1 AEWP-2000-10 Analysis of a Climatically Variable Production Season Susan Materer and Corinne Valdivia1 December 2000 “Working Papers generally have not been reviewed by anyone other than the author. They are a part of the process of obtaining from readers comments and suggestions for the author to use in subsequent revisions of the paper and to inform colleagues of work in progress. Working papers should not be quoted or cited as a reference without approval of the author.” 1 Corinne Valdivia is a Research Assistant Professor, Department of Agricultural Economics- University of Missouri-Columbia and the Principal Investigator of the NOAA project. Susan Materer is a Masters’ student in Agricultural Economics as the University of Missouri-Columbia and is a Research Assistant for the project. Send correspondence to Corinne Validivia, Department of Agricultural Economics, Social Science Unit, University of Missouri-Columbia 65211. Tel (573) 882-4020, Fax (573) 882-3958 and E-Mail [email protected]. Or contact Susan Materer at [email protected] OF CONTENTS 1. ABSTRACT……………………………………………………….1 2. INTRODUCTION…………………………………………………2 3. LITERATURE REVIEW………………………………………….3 4. RESEARCH PROBLEM…………………………………………..5 5. IDENTIFICATION OF DEPENDENT AND INDPENDENT VARIABLES………………………………………………………6 6. ANLAYSIS OF INDEPENENT VARIABLES IN EXPLAINING TOTAL INCOME…………………………………………………..8 7. ANALYSIS OF FINAL MODEL………………………………….11 8. TECHNICAL DISCUSSION………………………………………12 9. FINDINGS AND IMPLICATIONS………………………………..13 10. REFERENCES………………………………………………………14 11. ACKNOWLEDGMENTS…………………………………………..16 LIST OF TABLES TABLE 1: Pearson Correlation of Coefficients of Variables…………….7 TABLE 2: Summary of Forward Selection Model………………………..8 TABLE 3: Summary of Forward Stepwise Selection with Labor Unit…….9 TABLE 4: R-squared and C(p) Analysis……………………………………9 TABLE 5: Summary of Subset B Regression Analysis……………………10 TABLE 6: Summary of Subset A Regression Analysis……………………10 TABLE 7: Summary of Final Model……………………………………….113 ABSTRACT Analysis of a small agropastoral community in Bolivia during a climatically vulnerable year, provided insight into how households diversify and change productions strategies in order to secure income and food consumption. Twelve independent variables that identified distinct production strategies were analyzed according to how well they estimated total income. A subset selection process determined the best combination of variables, or production strategies. The four variables chosen were food plots, shared land, off farm income and number of native animals. These variables characterize traditional farming practices and have social and human capital embedded within them. It is important to understand how households mitigate and cope with climatic risks, in order to best develop methods that can help households during risky production seasons.4INTRODUCTION Data used in this analysis was extracted from the small peasant community of San José Llanga in the Bolivian Altiplano. San Jose participated in the Small Ruminant Collaborative Research Support Program-Bolivia (SR-CRSP), which was conducted from 1992-1995. SR-CRSP focused on understanding how the introduction of improved varieties of sheep and cattle affects household welfare. The purpose of using this data in this research differs from the initial analysis. It is the hopes of this research project to identify household strategies and characteristics that aid farmers in securing income and food consumption in a harsh climatic zone. San José Llanga is particularly susceptible to climatic changes, since it is located between 3,725 and 3,786 meters above sea level. Frosts, droughts, and wind erosion are prevalent climatic events that affect production decisions and food security. Data from 1995 will be used since this was a drought year and it is assumed that household will secure income and food consumption, by employing traditional agricultural practices that are less vulnerable to the perturbations of climate. First an examination of literature that relates to household production systems and climatic vulnerability will be discussed. Second the research problem, setting and design will discussed. In the third part the independent and dependent variables will be analyzed and the correlation matrix examined. In the fourth section initial subset selection procedures will be tested, followed by a discussion of the final model. Then in the technical discussion section possible caveats to the model will be explored. Finally the findings and implications of the research will be examined.5LITERATURE REVIEW Constrains to choice sets, climate, knowledge systems and social and human capital have influential roles in the ability of a household to secure income and food and the livelihood strategies employed. The political, social and physical environment of rural households’ surroundings also influences the stability and decisions of the households. Any changes in these factors can affect the stability and capabilities of the household to secure income and food for present and future consumption. Therefore households construct diverse portfolios of economic activities and secure social support systems in order to stabilize or increase household welfare (Ellis, 1998). The strategies employed by households are not stable and fluctuate depending on the given parameters. The economic activities of the household portfolio characterize it as both a producer and a consumer. The household must produce to secure income but also to ensure that there is an adequate supply of food throughout the agricultural year. Ellis (1993) describes this as a partial engagement of the household in market activities, which will fluctuate depending on market prices, agricultural productivity, and household needs. The partial engagement of households in the market place is necessary because of