,457 $764,486 $1,936,525 $4,907,124 $5,036,739 4 53.9027% $1,232,608 $1,543,616 $3,910,147 $9,908,251 $10,169,963 5 58.8726% $1,346,257 $1,685,942 $4,270,672 $10,821,816 $11,107,659 6 59.3257% $1,356,618 $1,698,917 $4,303,539 $10,905,102 $11,193,144 7 59.3636% $1,357,484 $1,700,000 $4,306,285 $10,912,058 $11,200,284 8 59.3667% $1,357,555 $1,700,090 $4,306,512 $10,912,635 $11,200,876 9 59.3670% $1,357,561 $1,700,098 $4,306,531 $10,912,683 $11,200,925 10 59.3670% $1,357,562 $1,700,098 $4,306,533 $10,912,687 $11,200,929 UDPe represents the percentage of planted area with UDP technology from the total area of Daule and Santa Lucia. They present the extra costs of producing rice with the UDP technology caused by the cost of briquette a pplications (CBA) and the c ost by harvesting the extra rice sacks generated by the UDP technology (IHC UDP). ** They represent the benefits of produci ng rice with the UDP technology; which are given by the money saved by the urea re duced with the UDP technology (MSUR) and the income generated by selling extra rice sacks produced with the UDP technology (Inc UDP). NI UDP that would be generated by producing rice with the UDP technology i n Daule and Santa Lucia.

PAGE 120

120 Table 4 52. Sensitive Analysis of Pessimistic and Optimistic Scenarios Scenarios BS ** BS + 10 % BS + 25% BS 10% BS 25% Max AR 59.37% 65.30% 74.21% 53.43% 44.53% Years NI by UDP *** NI by UDP NI by UDP NI by UDP NI by UDP 1 $62,445 $68,689 $78,056 $56,200 $46,834 2 $710,042 $781,046 $887,553 $639,038 $532,532 3 $5,036,739 $5,540,412 $6,295,923 $4,533,065 $3,777,554 4 $10,169,963 $11,186,959 $12,712,454 $9,152,967 $7,627,472 5 $11,107,659 $12,218,425 $13,884,573 $9,996,893 $8,330,744 6 $11,193,144 $12,312,459 $13,991,430 $10,073,830 $8,394,858 7 $11,200,284 $12,320,313 $14,000,355 $10,080,256 $8,400,213 8 $11,200,876 $12,320,964 $14,001,095 $10,080,789 $8,400,657 9 $11,200,925 $12,321,018 $14,001,156 $10,080,833 $8,400,694 10 $11,200,929 $12,321,022 $14,001,162 $10,080,836 $8,400,697 Total $83,083,006 $91,391,307 $103,853,758 $74,774,706 $62,312,255 *Max AR represents the maximum adoption rate of the UDP technology achieved by year 10. ** BS represents the base scenario estimated with the logistic function. *** NI by UDP represents the net income that would be generated by the UDP technology in Daule and Santa Lucia.

PAGE 121

121 CHAPTER 5 CONCLUSIONS Summary The benefits and costs associated with the adoption of UDP are closely correlated In general, farmers with the smallest amounts of land in rice production had the highest reported production costs. As detailed in Chapter 4, the baseline scenario documents that for most pro duction variables (that is, Cost of Soil Preparation/ha, Seed Cost/ha, Urea Cost/ha, Cost of other Fertilizes/ha, Herbicide Cost/ha, Insecticide Cost/ha, Cost of Products Applications/ha, Cost of Fertilizer Applications/ha, Irrigation Cost/ha, and Harvesti ng Cost/ha ), it was statistically determined through the Chi Square Tests and with Crosstab A nalysis that most of the smallest farm s (in terms of hectares of rice grown) are strongly associated with the highest production costs. Exceptions to these general findings include the Rent cost/ha and Seeding Cost/ha, for which no statistical significance across farm size was found. Also, opposite of what was expected, farmers with the largest land size had the highest costs with the Seeding Cost/ha variable and th e Cost of Product Application/ha variable. However, for the rest of variables (as already listed), ,most of the farms in the largest farm size category had the lowest reported production costs. In the analysis of the post harvest variables, it was clearly determined that most of the smallest farmers had the highest per hectare rice yields. For farmers in the other four land size categories there are no statistically significant associations between post harvest variables and land size categories. However, in the analysis of the income generated by selling the harvest as paddy rice it was determined that most of the largest farmers had the highest incomes. This outcome is expected, but it is also

PAGE 122

122 important to note that the reported value of this variable is very dependent upon the suggesting that income from the sale of paddy rice is not an entirely accurate measure of the relationship between farm size and income generation unless the value of on farm co nsumption is included in the analysis of farmers income. However, such an analysis would require the valuation of rice kept for consumption, and that would require assumptions about whether or not farm size could influence price received. Since this study did not collect data to justify a specific assumption about this factor, an average market price, regardless of farm size, was used in the analysis of potential benefits and costs. Another key finding from the farmer survey and subsequent analysis is that farmers in the smallest land size category express the highest willingness to adopt the UDP technology. In addition, the analysis determined that the UDP adoption rate Consistent with these findings is the fi nding that most of the smallest farmers had the highest per hectare urea cost Hence, the smallest farmers would benefi t the most if a new technology would reduce the quantity of t he urea applied which is the case for the UDP technology. Further, the anal ysis indicates that the smallest farm s have the highest per hectare cost of briquette applications, and the highest increment al increase in harvesting cost as a result of UDP adoption ; yet, the benefits from the extra net income resulting from UDP adoption exceed these increased costs. Finally, as farm size increases, the incremental net increase in income per hectare decreases, mainly because per hectare costs of UDP adoption increase incrementally more than per hectare income does.

PAGE 123

123 In the macroeconomic a nalysis, the labor cost of briquette applications is not seen as something negative because it is highly related to employment generation for farm w orkers It is estimated that the UDP technology would be adopted for use on about 60 percent of rice land, a nd as a consequence, this would generate demand for several hundred thousand person days of labor after the first two years that t he UDP technology is introduced in the Daule and Santa Lucia rice growing regions Other macroeconomic benefits of the introdu ction of the UDP technology are the impacts of the urea reduction which would prevent the application of thousand s of tons of urea which would in turn generate thousand s of dollars of saving s for rice farmers In addition, those urea savings would also lead to reduced nitrogen loadings associated with nitrogen run off and volatilization that now occurs with current rice production practices (i.e, broadcasting prilled urea). Finally the UDP technology would promote the reduction of capital outflow from Ecuador caused by urea imports. Finally, the increment al increase in rice production would ensure the annual carbohydrate needs of 1,555 persons in the first year that farmers incrementally adopt the UDP technology. This benefit would increas e over time, and would be around 276 5 00 persons after year 5 based on the Ecuadorian per capita rice consumption reported in 2009 of 122.41 kg (INEC 2012) Politics I mplications T he introduction of the UDP technology in D aule and Santa Lucia will depend upon support from national and international institutions This has been the case in other countries, where UDP technologies have been successfully adopted and diffused (e.g, in Indo nesia ( Pasandaran el al 1999 ) and in Ba ngladesh ( IFDC 2008) ). I n order to promote and manage an efficient diffusion of the technology a coordinated effort to

PAGE 124

124 build the capacity to manufacture briquettes (e.g., subsidizing the purchase and introduction of the machines used to make the briquettes) and to educate farmers about the use and benefits of the UDP technology are most likely to two most critical points of intervention o provide this coordinated effort can be based smallest producers, along with the added benefits identified in the macroeconomic analysis. For example, t he government could directly benefit from the reduction of urea imports and the reduced cost of providing urea subsid ies to smaller farmers (i.e., since 2007, the Ecuadorian government has subsidized fertilizer costs for rice with fewer than 15 hectares Under this po licy, t hese farmers c an receive 2 urea sacks per hectare at a price of 10 USD/sack up to a maximum of 20 sacks ).

PAGE 125

125 REFERENCES Aguirre, D., and I. Medina "Evaluacin de Diferentes Niveles de Nitrgeno Mediante la Aplicacin de Briquetas de Urea como Alternativa Para Pequeos Productores de Arroz (oriza sativa), en la Parroquia San Juan, Cantn Pueblo Viejo, Provincia de los Ros." Bachelor Thesis, Escuela Superior Politcnica del Litoral, 2010. Azam, F ., C. Mller, A. Weiske, G. Benckiser, and J. Ottow Nitrification and Denitrification as Sources of Atmospheric Nitrous Oxide Role of Oxidizable Carbon and Applied Nitrogen. Biology and Fertility of Soils 35 (2002):54 61. Barzola, L., and P. Herrera "Adopcin de la Aplicacin Profunda de Briquetas de Urea (APBU) por Parte de Tres Pequeos Agricultores de la Asociacin Amrica Lomas Cooperativa Nueva Estancia en Sistemas de Produccin de Arroz de la Provincia del Guayas." Ba chelor Thesis, Escuela Superior Politcnica del Litoral, 2010. Borbor, M. Modeling How Land Use Changes Affect the Nutrient Budget in the Guayas Basin Ecuador: Ecological and Economic Implications PhD dissertation, College of Environmental Sciences and Forestry at State University of New York, 2004. Bowen, W., R.B. Diamond, U. Singh, and T.P. Thompson Urea Deep Placement Increases Yield and Saves Nitrogen Fertilizer in Farmers' Fields in Bangladesh. Rice is Life: Scientific Perspectives for the 21st Century. Proceedings of the World Rice Research Conference held in Tokyo and Tsukuba, Japan, 4 7 November 2004 Toriyama, K, Heong K. L., and Hardy B., ed., pp. 369 372. Los Baos, Philippines: International Research Institute, 2005. Calle, O., and I. Med ina. "Anlisis De La Aplicacin Profunda De Briquetas De Urea En El Suelo Como Fuente De Lenta Liberacin De Nitrgeno En La Produccin De Arroz Bachelor Thesis, Escuela Superior Poltcnica del Litoral, 2010. Cao, Z.H., S.K. De Datta, and I.R. Fillery Effect of Placement Methods on Floodwater Properties and Recovery of Applied Nitrogen ( 15 N Labeled Urea) in Wetland Rice ." Soil Science Society of America Journal 48 (1984):196 203. Cho, J.Y. "Seasonal Runoff Estimation of N and P in a Paddy Field of Cent ral Korea." Nutrient Cycling in Agroecosystems 65 (2003):43 52. Choudhury, A.T.M.A., and Y.M. Khanif Effects of Nitrogen, Copper and Magnesium Fertilization on Nutrition of some Macro and Micro Nutrients of Rice Crop in Malaysia 2011. Choudhury, ATMA, and N. Bhuiyan "Effects of Rates and Methods of Nitrogen Application on the Grain Yield Nitrogen Uptake of Wetland Rice." Pakistan Journal of Science and Industrial Research 37 (1994):104 107.

PAGE 126

126 Choudhury, T. M. A., and Y. M. Khanif "Evaluation of Effects of N itrogen and Magnesium Fertilization on Rice Yield and Fertilizer Nitrogen Efficiency Using 15 N Tracer Technique." Journal of Plant Nutrition 24 (2001):855 871. De Datta, S. Principles and Practices of Rice Production New York: John Wiley and Sons, Inc, 1 981. De Datta, S.K., and E T. Craswell "Nitrogen Fertility and Fertilizer Management in Wetland Rice Soils." Rice Research Strategies for the Future. Anonymous pp. 283 316. Los Bafios, Philippines: International Rice Research Institute, 1980. De Datta, S .K., and R.J. Buresh "Integrated Nitrogen Management in Irrigated Rice." Advances in Soil Sciences 10 (1989):143 169. Food and Agriculture Organization of the United Nations http://faostat.fao.org/ edFAO, 2012. IFDC Expansion of Urea Deep Placement (UDP) Technology in 80 Upazilas of Bangladesh during Boro 2008: An Assessment of Project Impact Bangladesh: IFDC, 2008. INEC Sistema Agroalimentario del Arroz Ecuador: Instituto Nacional de Estadsticas y Ce nsos, 2012. Mayorga, J., and P. Herrera "Adopcin de la Aplicacin Profunda de Briquetas de Urea (APBU) por Parte de Dos Pequeos Agricultores de la Cooperativa 25 de Abril y Alianza Definitiva en Sistemas de Produccin de Arroz (Oriza sativa)." Bachelor Thesis Escuela Superior Politcnica del Litoral, 2010. Mora, S., and P. Herrera "Comparacin de Dos Tecnologas De Aplicacin De Nitrgeno (Urea) en Diferentes Niveles en el Cultivo De Arroz: Aplicacin Profunda de Briquetas de Urea y la Aplicacin Tradi cional Al Voleo." Bachelor Thesis, Escuela Superior Politcnica del Litoral, 2010. Pasandaran, E., B. Gulton, J. Sri Adimingsih, H. Apasari, and S ri Rochayati "Government Policy Support for Technology Promotion and Adoption: A Case Study of Urea Tablet T echnology in Indonesia. Nutrient Cycling in Agroecosystems 53 (1999):113 119. Phupaibul, P., N. Chinoim, and M. Toru "Nitrate Concentration in Chinese Kale Sold at Markets Around Bangkok, Thailand." Thai Journal of Agricultural Science 35 (2002):295 302

PAGE 127

127 Reeves, T.G., S.R. Waddington, I. Ortiz Monasterio, M. Banziger, and K. Cassaday "Removing Nutritional Limits to Maize and Wheat Production: A Developing Country Perspective." Biofertilizers in Actions. A report for the Rural Industries Research and Dev elopment Corporation. Kennedy, I., and Choudhury Abu T. M. A., ed., pp. 11 12 36. Australia: Rural Industries Research and Development Corporation, 2002. Saenz, C., and P. Herrera "Adopcin de la Aplicacin Profunda de Briquetas de Urea (APBU) por Parte d e Tres Pequeos Agricultores de la Asociacin America Lomas en los Sectores Brisas de Daule, Huachichar y la Rinconada en Sistemas de Produccin de arroz (Oriza sativa)." Bachelor Thesis, Escuela Superior Politcnica del Litoral, 2010. Savant, N. and P. S tangel "Urea Briquettes Containing Diammonium Phosphate: A Potential New NP Fertilizer for Transplanted Rice." Nutrient Cycling in Agroecosystems 51 (1998):85 94. SINAGAP III Censo Nacional Agropecuario, http://www.magap.gob.ec/sinagap/ index.php?option=c om_wrapper&view=wrapper&Itemid=224 ed. Ecuador: Ministerio Nacional de Agricultura, Ganadera, Acuacultura y Pesca MAGAP, 2012. Valiela, I., and J. Bowen "Nitrogen Sources to Watersheds and Estuaries: Role of Land Cover Mosaics and Losses within Watersh eds." Environmental Pollution 118 (2002):239 248. Vitery, V. "Aspectos Econmicos del Cultivo del Arroz en Ecuador." Manual del Cultivo del Arroz Manual No. 66. Anonymous pp. 145 146 161. Guayaquil, Ecuador: Instituto Nacional Autnomo de Investigacio nes Agropecuarias, 2007.

PAGE 128

128 BIOGRAPHICAL SKETCH Samuel Mora was born in Guayaquil, Ecuador in 1987. After graduating from udies at Escuela Superior Polit cnica del Litoral ESPOL in Guayaquil, Ecuador from 2005 to 2010 w h ere he obtained a bachelor in science degree in agricultural and biological engineering. Food and Resource Economics Department at t he University of Florida. Samuel worked with Dr. James Sterns and Dr. Pilar Useche in a Cost B enefit Analysis study for the Adoption of the Urea Deep Placement Technology by the rice farmers of Daule and Santa Lucia, Ecuador. He graduated in December 2012 and resource economics and is prepared for new challenges and opportunities in the agricultural and economic field s