-umMmmmmrqyumzflurul -..Awum. ..w . .....wm. .. fiwfimfigwfi. .3 c: 1. flu. . a 3?. . w ...» ...: .59}... .1- .u.. csofino . n W%% . 1 u r . ...... . . “my... A. q. . 3.5.314. 49 . 3...... ...... a... a ....ma. ,. _ 1%.”. aflfié . . . . . . h. I .ah .....I .3Lu.c$ .1, '4. flu... ..r. . . 4W . ....fi. :me «r? «m o . y I» , 3... ... . 10.. 5? SM}... ..15. , w. $3»? I» “.1. . w.“ . .. ..-.mudvfifi. . in...” . .W .3: ...]e in,» . i .. .Mwuwn um. .umi ,. . . Hahn". . ax... .. 4.93555uw . .gnsxnu: .. . . .1". . ,. . , .... .. . . L It 4 «judwwafluw... 5.... a . nu... ...na. ”v.14. 5 .fi' ‘1. fl U “7‘1“: a :V 1 U12" 7 “tlvi’lt h‘ . [A -'L- 1 M! t o “I; erI I A v 6| 8 (I 2‘? ‘0 by“... . 4.. .I . . v. I”. a: .0 w u... ..c f Jab..sr¢msmaq . (56......WWJ . . . I :- It . . . . “in. .. ...LBHJ . . . . , .‘varvuur. . Lflfn/ wJAflIHNMrHS . . ...": 14m... .4...» . . . .. .31. ...Il. . . . . . ..W.“: 3.4.. . . . . «.-.. , . . . . “Ndmnu . a L9,}... H... .. ... .-.. H . . . . ..fic‘flwfimiw . u . . . . . . .. . ......c... . . . . . . I . . . .6514“. . l . .l . : u§Jv>.hl..1.hhul.3 .d . u. x . .. z“... «www.m‘. MUMWHWN.» He .. -5 l l,. .55 1‘ I. § 4! an. 3 u! ti 4 . $ AM ‘ p f. . . 1' . ’1"d|.l\[ ...“..oos..u I 1.31%.... lvflvlabcfibu.ix. . .... . fl... . ulu..x.i~ QIA}¢..PD£ :1... v r... . ... A.Ih1.\ ‘ll. . . 7 .1... r4 ...va3 0 ..I ‘ t. L 1V .LHQIO I .. . 1.. van. v u: a ..l‘ rill. . 1.. 1:48? Ibis... '.".i‘:‘“J-l ;ll.:‘ “"4,“ ht'. .vblp.lr| _ C .....u .HMML .v. . . All.) 1-4... M u I". rHESlS llllllllllilllllIllllllllIlllHillllllllllllllllllllllllllllll 31293 01691 4313 This is to certify that the dissertation entitled Impact of Large Irrigation Projects on Participant Ejidatario-Farmers Over Time: The Panuco River Irrigation Project Between 1987 and 1995 presentedby Esteban Valtierra-Pacheco has been accepted towards fulfillment of the requirements for _M_demee in We] opment Date Mfli7 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Michlgan State Unlverslty PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINE. return on or before date due. DATE DUE DATE DUE DATE DUE jér . z 312911.: ma crew-mu IMPACT OF LARGE IRRIGATION PROJECTS ON PARTICIPANT EUIDATARIO-FARMERS OVER TIME: THE PANUCO RIVER IRRIGATION PROJECT BETWEEN 1987 AND 1995 BY Esteban Valtierra-Pacheco A DISSERTATION Submitted to Michigan State university in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Resource Development 1997 ABSTRACT IMPACT OF LARGE IRRIGATION PROJECTS ON PARTICIPANT EUIDATARIO-FARMERS OVER TIME: THE PANUCO RIVER IRRIGATION PROJECT BETWEEN 1987 AND 1995 BY Esteban Valtierra-Pacheco This research looks at Mexican large irrigation projects can deliver the intended benefits to poor participant farmers (i.e., ejidatario—farmers). This study focuses on the Panuco River irrigation Project (PRIP). The research design for is a panel design. The research included a survey conducted in 1995 in which 150 ejidatario- farmers. The results of this survey were compared with a survey conducted among 238 ejidatario-farmers in 1987. These surveys had a common sample of 57 ejidatario-farmers who were interviewed in both years. The null hypothesis for this study is "The PRIP did not meet its initial objectives: a) to develop an intensive irrigation system, b) to contribute to national food security, c) to reduce the uneven income distribution between the rural and urban populations, and d) to reduce the unemployment among landless peasants." The variables used to test the null hypothesis include perceived efficiency of the irrigation system, amounts of food crops produced by ejidatarios in 1987 and 1995, ejidatarios’ income (i.e., annual, on-farm and off-farm incomes), and land rent. The results of this study indicate that the PRIP failed to achieve its initial objectives. The PRIP did not develop an efficient irrigation delivery system. The production of food crops cultivated by ejidatarios experienced a dramatic reduction between 1987 and 1995, except for sugar cane. The ejidatarios total annual income did not increase significantly. The average ejidatario’s income was below the poverty line both in 1987 and 1995. Ejidatarios on-farm income had a significant reduction between 1987 and 1995. The ejidatarios’ farm lost its importance as primary source of income and employment because ejidatarios worked more off-farm in 1995 than in 1987 and many of these ejidatarios rented their land to outsiders in 1995. This study indicates that large scale irrigation projects, like the PRIP, did not always help ejidatarios to overcome poverty. Large scale irrigation projects are not always the best ways to address poverty among the nation's poor or increase the national food security. Copyright by Esteban Valtierra-Pacheco 1997 My degree is an achievement that belongs to my family. Their love kept me going. To my wife Ignacia To my daughter Nancy To my son Erick ACKNOWLEDGEMENTS I thank Dr. Scott G. Witter who served as my dissertation director and the chairman of the guidance committee for his support that I received from him during the four years I was in MSU. I would like to thank him for being always ready to advise me and for all the continuous favors he gave me. A special appreciation goes to Dr. Scott Whiteford who was more than a advisor. I always counted with his friendship and words of encouragement. Thanks to Dr. George Axinn and Dr. Mary Andrews for their contributions to this dissertation and their patience to review this dissertation. I am indebted to the institutions that make my studies and this dissertation possible, particularly my home institution in Mexico, the Colegio de Postgraduados. My studies were possible, thanks to the support from the W.K. Kellogg Foundation that awarded me a scholarship to support my family during these four years and funding to my dissertation research. I would like to thank all the W.K. Kellogg Foundation’s staff for their support and kindness, especially to Rebecca Hernandez, Robert DeVries, Maria Martin, and Heliodoro Diaz. Additional funding for the field research of this dissertation was provided by the "Ejido Reform Research Project (ERRP)" of the Center for U.S.-Mexican Studies at the University of California, San Diego. This center awarded me a Grant for Graduate Student Field Research on the Transformation of Rural Mexico. I appreciate the revision of the grammar made by Emily Holley, who had the patience to get through unintelligible sentences. Thanks to all my friends who helped me to make easy these four years at Michigan State University. vi TABLE OF CONTENTS LIST OF TABLES ........................................ ix LIST OF FIGURES ....................................... x LIST OF APPENDICES .................................... xi ABBREVIATIONS ......................................... xii CHAPTER I. INTRODUCTION ..................................... l A. Problem Statement ........................... 1 B. Research Objectives ......................... 9 C. Hypotheses .................................. 9 D. Study Methods ............................... 12 II. LITERATURE REVIEW ................................ 13 A. Contribution of Irrigation to Food Production 13 B Large Irrigation Projects ................... 15 C. Irrigation Project Objectives ............... 18 D Stakeholders Participation on Irrigation Projects ................................. 30 E. Evaluation of Irrigation Projects ........... 39 III. STUDY SITE ....................................... 44 IV. METHODOLOGY ...................................... 53 A. Research Design ............................. 53 B. Methods of Data Collection .................. 54 C. Operationalization of the Main Concepts ..... 55 vii D. Sampling Methods ............................ E. Data Collection Process ..................... F. Software used for Statistical Data Processing G. Advantages and Limitations of the Present Study .................................... V . FINDINGS ......................................... A. Development of an Intensive Irrigation System B. Contribution to National Food Security ...... C. Reduction of the Uneven Income Distribution between Rural and Urban Population ....... D. Reduction of Unemployment among Landless E. Summary ..................................... VI. CONCLUSIONS AND RECOMMENDATIONS ................. APPENDICES ............................................ BIBLIOGRAPHY .......................................... viii 6O 63 65 65 67 69 82 89 108 122 Table Table Table Table Table Table Table Table Table Table Table Table LIST OF TABLES 1 - Ejidatarios’ Perception of the Efficiency of the Irrigation System in 1995 .............. 70 2 - Adequacy of the Amount of Irrigation Water in 1995 .................................... 76 3 - Timeliness of the Irrigation Service Delivery in 1995 ........................... 78 4 - Main Crops Cultivated by Ejidatario-Farmers in 1987 and 1995 (Paired Subsample) ........ 84 5 - Synopsis of the Ejidatario-Farmers’ Net Income in 1987 (General Sample, n=238) ..... 92 6 - Synopsis of the Ejidatario-Farmers’ Net Income in 1995 (General Sample, n=150) ............ 93 7 - Synopsis of the Ejidatario-Farmers’ Net Income in 1987 (Paired Subsample, n=57) ........... 100 8 — Synopsis of the Ejidatario-Farmers' Net Income in 1995 (Paired Subsample, n=57) ........... 101 9 — Ejidatarios’ Annual Mean Income Differences between 1987 and 1995 (Stratified Paired Subsample) ................................. 105 10 - Ejidatarios’ On—Farm Mean Income Differences between 1987 and 1995 (Stratified Paired Subsample) ................................ 107 11 - Ejidatarios’ Off-Farm Mean Income Differences between 1987 and 1995 (Stratified Paired Subsample) ................................ 117 12 — Ejidatarios' Mean Income Differences for Land Renting between 1987 and 1995 (Stratified Paired Subsample) ......................... 121 ix LIST OF FIGURES Figure 1 - Diagram of a Quasi-Experimental Model of Evaluation and Monitoring of an Agricultural Development Project .......... 42 Figure 2 - Map of the Panuco River Irrigation Project .. 45 LIST OF APPENDICES Appendix A - Survey Questionnaire ...................... 139 Appendix B - Demographic Characteristics of the Ejidatario—Farmer Families in 1987 at the Panuco River Irrigation Project ........... 165 Appendix C - Demographic Characteristics of the Ejidatario-Farmer Families in 1995 at the Panuco River Irrigation Project ........... 166 xi ANAGSA AGROASEMEX BANRURAL CBDA CP CNA CONASUPO FAO INEGI NAFTA ODA PIDER PRIP SARH UNDP ABBREVIATIONS National Insurance Company for Agriculture and Livestock Aseguradora Nacional Agricola y Ganadera SMA. Mexican Agricultural Insurance Company Aseguradora Agricola Mexicana National Bank of Rural Credit Banco.Nacional de Crédito Rural Chad Basin Development Authority Graduate College on Agricultural Sciences Colegio de Postgraduados en Ciencias Agricolas Water Commission Nacional del Agua Nacional Cbmision Nacional Compafiia Company of Popular Supplies Nacional de Subsistencias Populares Food and Agriculture Organization National Institute for Statistics, Geography and Informatic Instituto Nacional de Estadistica, Geografia e Informética North American Free Trade Agreement Overseas Development Agency Programa Integral de Desarrollo Rural Integral Integrated Rural Development Program Panuco River Irrigation Project Proyecto de Irrigation de la Cuenca Baja del Rio Panuco Secretariat of Agriculture and Water Resources Secretarfa de Agricultura y Recursos Hidrafilicos United Nations Development Programme xii CHAPTER I INTRODUCTION A. Problem Statement Problem Background For many years, construction of irrigation projects was considered the most suitable way to increase agricultural productivity and reduce poverty in many Third World countries. More than twenty years ago, Orive-Alba (1970: 235) indicated that "the most important improvements on agriculture in the twentieth century have taken place in irrigated lands and new irrigation projects were good alternatives to alleviate rural poverty and to redress agrarian problems." As time has passed, optimism about how irrigation could redress problems of poverty, hunger and food security has declined. More than a decade ago, a Food and Agriculture Organization (FAO) document (Sagardoy, 1982) expressed some concerns about rapid increases in the world’s population which had made the efficient use of irrigation water critically important, particularly in poorer countries where the greatest potential for increasing food production and l 2 rural incomes was often found in irrigated areas. However, the poor performance of many irrigation schemes became a matter of serious concern. Despite their very high cost, many irrigation projects fell far short of their initial expectations. Recently, the World Bank (1994: 2) noted that the previous optimism has become a great concern. Major investments were made in irrigation infrastructure stocks, but in too many developing countries these assets did not generate the quantity or the quality of services demanded. The costs of this waste -—in forgone economic growth and lost opportunities for poverty reduction and environmental improvements-— were high and unacceptable. The main failures in irrigation infrastructure projects have been inadequate design, deficient drainage, operational inefficiencies, inadequate maintenance, excessive dependence on fiscal resources, unfair practices of water distribution, lack of responsiveness to users' needs, limited benefits to the poor, lack of equity objectives, lack of stakeholders’ participation and insufficient environmental responsibility (World Bank, 1994: 4-5). The World Bank attributes these failures to the direct participation of governments in the economy. Government intervention distorts the conditions of a free market, resulting in unwarranted allocation of resources and distribution of benefits among the economic sectors of a society. However, these conclusions are product of the change in the World Bank’s policy. In the 3 past, the World Bank encouraged the direct participation of government in the economy for many years. In addition, many of the irrigation project failures were a direct result of the way in which the World Bank provided the loans. Sometimes those loans were a direct intervention of the World Bank in national affairs of Third World countries. Those limited benefits to the poor have been continually reported, but governments have done little to understand the problem, explore the causes, or readdress the problem identifying the poor as project priority. Research Question Who is getting the benefits of irrigation, the poor or the rich? Perhaps nobody is benefitting when irrigation projects fail. Since there is a lack of research information regarding these questions, this research looks at the following: Can Mexican large irrigation projects, like the Panuco River Irrigation Project, deliver intended benefits to poor farmers (ejidatarioSU, especially as the new government initiatives underway reduce government 1 Ejidatarios or ejidatario-farmers are small farmers organized in a community based group, called ejido. Originally, ejidatarios could own ejido land collectively, but they could cultivate their land collectively or individually. Ejidatarios could use ejido land but not sell it. In 1991, a constitutional amendment allowed ejidatarios to transfer their land tenure from collectively owned land to individual privately owned land and, therefore, to sell their land. 4 agricultural support services? Previous Studies There are no studies in Mexico that have assessed the impact of large irrigation systems on poor ejidatario- farmers across time, especially when government support for irrigation has become more scarce. It is important to know if poor farmers have increased or reduced their income over time; if the poor farmers’ economic returns have been sustainable from the use of irrigation water; how poor farmers have balanced their income between on-farm and off- farm economic activities; and how the poor farmers have survived, while the Mexican economy has undergone one of its greatest structural transformations ever. In Mexico, most of the recent social and economic studies on irrigation have centered their attention on how the transference of infrastructure management has improved the operative and economic efficiency of irrigation districts. However, they have put aside important aspects such as who is having access to water or who can pay the new water fees. The Mexican government justified the transference of irrigation infrastructure management from the government to user associations because the operational cost of irrigation projects were not covered by water fees (Gorriz et al., 1995). In 1988, the water fees only covered 15% of the operative water costs and the Mexican government 5 maintained a large percentage of subsidies. At the beginning of the 19908, the situation was unbearable because the fiscal deficit crisis resulted in an increasing deterioration of the irrigation infrastructure resulting from the lack of maintenance. As a response to this problem, the World Bank and the Mexican government suggested that the only way to assure water irrigation supply was to transfer irrigation districts to users (Palacios, 1994: 113- 115)2. Irrigation management transference is not an exclusive Mexican government policy, it is a global policy being encouraged by the World Bank and implemented by many countries around the world (World Bank, 1994; Johnson, 1994; Jones, 1995). The transference can be a good way to encourage efficiency and achieve users’ expectations, the problem is that the current policy to transfer irrigation systems to users is a product of international pressures to reduce government expenditures. For this reasons, the transference of irrigation management has been mainly analyzed in terms of how much money governments have saved by eliminating subsidies and increases in water fees (Johnson, 1994). However, the current and potential impact on human capital has received little attention from donors 2 By March of 1994, the Mexican government had partially or totally transferred infrastructure management in 51 irrigation districts, which encompassed 2,314,870 hectares --75% of the total irrigated land in irrigation districts. and host governments. The transference of the irrigation infrastructure and reduction of government support have been successful in improving the efficiency of the irrigation management and service delivery system. The question is who gains and who loses? The studies noted above do not consider equity or distributional objectives as a measure of success in these irrigation projects, even though, it has been demonstrated in other countries that when equity is achieved, other indicators of project success are achieved (e.g., water delivery) (Shivakoti, 1991). Water fees and water rights are important parts of the recent government irrigation policies. Different types of water fees can have a great impact on the performance of irrigation schemes. Marginal productivity has been suggested as a fair criterion to charge farmers for irrigation water (Bernal, 1994). However, in Mexico, water fees have never been established in agreement with irrigation users. When irrigation users are not in agreement with water fees and are therefore unwilling to pay, the funds available for maintenance and operation may be drastically reduced (Valdivia, 1994; Levine & Garces- Restrepo, 1994). Technical problems are still a great concern for users, the Mexican government and donor agencies (Sotomayor, 1994; Palacios-Sanchez, 1994; Chacon, 1994). The lack of control of ordinary operations can be a very important threat to 7 irrigation productivity. Salinization, waterlogging, and erosion perturb large portions of the irrigation districts in Mexico (Cortez, 1994). Few studies focus on income distribution and irrigation users’ well-being (Valdivia, 1994). However, these studies have limited their scope to the current situation without establishing how farmers' situations improve or deteriorate over time. In addition, these studies do not focus on the rural poor. Chambers (1991) suggests some possible explanations for this situation, such as the fact that these studies have been conducted by "positive practitioners", who only see the potentially positive benefits of the irrigation infrastructure on agricultural productivity, but do not see the overwhelming problems of the rural poor users; or the poor are too weak and are, thus, unknown and unseen; or knowing the overwhelming problems of rural poverty, researchers feel powerless to offer a solution or to participate in seeking one. To overcome the limitations of these studies, this research assesses the impact of government-run irrigation projects on poor farmers in Mexico. Thus, anyone interested in improving the income and well-being of poor irrigation users in Mexico may gain important insights from this study. Representatives of irrigation user organizations need to understand how the recent government policies on irrigation may affect the economy of their affiliated farmers. At the academic level, it is necessary for scholars to bring the 8 poor into the intellectual discourse because the opinions of the poor have rarely been taken into account. They have disappeared from the intellectual discourse or have appeared in pseudo-theories such as the "Social Liberalism"3 that was coined by the former Mexican President Salinas for electoral purposes. This study assesses the impact of a large government- supported irrigation system on ejidatario-farmers’ households, comparing their situation over two time periods, 1987 and 1995. A survey of 238 ejidatarios was performed in the Panuco River Irrigation Project in 1987 (Diaz, et al., 1988; World Bank, 1989; Diaz & Valtierra, 1992). To make the comparison possible, a similar study was administered to 150 ejidatarios in PRIP in 1995. This study encompasses a time period of eight years in which the Mexican government drastically reduced its support for irrigation. The study site selected was the Panuco River Irrigation Project (PRIP) which is located about two hundred miles south from the Mexican-U.S. border, at 108° WG, 22° N. PRIP is composed of three irrigation units which encompass 144,000 ha. This irrigation project gave land to 8,500 ejidatarios. The PRIP was created in 1974, when the World Bank approved a loan to be disbursed between 1974 and 1982. 3 This theory proposes that free market driven economies promotes social justice for most of the people. If it does happen, a direct subsidy (i.e., money in hand) should be given to the poor. 9 B. Research Objectives The research objectives for this study are: To estimate the extent to which the benefits from large scale government supported irrigation projects have been sustainable, using the Panuco River Irrigation Project as an example. To estimate whether the Panuco River Irrigation Project was able to meet its original objectives: a) to develop an intensive irrigation system, b) to contribute to national food security, c) to reduce the uneven income distribution between the rural and urban populations, and d) to reduce unemployment among landless peasants (World Bank, 1989: i). C. Hypotheses Hgl: The Panuco River Irrigation Project did not meet its initial objectives: a) to develop an intensive irrigation system, b) to contribute to national food security, c) to reduce the uneven income distribution between the rural and urban populations, and d) to reduce the unemployment among landless peasants. Hgl: The Panuco River Irrigation Project met its initial objectives: a) to develop an intensive irrigation system, b) to contribute to national 10 food security, c) to reduce the uneven income distribution between the rural and urban populations, and d) to reduce the unemployment among landless peasants. Hypothesis Testing For operational purposes the null hypothesis is broken down into four subhypotheses; each null subhypothesis corresponds to each objective of the PRIP. The subhypothesis were formulated following the reasoning of an evaluator who expects that the initial expectations were covered and the objectives were accomplished. The order for the hypothesis testing, the null subhypothesis statement and the variables used for hypothesis testing are the following: a) The first subhypothesis is related to the objective "to develop an intensive irrigation system". This null subhypothesis states that ejidatarios perceive that PRIP did not develop an efficient irrigation delivery system (Ho:1nnm). The variables selected are related to the ejidatarios’ perception of the state of the irrigation infrastructure and the efficiency of the irrigation system, farmers' participation and the ejidatarios' willingness to pay. b) The second relates to the PRIP contribution to the national food security. The null subhypothesis states c) d) 11 that ejidatarios did not produce more food crops for the national food security in 1995 than in 1987 (Ho: Ab=#1)- The variables included in the hypothesis testing were regarding the improvement of yields per hectare and total amount of food crops per ejidatario and value of crops per ejidatario between 1987 and 1995. The third objective states that PRIP would contribute to the reduction of the uneven income distribution between the rural and urban populations. The null subhypothesis tested is that ejidatarios did not increase their income from 1987 to 1995 (H0:;h>pfl. Student t—tests are performed on the annual and on-farm income mean differences between 1987 and 1995. The fourth objective states that PRIP would contribute to the reduction of unemployment among landless people. The null subhypothesis states that the ejidatarios off- farm work increased from 1987 to 1995, Ho:;h>>>>> Beneficiaries >>>>> Beneficiaries ........ A.....................A...................A........ A A A A A A Base Line Impact Ex-post Study Evaluation Evaluation v v v V V V ........ v.....................v...................v........ Non- Non- Non- Beneficiaries >>>>>> Beneficiaries >>>>> Beneficiaries I ...................... WITHOUT PROJECT....................: Figure 1. Diagram.of a Quasi-Experimental Model of Evaluation and Mbnitoring of an Agricultural Development Project Evaluation research advocates claim that evaluation research promotes the combination of qualitative and quantitative methods of data collection (Cook and Reichardt, 1986; Patton, 1987; Reichardt and Rallis, 1994). However, there is a general tendency to favor quantitative methods over qualitative (Casley, 1987; Casley, 1988; Casley and Kumar, 1989). Surveys have been the most broadly used method for evaluation research (Rossi et al., 1983; Smith, 43 1992; Valtierra, 1989). A survey is an appropriate method of research when individual people are the units of analysis, and when there is a large population which is difficult to observe (Barbie, 1989: 237). This study mainly relies on two surveys, but other methods were used, such as group interviews and interviews with qualified informants. One of the main advantages of survey methods over qualitative methods is that the comparison of data in different times is easier, even though researchers or evaluators may be different every time. This is demonstrated in the evaluation of the Puebla Project. This evaluation has been probably the most ambitious process of evaluation ever performed on a development project in Mexico. Four major evaluation surveys collected the same basic information in 1967, 1970, 1983, and 1985. In 1985, the four surveys were reanalyzed and compared (Diaz et al., 1992; Diaz et al., 1993). The comparison of the four evaluations would have been more difficult if qualitative methods were utilized. However, this does not imply that it is impossible to use qualitative methods for a panel study. CHAPTER III STUDY SITE The Panuco River Irrigation Project (PRIP) is located in northeastern Mexico, about two hundred miles south from the Mexican-U.S. border. The geographic coordinates are 108°‘WG, 22° N. The three irrigation units that compose the Panuco Projects are situated on the border of three Mexican states: Tamaulipas, Veracruz, and San Luis Potosi (see figure 2). The climate is tropical semiarid with 900 mm of rain a year and a mean temperature of 25%L The Panuco River Basin is the fourth largest basin in Mexico. It covers 85,000 square kilometers and drains 18,500 million cubic meters of water per year. Its main tributaries are Moctezuma, Chicayan, Tantuan, Tampaén and Guayalejo Rivers. PRIP has been one of the largest irrigation projects that the Mexican government has built in the last three decades. The Mexican government built most of the irrigation infrastructure between 1926 and 1966 (2,543,302 hectares) (Drive, 1970: 179). During the "mega irrigation projects" period between 1941 and 1958, Mexico expanded irrigation on 1,922,432 ha of land. In 1990, there were 3,346,936 ha of irrigated land (Morett, 1992: 175). If the full potential of PRIP is considered (144,000 ha), it 44 45 United Slales Eslacion Gonzalez Eslacion Manuel III "' IIII “El III/IS ANIMAS UNIT , . “II“ a! Inn IIIII' III' - u. . ".an Bank "I .IIIIII . “II Cuauhlemoc 5&6- \\ ‘. doqoon (Harland / PanUCo «D Iomuin _ . 4, . ”er \ 0°“ anal \ 10m? IRRIGATION UNITS (mm) Animas UniI \ \ Rivers . ,‘ C C UNH S Chicayan Unil Mam Canals 7%,/¢ — Dams ‘ i ’ . @ PuIaI—Coy UnII ,. Roads cn' Da”// 3580“ S Kayo" "I; ’3 Reservoirs 4—4—4—4 Railroads '3 —————— Stale Boundaries ‘ Lagoons O 5 10 I Towns Chicoyan Kilometers C) O orxaw :IO 17 r5 I \, Figure 2. Map of the Panuco River Irrigation Project (Base Map from World Bank (1989) modified after maps courtesy of the Secretariat of Agriculture and Water Resources (1994)) 46 represents 18% of all the land incorporated for irrigation between 1967 and 1990. Instead, if the actual irrigated land is considered, PRIP represents 10%. The Panuco River Irrigation Project was chosen for this research because of its complexity. PRIP shares many of the structural features and problems of the irrigations systems in other regions of Mexico, specifically those irrigation project built by the Commissions of Hydrological Watersheds5 (Barkin and King, 1978). PRIP’s complexity includes several redesigns of the original physical infrastructure project of two dams that irrigate the "Las Animas" and "Chicayan" units, and one very large pump house to irrigate the "Pujal-Coy" unit. However, PRIP’s social situation is even more complex. For instance, the dams along the Panuco River Basin were studied during the 19608, but constant opposition from big landlords and ranchers delayed the construction of PRIP until 1974, when the World Bank approved the loans (Orive-Alba, 1970: 235; World Bank, 1989: 33). The Panuco River Irrigation Project was designed in 1971/72 by the Mexican Government. On February 19, 1974, 5 Commissions of Hydrological Watersheds were regional agencies created to coordinate the integral development of the main Mexican river basin areas (Panuco, Balsas, Papaloapan, Lerma-Santiago, etc.). These commissions followed the model of the Tennessee River Valley Authority (Barking and King, 1978). The commissions were created between the 19508 and 19608. However, in the 19808, most of them were dismantled, when the Mexican government abandoned the development approach by hydrological watersheds. 47 the World Bank approved a loan. The initial estimate of the project cost was US$ 197 million to be disbursed between 1974 and 1980. In 1978, a World Bank evaluation estimated that the construction costs were 35% over the budget, although only 65% of the infrastructure was constructed at the time. The World Bank provided an additional loan of US$ 25 million to complete the construction of PRIP. The final total cost was US$ 446 million, 126% more than the initial budget (World Bank, 1989: 33-34). In PRIP, the Mexican government expected to overcome most of the physical and social problems of prior irrigation projects. PRIP had four basic objectives (World Bank, 1989: i): a) To develop an intensive irrigation system, b) To contribute to the national food security, c) To reduce the uneven income distribution between the rural and urban populations, and d) To reduce the unemployment among landless peasants The four objectives were closely related. The first was the starting point, the construction of irrigation infrastructure was a means to achieve the other three objectives. The second was a societal objective to produce more food for an increasing Mexican population. To achieve the third objective, the Mexican government helped ejidatarios increase their income opportunities on their farms. The fourth objective was related to how the ejidatarios’ farm provided permanent employment to 48 ejidatario families. The implementation of PRIP had two main components: the construction of irrigation infrastructure for 144,000 ha6 and the agrarian reform process to give land to 8,500 ejidatario-farmers. Both processes had to be completed in eight years, between 1974 and 1982 (World Bank, 1989: 34). PRIP involved one of the larger agrarian reform processes ever7 for a Mexican irrigation project. The Mexican government and the World Bank provided ejidatarios with enough resources to create a prominent agricultural development enclave close to the Mexico-U.S. border market. Each ejidatario received 10 hectares of irrigated lands (the national average of land per ejidatario is 6.3 hectares, 4.4 hectares of them rainfed, and only 1.09 hectares irrigated). The PRIP policy makers hoped to provide enough land to generate the income needed to maintain an average ejidatario family. Even though, the family has changed after the beginning of the PRIP, the main structure of the family has ‘ The initial PRIP area was 137,000 hectares. Later, the Animas Unit was expanded by 7,000 hectares. 7 According to Orive-Alba (1970: 199-212) the 1969 amendments to the Water Law allowed the Mexican government and private owners to trade water for land in the new irrigation districts, but not within the old districts. With this measure, the Mexican government could extend the benefits of irrigation to more people. In 1969, the new legislation was only applied to the Carrizo Irrigation District. 49 been maintained“. PRIP promoted a drastic change in the previous land use pattern. The previous land use was: 44,800 hectares under cultivation (4,800 hectares were irrigated and 40,000 hectares rainfed), the grazing area was 38,800 hectares of natural grasses, and the area with native vegetation covered about 53,400 hectares. The traditional farming system of subsistence crops and non-intensive cattle production did not utilize the real potential of land. Planners estimated that this land could produce very high yields of cash crops, for example onions and vegetables, which had a good market on the US border. The area of crops was expanded from 44,800 hectares to 115,922 hectares in 1982 (49,229 rainfed hectares and 66,693 irrigated hectares) (World Bank, 1989: 40). The cropping intensity3 was calculated at 61% before the PRIP. It means that only six out of every 10 hectares were cultivated from year to year and the other four were not cropped. Planners anticipated that cropping intensity would increase up to 127% because many of the new irrigated lands could be cultivated twice a year (World Bank, 1989: 8 There is no information regarding the demographic characteristics of the PRIP participants before or at the beginning of the project. The appendices B and C contain the participant characteristic in 1987 and 1995. ’ Cropping intensity refers to the numbers of crops per year in an area, if all the land in the area is cropped once a year, the cropping intensity is 100%. If it is under 100%, some land is not cropped. If it is above 100%, some land is being cropped more than once a year. 50 42). In 1983, PRIP reached a peak of cropping intensity at 118%. After that year, there has been a constant decline. The crop pattern has reverted back to the previous PRIP configuration because many farmers converted their crop land into grazing land and cash crops (e.g., tomatoes) have been substituted by subsistence crops (e.g., corn) (World Bank, 1989: 40). Another important component of the PRIP strategy was the compulsory organization of all the ejidatarios into collective ejidos. Under this process, groups of approximately 50 ejidatarios were organized into collective ejidos to constitute medium enterprises. The purpose of those collective ejidos was to utilize all the resources provided by the Mexican government and the World Bank in an efficient way. Land, infrastructure, water, credit, machinery, fertilizers, etc., were provided for collective work. During the cropping season, every ejidatario received a salary for each day of work. At the end of the cropping season, each member of the collective ejido received a proportional share of the harvest. After a few years of working together, ejidatarios began to abandon the collective work after the bankruptcy of most collective ejidos. The first ejidos started to distribute their land among individual ejidatarios starting at the beginning of the 19808. In the 19908, most of the ejidatarios were working alone. In 1995, a few ejidos still 51 worked collectively. The PRIP was created under a paternalistic development approach. Under this approach, the Mexican government provided some other agricultural services, such as credit, crop insurance, administration of irrigation infrastructure, agricultural research and extension, commercialization, input supply, and machinery lease. In 1995, only commercialization of grains was provided on a limited basis by a state company, CONASUPO. The rest of the services are no longer being provided by the government. Credit was a great means of support for ejidatarios, who did not have economic resources. After ejidos abandoned collectivism, a government bank (BANRURAL) organized groups of eight to ten ejidatarios into grupos solidarios (solidarity groups). The objective of solidarity groups was to guarantee the repayment of BANRURAL loans. In the period between 1992 and 1993, the federal government stopped providing agricultural credit for all the ejidatarios. Currently, farmers can obtain credit if they have collateral to guarantee the repayment of loans. The crop insurance -while it existed- was always a matter of dispute between ejidatarios and government. The National Agricultural Insurance Company (ANAGSA) was dissolved by President Salinas at the end of the 19808 because of corruption among the insurance company employees. From that point, most ejidatarios have not had insurance for their crops and livestock. 52 In the case of the extension service, after several years of relatively good performance, most of the extensionists were fired between 1992 and 1994. All extensionists who kept their jobs were assigned to bureaucratic activities far away from the farmers' fields. In 1990, the Mexican government initiated a program to transfer the irrigation infrastructure management to the user associations. According to a former official of the Water National Commission, the PRIP is one of the last irrigation districts programmed to be transferred to users’ associations because it is one of the most problematic irrigation districts in Mexico. On a national level, in March of 1994 more than 75% of the irrigation districts had been transferred to user associations (Palacios, 1994:115). However, by July of 1994, only one out of more than ten PRIP modules10 had been transferred, "Jaboncillo" (2,872 hectares). “ The Mexican government administers irrigation throughout Irrigation Districts. PRIP is administrated by the 092 Irrigation District Office which is located in the city of Tampico. Irrigation districts are divided into Irrigation Units. Irrigation units have subdivisions called modules. A module diverts water from a secondary canal and its users are scheduled to receive water at the same time. CHAPTER IV METHODOLOGY This research is based on surveys of poor farmers, i.e. ejidatarios, at the Panuco River Irrigation Project (PRIP) in 1987 and 1995. A survey method based on a structured questionnaire was chosen for this research because it is assumed that ejidatarios are able to measure differences in their living conditions and in the irrigation project conditions from 1987 to 1995. A. Research Design The present research was designed as a longitudinal study. This type of research looks at the changes of a particular population over time. Longitudinal studies have some advantages compared to cross-sectional designs because they are explanatory instead of descriptive. This research specifically is a panel study. In 1987, an evaluation research was performed to study a sample of 238 ejidatarios on the Panuco River Irrigation Project (Diaz, et al., 1988; World Bank, 1989; Diaz and Valtierra, 1992). A similar study was performed with a sample of 150 ejidatario-farmers in 1995. These samples had a 57 paired 53 54 subsample of ejidatarios who were interviewed in 1987 and 1995. In a panel study, a collection of the same data is gathered from the same sample of respondents at least twice across time. B. Methods of Data Collection This study was based on a structured survey. A structured questionnaire was selected because: a) a structured questionnaire was used in the 1987 data collection, therefore, using a questionnaire in the 1995 survey would make the comparison between both sets of information more reliable; b) a questionnaire helped to overcome time and financial constraints of this type of research for a dissertation; c) questionnaires also offer some advantages when information is collected from large random samples because they standardize variables for computerized statistical processes. This questionnaire was used in both surveys during direct interviews with each farmer selected. The questionnaire was broken down into three sections: the first section dealt with social and welfare aspects; the second looked at several irrigation aspects of the irrigation system; and the third helped define the type of farming system according to ejidatarios’ annual net income and its breakdown (agricultural, livestock, off farm, and various incomes). The questionnaire included three types of 55 questions: open—ended questions, closed-ended questions, and tabular questions. In the open-ended questions, a list of possible answers was provided, and only one had to be selected. There were a few open-ended questions that generally asked "why" and followed a closed-ended question. Some tables were designed to gather data on ejidatarios’ productive activities (e.g., crop production and livestock production) (see Appendix A). The questionnaire was designed to be interactive between the interviewer and the respondents. The function of interviewers was to help respondents in a joint effort to remember the data of interest. The most difficult part was gathering data on agricultural costs (labor, inputs, services, etc). One of the main challenges was to match the 1995 questionnaire with the 1987 questionnaire, given the specific objectives of each survey. Conceptual, wording and formatting problems needed to be overcome. In some instances, it was necessary to aggregate and disaggregate data to make the surveys comparable. C. Operationalization of the Main variables The following concepts are main variables selected for this study. The definitions of concepts explain the sense in which the variables were considered for this study. Many other variables are influencing the complex situation that 56 is studied in this research. These variable are not ignore but they could not be studied at this time, such as the influence of the North American Free Trade Agreement (NAFTA) and the federal policy in this regional development project (e.g., credit, insurance, agricultural product pricing, etc.). Family Household. In this study the ejidatario family was examined as a socioeconomic unit that organizes its life around agricultural cycles, farming and growing domestic animals. The family makes its living from agriculture. It was not part of this research to delineate a detailed description of the characteristics of the ejidatario family. However, it was necessary to be able to profile the farming system in the selected periods (See appendices B and C). For the purpose of this research, a practical definition of household was needed to collect the field information, particularly information regarding the household income. The prototype of a "Western" nuclear family integrated by a married couple and their children does not fit the reality of most of the Mexican farmers. Most of the rural families in Mexico are extended families. A typical rural family includes a married couple (i.e., husband and his wife) with children, husband’s relatives, wife’s relatives, and other members without parental relationships. In this sense, all persons who contributed to or depended on a common family fund, or gasto familiar, were included as members of the household. The contribution 57 to the gasto included working on the family parcel, taking care of the family livestock, or giving money earned from a salaried activity off the farm. In some cases, it was a little difficult to include or exclude a person from the family. For example, children who were studying in other cities far away from home; children who were working in the U.S.A or Mexico City; married children who lived in the same house, but contributed nothing to or derived nothing from the family fund. It is understood that the structure of the family household has changed over the time period studied, however the information about the household was collected using the same parameters to make the comparison between 1987 and 1995 possible. Household Inggme. Household income is constituted by all the income from family members, monetary or in kind. Calculation of the ejidatarios’ income was a challenge because poor farmers do not keep any record of their cash flows from within or out of the farm. However, the 1987 survey11 demonstrated that ejidatarios could normally remember most of what they did in the previous year, how much money and time they spent, and how much they earned on their on-farm economic activities. It is just a matter of asking the right questions in the right way. To help ejidatarios remember, this study chose to break down annual “ Similar surveys were successfully performed to evaluate the Puebla Project and the PIDER in Ometepec, Guerrero. (Diaz, 1992; Diaz and Valtierra, 1985). 58 income by sources of income (crops, livestock, off-farm, and other incomes). Detailed breakdowns of crop and livestock production were also prepared. Crop production processes were divided into different stages: land preparation, sowing, fertilization, pesticide spraying, and harvest. Data collection included purchased and owned inputs (fertilizers, seeds, etc.), purchased and owned services (e.g., machinery), family and salaried labor, and opportunity cost of land (see Appendix A). In terms of crop production, this study encompassed data from the last two cropping seasons, 1994 spring-summer and the 1994-1995 fall-winter. In the case of other economic activities (e.g, salaried jobs), data included all costs and gross incomes from the previous twelve months. Opportunity Cost. The concept of opportunity cost played an important role in this survey, since many poor ejidatarios consume or produce several non-traded goods. Opportunity cost was calculated when ejidatarios used their own resources instead of purchasing them in the free market. For example, an opportunity cost was assigned to family labor. An opportunity cost was also assigned to products harvested and consumed within the household. For example, if a family consumes their chickens, the opportunity cost of these chickens was added to the value of livestock production. The opportunity cost assigned to each non-traded good was similar to the value that the product has on the local 59 market. The logic was based on how much money farmers would have to spend for an input if they did not have this input, or how much money they would have to pay a worker for applying fertilizers if farmers and their families did not. This approach differed from a classic economics approach.12 Fsrmsrs’ Perseption pf Wslfare Imprpvsmsnts. Farmers were asked if they have perceived any improvement or deterioration in the last eight years. They were asked to make an overall assessment taking into account several aspects, such as food consumption, house improvements, education, and others that they wanted. The questions were based on the assumption that farmers could make an overall judgment regarding their welfare indicators based on their own parameters and preferences for what is better or worse. There was no empirical checking to see if the farmers’ perception agrees with the researchers’ perception, which may imply a different research project beyond the scope of this study. Psrgsivsd Effigisngy pf Irrigspion ssrviss. Ejidatarios’ perceptions include several aspects of irrigation management, such as timeliness, adequacy”, and “ A classic approach on opportunity cost can be found in: Gittinger, Price. 1994. Egonpmig Analysis of Agrigpltural Projssts. The World Bank. Washington, D.C. ” Plusquellec et al. (1994: 10) define water timeliness as the frequency and duration of the water delivery and adequacy as the sufficient volume of water to irrigate crops. 60 water pricing. In the 1987 and 1995 surveys, the farmer’s opinions and assessments were the only parameters collected to know the adequacy and timeliness of irrigation water“. During the field research no further empirical evidence was collected, like water volume measurement, evapotranspiration, or other indicators to check if farmers’ judgments were accurate. Additionally, there were several questions in which ejidatarios were asked for "complex" evaluations, for example, "Is irrigation water delivered on time?" It was assumed that the farmers’ answer involved overall assessments based on their experience and knowledge of their crops, environment and irrigation system. D. Sampling Methods Sampling is part of any social research; however, it is a central piece of a panel study, because a great deal of theoretical and operative aspects of a panel study depend on the sampling methods. This study was based on two surveys performed in 1987 and 1995 respectively. The first survey was contracted for the World Bank and performed by a team of researchers from the Graduate College (Colegio de Postgraduados (CP)). The “ This study has similar approach on adequacy and timeliness as Shivakoti’s study (1991). 61 World Bank set a sample size of 23815 ejidatarios for the 1987 survey. The World Bank decided that 50% of the ejidos had to be surveyed, in approximately 80 ejidos. In each ejido one member of the board of representatives and two randomly chosen ejidatario-farmers were interviewed. The CP took responsibility for the sample representativeness. First, the number of ejidos was divided according to the number of beneficiaries in each irrigation unit. Second, ejidos were randomly selected from the PRIP area. The third step was to randomly select three ejidatarios in each ejido. The 1987 sample of ejidatarios was selected taking into account three aspects of the ejidatario population: a) geographic distribution, b) ejidos were classified into old ejidos and new ejidos“, and c) the sample was divided into four groups according to their level of development“. It is important to note that the 1987 sample consisted of 238 ejidatarios, plus twelve pequefios propietarios (private owners) and nine colonos (new private settlers). Both pequefios propietarios and colonos were not included in ” The original sample was 239 ejidatarios, but one questionnaire is missing from the file. “ Old ejidos were established before the creation of the PRIP in 1974, and new ejidos were formed by the PRIP after 1974. “ The level of development was determined empirically with the collaboration of technicians of the Secretariat of Agriculture and Water Resources. The levels of development were incipient, in process, almost developed, and fully developed. 62 this study because the 1987 sample size of both groups was too small to be statistically representative. The 1995 survey sample was of 150 ejidatarios. The sample was broken into three subsamples: 1) a random paired sample of 57 ejidatario-farmers interviewed in the 1987 survey; 2) a random sample of ejidatario-farmers who were not interviewed in the 1987 evaluation; and 3) a random sample of members of an irrigation user association that manages irrigation infrastructure. The sampling procedure was based on the 1987 annual income as the most representative variable of the variation among individuals of ejidatario-farmer population. The following formula was used to determine the sample size. Nd? + 22m s2 Where: n = size sample N = population of ejidatarios at the beginning of PRIP me = Value of a normal random variable from normal table at 100 (1- a)%. s2 = income variance of the population of ejidatarios interviewed in 1987. d = precision or maximum expected difference between the actual value and the estimator (6 - 3). Taking into account the variability of income, the precision was calculated at 21% from.the mean income, d,21 = 15,540 x 0.21= 3,263. It means that the true mean value will be in the interval 15,540 +/- 3,263. The confidence level of 90% (1 - a = 0.90) was used to 63 calculate the sample size. Substituting the values from the 1987 survey in the formula. 8,500 (1.64)2 (24,816)? 1’]: =153 8,500 (3,263)’ + (1.64)’ (24,816)? N = 238 z“,2 = 1.64 s = 24,816 R = 15,540 c1,21 = 3,263 The final size sample for the 1995 survey was 150 ejidatarios, which is 98% of the 153 ejidatarios indicated using the previous formula. B. Data Collection Process The 1987 survey contracted for the World Bank can be considered the first stage of the present study. The preparation for the 1995 survey began by reviewing results from the 1987 survey (Diaz, et al. 1988) and two evaluation reports (World Bank, 1989; Diaz and Valtierra, 1992). During the summer of 1994, an exploratory visit to the PRIP region helped to perceive the current situation of ejidatarios. During this field work, several activities were undertaken: interviews with officials of government institution and farmer leaders, field transects, and the collection of collateral information (maps, statistical information, previous studies, etc). This information 64 helped to design the survey questionnaire for ejidatarios. Once the questionnaire was approved by Michigan State University”, the field work of this research was executed in August of 1995. Four interviewers besides the researcher participated in the conduction of direct interviews with ejidatarios. Even though interviewers were agronomists and had previous experience in survey interviews, they were trained in managing the questionnaires, coding the answers, conducting an interview in the Panuco region, and how to handle critical situations before or during the interviews. The mechanics of the field work consisted of the following: a sample of ejidos was selected; once interviewers arrived at a village, they sought out the ejidatarios interviewed in 1987. If the selected ejidatarios were not in the village or unwilling to participate, they were replaced by an ejidatario who lived close to them. The interviews were completed primarily at the respondents’ homes. There were a few cases in which interviews were conducted at a different setting. An average interview lasted one hour and thirty minutes. Only two of the ejidatarios selected refused to participate. The questionnaire was designed to have most of the closed-ended questions coded and ready to be entered into a computer database. The most difficult sections of the questionnaire were cropping and livestock costs. These “ The University Committee on Research Involving Human Subjects (UCRIHS) at Michigan State University. 65 sections generally needed some additional mathematical calculation with the ejidatarios’ raw data before the questionnaire was ready to be coded. Once interviewers coded the questionnaires, the researcher checked them to see if they were properly filled out and free of inconsistent information. If any major disagreement over the data collected arose, interviewers were sent back and to check the information in question with the farmer. F. Software used for Statistical Data Processing The software used for statistical data processing was DBASE III, SAS (Statistical Analysis System) version 6.04 for DOS, and SPSS (Statistical Packet for Social Science) version 7.0 Win 95. G. Advantages and Limitations of the Present Study A longitudinal study has numerous advantages compared to a cross-sectional study. The achievements of an agricultural development project can be affected by many factors, especially the weather. For instance, weather may produce dramatic variations from one agricultural season to another, even in the cases where agriculture is irrigated. If the longitudinal study counts with two data sets, one set could be gathered in a good agricultural season and the other in a bad agricultural season or in an average season. 66 This situation may affect the conclusions and lessons that can be obtained from the longitudinal study with only two data sets. The eight years between both surveys may be considered as sort time in terms of the project irrigation life time. However, in this case many things happened at the regional and national level during those eight years. One limitation of this study was that there was not a baseline study conducted in 1974 by the World Bank at the time the project was first implemented. While there was some information collected (i.e., soils, water, previous agricultural production functions, etc.) none was collected regarding the participants themselves. Because of this, the first evaluation study in 1987 is used as the base line study to measure the impact of the project changes between 1987 and 1995. Another limitation for this study to become a comprehensive evaluation was that it did not include an environmental impact assessment. There were two reasons for this situation: a) PRIP did not set any environmental objectives and b) this study had important limitations in time and money. CHAPTER V FINDINGS The null hypothesis for this study states that "the Panuco River Irrigation Project (PRIP) did not meet its initial objectives". The testing of this hypothesis is divided into four subhypothesis, one for each of the four PRIP objectives. The first two sections of this chapter deal with the general hypotheses: (1) PRIP did not develop an efficient irrigation delivery system (Ho:‘n=.5) and (2) ejidatarios did not produce more food crops for the national food security in 1995 than in 1987 (Ho=lh=#1)- This analysis is followed by the hypothesis testing of two subhypotheses related to the ejidatario participants: ejidatarios did not increase their income from 1987 to 1995 (H°:;h>u2), and ejidatarios off-farm work increased from 1987 to 1995, (H0: u,30). The parameters for this test are: n=.545, n,=.5, and n=57. The following formula was used for the test: “ The value of .5 was selected because the actual proportion is unknown and the ejidatarios’ answers were grouped as binary. ” A chi-square test was used for this hypothesis testing because most of the cells in the table have frequencies lower than 5 observations, thus the results may not be valid. 72 Where: fi' = the observed proportion n, = the percentage selected to detect a deviation from r1 = sample size Substituting the values into the formula: .545 - .5 .045 Z = 7 = ————-— = 0.681 .066 .5(1-.5) 57 This calculated Z-value is smaller than ZJB=1.64 from the Z-distribution table. As a result, the null hypothesis is accepted23 and thus the alternative hypothesis is rejected. The percentage of ejidatarios who think that PRIP is an efficient irrigation system is not high enough to say that most of the ejidatarios have received an efficient water delivery service. The binomial test was selected to test the null hypothesis for each of the 9 strata because there are less than 5 ejidatarios for each cell in the table. The conditions to apply the binomial distribution state that "if the probability of being classified in a given category equals n for each observation, then the probability that x out of n independent observations are classified in that ” The statistics literature states that a null hypothesis is rejected or is not rejected (Agresti and Finlay, 1986: p.146; Bhattacharyya and Johnson, 1977: p. 175). For the purpose of this dissertation the null subhypotheses are considered rejected or accepted. 73 category, denoted by P(X), equals" (Agresty and Finlay, 1986: 143) n! P(X) = 11'x (1-1T)""x X = 0,1,2 ..... n X!(n-X)! The test consists of adding the probabilities of all the numbers equal or greater than the X number found in each cell. For example, the X number is equal to 7 in the second stratum, the test requires adding P(X) when X equals 7 and 8. The precalculated values from the binomial distribution tables (Hamburg and Young, 1994: A6-A10) when n=8 and n=.5 are : P = P(7) + P(8)= .0312 + .0039 = .0351 The calculated probability is lower than .5 which means that the probability that the null hypothesis may be true is low, thus the null hypothesis is rejected for the second stratum. Likewise, the probabilities were calculated for the rest of the strata. The results are displayed in the eight column of Table 1. These results show that the P- value is lower than .5 for the first and second stratum. Most of the ejidatarios in these two strata felt that they had an efficient water delivery service in 1995. On the other hand, the calculated P-value is higher than .5 for the other seven strata (from the third to the ninth strata), indicating that there is a high probability that this null 74 subhypothesis could be true. The reaction of the ejidatarios to the lack of efficiency was that they felt overcharged for lack of service. These farmers are not willing to pay higher water fees under the current condition of the water delivery service. Some of these ejidatarios expressed that they would be happy to pay the annual water fee if the PRIP water delivery service became more efficient. In PRIP, 49% of the ejidatarios felt that they were overcharged for water because of the poor-quality of water delivery service. Most of these farmers were opposed to any water price increase. A similar proportion of ejidatarios (47%) felt that they were fairly charged and were willing-to-pay a higher price. This is especially important at this moment when the government is transferring the infrastructure management to water user associations that do not receive any governmental subsidy to operate irrigation infrastructure. The "Jaboncillo" module that was transferred to a water user association in 1994 had the highest water fee in the PRIP area in 1995. An increase in the water fees may constitute a threat to the efficiency of the water delivery service provided by the water user association because not all of the users are willing to pay higher water fees. Adequacy of Water Supply Adequacy of water supply is one aspect used to measure 75 the efficiency of water delivery service. Adequacy was defined as the volume of water needed to irrigate crops. Adequacy is sometimes a critical factor, especially when the seasonal weather is dry. Ejidatarios pay an annual fee for a prescribed amounts of water, but they can buy more water if they need more water. The water fee and the amount of water received per hectare varied from one irrigation unit to another, depending on the operational cost of each irrigation unit. In 1995, the annual fee in the Pujal Coy Unit was $70 pesos per 2,000 HP of water, $70 pesos per 5,000 a? in the Animas Unit, and $125 pesos per 3,000 a? in the Chicayan Unit. Additional water could be purchased at $30 pesos per 1,000 n9, however ejidatarios rarely bought additional amounts of water. The same normal approximation to the binomial test was used to analyze the proportion of ejidatarios who considered the amount of water received adequate for cropping (Table 2). The parameters for the test are 2:.685, n=54, and n,=.5. Substituting these values into the formula of the normal distribution: .685 - .5 .185 Z = = -————— = 2.72 .068 .5(1-.5) 54 76 The calculated Z-value is larger than ZJB=1.64, indicating that ejidatarios received an adequate amount of water for their crops. Most of the ejidatarios had a favorable opinion about this aspect of the water delivery service efficiency. Adequacy appears to not be a crucial factor since ejidatarios can purchase additional quantities of water as they need it. Some of the ejidatarios felt that they did not receive the volume of water for which they have paid for. They felt that the volume of water was not accurately measured by PRIP operators. Table 2. Adequacy of the Amount of Irrigation Water in 1995 ADEQUATE(1) NON-ADEQUATE ROW TOTAL Binomial STATUS freq row % freq row % freq row % P-Yiiue 1 2 100.0 2 100.0 .2500 2 8 100.0 8 100.0 .0039 3 9 75.0 3 25.0 12 100.0 .0729 4 2 25.0 6 75.0 8 100.0 .9649 5 6 85.7 1 14.3 7 100.0 .0625 6 3 60.0 2 40.0 5 100.0 .4999 7 3 60.0 2 40.0 5 100.0 .4999 8 2 50.0 2 50.0 4 100.0 .6875 9 2 66.7 1 33.3 3 100.0 .5008 COLUMN 37 68.5 17 31.5 54 100.0 TOTAL A binomial test was performed to evaluate the answers of ejidatarios in each stratum. The eighth column of Table 77 2 shows the results of the binomial test regarding the adequacy of the water in the PRIP. There is a clear tendency in the lower strata (first, second, third and fifth strata) to show that the percentage of ejidatarios who considered the water supply adequate was significantly higher than 50% of the ejidatarios. In the other strata (fourth, sixth, seventh, eighth, and ninth), the percentage of ejidatarios was lower than 50%. Timeliness of Water Delivery After more than ten years of agriculture in the Panuco Region, ejidatarios have learned when and how often their crops need water. A PRIP rule states that water users need to request water three days in advance to receive the water on time. The ejidatarios’ opinion about the timeliness of the irrigation service was divided into two categories, timely and non-timely. A normal approximation to the binomial test was performed on the 57 paired sample and the strata values. The parameters for the Z-test are n=.593, n,=.5, and n=54. Substituting this values in the formula: .593 - .5 .093 Z = = ————-— = 1.37 .068 .5(1-.5) 54 78 This calculated Z-value is smaller than ZJE=1.64. Therefore, there is no evidence that .593 would be statistically higher than .5. According to this result the null hypothesis is accepted. The PRIP did not offer a timely water service delivery to most of the ejidatarios. The binomial test results are displayed in the eighth column of Table 3. There is a clear tendency that lower income ejidatarios received water on time in 1995. More than 50% of the ejidatarios in the first, second, third, and fifth strata received water service delivery on time. In the opposite situation were those in higher income strata. A significantly low number of ejidatarios in the fourth and sixth through ninth strata felt that they received irrigation water to their lands in a timely manner. Table 3. Timeliness of the Irrigation Service Delivery in 1995 TIMELY(1) NON-TIMELY ROW TOTAL Binomial P-value STATUS freq row % freq row % freq row % (1) l 2 100.0 2 100.0 .2500 H 2 6 75.0 2 25.0 8 100.0 .1445 3 9 75.0 3 25.0 12 100.0 .0729 4 3 37.5 5 62.5 8 100.0 .8555 5 5 71.4 2 28.6 7 100.0 .2266 6 3 60.0 2 40.0 5 100.0 .4999 7 1 20.0 4 80.0 5 100.0 .9686 8 2 50.0 2 50.0 4 100.0 .6875 9 1 33.3 2 66.7 3 100.0 .8750 COLUMN 32 59.3 22 40.7 54 100.0 TOTAL 79 In 1995, during the survey, a remarkable difference was observed between the opinions of those users living at the head of the canal and those living at the end regarding timeliness of water delivery“. Head-end-canal users almost always felt that they received water on time, while the tail-end-canal users felt that they almost never received their water on time. In some of the tail-end-canal sections, water delivery was so unreliable that many ejidatarios did not want the PRIP irrigation service. Perceived Deterioration of the Irrigation Infrastructure PRIP has not been as efficient as it was planned to be. One of the main reasons has been the increasing deterioration of the irrigation infrastructure. The efficiency of the irrigation water delivery greatly depends on the conditions of the infrastructure. If the infrastructure is deteriorated, the system is unable to deliver the requested quantity of water on time. This situation has a significant impact on crop yields at the PRIP. In 1995, irrigation infrastructure suffered from varying degrees of deterioration. According to the perception of 60% of the ejidatarios, the primary “ This was an empirical observation that was not documented with information from the surveys. 80 infrastructure25 was seriously damaged. A higher percentage of ejidatarios (71%) assessed that secondary infrastructure26 was seriously damaged. In the case of tertiary infrastructure“, a lower percentage of ejidatarios (42%) felt that this type of infrastructure was seriously deteriorated. Tertiary infrastructure is not as badly deteriorated as the other types of infrastructure because it has benefitted from work by ejidatarios, who maintain some parts of these structures. In interviews with government officials, most of them felt that the entire irrigation infrastructure needed to be reconstructed. A former federal official of the CNA felt that the infrastructure of the Panuco River Irrigation Project was so deteriorated that it was going to be difficult to transfer the irrigation management to the user associations. One local official of the CNA confirmed that practically all of the PRIP irrigation infrastructure needed repair. The question is: Who will pay for the rehabilitation of the infrastructure, since federal budgets are too small and the ejidatarios do not have money?. ” Primary infrastructure includes dams, main canals, house of pumps, ditches. “ Secondary infrastructure include all the diversion canals, gates, weirs, and offtakes. ” Tertiary infrastructure is the irrigation infrastructure in the ejidatarios’ fields. 81 Ejidatarios’ Participation in Maintenance The PRIP infrastructure is suffering from many years of neglected maintenance by the government and the lack of beneficiaries’ participation in infrastructure maintenance. The main cause of this situation was the paternalistic development approach of PRIP that dominated the relationships between government and PRIP beneficiaries. The lack of users’ participation was not new in 1987 or in 1995. The government constructed the PRIP infrastructure without beneficiaries participation. In the 1995 survey, most ejidatarios (95%) expressed that they did not participate in the construction of irrigation infrastructure, while 5% participated as salaried workers in the construction of canals (e.g., as bricklayers). In the past few years, there has been an increase in the beneficiaries participation, however, it is still far from what is needed. In 1995, only 43% of the ejidatarios contributed with an average of 29 hours of work per year to maintain tertiary infrastructure, while more than half of the ejidatarios (53%) contributed nothing to irrigation maintenance and operation. Ejidatarios assumed that water fees covered the total cost of operation and maintenance. However, according to local officials of the National Water Commission (CNA), water fees were too low and only covered 30% of the PRIP 82 operational cost in 1994“. By design, the government subsidized most of the PRIP operation. As a result, the infrastructure continued to deteriorate because there were never enough resources for maintenance. In 1995, a series of group interviews were conducted as part of this study. Ejidatarios expressed some concerns with the idea of transferring the infrastructure management from the federal government to the users. Ejidatarios were not willing to take the responsibility for managing and operating the irrigation project. They felt that the infrastructure was too deteriorated and they did not have the money to repair it. Ejidatarios also expressed a remarkably strong feeling of lack of ownership of the irrigation system because the government has always managed the irrigation project without the farmers’ participation. In 1995, most ejidatarios (69%) felt that government should be responsible for infrastructure maintenance and few of them (7%) felt that infrastructure maintenance should be the users’ responsibility. B. Contribution to National Food Security The Panuco River Irrigation Project was created to produce cash crops (e.g., vegetables) and high yield grains (e.g., sorghum). Once PRIP started to operate, the ” In 1987, according to the World Bank (1989:58) the water fee only represented 9% of the operational cost. 83 traditional system of grazing on natural pastures was rapidly substituted with annual crops and the traditional big ranchers were replaced with collective ejidos. The second null subhypothesis states that ejidatarios did not produce more food crops for the national food security in 1995 than in 1987 (Ho:;n=u1'vs. Hauhpflq). The main variables selected to test this subhypothesis were the total amount of food (kilos) per crop produced per ejidatario” and the value of this produce per crop per ejidatario (the fourth and fifth columns in Table 4). T- tests were used to compare the mean differences to determine if the ejidatarios produced more kilos of each crop in 1995 than they did in 1987 and if ejidatarios made more money per crop in the same period. Other two related variables were chosen to complement the statistical analysis: the percentage of ejidatarios cultivating a crop and the area per crop per ejidatario. A normal approximation to the binomial test was performed on the first variable and a t- test on the mean difference was performed on the second variable. The following formulas were used to perform the t-test and the normal approximation to the binomial distribution test: D 8 /‘V5 ” No differentiation was noted between the selfconsumption and the production for selling because most of the PRIP ejidatarios do not produce for selfconsumption. 84 mOmod unnumcoo mmm0 80 mum mmDHM> >umuocoe one «00 mowhmbmfiwho 0m >0 ooo0>0o was mmumuomn mo ucnosm 08000 one «0 oHQEMm ooufimm 0m mnu mo mowHMumpwho mo ommucmouom 8 m0 m0£9 0 00.000.0 000.00 00.0 0.00 00.000 000.0 00.0 0.0 mz0u0so . 600: .0 00060 85 Where: D = the sample mean of the difference income 8 = standard deviation of the difference income n = number of observations in the paired subsample fl - no 2 = no(1-no) , n Where: 8' = the observed proportion n, = the percentage selected to detect a deviation from r1 = sample size The crop production was disaggregated by crop to look at the variability of the different crops from 1987 to 1995. In the following analyses, detailed descriptions of the results are completed on corn and sugar cane because of their importance. The rest of the crops are grouped under a subheading. Corn Corn was the most popular crop among ejidatarios. Ejidatarios planted corn because corn is highly tolerant to harsh weather conditions, with few inputs and little work. In addition, corn was easy to sell to the state commercialization company, CONASUPO (The National Company of Popular Supplies) and to private brokers. Corn was the ejidatarios’ most important crop in 1987 and in 1995. However, the percentage of ejidatarios who cultivated corn experienced a dramatic reduction from 75% in 86 1987 to 49% in 1995. A normal approximation to the binomial distribution was performed on this variable. The calculated Z-value was -4.61. This is lower than the z-value from tables (-ZJB=1.16). Even though corn remained the ejidatarios’ main crop, the percentage of ejidatarios growing corn significantly decreased from 1987 to 1995. The area cropped per ejidatario decreased by 1.26 hectares from 4.56 to 3.30 hectares. However, the t-test performed on this variable shows that this difference was not statistically significant at .05 level. The calculated t-value was 1.55, which is lower than the t-value from tables at .05 level (1.64). In this case, the variability of the area cropped from one ejidatario to another made the mean difference not significant, yet it did reflect a 28% reduction in total area cropped. The amount of corn produced per ejidatario decreased by 121 kilos per ejidatarios, from 4,040 to 3,919 kg. The t- test calculated on the mean difference shows that the amount of corn produced contributed to the national food security, however, was not significantly different from 1987 to 1995. Ejidatarios' earnings for corn decreased, each corn grower made $909.12 pesos in 1987 and $403.67 pesos in 1995. The mean difference of $505.45 pesos was tested. The calculated t—value (-2.25) was lower than the negative t- value (—1.64) from the tables at .05 level with df=57-1. The statistical analysis showed that there was a significant decrease in the ejidatarios’ earnings from 1987 to 1995. 87 At the level of the 9 strata, the number of corn growers increased in the first and the ninth strata while decreasing in the other seven clusters. The most dramatic changes occurred in the second and ninth strata. In the second stratum, there was a significant reduction in the number of ejidatarios growing corn, from 8 to two, and in the average number of hectares of corn, from 9.96 hectare per ejidatario to 3.5 hectares. In the ninth strata, the number of ejidatarios increase from 1 to 3 and the hectares per ejidatario from 2.8 to 8.33. Corn lost importance because 67% of the ejidatarios received credit in 1987, however by 1995 none ejidatario had credit to cultivate corn. In addition, in 1987 most of the ejidatarios’ corn was traded by a state company (CONASUPO), by 1995 this company reduced its operations in the region to very low levels. Sugar Cane Sugar cane cultivation was completely different from all other crops. Sugar cane production experienced a "boom" in the Chicayan irrigation unit between 1987 and 1995. Many ejidatarios felt that sugar cane is the most profitable crop and, thus, their best alternative to crop their land. The percentage of ejidatarios growing sugar cane increased from 5.3% in 1987 to 12.3% in 1995. The normal approximation to the binomial test indicates that there was 88 a significant increase during this eight-year period. The average number of hectares of sugar cane per ejidatario increased from .09 to 1.17 hectares, in 1987 and in 1995 respectively. The results of the t-test demonstrate that the calculated t-value is equal to 2.64, which is higher than the tjn=1.64 from the t-distribution tables indicating that the difference was significant. The sugar cane production increased from 7,579 kg to 90,368 kg per ejidatario. The mean difference between these years was 82,789 kg per ejidatario. The calculated t-value was equal to 2.58, which is higher than the t-value at .05 level. The amount of sugar cane contributed to the national food security significantly increased during this period. The main reason for this change in sugar cane cultivation was that in 1987 the local sugar mill was owned by the Mexican Government. This sugar mill had a lot of problems with its operation and the promotion of sugar cane production. Between 1987 and 1995, the sugar mill was sold to private investors as part of the national policy to downsize the government. Since then, the sugar mill has operated more efficiently and has benefitted the ejidatarios who grow sugar cane. In 1995, growing sugar cane was the only alternative for ejidatarios to get credit and to make a profit within their farm. Unfortunately, the sugar mill only operates in the Chicayan irrigation unit and does not purchase sugar cane from other regions. 89 Other Crops Four other crops were important in understanding the ejidatarios’ cropping pattern in 1987: soybean, sorghum, beans, and carthamus. In 1995, beans and carthamus were no longer part of ejidatarios’ cropping pattern and soybeans almost disappeared from the list as well. Finally, the percentage of ejidatarios who cultivate sorghum increased from 8.8% to 14%, which was not a significant increase. The amount of sorghum produced per ejidatario increased by 293 kg between 1987 and 1995, again not statistically significant. The average earnings from sorghum was also not significant during this time frame. C. Reduction of the uneven Income Distribution between Rural and Urban Population The fourth objective of the PRIP was "to reduce the uneven income distribution between rural and urban populations." This objective was to be achieved by helping project participant ejidatarios increase their income. PRIP anticipated that the ejidatarios’ economic income would increase each year from the initiation of the project (1974- 1977) through the full development of the irrigation system, between 1990 and 1994 (World Bank, 1989). After this point, the ejidatarios’ economic benefits from agriculture were expected to be sustainable. 90 The null subhypothesis states that "ejidatarios did not increase their income from 1987 to 1995 (Ho:;n>psz." The rejection would be determined by comparing the ejidatarios’ annual net income and on-farm income in 1987 and in 1995. This comparison was performed at three levels. In the first, the two larger samples were compared (i.e., 238 ejidatarios in 1987 and 150 ejidatarios in 1995). In the second level, the comparative analysis was performed on a 57 ejidatario paired subsample. In the third level of analysis, the 57 ejidatario paired subsample was divided into nine clusters based on their 1987 annual income. Comparison of the Ejidatarios’ Income between 1987 and 1995: General Samples Most of the Mexican ejidatarios have small farms and are low-income farmers. They basically depend on agriculture and animal production for their livelihood. When ejidatarios do not make enough money within their own farms they look for off-farm jobs (e.g., urban salaried job) to earn extra money to cover their basic needs. On-farm activities provide income stability, but not necessarily the ‘majority of the money needed for family subsistence. ” For a question of time sequence, yq‘was chosen to designate the 1987 ejidatarios’ income and up for the 1995 Ejidatarios’ income. 91 Annpsl Ns; Inspme The annual net incomes for 1987 and 1995 were very similar as shown in Tables 5 and 6. The ejidatarios’ annual income in 1987 was $15,539.54 pesos. In 1995, ejidatarios made $15,356.02 pesos. The difference between both years was only $183.52 pesos. A t-test was performed to determine if the income difference was statistically significant. The null hypothesis tested was "ejidatarios did not increase their income from 1987 to 1995, Ho:u1>p2" while the alternative hypothesis was H,:plsp2, where p, was the ejidatarios’ 1987 annual income and.uq*was the ejidatarios’ 1995 annual income. The alternative hypothesis reflected the planners’ expectation that PRIP would help the ejidatarios sustain or increase their income from 1987 to 1995. A one-tailed t-test was performed for the hypothesis testing. The parameters for the t-test were the 1987 mean income (2,) of $15,539.56 pesos, its correspondent standard deviation (8,) of $24,816.03 pesos and the sample size (N3) of 150. The ejidatarios’ 1995 annual net income (2;) was $15,356.03 pesos, the standard deviation (8,) was equal to $21,547.96 pesos and the total sample (N,)‘was 150 observation units. 92 Table 5. Synopsis of the EjidatariO-Farmers’ Net Income in 1987 (General Sample, n=238) INCOME CONCEPT N MEAN PERCENTAGE INCOME (1995 pesos) AGRICULTURAL PRODUCTION GROSS AGRICULTURAL INCOME 225 10,035.24 TOTAL AGRICULTURAL COSTS 225 6,789.49 NET AGRICULTURAL INCOME 225 3,245.76 (238) (3,068.47)* ANIMAL PRODUCTION GROSS LIVESTOCK INCOME 183 12,886.86 TOTAL LIVESTOCK COSTS 183 5,688.72 NET LIVESTOCK INCOME 183 7,198.14 (238) (5,534.70)* T -FARM I ** 230 8,902.41 1238) (8.603.17)* 55.36 FF-FARM I ME NET OFF-FARM INCOME 178 6,650.76 1238) (40974109)* 32.01 VARI I ME VARIOUS NET INCOME 65 7,185.00 (238) (11962,2§)* 12.63 AL T I ME 218______l§liial§§____lQQlQQ * Numbers within parenthesis are the means for the total ejidatario sample (238) ** Net On-Farm Income is the sum of net agricultural income and net livestock incomes. 93 Table 6. Synopsis of the Ejidatario-Farmers’ Net Income in 1995 (General Sample, n=150) INCOME CONCEPT N* MEAN PERCENTAGE INCOME (1995 pesos) AGRICULTURAL PRODUCTION GROSS AGRICULTURAL INCOME 103 30,687.13 TOTAL AGRICULTURAL COSTS 103 25,750.80 NET AGRICULTURAL INCOME 103 4,936.33 (150) (3,389.62)* ANIMAL PRODUCTION TOTAL GROSS INCOME 130 7,017.25 TOTAL LIVESTOCK COSTS 130 4,916.18 NET LIVESTOCK INCOME 130 2,101.08 (150) (1,820.92)* NET QN-FARM INQOME** 144 5,427.65 (150) (5.210.54l* 33.93 FF-FARM I OME NET OFF-FARM INCOME 86 14,062.69 (150) (8,062.61) 52.51 VARIOUS INCOME VARIOUS NET INCOME 68 4,594.59 (150) (2,082.87) 13.56 AL NET INCOME 159 151356gflgi 100.00 * Numbers within parenthesis are the means for the total ejidatario sample (150) ** Net On-Farm Income is the sum of net agricultural income and net livestock incomes. 94 i, - 3?, t = 321 322 + n1 n2 Substituting the values in the formula: 15,356.03 - 15,539.56 t = = ‘0.077 24,8162 21,5482 —————+———— 238 150 The negative value of tJn=-1.64 with df=n-1=56 was lower than the calculated t-value=-0.077. The 1987 ejidatarios’ annual income was statistically similar to the 1995 annual income. The null hypothesis was rejected, instead the alternative hypothesis was accepted. Ejidatarios were able to sustain their annual income from 1987 to 1995. Based exclusively on this mean value, it appears that nothing changed in the ejidatarios' income. However, a mean income value is often a zero-sum trade-off as some increased their income and other decreased in similar proportion. This zero-sum gain often hides large economic inequalities within a population. Uneven income distributions can have the same mean income as fair income distributions. Therefore, it was necessary to conduct some further statistical tests to determine how specific parts of the participants’ annual income make up have changed over time. 95 Student t-tests were performed on the ejidatarios’ on-farm income (i.e., cropping and livestock) and off-farm income (salaried jobs and their own businesses) to identify particular variations within the annual net income mean. On-Fsrm Ns; Ingpme A major goal of the Panuco River Irrigation Project was to improve the income of landless people by giving them ten hectares of irrigated land and providing them with support services to run their new farms. The government assumed that the ejidatario' farm would be the primary source of income, while the off-farm activities would complement the ejidatario's income. The mean on-farm net income decreased $3,392.63 pesos from 1987 to 1995 (Tables 5 and 6). A student t-test showed that in the ejidatarios’ on-farm income was statistically significant. The point estimates of the 1987 on-farm mean income were i,=8,603.17, s,=20,344.91, n1=238 and the 1995 on-farm income estimates were ig=5,210.54, s,=13,150.48 and 15:150. Substituting the values in the t-test formula: 5,210.54 - 8,603.17 t = 7 = -1.78 24,3442 13,1502 —_+—— 238 150 96 When this t-value (-1.78) is compared to the t-value from the distribution table, -tJm=-1.64 for df=n-1=56, it falls out of the rejection region (RR2-1.64). Therefore, the null hypothesis (Ho:;h>ng) is accepted. The ejidatarios’ crop and livestock production were negatively impacted in the Panuco Region between 1987 and 1995. PRIP failed to help ejidatario-farmers increase their agricultural incomes. Comparison between the PRIP Ejidatarios’ Income and National Income Parameters The mean income of a sample is more meaningful when it is compared to national parameters. The sample parameters of this study are compared to the national average income and the poverty line for the Mexican rural population. The results show that the PRIP ejidatarios are poor, according to the national parameters. The most recent information was the "Ehcuesta Nacional de Ingresos y Gastos de los HOgares, 1992" (National Survey of Household Income and Expenses, 1992) (Tellez, 1994: 48). The results of this national survey reported that the 1992 annual net income per rural household was $12,604 pesos, which is equivalent to $21,049 pesos in 1995 constant pesos. The PRIP ejidatario annual household income was 97 $15,539.56 pesos in 1987 and $15,356.03 pesos in 1995“. In both years, the PRIP ejidatario households made less money than the national average income of rural households”. The ejidatarios’ annual net income was also compared to the poverty line of the Mexican rural population. The poverty line in 1995 pesos was calculated to be $3,292.44 pesos per person per year”. The household poverty line was calculated by multiplying the per-person poverty line by the number of family members. In 1987, an ejidatario household had 5.14 members, therefore, the poverty line was $16,923.14 in 1995 pesos. In 1995, the average size of an ejidatario family was 5.46 members, resulting in a poverty line of $17,976.72 pesos. In 1987, ejidatarios made $15,539.56 pesos, which was below the poverty line. A similar situation occurred in 1995, when ejidatarios made $15,356.03 pesos, which was also below the poverty line. However, not all the ejidatarios fell below the poverty line. In 1987, almost three quarters (74%) of the ejidatario households were below the poverty line. In 1995, “ The rural poverty line was calculated with data provided by Jarque (1992). The poverty line in 1992 was 164,291 pesos per person per month. Converting this poverty line into 1995 constant pesos, the poverty line was equivalent to $3,292.44 pesos per person per year in 1995. ” All incomes were calculated in 1995 constant pesos. ” This poverty line was calculated with data reported by Jarque, 1994: 472. 98 the situation of the ejidatario families was relatively similar, since 69% of them were under the poverty line. At the national level, only 16% of the population was under the extreme poverty line (Tellez, 1994: 46). There was no available information regarding the participant’ income before the creation of PRIP, they might have been even poorer. The only reference made to the ejidatarios’ prior situation was a question on the 1987 survey. Most ejidatarios (58%) felt that they were better off in 1987 than they were before the PRIP and a small portion of them (15%) believed that they were worse off. Comparison of the Ejidatarios’ Income between 1987 and 1995: Paired Subsample This research was designed as a longitudinal study in which the same population was studied in surveys administered in 1987 and 1995. Both surveys included a 57 ejidatario paired subsample of individuals who were interviewed in 1987 and 1995. This paired subsample allowed for a comparison of the incomes of the same individuals in two different years. Annusl Nst Incpme The mean income difference of the paired subsample was calculated by subtracting the 1987 annual net income from 99 the 1995 annual net income (Tables 7 and 8). When the income difference was a negative number, there was an income reduction between 1987 and 1995 and a positive number meant an income increase for the same period. The null subhypothesis stated that "ejidatarios did not increase their income between 1987 and 1995". According to the paired difference t-test the income mean difference had to be significantly smaller than zero (Ho:;n-p1<0) (Agresty and Finlay, 1986: 176). t= 8 /‘V5 Where: D = the sample mean of the difference income 8 = standard deviation of the difference income n = number of observations in the paired subsample The parameters of the paired subsample used to perform the t-test were the annual mean income difference of $3,665.77 pesos, and a standard deviation of the income difference of $37,905.27 pesos. Substituting the values in the formula: 3,665.77 t = = 0.73 37,905.27 / V57 The calculated t-value was larger than the -tJn=-1.64 from the t-table, falling within the rejection region. In 100 Table 7. Synopsis of the Ejidatario-Farmers’ Net Income in 1987 (Paired Subsample, n=57) INCOME CONCEPT N MEAN PERCENTAGE INCOME (1995 pesos) AGRICULTURAL PRODUCTION GROSS AGRICULTURAL INCOME 53 9,478.11 TOTAL AGRICULTURAL COSTS 53 7,080.94 NET AGRICULTURAL INCOME 53 2,397.17 (57) (2,228.94)* ANIMAL PRODUCTION GROSS LIVESTOCK INCOME 44 16,757.95 TOTAL LIVESTOCK COSTS 44 8,170.45 NET LIVESTOCK INCOME 44 8,587.50 (57) (6,628.96)* NET QN-FARM INCOME** 55 9,180.00 (57L 18.857190)* 61.2_ OFF-FARM I ME NET OFF-FARM INCOME 45 5,482.00 (57) (4.327.89)* 29.94 VARI I ME VARIOUS NET INCOME 16 4,528.44 (57) (1.271.14)* 8.79 ANNQAL NET INQQME 57 14,456.93 100.09 I * The numbers within the parenthesis are the means for the total subsample (n=57). ** Net On-Farm Income is the sum of crop and livestock incomes. 101 Table 8. Synopsis of the Ejidatario-Farmers’ Net Income in 1995 (Paired Subsample, n=57) r TINCOME CONCEPT N* MEAN PERCENTAGE INCOME (1995 pesos) AGRICULTURAL PRODUCTION GROSS AGRICULTURAL INCOME 35 23,710.60 TOTAL AGRICULTURAL COSTS 35 22,208.71 NET AGRICULTURAL INCOME 35 1,501.89 (57) (922.21)* ANIMAL PRODUCTION TOTAL GROSS INCOME 48 10,695.10 TOTAL LIVESTOCK COSTS 48 6,572.08 NET LIVESTOCK INCOME 48 4,123.02 (57) (3,472.02)* NET QN-FARM INQOME** 53 4,725.87 (57) (4.394.23)* 24.2_ FF-FARM I ME NET OFF-FARM INCOME 37 17,123.05 (57) (11.114.96)* 61.33 N VARIOQS INCOME VARIOUS NET INCOME 31 4,805.48 (57) (2 1. . 1 * .4 42 ANNQAL NET INQOME 57 18,122.70 IQQyQQ * The numbers within the parenthesis are the means for the total subsample (n=57). ** Net On-Farm Income is the sum of crop and livestock incomes. 102 this case, the null hypothesis was rejected as the ejidatarios’ 1995 income was not significantly lower than their 1987 income. However, this result was partially influenced by the great income variability among ejidatarios as the standard deviation was ten times larger than the mean. The 1995 annual net income was not significantly different from the 1987 annual net income. In terms of sustainability, the result of the t-test illustrate that ejidatarios sustained their annual net income level from 1987 to 1995. This confirms the results of the t-test performed on the general samples. On-Fsrm Nsp Inpgms A similar procedure was performed on the ejidatarios’ on-farm income in order to test whether there was a significant difference between the 1987 and 1995 income levels or not (Hozfib-pq<0 vs. H,,,:;12-;1.1 20). The point estimates of the ejidatarios’ on-farm income difference were: a mean of -$4,463.67 pesos, the standard deviation of the income difference was $19,722.24 pesos and a sample size was 57 ejidatarios. -4,463.67 t = = -1.71 19,722.24 / V57 103 The t-value for the mean difference (-1.71) was significantly smaller than the t-value —t,m= -1.64. This value falls out of the rejection region (rr 2 -1.64) at 0.5 level with df=n-1=56, thus there is evidence that the null hypothesis may be true, thus the alternative hypothesis is rejected. Ejidatarios made significantly less money on their farms in 1995 than they did in 1987. This test confirms that PRIP failed to help ejidatarios increase their farm income. Comparison of the Ejidatarios’ Income between 1987 and 1995: Stratified Paired Subsample Most development projects are designed to benefit specific groups called target populations. This is especially true in all development projects that have a regional influence. However, frequently it is found that a few people receive most of the project benefits. The following statistical analysis shows who improved and who decreased their income from 1987 to 1995. The 57 ejidatarios paired subsample was stratified to determine whether the whole population of ejidatarios received benefits or only a small group took advantage of the PRIP support. T-tests were performed on the annual and the on-farm income differences for each cluster. The results of the t-tests are shown in Tables 9 and 10. 104 Annpsl Ns; Ingoms The annual household income of the stratified paired sample is shown in Table 9. A t-test was carried out to determine if ejidatarios of different income strata improved their annual income from 1987 to 1995. The alternative hypothesis for all of the strata is that "ejidatarios increased their annual income from 1987 to 1995 (H‘:p.,-,u1 20." The t-value was calculated with the following formula: \\ Cl 16 The seventh column in Table 9 shows the calculated t- values, that needed to be compared to the t-values from the seventh column. The comparison between these values illustrates that most ejidatarios sustained or improved their income levels between 1987 and 1995. On the other hand, ejidatarios in strata 6, 8 and 9 made significantly less money in 1995 than they did in 1987. Theories of social polarization argue that the poor get poorer and the rich get richer. The numbers in Table 9 contradict this theoretical statement. The poorer ejidatarios in 1987 increased their annual income in 1995, while ejidatarios in the higher income levels reduced their annual income during the same period of time. However, it does not mean that poorer ejidatarios in 1987 became rich in 105 .mommm unmumsoo mama Ga mam monam> mumuosos Ono aad 0maa on» Eouu mama osb msauomaunnm commasoamo mos 8808 080080 mosoawumao one 0 .0008umpwhm nomm How mEoosa 000.0- 00.0 00.000.00 00.000.0 00.000.m0 00.000.00 00 00000 000.0- 00.0- 0.000.00 00.000.00- 00.000.00 00.000.000 0 0 000.0- 00.0- 00.000.00 00.000.00- 00.000.00 00.000000 0 0 000.0- 00.0- 00.000.00 00.000.0- 00.000.80 00.000.00 0 0 000.0- 00.0- 00.000.0 00.000.0- 00.00a.0 00.000.00 0 0 = 000.0- 00.0- 00.000.00 00 000.0- 00.000.0 00.000.00 0 0 000.0 00.0 00.000.00 00.000.0 00.000.00 00.000.0 0 0 000.0 00.0 00.000.00 00.000.00 00.000.00 00.000.0 00 0 000.0 00.0 00.000.00 00.000.00 00.000.00 00.000.0- 00 0 000.0 00.0 00.000.00 00.000.00 00.000.0 00.000.0- 0 0 o o mo EOmemM00o 0000 0000 0-suwo msam>-O .2000¢0>mo 000 £003 .0 omeaaooaso oddozmem 00 200: mzoo20 200: aeozza 2 2800000 300050050 van on 0030008300 0000 van 0000 6063000 uuunouuuuan «sauna anus 008:64 .00000006008 .0 00000 106 1995, as the next analysis shows. Ejidatarios improved their income, however most of them still remained below the poverty line. The annual income of each strata was compared to the poverty line --the number of family members multiplied by the poverty line per person per month ($3,292.44 pesos in 1995). In 1987, the average annual incomes of the first six strata were under the poverty line while the income of the last three strata were above the poverty line. In 1995, only the second and the ninth strata were over the poverty line, while the rest of the strata were not. The third stratum had a particularly interesting situation in 1995. Ejidatarios in this cluster were below the poverty line in spite of their high income ($22,459.75 pesos). They also had the largest number of family members (7.17). On-Farm Net Income The PRIP focused its activities on improving ejidatarios income within their farms. Ejidatarios’ on-farm income had a positive change in the first three strata, while the ejidatarios in the last six strata reduced their on-farm income from 1987 to 1995 (Table 10). A student t- test was performed to determine if ejidatarios improved their on-farm income from 1987 to 1995. 7 0 1 .00000 00000000 0000 G0 000 0050m> 00000008 000 000 .00000mpwhm £000 000 080000 0000 030 Sony 0000 000 mn0uomuun00 0000050000 003 0005 080000 0000000000 000 0 000.0- 00.0- 00.000.00 00.000.0- 00.000.0 00.000.0 00 00000 000.0- 00.00- 0o.000.0 0.000.00- 00.000.00 00.000.00 0 0 000.0- 00.0- 00.000.00 00.000.00- 00.000.00 00.000.00 0 0 000.0- 00.0- 00.000.00 00.000.0- 00,000.00 00.000.00 0 0 000.0- 00.0- o0.000.0 00.000.00- o0.000.0 00.000.00 0 0 000.0- 00.0- 00.000.0 00.000.0- oo.000.0 00.000.0 0 0 000.0- 00.0- 00.000.0 00.000.0- 00.00 00.000.0 0 0 000.0 00.0 00.000.00 oo.000.0 00.000.0 00.00- 00 0 000.0 00.0 0o.000.0 o0.000.0 oo.o00.0 00.000.0- 00 0 000.0 00.0 0o.000.0 00.000.00 o0.000.0 00.000.0- 0 0 0 00 «m 0000 0000 0-:umu 0200000-0 .200000>00 0020000000 0003 00 0000000000 00002000 000 00 2002 .2002 020020 2000-20 2 000000 0009360000 000000 0000000000. 0000 van 0000 0003000 nounoummu00 oauunn 0000.3000-00 ~000Muuuuwnu .00 00000 108 The comparison of the calculated t-value with the t- values from the tables with n-l degrees of freedom shown in Table 10 indicate that the ejidatarios in the fifth, sixth and ninth strata suffered a significant reduction in their on-farm income between 1987 and 1995. Ejidatarios in the first through fourth clusters, and the seventh and eighth clusters either maintained or increased their income during the 1987-1995 period. D. Reduction of unemployment among Landless The null subhypothesis states that "the ejidatarios’ off-farm work increased from 1987 to 1995, Ho:;00p2." The PRIP beneficiaries’ background helped to determine if the PRIP effectively selected landless and unemployed people. The ejidatarios’ off-farm employment and the land renting were used as indicators to test the null hypothesis. Three aspects of the off-farm employment were analyzed: the percentage of ejidatarios with an off-farm job (temporary or permanent), the number of days worked off-farm in a salaried job and the off-farm income difference between 1987 and 1995. The analysis of land renting in this study includes the percentage of ejidatarios who rented their land to others and the amount of money received for land renting. 109 Background Characteristics of the PRIP Participants The recruitment of participants started with some groups of landless people in the Pujal-Coy area. These groups were demanding that the federal government provide them with land. Beneficiaries had very heterogenous backgrounds, skillful and unskillful people, farmers and laborers, people from irrigated agricultural areas and from rainfed agricultural areas. At the beginning, the desertion rate among ejidatarios was very high. Many participants were not interested in farming and only looked at PRIP as an opportunity to obtain free land. Once they received the land they sold it, even though it was illegal. Others simply abandoned their land, while others rented their land out because they had businesses in the nearby cities. The previous job experience had an impact on the success of many of the PRIP beneficiaries. The beneficiaries had very different job experiences. Some of them were working in jobs related to agriculture and others had urban jobs. Landless people were the main beneficiaries of the land reform, 65% of the participants were agricultural laborers, called peones agrfcolas. Most of them were working on the big ranches in the Panuco region. Some others were migrant 110 agricultural laborers“ who were temporarily in the Panuco region when they received the land from PRIP. Most of these people did not have experience in running a farm by themselves. Some participants of the PRIP (21%) were previously farmers and owned land in other places, but they came to Panuco to obtain more land. These people came from different places, but they at least had some farming experience. The third group of participants (14%) were the most heterogeneous. This groups was composed of urban workers (e.g., bricklayers, artisans, bureaucrats, etc.). The majority had no experience as farmers. Previous experience in irrigation management was another important background characteristic that had a great impact on the success of the PRIP beneficiaries. Some beneficiaries had farmed in large irrigation systems, like the PRIP. Others were farmers in small irrigation systems and some did not have experience in farming or irrigated farming systems. Ejidatarios were divided into three categories. The first category included farmers (43%) who had previous experience with large irrigation projects, like the Panuco River Irrigation Project. The second category consisted of “ In Mexico, the migrant agricultural laborers are workers who move from farm to farm for the purpose of harvesting the crops. "h, 111 a group (14%), who had experience with small irrigation systems (e.g., well irrigation systems). The third group was (43%) made up of individuals with no prior working experience with irrigation. Even though some groups needed training in irrigation management, only a small group of ejidatarios (17%) received some training at the beginning of their settlement. After more than ten years, most ejidatarios (64%) felt that they had acquired expertise in irrigated farming systems. However, their learning process was slow and often distressful, relying primarily on "trial and error". Ejidatarios' Off-Farm.Jobs The PRIP planners initially assumed that ten hectares of land along with government support were sufficient to allow ejidatarios to maintain their families without having an off-farm job. Off-farm jobs could help ejidatarios to earn extra money, but they would not be the main source of income. Off—farm jobs played an important role in the ejidatario family’s subsistence. Whenever ejidatarios lost their agricultural produce, they looked for an off-farm job. Ejidatarios often use their income from salaries to subsidize on-farm inputs. For example, ejidatarios would use income gained from work as bricklayers during the off- season to buy fertilizers for their crops. In 1987, a high proportion of ejidatario families (79%) 112 held one or more off-farm jobs --58% had temporary jobs, 9% permanent jobs, and 23% their own businesses. The situation changed in 1995 when the percentage of households with off- farm jobs decreased to 65% (total population). A normal approximation to the binomial test was applied to determine if the percentage of ejidatarios with off-farm jobs significantly increased from 1987 to 1995. fl - no 2 = no(1-no) n Where: fi' = the observed proportion 15 = the percentage selected to detect a deviation from 11 = sample size Substituting the values into the formula: .79 - .65 .14 z = =-—————— = 2.22 .063 .65(1-.65) 57 The calculated Z-value (2.22) is greater than the ZJB=1.64. Thus, the percentage of ejidatarios with an off- farm job decreased significantly from 1987 to 1995 at .05 level of significance. In 1987, ejidatarios of the paired subsample worked 141 days in off-farm salaried jobs (temporary plus permanent) and 214 days in 1995. It represents an increase of 52% in 113 the number of days worked off-farm by ejidatarios from 1987 to 1995. A t-test was performed on the mean number of days worked off-farm. The z—value calculated was 2.71 which is greater than zjm=1.64. There is a significant difference between the number of days worked off-farm in 1987 and in 1995. In the case of the 9 strata of the paired sample, there was a significant increase in the number of days worked off-farm in the strata number 1, 2, 3, and 4 from 1987 to 1995. The strata 5, 6 and 7 did not have a significant difference in the same period. Finally, the 8 and 9 strata had a significant decrease in the number of days worked off-farm. There is evidence that lower income strata had a significant tendency toward an increase in the number of days in off-farm jobs while the higher income strata showed a tendency toward a reduction of their off- farm work. The reason for these changes is that the composition of the off-farm employment changed substantially from 1987 to 1995. The major change occurred in the percentage of households with one or more permanent salaried jobs which increased from 9% to 23%, especially in the lower strata. In contrast, the percentage of ejidatario households with temporary salaried jobs went from 58% to 40%, in 1987 and 1995 respectively. These changes suggest that ejidatarios and their families spent more time outside of their farms in 1995 than in 1987. 114 Off-Farm.Net Income (general samples) In Mexico, it is difficult to find a rural area where farmers’ households exclusively depend on income from their own farms. However, if farmers earn enough money on their farms, few look for outside employment. Throughout Mexico, farmers and their families prefer to work on their own farms, but the weather and other circumstances negatively affect agricultural production and force ejidatarios to seek off—farm jobs. Many ejidatario— farmers get off-season jobs to survive during the rest of the year when they do not have work on their own farms. This is especially true when they do not have a large enough harvest to recover their production costs. Temporary jobs provide farmers with some cash income to purchase food and other farming inputs. Scarcity of resources eventually compels families to look for temporary or permanent salaried jobs to cover the temporary or permanent cash flow deficits of their farms. Off—farm income increased from $4,974.09 pesos in 1987 to $8,062.61 pesos in 1995, as it was shown above in Tables 5 and 6. A t-test helped to determine if the increase was statistically significant. The parameters of the 1987 off- farm income are §1=4,974.09, s=18977.22, n1=238 and for 1995, i,=8,062.61, 82:17,734.18, n2=150. ————— 115 Substituting these values in the formula: 8,062.61 - 4,974.09 t = = 1.98 8,9772 17,7342 ——+——— 238 150 The calculated t-value (1.98) is larger than the tjn=1.64. Ejidatarios made significantly more money from off-farm sources in 1995 than they did eight years earlier, in 1987 from similar off-farm sources. Off-Farm Net Income (paired subsample) A t-test was performed on the mean off-farm income difference of the paired subsample to confirm what was found in the analysis of the general samples. The calculated estimates for a paired difference t-test were 5 = 6,787.07, 3 = 25,511.23 and n=57. 6,787.07 t: =2.01 25,511.23 / V57 The results indicate that the calculated t-value (2.01) is much higher than the t-value at .05 level (1.64) with df=56. This indicates that the income difference is greater than zero. The t-test showed that there was a significant increase in the off-farm income from 1987 to 1995. This 116 corroborates the previous conclusion that off-farm activities had increased in importance during the 1987—1995 period. Off-Parm.Net Income (stratified subsample) The off-farm income was significantly higher for ejidatarios in 1995 than in 1987 (Table 11). However, not all of the income clusters increased their off-farm income in the same proportion. The first five strata had a positive increase in their off-farm income from 1987 to 1995, while the last four strata made less money from off- farm income sources. Paired difference t-tests were performed to determine if these income differences were significant. Comparing the calculated t-values in the seventh column against the t- value from tables with df=n-1 in the eight column, it was found that the ejidatarios of the third and fourth strata significantly increased their off-farm income from 1987 to 1995. Most of the strata (1, 2, 5, 6, 8 and 9) did not have a significantly different off-farm income at .1 significance level with df=n—1. The ejidatarios of these strata sustained their off-farm income. Only the seventh cluster experienced a significant reduction of their off-farm incomes between 1987 and 1995. No tendency was found between lower or higher income strata regarding their off- farm income. .mommm uncommon mmma CH mum mwSHm> kHMuwooE won Had .OwHMumpwhm comm now mEoooH bmma mnu Scum mama man msfluumuuQSm omuMHSUHmo mos some mEoocH moomumuufip one « m mmm.a- Ho.m mm.aam.mm so.bm>.m mm.¢HH.HH mm.smm.¢ hm oases mmm.a- mn.H- HH.mmm.m oo.mmn.m- oo.o oo.mms.m m m mmm.H- os.a- mm.smm.m mm.Hmm.m- mm.amm.m om.mam.m a m mmm.fl- mm.m- Hm.moo.s oc.amm.m- oo.msm.a oo.mma.oa m s mmm.a- mm.o- om.sms.m oo.mms- oo.¢ma.w oo.mmm.¢ m m 06¢.H bm.o mm.mmm.b m~.mmo.a sm.mmm.m mm.msm.¢ s m smm.a ha.m mm.as>.ma mm.mmm.m sm.mflm.ma mm.mmm.¢ m a mmm.a FF.H mm.mam.mm mm.mm¢.va oo.mmo.sa sm.mom.m NH m mmm.H mm.H mm.¢om.s¢ oo.omm.om oo.m¢m.am oo.mam.a OH m mso.m oo.H as.mmm.m oo.oom.a oo.oom.m oo.ooo.H m H a mo m mama ammo Tonto Emopumb 23.233 MUmemmmHo an? H u omyfipuqou 385 m5. mo 2%: 2.4m: mzoqu amino 2 macaw Auflmauunum uuuflum ominounuumo mama can scan nousuon uwunonouufln uaauan noo:_anou-uuo .mowuuuunwmm .HH wanna 118 Land Renting Land renting included the rent of ejido land for cropping and the rent of grasses for cattle grazing. Normally, outsiders rent land from ejidatarios, but land renting can happen between an ejidatarios and another ejidatario member of the same ejido. Without owning land, land renters can manage thousands of hectares. There were cases in which all the ejidatarios of one ejidb rented their land to outsiders. Land renting is the effect of multiple causes, but it indicates that ejidatarios’ farms were not profitable and/or ejidatarios did not have sufficient resources to operate their farms. Despite the fact that ejido land renting was prohibited by law, ejidatarios rented their land to other farmers since the beginning of the PRIP. In 1991, a constitutional amendment allowed ejidos to sell or to rent their land. Since then, land renting has increased tremendously, even though ejidatarios receive very low rent fees for their lands. In 1995, the rent fee of one hectare of irrigated land ranged from $300 to $400 pesos a year (approximately $60 U.S. dollars in 1995). Ejidatarios were aware that the rent fee was low, but it was secure money. In 1987, few ejidatarios (12%) partially or totally rented their land for cropping or grazing. By 1995, 37% of the ejidatario beneficiaries of the PRIP rented land to private farmers or international companies. A normal 119 approximation to the binomial test was performed to determine if the percentage difference of ejidatarios who rent their land was statistically significant. .37 - .12 .25 Z = = -————— = 5.81 .043 .12(1-.12) 57 The resulting difference is significant at .05 significance level. A significantly higher percentage of ejidatarios rented their land in 1995 than in 1987. Ejidatarios made $517.54 pesos in 1987 and $1,779.30 pesos in 1995 for land renting, an income difference of $1,261.76 pesos. A t-test was performed on the land renting income difference of the 57 paired subsample. The parameters used to perform the test were 5 = 1,261.76 pesos, s = 3,274 and n=57. 1,261.76 t = = 2.91 3,274 / V57 The calculated t-value is higher than the t.05 with df=n—1 from tables, therefore, ejidatarios made more money from land renting in 1995 than in 1987. The analysis of land renting of the 9 income strata (Table 12) shows that the percentage of farmers who rent 120 their land to outsiders significantly increased in most of the strata (from the second to the eighth stratum). The ninth cluster had a significant decrease in the percentage of farmers renting their land. The only strata without significant change was the first stratum, which did not have income for land renting in either 1987 or 1995. The mean income for land renting increased in seven of the nine strata (from the second to the eighth stratum). The first stratum did not have income from land renting in 1987 and in 1995. The ninth stratum decreased their income. A t—test was performed on each of the mean income differences to determine if they were significant. The results of the t-test are shown in the eighth column of Table 12. The results were compared to the t1 with df=n-1. The mean income difference was significant in the second, third and sixth strata. In the rest of the strata the income differences were not significant. The paradox was that the highest ratio of land renting occurred in the "Las Animas" irrigation unit, which has the best natural resources of the three irrigation units in PRIP. In 1995, 51% of the ejidatarios rented their lands in Las Animas, 38% in Pujal Coy, and 18% in Chicayan. Coincidentally, the "Las Animas" unit had the highest percentage of ejidatarios who were resettled from other states, followed by the Pujal Coy unit and then by the Chicayan unit. 1 2 l .mOmma ucmumcou mama CH mum mmzam> humumcoe mnu HH< s Hm.m mm.vbm.m mb.HmN.H om.mbb.H H.Nv mm.bam o.¢H AdHOB oo.H- mm.omm.m oo.ooo.mu 00.0 0.0 oo.ooo.m m.mm m oo.H oo.omm.m oo.mNH.H oo.mNH.H o.mm 00.0 0.0 m am.H hm.mmv.m oo.omm.H oo.omm.a o.ov 00.0 0.0 a bH.~ mm.mam.m oo.oom.m oo.ooo.m 0.0m oo.ooo o.om m mm.H Hm.bN¢.H Ha.me Hm.me m.mN 00.0 0.0 m 0H.H om.mHm.m mm.mmv.H mm.mmm.m m.mm av.¢mm m.mm a mm.m mm.mma.m ha.mmm.a om.mam.a 0.0m mm.mm m.m m mm.m ¢H.om¢.m oo.mmH.N oo.obm.m 0.0m oo.mov 0.0m N . 00.0 00.0 00.0 0.0 oo.o 0.0 H meoucfl m0ahpu meoucfi moflumu : a boon -mpflflm » a sump -mpfiflm » SDdeHm pmumaooamu coflumfl>mp mOCmHmMMHp moam>-u pumpcmum mEoucfi mama mama mama can puma seasons annuaom onus you nouaououuao cannon can: .uouuuuuuwhu Aoaasuunsm condom uofluauuuumo .NH OHAdB 122 E. Summary The limited benefits to the poor farmers in irrigation projects have been continually reported, but governments have done little to understand the problem, explore the causes, or address the problems of the poor participants in irrigation projects. The Panuco Project is an excellent example of the situation of poor farmers in large irrigation projects which is similar to the situation of poor farmers in other parts of Mexico and around the World. PRIP was created in 1974. .It is composed of three irrigation units Las Animas, Pujal— Cmoy and Chicayan. It intended to irrigate 144,000 ha and kaenefit 8,500 ejidatarios. This research project looks at the following: Can Mexican irrigation projects, like the Panuco River Irrigation Project, deliver intended benefits to poor farmers, called ejidatarios?. The research objectives of this study are: 1) To evaluate if the benefits from large scale government supported irrigation projects have been sustainable, using the Panuco River Irrigation Project as an example and 2) To evaluate whether the Panuco River Irrigation Project was able to meet its original objectives: a) To develop an intensive irrigation system, b) to contribute to national food security, c) To reduce the uneven income distribution between the rural and urban populations, and d) To reduce 123 unemployment among landless peasants. The overall null hypothesis was that the Panuco River Irrigation Project did not meet its initial objectives and was shown to be true. Four null subhypotheses were also formulated, for each of the project’s objectives: a) Ejidatarios perceived that PRIP did not develop an efficient irrigation delivery system, b) Ejidatarios did not produce ‘more food crops for the national food security in 1995 than in 1987, c) Ejidatarios did not increase their income from 1987 to 1995, and d) Finally, the fourth null subhypothesis is that the ejidatarios off-farm work increased from 1987 to 1995. The results of the hypothesis testing indicate that the four null subhypothesis were accepted. These results and the previous results presented in this chapter regarding on-farm income show that ejido farm lost its importance as the primary source of employment and income for the ejidatarios in the Panuco Project. This was envisioned by the planners of the PRIP when developed the project. CHAPTER VI CONCLUSIONS AND RECOMMENDATIONS Conclusions The Panuco River Irrigation Project is a good example of what happened with many of large irrigation projects in Mexico. Some of the of important characteristics of these projects were the construction of large scale irrigation infrastructure, huge government investments with international financial aid, settlement and relocation of large groups of people, complete transformation of the landscape, and governmental assistance to project beneficiaries. The Mexican government hoped these large irrigation projects would benefit their participant farmers by improving their income and well-being and the rest of the country by providing additional amounts of food. PRIP is probably one of the most dramatic failures of this approach in Mexico. The main purpose of this study was to determine whether or not the Panuco River Irrigation Project met its initial objectives. The specific interest of this study was to determine the impact of the PRIP on participant ejidatario- farmers, who were the main beneficiaries of this project. 124 125 The PRIP planners established four basic objectives: a) to develop an intensive irrigation system, b) to contribute to national food security, c) to reduce the uneven income distribution between the rural and urban populations, and d) to reduce unemployment among landless peasants. This study was based on the overall null hypothesis that states "the PRIP did not meet its initial objectives", plus four subhypotheses based on each PRIP's objectives: a) PRIP did not develop an efficient irrigation delivery system (Ho: w=.5), b) ejidatarios did not produce more food crops for the national food security in 1995 than in 1987 (Ho=lh=#1): c) ejidatarios did not increase their income from 1987 to 1995 (Ho:;h>pq), and d) ejidatarios off-farm work increased from 1987 to 1995, (Ho:;h