Agricultural Commercialization, Nutrition, and the Rural Poor: A Study of Philippine Farm Households 9781685858629

Table of contents :
Contents
Tables and Figures
Foreword
Acknowledgments
Map of the Philippines, Indicating the Study Area Location of the Survey Barios Relative to BUSCO, Southern Bukidnon Province Study Area
1 Agricultural Commercialization, Nutrition, and Government Policy in Less Developed Countries
2 Research Design, Conceptual Framework, and Sample Selection
3 Changes in Land Tenure Patterns
4 The Corn Production System
5 The Sugar Production System
6 Comparing Corn and Sugar Production: Economic Profits, Changes in Labor Allocation Patterns, Other Sources of Income
7 Incomes, Expenditures, and Calorie Intakes at the Household Level
8 Individual Calorie Intakes
9 Heights and Weights of Preschool Children
10 Conclusions and Policy Recommendations
Appendixes
Appendix A Prices for Selected Inputs by Crop-Tenancy Group
Appendix B Carabao Input Demand Functions
C Detailed Production Function Estimations
D Family Participation Rates in Corn Processing, Disposition of Corn Bran, and Prices Paid and Received at Various Stages of Corn Processing
E Total Expenditures and Household Calorie Availability as Proxies for Income and Household Calorie Intakes
F Descriptive Statistics for Variables Used in Regression Estimates
G Results of Tests for Equality of Coefficients Between Corn and Sugar Households
H Results of Tests for Exogeneity
I Should the Model be Treated as Simultaneous or Recursive?
Bibliography
Index

Citation preview

Agricultural Commercialization, Nutrition, and the Rural Poor

This book is published in cooperation with the International Food Policy Research Institute and is one result of a research project undertaken in collaboration with the Research Institute for Mindanao Culture and the Institute of Market Analysis, Xavier University, Cagayan de Oro, Philippines.

Agricultural Commercialization, Nutrition, and the Rural Poor A Study of Philippine Farm Households Howarth E. Bouis Lawrence J. Haddad

Lynne Rienner Publishers • Boulder and London

Published in the United States of America in 1990 by Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 and in the United Kingdom by Lynne Rienner Publishers, Inc. 3 Henrietta Street, Covent Garden, London WC2E 8LU © 1990IFPRI All rights reserved L i b r a r y of Congress Cataloging-in-Publication Data Agricultural commercialization, nutrition, and the rural poor : a study of Philippine farm households /Howarth E. Bouis and Lawrence J. Haddad. p. cm. Includes bibliographical references ISBN 1-55587-206-9 (alk. paper) 1. Agriculture—Economic aspects—Philippines. 2. Rural poor— Philippines—Nutrition. 3. Produce trade—Philippines. 4. Farm income—Philippines. 5. Household surveys—Philippines. I. Bouis, Howarth E. II. Haddad, Lawrence James. HD20887.A37 1990 90-8074 338.1'09599—dc20 CIP British Cataloguing in Publication D a t a A Cataloguing in Publication record for this book is available from the British Library.

Printed and bound in the United States of America The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48-1984.

Contents

List of Tables and Figures Foreword, John W. Mellor Acknowledgments Map of the Philippines, Indicating the Study Area Location of the Survey Barios Relative to BUSCO, Southern Bukidnon Province Study Area 1

2

Agricultural Commercialization, Nutrition, and Government Policy in Less Developed Countries

vii xi xiii xv xvi

1

Research Design, Conceptual Framework, and Sample Selection

8

3

Changes in Land Tenure Patterns

19

4

The Com Production System

35

5

The Sugar Production System

53

6

Comparing Corn and Sugar Production: Economic Profits, Changes in Labor Allocation Patterns, Other Sources of Income

67

7

Incomes, Expenditures, and Calorie Intakes at the Household Level

78

8

Individual Calorie Intakes

97

9

Heights and Weights of Preschool Children v

108

Vi

CONTENTS

10

Conclusions and Policy Recommendations

134

Appendixes A B C D

E F G H I

Prices for Selected Inputs by Crop-Tenancy Group Carabao Input Demand Functions Detailed Production Function Estimations Family Participation Rates in Corn Processing, Disposition of Corn Bran, and Prices Paid and Received at Various Stages of Corn Processing Total Expenditures and Household Calorie Availability as Proxies for Income and Household Calorie Intakes Descriptive Statistics for Variables Used in Regression Estimates Results of Tests for Equality of Coefficients Between Corn and Sugar Households Results of Tests for Exogeneity Should the Model be Treated as Simultaneous or Recursive?

Bibliography Index

153 157 164

170 172 178 180 181 183

185 189

Tables and Figures TABLES

1.1

Area planted to cash crops and total crop area for 78 developing countries, 1982 2.1 List of topics covered by survey questionnaires, 1984/85 2.2 Selected data for respondent households, by crop-tenancy group, 1984/85 3.1 Percentage distribution of sugar and corn farms and area harvested in Bukidnon Province, by farm size, 1971 and 1980 3.2 Present primary occupations of heads of households, all municipalities, April 1984 3.3 Present primary occupations of heads of households, by municipalities close to and far away from the mill 3.4 Sugar-producing households, by land tenure and previous occupation, April 1984 3.5 Household employment and land tenure status during round 4 of survey and before sugar mill operation, July 1985 3.6 Area cultivated during survey period, area lost since establishment of BUSCO, size of sugar farm on which agricultural wages were earned, per capita expenditures, and per capita income, by sugar and com households and change in tenure status since BUSCO was established 4.1 Comparison of past and present corn yields, by crop-tenancy group and farm size, 1984/85 4.2 Trend in share of harvest paid to com laborers 4.3 Total labor inputs per com crop, by family and hired labor, crop-tenancy group, and farm size, 1984/85 4.4 Total labor inputs per com crop, by task, crop-tenancy group, and farm size, 1984/85 4.5 Total labor inputs per com crop, by crop-tenancy group, family and hired labor, and farm size, 1984/85 4.6 Women and children participation rates for various tasks in com production, by family and hired labor, all households, 1984/85 4.7 Total production expenditures per com crop, by crop-tenancy group, cash and in-kind wages, cash paid for nonlabor inputs, and farm size, 1984/85 4.8 Shelled com production and consumption of own production, by crop-tenancy group and farm size, 1984/85 4.9 Total and net revenues per com crop, by crop-tenancy group vii

7 19 20 31 32 33 34 35

36 46 47 48 49 50

51

52 53

Viii

5.1 5.2 5.3 5.4 5.5 5.6

5.7 5.8 5.9 6.1 6.2 6.3 6.4 6.5

6.6 7.1

7.2 7.3 7.4 7.5 7.6

TABLES & FIGURES

and farm size, 1984/85 Sugar yields, by crop-tenancy group and milling season Total labor inputs per sugar crop, by crop-tenancy group and family and hired labor, 1984/85 Total labor inputs per sugar crop, by task and crop-tenancy group, 1984/85 Total labor inputs per sugar crop, by family and hired labor and crop-tenancy group, 1984/85 Women and children participation rates for various tasks in sugar production, by family and hired labor, all households Total production expenditures per sugar crop, by cash and in-kind wages, cash paid fornonlabor inputs, and crop-tenancy group Total production expenditures per sugar crop, by plant and ratoon crops and crop-tenancy group Total and net revenues per sugar crop, by crop-tenancy group, 1983/84 milling season Total and net revenues per sugar crop, by crop-tenancy group, 1984/85 milling season Corn production profits per hectare per crop, by crop-tenancy group and farm size, 1984/85 Sugar production profits per hectare per crop, by crop-tenancy group and farm size, 1984/85 Labor inputs for corn and sugar, by family and hired labor Family and hired labor inputs for corn and sugar production for selected tasks, total sample average, 1984/85 Family and hired labor inputs for weeding by hand for com and sugar production, by crop-tenancy group and farm size, 1984/85 Sources of income, by expenditure quintile and crop-tenancy group, 1984/85 Income, expenditures, and calorie availability and intake, by income and expenditure quintiles and crop-tenancy group, 1984/85 Nonfood expenditures and budget shares, by expenditure quintile and commodity group, 1984/85 Nonfood expenditures for selected commodity groups, by expenditure quintile and crop-tenancy group, 1984/85 Allocation of per capita food expenditures, by source, expenditure quintile, and crop-tenancy group, 1984/85 Com and rice expenditures, by source, expenditure quintile, and crop-tenancy group, 1984/85 Regression results for the relationship between income and food expenditures

54 63 63 64 65 65

66 66 67 68 75 76 77 77

78 79

90 91 92 93 94 95

TABLES & FIGURES

Calories purchased per peso, by crop-tenancy group and expenditure quintile, 1984/85 7.8 Allocation of food expenditures, by food group, expenditure quintile, and crop-tenancy group, 1984/85 7.9 Regression results for the relationship between calorie intakes and food expenditures 8.1 Calorie adequacy, by family member, expenditure quintile, and crop-tenancy group 8.2 Preschooler calorie intakes, by whether preschooler was sick with any symptom in the past two weeks, had a fever, and had diarrhea, 1984/85 8.3 Regression results for preschooler calorie intakes as a function of household calorie intakes 8.4 Coefficients on ratio of parents' calorie intake over household calorie intake for preschooler calorie intake regressions, by crop-tenancy group and household size 9.1 Z-scores for height-for-age, weight-for-age, and weight-forhcight of preschoolers, by age and expenditure quintile, 1984/85 9.2 Height-for-age Z-scores for preschoolers who no longer breastfeed, by crop-tenancy group and age tercile 9.3 Mothers' weight and height, by physiological status, expenditure quintile, and crop-tenancy group 9.4 Allocation of mothers' time, by expenditure quintile and crop-tenancy group 9.5 Allocation of mothers' time by physiological status, expenditure quintile, access to land, and primary crop grown 9.6 Mothers' time away from home, time spent in strenuous activities, calorie adequacy ratios, and weights, by physiological status, expenditure quintile, access to land, and primary crop grown 9.7 Prevalence of sickness among breastfeeding and nonbreastfeeding preschoolers, by expenditure quintile and crop-tenancy group, 1984/85 9.8 Sanitation characteristics, by expenditure quintile and crop-tenancy group, 1984/85 9.9 Infant feeding practices, by expenditure quintile and croptenancy group, 1984/85 9.10 Regression results for the relationship between preschoolers ' calorie intake and weight-for-height 9.11 Regression results for the relationship between preschoolers' calorie intake and height-for-age 9.12 Summary of selected coefficients for height-for-age regressions, by age tercile and crop-tenancy group

ix

7.7

96 97 98 106

107 108

109

122 123 124 125 126

128

130 131 132 133 134 135

X

TABLES & FIGURES

FIGURES

2.1 4.1 5.1

Household resource allocation and nutrition Seasonal cropping pattern for corn production Seasonal cropping pattern for sugar production

16 45 62

APPENDIX TABLES

A. 1 Wages paid to com laborers, by task and crop-tenancy group, 1984/85 A.2 Wages paid to sugar laborers for selected tasks, by croptenancy group, 1984/85 A.3 Rents paid for use of carabaos and tractors, by crop and crop-tenancy group, 1984/85 A.4 Prices paid for fertilizer, by crop and crop-tenancy group, 1984/85 B. 1 Carabao inputs per hectare per com crop, by crop-tenancy group, 1984/85 B.2 Number of carabaos owned per household and per hectare cultivated, by crop-tenancy group, 1984/85 B.3 Regression results for carabao input demand functions for com production B.4 Carabao inputs per hectare per sugar crop, by crop-tenancy group, 1984/85 B.5 Regression results for carabao input demand functions for sugar production C. 1 Regression results for com production function (linear functional form) C.2 Regression results for the com production function (variation of linear functional form) C.3 Regression results for sugar production function (linear functional form) D. 1 Percent of com shelled, dried, and hauled, by crop-tenancy group, 1984/85 D.2 Production, use, and processing costs of com bran, by crop-tenancy group and prices received for com output, 1984/85 E. 1 Income elasticities of household calorie consumption implied by the two-way relationship between income and household calorie intakes F. 1 Descriptive statistics for variables used in regression analysis G. 1 Results of testing for equality of coefficients between com and sugar households H. 1 Results of Hausman tests for exogeneity of right-hand side variables

155 156 157 158 161 162 163 164 165 169 170 171 172

173

179 180 182 183

Foreword

Over the past several years the International Food Policy Research Institute (IFPRI) has undertaken research on the production, consumption, and nutrition effects of agricultural commercialization in Gambia, Guatemala, Kenya, the Philippines, and Rwanda. While it is widely recognized that the commercialization of agriculture is essential to overall economic development, various rural population groups adapt differently to the process of commercialization, depending on the resources available to them, economic and social conditions, and government policies. Many households benefit in the form of higher incomes; others may suffer a decline in income. A particular concern of policymakers has been the effect of commercialization on nutrition. The purpose of these five studies has been to analyze the process of commercialization in order to identify key factors that determine nutritional outcomes, with the objective of formulating policies to enhance the beneficial effects of commercialization and to minimize the harmful effects. As discussed in Chapter 1 of this book, interregional and international trade, which are implied by the process of agricultural commercialization, may have important benefits for overall economic development. Government macroeconomic and trade policies may have a decisive influence on how quickly agricultural commercialization proceeds and how the potential benefits from commercialization are distributed. While recognizing the significance of these broader policies (which are addressed by other IFPRI research projects), the five commercialization studies have concentrated instead on microlevel concerns. To do so required collection of detailed household and individual data. The benefits of commercialization can be very unevenly distributed, as documented in this book, by the increased concentration of landholdings resulting from the introduction of a profitable new crop. Decisionmakers have an obvious need to know how various constituencies are affected by the policies they undertake. The data sets laboriously gathered for this purpose

xi

Xii

FOREWORD

also increase our understanding of the development process. While at an aggregate level we can measure shifts in supply and demand in response to changes in prices and other factors, sound policy formulation ultimately requires a careful understanding of the human behavior that generates particular magnitudes. To attain such an understanding often requires a substantial investment of time and other resources. Work on this book began in 1982, with a trip to the Philippine island of Mindanao to identify an appropriate research setting and to gather other information for writing a research proposal—after consultation with persons from several less-developed countries had identified the effect of commercialization on nutrition as a high-priority research area, and after a common methodology for studying this issue in several countries had been researched and discussed within IFPRI. As with all the commercialization studies, several months of fieldwork were required to develop survey instruments even before data collection could begin. The considerable task of inputting the survey information into a computer software system, and subsequent review and organization of the variables before analysis can begin, are often underestimated by those unfamiliar with the process. Howarth Bouis and Lawrence Haddad present the detailed findings from the case study of Mindanao, where a substantial number of households converted lands from corn to sugarcane production after construction of a sugar mill. The main effects of the introduction of export cropping in this area were a significant deterioration in access to land, as smallholder com tenant farms using primarily family labor were consolidated into larger sugar farms using primarily hired labor, an increase in incomes for households that grew sugarcane; a decline in women's participation in own-farm production; and very little improvement in nutritional status as a result of increased incomes from sugarcane production, primarily because of the high levels of preschooler sickness in the sugarcane-growing households. The difficulty of generalizing as to the varied effects of agricultural commercialization is brought out by a comparison with the case study in Kenya, where farmers also switched from maize to sugarcane production. In that African setting, where land is often relatively abundant and labor scarce compared with many places in Asia, women increased their participation in o w n - f a r m production as cultivated area e x p a n d e d ; in the P h i l i p p i n e s women's participation in farm production decreased as land planted to sugarcane replaced land previously devoted to com production. Yet there are important similarities as well. As all of the commercialization studies have confirmed, poor health and sanitation conditions are serious constraints to the improved nutrition that increases in income might otherwise have made possible. John W. Mellor

Acknowledgments

This book has been many years in the making, and we would like to acknowledge several people who gave more of themselves than we could have reasonably expected and without whom this book would not have been possible. We would like to thank all the staff at the Research Institute for Mindanao Culture, especially the supervisors under the direction of Lourdes Wong. The project was literally in their hands for a year and a half and we learned a lot from them about how a survey should be run. We are indebted to Father Francis Madigan for putting together such a professional staff over so many years of dedicated work at RIMCU. We are deeply grateful to Azucena Limbo of the Nutrition Foundation of the Philippines, who provided training for collection of our anthropometric and 24-hour food-recall data—no doubt the most difficult portion of our questionnaire to administer and code. Her competence and enthusiasm helped us through some difficult times. We thank Joel Asentista, head of the Xavier University extension program in Bukidnon, and Roberto Montalvan, president of the Bukidnon Sugar Planters Association, for giving up so much of their personal time to familiarize us with the study area; as well as Father Antonio Ledesma and the graduate students of the Institute for Market Analysis at Xavier University, whose background knowledge of the area and logistical support were essential inputs into our survey work. The help of Per Pinstrup-Andersen, whose efforts provided the initial impetus for all of the IFPRI commercialization studies, is gratefully acknowledged. Discussions that compared the Philippine setting with the other four country case studies, undertaken by Joachim von Braun and Eileen Kennedy, provided valuable insights. For detailed comments on various drafts of the entire book, we would like to thank Romeo Bautista, Rodolfo Florentino, Yujiro Hayami, Francis Madigan, John Mellor, Earl Swanson, and Steve Vosti. For helpful comments on specific components of the

xiii

Xiv

ACKNOWLEDGMENTS

research, we would like to thank Julie Anderson, Gelia Castillo, Reynaldo Martorell, Marc Nerlove, and Anne Peck. Any errors and omissions remaining in the book are, of course, our responsibility alone. Finally, we would like to thank the U.S. Agency for International Development for giving us the opportunity and freedom to undertake the research by funding the project (under Grant No. OTR-OOOO-G-55-3513-00). Howarth E. Bouis Lawrence J. Haddad

Map of the Philippines, Indicating the Study Area

XV

- Primary provincial road Secondary provincial road with regular bus service

Ttertiary access road with occasional local transportation (All roads are graded and unsurfaced.)

Zone 1—Sugar production common Zone 2—Sugar production sparse Zone 3—Sugar production nonexistent or rare

Scale I

I 10 kilometers 10 miles

Location of the Survey Barios Relative to BUSCO, Southern Bukidnon Province Survey Area XV i

1 Agricultural Commercialization, Nutrition, and Government Policy in Less Developed Countries

The commercialization of agriculture, and in particular export cropping, has often been blamed as a cause of poor nutrition. In many less developed countries, vast amounts of land and other inputs arc devoted to the production of agricultural commodities for export. This production, whether of staple or nonstaple foods or inputs for manufacture, whether sent abroad or to other regions of the same country, often coexists with significant malnutrition. Might not the resources that were used to produce these exports have been used instead to produce food for the local economy to reduce, or even to eliminate, the problem of malnutrition? Many economists would argue that the process of commercialization, by raising incomes, actually improves a nutritional situation that might have been worse otherwise. 1 Specialization, the development of markets, and trade, which characterize commercialization, are fundamental to economic growth. But how are higher average incomes distributed among various economic and social groups as commercialization takes place? Does a higher household income necessarily mean better nutrition for all household members? Could a different approach to agricultural development, one of regional, or village, or even household food self-sufficiency, better alleviate the particular problem of malnutrition and still meet the objectives of economic growth and higher incomes? Because there are so many possible policy variations within the competing paradigms of specialization and self-sufficiency, because economic and social conditions vary so much across countries and regions, and finally because there are inevitably winners and losers in any process of change, unfortunately it is impossible to answer such crucial questions in any general and definitive way. In order to provide some guidance for policy formulation in this area, however, what is possible is to study the process of commercialization in specific contexts and to identify key factors that appear to lead either to beneficial or to detrimental outcomes in terms of nutrition. In designing and implementing future projects and policies, then, our objective 1

2

AGRICULTURAL COMMERCIALIZATION

would be to find ways for policymakers to enhance the beneficial factors and minimize the harmful ones.2 This book presents the detailed findings of a case study research project undertaken in Bukidnon Province (on Mindanao, an island in the southern Philippines), an area primarily engaged in semisubsistence corn production until the establishment of a sugar mill in 1977 led to a rapid expansion of sugarcane production.

THE EXTENT OF AGRICULTURAL COMMERCIALIZATION

How widespread is agricultural commercialization in developing countries? A recent study by Joachim von Braun and Eileen Kennedy (1986) estimated that, in 1982,22 percent of total cropped area in 78 developing countries was devoted to cash crops (see Table 1.1, at the end of this chapter), a figure that, as the authors note, understates the extent of commercialization because a considerable percentage of food crops are also marketed. In 32 out of the 78 countries studied, the ratio of cash crop area to basic food crop area was greater than 50 percent. A more significant aspect of their findings, however, was the patterns in shifts in land devoted to cash crops over time. Between 1968 and 1982 about 25 percent of the countries studied experienced a considerable decrease in land used for cash cropping; in another 25 percent of the countries studied, land used for cash cropping increased significantly. Most of the countries that witnessed a decline in land devoted to cash cropping—a disproportionate number of which were in the low-income group—were in Africa, whereas a majority of Asian and Latin American countries experienced constant or increased shares of land used for cash crops. These data contradict a frequent assumption that expansion of cash cropping necessarily means a decline in food crop production. In their study, von Braun and Kennedy found that growth in the area allocated to cash crops was generally positively associated with per capita staple food production. They conclude that most countries either manage growth in both cash cropping and food production or fail to manage either.

GOVERNMENT MACROECONOMIC POLICIES

The study by von Braun and Kennedy (1986) provides useful evidence on some empirical relationships between cash crop and food crop production at the national level. Although economy-wide or macrolevel effects of agricultural commercialization are not the focus of this book, it is nevertheless important to keep in mind that how governments manage their macro policies may well determine how the potential benefits and/or unwanted effects of

COMMERCIALIZATION, NUTRITION & POLICY

3

cash cropping impact both those directly involved in the production of these crops and those not directly involved. For example, trade economists have long argued that free trade between countries can result in higher incomes for all parties involved, as each nation exploits its own particular comparative advantage, producing more of what it produces most efficiently as compared with other products that it then can afford to import. Border taxes on agricultural exports often provide a convenient, "ost-efficient mechanism for collecting taxes in agriculturally dominated economies, which the government can then spend for education, infrastructure projects, and other investments for the development of any or all sectors of the economy. In practice, of course, such a system may work in less than an ideal way to the extent that barriers to free trade exist, often to the benefit of urban industry.3 Excessive export taxes themselves can significantly distort the flow of free trade, and tax monies are not always spent wisely and in the general interest. With respect to nutritional outcomes, a particular concern of critics of cash cropping is that if export cropping replaces food production and if export revenues are not used to import sufficient quantities of staple foods, then nationwide food prices will rise to the detriment of the poor and their nutrition. Although the von Braun and Kennedy (1986) finding that cash crop and food crop production in most countries seem to proceed in tandem should mollify some of these concerns, this does not necessarily occur in every country. Also, even though domestic food production outpaces population growth, food prices may still rise because of increased incomes, reduced food imports, or increased food exports; food prices might have been lower if domestic resources used to produce export crops had been used to produce staple foods instead. The final food price outcome in the short run depends critically on government trade and buffer stock policies. Even though such macrolevel concerns are justifiably an important component of the debate over the efficacy of export cropping and agricultural commercialization, this book concentrates instead on microlevel concerns. How are persons and households directly involved in the process of agricultural commercialization affected by this process? Whose incomes go up? Do some incomes decline? Do incomes become more or less diversified? How do households reorganize use of their time and farm assets? Do they spend their money differently, and do the foods they eat change? Does the individual nutritional status of particular household members get better, deteriorate, or remain unchanged? Why or why not? Apart from macrolevel considerations, these issues are also important concerns to government policymakers because they directly reflect the impact that government policies have on the current welfare of particular rural constituencies. What is perhaps less obvious, however, is that such microlevel analyses of how household behavior adapts to commercialization

4

AGRICULTURAL COMMERCIALIZATION

provide important information about structural changes that m a y be occurring in aggregate rural supply and demand behavior. As a sufficiently large number of households react in the same way over the longer run to a particular economic stimulus or policy, microlevel effects have macrolevel consequences.

THE PHILIPPINE POLICY SETTING

A l t h o u g h it is o f t e n the o b j e c t i v e of m i c r o l e v e l s t u d i e s to d e r i v e generalizable findings that can be usefully applied to other situations, it is in the nature of microlevel studies that relevant, complementary information also becomes available for the specific context in which the research is taking place. It is in this sense that this book is also about agricultural policies in the Philippines. In order to gauge the potential benefit of this book for agricultural policymaking in the Philippines, it is useful to provide some background information on the macro policy setting in which the study area in Mindanao is situated. The overarching issue that influences debate about all economic policies in the Philippines today is how to reduce the massive foreign debt, estimated at about US$28 billion—roughly US$500 of debt per capita for a country with a per capita annual income of about US$600. Debt service payments are about 10 percent of the total owed, or about 40 percent of export earnings. An economic crisis, precipitated in October 1983 by the Marcos administration's inability to meet its foreign debt obligations, is still largely unresolved. Agriculture has emerged from the ensuing policy debate as the key sector in economic recovery. Discussion of agricultural policies during the last years of the Marcos regime and through the first year of the Aquino administration centered on ridding the agricultural sector of monopolistic control by close associates of Marcos and on adjusting macroeconomic trade and fiscal policies so as not to be biased against agriculture. These issues address the problems of agriculture in a general way, focusing attention on some very necessary conditions for efficient allocation of private sector resources. There has been broad consensus on the necessary corrective measures, and some policies have already been put in place to rectify mistakes of the past. Public attention to agriculture shifted dramatically to the issue of land reform in January 1987, when several persons who were demonstrating for land reform near the presidential residence were killed by security forces. Land reform has since become the political litmus test of the ability of the Aquino administration to provide a better life for the rural poor. Agriculture is expected not only to generate much of the growth for the economy as a whole (and in the process contribute increased exports to help pay off the

C O M M E R C I A L I Z A T I O N , NUTRITION & POLICY

5

burdensome foreign debt) but also to accomplish this in the context of a significant redistribution of wealth through land reform. The government investment strategy in rural areas is a crucial component in achieving sustained high agricultural growth rates. The term "investments" is used here in a very broad sense to include expenditures for agricultural research and extension as well as expenditures for irrigation, roads, and other physical infrastructure. Technological change is essential for raising agricultural productivity; it is the sine qua non of high agricultural growth rates where land is a constraint, as is the case in the Philippines. Perhaps because of the understandable desire for getting government out of agriculture after the experience of the Marcos years, perhaps because what investment resources are available are very limited, and perhaps because of a preoccupation with generating short-run increases in exports to keep up with interest payments on the foreign debt, there has been relatively little discussion of the government investment strategy for agriculture. Much of the growth in cereal and export crop production in the past decade has occurred in the southern region of Mindanao. Because of its relatively even distribution of rainfall throughout the year, its position outside of the path of typhoons, and its lower population densities, Mindanao is better situated than Luzon for realizing rapid increases in agricultural productivity. Over the past 20 years rice yields in Mindanao have grown rapidly enough to now surpass average yields in Luzon, although corn is more widely grown than rice. Mindanao has also witnessed a rapid expansion of alternative export crop production, which includes the production of bananas, cacao, rubber, palm oil, coffee, and pineapples; this has reduced dependence on the traditionally preeminent agricultural exports of coconut products and sugar. Much of this expansion has taken place on large-scale operational units (Lim 1987). Not only will the growth of agriculture in Mindanao determine to a significant extent whether or not the high expectations for agriculture as a stimulant to economic recovery will be realized but also the policy choices to be made there are a microcosm of those confronting national agricultural policy. Now that a land constraint has been reached in Mindanao, should the government continue to promote the expansion of large-scale export crop p r o d u c t i o n as a m e a n s to earn f o r e i g n e x c h a n g e ? A l t e r n a t i v e l y , if distributional objectives are given precedence by encouraging smallholder export crop production, how much growth, if any, would be sacrificed? A third strategy would be to emphasize increased rice and corn production, which typically has been undertaken on smaller operational units and which may need to be imported in larger and larger quantities in the years ahead (Bouis 1989a). Under any of these three options, what would be the consequences for income levels and the nutritional status of the poor? As just discussed, any complete evaluation of these three broad alterna-

6

AGRICULTURAL COMMERCIALIZATION

tives would require construction of a multisectoral macroeconomic model that could determine agricultural supply and demand responses to marketclearing prices, which is well beyond the scope of this research. However, this book does provide a detailed household-level and individual-level look at what happened to land tenure patterns, incomes, and nutrition in an area in Mindanao that was primarily engaged in semisubsistence corn production and then switched to export cropping with the establishment of a sugar mill. In the Philippines as a whole, over three million hectares of corn are harvested each year—about the same area harvested as rice. Yet its production and consumption patterns at the household level have been studied relatively little, just as Mindanao has been relatively neglected in the socioeconomic literature. Much has been said and written about the decline of the sugar industry in the Philippines in the wake of low world prices, especially with reference to Negros, where most of the nation's sugar is produced. This research provides some hard evidence on net returns to sugar production of smallholder producers and on their nutritional status in a nontraditional sugar-growing area with a more diversified agricultural economy than exists in Negros.

NOTES 1. Well-known critics of export cropping are Frances M. Lappe and Joseph Collins (1977) and Susan George (1976). Although Jimmye S. Hillman (1981) cautions that a case-by-case examination is needed, he provides strong evidence in favor of export cropping. See von Braun and Kennedy (1986) for an assessment of the competition between cash crop and food crop production using aggregate country data. 2. Toward this end, the International Food Policy Research Institute (IFPRI) has conducted microlevel studies in five countries—the Gambia (von Braun, Puetz, and Webb 1989), Guatemala (von Braun, Hotchkiss, and Immink 1989), Kenya (Kennedy and Cogill 1987), the Philippines (Bouis and Haddad 1990), and Rwanda (von Braun, de Haen, and Blanken forthcoming)—in rural areas in which farm households have recently undergone a switch from semisubsistence staple food production to production of crops primarily for sale in the market. 3. For a general discussion of the effects of macroeconomic policies on agriculture and the food system, see chapter 5 of Timmer, Falcon, and Pearson (1983). Specific examples of these effects in several Asian countries are examined in Sicular (1989); the Philippines is discussed in chapter 5. Bautista (1987) also analyzes the effects of trade and exchange rate policies on agricultural production incentives in the Philippines.

COMMERCIALIZATION, NUTRITION & POLICY

TABLE

1.1

7

Area planted to cash crops and total crop area for 78 developing countries, 1982

Reg i on/1ncome

Number of

Classification

Countries

Cash Crop Area

Total Crop

Percent

Area

Cash Crop

( thousand h e c t a r e s ) Asia Low i ncome Lower-middle i ncome Upper-middle

327,,786

*3,,12 E > a. o < o i o j

(0 >v- — - "O L. • Q. 4>fli4> Q. Q. L. CJ * x 3

om O CSJ ro

4> TJ 0) CL C (_ 4-> 3 4) < X

ro tr> ro kO

J- r-- O in cri

iX) in in .3-

in O (N NrtO «— ro r-j r-

rO V) V) w (0 v) V) C C 4) < i_i_Cl-C0(-4)4>C+J0" CC4>T>4)-DflJCC4)C.O oo w) d o o a: in c

i-

3 Changes in Land Tenure Patterns

The introduction of sugarcane production in Bukidnon apparently led to a significant deterioration in the access to land. This is unfortunate for at least two reasons. First, because access to land is such an important determinant of income in rural areas in a land-constrained, labor-surplus country such as the Philippines (and the survey data to be reported later confirms this), income distribution was skewed, and the plight of low-income groups worsened. Second, if a larger proportion of the higher incomes possible from sugar production had gone to lower-income groups instead, the linkages with other sectors of the rural economy would have been stronger, stimulating more local business and service activities, and so generating higher regional employment and economic growth (see Hazell and ROell 1983, Johnston and Kilby 1975, Mellor 1976, and Ranis and Stewart 1987). Several variables are treated as exogenous in the conceptual framework presented in Figure 2.1; most important are prices and the factor inputs available to the household for agricultural production, in particular land. Such assumptions are appropriate for analyzing household data collected over a period of one or two cropping cycles, and are a useful device for understanding the household decisionmaking process. In formulating policies, however, government decisionmakers are forced to consider how their actions will affect prices and the distribution of the factors of production, which are endogenous to their decisions, even in the short run, when a large number of households react quickly and in the same way to a particular policy stimulus. The effect of changing prices, including wages, from the introduction of export cropping (as has already been discussed) is primarily an issue at the national aggregate level and is beyond the scope of the research reported here (see Bouis 1989a for a simulation of an agricultural sector model). At the local level, Bukidnon has remained a net exporter of corn, the main staple food there, in a country where cereal markets operate efficiently (Bouis 1983). Total labor inputs per hectare per year for corn and sugar in the 19

20

AGRICULTURAL COMMERCIALIZATION

study area are nearly identical (as will be discussed in Chapter 6). Sugar production occupies less than 10 percent of the total cropped area in Bukidnon (one of 17 provinces in Mindanao), and labor mobility is high, as evidenced by high rates of migration. It is therefore unlikely that the introduction of sugar had much of an effect at all in the survey area on the price of com, or the wage paid to labor. This chapter will focus, then, on changes in access to land induced by the introduction of sugar.

1971 AND 1980 AGRICULTURAL CENSUSES The evidence that the expansion of sugar production has resulted to some extent from a consolidation of smaller operational units comes from three sources: two agricultural censuses conducted by the National Census and Statistics Office in 1971 and 1980 (National Economic and Development Authority 1974, 1985), the presurvey of a random sample of 2,039 households in the study area in 1984, and the four survey rounds. Table 3.1 shows the distribution of sugar and com farms, by number of farms, area harvested, and size of farm, for 1971 and 1980 for the whole of Bukidnon Province. In 1971 sugar production was negligible, but it had expanded to more than 9,000 hectares by 1980. Two-thirds of total sugar area was accounted for by farms larger than 25 hectares, which constituted only 12 percent of all sugar farms. By contrast, com is a smallholder crop. In 1971 nearly three-fourths of com farms were less than 5 hectares and accounted for 40 percent of all c o m area. Between 1971 and 1980, com area harvested increased by 51 percent and the number of com farms by 68 percent, implying (assuming no change in the cropping intensity) a modest reduction in average corn-farm size and a rapid expansion of population. By 1980 nearly 50 percent of com area was on farms of less than 5 hectares. Although these figures strongly suggest that smallholders participated only marginally in the sugar expansion, it is not clear to what extent the expansion resulted from a consolidation of smaller farms, if at all, or from the decision by large landowners to convert lands to sugar production that they already owned and were either cultivating themselves, leaving fallow, or renting out to other farmers. The census data show that there was apparently some expansion of total cropped area onto previously unused land during the 1970s. Unfortunately, the analysis that can be undertaken with the census is limited because data on operational farm size are not disaggregated by type of tenure; the census data refer to the entire province of Bukidnon, whereas our surveys focus on the southern half of the province; and the census data cover only the period up to 1980 before the expansion of the sugar mill's capacity. The presurvey of 1984 offers more precise evidence.

LAND TENURE PATTERNS

21

PRESURVEY OF 2,039 HOUSEHOLDS A 10- to 15-minute questionnaire was administered to 2,039 randomly selected households in April 1984, covering the main occupation of the head of household, secondary sources of income, crops produced, and basic demographic information. Table 3.2 presents the distribution of primary occupations of the heads of household recorded from this presurvey. Eighty percent of the respondents identified themselves as either landowners, tenants, or agricultural laborers. Seventy-nine percent of these respondents directly employed in agriculture were engaged in corn production, and only 7 percent were primarily employed in sugar production. Table 3.3 subdivides the sample into two groups: respondents from the three municipalities closest to the sugar mill and those from the remaining seven municipalities. As e x p e c t e d , s u g a r e m p l o y m e n t is f i v e t i m e s as p r e v a l e n t in the t h r e e municipalities closest to the mill. However, even in those municipalities, sugar production was the primary occupation of less than 20 percent of farming respondents, compared with 70 percent for com. More important, a comparison of the tenure patterns between corn, rice, and s u g a r p r o d u c t i o n in T a b l e 3.2 s h o w s that c o r n and rice l a b o r e r households account for a very small fraction of all households engaged in cereal production, with the number of landowners and tenants about equal. In sugar production, however, laborers account for half of the households, with a m u c h lower percentage frequency for tenants and a s o m e w h a t lower percentage frequency for landowners, as compared with corn. If the same distribution of c o m landowners, tenants, and laborers existed before the introduction of sugar as now, this implies that some former com landowners —and especially former com tenants—must have become sugar laborers. Table 3.4, which presents data for the previous occupations of heads of households presently engaged in sugar production, shows that this is indeed the case. For a majority of households, tenancy status does not change. However, 40 percent (21 out of 52) of households presently identified as sugar laborer households were com owner or com tenant households before the BUSCO mill was built. Another 40 percent (22 out of 52) inmigrated to the area after BUSCO began operations (typically sugar laborers from the islands of Negros and Panay, who were recruited by the sugar hacienda owners). Thus, only about 15 percent of present sugar laborers (8 out of 52) came from the preexisting pool of com laborers. Except for the inmigrants, almost all households that switched to sugar production had previously been involved in com production. It is especially important to note that the previous tenure distribution of those who switched from com to sugar production is very similar to the present tenure distribution of households primarily engaged in com production. This suggests that the situation of com households in areas further away from the mill is

22

AGRICULTURAL COMMERCIALIZATION

representative of the situation that households in the vicinity of the mill would be in if the sugar mill had never been built.

CONSOLIDATION OF EXISTING OPERATIONAL FARM UNITS It is evident from Table 3.1 that on average, sugarcane is grown on much larger farms than corn is. But was there a consolidation of smaller farms into larger farms? Might not the expansion of sugarcane have occurred simply through the adoption of sugarcane production by larger farms, which might have been growing any of several crops previously, including com? If consolidation was not an important phenomenon, Table 3.4 raises two questions: First, why then did so many households lose access to land ("displaced" households)? Second, how was so much additional labor from these displaced households absorbed by the larger sugar farms? The presence of displaced households could be explained simply by normal attrition. As will be seen in Chapter 4, corn productivity is so low that many com farmers, especially tenants, become discouraged by the low profits that their hard work produces, and thus they look for alternative ways to earn income. Also, some individual farmers are no doubt poor managers of the resources that are available to them, or poor harvests because of weather or pest infestation may force them off the land. The key question then becomes whether these "abandoned" farms are passed on more or less intact to other households (farm size remains constant), or whether they are absorbed into larger operational units as sugar is adopted. Data will be presented in Chapters 4, 5, and 6 to show that sugarcane and com production use almost identical amounts of labor on a per hectare per year basis.1 Thus, if the labor of displaced com farmers is absorbed by sugar farms (as indicated in Table 3.4) and farm consolidation did not occur, some previously idle land must have been available to the larger sugar farmers, which was put into use when sugar was adopted. Some idle land was available to the largest hacienderos for sugarcane production when BUSCO began operations. These idle lands, when converted to sugarcane production, provided some additional employment to local residents and to experienced sugar workers who were brought in from Negros. However, evidence shows that the survey respondents who earned wage incomes from sugar production worked mostly on small sugar farms. A second important issue, then, is the extent to which idle land was available to these medium-sized farmers for expansion of sugar area. To demonstrate more conclusively that there was some consolidation of existing operational farm units, it is necessary to turn to the survey data. Regardless of whether consolidation of existing farm units took place or not, however, it is already clear from the evidence presented that smallholders

LAND TENURE PATTERNS

23

participated only marginally in the income-augmenting opportunities made possible by introduction of the new export crop. The outcome might have been different if policies had been pursued to convert idle and already cultivated lands to sugarcane production on a smallholder basis.2

EVIDENCE FROM THE FOUR SURVEY ROUNDS In round 4 of the surveys, respondents were asked detailed questions about changes in their tenure status since the establishment of the BUSCO sugar mill. Table 3.5, which is conceptually identical to Table 3.4 (showing previous and present tenure status), is constructed from these responses. All 448 households are included in Table 3.5—not just those households presently engaged primarily in sugar production, as in Table 3.4. For purposes of comparison with Table 3.4, tenure categories used are those attributed by the households to themselves, not the categorizations used later in the report (see Table 2.2) that are based on reported landholdings and sources of income.3 Table 3.5 divides the respondents into four employment categories: 1) those engaged primarily in corn production; 2) those engaged primarily in sugar production; 3) those engaged primarily in agricultural production but who declined to specify a particular crop as dominant; 4) and those engaged primarily in nonagricultural employment.4 Comparing Tables 3.4 and 3.5, note that the percentages of present sugar laborers who specified a tenure status as a corn producer when BUSCO was established who lost access to land are nearly equal between the two tables (21 out of 52 in Table 3.4 and 8 out of 19 in Table 3.5). The percentages are also similar for present sugar landowners (2 out of 42 in Table 3.4 and 3 out of 28 in Table 3.5) and for present sugar renters (-2 out of 16 for Table 3.4 and 1 out of 9 for Table 3.5) who bettered their past tenure status. Table 3.5 shows that 56 out of 64 households that became involved in corn production after the establishment of BUSCO were able to acquire access to land, so the overall tenure pattern remained stable over time (many older residents improved their status from tenant to landowner, while newly married or newly resident couples embarked on com production in disproportionate numbers as tenants). For households engaged primarily in sugar production, however, only 4 out of 21 newly married or newly resident households got access to land, and none as landowners. A high percentage of households presently engaged primarily in nonagricultural employment who were married Bukidnon residents when BUSCO was established (14 out of 27) had lost access to land. Only one household moved in the other direction, from nonagricultural employment to corn production. The survey data, then, also present a contrast between

24

AGRICULTURAL COMMERCIALIZATION

sugarcane production and corn production: relatively easy access to land for corn production and a decline in access to land as sugarcane is adopted. From Table 3.5, it is possible to identify 38 households whose tenancy status improved, 34 households whose tenancy worsened, and 236 households (with previous access to land) whose tenancy status remained the same. Sixty-seven newly in-migrant or newly formed households gained access to land. The remaining 73 households (including both old residents and new households) have never acquired access to land at any point in time. In Table 3.6, these five groups (totaling 448 households) are classified by change in tenure status and are further subdivided into households that are presently engaged in sugar production (sugar households) and those that are not (com households). 5 Data are presented on area cultivated, area lost, size of sugar farm on which sugar wages were earned, per capita expenditures, and per capita income, by these ten groups of households. For households engaged in corn production, the sizes of farms being converted to and taken out of corn production appear to be in rough equilibrium. The average area cultivated by households whose tenancy status was unchanged at 2.5 hectares is almost equal to the average area lost of the 25 households that reported a decline in tenancy status at 2.4 hectares and to the average 2.8 hectares cultivated by the 26 households whose tenancy status improved. The size of the average corn farm appears, however, to have been decreasing over time, as the farms of new households at 1.6 hectares are disproportionately tenant households, which tend to be smaller than owner corn farms. The dynamics of change in farm size for sugar production appear to be quite different. All sugar households with access to land were new adopters of sugarcane production in the time between BUSCO's establishment and the 16 months covered by our surveys. The average area cultivated by adopting households was about 5.5 hectares—more than twice the area cultivated of the average corn household. The average area lost by households whose tenancy status declined and whose land was converted to sugar production was only a third that size at 1.8 hectares. Somewhat surprisingly, for com households in the five change-in-tenure groups, there are not wide disparities among the per capita expenditures or per capita income levels. One might have expected those whose tenancy status declined, for example, to have much lower expenditures and incomes than those who maintained access to land. Chapter 6 will show that com productivity is so low that on average, com share tenants would earn about the same income by renting out or selling what inputs they own (such as carabaos) and working in the agricultural labor force (assuming employment were available) as they would by engaging in com production. 6 The incomes of all com-tenancy groups are relatively low. The larger farm sizes and greater profits possible from sugar production (see Chapter 6) lead to much higher per capita incomes and expenditures for

LAND TENURE PATTERNS

25

sugar households. When evaluated using per capita expenditures, income, and other measures of welfare as bases for comparison, sugar laborer households are among the poorest households surveyed. Finally, in each month in which wages from sugar production were earned, respondents were asked to identify a farm size category on which this labor was performed, ranging from a score of 1 for farms larger than 50 hectares, to a score of 5 for farms smaller than 5 hectares (see the notes at the bottom of Table 3.6 for other category classifications). An average score of about 4 was reported by the respondents, indicating that most of the off-farm sugar labor provided by our respondent households was hired by farms of medium size relative to all farms, though small farms relative to those farms engaged primarily in sugar production. Did sugar e x p a n s i o n o c c u r on previously unused land? Certainly whatever expansion onto previously fallow land that occurred on the largest sugar haciendas contributed little to the employment of local residents whose tenancy status had declined. They tended to be hired instead by relatively small sugar farmers. The 80 sugar households in Table 3.6 with previous access to land and whose tenancy status remained the same reported that in 1977 they grew mostly com and that only about 10 percent of their land was left fallow, a percentage similar to that surveyed during 1984/85. Over the same period, these 80 households reported a 28 percent increase in the amount of land that they cultivated, an absolute increase of 1.3 hectares from 4.6 hectares in 1977. 7 The farms of smallholder sugar adopters (small relative to all sugar adopters, large relative to the average corn-producing farm) increased in size, as sugarcane production replaced corn production. T h e evidence f r o m households with access to land, then, supports a conclusion that there was some consolidation of existing operational farm units. Data collected from households that lost access to land supports a similar conclusion, although the number of observations is small. Of the 34 households that reported a decline in tenancy status, only 7 of 9 households whose land was converted to sugar production and 9 out of 25 households whose land was not converted to sugar production were able to indicate a farm size category for future use of the land that they had lost. 8 For land converted to sugar, 3 out of 7 (43 percent) responded that their land was absorbed by farms larger than 10 hectares. For land not converted to sugar, only 2 out of 9 (22 percent) reported that their land would be absorbed by farms larger than 10 hectares.

CONCLUSION As discussed in Chapter 1, land reform is central to the agricultural policy debate presently taking place in the Philippines. In the context studied here, a

26

AGRICULTURAL COMMERCIALIZATION

kind of land reform in reverse has taken place with the introduction o f export cropping. This raises at least two important questions. The first is the empirical question (which is beyond the scope of the present study) o f whether a similar deterioration in access to land has been occurring all over Mindanao, where various other export crops have been newly introduced. S e c o n d , what are the f o r c e s driving this redistribution o f land? To answer this second question, w e will need to compare the corn and sugar production technologies in s o m e detail, as undertaken in the following two chapters. The analysis that follows will show that these forces appear to be the declining productivity of corn lands, and the k n o w - h o w and financial resources o f the wealthier families w h o are in a position to take advantage of a n e w i n c o m e - e a r n i n g opportunity. G o v e r n m e n t p r o g r a m s n e e d to be introduced to make these opportunities available to smallholders as well.

NOTES 1. Informal surveys indicated that large-scale sugarcane production is more mechanized so that it may use less labor than smallholder production. Our surveys covered only landowner and renter farms that were smaller than 15 hectares. 2. It is difficult to judge the effcct of the Marcos administration land reform program on this process of land consolidation. The land reform law, which went into effect shortly after martial law was declared in 1972, specifically covered only land planted to rice and corn. The extent to which the government took active steps to implement the law in Bukidnon, and the expectation that the government might take future action, created an incentive for landowners to switch to the production of noncereal crops, such as sugar. There is evidence that the Marcos land reform program effectively redistributed land in rice-growing areas, especially in Luzon (Herdt 1987 and Ledesma 1982). There are no studies (of which we are aware) of the effectiveness of the land reform program in corn-growing areas, or in Mindanao specifically. Whether land consolidation would have happened anyway in the absence of this type of land reform (which exempted land planted to export crops) is an important question. Given the longer growing period of sugarcane and its greater resistance to short periods of unfavorable weather, the precise timing of planting, harvesting, and other steps in production is less crucial, so sugarcane production is more neutral to scale than com production, for which there are apparently diseconomies of scale in the Bukidnon context. The evidence presented in this report that sugarcane production is more profitable and more conducive to use of hired labor than corn production suggests that consolidation would have taken place anyway, whether or not there was a land reform law. A related question is whether land consolidation would have taken place because of the land reform law, even if sugar production had not been introduced. The importance of sampling households from areas near the sugar mill and far enough away that conversion to sugar was not economically feasible is underscored in answering this question. Both areas were subject to the same land reform law, yet

LAND TENURE PATTERNS

27

Table 3.5 indicates that com land tenure patterns and farm sizes were in some form of rough equilibrium over time, suggesting that the land reform program in a traditional corn-growing area did not result in land consolidation. Adoption of some other noncereal crop should also have resulted in consolidation in areas farther from the mill if the land reform law was a significant force driving consolidation. 3. The advantage of using the survey data over the presurvey data is that more detailed follow-up questions could be asked of households whose tenure status had either improved or deteriorated. A disadvantage of the survey data is that a stratified sample was taken. Sugar owner and sugar renter households are overrepresented relative to the population in southern Bukidnon; persons in nonagricultural occupations are underrepresented; and sugar laborer households are underrepresented within the aggregate group of all sugar households. Thus, the number of households appearing in Table 3.6 who improved their tenancy status is biased upward relative to the number of households whose tenancy status worsened. It should also be kept in mind that only households with at least one child less than six years of age were eligible for selection, which is an additional factor contributing to an unrepresentative sample. 4. There is a strong tendency for nearly landless households to classify themselves as corn tenants even if only a small plot of land is cultivated and agricultural labor is a more important source of income, so that fewer landless households appear in Table 3.5 than in Table 2.2. Also, a relatively low percentage of households appear as sugar households in Table 3.5 as compared with Table 2.2, apparently for two reasons. First, a household that produced any sugar was classified as a sugar household in Table 2.2, but a number of households chose (appropriately) to classify themselves as corn households when the percentage earned from sugar out of total income was low. Second, as will be seen in Chapter 6, income sources are quite diverse, so in many households one particular crop does not account for a dominant percentage of total income. This resulted in 80 households that declined to specify a particular crop and designated themselves merely as agricultural households. Other than these understandable discrepancies, there was a close correspondence between the respondent characterizations of their crop and tenure statuses and the classifications assigned to these households in Table 2.2. Of the 448 households surveyed, 343 reported that husband and wife were married and residing in Bukidnon in 1977 when BUSCO began operations. The remaining 105 households either migrated to Bukidnon after 1977, or husband and wife were married after 1977. Round 4 of the survey took place in August 1985. 5. For the "better," "same," and "new" groups, the criterion used to classify a household as "sugar" or "com" was whether or not any land was planted to sugar during the 16 months covered by our surveys. For the "worse" group (households that, with a few exceptions, are presently landless), the criterion used was whether or not the land to which they had lost access was converted to sugar production. For the "landless" group, a household was classified as "sugar" if wages from sugar crop production were greater than agricultural wages from all other crops, and all agricultural wages were greater than nonagricultural income (that is, if the household was classified as a sugar laborer household in Table 2.2). The landless com group in Table 3.6 includes both com laborer households and households engaged in other occupations, as defined in Table 2.2.

28

AGRICULTURAL COMMERCIALIZATION

6. An additional problem in evaluating per capita expenditures and incomes in this table is that household sizes vary across groups. Newly formed households have fewer children, so their incomes appear high on a per capita basis. 7. This may be compared with an absolute increase of 0.25 hectare for the 156 corn households who had previous access to land and whose tenancy status remained the same, a percentage increase of about 10 percent. 8. The remaining two households whose land was converted to sugar reported selling or giving up their lands to middlemen or relatives. Eight households whose land was not converted to sugar gave the same two responses.

LAND TENURE PATTERNS

—

o

29

o

ai 00 '«) « E 0 w w

1

l i f .

T3 l/l i/l

g > g 0.

fi

1 o • C

W

X C (. ID 0)

4> O

C. C I D 4) N O -U 1 •X) RSJ ®

•

(T> C


C
C 4>

3 •$>- C — 3 C

c

c

4) 4) 4>

-U 3 -C «- — O 4-> 4> • IB .C W 3 — J C R I > X> — 4) 4) 3 C C .O *->

4> to

"D

C

I I E U 0) c -»-) 4> O

> 4> IB -O O >!•— U E - - CHL O - I - " - O 4) O - C W) " P OH) 3 V 4) : I FL) « (0 *J I O *•> 3 IB ¡0 4"J '

J

c *->

¡5

. c H> E 3

s a a. ® "8 c O £ C 0) « CD D) a) C *c 3 D CO T> a> TJ D> CO

•»-> A I (

o c

O CR>

Q. CO o k CO D iZ O Z3 O (0 •c S O) < CO

o to

C " 0 —3 4) 4> W ^ L

ra « 2

O 4J "D 4)

4> 13

>. a

>B E < f l * > E . 5 v> . c _) v> O

f> I. D l_ "O O C Q < o _J

X> C 10

ID X L. 0) JZ *->

o

. o in in to ct** _

o

O fN O -3"o O o o o o o o oo o o o

o o Lfl o o Oo

Oo Oo

o

""

""

'—

'—

CT)

C N 7Z

4-» U 4>

fflfO-t i-^ ^

r- o o r-. r-- r- r-o r- O o o o o

co CTI CD to rsi rl (SI o O O

o

in rs.

m

^

m CD co 01 4>>>> D

v>

•

E «

•o c« ci o o> >CO c

S

C 4) •"< - J4T-> — • O V u JÉ 0) 5 "" CTI C "O "O >~i c 4> X) (0 O • «r- l_ W 0) 5 * o CO in ro r-- kd r-. in O » -S" < —U" (SJ C vJ CO tO Csi m ko ro CO -3" J" i/1 •d- •d- ro•d- -d- d-

J- -à- -i-

cn Is-. O r»» to rO rO j-j- ro -i-

J- Jo o O (SI

LDJ- CSI (T>CO 01 rO rrO COo r-. i/l ro j- vOLTl

in rO

ro Cvl CM LTI in J-

COJ" rO

ro

01 ro

CO

ro O

in # o O

in o

f*-» CO CSI COO -J O