UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Rural out-migration and rural development in Iran : implications for the roles of infrastructure in case.. Sarrafi, Mozaffar 1995-12-31

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


ubc_1995-983450.pdf [ 7.72MB ]
JSON: 1.0088910.json
JSON-LD: 1.0088910+ld.json
RDF/XML (Pretty): 1.0088910.xml
RDF/JSON: 1.0088910+rdf.json
Turtle: 1.0088910+rdf-turtle.txt
N-Triples: 1.0088910+rdf-ntriples.txt
Original Record: 1.0088910 +original-record.json
Full Text

Full Text

RURAL OUT-MIGRATION AND RURAL DEVELOPMENT IN IRAN: IMPLICATIONS FOR THE ROLES OF INFRASTRUCTURE IN CASE OF HAMADAN PROVINCE  By: MOZAFFAR SARRAFI MArch. Tehran University, 1978 M.Arch. University of California, Los Angeles, 1982  A THESIS SUBMITTED N PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES  (School of Community and Regional Planning)  We accept this thesis as conforming to the required s ndard  THE UNIVERSITY OF BRITISH COLUMBIA  February 1995 © Mozaffar Sarrafi, 1995  In presenting this thesis in  partial fulfilment of the requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or  by  his  or  her  representatives.  It  is  understood  that  copying  or  publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Deptrñt  *  of  e6v4(  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  MI1VtL1  (5, I  ‘iøij  In the Name of the Creator  ABSTRACT Large scale rural out-migration has gained momentum over the past four decades in Iran, contributing to urbanization at unprecedented rates. In the wake of the Islamic Revolution, it was recognized that in order to reduce reliance on oil revenues and foster self-sufficiency and social equity, it was essential to ensure the viability of agriculture and rural settlements. As a part of this new strategy, a rural infrastructure provision policy (RIPP) was undertaken in order to bring about rural prosperity and to curb out-migration. Yet, the plight of villagers and out-migration persist. This dissertation focuses on the village end of the problem, and on permanent outmigration in post-revolutionary Iran. It investigates the causes of rural out-migration and their impacts on the remaining rural households. Further, it examines the potential of RJPP to reduce out-migration and enhance village viability. In terms of methodology, a cross-analysis was conducted at the levels of individual, household, and community. Both qualitative and quantitative research methods were employed. Data were collected from primary and secondary sources. While the latter served analysis needs at the macro-level, the former, which included case studies in five villages in Hamadan Province, served those at the micro- and meso-levels. The macro-level analysis reveals population pressure on agricultural resources and ruralurban disparities as the overriding causes of rural out-migration in Iran. Correspondingly, the micro- and meso-level analyses: (a) highlight the critical importance of the middle strata (MS) for the fhture viability of rural Iran; (b) identify household insecurity, resulting from precarious and uncertain rural livelihoods as the root cause of out-migration for MS; and (c) suggest that the ongoing migration of youth from MS must be contained to ensure the next generation of farmers. Finally, five roles are identified for RIPP to target the overriding causes as well as those pertaining specifically to MS. While there is need for policy changes in the macro-economic sphere in Iran, RIPP has the potential to reduce rural out-migration. More fundamentally, it suggests that it is not merely the presence of physical infrastructure and its direct role, but rather an effectively functioning social infrastructure and its intermediary roles that are vital to curbing excessive out-migration and ensuring village viability.  II  TABLE OF CONTENTS  ABSTRACT TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES NOTE ON THE IRANIAN CALENDAR YEAR GLOSSARY AND ABBREVIATIONS ACKNOWLEDGMENTS DEDICATION  ii iii viii xi xi xii xv xvi  CHAPTER I: INTRODUCTION 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9  THE QUANDARY IN THE FIELD OF iNQUIRY STATEMENT OF THE PROBLEM IN IRAN A PRINCIPLE, A PREMISE AND AN ETHICAL ISSUE RESEARCH RATIONALE RESEARCH OBJECTIVES RESEARCH QUESTIONS RESEARCH SCOPE RESEARCH STRATEGY ORGANIZATION OF THE DISSERTATION  2 4 7 8 9 10 11 12 15  CHAPTER II: DEVELOPM]NT AND RURAL-URBAN MIGRATION 2.0 INTRODUCTION 2.1 RURAL-URBAN MIGRATION AND DEVELOPMENT THEORIES 2.1.1 Modernization Theory 2.1.2 Dependency Theory 2.1.3 Sustainable Development Social Equity o Ecological Balance o Implications for Migration in the South 2.2 ASSESSMENT OF RURAL-URBAN MIGRATION 2.2.1 Trends of Population Distribution in the South 2.2.2 A Conceptual Framework for the Assessment 2.2.3 Rural-Urban Migration in the South: A Preliminary Assessment  . .  17 ..17 18 20 22 22 23 25 27 28 30 32  in  CHAPTER ifi: RURAL-URBAN MIGRATION AND NATIONAL POLICIES IN IRAN 3.0 INTRODUCTION 3.1 RURAL-URBAN MIGRATION TN IRAN 3.1.1 Rural and Urban Distribution of Population 3.1.2 Rural-Urban Migration 3.1.3 Settlement Pattern 3.2 STATE INTERVENTION TN POPULATION DIST1UBUTION 3.2.1 The Need of Policy Intervention 3.2.2 Policy Classification 3.2.3 Significance of the Government’s Role in Iran 3.3 A BRiEF HISTORY OF RURAL-URBAN MIGRATION TN IRAN 3.3.1 Stage One: Modernization Approach 3.3.2 Stage Two: Oil-Driven Urbanization 3.3.3 Stage Three: Inconsistency and Turbulent Movement 3.4 POST-REVOLUTIONARY MIGRATION POLICIES 3.4.1 Restrictive Policies in Urban Centers 3.4.2 Retentive Policies in Rural Areas 3.5 AS SES SMENT OF RURAL-URBAN MIGRATION TN IRAN  37 37 39 40 43 45 47 48 52 52 52 54 57 61 62 64 67  CHAPTER IV: MIGRATION THEORIES AND THE ROLES OF INFRASTRUCTURE 4.0 INTRODUCTION 4.1 THE MACRO-LEVEL MIGRATION THEORIES 4.1.1 The First Branch 4.1.2 The SecondBranch 4.1.3 The Third Branch 4.1.4 Summary 4.2 THE MICRO-LEVEL MIGRATION THEORIES 4.2.1 The First Branch 4.2.2 The Second Branch 4.2.3 Summary 4.3 THE MESO-LEVEL MIGRATION THEORIES 4.3.1 The First Branch 4.3.2 The Second Branch 4.3.3 Summary 4.4 THE IRANIAN LITERATURE IMPLICATIONS 4.5 IMPLICATIONS FOR THE ROLES OF 11’IFRASTRUCTURE IN RURAL OUT-MIGRATION 4.5.1 A Conceptual Framework for the Roles of Infrastructure  72 74 75 78 79 81 82 82 87 89 90 91 94 97 99 101 103  iv  CHAPTER V: RESEARCH METHODOLOGY 5.0 INTRODUCTION 5.1 APPROACHTOMIGRATION STUDY 5.1.1 On Searching for Cause of Migration 5.1.2 On Searching for the Infrastructural Cause 5.2 LEVEL OF ANALYSIS 5.3 RESEAR.CH DESIGN 5.3.1 Units of Analysis 5.3.2 Variables and Measurement 5.3.3 Data Collection 5.3.4 Research Strategy 5.3.5 Cases Selection and Sampling Method 5.4 CONCLUSION  108 108 109 112 113 113 114 114 115 119 119 127  CHAPTER VI: SURVEY FINDINGS 6.0 INTRODUCTION 6.1 THE CHARACTERISTICS AJ’1]) BEHAVIOR OF RURAL OUTMIGRANTS 6.1.1 Sex, Age, and Marital Status 6.1.2 Education 6.1.3 Destination and Presence of Relatives 6.1.4 Occupational Change 6.1.5 Collaboration and Contacts with the Original Household 6.1.6 Summary 6.2 CHARACTERISTICS OF THE RURAL MIGRANTS’ HOUSEHOLDS 6.2.1 Size and Composition 6.2.2 Literacy 6.2.3 Occupation of Head of Households 6.2.4 Land and Livestock Ownership 6.2.5 Sources of Income 6.2.6 Remittances and Spending Patterns 6.2.7 Migration Intention 6.2.8 Housing and Economic Status 6.2.9 Summary 6.3 STATED CAUSES OF MIGRATION 6.3.1 Stated/Perceived Causes for Out-Migrants 6.3.2 Stated Causes for Would-be Individual and Household Migrants 6.4 CONCLUSION  129 129 130 131 133 136 139 140 141 141 143 144 145 146 146 147 148 150 153 153 156 158  V  CHAPTER VIE: GROUP INTERVIEWS’ FINDINGS 7.0 7.1 7.2 7.3 7.4  INTRODUCTION ON THE VILLAGE OUT-MIGRANTS ON THE CAUSES OF OUT-MIGRATION ON THE CONSEQUENCES OF OUT-MIGRATION THE KEY iNFORMANTS’ APPROACH TO THE FUTURE OUTMIGRATION 7.5 ON THE RURAL DEVELOPMENT AN]) INFRASTRUCTURAL SERVICES 7.6 IMPLICATIONS FOR INFRASTRUCTURE’S ROLES 7.7 CONCLUSION  162 163 166 171 173 175 179 181  CHAPTER Vifi: CONCLUSION AND POLICY IMPLICATIONS 8.0 INTRODUCTION 8.1 A SUSTAINED MOVEMENT TOWARD UNSUSTAiNABLE DEVELOPMENT 8.2 THE CAUSES AND CONSEQUENCES OF RURAL OUTMIGRATION 8.2.1 MACRO-LEVEL 8.2.2 MICRO- AND MESO-LEVEL 8.2.3 THE MIGRATION INTENTION OF RURAL DWELLERS 8.2.4 SYNTHESIS 8.3 IMPLICATIONS FOR THE ROLES OF INFRASTRUCTURES IN THE RURAL AREAS 8.3.1 SERVICE DELWERY FACILITATION 8.3.2 EMPLOYMENT GENERATION 8.3.3 INCOME UPGRADING 8.3.4 RURAL-URBAN INTEGRATION 8.3.5 RURAL INSTITUTIONALIZATION 8.3.6 SYNTHESIS 8.4 EPILOGUE: RURAL DEVELOPMENT, OUT-MIGRATION, AND INFRASTRUCTURE  BIBLIOGRAPHY  185 186 190 190 193 196 199 201 202 204 205 207 210 213 215  219  vi  APPENDICES APPENDIX (A): APPENDIX (B): APPENDIX (C): APPENDIX (D):  Household Questionnaire .257 Group Interview Questions and List of Participants 260 Additional Survey Tables 264 Summary of Livestock Ownership for the Sample Migrants’ Households 277 APPENDIX (E): Abridged Record of Group Interviews 279 APPENDIX (F): Data on the Five Sample Villages 302 APPENDIX (G): Out-Migration & Infrastructure Availability in the Dehestans of Hamadan Province 305 APPENDIX (H): Labour Force and Employment in the Urban and Rural Areas of Iran in the Census Years 312 APPENDIX (I): Population Pressure on Rural Resources and Rural-Urban Disparities in Iran 317  POSTSCRIPT  321  vii  LIST OF TABLES Table 3.1: Distribution of Iranian Residing in Places Other Than Their Birthplace in Urban and Rural Areas 38 Table 3.2: Iran’s Urban & Rural Population Changes during 1956-199 1 39 Table 3.3 Estimation of Iran’s Rural-Urban Migration during 1956-199 1 41 Table 3.4: Changes in the Size Distribution ofUrban Centers in Iran: 1956-1986 44 Table 3.5 Changes in the Size Distribution ofRural Settlements in Iran: 1956-198.6 44 Table 3.6: Classification of Migration Policies & Their Emphases 50 Implications for the Roles of Infrastructure in Rural Out-Migration Table 4.1 according to the Macro-Level Theories 105 Table 4.2: Implications for the Roles of Infrastructure in Rural Out-Migration according to the Micro-Level Theories 105 Table 4.3 Implications for the Roles of Infrastructure in Rural Out-Migration according to the Meso-Level Theories 105 The Sample Villages (Population, Location and No. of Questionnaires).. 123 Table 5.1 Table 6.1 Relationship of Migrants to Household Heads 129 Table 6.2: Distribution of Rural Out-Migrants by Sex 130 Table 6.3 Distribution of Rural Out-Migrants by Age Groups 131 Table 6.4: Marital Status of Out-Migrants at Present and at Departure Time 131 Table 6.5 Distribution of Out-Migrants by No. of School Years Attained 132 Table 6.6: Distribution of Out-Migrants Sex by Literacy Status 133 Table 6.7: Distribution of Migrants Education by Age Groups and Sex 133 Table 6.8: Distribution of Out-Migrants’ First and Final Destinations 135 Table 6.9: Distribution of Migrants Who Have Had Relatives at Destination by Sex 136 Table 6.10: Distribution of Migrants’ Occupations Prior to and After Migration 137 Table 6.11: Distribution of Migrants’ Sex by Their Assistance in Agricultural Work .139 Table 6.12: Distribution of Migrants’ Visits to Their Origin by Marital Status 140 Table 6.13: Distribution of Rural Households by the No. of Out-Migrant Members .142 Table 6.14: Distribution of Rural Migrants Households by Size 142 Table 6.15: Literacy among Members of Rural Migrants’ Households 143 Table 6.16: Distribution of Rural Migrants’ Household Heads by Education 143 Table 6.17: Distribution of Rural Migrants’ Household Heads by Occupation 144 Table 6.18: Average Land Ownership among the Sample Households by Agricultural Type 146 Table 6.19: Distribution of Sample Households by Recipient/Non-Recipient of Remittance 147 Table 6.20: Rural Migrants’ Household and Respondents’ Migration Intention 148 Table 6.21: Distribution of Rural Out-Migrants’ Causes of Migration 154 Table 6.22: Distribution of Rural Respondents’ Causes of Migration Intention 156 Table 6.23: Distribution of Sample Households’ Causes of Migration Intention 157 Table 8.1: Typical Social Strata and Out-Migration Status in Rural Hamadan 198 Table 8.2: The effectiveness of Infrastructure’s Roles on Out-Migration Causes 214 Table C. I: Distribution of Sample Households by the Number of Male Child Migrants 265  viii  Table C.2: Distribution of Sample Households by the Number of Female Child Migrants 265 Table C.3: Distribution of Sample Households by the Number of Female Members .266 Table C.4: Distribution of Sample Households by the Number of Male Members 266 Table C.5: Distribution of Sample Households by the Number of Aged 6 Years and Over Members 267 Table C.6: Distribution of Sample Households by the Number of Literate 267 (Aged 6 Years and Over) Members Table C.7: Distribution of Household Heads’ Occupations by Education 268 Table C.8: Distribution of Sample Households by Their Amount of Rain-Fed Farmland Ownership 268 Table C.9: Distribution of Sample Households by Their Amount of Irrigated Farmland Ownership 269 Table C. 10: Distribution of Sample Households by Their Amount of Orchard Ownership 269 Table C. 11: Distribution of Sample Respondents’ Migration Intention by Their Relationship to Household Heads 270 Table C. 12: Distribution of Sample Households’ Migration Intention by the Occupation of Household Heads 270 Table C. 13: Distribution of Sample Households by Their Available Rooms 270 Table C. 14: Ownership of Certain House-Ware/Transportation Items by Sample Households 271 Table C. 15: Distribution of Sample Households’ by Their Main Sources of Income ....271 Table C. 16: Distribution of Sample Households by Remittances’ Spending Categories 272 Table C. 17: Distribution of Male Rural Out-Migrants’ Causes of Migration by Their Previous Occupation 272 Table C. 18: Distribution of Female Rural Out-Migrants’ Causes of Migration by Their Previous Occupation 272 Table C. 19: Distribution of Rural Out-Migrants’ First Causes of Migration by Their Education at the Time ofDeparture 273 Table C.20: Distribution ofRural Out-Migrants’ Second and Third Causes of Migration by Their Education at the Time of Departure 273 Table C.21: Distribution of Rural Out-Migrants’ First Causes of Migration by Their Relationship to Household Heads 274 Table C.22: Distribution of Rural Out-Migrants’ Second and Third Causes of Migration by Their Relationship to Household Heads 274 Table C.23: Distribution of Rural Respondents’ First and Second Causes of Migration Intention by Their Relationship to Household Heads 274 Table C.24: Distribution of Rural Respondents’ First and Second Causes of Migration Intention by Their Household Heads’ Occupation 275 Table C.25: Distribution of Sample Households’ First and Second Causes of Migration Intention by Their Household Heads’ Occupation 275 Table C.26: Distribution of Rural Respondents’ First and Second Causes of Migration Intention by Their Household Heads’ Education 276 ..  ix  Table C.27: Distribution of Sample Households’ First and Second Causes of 276 Migration Intention by Their Household Heads’ Education 303 Table F. 1: General Data on the Five Sample Villages Table F.2: Availability/Non-Availability of Services and Institutions in the 304 Sample Villages 1992 311 Table G. 1: Correlation Coefficients Among Dehestan-Level Variables Table H. 1: Distribution ofEmployed Population of 10 Years of Age and Over in the Urban and Rural Areas of Iran by Economic Sectors, 1966 & 1976 ..3 15 Table H.2: Distribution ofEmployed Population of 10 Years of Age and Over in the Urban and Rural Areas of Iran by Economic Sectors, 1976 & 1986 ..3 15 Table H. 3: Urban and Rural Labour Supply and Employment in Iran, 1966 & 1976 316 Table H.4: Urban and Rural Labour Supply and Employment in Iran, 1976 & 1986 316  x  LIST OF FIGURES Figure 1.1: The Location of Hamadan Province in Iran Figure 2.1: Rural-Urban Migration Assessment Model Figure 3.1: Internal Forces of Uneven Development and Resultant Population Mobility Figure 4.1: Multi-Level Contextual Factors in Migration Decision-Making Figure 5.1: Diagram of the Research Major Variables Figure 5.2: Hamadan Province: Distribution of Human Settlements, Major Roads, and Highlands Figure 5.3: The Locations of Sample Villages in Hamadan Province Figure 8.1: Mechanism of Rural Labour Migration in Iran (Macro-Level)  14 31 46 103 115 122 124 192  NOTE ON TIlE IRAKEAN CALENDAR YEAR: The Iranian solar year (Sal-e-Shamsi), which starts on March 21 (the Spring equinox), has been converted to the Christian year by adding 621 years to the solar year. For example, the census years of 1355 and 1365 are converted to 1976 and 1986, ignoring the January 1 to March 20 difference. In other words, the precise conversion would be March 21, 1986 to March 20, 1987.  xi  GLOSSARY AND ABBREVIATIONS Beh Sazi-ye Roosta  A national program for rural renovation with special emphasis on infrastructure provision, embarked on circa 1985 in Iran.  Boneh-kan  Refers to the wholly out-migrated rural household  Dehestan  The next highest level administrative/jurisdictional boundary after the rural settlement itse1f comprised of farms, hamlets, and villages, but not towns.  Household  “A group of persons who wish to live together and therefore share living quarters and their principal meals.” (Goodall 1987, 215) Some of the members may be temporarily away. It is not necessarily equal to family and can be a single person or multi-family.  Homemaker  A person whose principal occupation is managing a household and taking care of domestic affairs. The same as housekeeper, but is not a hired person.  Infrastructure  Installation and facilities that provide a fundamental framework for a community and which, therefore, facilitate all forms of development such as: agricultural, industrial, and It includes the provision of transport, educational. telecommunications, power supplies, water and sanitation, waste disposal, educational and health care facilities, and other public utilities. These are considered as social over belonging to the community as a whole head capital (Bannock et al. 1987, 206; World Bank 1994, 2). ,  Institutionalization  “A process, as well as the outcome of the process, in which social activities become regularized and routinized as stable, social-structural features” (Jary & Jary 1991, 239).  Key Informant  A community person who is most likely to be well-informed about the topic of interest, e.g., village councilor, elderly person, local teacher (Bilsborrow 1984, 428).  Khaneh Behdasht  The lowest order of health care centers in the health care network for the Iranian rural areas, which has at least one health care worker/nurse, being visited once a week or more by a medical doctor.  xii  Khod-Yari  Voluntary contribution of community residents, often exclusively for funding and physical labour, for the construction of public facilities (e.g., public bath, school, telecommunication center) by the government sector.  Khosh-neshin  Same as Aftab-neshi, a landless villager “without rights to cultivation or permanent employment on the land” (McLachlan 1988, xx).  Markaz-e-Shahrestan  An urban center which is the administrative center of the township jurisdiction (i.e., Shahrestan).  Markaz Khaclamat e-Roosta’ei  offers agricultural service center which Rural extension services and some subsidized inputs (e.g., seed, fertilizer, pesticide, tractor tires) by the Ministry of Agriculture. A nation-wide network has been established since the Islamic Revolution in Iran.  Migration  A permanent population displacement or change of residence to a new population center (Lee 1966, 49). Thus, other kinds of population mobility such as: circulation, commuting, and seasonal movements, are not addressed in this dissertation.  RCC  The Rural Construction Crusade, Jihad-e-Sazandegi, established after the Islamic Revolution in Iran in 1979.  Rial  The official exchange rate is The Iranian currency. 1750 rials to $1 U.S. (as of July 1994).  RIPP  Rural infrastructure provision policy.  Remaining Household  Always refers to the migrant’s household of origin (i.e., village) and excludes the migrant. Same as rural migrant household or originating household.  Rural Area  Area classified in each country as rural according to its own jurisdictional definitions. In Iran, any human settlement which is not classified as an urban area, is considered to be a rural area.  Rural Development  Rural development is defined as a sustainable process which leads to a continuous rise in the capacity of rural community to attaine its own values, to control reaources, and to  xlii  conserve environment, accompanied by a wider distribution of benefits resulting from such a process. (Weber & Abeyramal984, 70; Misra 1981, 116) Rural Migrant Household  Always refers to the migrant’s household (prior to leaving the village) which has at least one member still living in the village. Also, the terms remaining and originating household may be used interchangably.  Rural  Out-Migrant  A villager who had out-migrated at the time of interview with the intention to stay elsewhere permanently. Tantamount to rural emigrant.  Rural Out-Migration  Permanent movement of individuals or households from a rural area to another rural or to urban areas in the country. Equal to emigration. Also, see the definition of migration.  SCI  Statistical Center of Iran, affiliated with Plan and Budget Organization.  Sevener Conmiission  Refers to seven-member committees of land devolution, established after the Islamic Revolution of Iran to carry out a limited land reform, mostly for confiscated lands of escaped big landlords (Schirazi 1993, 161; Lahsaeizadeh 1993, 257)  Stayer  A rural person or household member who has remained at the village.  Sustainable Development  “Improving the quality of human life while living within the carrying capacity of supporting ecosystems.” (IIJCN/ UNEPI WVTF 1991, 211) A development process which could be sustained through the promotion of social equity, economic viability, ecological balance, and all together, their integration in the ecosphere.  The North  Same as: developed countries or industrialized countries.  The South  Same as: Third World or developing countries.  Urban Area  Area classified in each country as urban according to its own jurisdictional definitions. In Iran, any population center with a municipality approved by the Government is considered to be an rural area. (SCI 1991, 32)  xiv  Acknowledgments In writing this dissertation, numerous individuals and organizations have assisted me. To begin with, I wish to express my deepest gratitude to my research supervisor, professor Setty Pendakur for his understanding character, compliant support, and the intellectual guidance he constantly provided me in the course of my research. I am also much indebted to the other members of my research committee, professor Peter Boothroyd and professor Terry McGee for their invaluable advice, comments, and encouragement during three years of study. In general, I certainly benefited from the educational environment and facilities available in UBC; especially, my thanks are due to the faculty and the staff of the School of Community and Regional Planning, Center for Human Settlements, and the UBC libraries, who responsibly responded to my research needs. My research could not have been completed without generous scholarship provided by the Iranian Ministry of Culture and Higher Education, during the four years, residence in Vancouver. In addition, I owe much to Canadian Universities Consortium for the grant which partially allowed me to undertake the field work in Hamadan Province. The arduous task of carrying out the field survey in the remote rural areas was eased by the kind support of many persons. The hospitality of five sample village households during the field work, their time, devotion, and enthusiasm, was a great moral energy, without it this research would never have been finished. I humbly feel they are the only authority to appraise my research in the future. I am particularly grateful to Mr. Majidi, director of Plan and Budget Organization of Hamadan Province for his goodhearted and full support. My special thanks are due to: Mr. Shobeiri, Mr. Farsh-chian, Mr. Farzaneh kari, Mr. Shokouh-e-Zanganeh and Mr. Firoozfar and his family who sincerely facilitated my research needs in the Province. The diligent assistance of Mr. Alishahi, Mr. Salehian, Mr. Mohebi, and Mr. Say yadpoor, was a decisive help during the administration of my survey in the villages. Warm appreciation is expressed for the support of all my colleagues in the Regional Planning Bureau of Plan and Budget Organization at Tehran, especially Mr. Karimi, Mr. Haghighi, Mr. Baroumand, and Mr. Kabiri. A special word of appreciation should go to Mr. Sabetghadam for his whole-hearted and invaluable consultation on the SPSS program and to Ms. Nazari and Ms. Alaindari for their kind assistance. Furthermore, thanks are due to many colleagues in IJBC, including: Priscilla Boucher, John Curry, Sudharto Hadi, linn Teetzel, and Mathis Wackernagel (with whom I started the curriculum), and Sam Afrane, Khalid Alskait, Madhav Badami, Michael Carr, Stan DeMello, Yoshihiko Wads, Pongsak Vadhanasindhu whom I enjoyed for their encouragement and insights. I extend my heartfelt gratitude to them for their goodwill, companionship, and collegiate support during my stay in Vancouver. Also, I would like to remember my late friend, Joseph Otto, a Ph.D. fellow from Ghana who willingly helped me during the first month of my arrival. I am also indebted to some Iranian families in UBC housing, particularly to the Ghofraniha family, whose compassionate friendship was a great help for me and my family. Special thanks go to Heather Ross whose editing assistance helped me to clarify my arguments. And last, but certainly not the least, my greatest debt in writing this dissertation goes to my family. My wife, Soheila, her insights, encouragement, and support, allowed me to realize this work. I cannot thank her adequately who suffered being away from her professional work and being engaged in homemaking all these years. Also, I would like to remember my late aunt, from whose compassionate support I benefited and who passed away while I was studying. My deepest gratitude goes to my children and especially my family in Iran: my mother, my sister and my niece, who painstakingly tolerated my negligence and absence. To all my family, I am forever indebted and grateful. I have dedicated this dissertation to the memory of my father and to my mother who raised me with great sacrifice and wisdom and taught me the best lessons of altruism, though I might not be a good disciple. There are no words that I can find to express my respect, love and indebtedness toward them.  xv  This dissertation is dedicated to the memory of my father: Mahmood Sarrafi, to my mother: Fatemeh Zand, and through them, to all the altruistic people of Iran.  CHAPTER I: INTRODUCTION  1  1.1 The Quandary in the Field of Inquiry  There is general agreement among researchers that the population distribution trends in the South’ are not desirable. Urban population in general, and the population of primate cities in particular, is increasing at a very high rate. 2  Acute urban problems such as  unemployment, overloaded infrastructure, environmental pollution, squatter settlements, social alienation, etc., have spread in most metropolitan areas (Qadeer 1983, Ch. 1). Ninety-four South countries have made “reversing or at least slowing the tide of ruralurban migration one of the top priorities of their population policies” (ICSC 1994, 5). While rural-urban migration is not typically the most important component of urban population growth, 3 its secondary effect becomes more significant should we take into account that a substantial proportion of natural increase is made up of migrants’ offspring.  However, there is little prospect that the rural areas of the South will retain their high rates of natural population increase in years to come. Environmental degradation, widespread poverty and rapid population growth (as three intersected global threats) are rampant in most rural areas (Sadik 1991, 16). Parallel to these, in wake of the growing prevalence of market-friendly policies (World Bank 1991, 1), the so-called structural adjustments (ADP 1992, 274) and market globalization in the South, it seems less attention will be paid to  job creation and social welfare in rural areas (IFAD 1992, 270-271; Afshar 1994, 271).  1  Throughout this study, the tenn the South or South Countries will be used interchangeably with: Third World, less developed or developing countries. Also, the tenn North has been used interchangeably with: developed countries or industrialized countries. 2  The average annual growth rate of the urban population in 1985-1990 was 5.5 times the rural one in the South (refer to: UN 1991, 154 & 160). Population growth rate for most mega-cities (with a population 8 million or more according to the UN defmition) in the South was more than 3 percent during the period 1980-1990 while the corresponding figure in the North was less than 1 percent (UN 1991, 30). The causes of urban population growth are, in typical order of importance: natural growth, rural-urban migration, and reclassification of rural centers or their assimilation to adjacent urban centers. The ruralurban migration component usually constitutes over one third of the increase.  2  Thus, rural-urban migration will persist, at least until the promise of trickle down, if possible at all, comes true.  At the same time, skepticism regarding the ongoing development at the world scale is growing (MacNeil 1991; Brown et al. 1992). More than at any time in recent decades, the quest for another concept of development [under the rubric of sustainable development] (WCED 1987) has occupied intellectuals and, by and large, the grass roots. 4 Sustainable development, though, is not a new concept and a large spectrum of previous concepts countervailing to the mainstream development, such as equity, self-reliance, and ecological balance, are gathered under its umbrella, gaining momentum and international attention. In general, urban growth in the South is currently unsustainable; the solution to this problem calling for, among other things, the conservation of rural areas and agricultural production without degrading the natural base.  As such, the population  distribution trend in the South is unfavorable. 5  Many scholars attribute the rural-urban migration 6 in the South to the interaction between the “pull” forces of cities and the “push” forces of villages conditioned by the context. For many years, the governments of the South have attempted to change this trend, with insignificant success.  Despite governments’ rhetoric, in practice, most of the remedies  have focused on the urban side of the problem, neglecting the rural side (UNCHS 1987, 17). Ironically, alleviating the negative effects on the urban side often intensifies rural outAs an indication, the social malaise in many South countries since 1992, from Mexico and Brazil, to Algeria and Rwanda, and to South Korea and Thailand portend the popular aspiration for an alternative ‘  development paradigm. In many cases where migrant labour in rural areas is substituted with energy-intensive technology or there is an abandonment of land, the negative results in rural areas are as important as in urban areas (see section 2.2.3). 6  Migration is defined as a pennanent population displacement or change of residence to a new population center (Lee 1966, 49). Thus, Other kinds of population movements such as: circulation, commuting, and seasonal movements are not addressed in this dissertation.  3  migration (Savasdisara 1983; Danisword 1984).  And the city has continued to be  perceived by the majority as being a better living place than is the village (e.g., in terms of quality of life, income, social mobility, and job opportunities), a notion nurtured by objective reality in conjunction with subjective presumptions.  Thereby, if governments  have been successful in tackling emergent urban predicaments, the glaring rural-urban gap has also widened, drawing more migrants to cities.  All together, the quandary about the growing urban problems and the continuity of rural-  urban migration in the future on one hand, and the necessity of rural revitalization and environmental conservation on the other hand, rationalize the inquiry into how to affect population distribution trends, specifically rural out-migration, towards a sustainable development in the South.  1.2 Statement of The Problem in Iran In a short period of time, during almost three decades, Iran has undergone a drastic change from being an agricultural and rural-dominated society to being a non-agricultural and urban society. While in 1956, 56.7 percent of the employed labour force was in the agricultural sector, and 68.6 percent of Iran’s population was living in rural areas, in 1986 these ratios were 29.1 percent and 45.7 percent respectively. 7  Considering the  quantitative aspect of this transition, it can be said that an urban explosion has taken place in the last three decades.  For the next three decades, the projection of Iran’s urban  population indicates prolonged immense growth.  It is projected that with a moderate  population control policy, the urban population will triple from approximately 32 million in 1991 to 99 million in 2021 (Zandjani 1991, 39). To illustrate the previous situation, a scholar estimates that in 1900 and also in 1926 fran’s population was 79 percent rural and the labour force was 90 percent and 85 percent in agriculture, respectively (Bharier 1971, 26).  4  A substantial part of the urban population growth in Iran is derived from rural migrants (around one third; see section 3.1).  While the limitation of agricultural resources  (specifically water) and the high natural population growth have a major push impact, the large rural labour out-migration implies that there are exogenous forces at work as well (see section 3.3). These forces are mainly due to the infusion of oil revenues into the urban economy of Iran, which has created higher economic opportunities, better social and physical infrastructures, and a higher standard of living in urban areas (Goldscheider 1983, 186-187).  In essence, cities in Iran attract rural inhabitants with the hope of gaining  8 access to the oil revenue-created opportunities, a fluctuating and unsustainable resource.  As a result, agricultural activities in the rural areas of Iran are under strain of labour shortage, abandoned fields, and under-utilization of resources and establishments. Moreover, other long-run problems can be foreseen, namely: loss of local experienced knowledge and skills, lack of the next generation of farmers, greater dependency on food stuff imports, and ecological imbalance. 9 As such, the historical rural settlement pattern in Iran has been gradually disintegrating and the viability of a majority of 65,000 Iranian villages is threatened.’°  It appears, therefore, that the population distribution trend is  detrimental to rural viability and sustainable development of the country. 8  Analogically, the “pseudo-urbanization’ in the South-East Asia (McGee 1967) with some reservations is attributable to urbanization in Iran (see section 3.3). For realizing the impact of exogenous factors in urbanization in the Middle East, including the case of Iran, refer to Costello (1977). In essence, the insightful inference of M.H. Pesaran about the pre-Revolution hasty industrialization, can be generalized to rapid urbanization of Iran which was “sustained exclusively by means of excessive exploitation of the country’s exhaustive oil reserves” (Pesaran 1985, 26).  Iran depends heavily on food stuff imports. The surge in imports started before the Revolution. In 1974 and 1978, the imports were 1.27 and 2.01 billion dollars respectively (Beaumount & McLachlan 1985, 41). The peak was in the year 1990, with 6.25 million tons of cereal alone. Iran is one of the top importers in the world (IFAD 1992, 27) A major share of oil revenues is spent on this kind of consumer goods. The combination of squeezing oil income, rapid population growth, adverse effects on domestic food production and natural constraints, warn of the unsustainability of this trend. 10  The historical pattern of rural settlements in Iran can be summed up as numerous scattered small settlements in accordance with the natural resource base, often subterranean water capacities (English  Since the 1979 Revolution, great  controversy has raged over the striking disparity  between Iran’s rural and urban areas and over rural-urban migration.”  The avowed  policies of the post-revolutionary State have been to stem the influx of rural emigrants, as it was believed to be deleterious to development (see section 3.3.3). Because of this conviction, from the outset of the establishment of the Islamic Republic, we have witnessed an enormous effort to make villagers satisfied with their way and place of living. To a large extent, a pivotal effort in this struggle has been the provision of physical and social infrastructures in the rural areas. 12 While there are some merits to these efforts, the rural out-migration has continued, and in some time periods, intensified in absolute terms.  In a nutshell, the problem is village viability and a problematic population distribution trend in Iran, associating with agricultural disruption and retardation in originating rural areas, dysfunctioning of receiving cities, and environmental degradation in both areas. The controversy about excessive rural out-migration while provision of public infrastructure as a method by which to remedy the plight of remaining villagers in Iran calls for investigation.  1966, Ch.3; Aresvilc 1976, 13; Misra 1978, 156; Ehiers 1985). However, in this century, non-natural factors (e.g., state policies and market factors) become more significant in the changing pattern of rural settlements (Sharbatoghlie 1991, 34 & 44). In fact, many scholars have argued that rural migrants, especially in squatter settlements around cities, have played a significant role in urban m.rnrest, leading to the 1979 revolution (Kazemni 1980; Ahmad 1981; Hooglund 1982; Kielstra 1987, 218; Kamrava 1990, 109; Lahsaeizadeh 1993, 245; Schirazi 1993, 25). They somehow consider the bulk of migrants as “classless” idlers living in shanty towns with revolutionary potential as the African thinker, Fanon has purported (Fanon 1966). However, some disagree as to this models applicability to Iran (Denoeux 1993, 223). Paradoxically, another school of thoughts refutes the revolutionary potential of the first generation migrants as a whole (Safa 1975, 8). 12 In this study infrastructure is defined as: installation and facilities that provide a fundamental framework for a community and which, therefore, facilitate all forms of development, such as: agricultural, industrial, and educational. (Bannock et a!. 1987, 235). It includes the provision of transport, telecommunications, power supplies, water and sanitation, educational and health care facilities and other public utilities. These are considered as social over-head capital, belonging to the community as a whole (Bannock et a!. 1987, 206; World Bank 1994, 2).  6  1.3 A Principle, a Premise, and an Ethical Issue  The issue of migration in this research is not pursued for understanding of its mechanism and effects alone. Rather, the purpose is to probe into the causes of migration in light of rural and national development (Stark 1991, 19).  This thesis is based on the premise that the village will be an indispensable type of human settlement for years to come in the South. The underlying reasons for this premise are: 1-  The considerable number of people who will continue to live in the South rural areas (Afshar 1994, 272).  2-  The crucial need for village-based agricultural production. (Korten 1992b, 72)  3-  The important nexus of rural community and natural milieu for ecological sustainability. (Lele 1991, 617)  However, by this premise I do not mean to pursue ways of ceasing rural out-migration. Freedom of movement and residence within the boundaries of each country, free choice of employment, and equal access to public service in each country, are all recognized as human rights by the Universal Declaration of Human Rights (Articles 13 & 23).  This  means migration must be open to and optional for all national citizens. Nevertheless, prevailing poor conditions in the rural areas of the South impel many villagers not to stay.  Thus, the research is not intended to provide answers of how to move or how to stop villagers, but What is the reason behind dissatisfaction and how to create a choice of stayingfor villagers. Enlarging the choices of people has been stipulated as necessary for human development (UNDP  1990, 10).  7  1.4 Research Rationale  Research on issues of migration has been neglected in Iran when compared to that conducted in the other countries of the South. With existing knowledge, it is difficult to comment on the general causes and consequences of internal migration in Iran, let alone on assessment of state policies. However, recurring debates on the success or failure of rural development policies after the Revolution in Iran resort to the simplistic criteria of 3 Despite insufficient data and a lack intensification or reduction of rural out-migration.’ of studies, the widely-held belief in Iranian politics is that rural out-migration ought to be reduced, if not stopped. Therefore, the primary rationale of this research stems from the discrepancy between the lack of studies on the issue of rural out-migration in Iran and the major efforts by the Iranian government to reduce it (see section 3.4).  As will be discussed later, it appears that the present population distribution trend in Iran is unsustainable and detrimental to rural development (see section 3.5). This may require us to change our development paradigm first, and consequently, to change our policies towards enhancement of balanced rural-urban development. Meanwhile, we should not downplay many urgent and immediate solutions that can be undertaken within the exercised policies (e.g., rural infrastructure provision policy). These solutions may sustain rural settlements in order to provide conditions favorable to long-term transformation.  Hence, the other rationale of this research is related to a prevailing rural development policy, i.e., infrastructure provision, which is intended to retain the rural population of Iran. Again, the potential of rural infrastructure for playing such a role is not known and needs investigation. Indeed, there are scant empirical studies about the relation of rural out-migration to infrastructure, even in the other South countries. 13  That is, any  policy which results in rural out-migration is wrong and vice versa is right (see section  3.3.3).  8  Moreover, this research is a response to some researchers’ call for research on the ruralside of migration: on the intention of potential rural migrants (Hugo 1981a, 222; Chang 1981, 324; Goldstein 1981, 339; Lee 1985, xviii; Permi 1990, 203; Boyer & Ahn 1991,56)14 and the developmental impacts of out-migration (Balan 1981, 13; Lipton 1982, 25; Stark 1991, 19).’5  1.5 Research Objectives Infrastructure is conceived to be a major component of rural development in the South, and to be provided mainly by states (World Bank 1988, 149; Asian Development Bank 1992, 265; Singer 1992, 116). It is supposed to support rural production and to facilitate social services. In so doing, it is presumed the infrastructure will satisfy the need of rural producers (mainly farmers) for village viability and thus bring about satisfaction and incentives to stay in rural areas (at least for rural producers, if not for surplus labour). This presumption, which was relied upon by Iranian State policy-makers, needs to be investigated.  Across a wide spectrum of theories, there is agreement that voluntary human migration is a purposive move in response to perceived spatial differences in opportunities for betterment in one or more aspects of life. Hitherto, mainstream theories of migration have sought to explain migration in economic terms (see Ch.IV). Also, the most consistent and 14  For instance, Goldstein maintains that: “We also need to know much more about those who stay and who constitute a massive potential, especially in rural areas, for future migration.” (Goldstein 1981, 339) Also, Lee insists on the importance of data on migration intention. (Lee 1985, xviii) 15  In this regard Lipton says: “We need not worry much, or mainly, about the rural-urban migrants themselves... We need not worry that rural-urban migration is massive or swamping... We must worry about the damage that even quite low emigration rates can do; to those who stay in the country-side... (Lipton 1982, 25)  9  recurrent finding of empirical studies is that migration occurs for economic reasons (e.g., Simmons et al. 1977, 12). While accepting that causes of migration are manifold and complex, non-economic reasons are subordinated to, even submerged in, the major trend of economic reasoning.  However, there has been strong evidence supporting the significant influence of non economic migration reasons in the rural South (e.g., Connell et al. 1976, 50; Hugo 1981a, 188; Mitchell quoted in: Oberai et al. 1989, 37). This is not to deny economic reasons, but to stress that it is simplistic to reduce the causes of rural migration to economic ones alone. Migration causes must be thoroughly and deeply understood.  This research attempts to explore non-economic reasons in conjunction with economic reasons. The role of infrastructure in rural out-migration is investigated for the interplay of non-economic and economic reasons.  The objective is to contribute to a humane  paradigm of development, not merely a physical one, and to shift from reductionism to an holistic approach to rural development: a development of people with the help of infrastructure, not vise versa (Chambers 1991, 532; Korten 1984).  The major objective here is to complement the common policies in practice in Iran with a multifaceted approach, integrating non-economical and economical roles of infrastructure in relation to village viability.  1.6 Research Questions The following principal questions will be answered by this study in the selected context: •  Why do people migrate from rural areas?  10  •  To what extent do infrastructure availability/non-availability play a part in the decision to migrate from rural areas?  In conclusion, the determinants of a successful policy for rural infrastructure provision to enhance village viability are logically discussed.  The subsidiary research questions are: 1. What are the characteristics of rural out-migrants and of those who intend to leave the village? 2. What are the causes of out-migration for different individuals and rural households? 3. What kind of rural households have the highest potential to migrate from the village? 4. What is the dominant reason for rural out-migration in Iran, rural push or urban pull? 5. What is the relative importance of various infrastructures in relation to the migration causes of rural households? 6. What is the impact of out-migration on the migrants’ remaining members of the household in the village? and on the agricultural activities? 7. Which rural groups are more crucial to village viability? 8. What are the policy implications of the migration causes of varied rural households for infrastructure provision?  1.7 Research Scope The scope of this study is limited as follows: 1. Iran as a medium income, oil-exporting mixed economy and Islamic country, has been selected for sampling rural out-migration in the South context. 2. While the interaction between rural and urban forces in the occurrence of migration is undeniable, this study undertakes only a one-ended view of the process of rural out migration by concentrating on the forces operating within rural areas. 11  3. Again, recognizing complex and interactive causes of rural out-migration, the study focuses on infrastructural causes. 4. Out-migration in the period of 1979 to 1992 is investigated. 5. Only permanent and intentional migration is considered; On the basis of my four years  experience in the Province before starting this research, other kinds of migrations (e.g. compulsory, circular, seasonal or temporary) are considered not important for this region and/or for this research topic. 6. While it is not intended to be male-oriented research, in practice, due to cultural barriers and a higher rate of participation by males, most of the addressees are maleheaded households and the female concerns may be underrepresented.’ 6  1.8 Research Strategy  To investigate the relationship of infrastructural provision and the causes of rural outmigration, and also, to probe the non-economic as well as the economic aspects together, the research uses a relational and explanatory approach via an empirical study (Northey & Tepperman 1986,10-17). A case study is employed using the conceptual framework of the roles of infrastructure, grounded in the South development and rural-urban migration theories.  The case study in rural Iran provides a basis for interweaving  the complex and  multifaceted causes of migration in relation to the availability or non-availability of the infrastructure. Multi-causality and interaction at different levels set forth the necessity of multi-level research.  Therefore, a macro-level outlook with a structural approach  (including: the country’s development history and state policies) is combined with both the  16  Notwithstanding, over 20 percent of respondents in our household survey were women.  12  meso-level and the micro-level research (including: village, household, and migrant units of analysis), using a cognitive-behavioral approach (Chant & Radcliffel992, 20-23; Uphoff 1992, 331). These are discussed at length in Chapter V.  The research has been conducted in a rural-based and chronically out-migrating region in the western part of Iran, called Hamadan Province (Figure 1.1). This province, with an area of about 20,000 sq. km, is located to the west of Tehran, the Capital. Roughly, two thirds of its 1.5 million people are rural (1986 census), and the rest live in 12 small-tomedium-sized cities.’ 7 The rural population is scattered among more than 1200 villages. However, in terms of the average population size of villages, those of the Province rank first in the country.’ 8  The capital of the Province is the city of Hamadan (population  272.5 thousand in 1986), a historic place on the Silk Road, on the foot of Alvand (an almost 4000m. altitude mountain), and about 300km away from Tehran.’ 9 The overall climate is cold and semi-arid with a lack of water being the main natural limitation of the Province.  As will be described in chapter five, five depopulating villages in Hamadan  Province were selected for the case study.  17 The urban center definition, since the 1986 Iranian census, has been: a population center with town hail or municipality. Other population centers are considered as rural centers. 18  In 1986, the arithmetic average of villages’ inhabitants in Hamadan Province was 782 as compared to 341 inhabitants for the same figure at the national level (Sarrafi et al. 1991, 94). 19 Harnadan encompasses the ancient site of Ecbatana, the second-largest city of Persia and ninth in the world around 650 to 430 B.C. (Chadler 1987, 460-461) It is estimated that the Ecbatana population reached 150,000 around 100 A.D. (Ibid. 570)  13  Figure 1.1: The Location of Hamadan Province in Iran S—i  SOVIET UNION  E  SOVIET UNION  -.  Chardzhoj.,lb...  KizylAivat A.1\Ill’AY Au ,JIAVAI1I  hy•  ‘U,  ,Tabiiz  a)  r” 1  .1 Ardebil  N  .Asbkhabad  •—.  Mary/  A7  I!iI  Gorgárr  nlar  Z’(  •..r—.  Sananda  ?Karkük  ••  ..arl ‘MAZANOARAN  -  5,—  -I-.--  ahi  Mashhad  )  .5..-.  —  SEMNA,J  KHORASAN  ày  \I  >•  Bdkhtarin  ‘  •  (aghd.d  DasMt Kay,,  CI  hard.  f  •.  -—  I  j,  r’ f’  ‘rhorramaba  hi4:  ( I  ..  ‘•  •i’  (AM  9W’  Kashàn  •.  Heui(’i  -  / /  Kavi,-e Narnak  AFGHANISTh  FA HAN  (f bàn  •5,’  ••%•  ,Farili  -•YAZD  •Zarand I  I” lAPIL4Al  r  ‘I  -Kermàa  ‘I.) II  f  -  •Shiraz  •iEI-l1\  FARS  •‘-c.:.  t..  —  ihedáà’\  —‘  Nob usdi  AUDI ARABIA 1  c  Iran  B LUHES TAN  VA SISi AN  Bartda-e  international boundary Province boundary National capital  .  S  Province capital Railroad Road  ‘a-. I  050 Kitomelers o  io  4,  Miles  5-’  --—--——-—--—-—-  14  1.9 Organization of the Dissertation The stream of discussions to be pursued in the following chapters, is as follows: The next chapter begins with an overview of development theories from a migration perspective. It concludes by considering the way in which we should look at the spatial mobility of a population and what developmental viewpoint should be used for assessing rural-urban migration in the South.  Following the description of rural-urban migration as a problem, in chapter three, state migration policies in Iran are explained in order to illuminate macro driving forces. The chapter provides the sets of policies contrived by the revolutionary government which affected rural out-migration. At the end of the chapter, an attempt is made to assess the migration pattern in Iran according to the previously mentioned developmental criteria.  Chapter four extensively reviews the migration theories and their explanation of ruralurban migration in the South. Then, a summary of Iranian migration literature is presented. At the end of this chapter the implications of migration theories for the role of the infrastructure are drawn and a conceptual framework is proposed.  The appropriate  research methods necessary to fulfill the research objectives and to find the answer of research questions are presented in chapter five. The research design and its components are discussed afterward.  The findings of the case studies of rural migrants, remaining members of households, and village key-informants are described in chapter six and chapter seven. The last chapter summarizes the findings in relation to the causes and consequences of rural out-migration in Hamadan Province and to the potential roles of infrastructure in affecting them. Finally, the end of the chapter concludes with the general implications for rural development and out-migration studies in Iran and similar South countries. 15  ChAPTER II: DEVELOPMENT AND RURAL-URBAN MIGRATION  16  2.0 Introduction This chapter explores ways to assess rural-urban migration from a broad view.  Since  development achievement is the mega-framework for assessing societal phenomena in the South, I will focus on the main theories of development and their relation to rural-urban migration. Modernization theory, dependency theory, and an evolving paradigm, called  sustainable development, will be reviewed, and finally, a conceptual framework for the assessment are drawn.  2.1 Rural-Urban Migration and Development Theories  Development is probably the predominant term used to describe the goals of the South countries. There is a lack of consensus about its meaning. “Development”, a normative term (Seers 1969, 2), can be viewed differently from various socio-economic and cultural viewpoints (Sachs 1992). However, it is generally accepted that “the ultimate purpose of development is to improve human well-being” in all aspects (UNCHS 1991, 1)20  Human migration, as change of migrant’s residence, is generally believed to mean a move to improve the migrant’s well-being (Adepojo 1992, 36). Thus, one may postulate that migration is consistent with the goal of development, i.e., that of achieving human well being. Although this view can be true in certain cases, it equates individual development of migrants with the development of society as a whole which is contentious (Oberai 1987, 51; Abu-Lughod & Hay 1977, 195). This contention, Development theories and  20  J found the following definitions to be appropriate in the South context: “Development is an increasing attainment of one’s own cultural values” (H. M. Jackson quoted in: Misra & Honjo 1981, 9). “Development is a process by which the members of a society increase their personal and institutional capacities to mobilize and manage resources to produce sustainable and justly distributed improvements in their quality of life consistent with their own aspirations” (Korten 1990, 67).  17  their interpretations of migration, will be discussed from a theoretical perspective in ensuing sections.  While migration is a divergent phenomenon, I will focus only on rural-urban migration. Using Western yardsticks of development, we see that for decades development has been associated with the reduction of the agricultural share of the overall economy and the increase in the urban population (Oberai 1987, 10; U.N. 1982, 10). As a matter of fact, this association has been seen in the industrialized countries, and to a lesser extent in the South as well (see Renaud 1979, 18).  However, as of two decades ago, numerous  scholars have characterized rural-urban migration as an obstacle to development in the South context (Mortuza 1992, 126; Hauser et al. 1985, 10; Bhaskara Rao 1980,1; Todaro 1976, 2). I will attempt to establish which theoretical framework best serves to conduct research on the South rural-urban migration.  To that end, an overview of the main  development theories will be provided.  2.1.1 Modernization Theory  Modernization theory (i.e., the theories of neo-classical economics) has been the most pervasive theory of development in practice.  In retrospect, this theory began to  overwhelm the South starting in the 1950’s. Economic growth was seen as the only force capable of overcoming underdevelopment. To this end, modernization via the Western technological and economic path was to be applied (Hume & Turner 1990, 24). Thus, saving money, investing, earning foreign currency (with which to import the modern technology) and replacing the traditional socio-economic organizations with modern organizations were of paramount importance. (Friedmann & Weaver 1980, 111)  Modernization theory postulates that “investment in industrial enterprise will stimulate growth in the modern sector, draw surplus labour out of subsistence agriculture into the 18  modern sector, and result in a significant increase in levels of living” (Midgley 1992, 23). The pivotal strategy was that of raising productivity through industrialization and it was believed that a “big push” could make the economy “take off’ into the developed stage (Rosenstein-Rodan 1961; Rostow 1960). Economist-theorists, such as Nurkse in his book on Problems of Capital Formation in Underdeveloped Countries (1953), prescribed rural out-migration as a means by which to achieve economic development (Bhaskara Rao et al. 1980, 7) In a nutshell: industrialization, urbanization and migration, were all seen to be ’ 2 concomitant to the modernization process.  Having regard for scarcity of resources and prevention of dispersed inefficiencies, the 22 process of modernization was to evolve gradually around a leading sector.  Hence, a  dualistic co-existence of modern and traditional sectors was accepted, with the hope of gradual dissolution of the latter (Lewis 1955; Higgings 1956).  The modern sector had its spatial preference and as mentioned, was conducive to urban growth (Mabogunje 1989, Ch.7).  The process, in turn, has created intra- and inter-  23 (Stohr & Taylor 1981). However, the idea of “polarization and regional disparities  As an example of this belief, a classic work on modernizing the Middle East prescribes that masses should move “from farms to flats and from fields to factories” and seek what the West is (Lemer 1966,  21  48). Also, the following conclusion about migration policies in Asia by a U.N. research center is selfevident: “Urbanization is the direct consequences of economic development and industrialization. Urbanization and migration are also positively associated” (Bhaskara Rao et al. 1980, 14). In hindsight, the relationship of industrial development and urbanization in the South cannot be explained as a priori. 22  These description could be oversimplifications due to the differences existing among the pioneers of economic development theory. For example, the notion of the leading sector by Rostow were weakened by the balanced growth of Nurkse proposal (ESCAP 1979, 24). Yet, for our general purpose, the description of the core attitude seems adequate. Emphasizing industrial sector and related services, the principles of economies ofscale tend to place the modern sector in the major urban centers of dual economies(Johnson 1970, Ch.5). In the same course, the strategy of growth poles was proposed for induced modernization in regional planning Haiiseii 1972). Combining with the high degree of political centralization in most South countries, primate cities have emerged as a spatial manifestation of these discriminatory policies. (McGee 1971; El Shakhs 1972) 23  19  trickle-down’ was formulated (Hirschman 1958) to suggest that the centripetal stage was transient and later a centrifugal stage would occur (Pryor 1975, 24).  As Walter Stohr  says: A basic assumption of neoclassical economics is that regional imbalances between supply and demand of production factors or commodities will even out automatically once the accessibility between regions and the mobility of production factors and commodities increased sufficiently. (Stohr 198 ib, 218)  Two conclusions are asserted by modernization theory: 1) Labour migration from rural/traditional sector to urban/modern sector is imperative for development; 2) Such migration is a transitory phenomenon which will cease in higher stages of development, when spatial equilibrium is supposedly attained. Generally, modernization theory holds a positive view of rural-urban migration, as fostering modern development and “transfer of prosperity” (Osterfeild 1992, 194).  2.1.2 Dependency Theory In the early 1970’s, because of the persistence and even aggravation of poverty and inequalities on one hand, and the sluggish attainment of national development goals on the other, some of the South leaders who had been disenchanted with the modernization approach, appealed for another development (Dag Hammarskjold Foundation 1975). This appeal was answered by a countervailing theory named dependency theory which to some extent influenced the mainstream practices of development.  Theoretical dissent about modernization outcome may be traced back to Gunnar Myrdal (1957). He argued that dual economies would be perpetuated in a developing situation  20  through backwash effects. 24 This set the stage for positing that the backward sector is re created by the modern sector. Building on this idea at an international level, as elaborated by Frank  (1967)  and  the  U.N.  Economic  Commission  for  Latin  America,  underdevelopment was explained by its linkage to developed regions (Armstrong & McGee 1985, 21).  Put another way, this theory is concerned with the structural relationships between the core [developed country] and the periphery [undeveloped country] (Friedmann 1966).  From this perspective, it is external impediments which are the decisive factors, as opposed to internal ones in modernization theory (Agazzi 1988, 45; ESCAP 1979, 26). Moreover, dependency theory refutes the necessity for South countries to follow the same economic path as Western countries, and advocates the need to delink from the world market system(Seers 1981). Overall, the emphasis was placed on self-reliance, equitable distribution, inward-looking growth, and the reduction of regional and sectoral imbalances in the country. 25  According to dependency theory: 1) Population mobility in the peripheral countries is motivated by dependent development and it increases existing inequalities; 2)The trend will not cease, and it will debilitate indigenous development. Dependency theory holds a  24  It is unfair not to acknowledge that Mahatma Gandhi was the first political leader who rejected the imitation of the Western model of development in a coherent manner and elaborated on the culturalspecific and self-sufficient path of development during his compaign for India’s independence in the 1940’s (For a brief summary of Gandhi’s developmental vision refer to Misra & Natraj 1981, 275-8). Unfortunately, this has not been sufficiently recognized in development theories. (Agazzi 1988, 18). 25  Some writers enumerate another point of view called structuralist or world system (Hajnal & Kiss 1988, 19; Welsh & Butorin 1990, 322) which I consider as a mixture of the preceding views in terms of discerning both internal and external impediments, though in an interwoven structure. Thereby it acknowledges: the truth lies somewhere between the two! This view has not been spelled out and it seems that the implied strategy resembles more the dependency one (Packenham 1992, 111). The prominent writers in this regard are Galtung (1980) and Wallerstein (1979).  21  negative view of migration, referring to the result as  26 and treating it as overurbanization  a symptom of economic underdevelopment (Shrestha 1990, 60).  2.1.3 Sustainable Development In addition to the aforementioned well-known classification of general development theories, it seems a third distinctive theory is emerging.  Unlike the preceding theories,  which in general were both based on classical economics and its ramifications in other social sciences, the new paradigm questions the bases as well as the results. Some almost concomitant tenets have been incorporated into this alternative paradigm under the banner  27 of sustainable development.  • Social Equity: A certain tenet contributing to this alternative, the social equity concern emphasizes non economic values. The archetypal idea could be found in Schumacher’s Small is Beautful (originally published in 1973):  26  Here the evaluation of migration is closely related to its consequences in the urban areas. The term overurbanization is used to show that urban growth has greatly outstripped industrial growth (Rogers 1977, 68) and the productive absorptive capacity of the South cities. With a slightly different meaning, Gugler argues that overurbanization should be viewed as the misallocation of labour between rural and urban areas in the South economy (Gugler 1986b). Correspondingly, the term hyperurbanization is found in John Friedmann’s book with the same comiotations (1973, Ch.5). In the same stream, considering the variance of Western course of urbanization with the South, T. G. McGee coined the term pseudourbanization (1967, p 1 9). Later, he revised it concerning the urbanization process in a few South countries -viz. Newly Industrialized Countries, e.g., South Korea, Taiwan, Brazil (Armstrong & McGee 1985, 2-5). Yet, the trust of cities’ roles in strengthening dependency and in siphoning-out the accumulated capital is strong in this stream of idea (Ibid. 213-219; Amin, 1976). 27  The well-quoted definition by the Brundtland Commission defines sustainable development as “development that meets the needs of the present without compromising the ability of future generations to meet their own needs.” (WCED 1987, 43) Bearing in mind a wide variety of interpretations from the concept of sustainable development, I use the term in its radical meaning (cf. Daly & Cobb 1989; Bookchin 1989; Rees 1990; Boothroyd 1991; lIED 1991; Brown 1993) which will be summarized hereinafter. The mainstream approach to sustainable development by international agencies maintains economic growth while it renders some conciliatory adjustments in the present economic system (cf. WCED 1987; IIJCN/UNEP/WWF 1991; World Bank 1992; UNCED 1992).  22  There is the immediate question of whether “modernization”, as currently practiced without regard to religious and spiritual values is actually producing agreeable results. As far as the masses are concerned, the results appear to be disastrous- a collapse of the rural economy, a rising tide of unemployment in town and country, and the growth of a city proletariat without nourishment for either body or soul. (Schumacher 1989, 65) if we could return to a generous recognition of meta-economic values, our landscapes would become healthy and beautiful again and our people would regain the dignity of man... (ibid. 124)  Negating the overemphasis being placed on economic growth, Schumacher admonishes: The starting point of all our considerations is poverty... The primary causes of extreme poverty are immaterial, they lie in certain deficiencies in education, organization, and discipline. Development does not starts with goods, it starts with people. (Schumacher 1989, 178-179)  On their part, these concerns are incorporated into the emerging paradigm of sustainable development,  emphasizing:  spiritual values,  human scale,  appropriate technology,  agricultural and rural sectors in achieving social inclusion and equity (see: Friedmann 1 988b).  • Ecological Balance: The other tenet which has been constructing the core of this alternative paradigm is 28 According to Goodland and Daly, it started with the source limit ecological concerns.  -  28  The climax of global concerns for the environment can be best seen in the Earth Summit at Rio, an unprecedented participation of 178 governments -including 106 heads of state- in June 1992 (Brown 1993, 3). Apparently, there is a growing consensus that the exercised course of development is leading to an ecological collapse and must be altered. Nonetheless, convictions about the degree of necessary changes drastically diverge. The following excerpts from Agenda 21 should be viewed as the bottom line of general agreement: “Underlying the Earth Summit agreement is the idea that humanity has reached a turning point. We can continue with present policies which are deepening economic division within and 23  i.e. depletion of natural resources like fossil fuels- and later the sink constraints -i.e. the overshooting of the absorptive carrying capacity- have been added (Goodland & Daly 1992a, 37). Accompanying these, a trust to conserve the natural capital and to cease the throughput growth is the cornerstone of ecological views (Ibid.).  Consequently, this  means a shift to environmentally friendly technologies, and as a whole, it denotes a necessity to change the consumption and production pattern. Moreover, a far-reaching equitable redistribution is intended to break the vicious cycle of poverty and environmental damage. 29  Overall, the third theory relies on a different paradigm with the all-encompassing concept of sustainable development. In the wake of challenging the embedded values and outlooks of conventional development, the new paradigm rejects the economic model of unlimited growth, consumerism and production-centered attitudes (Korten 1990; Ekins 1992). Instead it builds upon a steady-state economics (Daly 1991), on conservation, and on a biophysical approach (Wackernagel 1992), on eco-community (Bookchin 1977), and on people-centered development (Korten 1990).  In summary a comparison of development approaches show that if the emphasis of the modernization was on growth and efficiency, and of dependency was on independence and national equity, the third approach places its priorities on ecological balance and social between countries -which increase poverty, hunger, sickness and illiteracy and cause the continuing deterioration of the ecosystem on which life on Earth depends. Or we can change course. We can act to improve the living standards of those who are in need. We can better manage and protect the ecosystem and bring about a more prosperous future for all. No nation can achieve this on its own. Together we can -in a global partnership for sustainable development” (UNCED 1992, 3). 29  This is a well-documented fact and widely accepted (for a recent reference see Jazairy et al. 1992). A cardinal policy in this regard is population control, which is being accepted overwhelmingly by the states. It should be noted that overconsumption by the rich -mostly in the West- is also considered as a major source of ecological threat, if not a greater one (refer to: Ekins 1991; Daly 1989). In fact, despite the need for ceasing growth economics in the North, growth is imperative in the South for sometime (Daly 1992, 12; SID 1992, 4). “It is ironic that significant environmental degradation is usually caused by poverty in the South and by affluence in the North” (UNDP 1991, 28). -  24  equity at a global scale. Here, the hallmark of development is the preservation of the lifesupport system for all at present and in the future.  Implications for Migration in the South: Bearing in mind the variety of migration patterns in the South (Nam et al. 1990, 8; Gugler 1986, 2 12-217), there is a pitfall of over-generalization in interpreting the pattern and its effects.  However, since the purpose of this section is to give a schematic view of  sustainable development implications for migration, some interpretation for a stereotyped pattern can be made.  From the viewpoint of sustainable development -particularly from its biophysical approach, the prevailing rural-urban migration in the South unquestionably hampers the development of the South by intensif,’ing the present form of urbanization and its implications (Rees & Wackernagel 1994; Wackernagel et al. 1993).  In fact, from  Schumacher’s view, the major problem arises from the economics principles which make people footloose (Schumacher 1989, 72). He contends that prior to the advent of mass transport, because of the community structure, “there were communications, there was mobility, but not footlooseness” (Ibid. 73). Schumacher states that destructuring of the poor countries, “produces mass migration into cities, mass unemployment, and as vitality is drained out of rural areas, the threat of famine” (Ibid. 75). He sums up by claiming that rural migration resulting from the dual economy is poisonous to both the rural and urban sectors. 30 30 The following excerpt from Schumacher is self-evident: Until recently, the development experts rarely referred to the dual economy and its twin evils of mass unemployment and mass migration into cities. When they did so, they merely deplored them and treated them as transitional. Meanwhile, it has become widely recognized that time alone will not be the healer. On the contrary, the dual economy, unless consciously counteracted, produces what I have called a ‘process of mutual poisoning’, whereby successful industrial development in the cities destroys the economic structure of the hinterland, and the hinterland takes its revenge by mass migration into the cities, poisoning them and making them utterly unmanageable” (1989, 177).  25  It may be stated that if the source limit and the sink constraints are both performing as a zero-sum game at the planetary level, then the place of people, on the macro scale, is not conducive to sustainable development.  However, the rural life style is virtually more  sustainable, considering its reliance on agricultural activities and its low level of consumption and waste.  This sustainability is endangered by ongoing trends,  overpopulation in general, and a growing number of impoverished rural people in particular (UNCED 1992, 31; Jazairy et al. 1992; World Bank 1990), application of inappropriate technologies in general, and expansion of fossil fuels-dependent agricultural methods in particular (Daly & Cobb 1989, 272).  In terms of urban areas, their concentrated growth often supersedes the absorptive capacity of the environment (Rees 1992). In addition, it causes inefficient use of resources due to automobile dependence (Roseland 1992, 26), urban squalor, conspicuous consumption and specious investment (lIED 1991, 31; IUCN/UNEP/WWF 1991, Ch.12; Gugler 1988, 86).  Therefore, in the status-quo the rural-urban migration in a typical  31 South country is detrimental to sustainable development.  In sum, the present rural-urban migration in the South, has been seen at the foremost as a result of wrong socio-economic policies, which intensify a non-viable pattern of settlement and environmental degradation. 31 There are optimistic views of the ongoing pattern which suggest the problem can be solved by the patterns themselves: “Human settlements, and particularly large urban agglomerations, are major contributors to environmental degradation and resource depletion. At the same time, human settlements, large and small, are also areas of unused opportunities: creativity, economic growth, communication; accessibility for transfer of knowledge; and an efficient and effective attack on waste and pollution.” (M. van der Stoel, ‘Statement of the chairman’ to the Intergovernmental Meeting on Human Settlements and Sustainable Development, The Hague, Nov. 1990; cited in: UNCHS 1991, 0). Another view holds that salvation from the negative aspects of urbanization lies in tackling fmancial shortcomings (World Bank 1992). And since, supposedly, there is a positive relationship between economic development and urbanization (Henderson 1987), ultimately, rural-urban migration is not a negative phenomenon.  26  2.2 Assessment of Rural-Urban Migration  The preceding section sought to highlight the relationship of development theories and their presumptions to rural-urban migration. While the interdependent nature of resource allocation and population distribution is obvious, in reality, priority is given to economically and politically-driven decisions for resource allocation, thereby, causing population displacement.  All development theories accept that South rural-urban migration is engendered by ruralurban inequalities, but the solution proposed within each theory varies. Accepting the reality of urbanization, modernization theory calls for continuing the current path, 32 However, according to several promising future equilibrium in a developed economy. researchers “in reality, this equilibrium mechanism does not materialize in most 33 (Stohr 1981a, 45; also: Abu-Loghod & Hay1977, 105; Rodinelli 1984, 222) countries.”  For dependency theorists the solution lies in cutting international ties and adjustment of national policies toward even development. Looking back, no country has experienced such a development strategy thoroughly.  Anyhow, the viability of this strategy in face  of economic hardships, and presumably, trade embargo is questionable. The ecological blind-point in this theory (i.e., not reckoning with the environmental impact of socio economic processes), as in the modernization theory, eventually makes it unsustainable.  A recent time-series study on Israel verifies the U-shaped curve in relation to the decrease of the second phase of development” (Alperovich 1992, t population concentration after progressing toward ‘ 63). However, the author concedes that international cross-sectional data fail to confirm this relationship 32  (ibid. 72). Preswnably, Taiwan, South Korea, and Israel are among the exceptional cases. The ensuing section describes the continuation of unbalanced trends in the South. In my opinion, policies of some socialist countries, like: Albania, Cuba and China before 1978 may be partly interpreted as the practice of this theory. -  -  27  Sustainable development theory evokes a paradigm different from that of neo-classical economics in practice. Consequently, the generative forces of rural-urban migration will occur on the basis of ecological constraints not economical exigencies and consistent with a healthy community-planet relationship. Rural-urban migration may thereby take place to balance the carrying capacity of bio-regions at a macro scale and/or to respond to microecological imperatives (Andruss et al. 1990; Sale 1985). Ultimately, in the next century the world population should be stabilized and thus, the footlooseness of people will be drastically reduced.  This overview of development theories does not provide sufficient data with which to draw up a list of advantageous and disadvantageous effects of migration. However, it acknowledges that assessing migration is embedded in underlying development theory and that “the overall implications of migration for national development  .  .  .  cannot be stated a  priori” (Oberai 1987, 70). Moreover, it poses the need for a conceptual framework and both global and local criteria to assess developmental impacts of rural-urban migration in a South country.  To find out the guidelines for assessing rural-urban migration, first, the trends of population distribution in the South are described. Next, a general framework for migration assessment is elaborated.  Finally, the assessment’s developmental tenet is  proposed.  2.2.1 Trends of Population Distribution in the South  Having summarized the theoretical spectrum of migration interpretation, I now turn to the consideration of the factual population distribution in the South. I intend to give a sense of the dimension of trends before posing a conceptual framework for the assessment. 28  First of all, it should be noted that the quantity and pace of population growth in the South are two of the driving forces of distributional changes. In 1990, over 77 percent of the world’s population was in the South and 93 percent of world population growth occurred therein (U.N. 1991a, 10&100).  Secondly, the mega-trend is undoubtedly one of rapid urbanization. The urban population of the South has risen from 17 percent in 1950 to 37.1 percent in 1990, and is expected to pass 62 percent in 2025 (U.N. 1991b, 106-107). This means, that since 1950 the volume of urban population has multiplied 5.3 times, until 1990, when it amounted to 1.5 billion. It is projected to multiply 15.3 times by 2025 (Ibid. 118-119).  Thirdly, urban agglomerations containing 5 million or more inhabitants, have been growing rapidly in the South. While 11 of the 20 world agglomerations in 1970 had been located in the South, the figure was 23 of 34 such agglomerations in 1990 (U.N. 1991b, 22). It is expected that 34 (more than three forth of the projected 45) agglomerations in the year 2000 will be in the South (Ibid.).  Finally, another trend in South urbanization is the emergence and proliferation of mega cities, defined by the U.N. as those with 8 million or more residents. In 1970, there were 10 world mega-cities; 5 belonged to the South (U.N. 1991b, 24). This figure changed to 14 mega-cities in the South, out of 20 in the world, and it is projected that by the year 2000, 8 more cities will have gained mega-city status, all located in the South (Ibid.). This urban tendency of over-concentration is materialized “at the expense of towns with populations below half a million” (U.N. 1982, 9).  29  The general perspective shown above, hints at the contribution of rural-urban migration to South urbanization, or as the IJNDP calls it, to the urban explosion (UNDP 1990, 85). As a matter of fact, the average annual rates of change of urban population between 198085 and 1985-90 have been 4.62 and 4.53 respectively as compared with 0.97 and 0.81 for rural population (U.N. 1991b, 154&160). Bearing in mind that the change in distribution of population has taken place through differential natural population growth, and international and internal net migrations. A crucial component of South urbanization is internal migration, particularly rural-urban one (Parnwell 1993, 18; Nam et al. 1990, 9; Bhaskara Rao et al. 1980, 4). It is estimated to account for an average of 40 percent of urban population growth (White & Whitney 1992, 18; Gugler 1988, 74; Hauser et al. 1985, 8; Todaro 1976, 10; also, for seeing the wide variety of contributing rates among countries, refer to: UN 1980, 24 and Renaud 1979, Annex Tables 2.3).  All the above trends herald aggravation of population distribution imbalances in most of the South countries, with the momentous share of rural-urban migration in the predicament.  2.2.2 A Conceptual Framework for the Assessment  Taking into consideration the difference in theoretical approaches, assessment of ruralurban migration requires a multifaceted approach, distinguishing between different causes of out-migration at the place of origin and between different consequences at the place of destination. The following diagram (Figure 2.1) depicts the hypothetical assessment of rural-urban migration (with positive and negative symbols as appropriate and inappropriate streams, respectively) relating “push” forces in the village to “pull” forces in the city. The diagram shows two overall contexts for the propensity to leave the village:  30  •  When there is excessive labour force, as the outcome of mechanization diffusion, saturation of the agriculture-sustaining natural resources, and overload of the non agricultural sector, (etc.); showing as a thriving state and positive [+] repulsion.  •  When there is an out-flow of people, due to the deterioration of the socio-economic system, perception of rural-urban disparities, degradation of life-supporting natural resources, occurrence of natural disasters, (etc.); showing as a declining state and negative {—j repulsion. 35  In the city, the rural in-flows can be categorized into two extreme situations: •  When the in-migrants are absorbed by the productive urban sectors which indicates a positive [+j attraction because ofgenuine demand in the urban labour-market.  •  When the supply of in-migrants are more than the city’s absorptive capacity and the unproductive urban sectors are inflamed by emigrants which imply a negative [—j attraction because offalse expectation on behalf of the migrants.  Figure 2.1: Rural-Urban Migration Assessment Model  /  Genuine Demand  Pull \  \  False Expectation  /  +  Thriving Situation  —I.. V’N + Declining / Situation \ —  Source: Author  It is also possible that due to overpopulation and the resultant pressure on natural resources, the out flow of population can be viewed positively in terms of its environmental impact in the village. Though, in retrospect, rarely has the out-flow acheived a level which has removed population pressure.  31  These categories identify four variants for rural-urban migration, only one of which may be assessed as merely negative [—j at both ends from a societal perspective (i.e., leaving the declining state of the village with false expectations or a lack of socio-economic opportunities in the cities). This conceptual framework gives the possibility of applying a variety of development theories. In other words, the normative evaluation for ascribing the declining state, genuine demand and so on, is changeable according to different theoretical explanations.  Generally speaking, there is a combination of all these variants in rural-urban migration in the South. Hence, policy-making should be differential and in accordance with concrete regional development cases and visions. In the absence of such visions, making decisions on migration is a political action, and is vague in developmental terms.  2.2.3 Rural-Urban Migration in the South: A Preliminary Assessment  Despite conceptualizing the above model, rural-urban migration cannot be assessed unless a specific development paradigm be employed for distinguishing the four situations in the model (i.e., declining situation, thriving situation, genuine demand, and false expectation). On the basis of our development theory overview, it has been argued that one of the approaches, that is sustainable development, has a promising and indispensable tenet. In this view, normative evaluation of rural-urban migration in the South ought to take into account the sustainability of the receiving and the originating communities as well as the global community.  Accordingly, when there is environmental degradation in the rural areas, whether because of source depletion (e.g., fall of water table, soil erosion, deforestation) or sink overflow (e.g., water pollution, soil poisoning, acid rain), the declining situation label is applicable.  32  However, from the mere economic point of view, it could coincide with thriving or slack periods.  The thriving situation is when the environment is saturated without human-made degradation, and thus, if the excessive population (i.e., more than the natural carrying capacity) stays, then there will be degrading pressure and/or loss of livelihood. Again, this could be simultaneous with thriving or recession economic periods.  Sustainable development requires social equity, economic viability, and ecological balance (BCRT 1992). These encompass a broad range of values. The following are suggestions from some of the scholars: •  36 Meeting the basic needs of all people  •  Broadening individual and community participation, and control over resources (Korten 1992b, 71)  •  37 Enhancing the social capacities to deal with scarcity and limitations  •  Protecting life support systems (Robinson et al. 1990; Korten 1992a, 97)  The declining situation or environmental degradation occurs when the above criteria have not been met in the originating village. Conversely, if these are met, a thriving or healthy environment could be declared to exist.  David Korten defines these needs as: “a humanly adequate standard of food, clothing, shelter, education, basic medical care, and decently remunerated employment.’ (Korten 1 992b, 71) I would add a psychological dimension, say self-esteem, and a social one, say belonging together and freedom, as well. (Goulet 1971) 36  Specifically, regarding the economy the criteria are: a) Reduction of throughput of energy and other natural materials. b) Waste emissions within the assimilative capacity of the local environment. c) Harvest of renewable resources within their regenerative capacity. (Goodland & Daly 1992a, 37 & 1992b, 66)  33  For example, in a given agricultural economy, if fossil fuel throughputs are increasing (as a direct result of, say, mechanization, or as an indirect result of application of fertilizers, pesticides, and herbicides) even though the production grows, this is an unsustainable way 38 However, in such a situation there of development, and of environmental degradation. could be no out-migration if the new employment opportunities absorbed the unemployed labour force.  Clearly, these criteria are too complex and too general to be precisely applied to ruralurban migration in the South. But my intention is to scan the major considerations in assessing the overall pattern of migration. In other words, I want to sketch criteria on the basis of holistic developmental effects, and to conclude whether there is a need for change in the prevalent migration patterns. Like those regarding other social phenomena, this conclusion will be conditional upon temporal and spatial conditions and may be varied greatly.  Drawing on my understanding of sustainable development criteria, it can easily be elicited that most of the rural-urban migrations in the South are not compatible with the 39 Unquestioningly, those who sustainable development paradigm which is outlined above. are leaving the rural areas at present are entering an unsustainable urban development, considering their ecological impact (White & Whitney 1992). The question remains open if this modern way of urban development can be transformed into a sustainable lifestyle and also, if the traditional way of rural development may be viable in face of global market forces. 38  This is due to inconsistency with the principle of “reduction of throughputs”, and it results in the limitations.  debilitating of the social capacities to deal with scarcity and  Undoubtedly out-migration has helped to ease pressure on the rural job market (Dasgupta 1981, 53), and in turn, has put a strain on urban labour markets in the South (Gugler 1986, 195). However, the unsustainablity implications in temis of production and consumption patterns are more vivid in the cities.  34  In fact, it is not certain whether the modern way of rural development is less harmful than the urban one. available,  The effect of migration on the community depends on the resources  the number of people living off of these resources,  the  rate of  exploitation/consumption, and the type of technology which they are using (Ehrlich & Ehrlich 1990, 58). Therefore, the diagnosis of unsustainability cannot be reduced to a rural/urban or agriculture/non-agriculture differentiation.  However, viewing rural out  migration as an index of dissatisfaction, invites us to investigate the viability of rural settlements as a whole. Here, my concern is about the majority of the South population, i.e., the rural people, and their livelihood, which is mostly based on agriculture. As such, the assessment of rural-urban migration is looked at from the village standpoint and from its effect on the sustainability of the community. This will be discussed concretely in terms of the Iranian case at the end of next chapter (section 3.5).  35  CHAPTER ifi: RURAL-URBAN MIGRATION AND NATIONAL POLICIES IN IRAN  36  3.0 Introduction This chapter outlines the Iranian context and discusses the problems of rural-urban migration. It begins with presenting the statistical dimension of the problem, covering the available data of four recent decades. It shows the significance of rural-urban migration as part of unprecedented urban growth.  The next section introduces the necessity of policy intervention by the South states in the population distribution.  It goes on to classifS’ varied policy approaches.  Lastly, the  decisive role of the Iranian state is demonstrated.  A brief history of migration in Iran continues the contextualization of the problem of ruralurban migration and its relationship to sustainable development. It is demonstrated how the internal and external intervening factors create migration streams, regardless of the equilibrium process. Subsequently, the policies of post-revolutionary Iran are reviewed. It is stated that infrastructure provision at the rural-end has become an important policy tool, though without proving to have successful results.  Finally, a general appraisal of  rural-urban migration in Iran is presented.  3.1 Rural-Urban Migration in Iran A staggering trend in the last four decades in Iran has been rapid urbanization and its concomitant influx of rural out-migrants to the large cities. To convey the quantitative meaning of this abrupt shift, the Iranian population components are illustrated here in statistical data. Whereas national and provincial data are only available for the census years, the data are presented in ten-year intervals from the first national census in 1956 up to 1986, augmented in some cases by a recent broad national survey in 1991.  37  Generally speaking, since 1956, the Iranian population mobility both in relative and absolute terms, has continued to increase. In particular, life-time migration (as the only type of internal migration that has been the subject of a question in all the four decennial censuses) has almost doubled within thirty years, from 1956 to 1986 (see Table 3.1). The high proportion of migrants in cities compared to villages is notable. It implies that there is more attraction to the cities. The Table also shows an accelerating rate of life time migration in the last 35 years, in which the recent jump in rural areas from 5.3 percent ° 4 before the Revolution to 15.3 percent after the Revolution is remarkable. Distribution of Iranians Residing in Places Other Than Their Birthplace in Urban and Rural Areas* (percentage) Year** Country Urban Areas Census Rural Areas 11.0 1956 1966 13.1 26.4 4.6 1976 15.5 27.2 5.3 1986 21.6 29.7 11.9 1991 25.3 32.2 15.3  Table 3.1:  Rural population, hereafter, includes unsettled population, e.g., nomads and gypsies, accounted for 0.5 percent of total population of Iran in 1986. ** Except 1991 which is the year for a national survey and the figures must be treated discreetly. Source: Calculated based on SCI 1992a, 48-2; SCI 1990a, 3. *  The following sections will explain the changes in: the national distribution of the rural and urban population,  the rate of rural-urban migration, and the settlement pattern of  population in Iran.  40  The life-time migration to rural areas, according tomy observation and inference, has been mostly migration to: 1- Frontier agricultural areas (e.g., Moghan, Jiroft, Gonbad and part of Khuzestan), 2Marginal settlements in the vicinities of major cities (e.g., new squatter settlements or commuter settlers in nearby villages around Tebran and Isfahan) and, 3- Frontier mineral lands (e.g., Bafgh in Yazd, Pabedana and Sarchesh-meh in Kennan). The sudden growth of migration to rural areas during 1976-86, in addition to what was pointed out, is due to the destruction of some 4000 rural settlements by Iraq’s invasion (Arnirahmadi 1990, 64) and the consecutive inhabiting of other villages. In addition, the Afghani’s settlement in rural areas is another contributing factor to the rise of non-indigenous villagers.  38  3.1.1 Rural and Urban Distribution of Population The unceasing exodus from rural society to urban society can best be seen in Table 3.2. Over a period of 35 years, the urban population of Iran grew more than 5.5 times while the rural population grew less than 2 times.  Correspondingly, the urban share of the  population raised quickly from 31.4 percent to 57 percent.  Overall, it is estimated that  near 33 percent of the total urban population growth in the period between 1956 and 1991 consisted of net rural-urban migration.  41  Table 3.2: Iran’s Urban & Rural Population Changes during 19561991* Average Annual Growth Rate (%) Urban Total Rural  Census Year**  Country (in 1000’s)  Urban (in 1000’s)  Rural (in 1000’s)  1956  18,955  5,954  13,001  (100%)  (3 1.4%)  (68.6%)  1966  25,788 (100%)  9,794 (38.0%)  15,994 (62.0%)  3.13  5.10  2.09  1976  33,709  15,855  17,854  2.71  4.93  1.11  (100%)  (47.0%)  (53.0%)  49,445  26,845  5.41  2.27  (100%)  22,600 (45.7%)  3.91  (54.3%)  55,837  31,837  24,000  2.46  3.47  1.21  (100%)  (57.0%)  (43.0%)  1986 1991  *The figures are rounded in thousands. ** Except 1991 which is the year for a national population survey, and its figures should be treated discreetly. Source: Computed from: SCI 1982; SCI 1985; SCI 1991; SCI 1992a.  41  The average annual growth rate of the urban population has been 4.91 percent, comparing to 3.13 percent for the whole country. Given the lower fertility rate of the urban population, it is estimated that 10.9 million is made up of the natural increase. Thus, the reminders of the increase in urban population during the last 35 years, i.e., 15 million, has originated from outside the cities of 1956. A rough calculation shows that around 6.5 million accounted for the reclassification of rural centers as new towns, and for their natural increase within this time period. Consequently, the remaining 8.5 million of the increase was caused by net rural-urban migration. In other words, the components of urban population increase since 1956 are: 32.8 percent net rural-urban migration, 42.1 percent natural increase of the original urban areas, 25.1 percent reclassification and pertinent natural increase. However, a major deficiency in this calculation is the unknown number of international migrants. The international outmigration from the large cities could be quite effective in changing our calculation of rural in-migrants to a higher number.  39  3.1.2 Rural-Urban Migration  As far as rural-urban migration is concerned, comparable census data is only for life-time migrants (i.e., residing in places other than their birthplace) as shown in Table 3.1. Only in the latest Iranian national census, has the duration of stay for life-time migrants and some characteristics and origins for the in-migrants within the census interval (i.e., since ten years before 1986) been augmented.  Therefore, the calculation of flow of migration  between rural and urban areas is an estimation by using the indirect method of natural growth assumption (Oberai 1987, 31).  The method assumes a slightly lower natural  increase in the urban areas, than in the rural areas.  Wherever it was possible, a  comparative study of other researchers’ estimation with this or other methods has been employed.  The net in-migration to the urban areas from 1956 to 1966 is calculated to be approximately 1.5 million, an average of 150 thousand a year, and 39.1 percent of urban population increase. 42  Because of the oil boom, the climax of city-ward migration  occurred in the next decade.  The volume of net rural-urban migration sprang to 2.4  million between 1966-1976, equal to 39.6 percent of urban growth. 43 The same migration figures for 1976 to 1986 are calculated at about 3.2 million, or a surge to an average of 320 thousand net urban in-migration and a relative decline to 29.1 percent of urban  42 For comparison’s sake, the following has been calculated independently by others for the 1956-1966 period: 1.33 million and 1.68 million Bharier 1972, 58 from two different methods), 1.2 million (Zahedi Mazandarani eta!. 1985, 83), 1.52 million (Karshenas 1990a, 283).  The independent calculations of net rural-urban migration between 1966-1976 are: 2.2 million (Majd 1992, 453); 2.337 million (Karshenas 1990a, 283), 2.578 million (Hakimian 1990, 136), 3.5 million Mohtadi 1990), 2.3 million Zahedi Mazandarani et al. 1985, 83), 2.363 million (Alizadeh & Kazerooni 1984, 71), 2.111 million (Kazemi 1980, 28). Danesh estimates the volume of rural-urban migration at 6.2 14 million during 1966-1982, on the basis of a lower natural growth rate and a higher population for the urban component (Danesh 1987, 48). Another researcher has claimed that more than a million rural migrants entered the Iranian cities every year between 1973-78 (Farazmand 1989, 155), a certainly incorrect figure.  40  population increase. 44  A milestone in this decade was the Islamic Revolution and its  dramatic socio-economic upheavals. As will be described later, two major groups of warstricken people and Afghani refugees intensified urban growth in this time period, as well.  From the recent national survey, it is estimated that roughly 1.4 million rural people have migrated to urban areas in the 1986-1991 period, yielding only a 28 percent urban population increase. Table 3.3 shows the aforementioned migration volumes.  Table 3.3: Estimation of Iran’s Rural-Urban Migration during 1956-1991 Time Period  Total No. of RuralAnnual Average of R-U Migration as % Migrants* Urban R-U IVligration of Urban Growth 1900-1956 700,000 12,500 16.5 1,500,000 1956-1966 150,000 39.1 2,400,000 240,000 1966-1976 39.6 1976-1986 3,200,000 320,000 29.1 1986-1991 ** 1,400,000 280,000 28.0 * See footnotes 42 to 44 for alternative estimations. ** See the text for dubious estimation on the basis of the 1991 figure. Source: Calculated based on: Bharier 1972, 55; Karshenas 1990a, 283; SCI 1982; SCI 1985; SCI 1992a.  , if correct, the trend shows that for 45 While there are serious doubts about the 1991 figure the first time in Iranian history, there has been both a relative and an absolute decrease in rural-urban migration. However, it should be noted that because of the migration surge in the 1976-86 period, a subsequent drop in the number of rural-urban migrants is  ‘  Other researchers t calculations for 1976-1986 are: 3 million (Lahsaeizadeh 1993, 288), 3.2 million (Schirazi 1993, 313). One researcher has stated the doubtful figure of 2.2 million for the 1979-1983 period (Haghayeghi 1990, 45). Some analysts have estimated that 2.93 million has been the net migration to the urban areas of Iran between 1976 and 1982, an average of 488 thousand a year which was sharply reduced after this period (Alizadeh & Kazeroom 1984, 72). One reason for this doubt is the result of preliminary extraction which shows about 1.3 million more urban population than the fmal results (see the preliminary results in: Shafigh 1992, 8). On that basis, the rural-urban migration would be more than 340 thousand a year. Another reason is the sharp drop of natural population growth of Iran from 3.92 percent a year to 2.46 percent (see Table 3.2). Finally, this figure is the result of a broad sample survey, not census, and consequently, susceptible to more errors.  41  46 expected.  Indeed, the annual average for the fifteen years is more than 306 thousand  migrants a year, unprecedentedly high. The other point to be noted is that the decrease of the rural migrants’ component in the urban population growth is partly due to the sudden jump in natural population growth (from 2.71 percent to 3.91 percent a year) after the Revolution. Moreover, the ever-increasing base figure, i.e., urban population, which from 1981 surpassed 50 percent of the country’s population, relatively, tends to reduce the inmigration portion.  47 the data Bearing in mind the predominance of the rural-urban type of migration in Iran, on inter-provincial migration also assists us to comprehend region-specific movements. Briefly, over 50 percent of the inter-provincial migrations were destined for the Central Province, including the Capital, during 1956-1976  (Tale’ et al. 1978, 36).  following decade, the same figure dropped to 42 percent (SCI 1990a,  7).48  In the  According to  the official statistics, Tehran’s share of the urban population and logically its share of urban immigration has been decreasing in the last decade. Generally, the dominant trend can be distinguished as one of rural emigration from the western provinces to the thriving central , 49 and coastal provinces  because of the administrative centralization, investment  46  Regarding the fall in rural out-migrants’ number, the prolonged economic crisis of urban centers (during and after the Revolution and in the War period) should be considered as a cardinal deterrent of rural-urban migration. Similarly, a study in Peru indicated that the slow-down of rural out-migration was due to the urban economic crisis and the importance of extended family in the absence of social security for the villagers (Osterling 1988, 173). There is no national census or other reliable source of data on the share of rural-urban migration in the total internal migration in Iran. A well-red study suggests a 90 percent share in the 1956-66 period (Bharier 1972, 58). Probably the most accurate data belongs to a national survey conducted by the Statistical Center of Iran in 1972 which shows 53.5 percent of all internal migrations have been of the rural-urban type. Another pilot study by the Center in 1974 found 66 percent for the same figure (quoted inDanesh 1987, 47). 48  On the contrary, some earlier studies indicated that the share of the Central Province & Tebran raised to an unprecedented two thirds of the country’s internal migration in the first four years after the Revolution (Alizadeh & Kazerooni 1984, 38). However, this trend assuredly subsided in the following years to the benefit of other regional cities. Aside from the impact of the war-damaged zones, the principal provincial out-migration occurred in: East-Azarbaijan, Hamadan, Kurdestan, Zanjan and Lorestan in the western part of Iran. The main  42  concentration, and foreign trade ports in their major cities. In addition, the higher density of western provinces and their higher ratio of rural population as compared to the rest of the country, denote their potential for out-migration.  Although the share of natural increase in the urban population growth of Iran has been growing faster than the share of net rural in-migrants, one cannot deny the decisive effect of such a migration in the last decades. Even the present secondary effect of in-migration is more evident should we take into account that a substantial proportion of natural increase is made up of migrant’s reproduction.  3.1.3 Settlement Pattern Having described the flow of migration, its effect on the settlement pattern will now be explained. Although the changing pattern of settlement may be partly attributed to other types of migration and to differential natural population growth, hitherto, it has been strongly shaped by the decisive share of rural-urban migration.  Changes in the urban component of the settlement pattern are illustrated in Table 3.4. The evident variance in the average annual growth rate of the four population categories in the Table indicates a higher growth rate for the large cities. This is reflected in the perpetual fall of the small cities’ share of the urban population, as opposed to the ever-increasing share of large cities. Nevertheless, the primacy index (i.e., the percentage of the primate city’s population to the total urban population) of Tehran has decreased from 30.5 percent in 1966 to 28.6 percent and 22.5 percent in 1976 and 1986 respectively (Amirahmadi & Kiafar 1987, 171; SCI 1991, 43). In other words, the polarization process, especially after  provinces for receiving migrants in the last decade were: Tehran,Isfahan, Hormozgan, Bushebr and Semnan in the central and southern part of Iran (SCI 1991,48; Zanjani 1991, 210).  43  the Revolution, has not been confined to the Capital, but has spread to other large cities as well. Table 3.4: Changes in the Size Distribution of Urban Centers in Iran: 1956-1986 (percent) * Share of Urban Population Average Annual Growth Rate in the Census Years______ of Population Population Ranges 1956 1966 1976 1986 1956-66 1966-76 1976-86 of Urban Centers Lessthan25,000 25,000 to 99,999 100,000to499,999 500,000 & more  25.8 23.3 25.7 25.2  20.2 21.9 30.1 27.8  17.8 19.4 22.1 40.7  12.7 20.3 22.3 44.7  2.48 4.39 6.71 6.04  3.64 3.64 1.72 9.03  1.83 5.95 5.51 6.39  The definition of a city in 1956 to 1976 censuses was: a population center with 5 thousand or more population and all “Markaz-e-Shahrestan” (i.e., township center) jurisdictional centers, regardless of their population size. The city definition for the 1986 census was changed to: a population center with town hail or municipality, regardless of its population size. (SCI 1991, 32) Source: Compiled and modified from: SCI 1991, 42; Alizadeh & Kazerooni 1984, 69; Madjd-abadi 1991, 169-202; Sharbatoghlie 1991, 70. *  Table 3.5: Changes in the Size Distribution of Rural Settlements in Iran: 1956-1986 (percent) Share of Rural Population Average Annual Growth in the Census Years______ Rate of Population Population Ranges 1956 1966 1976 1986 1956-66 1966-76 1976-86 of Rural Centers 1 to 99 100 to 499  06.8 44.5  07.9 42.1  06.4 36.8  04.7 27.6  +3.35 +1.28  -0.76 -0.02  -0.69 -0.41  23.0 23.7 24.7 22.2 +2.16 +1.71 +1.44 500to999 18.9 19.7 22.9 26.3 +2.23 +2.86 1000to2499 +3.96 06.8 06.7 09.2 19.2 +1.68 +4.63 Over 2500 10.29 Note: Up to the 1976 census, the over 2500’ population range of villages did not include villages with over 5000 inhabitants. In the 1986 national census, however, 194 villages with more than 5000 inhabitants were included. See the footnote of Table 3 for the definition of city in each census. Source: Compiled and modified from: SCI 1985, 58: SCI 1991, 42; Vadi’ei 1974, 134; Sharbatoghlie 1991, 69.  Table 3.5 illustrates the changes in the rural component of the settlement pattern. Here, it can be elicited that villages with a population of under five hundred have been facing  44  severe out-migration. ° In contrast, villages with over one thousand inhabitants have been 5 receiving migrants. The abrupt increase in the growth of villages with a population of over 2500 during the last decade (i.e., from 4.63 percent to 10.29 percent) is quite ’ 5 striking.  A ftirther fact that should be taken into consideration is the accelerating  number of depopulated villages, from 15,925 to 38,571 during 19661986.52  In short,  these changes in the settlement pattern manifest the recurring trend of rural out-migration from smaller to larger centers, leading to polarization.  3.2 State Intervention in Population Distribution Before discussing the case of Iran, it is worth paying attention to how development theories affect South migration in general.  During the recent century, the process of so-called modern development has accrued in an uneven fashion (Cook 1983, 20), manifested predominantly in a spatial concentration of people and activities in the South countries (Kuklinsky 1972; Frank 1967; Friedmann 1966). The unequal distribution of people and activities is partly explained by the varied endowment of natural resources in the South and by the occurrence of natural phenomena affecting their availability -e.g., drought, earthquake, flood (Barke & O’Hare 1984). Partly the unequal distribution sterns from historical and contemporary human-made events, whether spontaneous or induced -e.g., demographic changes, the state’s policies 50  The small rural centers have significant contribution to the country’s agricultural production. As the minister of RCC has acknowledged more than one third of arable lands of Iran are located close to villages with less than 250 inhabitants that are threatened by rural depopulation (cited in: Schirazi 1993, 314). The number and the population of these villages are roughly 62 percent and 20 percent of all the rural settlements, respectively (SCI 1990C). 51  To my knowledge, the majority of the growing villages are on the outskirts of cities in a transitional phase. 52  These depopulated settlements may have formerly contained one or more persons or may have been established as non-resident farms. Thereby, it cannot necessarily be elicited as abandonment of villages or agricultural lands. This point will be elaborated in the last chapter.  45  (Mabogunje 1989; Morrill & Dormitzer 1979). Reciprocally, the population distribution affects the spatial variation of previously causal factors: resources and human interventions (see Figure 3.1). Figure 3.1: Internal Forces of Uneven Development & Resultant Population Mobility  Development Theory  Sectoral Priorities  State Intervention >  Spatial Disparities  I  —  Differences in Natural Resources  Natural Calamities  ii  Source: Author  [  Increasingly the impact of the state’s intervention policies, specifically by scarce resource allocations, permissions and land use control, determine the trend of distribution of people and activities in the South (Stohr 1981a, 42; Singh 1991, 42).  As a matter of fact,  numerous decolonized states after the second World War, in addition to other pre industrial countries aspired to catch up with the modern industrialized countries by state intervention (Friedmann & Weaver 1980, 108-110). National development planning were in vogue and considered to be the imperative method of conscious intervention to overcome backwardness (Hulme & Turner 1990, 100).  In so doing, the influential  theories of development have played a decisive role as a mindset for the state’s policy  46  planning. Consequently, it is inferred that the political priorities manifested in through economic strategies are the root cause; rural-urban migration is the observed malady” (Stark 1991, 16).  3.2.1 The Need of Policy Intervention  Whichever development is perceived, it appears the ongoing rural-urban migration in most  of the South countries and especially in Iran is detrimental in terms of socio-economic and environmental sustainability (Roseland 1992, 22-6; White & Whitney 1992; WCED 1987, Ch.9).  In this sense, migration is the reflection of premature urbanization and mal  development (Amin 1990) and exacerbates the already acute problems, such as: environmental degradation, hornelessness, overloaded infrastructure, unemployment and unproductive employment, cultural alienation and social disintegration, particularly in large cities (Qadeer 1983; Drakakis-Smith 1987).  In fact, despite broad criticism, modernization theory has been widely practiced in the South. It has, though, been subjected to repetitive revisions. As already explained, this approach has brought about huge rural-urban migration streams. The resultant population distribution is highly unsatisfactory as expressed overwhelmingly by South governments in 1978 and 1983 United Nations’ surveys  Seemingly, the governments consider their  population distribution patterns as a deterrent to offsetting present disparities. Thus, they tend to intervene in the trend of population distribution for facilitating the process of balanced development.  In a 1978 survey, 121 South Countries out of 136 respondents perceived the spatial distribution of their population as being unacceptable (U.N. 1982, 19). Specifically, 116 out of 122 had devised policies to reduce the rural-urban migration (Parnwell 1993, 130). And in 1983, “the governments of 123 developing nations out of 126 respondents indicated that they considered their population distribution to be partly or wholly inappropriate; the remaining three were relatively small island states” (Oberai 1987, 13).  47  A further point worthy to be mentioned is that while the development theories summarized above present contrasting views about the future of ongoing migration, almost all agree on the predicament of South urbanization, and by and large, on the failure of laissez-faire policies at the present time (Wood 1982, 304).  Hence, all-in-all, the theoretical  interpretations, governments’ perceptions and factual evidence, uphold the necessity of government intervention in the status quo.  3.2.2 Policy Classification In order to investigate policy interventions by the state, it is important to understand its variety of approaches.  A wide spectrum of socio-economic policies directly and/or  indirectly influence the population distribution between rural and urban areas (Todaro 1976, 3).  However, for the scope of this study, those policies which are designed  explicitly to alter rural-urban migration behavior are discussed.  Whereas most of the rural-urban migration streams in the South are toward metropolitan cities, the definite objective of policies may be traced in one or more of the following categories: 1-  Restrain the migration stream at the rural source or at the cities’ gates.  2-  Divert the stream to other destinations, such as rural frontier areas, small or intermediate cities.  3-  Return the migrants from cities to their rural homes.  4-  Accommodate the migrants at their destination. (Simmons et al. 1977, 103; Skeldon 1990, 193; Oberai 1983, 11)  Needless to say, any of the above categories may be carried out through direct or indirect and voluntary or compulsory leverages.  48  The underlying policy approach to select each of the above objectives is varied. A useful classification of policy approaches, as Stohr put it, is to distinguish between adaptive and normative interventions.  The two approaches may be seen as reactive and proactive  responses, respectively. The objective of adaptive intervention is “to adapt population distribution to the consequences of functional societal change, e.g., of industrialization, modernization, [and] spontaneous growth centers” (Stohr 198 la, 42). The objective of normative intervention is “to use government intervention as an instrument to influence the rate and direction of functional societal change” (Ibid.).  The adaptive policies are mostly implemented in cities and in their fringes, and are aimed at responding to the resultant needs of in-migrants -i.e., the accommodating objective. This kind of policy is more ‘curative’ than ‘preventive’. The normative policies are carried out in both urban and rural areas, often relying on negative leverages in cities, e.g., bans on in-migration, and on positive ones in villages, e.g., rural development programs.  To put it more precisely, in terms of policy leverage, normative policy can be divided into two groups: negative and positive. restrictive (Skeldon 1990, 193).  Negative policies are mainly discouraging and/or  Conversely, positive policies place the emphasis on  encouragement and/or facilitation.  Probably, the most exercised positive-normative policies to control South rural-urban migration are rural development programs. A substantial component of these programs is social and physical infrastructure provision (e.g., the case of Pakistan in: Aziz 1988, 113).  From the aspect of geographical points of stress, the rural-urban migration policies may be classified into three groups. The first group underlines the push forces in the rural origin,  49  consequently devising rural development. The second group accentuates the pull forces in the urban destination, mostly using disincentives such as high taxation.  The third group highlights the origin-destination relationship and attempts to find solutions for a structural balance between rural and urban areas.  These policies are often of an  economic nature, in Todaro’s words, and aim at dismantling urban-biased policies (Todaro 1981, xiii), such as changing the terms of trade or wage regulations to the benefit of the rural sectors.  In the first policy group, as already mentioned, infrastructure provision in rural areas is crucial to enhance socio-economic performance.  Similarly, for the third group, the  existence of infrastructure in rural areas is an integral part of attempts to mitigate disparities.  Considering the arduous and long-term task of rural-urban balanced  development, and despite its questionable outcomes, the practicality and visibility of rural infrastructure-building by governments make it virtually common.  Table 3.6 demonstrates the geographical points of stress for adaptive/normative policy classification. It is needless to say that the above policy groups are not mutually exclusive and could be combined as well. Table 3.6: Classification of Migration Policies & Their Emphases  Policy Emphasis on: Origin Push Destination Pull Structural Relations Source: Author  Adaptive Policies  Normative/ Positive Policies  Normative! Negative Policies  0  0 0  0  50  3.2.3 Significance of the Government’s Role in Iran The need for state intervention in prevalent trends of population redistribution and different policy approaches were discussed above. Now the authority of the Iranian state is underscored. The significance of the State role emanates from two specific bases. First, since the movement of nationalization of the oil industry in 1953 and the clear recognition in the constitution of the Islamic Republic of Iran in 1979, the State has enjoyed exclusive rights over oil revenues, among other things. In an oil-dependent country with a weak private sector [at least in terms of willingness for productive investments] the momentous investment and expenditure of the oil-rich State is clearly decisive 54 (Karshenas 1990a, 239).  Second, taking into consideration the historic tradition of centralization in the Middle East, the State of Iran has the constitutional right to intervene in the market, and it authoritatively exercises its power, consistent with Islamic jurisprudence. 55 The political system of Iran is highly centralized and there is no local government whatsoever. On this ground, the importance of investigating the State’s policies as a determinant of migration in Iran becomes more evident. 56  The State’s share of investment in Iran during 1973-78 was 66.1 percent of the total as compared to 32.4 percent in 1963 (Ehteshami 1993, p. 21 5). The following figures of the peak years, before and after the Islamic Revolution, depict the importance of government income from oil exports: In 1975, oil revenues were 20.9 billion dollars, accounted for 96.9 percent of fran’s exports income and 50.6 percent of GDP (SCI 1981, 798 & 1001; CBI 1975, 22) In 1983, oil revenues amounted to 20.2 billion dollars, about 98 percent of the exports’ income. The government share of gross domestic capital fonnation was more than 46 percent (SCI 1985, 515 & 699 & 785). It is worth mentioning that due to the sharp fall of oil prices in the international market, in 1986, han’s revenues dropped to 5.9 billion dollars (SCI 1991,410). -  -  -  In part of article 44 of the Constitution it has been stated: The state sector is to include all largescale industries, foreign trade, major minerals, banking, insurance, power generation, dams and largescale irrigation networks, radio and television, post, telegraph and telephone services, aviation, shipping, roads, railroads and the like; all these will be publicly owned and administered by the State.” (Constitution of the Islamic... 1990, 44) “...  56  In terms of employment, the state sector comprises roughly one third of all the jobs (SCI I 992a, 74).  51  3.3 A Brief History of Rural-Urban Migration in Iran Migration cannot be isolated from its socio-economic context. In this section, I do not intend to provide a historical account, but rather to delineate those factors that determined rural-urban migration flows.  These determinants in the last three decades or so are  reviewed to portray migration patterns, by and large as inevitable spatial manifestation of adopted policies by the government. Nonetheless, migration streams, in return, have been affecting policy making through their pressing needs and backlashes.  In terms of driving forces, three distinct stages can be distinguished in the era of migration intensification in Iran, as follows:  3.3.1 Stage One: Modernization Approach Rural-urban migration is not a recent phenomenon in Iran (Khosravi 1978, 99). Albeit, it gained momentum in the mid 1950’s during the reign of the Shah, 57 when the prevailing strategy for development, like that of many other Third World states, was the modernization approach (Bharier 1971, 27).58 Consecutive National Development Plans placed enormous emphasis on urban-based infrastructure and industrialization. 59  The  5  Interesting to know, between 1900 and 1956, according to Bharier’s calculations, city-ward migration between 685 and 728 thousands, or an average of 13 thousand annually (Bharier 1972, 57). 58 The inception of modernization can be traced back to the establishment of the first modem school (i.e., Dar-ol-Fonoon school) in 1851 (Menasliri 1992, 5) or the Reza Khan’s coup d’etat in 1921 (Karshenas 1990a, Ch.3). The era and its pervasive influence on the country is still seen in the stated  was  period. For example, during the years of the Second Development Plan (1955-1962), 49 percent of the whole budget was allocated to industrial development and to major physical infrastructures, such as airports, seaports, railroads and highways. Also, in 1965, government investment in the secondary sector (i.e., industry, mining and construction) and tertiary sector (i.e., all services and infrastructures) were 8.4 and 5.2 times that in the primary sector. The same figures changed to 11,8 and 5.7 respectively, in 1976, during the second stage of rural-urban migration (calculated on the basis of data in: MEB 1990). An alternative study during 1962-77 by Karshenas (1990b) argues for the inefficiency of resources allocated to the agriculture sector, rather than inadequacy.  52  national economy conspicuously disintegrated into dualistic sectors, namely the modern sector with its spatial preference for principal cities, versus the traditional sector in the rest of the country, basically in rural areas.  In the early 60’s, the government embarked on an extensive land tenure reform which developed capitalist relationships in rural areas and thus dissipated the traditional ties of landlords and peasants (Lahsaeizadeh 1993, 305; Hakimian 1990, 79).  But the quasi-  feudal institution and the subsistence economy in rural areas were not substituted by an 60 New landowners, on average, did not receive enough land organized market economy. and lacked supportive services. Disenchanted by this land reform, many farmers joined the already-on-the-move landless villagers toward large cities, where unskilled labour was ’ 6 hailed for construction work.  In light of the prioritized industrialization and incomplete land reform on one hand, and 62 a large gradual increase of oil revenues and related urban-biased policies on the other; portion of the agricultural sector, and consequently of rural areas, was marginalized, if not  This analysis about the consequences of Land Reform in Iran has been docwnented in many studies (Cf. Misra 1978, 153; Hooglund 1982; Afshar 1985; Danesh 1987; McLachlan 1988; Amid 1990; also in  60  Persian: Ra’is-dana 1979). It must be mentioned that while there is a vast literature on Iranian land reform, two contrasting analyses of its consequences are observed. Mainstream criticism emphasizes the overwhelming negative impacts on separating peasants from the land, intensification of rural out-migration and total failure to benefit the peasants (for example see: Adibi 1977, 176; Katouzian 1981, 308; Hooglund 1982, Ch.6; Azkia 1986, 124; Danesh 1987, 81; Lahsaeizadeh 1987; Farazmand 1989, 108; Araghi 1989, 1049; Vosoughi 1990, 193). The alternative analysis poses that the land reform has been “highly advantageous” to the majority of peasants and has not driven major migration streams (Majd 1987, 847; Majd 1989, 1053; Majd 1991, 76; Majd 1992, 455). Some scholars held a centrist view like Ashraf (1991). 61  Iran oil income rose from 11 percent of the GNP in 1959, to 18 percent in 1969, while its value in fixed price became almost 3.5 times greater. In the second stage of migration, described in the following section, oil revenues jwnped to 50.6 percent of the GNP in 1973 and dropped to 34.7 percent in 1978 (Katouzian 1981, 257; MEB 1990, 1-2).  62  53  63 At the same time, large cities, which were deemed to perform as growth poles, isolated. became the focus of government infusion of modernization. 64  3.3.2 Stage Two: Oil-Driven Urbanization  The second stage in the accelerated pace of rural-urban migration started in the early 1970’s, and can primarily be related to the oil boom.  In fact, Iran’s per barrel oil revenue  skyrocketed from $0.98 to $9.49 between January 1971 and January 1974.65  As a result,  the petroleum revenues of the government in 1974 were about 18 times that of 1970 while the GNP increased at the rate of 33.3 percent per annum in the 1970 -1977 period (Ghods  1989, 202).66 This boost in revenue provided the government with a capital surplus with which to hasten and expand the aforementioned modernization approach, using more  capital-intensive technologies which had smaller labour needs (Katouzian 1978, 367). A considerable leap in the amount of investments in construction and an enormous growth of urban services took place in these years, 67 both offering a great number of jobs to unskilled migrants 68 (Kamrava 1990, 107; Hakimian 1990, 124; Kielstra 1987, 218). 63  On the narrow-minded policies in this period, refer to Jacobs’ sociological research, which shows that the Shah had substituted linear economic growth for a genuine development (Jacobs 1966, Cli. 12). M. Vosoughi suggests two different sub-stages within the first stage, that is prior to the Land Reform and after it. He characterizes the first sub-stage with dominance of rural-rural and seasonal migration of the landless individuals as opposed to the expansion of rural-urban and permanent migration of rural families (Vosoughi 1990, 46-49). For a rare study, briefly mentioning rural-rural migration in 1960’s 64  refer to Safi-nejad (1976, 96) 65  Interesting to know that the price rise in “per barrel” oil of Iran during 1960-70 was only $0.06 (Pesaran 1985, 24 & 31), compared to more than $36 during 1970-1980 (Amirahmadi 1990, 71).  66  The national income at the 1974 fixed price grew astonishingly on average by 21 percent per annum over the period 1970-1977 (calculated on the basis of: MEB 1990, 4). -  -  67  From 1970 to 76, the investment in the construction sector and the service sector (excluding transportation and communication) increased 3.2 and 5.5 times respectively (calculated on the basis of data in: MEB 1990, 10). In the same years, the share of domestic fixed capital fonnation, in construction at the 1973 fixed price, grew from 13.3 percent to 31.2 percent, while its volume became 4.2 times larger (SCI 1982, 1093). The construction sector’s share in the total employment of the country almost doubled: from 7.2 percent in 1966 to 13.5 percent in 1976 (Hakimnian 1988, 223). 68  To underscore the importance of the booming construction sector, a glance at “prestigious” urban projects in the Capital is convincing: Shahestan-e-Pahlavi (a huge new city center), Shahrak-e-Gharb (a new city adjacent to Teliran), many satellite “bedroom” towns, high-rise complexes and freeway  54  At the same time, urban demands for consumer goods rose sharply by virtue of the higher purchasing power of the urban population. In 1974, the government started to subsidize staple goods for the first time (Afshar 1985, 70). Really, the extravagant imports bought 69 For instance, food stuff imported with petro-dollars benefited the urban dwellers most. with an open-door agricultural policy (Farazamand 1989, 155), inhibited the price rise of . Otherwise, the high demand could have stimulated balanced 70 domestic food production growth in the agricultural sector (Katouzian 1978, 350).  Thereby, the urban-driven  sectors and rural-based ones did not accrete simultaneously, embodied as they were in uneven spatial development of urban and rural areas.  Another government policy which adversely affected rural emigration was the expansion of agro-businesses, farm corporations, and in essence, spatially concentrated investments in pursuit of “efficiency and productivity”  71  Agricultural poles received easy credits for  72 (Aresvik 1976, 124; Atash 1988, labour-saving technologies and cash-crop cultivation  construction, an underground mass-transit system, an Olympic sport complex, etc. The construction and service activities account for over 70 percent of the employment increase in urban areas between 1973 and 1978 (Nattagh 1986, 63). 69  In this respect, the open-door policy and lavish imports brought about a modern consumption pattern far ahead of the capabilities of the country’s economic sectors to meet. The economy, in terms of production (e.g., because of needed intermediate goods for the performance of assembling /processed manufacturing) and consumption (e.g., the consistently increasing imports of food stuff and other staple goods) was addicted to petro-dollars and became quasi-mono-base (Jazayeri 1988, 186). 70  Some noteworthy figures are: during 1970-78, the demand for food grew 10 percent yearly comparing to 4.6 percent growth in domestic production, and per capita yearly meat consumption rose from 10.5 Kg. to 24.3 Kg. (Majd 1991a, 74), 71  Government policy was shifted in favor of large-scale production after the failure of the co-operative program in the wake of the 1962 land reform (Najafi 1991, 327; Najmabadi 1987). In spite of heavy supports, the contribution of the modem sector to agricultural output accounted for only 4.1 percent in 1975 (McLachlan 1988, 136), while the modem sector gained many times of this share from state-run agricultural banlcs (Amid 1990, 113-119; Ghahreman 1982, 135-154). 72  As a matter of fact, the application of machinery in agriculture ever since modernization’s approach, has been widely diffused throughout the entire country. The number of tractors in 1977 had increased nearly nine times since 1962 (Amid 1990, 122). For a complete analysis see: Okazaki (1985).  55  103; Salehi-Isfahani 1989, 374).  Overall, however, the state was reluctant to invest  proportionately in the agricultural sector (Aresvik 1976, 239).  Again, these changes  fostered new push factors for agricultural workers 73 as well as peasant land-owners who had been assimilated into these large-scale, bureaucratic establishments. 74  Overall, the oil-induced “boom” was channeled mainly to major urban centers and to a 75 The inequalities between rural and urban areas, lesser extent to a few agricultural poles. as well as within the rural areas deteriorated (Amid 1990, 139; Kamiar 1988, 333; Nattagh 1986, 61; Amirahmadi  1986, 501).  However, a tendency toward conspicuous  consumption and the new urban life-style was ubiquitous. 76 Accordingly, the bright light of metropolitan areas, as well as cash and higher wages for unskilled workers, 77 attracted an unprecedented flow of rural migrants to urban areas. 78  At the same time, continuing  rapid population growth was the decisive force for all these displacements of the villagers Many studies relevant to the cause of rural-urban migration in this stage clearly conclude that the rural push factors make for a better explanation than do urban pull factors for the pattern of migration in Iran (Mohtadi 1990, 842; Rarnin 1988, 33; Danesh 1987, 49; Kazemi 1980) However, some held to the dominance of pull factors (Hakamian 1988, 218) As it is discussed in footnote 122, push and pull segregation is not clear-cut reasoning. Regarding the negative impact of rural bureaucratization in the Shah’s regime, especially after the oil boom, refer to Farazmand (1991, 551-565). On the neglect of the rural areas at large and the damaging effect of the oil boom on Iranian agriculture, most scholars agree (for example: Katouzian 1978; Hooglund 1982; McLachlan 1988). However, some scholars recently have challenged the view and asserted that the faster growth of urban sectors, not the neglect of agriculture, has been the cause (Majd 1987 & 1988 &1991b; Hakimian 1988; Karshenas 1990b). 76  A cardinal catalyst of this tendency, undoubtedly, was the Westernization campaign of the mass media before the 1979 revolution. (Kamrava 1990, 127) For the necessity to gain cash as a supplementary means of support even for peasant land-owners see: Hooglund (1982, 93-94), Azkia (1986, 124),. Also, for the high involvement of peasants in urban constructions which were calculated at 15 percent of the urban labour force and more than 1 million from 1975 to 1977, see: Hoogland (1982, 116). 78  This wave of migration is seen in the latest national census data which shows a sudden increase in life-time migrants to urban areas who had been migrated nineteen years prior to the census (i.e., 1967SCI l993a, 6). The surge of city-ward migration is also obvious in Hamadan Province. Here the number of life-time migrants to urban areas who arrived between 1967 and 1976 is 3.2 times the same number for 1957 to 1966 (SCI 1992b, 6)  56  (Ashraf 1991, 289). The general perception of migrants was simple: to partake of the oil wealth, one must go to the core cities.  3.3.3 Stage Three: Inconsistency and Turbulent Movements The modernization approach was disrupted by the Islamic  Revolution in February of  1979. Suffice it to say that animosity towards urbanization as a phenomenon which was deemed to create dependency, to usurp oil windfalls to the benefit of urban elites, and to ruin cultural values, prevailed among the leading politicians (Etemad 1984). In this view, even industrialization, due to its structural dependence on oil revenues, was rejected and rural emigration was condemned and seen as the cause leading to agricultural decline and urban predicaments. The opposition to the Shah’s plans and the “back to the roots” spirit left virtually no room for a modernization approach.  Policy-making attitudes were  79 fundamentally changed at this time.  Two of the most emphatic slogans of the Revolution were independence and social justice. To realize these, the new regime vowed to achieve self-sufficiency in food stuffs 80 and also to eradicate deprivation.  Contrary to the toppled regime, the agricultural  sector was proclaimed to be the pivot of national development.  The new government  called for small-scale projects and increased loans for small farmers (Weinbaum 1982, 44). In recent years, many of these views, e.g., the anti-urbanization attitudes or underrating of industrialization, have been fading away among the politicians. During the years of the First Five Year Plan (1989-1993) mainstream economic policy has been in the direction of so-called restructuring, adjustment and privatization policies (see Askari 1994). Due to these shifts, I tend to assert that a fourth stage of population movements has been going on since 1989, the beginning year of the First Plan; though, it is too soon to analyze this. ‘  80  In retrospect, increasing dependency on the imports of agricultural goods followed the 1973 oil boom.  The revolutionary leadership stressed self-sufficiency as a requisite for national sovereignty and as bargaining power in selling oil at a fair price. Iranian agricultural imports, however, did not cease after the Revolution and reached more than 8 million tons of grain and an estimated $3.8 billion in agricultural imports in 1989. These figures in 1976 were 2.7 and 1.3 respectively (SCI 1990a, 392-393; SCI 1981, .997; McLachlan 1988, 232). To convert Rial figures into U.S. Dollars, the somewhat official exchange rate of $1 = 70 Ris. has been used.  57  In the same course, the Rural Construction Crusade was hailed as a revolutionary institution (it later became a ministry) all over the country. Effusive youths volunteered to work in rural areas to build basic services and to help agricultural production in a nonbureaucratic form (Sharbatoghlie 1991, 75).  Furthermore, a new land reform program was launched, mainly aimed at barren lands left by owner absentees or fugitive landlords, but was not continued in order to include a broader range of lands (Ashraf 1991, 306; Farazmand 1991,  560).81  The agricultural  policies of the State lacked consistency, particularly in regard to ownership, and consequently, negatively affected the incentives for new investments on farmlands (McLachlan 1988, 221; Schirazi 1993). Besides, for those agricultural outputs which used to be purchased exclusively by the government, the prices increased well but the market determined prices of inputs, and thwarted their positive effects on agricultural households’ incomes (Alizadeh & Kazerooni 1984, 24). Many inputs for the farmers were subsidized by the government, but only for a part of their needs, and also the amount of subsidy has been dependent upon oil revenues -as the cardinal determinant of the State income- with immoderate fluctuations (Mojtahed & Esfahani 1989; Najafi 1991).  In this stage, from 1979 on, some momentous events seem to have had more impact on population movements than have policy-related issues, including: The drastic decline of oil income since the Revolution (except in 1982 and 1983): Owing to this, the dependent structure of manufacturing (e.g., dependence on importing intermediate products) was paralyzed, which caused a prolonged recession  81  Taking into account that only 1.7 million hectares of agricultural lands of the total 18 millions were subject to the reform up to 1987 (PBO 1 990a, 4/1 & 4/2), it means only about 3 percent of the total (Ashraf 1991, 306).  58  for the already contracting economy.  Correspondingly, cities could not be as  attractive as before with respect to economic opportunities. 82 •  The imposed war by Iraq: It is estimated that about 1.5 million war-inflicted people took refuge in urban areas, most of whom are still living there (Alizadeh & Kazerooni 1984, 35).  •  Domestic um-est and strife in the western region: In a short period of time (from 1979 to 1982) many villagers (estimated at as many as 200 thousand), fled from the insecure rural areas to the provincial cities in three western provinces. 83  •  The Afghanistan Civil war and Iraqi Refi.igees:  More than 2 million refugees have  entered Iran, approximately 50 percent of whom have resided in city centers. 84 •  International Emigration: As of 1978, it is estimated that at least 2 million Iranians, mostly residents of large urban centers, have emigrated abroad (Danesh 1987, 30).  Another crucial factor, contributing to the migration in this stage, was the lack of birth control policy or, as some may interpret it, an implicit encouragement of larger family size and young marriage. 85  In effect, the average annual growth rate of the population,  dissimilarly to the two previous decades, well exceeded 3 percent. 86 More than 22 million 82  Oilrevenues from export dropped from $20.5 million in 1977 to $11.6 million in 1980 and then went up to $20.5 in 1983 and fell again to $5.8 million in 1986 (Arnirahmarh 1990, 225; SCI 1991, 410). 83  The 200 thousand figure was roughly calculated as the difference between the 1986 urban population and the expected rural population in 1986 (i.e., the extrapolation 1966-76 trend) of the pertinent provinces, subtracting the war refugees. 84  The 1986 census shows that almost 840 thousand Afghanis and Iraqis are residing in Iran, which is certainly a miscount (see: Zanjani 1991, 179). Regarding this, many official speakers have  acknowledged that more than three million were living in the country (Kayhan-e-Havai 1992, 4). Or recently, only for Afghams, the United Nations High Commissioner for Refugees announced that: “By mid-1993, the number of Afghans in Iran stood at 2.4 million.” UNHCR 1993, 31) The estimation of refugees in Iran for 1992 was 4.2 million (by Population Action) which distinctly ranked first in the world (Vancouver Sun 1994, A14). 85  At least in the first five years of the Revolution, the principle of the priority of most populated families governed the State laws and by-laws for subsidized and short-of-demand allocations (e.g., Lowcost housing and urban land cession). 86 The 1986 census showed the average annual growth rate of population during 1976-86 as 3.9 percent. Bearing in mind that a huge number of international emigrants entered the country, population studies  59  were added to the 1976 population, an average increase of 1.47 million a year till 1991. On the assumption of a natural increase at 3.2 percent per annum, in 1991, 11.5 million should have been added to the 17.8 million rural population of 1976 (the last census year under the previous regime), a significant 64.6 percent growth.  Essentially, the major  driving force of rural population for decades. i.e., rapid population growth, has remained intact.  Yet, in the last five years, the govermnent has launched a broad campaign on  family planning programs. It seems that these efforts effectively reduced the population growth rate to well below 3 percent. 87  Despite salient changes in rural and agricultural policies, ironically, the practice of ruralurban migration persisted (Schirazi 1993, 312). 88 contradictory policies on the urban side  Part of this has been rooted in  which still provide better economic  opportunities in the cities and eventually, perpetuate urban-rural disparities (Sharbatoghlie 1991, 104). For this, the unhealthy growth of employment in unproductive urban services is mostly responsible. Indeed, the highest value added per capita and the greatest growth in job numbers, have been in the urban service sector, evidently booming the urban pull. 89  The accelerated trend of rural out-migration, in spite of policy shifts, has continued (Adibi 1989, 279). The rural-urban gulf in employment and income which was slightly narrowed percent as the right figure for the natural increase (Zandjani 1991, 36 & 156; Sharbatoghlie 1991, 204).  suggested 3.23 87  The latest national population survey in 1991 shows the average annual growth rate of the population as2.46 percent for the 1986-9 1 period which seems under-counted. Some Iranian population experts estimate the natural growth rate of that period to be around 2.9 percent per annum which was the target of the First Five Year Plan for the year 1993 as well (Assadpour 1992). 88 An example is a top official declaration of the urban land grants for the deprived migrants at the beginning of the Revolution (Adibi 1989, 270). No need to say that it was not put into effect. 89  Another widespread urban policy has been that of the projects similar to site-service schemes (Arnadeh Sazi-e-Zamin) for low and middle income housing in majority of cities and towns. So far, it seems that the effect on the increase of urban land ownership and to some extent decentralization of urban population has been positive. However, one more time it sends the message of hope and fulfillment in the cities to the blighted villagers.  60  (Atash 1988, 99; Behdad 1989, 351; Sharbatoghlie 1991, 209; Schirazi 1993, 309), is still quite wide (Amirahmadi & Atash 1987, 155; Shafigh et al. 1992).  However, the  migration pattern has been changed to the benefit of better distribution of migrants in cities (as opposed to heavy concentration in the Capital). But once again the incentive for villagers to emigrate is strong. The perception could be simplified as this: Prosperity in the city is plausible but not likely in the village. Even to gain the promised fruits of the Revolution, wisdom calls for being close to the center of power: the city!  3.4 Post-Revolutionary Migration Policies  As briefly mentioned in the preceding section, post-revolutionary governments in Iran, have used numerous policies in order to reduce the outflow of migrants from rural areas. 90 In retrospect, the policy instruments prescribed in Iran have been more normative, though encouraging and directive, rather than being coercive and indicative. They mostly rest on the normative and restrain category of the previously mentioned classification, with very little attention to the other categories (i.e., divert, return, accommodate).  However,  especially at the outset of the Revolution, some adaptive policies (e.g., land cession to the city squatters, providing infrastructure for shanty towns) were made. ’ 9  An overview of the policies carried out after the 1979 Revolution is presented in the following categories: a-  The restrictive policies in urban centers, which attempted to discourage migrants from staying at the destination (i.e., preventing in-migration and returning them to  90  To realize the priority which was given to solving migration by the government, it suffices to quote  the then-successor of the spiritual leader of the Islamic Republic of Iran who said: “After the War [with Iraq] our most important problem is migration” (Kayhan 1985, 15). 91  See section 3.2.2  for clarification of policies.  61  their rural homes). These policies are negative-normative, such as: banning peddlers or not entitling the in-migrants to rationed foods. b-  The retentive policies in rural areas, which aimed at holding back migrants at their place of origin (i.e., restraining rural out-migration).  These policies are positive-  normative, such as: providing more services or higher agricultural pricing to rural areas.  3.4.1 Restrictive Policies in Urban Centers Without spelling out the policies or making a chronological summary (which is not intended for this study), the principal government policies are explained here. First, and of the paramount importance, is the rationing system for both urban and rural areas which began after the Iraqi invasion in September 1980. The system introduced a registered certification-- a passbook, on the basis of the birth place/residence of each individual. This passbook became essential for obtaining many rights and privileges. Receiving rationed staple goods (which these days have gradually disappeared), obtaining work and business licenses (for the formal sector, excluding most of construction and other blue collar workers), registering children at school, applying for some home appliances and cars provided by the state or the cooperatives (at subsidized prices and by lottery-- a process which have been almost stopped since 1988), receiving bank loans, obtaining urban 92 all required the submission of the local property and even renting a house legally, passbook by the residents of large cities. Thus, in deciding weather or not to move to large cities, would-be migrants had to take into account the difficulties of obtaining the foregoing goods and services without a passbook for that particular city.  92  The  requirement of local passbook for purchasing a house or an urban land was removed in 1985.  62  While rationing-local passbooks were an obstacle to easy population movement, in reality rural migrants were not halted from entering the cities for a number of reasons. First, in terms of schooling, many migrants were alone or had no children of school age. Also, sharing residence with a previously villager and now city resident, or renting a room without the notarized lease contract were among their options to cope with the constraints imposed by the local passbook. Second, the availability of all rationed items in the black market at exorbitant prices (Amirahmadi 1990, 185), attracted many rural migrants to present their saved rationed rural goods in this market and/or to engage in middleman activities in large cities.  Third, many migrants could join the informal sector and by  profiteering could offset the legal restrictions of the cities.  Another cardinal policy that indirectly influenced the rate of migration to cities was the 93 In effect, the city taxes increased major cutbacks in the state subsidies for large cities. rapidly, affecting day to day expenses, especially in recent years.  Coinciding with the  prolonged economic recession of the war time and the consequent post-war years, the contracting urban economy, did not at least magnetize those would be migrants who could make a living in their villages.  In addition, some pre-Revolutionary policies for curbing city expansion were adopted and modified, namely: placement of restrictions on new industrial establishments within a 120 Km. radius of Tehran, enactment of city master plans to control physical development, conditioning of the transfer of government employees to large cities, and requirement of certain professionals to work outside large cities for a number of years (Hemmasi 1980, 224; Amirahmadi 1990, 216). Resuming the harassment of large city peddlers and other To give an idea of this trend, in 1977, the ratio of urban to rural households expenditures was 2.1 and the ratio for their incomes was 2.7. In 1987, these ratios changed to 1.64 and 1.59 respectively which implies that income disparities had lessened (Amirabmadi 1990, 200). It also indicates that the growth of urban households, expenditures was faster than their income growth, in comparison to rural areas.  63  unlicensed shopkeepers were among the other perpetual restrictive policies against rural emigrants.  3.4.2 Retentive Policies in Rural Areas  While lacking clear-cut policies from the inception of the Islamic Republic, offsetting the rural-urban disparities and meeting the particular needs of rural people (which could suitably influence villager& out-migration) were on the states agenda. To begin with, the revolutionary institutions, particularly the Rural Construction Crusade (RCC) and the Housing Foundation, made the effort to develop physical and social infrastructures in the rural areas. Despite war time shortcomings, these institutions and conventional relevant ministries, immensely expanded the following rural services: electrification, potable water pipework, road, school and public bath building, networks of health care and agricultural service centers (Sharbatoghlie 1991, 75-78). The state dominantly invested in all these projects with a less than one third contribution of local people in most cases. Overall, the physical constructions proceeded well, while the functional operation (especially in the absence of adequate skilled and higher-educated manpower and in face of the shortcomings in the government current expenditures) was questionable in many areas. 94 For example, lack of qualified health practitioners or agricultural experts with respect to the rural health care network or the agricultural service centers network were a major deficiency (Vosoughi 1990, 252) which in some extreme cases led to abandoned structures that were newly built. 95 To give an idea of the enormous efforts by the RCC for physical projects in rural areas, the following data are self-evident: In 1989, 56.6 percent of rural households had piped water, 70.8 percent had electricity, compared to oniy 14 percent and 22 percent for the same figures in 1978, the last year before the Revolution (Shafigh et al. 1992, 31-2). “Only 8,000km of village roads were built before the revolution, according to the minister, but Jahad had constructed 43,000km of village roads by the summer of 1988.” (Schirazi 1993, 307) Najafi’s observation about rural agricultural service centers is to some degree different. He maintains, at least in the Fars province, that the centers “have been successful because the farmers are not obliged any more to travel to nearby towns for required services.” Based on my own four years of research in Hamadan and north Khorasan province, this was not the case. Usually, to obtain the allocated quota of  64  In the same course of infrastructure provision, since 1985, the ambitious national program of rural renovation (called: Beh Sazi-ye Roosta) was embarked upon. The program mainly consisted of physical master plans for large villages and resultant projects like: widening and paving village roads, allocating land for mosques and other public needs, building drainage and sewage systems. The selection of villages and the programming devolved to provincial authorities. A 50 percent local contribution was a prerequisite for financing the rest of the budget for the proposed program by the government in each selected village. The compensation for damaged buildings in the village (as a result of renovation programs) has been negotiated and mediated by the revolutionary institutions and the 96 village Islamic Councils, successfully in many reported cases.  The above-mentioned efforts can be labeled as physical remedies to rural backwardness, the objectives of which were to diminish the attractiveness of infrastructure availability in cities for potential rural emigrants and to a lesser extent, to provide supportive services for agricultural activities. As the years are passing, there is a tendency to pay more attention to the production supporting role of infrastructure as opposed to the overemphasized social service delivery role. This is probably due to the low return of the latter and to the recent budget cutbacks.  fertilizers, tractor spare parts, etc. by the service centers and also, to purchase their remaining needs [often more than half of their needs] the farmers had to go to cities. Regarding the shortage of health personnel, refer to Malek (1991). As an attempt at encouragement, the provincial authorities have given a quota of subsidized construction materials (e.g., cement and iron beam at formal prices) to those who have incurred a loss. In a few cases, I observed that the damaged property has been given the commercial land use designation (i.e., shop) and business permission for the owner. 96  65  The other lines of efforts in rural areas were numerous economic policies, of which the new land reform and subsidization of agricultural inputs are already mentioned.  The  general impacts of these economic policies may cautiously be drawn as follows: •  Despite a better access to credits and loans (Amirahmadi 1990, Ch.3) and a better provision of supportive services, the small holdings of most peasant farmers are still under economic strain and far behind their needs.  •  Subsidized agricultural inputs and government pricing of some agricultural outputs were inadequate to economize staple cropping (i.e., particularly wheat and barely). Besides, free market pricing made cash cropping and livestock breeding more lucrative (Mojtahed & Esfahani 1989).  Thus, where it was feasible, the commercialization  process was spread out. •  The supply of rationed goods through rural cooperatives (although lesser than what was supplied for the cities) was beneficial in helping to keep the level of rural absolute poverty down (Shafigh et al. 1992, 28-31).  •  Income distribution in the rural areas of Iran, compared to 1977 (before the 97 Revolution) is less unequal.  However, the new land reform did not radically  mitigate the land distribution inequalities (McLachlan: 1987, Ch.8). •  The service sector in general and the public employment sector in particular, expanded rapidly in the rural areas, fostering a limited diversification in the employment 98 structure.  Nevertheless, the shortcomings of supportive services for agricultural  production were more remarkable than those of social services.  On the whole, the Gim Coefficient has followed a decreasing trend in rural areas after the Revolution. 0.448 in 1977, 0.416 in 1982 and 0.408 in 1987 (Shafigh eta!. 1992, 34-37). Census statistics show an average annual growth rate of 9.2 percent for service sector jobs in rural  It was 98  areas and 5.3 percent in urban areas during 1976-1986. The share of public employees in total rural jobs rose from 5.7 percent in 1976 to 17 percent in 1986 (SCI 1981, 83; SCI 1990a, 65).  66  Equally important were the social impacts of some other types of policies (e.g., social, cultural, administrative, etc.) such as: •  The establishment of the rural Islamic council.  •  The implementation of new rural jurisdictions (i.e., regrouping villages into a new district with a central village, called dehestan).  •  Welfare payments/pensions for elderly villagers (albeit quite short of their needs, and hitherto non-existent).  •  The broad cultural campaign for the value of a rural life-style and of agricultural works.  All the selected policies pointed out here, in essence, have influenced rural out-migration with varying degrees of effect. Yet, they were not sufficiently effective to alter, as the State intended, the predominant pattern of urban-bound migration. Aside from this, now let us briefly assess rural-urban migration in Iran, to see if there is such a need of intervention.  3.5 Assessment of Rural-Urban Migration in Iran  So far, the context and dimension of rural-urban migration in Iran and the postrevolutionary policies to cope with it have been delineated.  Subsequently, it can be  deduced that the occurrence of Iranian rural-urban migration has been inevitable due to a number of determinants created by the foregoing policies.  Recapitulating the  determinants, I highlighted the driving forces of rural-urban migration in three stages as being: 1-  The breakdown of the village-city relationship and the traditional organization of agriculture (without sufficient substitution) in the name of modernization  67  2-  Oil-driven urbanization and the Westernization of consumption in the name of industrialization.  3-  The revolution of rising expectations among the rural masses after the Revolution and consecutive waves of refugees.  Notwithstanding, other non-policy factors which contributed involuntarily and historically to the rural out-migration trend in Iran should not be underrated. The most significant of these are the demographic traits (e.g., very high natural growth rate) and the natural 99 resource constraints (e.g., severe water shortage).  The widespread common view about rural-urban migration in Iran is negative and is derived from two misconceptions about the consequences of rural-urban migration. It is assumed that this kind of migration has: 1) devitalized the agricultural sector, and 2) created urban predicaments. While there are some truths in these assumptions, their flaws are not taken into consideration.  In fact, much evidence inversely suggests that certain portion of rural out-migration makes agriculture viable in most parts of Iran. The statistics demonstrate that the increase in arable lands has not kept up with the natural population growth in Iranian rural areas during the last three decades.’°°  Considering the low productivity per capita of the  These factors are pointed out in section 4.4, but not discussed in detail due to the scope of this research. For the combined effect of demographic pressure on the resources in Iranian mountains and highlands refer to the study by Xavier de Planhol (1970) and notes of Bowen-Jones in Costello (1977, 47). The general description of land and water constraints in Iran are found in: Mclachlan (1988, Ch.2) and Nattagh (1986, Ch.3) 100  Between 1971 and 1978 there was 1.4 percent growth of arable lands; then a sluggish growth of 0.4 percent on average annually uiitil 1987 (PBO 1990b, 2-4). During this period, the actual population living in rural areas increased at an average rate of 1.8 per annum whereas their natural growth rate is estimated to be at least one percent above that.  68  101 and also the higher household size agricultural sector in comparison with other sectors, 102 it is virtually unlikely to have considerable and commensurable growth in in rural areas, agricultural employment opportunities. Even though the non-agricultural jobs rose sharply in the years after the Revolution, they could not hinder the increase in population/job ratio in the rural areas. 103 As a result of the growing pressure of population on land and the inadequate growth of non-agricultural sectors in the rural areas of Iran, probably we can conclude that emigration to a large extent was helpful from the point of view of economic viability at its origin: the village.  Nonetheless, from the broader point of view of sustainable development, rural outmigration in Iran probably debilitates the rural community because of its selectivity (Hakimian 1990, 137).  The most possible negative impacts are: skill drain, exodus of  educated and entrepreneurial persons, demoralization of staying villagers, and capital flight (Hooglund 1982, 119).  In many environmentally sensitive areas, it has led to  abandonment of land and the aggravation of soil erosion and desertification. It is not certain if correlation of the growing migration with the increasing imports of agricultural goods in Iran is causally linked (Nattagh 1986, 63). However, the out-migration certainly has caused the shortage of household labour which adversely affected agricultural production (Hakimian 1990, 146; Amid 1990, 140). In addition, it is strongly believed that environmental degradation, particularly soil erosion, overgrazing, desertification and  In 1960, the share of the agricultural sector in the GDP (excluding oil) was 37 percent with 52 percent of the country’s employed workforce. In 1976, the same figures were 14 and 36 percent as compared to the service sector having 63 percent of the GDP and 31 percent of employed workforce. in 19 87, while about 28 percent of the employed people were in the agricultural sector, they produced 19.5 percent of the country’s surplus value (SCI 1991, 537; National Accounts, Central Bank ofIran. cited in: ZahediMazandaranietal 1985, 135-137). 101  102 The average rural household size was almost 0.7 persons more than in the cities in 1986 vs. 4.77 (SCI: 1990, P.33).  --  i.e., 5.45  The ratio of populationljob in the rural areas grew from 3.55 in 1966 to 3.81 and 4.46 in 1976 and 1986 respectively (refer to Appendix H.).  103  69  aquifer decline, is proceeding hastily in Iranian rural areas. In essence, there is a declining situation at the rural end.  On the other side, the migrants’ drift aggravated urban problems in Iran, because of transferring unemployment, underemployment and the need for infrastructure to the cities. The unemployment rate and the unproductive service sector are rampant in metropolitan areas (Danesh 1985). Holding back the emigrants from the cities could mitigate the urban crisis but would hide the acute rural problems as well. Therefore, at the national level, intensification of cities’ shortcomings is due to a spatial transformation of problems which must be confronted anyhow.’° 4  This evidences suggests false expectations for in-  migrants of cities.  So, referring to the proposed conceptual framework for assessing migration in section 2.2.2, the migration from a declining situation in the village to the city with false  expectations, is a negative movement all together (see Figure 2.1). Undoubtedly, a shift from a weak agricultural base to a specious urban economy on the basis of oil revenues can neither be assessed as positive nor as sustainable. In lieu of extensive studies on the impact of rural out-migration in Iran, this rough generalization brings us to the crucial need of intervention in the general trend, which must yet be investigated case by case.  104  As a matter of fact, the influx of migrants to the cities (particularly to the Capital) forced the goverifinent to pay more attention to the needs of this potentially disturbing group, reciprocally encouraging more rural-urban migration. Hence, the cycle will continue. So, the question remains for the national strategy to delineate in which places, rural or urban, and at what level should the needs of potential migrants be met.  70  CHAPTER W: MIGRATION ThEORIES AND THE ROLES OF INFRASTRUCTURE  71  4.0 Introduction  Despite considerable theoretical work on migration, the focus area of this research, namely, the role of infrastructure in rural out-migration, cannot be drawn explicitly from these theories.  By reviewing major migration theories in this section, the causes and  determinants of rural out-migration will accordingly be summarized to be related to the potential roles of infrastructure. This is how the conceptional framework for the field research will be outlined.  At the beginning, two caveats are necessary.  First, the focus of this research is on  voluntary migrations, excluding compulsory migrations whether driven by natural [e.g., drought] or artificial forces, [e.g., war].  Second, while migration theories have been  treated universally, the presupposition is that they are applied to the South context in relation to rural-urban movement.  There exists a huge migration literature which emphasizes the descriptive aspects of migration with who and how questions on an empirical level, rather than with why questions on a theoretical level (cf Skeldon 1990, 126).  And as many authors have  5 (Shrestha discussed, it seems that, if possible at all, the field lacks a general theory’° 1990, 59; Selier 1988, 20; Oberai & Bilsborrow 1984, 21; Ogden 1984, 5; Wood 1982, 298; Peek 1981, 59; Portes 1978, 5; Simmons et al. 1977, 9; Goldstein 1976, 427). The overriding reason for this lack could be the interwoven multi-disciplinary (economic, demographic, ecological, cultural, etc.) and multi-level (individual, household, community, regional, etc.) nature of migration study. Bearing in mind that there is a dire need for policy guidelines in relation to inappropriate migration trends, it is logical to put research As has been pointed out comprehensively by some scholars, “a general theory of miation must be able to answer: Who are the migrants? Why do they move, stay, or return? How and where do they move? When do they move? What are the effects of such actions on the migrants and on others?” (Chang 1981, 105  305-6).  72  efforts towards grounded theory and praxis, contrary to trying to construct a general theory.  Therefore, the aim of this overview is to compile the implications of existing  theories for the roles of infrastructure in the South.  For the purpose of this overview, it is convenient to summarize the migration theories in three categories, namely: macro, micro and meso-levels. Macro-level of theories posits migration as a structurally determined social process at the aggregate level, usually in terms of nations or regions. At the other end, micro-level theories of migration treat migration on personal level, seeking to explain the migrant’s decision to move. And at the meso level, there is a growing interest to concentrate on the household/community level as the interface of both micro and macro migration theories.  These are discussed in the  following sections.  Prior to proceeding with the review of migration theories, it should be mentioned that some classic and general theories of migration, namely, Ravenstein’s laws of migration’° 6 (1885 & 1889), Lee’s theory of migration’° 7 (1966), Gravity and Spatial Models’° 8 (refer On the bases of Britain’s data between 1871 and 1881, Ravenstein suggests the laws of migration. As Ogden has summarized these are: “1) The majority of migrants go oniy a short distance. 2) Migration proceeds step by step. 3) Migrants going long distances generally go by preference to one of the great centers of commerce or industry. 4) Each current of migration produces a compensation counter-current. 5) The natives of towns are less migratory than those of rural areas. 6) Females are more migratory than males within the kingdom of their birth, but males more frequently venture beyond. 7) Most migrants are adults; families rarely migrate out of their country of birth. 8) Large towns grow more by migration than by natural increase. 9) Migration increases in volume as industries and commerce develop and transport improves. 10) The major direction of migration is from the agricultural areas to the centers of industry and commerce. 11) The major causes of migration are economic” (Ogden 1984, 17). Admittedly, many of the stated laws are subject to enormous modification, if not abolition, in terms of population mobility, e.g., female predominance, short-distance migrations, and step-by-step procedure. 106  107  Lee’s formulation is further development of Ravenstein’s work. In a nutshell, he proposes a pushpull model (Lee 1966) comprised of negative, positive, and indifferent factors associated with the area of origin and the area of destination, all are governed by personal factors and intervening obstacles. The application of this conceptual framework is useful for analysis but not suggestive of causation. 108  The gravity model postulates that migration is directly related to the population size and inversely related to the distance between origin and destination. Stouffer (1940 & 1960), by proposing the idea of intervening opportunities, incorporates some other spatial factors into the model. As stated by Huw Jones, Stouffer argues that linear distance is a less important a determinant of migration patterns than is 73  to: Zipf 1946, Stouffer 1940 & 1960, Hagerstrand 1962, Morrill 1963), have been excluded because of their descriptive approach.  Put another way, these theories  essentially address how, not why, migration occurs. They tend to treat migration as a black box at the aggregate level.  Let us briefly states that: distance  ,  momentum of  population movement, gravity of population size and mental-maps, among others, considered as deterrent or motive, are all of more of an intervening nature, than root causes of rural out-migration in the South countries. Consequently, the interpretation of causal relation is faulty when determining the role of infrastructure. Thus, these theories have not been reviewed in this chapter.  4.1 The Macro-Level Migration Theories  The theories in this category hold migration to be a causal outcome of socio-economic development and ultimately, seek migration determinants at a superior level.  The key  question at this level of theory is: why does migration occur rather than the movement of individual migrants? (Shrestha 1990, 63). Although the developmental paradigm of these theories has been discussed earlier in relation to the migration assessment in the second chapter, here their concrete migration theories are reviewed. Three contrasting branches of theories can be classified within this level. First, those which have been drawn from neo-classical dual economy and modernization theory and secondly, those derived from Marx’s views and dependency theory. A third branch may also be recognized on the basis of an ecological approach and Maithusian theory.  the nature of space; that distance should be regarded in socio-economic rather than geometric terms; and that because migration is costly (socially as well as frnancially), a mobile person will cease to move when he encounters an appropriate opportunity (in Ogden 1984, 21-2).  74  4.1.1 The First Branch  The basic tenet of the first branch is migration as an equilibrium mechanism, which shifts unutilised and unlimited surplus labour of the subsistence/agricultural sector to the labourdeficit modern/industrial sector (Dasgupta 1981, 52-53; Bhatia 1992, 15-16). This was the conventional economic approach to migration, known as the Lewis-Fei-Ranis model (Lewis 1954; Fei & Ranis 1961) and a cornerstone for then mainstream micro-level theories of migration. It was believed that the process of labour transfer and the growth of employment in the modern sector were concomitant and were brought about by high industrial capital accumulation and its reinvestment in the modern sector (Todaro 1976, 21).  Therefore, migration was conceived to be a necessary component of structural  transformation to a modern state, mutually beneficial to urban and rural areas, and 109 eventually bringing about an equalization of income and productivity in two sectors (Oberai 1987; 38)  In spite of its appeal, the model has been found unsatisfactory. There are major inherent suppositions in the model which much empirical evidence does not confirm.  First, it  implies that lost agricultural production (as a result of rural migration) is none and thus, attributes zero or even negative marginal productivity to the migrants. Second, the model postulates that full employment in cities prevails and that continuous investment is enough to absorb the increasing labour supply.  Third, the supposition is of the continued  existence of constant real urban wages until the supply of rural surplus labour is exhausted (Todaro 1976, 23-25; Dasgupta 1981, 52; Cherunilam 1984, 41-42). None of these in the  109  To elaborate on this, it is supposed that the migrants will be absorbed with a constant wage premium up to the point where there is excess labour in cities and a shortage of that in the countryside. Corollary to this, since it is thought that there is a cumulative reinvestment in the city-based modem sector, the demand for agricultural production will increase. Thus, there will be a higher agricultural wage and better tenns of trade for rural-based sectors. (see Bilsborrow et al. 1984, 16)  75  South countries have been realized, and with other theoretical developments are no longer 10  A good reflection of this branch of theory is found in the work of Zelinsky who introduced the mobility transition theory. He has hypothesized that patterns of movements change according to stages of development. In his words, “there are definite, patterned regularities in the growth of personal mobility through space-time during recent history, and these regularities comprise an essential component of the modernization process.” (Zelinsky 1971, 221-222)  Zelinsky delineates five evolutionary phases from traditional to super advanced society. Rural-urban migration, in this model, will increase during transition and finally will decline with improvements in transport and communications and diffusion of developmental space” (Pryor 1975, 24;Guest 1989, 7).  110  Due to the importance of this theory, it is noteworthy to mention the serious shortcomings according to different scholars: Todaro enumerates the conspicuous consumption and the transfer of capital to overseas as the outlet for profits (inhibiting reinvestment in the South cities) and also argues that most modern technologies are labour-saving and do not create jobs proportionate to the rate of investment(1976, 24-5). Peek adds that a positive rural-urban wage difference is not enough to attract subsistance farmers and that there is a need for preferntial government policies to pave the way( 1981, 612). Similarly, Bhatia argues that migration from rural areas is not solely induced by underemployment. He points to the adverse effects of labour/skill drain on rural areas of the South which lead to a fall in production (1992, 16-7). McGee shows the dynamic interaction of dual sectors and the existence of both peasant and capitalist workforces in both rural and urban areas of the South. Thus, he draws the attention to the different implications for inter-sectoral mobility (1977, 207-11). Grindle emphasizes the inability of the model to explain temporary migration, a growing pattern in the South (1988, 26). And as a whole, Lipton swnmarized the equilibrating failure because of the migrants’ characteristics, migration process, the impact of absence on the village, remittances, and the effect of return migrants (1982, 202-15). 11  Zelinsky deals with mobility in a broad sense including migration and various forms of circulation. Mobility transition in his view parallel vital transitions, specifically fertility and mortality(Kosinski & Prothero 1975, 10). He suggests an evolutionary model of spatial behavior with the ensuing traits: Stage one: low residential mobility and natural increase. Stage two: a sudden increase in fertility, accompanied by large scale rural-urban migrations. Stage three: a decline in the rate of natural increase, rural-urban and rural-rural migrations, but inter- and intra-urban mobility increases. Stage Four: leveling off rates of natural increase and further decline in rural-urban and rural-rural migration. Stage Five: a substantial decline in residential mobility because of communication developments allowing closer place of work to place of residence and their integration (McGee 1977, 200).  76  The mobility transition theory hypothesizes a unilinear process of modernization occurs everywhere (McGee 1977, 201) and the underlying premise is the similar accessibility of each settlement to major urban centers and its resource endowment for all social groups (Brown 1991, 52).  Furthermore, it treats migrations as a homogenous phenomenon  (Standing 1984a, 53). The empirical evidence is mixed and generally, does not support  the patterned regularities consistent with the development process in the South (Colferl985, 248; Lait 1985, 117; Brown 1991, 52). The theory gravely underestimates the volume of circulatory migration, and overall, poorly articulates the cause of migration variation from region to region  (  Woods 1985, 2;Guest 1989, 8).  Reliance on  modernization theory, undermines the ability of the mobility transition theory to acknowledge the coexistence of rural/urban, modern/traditional, and formal/informal 2 economic sectors in a dualistic society with irregular mobility implications.”  In general, this branch fails to explain the increasing flows of migration in the face of increasing urban unemployment  and  does not recognize migration as a result of  3 (Paul 1989, 24). As such, it is concluded that rural-urban continuous disequilibrium” migration in the South can no longer be seen as transfer of surplus labour to more productive urban employment, rather it is a shift of under- and unemployment from rural to urban areas (Oberai 1987, 39; Dasgupta 1981, 53).  112  It is stated that “the transfer of population alone, without economic development is not a solution...  [it adds] to population-resource problems... in developing countries the majority of migration is not the result of sustained socio-economic growth, but of pressure associated with population increase, limited economic development, and rising expectations...” (Kosinski & Prothero 1970, 255-6). The critiques say this requires recognition of the national process of capitalism penetration (McGee 1977, 202) and the dependence on the external market as well (Peek 1981, 62).  77  4.1.2 The Second Branch In the second branch, the theories utilize historical-structural factors to explain migration, specifically the changes in the organization of production (Wood 1981, 339; Shrestha 1990, 55; Chant & Radcliffe 1992, 18). Wood describes their difference as being that: “migration is conceptualized as a class phenomenon, where the unit of analysis is the stream, as opposed to the atomistic approach that treats migration as the sum of individual choices.” (Wood 1982, 302)  It is stated that “proletarianization” and penetration of capitalist relationships in rural areas of the South generate streams of rural-urban migration (McGee 1977, 205; Roberts 1978, Ch.4; Peek 1981, 62; Standing 1981, 201; Friedmann 1992, 14-18).  Consequently, a spatially uneven structure has been brought about “by geographically separating capital (the employer) from capital-dependent labour” (Shrestha 1990, 54). Given structural imbalances, many scholars pose that there is a necessity of labour displacement from backward rural areas to thriving urban areas (Shrestha 1990, 42-44). 114 (Parnwell 1993, As such, people vote with their feet to gain access to the opportunities 74; Lipton 1976, 217).  Nevertheless, the notion of proceeding toward equilibrium is  negated by deducing the recurring mechanism of uneven population distribution in the South (Lipton 1976, Ch.9 Amin 1974).  To exemplify these views, Amin, a prominent researcher in this category, has studied the migratory phenomena in West Africa “which accompanied the consolidation of colonial  On this theory, Portes explains: “In general, the structure of economic forces in core and periphery tends to be arranged so as to condition migrants to sell their labour in places where needed and at the cheapest possible price... The phenomenon of migration thus stands at the crossroads between national and regional inequalities and class exploitation” (Portes 1978, 47-8). Other authors in the same vein suggest: “Migration to cities reflects merely a demographic adjustment to change in the spatial structure of economic and social opportunities... in a developing economy where foreign capital investments are important” (Friedmann & Wulff 1976, 26-7). 114  78  presence”(Amin 1974, 69).  He maintains the determinant role of external forces and  emphasizes unequal spatial development as being the root cause of migration (Ibid. 120). Amin points out that this migration reciprocally perpetuates the inequalities.  Another  researcher studying the same region notes the trivial effect of population pressure in pushing the villagers to migrate, and underscores the state-driven unequal concentration of opportunities in the cities (Hart 1986, 72-76).  The major critiques of the second branch of macro-level theories are: they tend to be deterministic (Skeldon 1990, 33), to lead to generalities over time without directly addressing acute migration problems (Oberai & Bilsborrow 1984, 20) and ‘pay little attention to the factors that motivate individual actors” (Selier & Karim 1986, 13). Particularly, this branch fails to address the variation of migrations which occur under the same structural conditions.  4.1.3 The Third Branch  Migration is believed here to be a symptom of the growing population pressure on resources, often arable lands (Grigg 1980, 63). Therefore, it is perceived that migration equilibrates the regional distribution of population or at least lessens the pressure in the originating region. Equilibrium is thence considered from the ecological point of view, not from the economic standpoint echoed in the first branch theories. The fundamental tenet arises from the Malthusian theory of population, stipulating that population growth outstrips the means of feeding it, i.e., there is a geometric/multiplicative progression of population vs. an arithmetic/additive progression of food supply (Goodall 1987, 282).  In the same vein, some geographers and ecologists use carrying capacity concepts to show from a broader perspective, that a specific area can support a certain number of people without deterioration of its eco-systems (Hagget 1979; Catton 1980).  Rationally, 79  the capacity is affected from the following patterns: population growth, distribution of resources, consumption, production, and technology (Hynes 1993).  By virtue of this  conviction, with respect to rural-urban migration it is elicited that “...with given technology there is a certain proportion of the labour force which can be absorbed by agriculture. As population grows, unless the rural non-crop husbandry sectors [dairing, poultry, forestry and fisheries] or cottage and small-scale industries expand so as to absorb the surplus, increasing number of people must move to the urban centres to find employment.” (Oberai & Bilsborrow 1984, 18) Hence, while the cause of rural out-migration can be justified, the urban consequences in all probability create more pressure on the eco-systems.  As such, migration is not a true  equilibratory mechanism from the ecological point of view.” 5  The stumbling block in this branch is preoccupation with natural forces and natural impacts.  Bearing in mind the importance of unemployment and poverty for rural out-  migration in the South, studies have demonstrated that “population pressure is not the only or even the principal cause of increasing unemployment and poverty of the rural population” (Ibid). It has been evident that there have been varied responses to increases in rural population in the South, “categorized as: a) economic, e.g., increase in arable land or land intensification, b) demographic: fertility decline, c) economic-demographic: rural out-migration” (Bilsborrow 1987,  198).  Forces like individual motives, types of  technology, socio-economic structures, and economic policies that affect the carrying  115  An interesting study by Boserup indicates that an increase in population density through population growth will hasten the adoption of more intensive methods of farming. She concludes: “According to the explanation offered here, population increase leads to the adoption of a more intensive system of agriculture in primitive communities and an increase of total agricultural output” (Boserup 1965, 118). She goes on to say that migration because of population pressure is not inevitable and “...the choice [is] between harder work in more intensive food production, or migration to urban areas” (Ibid. 120). Also, in studying migration in the Philippines, Simkins argues that “migration is not an automatic response to increasing population pressure- there are a number of other alternatives open that may be more acceptable to the persons invoved” (Simkins 1970, 266).  80  capacity have been downplayed (De Jong & Fawcett 1981, 19; Meyer 1993, 1516).h16 Of particular weight is the system of resource distribution, which exerts heterogeneous population pressure due to social classification.  4.1.4 Summary  In short, all the branches of macro-level theories view migration as a dependent variable and are concerned with the macro contextual constraints. The first and the second branches share the focus on economic factors that are virtually determined within urban areas (Guest 1989, 7). Both rely on the conditioned behavior of individual migrant due to 7 (Selier 1988, 23-24). In modernization-led theories, rural-urban development structure” migration is caused principally by the incapability of rural sectors to absorb more productive labour. Dependency-led theories in essence attribute rural out-migration to the “skewed distribution of capital resources” (to the benefit of cities) and to structural constraints for resource accessibility(Shrestha 1990, 60; also see Lipton 1976). Generally, as Pryor has mentioned, “assuming that migration is always economically purposive behavior” and related to macro-level variables, overlooks “the significant proportion of migrants who move for social, idiosyncratic and particularly multiple reasons” (Pryor 1975, 30).  The third branch of macro-level theories attributes the cause of rural out-migration chiefly to deterioration of the environment and/or to shortage of the natural resource capacity. Anyhow, it should be taken into consideration that despite the suggestive rationality at the Analogy with international migration may be helpful. It has been noted that: “Enviromental degradation is but one element in the interconnected web of forces rending to generate new patterns of 116  South-to-North movement” (Simmons 1992, 32). Simmons indicates that the globlaization of economies is the chief cause of increased differentiation and uneven development, generating increased human migration (Ibid.).  In other words, it is asserted even in the free-market situation “the way individuals respond to market conditions is itself an artifact of development” (Brown 1991, p.44). 117  81  macro-level, the majority of rural people in the South decide not to migrate (Parnwell 1993, 91). Hence, the micro- and meso-level theories complement macro-level theories in their application towards an holistic understanding of migration cause.  4.2 The Micro-Level Migration Theories  In contrast to macro-level migration theories, “the micro-level approach looks at the issue from the perspective of the individual unit, which could be the separate individual or the family” (Chang 1981, 309). The migration theories in this category pursue the question of why individual migrants move instead of why migration occurs. And they deal with the perception and the behavior of migrants (called cognitive-behavioral approach) as opposed to the historical-structural approach at the macro-level (Chant 1992; Boyer & Ahn 1991). Two branches in the micro-level can be discerned as follows: the first branch, in general, emphasizes the individual migrant-decision making in the  framework of neoclassical  economics and virtually, probes the economic determinants of migration.  The second  branch, which by no means is mainstream, underpins the non-economic cause of migration, still on the basis of individual decision-making.  4.2.1 The First Branch  As pointed out, the focus here is on economic reasons for migration, and overall it rests on “the premise that the direction of the progression or path of development is towards equilibrium in the price of factors of production and in living standards” (Therom & Graaff 1987, 2). Similarly to the first branch of the macro-level, labour mobility is the core of reasoning. The labour mobility studies of economists postulate that “as labour demand and supply are always in equilibrium in a classic competitive framework, differences in net economic advantages, chiefly differences in wages,” are the main cause of migration.(De Jong & Fawcett 1981, 22) However, the criticism that the macro-level theories consider 82  the mobility as a passive response to a variety of stimuli and do not differentiate the rural source areas, cannot be repeated at the micro-level (Parnwell 1993, 94).  The main contention in the first branch is based on human-capital migration theory, formulated by Sjaastad (1962). The theory treats migration as an investment decision in light of cost-benefit calculation by a potential migrant. In effect, if returns of migration over time (i.e., both monetary and non-monetary) outweigh the related costs, then the theory assumes that the individual will rationally  18  (De Jong & Fawcett 1981,  23-24; Oberai 1987, 39). It is taken for granted that the “people desire to maximize their net real incomes over their productive life and can at least roughly compute their lifetime 9 ” t income streams in the present place of residence as well as in all possible destinations (Oberai & Bilsborrow 1984, 16).  In the same strand of theory, the most prominent migration model is known as Todaros model (Todaro 1969; Harris & Todaro 1970). He copes with the seemingly contradictory phenomenon of continued in-migration while unemployment is being increased in the South cities. Todaro writes: As Lee describes: “The costs of migration are the costs of moving -often represented by distance as a proxy- earnings lost while moving and fmding a new job, and the psychic costs associated with the disruption of leaving familiar surroundings. The returns of migration are better employment opportunities, increased job earnings in the new location and the value of amenities or public services that might be superior to those in the location of origin’ (Lee 1985, 15-6). From this viewpoint, the critical amount to justify or reject the migration decision is “the difference between the expected utility of the present discounted value of the lifetime real-income stream the migrant expects to receive if he or she moves to that destination and the expected utility of the present value of the lifetime real-income stream the migrant expects to receive if lie or she does not move, less the costs of moving” (Da Vanzo 1981, 93). 18  119 Sjaastads contribution is more successful in rationalizing the characteristics of migrants. Accordingly, it can be justified that most of the migrants must be: youth [thus having a longer period of harvesting the returns], single [thus having lower moving costs], educated [thus having a better chance of obtaining a high wage job] and following the known path [thus reducing the cost of residence at the destination] 1Paul, 1989,25-6). The operational problem for testing this theory is in respect to generalization from aggregate data analyses to the individual level (De Jong & Fawcett 1981, 23-4). Also, it is noted that because of less stability and more uncertainty in the South countries, probably potential migrant attach less value to future gains than immediate ones (Da Vanzo 1981, 94).  83  “migration proceeds in response to urban- rural differences in expected rather than actual earnings.. Expected income gains are measured by [a] the dfference in real incomes between rural and urban job opportunities migrant obtaining an urban job.”  and [b] the probability of a new  (Todaro 1976, 28-29)  Actually, Todaro considers migration as a direct response to real-wage and opportunity differentials. In other words, migrants participate in a lottery of relatively high-paid jobs in towns (Gillis et al. 1987, 187). This formulation of the role of economic incentives has been quite influential in migration policy-making -e.g., drawing attention to the migrational implication of urban wage determination and agricultural pricing.  The findings of empirical studies in the rural South are not supportive of this theory at best (Parnwell 1993, 87; Bhatia 1992, 20-21; Simmons et al. 1977). Though the income gap between rural and urban areas has been widely inferred as the overriding cause of migration, there are numerous criticisms in relation to the undervalued role of non economic factors on one hand and the neglected characteristics of economies and communities of the South on the other.  Specifically, the criticism about the theories in this branch are:’ 20 •  Family ties and commitments, and the migrant’s stage in the life cycle are crucial in migration decision-making in the South, but have not been accounted for (Gillis et al. 1987, 190; Grindle 1988, 27).  •  While the cost-benefit theory of migration “is developed in terms of the individual decision-maker, most tests are conducted with aggregate level data” (Speare et al. 1988, 108) which suggests “misleading aggregation and artificial separation of variables” (Chang 1981, 320).  Because of the prevalence of this branch in migration studies, a more detailed criticism of this branch is discussed here. 120  84  •  lVligrants tend to optimize, not necessarily maximize utility as assumed -e.g., the case ’ (Chang 1981, 314; Grindle 1988, 26). 12 for circulation between rural and urban areas  •  The implicit assumption of complete information and rational evaluation of alternatives is not valid and the antecedence of migration streams and establishment of networks with urban areas over time chanalized the information -e.g., the case for chain migration (Cherunilam 1984, 44; Oberai & Bilsborrow 1984, 17; Paul 1989, 31-32; Singh 1991, 37; Parnwell 1993, 93).  •  The models assume homogenous migrants’ labour while in reality it has been observed universally that migration is a highly selective process in respect of skills and attitudes (Oberai 1987, 40; Paul 1989, 31 cited Sabot).  •  The access to the urban labour market is not equal and friends and relatives play an important role in who, where, whatjob (Speare et al. 1988, 309; Singh 1991, 35).  •  A large informal sector, in addition to dual economic sectors, exists in the South, is being ignored (Paul 1989, 31).  •  The explicit assumption that migration is the only way in which to compensate for geographical wage differences is invalid. (Bhatia 1992, 20-21) In-situ strategies, e.g., diversification, specialization and circulation, are possible in most cases as well (Guest 1989).  •  The theories in this branch are not helpful in explaining particular choices of destination (Grindle 1988, 27).  •  The aspiration to a higher quality of life and social mobility via migration to cities has been overlooked (Lee 1985, 16-7; Parnwell 1993, 87).  121 The premise of utility maximization and income as its proxy has been strongly questioned in the literature. For instance, Chang states: “Use of the income maximization principle could produce  misleading conclusions if the migrant does not maximize all the components of this aggregative variable. The demand for urban income by in-migrants could mean a demand for the options conferred by purchasing power. These other demands may include economic and noneconomic goods. But they have now all become disguised economic aggregative variables.” (Chang 1981, 322)  85  •  out-migration’ (Sovani In the South, with the importance of push factors in rural 22 1984; Gillis et at. 1987, 189; Oberai 1987, 12; Ghose 1990, 48; Brown 1991, 49; Singh 1991, 226), many villagers migrate “not because of expected income differences but because they are unable to earn a subsistence” (Dasgupta 1981, 64) and there is no real alternative (Wood 1982, 305). Put another way, weighing the decent lfe chance in two environments is more relevant to the majority of migration decisions than a cost-benefit analysis.  From a different perspective, proponents of macro-level theories raise a meta question about the cause of differentiation of wages: “Capital is much more mobile than labour”, so why does capital not move for equilibrium? (Amin 1974, 85) Indeed, in this branch the unequal geographical distribution of the factors of production except for labour (i.e., capital, land, and natural resources) have taken for granted that which, among others, 23 (McGee 1977, 197). determines the unequal remuneration of all the factors’  In addition, advocates of meso-level theories pose the problem of overemphasis on the individual at the micro-level and ascribe it to “the consequence of influence from Western scholarship, which stress the Benthamite atornistic individual economic man.” (Chang 1981, 317).  It should be noted that the distinction between push and pull factors, as elaborated by Lee (1966), is not a convincible polarity. To illustrate this, while urban employment opportunities are necessary as the pull factors, they do not suffice to cause rural out-migration unless the push factors [e.g., dwindling incomes, unemployment and landlessness] are present (ILO 1960). Thus, researchers argue for a quasipush and quasi-pull distinction (Connell et al. 1976, 198; Pryor 1979, 326). A further point worthy to be mentioned is that for better-off rural migrants it has been elicited that pull factors are the main cause of migration (Lipton 1980, 15), the assertion of push prevalence must be regarded relatively and discretely. For instance, in some part of the South, e.g., West Africa and probably in oil-rich states of Middle East, seemingly, the urban pull is more explanatory of rural out-migration than the rural push (Hart 1986, 75). 122  123  fact, remuneration varies mainly becouse of productivity. And the detenninants of productivity of technology, the workforce skills and the reproduction /enhancement of its quality. Thus, level are the migration due to unequal remuneration is a superstructural explanation.  86  4.2.2 The Second Branch  The thrust of this branch of theory is on the preponderance of non-economic factors in migration behavior. Ever since the economic discipline prevailed the migration studies in the 1960’s, this branch has been on the wane and not an articulated theory can be found. While most studies of migration acknowledge the presence of a wide variety of non economic factors in any decision to move, they often ascribe to them a peripheral role and submerge them in the economic factors’ supremacy. Dissimilarly, some studies in this branch negate any explicit economic calculations for migration (e.g., Speare 1971) and instead search for non-economic motivations.  Regarding non-economic factors in the migration literature, there are studies which focus on the individual perception of the utility of the environment, mostly studying intra-urban mobility and residential location preference in the North countries (Shaw 1975, Ch.5). As an example, on the basis of the subjective evaluation of place utility by a potential migrant, the stress-threshold model views migration as “a form of individual or group adaptation to perceived changes in environment” (Wolpert 1965, 161).  It conceptualizes a stress  threshold for the potential migrant to bear or modify the environment of origin, and beyond the threshold, to seek a new place of residence (Shaw 1975, 110, Wolpert 1966). Considering another example, the residential satisfaction model postulates that satisfaction with one’s home area is inversely related to the migration intention and this satisfaction is proposed to be in the bonds to other people in the community (Lee 1985, 15; Spear 1974). To operationalize the level of place satisfaction, amenities of the origin infrastmcture.’ have been used (Lee 1985, 15) which have implications for the role of 24  is noteworthy that there are studies which incorporate the role of location-fixed amenities in human migration from the point of view of neoclassical economics (see Knapp & Graves 1989). They argue that “in an equilibrium setting, rising per capita income levels lead to changing demands for 124  It  location-specific amenities. These changing demands lead to migration flows to more desirable locations over time... amenity values are capitalized into wages, rents, or other local prices” (Ibid. 71).  87  Parallel to these studies, the most famous formulation with some proof in the South ruralurban migration, is the so-called bright lights of the city (Gulliver 1955).  It has been  hypothesized that the rural youths, being aware chiefly through mass media and former 25 migrants’ stories, were attracted to the urban entertainment and life-style.’  A scholar  indicates: ‘The city is thus portrayed and seen as the place to find fun  and excitement, and in the  mind’s eye of young impressionable villagers contrasts sharply with the generally slow and unexciting pace of life in the countryside.” (Parnwell 1993, 89)  In general, however, few empirical studies in the South have supported the theory (reviewed in: Findley 1977).  The fact of the matter is that many people do migrate for a wide variety of non-economic 26 It has been observed that education, health and housing facilities, the reason, as well.’ abandonment of traditional norms of rural life, aspiring social mobility, and escape from village disputes cause many migrations (Bilsborrow et al. 1984, 19; De Jong & Gardner 1981, 50; Shaw 1975, 110-115).  It has been argued that non-economic factors are  difficult to truly investigate and that in any case they follow primary economic cause or at best “become more important relative to economic factors at higher levels of development” (Bilsborrow et al. 1984, 19). So, Parnwell in relation to the bright lights theory, suggests that these factors influence the direction, rather than the incidence, of migration (Parnwell 1993, 91).  In the rural North, it is worth mentioning that an extensive survey titled: The Cause of Rural-Urban Migration among the Poor in the USA, has summed up its findings: “The urban amenity variable which we have interpreted as a measure of the bright lights attraction of cities is the chief predictor of migration for all groups, not just for the young” (Abt Associates 1970, 73). 125  126 The importance of some non-economic causes, such as family and community ties, and detenninants such as community context and cultural system, will be discussed with the meso-level theories, later.  88  All in all, it is believed in this branch that migration is more than a direct response to the objective economic factors (Wolpert 1965, 161). The subjective accounts of migration motives are pursued with a view to determining difficulties in operationalization. Even with the acceptance of the primacy of economic objectives, the individual subjective perception is indispensable (Shaw 1975, 107).  4.2.3 Summary  The theories at micro-level maintain that migration can be understood by investigating individual decision-making, in terms of both the perception and the behavior. Different from the macro-level theories with an historical-structural approach, they apply a cognitive-behavioral approach. For the first branch which stresses the economic reason of labour mobility, a rational decision to move is taken when the stream of life-time income in another place exceeds that available in the place of residence (Sjaastad 1962). Further theoretical work suggests that the migrant’s calculation is discounted at present rates and in addition, modified by the probability of getting a job in the new place (Todaro 1969).  In the second branch, the non-economic reasons of migration are underscored. A group of them reckons the perceived place utility as the motive to consider moving (Wolpert 1965; Spear 1974). And another theory in the same branch, puts forward the perceived attractiveness of urban social facilities, especially its entertainment, for the villagers (Connell et al. 1976, 51).  The empirical findings for both branches are non-conclusive and in many cases, contradictory. Nonetheless, the mainstream migration policy-making, by and large, relies on the first branch theories of this level of analysis, to the extent that some scholars conclude: 89  An unchallenged assumption in all literature is that differences in the price of labour among regions and sectors provide the principal incentives for mobility of economically active members of the household. (Sabot 19, 232)  Both branches assume freedom of movement in response to individual aspirations or socio-economic forces.  And both abstract from the socio-economic structural context  (Oberai & Bilsborrow 1984, 17; Standing 1981, 173)127  4.3 Meso-Level Migration Theories Having criticized the neglect of concrete contextual factors in foregoing migration theories, now we turn to consideration of meso-level migration theories.  This level of  theories is characterized by recognition of household as the unit of analysis and paying more attention to the community context.’ 28 It is claimed by the proponents of this level that relying on household gives the potential of linking micro- and macro-level analysis (Wood 1981, 339; Bilsborrow et al. 1984, 20; Goldscheider 1984, 294; Selier 1988, 27). Both questions of why individual migrants move and why migration takes place, are addressed in this approach.  Since in almost all South countries the prevalent unit of production and control of property in rural areas is household -i.e., usually extended families with three generations (Gugler 1988, 57)- it is believed that “migration is rarely a solitary affair” (Gugler 1 986a, 204).  By virtue of this, rural-urban migration is viewed as a household decision,  127 It is safe to reiterate the claim that: “atomistic, ahistorical free social agents do not accurately reflect the realities involved in the process of migration” (Sherestha 1990, 48). 128  Household is defined as a group of persons who wish to live together and therefore share living quarters and their principal meals” (Goodall 1987, 215). In other words, household consists of a group of people who “ensures its maintenance and reproduction by generating and disposing of a collective income fund.”(Wood 1981, 339) The key difference with family is the kinship which is not necessary in the case of household. As we will see in case of the first branch at this level, there is an ambivalent emphasis on both family and household.  90  responding to changes in its conditions (Goldstein 981, 337; Harbison 1981, 228). Understanding these changes in the context of local community is called for when seeking the motives of migration 129 (Hugo 1981a, 222).  In a sense, no articulated theory which explains the cause and pattern of migration can be found at this level. The novelty of this kind of study and its recognition of contextual and situational factors, make the formulation of theories at this level methodologically complex. Nonetheless, two branches of theoretical works may be discerned. One branch is more concerned with the economic rationale of migration, mainly through labour migration, though in a limited way looks at some contextual causes as well. The second branch utilizes an inter-disciplinary approach (psychology, sociology, and to a lesser extent economics) and interprets migration as one of the household sustenance strategies.  4.3.1 The First Branch  Here the diversion from conventional economic approaches, reviewed as the first branch at the micro-level, started with consideration of family/household as the basic unit of migration decisions. To differentiate this from human capital and Todaro’s models, it has been stated that: decisions to migrate are generally made on the basis of a calculus of household income streams from a variety of sources [agriculture, wage labour, petty commerce, artisan manufacturing, and remittances from labour migration] and so levels of wages may not be as important as the extent to which employment opportunities combine efficiently or effectively with other income-generating pursuits 129  ...  Households attempt to  To illustrate this, Saint & Goldsmith (1980) in their study in Brazil, have shown that as family  labour supply increases, more labour intensive crops predominate, which emphasize total rather than per capita output. When demographic pressure on the farm exceeds the capacity of the most labour intensive system, then out-migration of family members occurs. The study hereupon suggests that individual  rational-decision explanations of migration are inseparable from structural social explanations. It should be added that this inseparability is manifested in household decision-making, i.e., the meso-level study.  91  maximize total family income at the same time that they attempt to minimize the risk to overall income level through diversification.” (Grindle 1988, 26-27)  Thus, the family investment, particularly in the agricultural context, instead of individual gain, is considered to motivate migration of persons or households (Mincer 1978, 750; Da Vanzo 1976, 7; Stark 1980, 367). Subsequently, there is a possibility of incurring loss (in income or utility) for one migrant while the migrants family is gaining in total (Clark 1986, 69).  A useful theoretical formulation in this branch is the theory of relative deprivation, elaborated by Odek Stark (1984).  Substituting an interpersonal and concrete income  differential in place of general and abstract comparisons, he writes: “The relative deprivation approach analyzes rural-to- urban migration as a response to measurable dissatisfaction within an origin reference group and as a means of reducing or eliminating such dissatisfaction.” (Stark 1984, 481-482)  The emphasis again is on labour migration as the chief reason, though not from an individualistic optimizing perspective (Stark  1991,  3).  Recently,  revisiting his  contribution, he has mentioned: “Placing the family, rather than the individual, at the center of the migration decision [this need not result in migration by the family] was a relatively new direction. This must not be interpreted to suggest that the behavior of individuals should be ignored, but rather that it should be analyzed in the context of a decision-making unit operating as a group. (Stark 1991, 5)  92  Stark explicitly set forth that rural-urban migration is more than a labour migration in response to wage differential and indicates that “the ill-functioning capital market” is the ° (Stark 1991, 4 & 18). 3 major cause of”a great many of migratory phenomenon”  In sharp contrast to a widely held view, he postulates that migration is a risk-aversion, not risk-seeking, behavior by the household who spread the risk through diversifying sources of income (Stark 1978 & 1991, Ch.4).  While there is no direct empirical test, many  dispersed findings in the rural South support the argument. However, the major criticism is that it does not pay enough attention to contextual determinants at the community level.  Another theoretical formulation which may be considered in this branch is named the  value-expectancy model (DeJong & Fawcett 1981). Respectively, migration is assumed as a purposive and instrumental behavior, emanating from a “cognitive calculus” of family costs and benefits to improve or enhance the quality of life (DeJong & Fawcett 1981, 47-  57). The intention for migration is the sum of “the value-expectancy products” (Ibid.). In this model, values are associated with objective goals and expectancies with subjective probabilities.  Attempting to link different levels of migration determinants, this model proposes that: “Migration behavior is hypothesized to be the result of [1] the strength of the valueexpectancy-derived intentions to move, [2] the indirect influences of background  individual and area factors, and [3] the modifying effects of constraints and facilitators  130  The relative deprivation theory has built on three premises: I. There is more to labour migration than an individualistic optimizing behavior 2. There is more to labour migration than a response to wage differential 3. A great many migratory phenomena would not have occurred if the set of markets and financial institutions were perfect and complete (Stark 1991, 3-4). However, the third premise overlooks the poor agricultural resource base in many South countries (as shown in Mexico by Grindle 1988), which cannot be helped by technical and capital infusion alone.  93  that become salient during the process of migration decision-making.” (DeJong & Fawcett 1981, 56)  Criticizing the narrow assumption of economic rationality and self-interests maximization, it is correctly postulated that maximization of values need not to be economic and from an individual perspective (Chan 1981, 313).  The value-expectancy model is more useful as a research tool and still has difficulties for operationalization.  Particularly, its cognitive approach makes it vulnerable to the  inaccurate expression of goals and expectations by the respondents.  Also, in spite of  incorporating contextual factors, the weight assigned to them, especially for macrosetting, is not as important as we observe it to be in the South countries. The structural and cultural attributes of society may primarily predispose the individual to move or not to move (Brake & O’Hare 1984, 196).  4.3.2 The Second Branch The salient feature of this branch is its clear stress on the household as the unit of migration decision-making and on trying to connect it with contextual constraints and possibilities.  To offset the major critiques of micro-level theories, i.e., overlooking  underlying structural factors, and of macro-level ones, i.e., ignoring individual choices and behavioral differences, it is claimed here that: “Migration is conceptualized as an integral part of the sustenance strategies the  household adopts in response to the opportunities and limitations imposed by conditions that lie beyond the household  unit...  In the  context of  rural areas, where the unit of  production and consumption is the household, an integration of individual and structural  approaches can be accomplished through the analysis of household behavior as the unit interacts with its environment.” (Wood 1981, 338-339)  94  This branch of theory thereby, providing a more holistic approach than other levels of migration theories, it is believed that migration is only one, though important, aspect of the adaptive strategy of the household and is conditioned on the effectiveness of other ’ (Wood 1982, 314). 3 strategies for maintaining or increasing the household quality of life’  Scholars in this branch have shown that in many of the South communities, a different kind of family/household is emerging as a result of rural-urban migration. For instance, in West Africa, migrants maintain strong ties with their homes (Gugler & Flanagan 1978). Or in the Philippines, Trager argues, the ties between the migrant member and the original family/household, regardless of where the families live, endure and this kind of functionally related but locally dispersed household does not meet the definition of co-residence and sharing a common pot (Trager 1988, 182-184). Consequently, the terms like expanded family (Selier 1988, 31) or shadow household (Caces et al. 1985) have been coined. This phenomenon implies that comprehending the decision of household in the areas of origin is crucial to the understanding of individual or family migrations.  In this respect, household decision-making to maximize its welfare is embedded in family values and community norms and carries moral commitments for its members.  So, the  household life-cycle stage plays an important role in determining which member, for example, should migrate. These kinds of decisions include reciprocal obligations under which members contribute their share of responsibilities at another stage (Trager 1988, 188). Therefore, considering migration as a survival strategy of households and having mentioned the contextual impact on choosing this strategy, the strand of studies in this  As Selier mentions: “Rural-urban migration is not the only option when the source is no longer sufficient to reproduce the household.” There are many other alternatives, such as: altering consumer needs, reinterpretation of the sexual division of labour to market production, seeking new local non-farm opportunities, commuting to nearby towns, etc. (Selier 1988, 29).  95  branch may be characterized by relying on household labour allocation and interactive contextual decision-making (Goldscheider 1984).  To mention a few theoretical and empirical studies in this branch, three studies published by the Brown University Studies in Population and Development are good examples. The study undertaken by Lee (1985), associates community-level factors of rural out-migration in the Philippines with individual ones and shows contextual influence on explanation of 132 He postulates that the competition for land in migration behavior (Lee 1985, Ch.2&6). the study area is “the most decisive factor determining the probability to move’ (Lee 1985, 122).  The other study by Findley (1987) stresses interactive contextual decisions made by families in the Philippines. She argues that migration for gaining a certain amount of cash is the family strategy when the local community cannot provide such a complementary income to the insufficient farm income (Findley 1987, 6). She then concludes: “From the farmers perspective, farming and migration are an integrated system, not discrete, exclusive choices” (Ibid.). Findley hypothesizes that: “Families of the same socio-economic status are more likely to migrate if they live in communities with a lower level of socio-economic development.” (Findley 1987, xxii)  The third study, conducted by Guest (1989) in Javanese villages, investigates the linkage between household labour allocation and rural out-migration. He contends that there are two strategies for household responses to dwindling income: production  [via  specialization or diversification] and demographic [via labour mobility, whether  132 The study shows that both the commitment to family, job, and place, and community resources of origin, as well as resources for moving, are the major determinants of migration intention (Lee 1985, 117-122).  96  commuting, migration, or fertility reduction] (Guest 1989, 32-33). Guest found that a substantial proportion of the differences among villages in the amount of labour allocated through migration could be explained through the patterns of land ownership and access. He writes about the findings in the cases: “...variation among villages in the amount of labour allocated through migration are not explainable by processes operating within households. Instead I have argued that the economic and social institutions within villages play independent roles in affecting patterns of labour allocation. (Guest 1989, 183)  Guest construes that rural non-agricultural income is the key to curtailing labour outmigration (Ibid. Ch.7).  Since this branch is rather new, not much criticism can be found. A point of strength may be enumerated in its elaboration through empirical works and developing grounded theory. The weak point of this branch lies in its vagueness in pinpointing what constitutes the cause of migration in general, and in discovering what the weight is of each level of forces in different contexts? After all, this branch has more capability than other branches in all levels to synthesize the broad reasons of migration, working at different levels.  4.3.3 Summary  Migration theories at the meso-level view migration as a household decision for its maintenance or  well-being. In pursuit of conceiving the mechanism and dynamism of this  decision-making, the macro social forces and micro individual forces are contextualized at the household level. The macro forces can be pursued on general societal and cultural terrain and/or on the local community terrain. The micro forces entail a wide range of individual motives acting on the basis of migrants’ characteristics.  97  I have categorized two branches of theories at this level. The first branch concentrates on the economic behavior of the household and, using a labour migration framework, stresses collective rationality instead of individual rationality. One theory in this branch explains migration in relation to the family perception of relative deprivation (Stark 1984, 1991). A perception which is emanated from the comparison with a reference group, and not all groups.  The major cause of rural migration is attributed to relative low-income,  precarious agricultural income, capital-market deficiency, and supremacy of urban values.  The second branch places more emphasis on community contextual forces and associate them with household labour allocation (Goldscheider 1984). Migration is interpreted as household sustenance strategy and as one option in combination with or in lieu of other options (Wood 1981). From this point of view, the major cause of rural migration may be cited as: the lack of local employment opportunity (especially in non-farm sector), population pressure, poor agricultural resources, limited access to land, and social inequality in general.  In sum, much controversy exists over the excellence of micro- or macro-levels of migration theories. As a matter of fact, by evidence of the preceding literature review, none of them could theoretically or empirically assert to be exhaustive and exclusive. To date, the most appropriate conceptual framework toward interrelating and probably synthesizing the causes of rural out-migration in the South, has been approached by what I called as meso-level theories of migration. Despite the lack of enough elaborated theory, the promising feature of the meso-level is its capability to interrelate different levels and to connect insights of all levels in a holistic approach.  However, the presumption that  migration is merely a household affair, should also give space to the individual and unlinked migrations (Connell et al. 1976, 90) which occur in some cases as well.  98  4.4 The Iranian Literature Implications  Back to the policy explanation in sections 3.3 and 3.4 and having reviewed the migration studies on Iran (almost exclusively studying the last pre-Revolution decade), the following causes for rural out-migration have been specifically accentuated among others: 133 •  Population pressure: population growth has outpaced the growth of agricultural resources and have been exacerbated by heritage traditions and land fragmentation (Ashraf 1977; Mahdavi 1982; Hamidi et al. 1983; Alizadeh & Kazerooni 1984; Zahedi et al. 1985; ILO in: Nattagh 1986; Karshenas 1990a; Majd 1992).  •  Employment opportunities: new employment opportunities in the cities have been greater than the rural ones and the very low rate of investment in rural areas has been observed (Adibi 1977; Ashraf 1977; Nik-Kholgh 1979; Kazemi 1980; Zahedi et al. 1985; Ramin 1988; Vosoughi 1990; Mohtadi 1990; Lahsaeizadeh 1993; Schirazi 1993).  •  Rural-Urban income gap: average household income, wages and agricultural profitability in the rural areas have been less than those in the urban ones (Keshavarz et al. 1976; Khosravi 1978; Kazemi 1980; Alizadeh & Kazerooni 1984; Azkia 1986; Hakimian 1988; Kamiar 1988; Shafigh et al. 1992; Schirazi 1993).  •  Unequal land distribution: mostly attributed to the incomplete and adverse effects of the Shah’s land reform, landlessness and insufficient land-holdings have been understood as the determinants of out-migration (Paymaan et al. 1976; Nik-Kholgh 1979; Alizadeh & Kazerooni 1984; Azkia 1986; Vosoughi 1990; Lahsaeizadeh 1993). Land fragmentation, mostly due to the inheritance tradition and population pressure, is also part of the inefficient land distribution causing diseconomy of scale (Papoli Yazdi & Hossein Poor 1992, 176).  133  In Iran, because of the huge varieties among the rural areas across the country, it is virtually  impossible to ascribe general  common causes for out-migration. The following causes should therefore be treated as the most recurrent ones.  99  •  Mechanization: as a result of inequalities in land distribution and government facilitation, inappropriate technological change in rural areas is biased against labour (Paymaan et al. 1976; Adibi 1977; Ashraf 1977; Harnidi et al. 1983; Zahedi et al. 1985; Danesh 1987; Lahsaeizadeh 1993).  •  Institutional and price mechanism: not only the terms of trade between urban and rural areas, but also, provision of credits and other services favored cities and reinforced the inequalities (Alizadeh & Kazerooni 1984; Rafi’e-Pour 1985; Danesh 1987; Karshenas 1990a; Schirazi 1993).  •  Awareness of urban opportunities: precedence of migration patterns, expansion of transportation, communications and literacy, have increased contacts and awareness and reduced the deterring effect of distance (Adibi 1977; Nik-Kholgh 1979; Kazemi 1980; Hemmasi 1980; Ramin 1988; Vosoughi 1990).  •  Cash Needs: the somewhat self-reliant and closed economy of villages was changed and the cottage industries and rural crafts production decreased due to new consumption patterns and rural female migration/education. Wage-labouring in cities became a device to earn cash income for new needs (Khosravi 1978; Mahdavi 1982; e-Pour 1985). t Rafi  •  Cultural Discrimination: the urban-biased policies of governments and its cultural components, such as conspicuous consumption pattern, inculcated the supremacy of city life-style before the Revolution and nurtured the aspiration of breaking away from the village norms (Paydarfar 1974; Keshavarz 1976; Mahdavi 1982; Rafi’e-Pour 1985; Vosoughi 1990).  •  Natural calamities: drought, flood, earthquake and agricultural pests are among the natural calamities causing mass migrations in some cases (Nik-Kholgh 1979).  Other activities such as: accompanying the head of the family, pursuing an education, escaping from village disputes, breaking away from rural norms, intermittent nature of agriculture, etc. have been considered as subordinate causes. 100  Given this interconnected causality, the array of foregoing causes of migration may be reduced to two root-motivations as following: 1. Subsistence motivation: encountering pauperization and degrading factors in the rural  area, many villagers are pushed to leaving the natal community for another place. 2. Enhancement motivation: in response to inequalities in opportunities among human settlements, some villagers are pulled to the conceivably better places, which are often urban centers. Subsistence motivation is more objective and related to absolute poverty, being often an act of desperation for the poorest villagers. On the contrary, the enhancement motivation is more subjective and associated with relative deprivation, being often an act of preference by wealthier villagers (Lipton 1980). A combination of both motivations are found in most migration streams in the South.  As to the cause of rural out-migration studied in this dissertation, the stated two rootcauses are probed in the Iranian case after the 1979 Revolution. This might be achieved through investigating prerequisite questions about characteristics of migrants and their households and logically locating their cause of migration in one or both of the categories.  4.5 Implications for the Role of Infrastructure in Rural Out-migration In reviewing the migration theories at different levels, we observed a wide range of causes for migration. It was demonstrated that all the theories have something to contribute,  while none might claim exclusive competency.  With reference to the new emerging  paradigm in social science, it seems a pluralistic approach can be more safe in face of the  101  34 accepted uncertainties and lack of a received theory in the field.’  Indeed, the  continuums such as: micro- to macro-level theories, economic to non-economic reasons of migration, and freewill to deterministic views, all call for an integrated and inclusive approach. Several scholars have reached the same conclusion from different aspects (for example: Singh 1991, 11; Findley 1987, 16; Gardner 1981, 88; Hugo 1981a, 222; Wood 1981, 338; Todaro 1976, 51).  In fact, it is self-evident that while migration streams are basically a response to change,  whether spontaneous or induced changes, and either for worse or for better, it will not occur unless certain conditions facilitate the actualization process. These conditions are distinguished at different contexts: an individual within the household context, and the household within the community context, and the community within the regional-national or even global context. That is to say, there is an embedded or nested set of contexts which influences the potential migrant decision. (Figure 4.1)  For instance, in the South the overall motivation for rural out-migration can be attributed to the macro-level factors (e.g., inequitable resource distribution). Still this needs to be conceived by the potential migrant (e.g., information flow by previous migrants) and facilitated by community factors (e.g., norms and structures) as well as by household (e.g., division of labour and family ties) and individual factors (e.g., personal characteristics and moral obligations).  Considering this interrelated contextual nexus, the complementary  Concerning the new paradigm in social science, refer to Uphoffs book which contends the me chaiistic and reductionist way of thinking should give way to a probabilistic one, avoiding polarity of concepts and using chaos theory (Uphoff 1992, Ch.14), Also, some alternative paradigms posit epistemological problems in knowing things as they “really are” and “really work” and refute simple causation because: “Any observed action is the instantaneous resolution of a large number of mutual, simultaneous shapers, each of which is constantly shaping, and being shaped by, all other shapers.” (Guba & Lincoln 1989, 106). Regarding the cause of migration, this implies a holistic/inclusive view. 134  102  nature of different levels of theories of migration is indispensable and should be incorporated into our research.  Figure 4.1: Multi-Level Contextual Factors in Migration Decision-Making  f  Source: Author  4.5.1 A Conceptual Framework for the Roles of Infrastructure The following comparative tables (4.1,4.2, and 4.3) present a concise translation of all migration causes, suggested by different theoretical branches, into the potential roles for infrastructure in the South rural areas at each level. In the first place, a taxonomy of the causes of rural migration according to varied theories is listed and then their implications for rural policies, in terms of coping with deficiencies, are inferred.  Subsequently, the  relevant roles of infrastructure in carrying out policies are drawn.’ 35 The interpreted roles 135 It should be noted that these roles are not necessarily intended to stein the migration, but to enhance the rural development. Indeed, the provision of some infrastructures may play a contradictory role by easing rural out-migration or as some argues reducing the deterrent effect of intervening obstacles (Lee 1966). For instance, in the absence of balanced sectoral and spatial policies, building a road or a school all alone in a deprived village is more probable to encourage city-ward migration than retarding it (cf. Rhoda 1983; Findley 1981; Connell et al. 1976, 15-6). These discourse will be elaborated in the  following chapters.  103  of infrastructure are by no means a comprehensive list; however, an attempt has been made to cover all the potential areas. As suggested at the beginning of this section, an inclusive and multi-level approach has been employed here.  In keeping with the major concern of this dissertation, the stated potential roles of infrastructures in the preceding comparative tables are categorized into the five ensuing categories. The parentheses at the end of each category show the merged roles: 1. Service Delivery Facilitation: Expanding of service accessibility, whether in cities or in rural areas, and enhancement of the service quality are offered by infrastructures. (Including: providing financial services, delivering family planning services, and health and literacy improvement roles) 2. Employment Generation: This is achieved through direct employment generation, as during construction or operation and maintenance of infrastructures, and indirectly by facilitating investment in rural sectors. (Including: economic diversification role) 3. income Upgrading: Infrastructures increase outputs, and decrease inputs, marketing, and transportation expenses.  Also, they provide technical skills for better resource  mobilization. (Including: facilitation of better resource mobilization role) 4. Rural-Urban integration:  A viable rural settlement pattern is brought about by  expansion of infrastructures and articulation with the city centers. Here the dichotomy of rural vs. urban and the isolation of rural activities are blurred. (Including: improvement of service accessibility role) 5. Rural  Institutionalization:  Infrastructures  play  an  important  role  for  institutionalization of rural capacities, creating self-reliance, and more bargaining power with cities through providing better communication and organizational arrangements. (Including: cultural/community programs enhancement role)  104  Table 4.1: Implications for the Roles of Infrastructure in Rural Out-Migration according to the Macro-Level Theories Theoretical Branch Cause of Rural Out-Migration Dual Economy and Low Productivity of Rural Modernization Sectors and High Theory Productivity of Urban (Modern) Sectors Unfavorable Resource Marx’s Views and Dependency Theory Allocation and Structural Constraints to Accessibility Ecological approach Limited Agricultural Lands relating to High and Malthusian Fertility & Environmental Theory Degradation Source: Author  Rural Policy Inference  Potential Role of Infrastructure Equilibrating Agriculture Employment Generation, with other Economic Rural-Urban Integration, Sectors and Diversifying Income/Efficiency Rural Economy Upgrading. Balancing Rural/Urban Improvement of Service Development and Accessibility, Socio-Political Empowering Rural Institutionalization. People Upgrading Resource Facilitation of Better Mobilization and Resource Mobilization, Promoting Family Health and Literacy Planning Improvement.  Table 4.2: Implications for the Roles of Infrastructure in Rural Out-Migration according to the Micro-Level Theories Theoretical Branch Cause of Rural Out-Migration Human-Capital and Income differences during lifetime, WageWage-differential! differences & Job-Probability Employment Probability Theories Dissatisfaction with Stress-Threshold! Place Utility and Place Utility & Rural Community, Urban LifeBright Lights Theories  Style Attraction  Rural Policy Inference  Potential Role of Infrastructure Improving Rural People’s Employment Generation, Income vs. Urban Economic Diversification, People’s, Investing in Income/Efficiency more Rural Jobs Upgrading. Upgrading Rural Quality Amenities/Service Delivery, of Life, Valuing Rural Cultural/Community Culture, Enriching programs Enhancement. Community Building  Source: Author  Table 4.3: Implications for the Roles of Infrastructure in Rural Out-Migration according to the Meso-Leve! Theories Theoretical Branch Cause of Rural Out-Migration Relative Deprivation Low Income, Unstable Theory and Value- Agri. Earnings, CapitalExpectancy Models Market Deficiency, Urban Values Household Labour- Insufficient Employment, Poor Allocation and Contextual Factors Agri. Resources, Social Inequality, Population Models Pressure Source: Author  Rural Policy Inference  Potential Role of Infrastructure Upgrading Rural Income, Income/Efficiency Improving Upgrading, Providing Banking/insurance, Financial Services, Cultural Resuscitating Rural Values Programs Enhancement. Expanding Non-Farm Employment Generation, Jobs, Undertaking Land Socio-Political Reforms, Promoting Institutionalization, Family Planning Delivering Family Plan Services.  105  All these potential roles are supposed to fulfill the underlying goal for the infrastructure, that is equal accessibility to opportunities, and thereby provide a choice for villagers: that 36 This accessibility can be prepared in two of not being compelled to leave their home.’ phases.  First, the physical requirements (e.g., establishments and equipment) must be  provided and secondly, the service must be functioned and maintained (e.g., supplying human resources and efficient management). These two phases are inseparable for the proper fulfillment of infrastructures’ roles. It is inferred that the process of infrastructure construction, in addition to the process of its operation, are important to its function in the aforementioned roles.  The case study of this research is basically intended to investigate the capabilities of the above-mentioned five categories in affecting rural out-migration. However, this might be achieved through investigating prerequisite questions, as is discussed in the next chapter.  Let us elaborate on a broader meaning of equal accessibility, or as some call it, spatial equity (Stohr & Todtlmg 1979). Often, the spatial outcome of conventional resource allocation in national planning is not reckoned and is left to spontaneous forces. In retrospect, such a spatial development tends to concentrate in few centers and the resultant concentration causes a better opportunity for those who have the advantage of spatial access. Hence, development space with regard to the general shortage of resources in the South should be considered as another scarce resource. This space must be allocated as equitably as any other resource to avoid regional disparities. However, even with equitable national allocations, many areas will still not receive many of the fruits of development. To increase the number of beneficiaries, the created opportunities ought to be accessible to more than the geographically immediate people. Expanding this accessibility (e.g., by reducing friction of space) is a major goal of infrastructures. 136  106  CHAPTER V: RESEARCH METHODOLOGY  107  5.0 Introduction  Reviewing a wide variety of migration theories in the previous chapter, it was clarified that a combination of the theories were needed for a holistic approach to the complex and multi-faceted issue of rural out-migration. necessity of  What this implies for methodology is the  a pluralistic approach to accommodate a variety of theories’ needs.  Accordingly, a combination of research methods were adopted with respect to multi-level needs which supplement each other.  As noted in Chapter I, the principal purpose of this research is to explain the causal relationship of rural out-migration and infrastructure availability/non-availability from a rural development standpoint. The purpose is not categorized as exploratory, descriptive or predictive research, but rather relational and explanatory research (Northey & Tepperman 1986, 21; Marshall & Rossman 1989, 78; Babbie 1991, 90-92). Keeping this in mind, this chapter presents the appropriate approach to the research method and the selection of the levels of analysis. Then, the research design is discussed and the research components including: units of analysis, variables, measurement, data collection, research strategy, cases selection and sampling are explained.  5.1 Approach to Migration Study  Within the three levels described for migration theories, I identified the meso-level as the most appropriate theoretical level (see section 4.3.3). This has certain implications for the research design in terms of units of analysis which will be discussed later. But before elaborating on this, the general approach to migration study needs to be selected.  Two distinctive approaches in studying causes of migration are distinguished.  First, a  structural approach, as was noticed in macro-level theories (section 4.1), seeks to 108  understand the reason of movement beyond individual will and from contextual and historical perspectives which are embedded therein. It stresses the effect of context on behavior (Uphoff 1992, 330). The structural approach is appropriate for examining the rationality and selectivity of broad movements of population (Sylvia 1992, 19).  Second, a cognitive approach, as was mentioned in micro-level theories (section 4.2), is concerned mostly with the cause of migration from individuals’ point of view including their ideas, values, and ideals (Uphoff 1992, 331). important role (Boyer & Ahn 1991, 56).  The migrants’ perception plays an  This approach provides better insights into  situational and cultural specifics in the migration process (Sylvia 1992, 21).  In the theoretical review it was elucidated that the migration process took place as the interface of different level factors.  Ostensibly, an individual makes a decision.  Upon  further investigation, the influences of household and community become evident. Continuing in more profound study, the regional and national level factors are linked and set the ground for population movements.  So, while migration intention can be  discovered with a cognitive approach, “the actual migration takes place only when the political, legal, and economic constraints have been examined” (Boyer & Ahn 1991, 56). Consequently, cognitive and structural approaches are complementary, and the concomitant use of them is conducive to a thorough understanding of migration cause.’ 37 This research relies on both approaches.  5.1.1 On Searching for the Cause of Migration  To understand the cause of migration, as a scholar states, 137 Empirical studies have well shown that individual characteristics are a very important factor in the explanation of migration behavior. However, the same literature also shows, after controlling for individual characteristics, that migration behavior differs widely due to social setting (Findley 1987, Guest 1989).  109  either we accept the migrant’s own statement of motive, or we infer motives from a study of objective structural determinants and then impute these motives to the migrants. The third possibility is that we combine the migrant’s subjective account of motives with our own account based on objective inference. (Taylor 1969, 99)  The present study has selected the third possibility. Indeed, as will be explained later, three different ways to find the cause of migration have been examined: 1. Surveying Individuals’ Perception 2. Investigating Key-Informants’ Perception 3. Making Objective and Statistical Inferences Finally, all these have been synthesized, and with a logical argument the overriding causes of rural out-migration have been postulated. Both quantitative and qualitative data are 38 included.’  The traditional way of approaching the cause of migration by asking the individual has been widely criticized (see De Jong & Gardner [eds.] 1981). The main flaws are: •  The respondent tends to state the last straw or triggering reason which may conceal the root causes.  •  The respondent is required to recall the cause(s) of migration about their prior behavior which may be forgotten or not correctly accentuated.  •  The respondent may state rationalized excuses to the decision to migrate according to his/her latest understanding, social values, etc., which may not be the actual cause at the time of migration (De Jong & Fawcett 1981, 34).  •  Categorization of the respondents’ terms is not accurate; this involves semantics.  138  Since there is a tendency to rely more on quantitative data, Patton’s critique of this method is  noteworthy. He asserts that it: “1) oversimplifies the complexities of real-world experience, 2) misses major factors of importance that are not easily quantified, and 3) fails to portray a sense of the program and its impact as a whole.” (Patton 1990, 49-50)  110  And in general, it requires an articulated response, which is difficult for many people in the rural South.’ 39  The next flaw in the above individual questioning is that it overlooks other causes which may be found if the household are interviewed.  Collective rationality could be quite  different from individual rationality (Goldscheider 1984, 302; Grindle 1988, 27). Seeking the household’s cause of migration, still does not necessarily include other factors contributing to the decision to migrate by the individual.  Household functions in  structural and community setting and is affected by related objective factors(e.g., existence of migration patterns, resource accessibility and local labour market opportunities) and subjective factors (e.g., community values, family ties and social bonds).’ ° 4  It bears repeating that migration is enmeshed within a very broad range of issues. And a complex interplay of a great variety of factors at different levels is involved (Hugo 1981b, 179; Wood 1982, 314).  Hence, a cross-level analysis of responses and inferences is  preferred for understanding the cause of out-migration. In spite of all rigor, realizing the cause of migration is problematic. Lee correctly has warned: ‘Since we can never specify the exact set of factors which impels or prohibits migration for a given person, we can in general, oniy set forth a few which seem of special importance and note the general or average reaction of a considerable group.” (Lee 1966, 49)  Some scholars talk about tacit knowledge “as the inner essence of human understanding, what we but can’t articulate” (Patton 1990, 72). This should be more of a problem for the illiterate villagers. For example, I could not use a semantic differential scale [choosing between two opposition  know,  positions] or a Likert-type scale [standardized responses] (Babbie 1992,180-181) for gauging the weight of each contributing factor in migration decision making. 140 With the same view, Grindle observes: “Households and the decisions they make are constrained by the development potential of their community, the policies historically adopted by governments that affect them, and the characteristics of both national and international political economies.” (Grindle 1988, 28)  lii  All in all, probably the most safe way to approach the cause of migration is to use crossvalidation to corroborate the finding at each level. That is why a holistic and multi-level approach has been adopted in this research.  5.1.2 On Searching for the Infrastructural Cause  Having discussed the complicacy of searching for causes of migration, now I focus on the main purpose of this research, i.e., searching for the importance of infrastructure motive in the decision to migrate from rural areas.  Here, in addition to the aforementioned  difficulties, we deal with the problem of weighing and priorizing the causes. Needless to say, this cannot be accomplished by questioning the individual alone and requires additional information gleaned with other levels of analysis. Finally, the integration of all this information and the building of a logical argument, will shed light on the inquiry.  A particular problem, probably not exclusive to Iran, is that if interviewer is seen as an outsider, either the villagers censure their ideas and resort to rhetoric when answering sensitive questions, or try to portray a bleak picture of the situation in the hope of getting more of the state allocations for infrastructural projects.  In fact, numerous previous  attempts (e.g., for income data) have shown the inaccuracy of this kind of direct soliciting. Instead, lateral questions which are the building blocks of decisions to migrate are more effective in obtaining the right motives.  The approach for the survey, in sum, should be  to acquire knowledge of the ingredient of migrant’s decision for our final inference, rather than simply the accumulation of direct responses.  In the case of community-level interviews, there are fewer constraints than in the individual interviews since quite often, the people expressing more precise data for co villagers and for the village in aggregate. The elders and community leaders especially feel a strong moral obligation to present the situation as truly and as fairly as possible. The 112  presence of other respected villagers fortifies this conscientious feeling.  Thus, the  approach for the group interviews, in contrast to the individual survey, could be through direct questioning, though associating the ingredients of typical decision-making regarding moves, as well.  5.2 Level of Analysis In a seminal work, Germani (1965) distinguished three levels of analysis in the study of causes of migration, as follows: 1. Objective level, comprising of the so-called push and pull factors, and also the nature and conditions of communications, accessibility, and contact between origin and destination. 2. Normative level, which contextualizes the objective influences according to the norms of the particular society of origin. 3. Psycho-social level, entailing the attitudes and expectations of individuals, different from the normative pattern (Hugo 1981a, 189190)141 This conceptualization highlights the necessity of investigating migration cause in different and interdependent levels. Once again, the multifold and multi-level causes of migration are stressed. This, in return, calls for employing an inclusive research method at different levels.  5.3 Research Design The main research questions involve the cause of rural out-migration and the significance of infrastructure motives therein.  Until now, the theoretical framework of meso-level  141  In this regard, Taylor (1969) developed the above discussion and argued that each level conditions the others, the objective level having priority While it apprears that priority of the objective level is valid, the interplay of other levels should not be undeffated.  113  theories has been adopted and also an inclusive research method in terms of general approach and level of analysis has been stressed.  Now, the research components are  delineated.  5.3.1 Units of Analysis Due to the explained inter-connection of causes of migration at different levels, there are four units of analysis in the present research, as follows: 1. Dehestan (a rural jurisdictional unit in Iran), for aggregate analysis of the relationship of independent and dependent variables at the macro-level, from its structural and objective aspects. 2. Community or village, as the unit for investigating the migration causes from its structural, objective and normative aspects. 3. Household, as the key unit at the meso-level which links the community and the individual units, particularly covering its cognitive and normative aspects. 4. Individual, as the base unit of migration at the micro-level for scrutinizing its cognitive and psycho-social aspects.  5.3.2 Variables and Measurement  As mentioned before, this study in its basic type is a relational study Northey & Tepperman 1986, 10) between one independent variable [i.e. infrastructure availability] and one dependent variable [i.e. rural out-migration] (Figure 5.1).  The importance,  however, of conditioning and intervening variables [e.g. the migrants personal characteristics, household status, kinship ties, village economic condition, accessibility and performance of infrastructures] that effectively mediate between the former variables, incorporate explanatory traits as well (ibid. 17).  114  Figure 5.1: Diagram of the Research Major Variables  (Independent Variable)  (  (Intervening Variables)  )  (Dependent Variable)  /  Determinants Motivation/ Perception  Behavior/ Intention  ant& Household Characteristics, Community Norms, Societal Structures, etc. Source: Author  There are several more variables regarding the subsequent questions (stated in section 1.6) which have been reflected in questionnaire design and will be described later in the next section.  The study relies mainly on quantitative methods for measurement at the household and the individual levels, while for the village level qualitative methods are utilized for a holistic conclusion (Marshall & Rossman 1989, 49). All the levels of measurement, i.e., nominal, ordinal, and ratio, are employed with respect to different variables.  5.3.3 Data Collection  The methods of data collection for social scientists are identified in four categories,  namely: experimental research, survey research, field research, and unobtrusive research (Babbie 1992, 234; Singletone et al. 1993,  179).142  The relevance and application of each  to the present research are discussed in the following. 142  Babbie also talks about a fifth kind which he calls evaluation research (Babbie 1992, Ch. 13). He admits, however, that this “refers to a research purpose rather than to a specific research method” (Ibid. 346). This kind of research, according to Babbie, is most pertinent to evaluating a social intervention in 115  Experimental research, usually associated with physical sciences, requires manipulation and control over the research environment to test the causal processes (Singletone et al. 1993, 181). It usually tests “the effect of an experimental stimulus on some dependent variables through the pre-testing and post-testing of experimental and control groups” (Babbie 1992, 258). This method is not relevant to the present research, whereas there exists no control over the research environment (i.e., rural communities, households, and individual migrants). Besides, no control group can be used to make sense of comparing the effect of infrastructure availability/non-availability on rural out-migration, for every village has its unique complex of socio-economic, historical and geographical factors which contribute with different weights at different times.  Survey research is useful when there is a population too large to observe directly (Ibid. 262).  A representative sample is selected for data collection through questionnaires,  interviews, etc. (Jackson 1988, 32). Typically, an individual should be asked to respond to the questions, even if the unit of analysis is other than individual.  This method is  appropriate for part of this research, because of lack of existing data for each village, and in general for rural households and individuals. For the household and the individual units, a questionnaire was developed to be filled by face-to-face questioning and with emphasis 143 A semi-structured interview also was designed for the village unit on quantitative data.  light of the intended results. This dissertation uses this kind of research as the outcome of answering the questions of the causes of migration and the significance of infrastructure. In other words, after concluding the relationship of infrastructure and rural out-migration, the policy implications are logically drawn. The other ways of administrating the survey are categorically negated. Because of high illiteracy, self-administered questionnaires exclude a large portion of the rural population (over 50 percent in 1986). Considering the lack of availability of telephones in the rural area of Hamadan, obviously, telephone interviews are irrelevant. Both types are outlandish for authentic communication in Iranian rural culture. 143  116  with much more emphasis on qualitative data (Fontana & Frey 1994). 144 In terms of the time dimension, both designs are cross-sectional with some questions included to approximate longitudinal aspects for certain variables.  Field research refers to the probing of social phenomenon in its natural setting (Singletone et al. 1993, 349). Using direct observation by the researcher, whether in a “participant” or “non-participant” manner, this method more typically yields qualitative data, and deeper and fuller understanding (Patton 1990, 10; Babbie 1992, 285).  Field research is  appropriate for studying a comprehensible-size community such as a village and for focusing on human behavior and groups’ attitudes (Ibid. 286). Thereby, it is useful in the present dissertation for researching migration behavior on village context.  My active  involvement for more than four years in a comprehensive development plan for the rural 45 prior to starting this research, enabled me to observe the areas of Hamadan Province,’ migration phenomenon, its causes and consequences in a longer period of time and in a broader geographical sphere. In addition, this helped me gain access to sampled settings without appearing to be such as an outsider.  Unobtrusive research or non-reactive research implies the study of social behavior without affecting it (Babbie 1992, 342). As opposed to the foregoing research methods, this method relies on the available data (Singletone et a!. 1993,  387)146  This dissertation  utilizes the unobtrusive research method as a complementary method at the macro-level 144  To make a valid case and adding to the rigor, breadth, and depth of analysis, use of multiple methods, or triangulation is suggested by many researchers (Denzin & Lincoln 1994, 2). In this sense, the supplementary data collection methods have been considered in this research. From 1987 to 1991, as the head of The Plan coordination committee, and also in charge of spatial development synthesis, the author and his collaborators prepared the first volume of The Comprehensive Rural Development Study of Hamadan Province. 145  146 The analysis of these available data could be of three types: analysis of secondary information, content analysis, and historical/comparative analysis (see Babbie 1992, 312; Singletone et al. 1993, Ch.12).  117  for laying the background of government policies and for studying cardinal historical events which affect population movements in Iran (see Chill). The existing data for a rural administrative unit called Dehestan, are useful for aggregate analysis of the effect of infrastructure availability/non-availability on rural out-migration trends from a quantitative point of view.  Moreover, the content analysis is appropriate regarding the analysis of  village-level interviews and their qualitative data (Marshal & Rossman 1989, 98).  Bearing in mind that these methods are not mutually exclusive, in sum, an integrated approach is taken to complement the data collection at various levels (Harrison 1987, 21). The levels and the data collection methods in this research are listed below: •  Macro-level with the whole country as the overall context: using national existing data and reviewing government policies and major events as were seen in Chapter III 47 [unobtrusive research].’  •  Macro-level with Dehestan as the unit of analysis: relying on existing provincial data as are displayed in Appendix G.  •  Meso-level with village as the unit of analysis: doing group interviews and field observation for contextual factors as are shown in Chapter VII [survey and field research].  •  Meso-level with household as the unit of analysis: conducting questionnaire surveys for the remaining migrants’ household at the village as is presented in Chapter VI [survey research].  •  Micro-level with Individual as the unit of analysis: as part of the household questionnaire survey for obtaining individual data and it is described in Chapter VI [survey research].  The integrity of these data is discussed in the research strategy. 147  Using a limited macro-level rural migration analysis with the help of national census in different  years, adds a longitudinal dimension to the so-far cross-sectional approach of this study.  118  5.3.4 Research Strategy  The principal research questions of this research may be reduced to: Why do people migrate from village? and How important is infrastructural cause in their decision? To investigate these, the case study strategy has been undertaken which requires no control over behavioral events and is appropriate for why and how questions (Yin 1989, 17).  Also, case study research has the ability to include a variety of research methods (e.g., qualitative and quantitative, survey, field, and unobtrusive research) in order to deal with 48 the contemporary phenomenon in its context.’ survey,  As mentioned above, questionnaire  community key-informants’ interview, and field observations have been conducted  in each selected case.  The integrity of the case study is suitable for synthesizing the  qualitative and quantitative findings for a holistic approach to the cause of migration (Stake 1994, 245).  It should be stated that the research strategy consists actually of embedded case studies (Yin, l989:p.46). To put it another way, the case of rural households is embedded in the case of sample villages which are embedded in the case of Hamadan Province of Iran.  5.3.5 The Cases Selection and Sampling Design  To investigate the research questions, a rural-based province named Hamadan, which has 49 Since the a recorded history of three decades of high rural out-migration, was selected.’ 148 In this regard, the Yin definition is elucidating: “A case study is an empirical inquiry that: investigates a contemporary phenomenon within its real-life context; when the boundaries between  phenomenon and context are not clearly evident; and in which multiple sources of evidence are used.” (Yin 1989, 23) All these fit precisely the objectives of this research. Interesting to know, Yin holds that “case study research is among the hardest types of research to do.” (Ibid. 61) ‘  The 1986 census showed that 62.6 percent of the Province’s population resided in rural areas, as compared with 45.2 percent for the whole country. Since the first national census in 1956, Hamadan Province has had a constantly lower rate of population growth than the national rate and, consequently,  119  Islamic Revolution, Hamadan Province has benefited from a higher resource allocation for rural development compared to other provinces, and the rural coverage of social and physical infrastructures is significantly better than the country’s average.’ 50 Moreover, an important rural development program (i.e., Beh Sazi-ye Roosta or rural renovation) was first initiated and implemented in this province, and later expanded to be country-wide. It should be added that the researcher had worked for more than four years (1986-1990) as one of the principal planners in charge of Comprehensive Rural Development Plan for Hamadan Province. Thus, the ample background information and access to the whole study area was a privilege for me.  In the case of Dehestan, the existing data were extracted from the latest two national population censuses in 1976 and 1986, and from the agricultural census of 1988.  The  village-level data was gathered through field observation and a semi-structured interview with open-ended questions from village key-informants in a panel-discussion setting (Yin 1989, 89). For primary data collection, a stratified and multi-stage random sampling for villages (Jackson 1988, 162), and a survey comprising of simple random sampling for individuals and households were employed (Babbie 1992, 211).  To find the stereotypes of out-migrating villages with implications for the infrastructure provision, the following steps were taken:  its population share has dropped from 3.8 percent of Iraii’s population to 3.2 percent between 1956 and 1986, respectively. 150  For instance, in the First Five Year Development Plan of the Islamic Republic of Iran, Hainadan’s rural areas, with 4.21 percent and 1.84 percent of Iran’s rural population and rural centers respectively, had been earmarked with 5.16 percent of public investments (computed on the basis of detailed allocations in the Plan by the author). As an example regarding infrastructures, 69.3 percent of Ha.rnadan’s rural population was covered by primary health care network as compare to the average of 53.5 percent for Iran’s rural population in 199 1(World Bank 1993b, 53).  120  a-  In the Province, all the villages which had lost part of their population during 197686 were marked on a map, also showing different population ranges (e.g., 1-99, 100249, 250-499,  b-  ...).  Due to easy access to urban infrastructure and the blurred boundary of urban and rural areas, those of the marked villages which were located within the 30-minute isochronal zones of cities, were eliminated.  c-  Using the findings of an analysis extensively undertaken for the Province (Sarrafi et al. 1991, 441-450), all the villages with over 19 kinds of public services (as the superior threshold to be surpassed) were identified.  Among five identified  depopulating villages in this service category, two sample villages were randomly selected from each of the two main geographical areas (i.e., the Northern and the Southern halves of the Province, separated by a high mountain range. (see Figure 5.2). d-  Because of the economy of scale in general, and resource constraints in particular, it is impractical to provide basic public services, substantively, for each rural center ’ Thus, for the remaining sample village 5 with a population of less than 250 people.’ selections, those with populations of under 250 people were eliminated.  e-  Overall, five peripheral sample clusters as the typical manifestation of rural outmigration were distinguished in the Province. Among them were the two previously selected sample villages. Three other sample villages were selected randomly from the remaining sample clusters.  Given a 50 percent-and-over occunence of each service (in any population range of rural centers) to be considered as a viable and feasible service in that range, there found only 3 public services (i.e., primary school, public bath and mosque) among the 25 needed ones in the 100-249 population range (Sarrafi et al. 1991, Ch.4). 151  121  Figure 5.2: Hamadan Province: Distribution of Human Settlements, Major Roads, and Highlands  Major Road Above 2000 m. Elevation  Hamedan Province  122  After selecting the five sample villages, 10 percent of their number of households were s.’ 52 determined as the sample size to which to administer the household questionnaire The sample villages are shown on the map (Figure 5.3) and the number of sample households are given in Table 5.1 (for detailed community-level data refer to Appendix F).  Table 5.1: The Sample Villages (Population, Location and No. of Questionnaires) Name of Sample  Number of Questionnaires  (1986)  Household Numbers  1469  328  1 190  177  340  59  322  59  304  56  Population  (Access Time)  Village  Maanizan  Nearest City  32  Malciyer (35 minutes)  Jorban-lou  18  Ghorveh (55 minutes)  Se-miran  6  Asad-abad (45 minutes)  Hassan-taimur  6  Mian-roudan  6  Kaboudarahang (75 mm.) Nahavand (35  minutes)  Source: Author & SCI 1990c. In each village, due to the lack of the list of households and the research constraints, with the help of a table of random sampling, the residential units were chosen by starting from 53 In each direction, first it was asked if the two vertical directions on the village roads.’ household has had an out-migrant member after the Islamic Revolution. In most cases the responses were positive.  Where the household did not have any out-migrant member, or  It is said that reduction in sampling error is possible by increasing the sample size and increasing homogeneity of the elements being sampled (Babbie 1991, 219). According to the aforementioned study of rural development planning in the Province, inter-village homogeneity is higher than intra-village one. Thus, the number of villages is reduced to the minimum number which could represent all five sub regions. The number of households in each case assigned: 10 percent; respecting the elimination of households without migrant members, it amounts to more than that. 152  153  All the five sample villages have a compact and nucleated settlement pattern. Using an elevated  standpoint or a walking reconnaissance, the principal researcher made effort to include the typical sections of the pattern in each direction of random sample surveys. Always one of the direction were started from the village center and the other from the fringe inward.  Ñ 123  Cl)  C  0  C  could not be contacted for any reason, it was excluded and replaced by the next counted house, according to random numbers and along the same road.  The household questionnaires were orally administered.  For this purpose, two  experienced interviewers were recruited from the staff of Hamadan Plan and Budget Organization, rural data collection division. They were instructed carefully in advance and were accompanied during the fieldwork in July of 1992. The respondent was chosen from amongst the adult members of the household, foremost the head of the household and then his spouse or in rare instances, her/his children.  The questionnaire contained questions on the rural households and their emigrants who had left the village after 1979, the year of Islamic Revolution (see “Appendix A” for the questionnaire format). Therefore, it should be noted that the method employed here was partially that of pursuing the respondent’s perceived reason(s) of migration for his/her household member(s).  This was deliberately chosen, whereas the study scope was  confined to the place of origin. Moreover, it had been designed to investigate the stayer’s view as an implicit proxy for his/her satisfaction with the village and to probe the attitude towards migration. The other option, which is to interview emigrants in the city, could be complementary, but not possible because of the research constraints.  Such a study,  however, requires more data than what is available for the identification of emigrants of a specific village, in order to have the same probability of inclusion in different destinations for all of them.  While the interviewers were filling out the questionnaires, in addition to supervision, I pursued my “participant observation” and personal interaction with the village leaders. At this time, the purpose and the content of research were fully described. These contacts laid the ground for my next year visits to the field and for conducting the in-depth 125  54 Indeed, without such a precedent of interviews with the community key-informants.’ trust-building, it was hard to communicate authentically and sincerely during interviews.  The village interviews were carried-out in field, over a period of one week by the author in September 1993. All interviews were taped and conducted in one of the villagers’ houses. For each interview, key community informants (mostly being male, elderly, and councilors) had been identified during my field observation in the previous year.  In  addition, at least one male youth who has been working in the village and has been vocal, was invited too. As opposed to the collection of household and individual data (which should be asked usually without the presence of others for personal information) the village interviews were designed to benefit from the dialogue, varied information, crosschecking and sometimes, conflicting interpretation among a group of villagers (Bilsborrow 1984, 427).  A set of questions to probe the causes, consequences, and patterns of out-migration in each village and the relevant RIPP impacts, as well as the socio-economic characteristics of community was raised (see “Appendix B” for interview questions). Nevertheless, due to the interactive dialogue among the participants and the open-ended nature of questions, circumstantial issues also have been addressed in each interview (Fontana & Frey 1994). ‘55  154 Ever since I designed the research, I had thought about the use of community-level data to  supplement and cross-check/complement the household survey. However, during conducting the survey I gradually arrived at the necessity of interviews with key informants as the best source of primary data for the villages. Other researchers on migration in the South, after extensive work, have stressed the same conclusion (Bilsborrow 1984, 426) 155 For quantitative data analysis, frequency tables, cross-tabulation, and correlation analysis, and for qualitative data analysis, analytic description and content analysis have been employed.  126  5.4 Conclusion In this chapter, it was shown that in response to the selected theoretical framework (in Ch.IV), there is a need for a variety of research methods to be employed at different levels of research, which finally, should be integrated. Generally, this research was categorized as explanatory and relational and was used to investigate the infrastructural causes of rural out-migration in light of rural sustainable development.  It was contended that both structural and cognitive approaches for seeking the cause of migration were essential and should be used for validation. Regarding the infrastructural causes, indirect inference for the individual and household survey, and both direct and indirect inquiry for the village interviews were discerned as being appropriate.  It was argued besides that multiple units of analysis were needed in the wake of multi-level research, including: individual, household, community, and dehestan. Correspondingly, a set of data collection techniques: survey, interview, field observation, and the use of existing data were found relevant.  These different modes of inquiry provide a better  understanding and increase the precision of research (Huberman & Miles 1994, 431).  The case study research strategy was employed to interconnect all these various approaches, units, techniques and methods. Also, the selection of embedded cases (i.e., case within case), community informants and random sampling of households in five depopulating villages in the Hamadan Province were described.  127  CHAPTER VI: SURVEY FINDINGS  128  6.0 Introduction  In this chapter, the survey findings from five sample villages will be discussed.  The  chapter begins with the characteristics and behavior of rural life-time out-migrants according to the description of the respondents (85 percent of whom being the heads of remaining households), and some of the characteristics of the originating rural households (with at least one life-time out-migrant) are summarized. Then, the stated reasons for rural out-migration are discussed. The survey findings and their analysis are discussed together. However, conclusions will be elaborated upon after the discussion of interviews, in the next chapter.  6.1 The Characteristics and Behavior of Rural Out-Migrants  In total, from 68 sample households selected from five sample villages of Hamadan Province, 238 migrants who had migrated after the 1979 Revolution were identified. The average number of emigrants was found to be 3.5 per household, of whom 3.1 persons (89.1 percent- see Table 6.1) were children of the heads of the remaining household (including 1.8 sons and 1.3 daughters- see Appendix C, Tables C. 1 & C.2). Table 6.1: Relationships of Migrants to Household-Heads  Relationship  Frequency  Spouse Child Parent Grandchild Son- or Daughter-in-law Neice or Nephew Sister or Brother Sister- or Brother-in-law Total:  Percent 0.4  1 212 2 10  89.1 0.8 4.2  8  3.4  2 2  0.8 0.8  1  0.4  238  100.0  Source: Author’s Survey in Hamadan’s Rural Areas  129  6.1.1 Sex, Age, Marital Status 57.6 percent of rural out-migrants were male; 42.4 percent were female (see Table 6.2). 156 for the rural population in 1986 was 108.1 (HPPBO 1992, 29) The provincial sex ratio as compared to 135.6 for the sample migrants.  Given the dominance of the head of  household’s decision in initiating migration, and the overwhelming majority of maleheaded households, male selectivity in migration is magnified.  Table 6.2: Distribution of Rural Out-Migrants by Sex Sex Male Female  Frequency  Percentage  137  57.6  101  42.4  238 Total Source: Author’s Survey in Hamacian’s Rural Areas  100.0  The urban migrants were mostly young, 54.2 percent were in the age group of 16-30, 33.2 percent were between 31 and 45 years, 7.6 percent were under 16 years of age and 10.5 percent were over 45 years of age (see Table 6.3). Considering that half of the Hamadan rural population is under 15.5 years age, and having reviewed the detailed age distribution of migrants, the dominant ages for city-ward migration is 16 to 32 years.  Indeed, the  mean, median, and mode are 28.5, 28, and 25 years respectively.  64.7 percent of migrants were single at the time of out-migration. However, 76.1 percent 57 (see Table 6.4). This indicates that while the of them are married at the present time’ first move is easier for singles, after establishment migrants tend to marry. The common  156  Sex ratio is calculated as number of men is divided by number of women, multiplied by one hundred.  157  Eliminating the not-responded category, the married cases are increased to 80.1 percent of total.  130  pattern is an arranged marriage with a villager from the originating place, which results in a higher proportion of married females among the out-migrants than single females. Table 6.3: Distribution of Rural Out-Migrants by Age Groups  Age Group  1-15 Years  16-30 Years  3 1-45 Years  46 & Over  Frequency  18  129  79  12  33.2  5  54.2 7.6 Percentage Source: Author’s Survey in Hamadan’s Rural Areas  Table 6.4: Marital Status of Out-Migrants at Present and at Departure Time Frequency at Percentage at Departure Time Departure Time  Marital Status of Out-Migrants  Frequency at Present Time  Percentage at Present Time  Single Married Divorced NotResponded  44  18.5 76.1 0.4 5.0  154 71 1 12  64.7 29.8 0.4 5.0  100.0 238 Total Source: Author’s Survey in Hamadan’s Rural Areas  238  100.0  181 1 12  6.1.2 Education 78.6 percent of migrants are literate (see Table 6.5). The literacy rate is increased to 79.2 percent among persons of 6 years of age and more. Comparing to 49.6 percent as the literacy rate for the rural population of Hamadan Province in 1986 (SCI 1990a, 115), out migrants are considerably better-educated (prior to migrating) than the remaining population of the sample villages.  The frequency column of Table 6.5 reveals that there is an upsurge in the number of migrants in certain groups. This is due to the end of a stage in the Iranian educational system.  The five and six years groups come at the end of the new and old primary  education curriculum and the eight years group is for those who have reached the end of  131  junior high school (namely guidance school). The results in the column reveal that high tendencies of out-migration among rural youth are created after finishing primary school and to a lesser extent after the end ofjunior high school. Table 6.5: Distribution of Out-Migrants by No. of School Years Attained  Frequency  School Years One Year Two Years Three Years Four Years Five Years Six Years Seven Years Eight Years Ten Years Eleven Years One Year ofAdult School Two Years ofAdult School None Formal Education Under Six Years ofAge Total:  3 8  5 7 61 44  Percent 1.3 3.4 2.1 2.9 25.6 18.5 2.1 19.3  Cum. Percent 1.3 4.6 6.7 9.7 35.3 53.8  0.4  55.9 75.2 75.6  2  1.3 0.8 0 .8 20.6 0.8  76.9 77.7 78.6 99.2 100.0  238  100.0  5  46 1 3 2 2  49  Source: Author’s Survey in Hamadan’s Rural Areas  The cross-tabulation of literacy and sex variables of sample out-migrants indicates that women were over-represented in the illiterate category. Women compose 81.6 percent of  illiterate migrants although they constitute only 42.2 percent of migrants (compare Table 6.6 and Table 6.2). It is true that the literacy rate of rural women in Hamadan Province (being 23.4 percent in 1986) dramatically lagged behind the corresponding figure for rural men (63.3 percent).’ 58  The same unsuitable situation is observed in the sample out-  158  The 1991 national survey shows a sharp increase for the female literacy rate in rural Harnadan, up to 58.8 percent (Hosn Bakhshan in: Sarrafi et al. 1991, 274), which cannot be confirmed in our out-migrants and households survey.  132  migrants. However, the illiteracy rate are quite higher than the aforementioned rates at the provincial level (60.3 and 93.3 percent for male and female migrants respectively).  Table 6.6: Distribution of Out-Migrants’ Sex by Literacy Status  Literate  Sex of  Illiterate  Under 6 Yrs  (Frequency)  (Percentage)  (Frequency)  (Percentage)  (Frequency)  126 61  67.4 32.6  9 40  18.4 81.6  2 0  100.0 Total 187 Source: Author’s Survey in Hamadan’s Rural Areas  49  100.0  2  Migrant  Male Female  Controlling sex variables in the cross-tabulation of age groups and school years of outmigrants shows that while the same age groups of men and women are likely to migrate (i.e., 16 to 45), the female migrants are less literate than the male (see Table 6.7).  Table 6.7: Distribution of Migrants’ Education by Age Groups and Sex Age Groups:  1-15 Years  School Years  Male  ]to4Years Sto6Years 7 to 8 Years 9to12 Years Adult School None Under6Yrs.  4 7 1 0 0 1 2  16-30 Years  Female Male 2 1 0 0 0 0 0  3 37  25 3 0 2 0  31-45 Years  Female Male 6 17 15 0 1 20 0  70 Total 15 3 59 Note: Shaded cells indicate notable over-representation. Source: Author’s Survey in Hamadan’s Rural Areas  46 &  Female Male  +  Years Female  Total  6 27 9 1 0 3 0  2 12 1 0 3 15 0  0 3 0 0 0 3 0  0 1 0 0 0  23 105 51 4 4  5 0  49 2  46  33  6  6  238  6.1.3 Destination and Presence of Relatives To find out the pattern of migration stages, the first and final destination of rural migrants have been questioned (Table 6.8).  The responses suggest a trivial change of the first  133  destination and thus, a step-wise migration pattern is not observed.  Greater Tehran  (about 300km away from the province capital) magnetizes the largest portion (47.5 percent of migrants), followed by the relevant township center (31.5 percent). Accordingly, the predominant type of permanent migration is the rural-urban type, with 95.8 percent of the movements belonging to this category. Rural-rural migration consists  of only 4.2 percent of the out-migration moves.  Interestingly, the capital of Hamadan Province, i.e., the city of Hamad an, has not attracted a good deal of rural out-migrants. In the last thirty years, the city’s economic activities (especially industry and construction) have not boomed.’ 59  This, as well as its limited  population scale, hindered competition with not only the unchallenged Tehran magnet at the national level, but also with the other regional centers;  the Township Center or  Neighbouring Province Cities categories in Table 6.8.  These observations lead us to conclude that the regulation of destination choice by migrants for distance and the negative influence of distance (a cornerstone of Ravenstein’s Laws of migration- 1885 & 1890) is not of importance in the sample villages. In other words, the proportion of out-migrants moving to a farther destination is quite greater than those moving to nearby cities; providing that there are abundant means of access, better economic opportunities, and also, that necessary information is disseminated by co ° 6 villagers in the far destinations (e.g., in Tehran).’  159  Despite the salient role of Hamadan City as the regional center up to the 1960’s, it lost its supremacy henceforth. Hamadan was bypassed by the important national highway which links the Persian Gulf ports to the Capital, and was not designated as the site of a few industrial centers in the region. The booming crescent of industrial centers around Hamadan City, including: Zanjan, Ghazvin, Saveh, and Arak, drained many of the required resources (see: Sarrafi et al. 1991, 31). 160  The same conclusion was found in Uttar Pradesh of India, studied by Najma Khan (1986, 117).  134  In the same vein, crossing these with the data of migrants’ visit of originating village (Table 6.12) shows that while 43.6 percent of migrants has gone to destinations close  enough for regular visits (including the central village, township center, other villages and other cities in the Province, and Haniadan City categories), 58.8 percent of migrants do regularly visit their original homes. This is due to the extension of transportation system between major cities in Iran and the low marginal cost for the distance increase.  Table 6.8: Distribution of Out-Migrants’ First and Final Destinations Migrants Destination (Geographical Categories)  Close Central Village Township Center * HamadanCity** Neighbouring Province Cities The Greater Tehran Other Villages in the Province Other Cities in the Province Other Cities in Iran *  **  Final Destination Frequency Percentage 3 75 17 10 113  7 2 11  1.3 31.5 7.1 4.2 47.5 2.9 0.8 4.6  First Destination Frequency Percentage 6 80 18 6 110 9 1 8  2.5 33.6 7.6 2.5 46.2 3.8 0.4 3.4  Total 238 100.0 238 100.0 Markaz-e-Shahrestan, including migrations to the city of Hamadan, if originating in the rural areas of the same township. Those intra-Provincial migrations which has occurred beyond the townships’ boundaries, excluding Hamadan township. Source: Author’s Survey in Hamadan’s Rural Areas  Many migration studies have suggested that existence of prior migration patterns intensify the stream of migration and generate chain-like movements. In a limited way, the data presented in Table 6.9, shows that more than half of the migrants (55.9 percent) have had relatives in their first place of residence after migration to rely on.’ 61  For the rest,  161  This is accentuated for those who have migrated because of arranged marriages, continuing education, and jobs in the urban formal sector. The precedence of migration to certain places with readily accessible jobs plays a significant role in decision-making about prospective destinations (i.e., chain nigration).  135  however, the source of information may still come from the previous co-villagers who have migrated, as will be discussed in the next chapter.  Another unexpected observation in Table 6.9 is that, in relative terms, male migrants have relied on relatives more than the female migrants. In fact, this can be misinterpreted if it is not taken into account that those females who have married a relative and have had relatives at destination, would not be considered as they have had relatives at destination. The relatives have been stated for the husbands only by the respondents. Table 6.9: Distribution of Migrants Who Have Had Relatives at Destination By Sex  Relative at Destination  Male Migrants Frequency Percent 51  37.2  51  84 2  61.3  49  1.5  137  100.0  No Yes MissingData Total  Female Migrants Frequency Percent  Total Migrants Frequency Percent 102 133  1  50.5 48.5 1.0  3  55.9 1.3  101  100.0  238  100.0  42.9  Source: Author’s Survey in Hamadan’s Rural Areas  6.1.4 Occupational Change  The change in the occupational pattern of sample out-migrants, before and after migration, as shown in Table 6.10, indicates the following points: •  The loss of family agricultural workers (from 17.3 percent to zero) is very high and could be detrimental to the household agricultural income; however, the probability of disguised unemployment of the family workers should be probed to determine if the shift is truly detrimental.  •  In the after migration occupations, the substantial growth of governmental jobs is indicated (from 2 to 18.4 percent). It demonstrates the key role of government in the  136  creation of employment opportunities in certain places and the consequent labour force movements. •  In addition, the share of daily wage earners and venders/brokers (including hawkers and sales persons) has considerably increased in the after migration composition (from 2 to 15.6 percent and from 1.3 to 8 percent respectively). These are low-skilled jobs which are subject to great fluctuation in demand and income.  Table 6.10: Distribution of Migrants’ Occupations Prior to and After Migration Occupational Category  Before Migration Frequency Percent  3 Farmer/Herder/Agricultural Worker 41 Family Agricultural Worker 5 Public/Army Employee 0 Private Sector Employee* 2 Driver (Self-Employed) 3 Shopkeeper/Vender/Broker 5 Constructional/Daily Wage Eai ner 0 Artisan (Self-Employed) 0 Soldier (lvlilitary Service) 16 Carpet Weaver 63 Homemaker (Home Economics) 97 Student 3 None (Six Years ofAge & Under) Total 238 * Including clerical, technical, and blue collar jobs. Note: Shaded cells indicate dramatic change. Source: Author’s Survey in Hamadan’s Rural Areas  •  1.3 17.3 2.0 0.0 0.8 1 3 20 0.0 0.0 6.7 26.5 40 8 1.3 100.0  After Migration Frequency Percent 0  0.0  0 44 2 8 19 42 3 3 12 74 29 2 238  0.0 18.4 0.8 3.4 8 0 17 6 1.3 1.3 5.1 31.1 12 2 0.8 100.0  Change Percent —  1.3  0.8 + 2.6 + 67 + 15 6 + 1.3 + 1.3 1.6 + 4.6 28 6 0.5 +  —  —  —  The sharp drop in students’ share of the after migration composition (from 40.8 to 12.2 percent) denotes two possibilities: first, the supremacy of a higher quality of life and better job expectations for the educated villagers over the desire to continue their education, and second, the poverty of rural students which deprive them from continuing their education. The second was not seen as predominant pattern. 137  •  The change in occupational status suggests that a small portion of migrants have been able to acquire new semi-skilled jobs and a relatively larger portion have acquired jobs in the public/formal sector of cities (usually life-time jobs).  •  The out-migrants have become more economically active mostly due to leaving the  student category. Looking at the jump in the homemaker category (from 26.5 to 31.1 percent) and the fall in the carpet weaver category (from 6.7 to 5.1 percent), we may postulate that, in conventional economic terms, women have become less active after migration than they were in original situation.  It should be recalled that female  migrants have had less education than male migrants, and that this hinders their 162 participation in the urban job market. Cross-checking the present occupations of the rural migrants with their final destinations we arrive at the following conclusions: 1- Most students (68.0 percent) go to their township center (Markaz-e-Shahrestan) or the city of Hamadan for a continuation of their studies (mostly secondary school). 2- The majority of low-skilled migrants (69 percent of construction/daily wage earners) work in Greater Tehran. 3-  The proportion of government employees for occupational composition of the  migrants who have resided in their township center or Hamadan City is more than the corresponding figure of those who have resided in the Greater Tehran or other cities. 4- Most of the migrants who are self-employed drivers or venders/brokers (75 and 77.8 percent respectively) live in the Greater Tehran.  162  Another aspect worthy of mention is socio-cultural barriers. In rural areas, young women can easily learn the agricultural and carpet weaving skills and work in their family unit of production in a congenial milieu. This is not easily possible in the urban areas, except in informal home-based employment. According to some rural people’s mentality, the idea of women working outside the home and having remunerated jobs is considered degrading in male-headed households (Moghadam 1993, Ch.6).  138  5- While all of the above observations are chiefly for men, female migrants’ occupations become more diverse for those who live in Greater Tehran and then, for those who live in their township center (still overwhelmed by housekeeping).  6.1.5 Collaboration and Contacts with the Originating Household  On the participation of out-migrants in agricultural work, the responses were that only 21.8 percent (or 21 percent of total in Table 6.11)  ,  basically at the times of sowing and  harvesting, have assisted their remaining households at the original villages. Granting that the migrant females hardly participate in this kind of work (2 percent), the share of participating males rises to 37.2 percent (omitting those in the not able category). Considering that most of sample households live on agriculture, as will be discussed in the next chapter, their assistance is still crucial to the remaining household.  Crossing these data with the number of migrants’ visits to the originating households reveals that nearly one third of those who regularly (from over two times a year to once a week) visit the home village assist the household in agriculture. The proportion sharply 63 drops when the visit numbers are reduced.’ Table 6.11: Distribution of Migrants’ Sex by Their Assistance in Agricultural Work  JVligrant  Male Female  Sex  Assisting the Originating Household’s Agricultural Works Yes No Not Able*  Total  Frequency  Row Percent  Frequency  Row Percent  Frequency  Frequency  81 98  59.1 97.0  48 2  35.0  8  2.0  1  137 101  Total 179 75.2 50 21.0 9 238 Including disablec, and aged persons, and those who were considered by respondents too young to work. Source: Author’s Survey in Hamadan’s Rural Areas *  163  Precisely, 34 percent of those migrants who visit once a week, 30 percent of those visiting over two times a year, and 6.1 percent of those visiting two times and less, collaborate in the agricultural works of their previous household at the home village.  139  Asking about the frequency of migrants’ home visits as shown in Table 6.12, it was found that one fifth of them do not visit their originating village yearly, whereas approximately three fifths have a sound contact with their previous households (three visits or more during a year). The out-migrants’ marital status slightly affects their frequency of visits. The single migrants are more prone to regular visits than are the married ones (5.4 percent difference for over twice a year categories).  Table 6.12: Distribution of Migrants’ Visits to Their Origin by Marital Status Migrants’ Single Migrants Visits Order Frequency Percent Once a Week  MarriedMigrants Frequency Percent  Divorc’d Unkno’n Row Total Frequency Frequency Frequency Percent  10  22.7  34  18.8  0  3  47  19.7  19  43.2  70  38.7  1  3  93  39.1  8  18.2  42  23.2  0  0  50  21.0  7  15.9  35  19.3  0  6  48  20.2  Col. Total 44 18.5 181 76.1 Source: Author’s Survey in Hamadan’s Rural Areas  1  12  238  100.0  Over Twice a Year Once/Twice a Year Less than OnceaYear  6.1.6 Summary In summary, the migration selectivity of young, single, educated male children, and to a lesser extent young, married, educatedfemale children, of rural migrant households has been recorded for the sample villages. The most important type of rural out-migration is toward the cities.  Their first and final destination is almost the same, being  overwhelmingly the Greater Tehran and their pertinent township center (Markaz-e Shahrestan). And more than half of migrants go to a place where they have a relative to rely on.  140  Prior to the migration, most of the would-be migrants were: student, family agricultural worker, and housekeeper.  After migration, the dominant occupation among the out-  migrants was: government/army employee, constructional/daily wage earner, and again homemaker (solely for female migrants). The occupations of those who have resided in the Greater Tehran, proportionally, required less skills-training than migrants to cities at shorter distance. Approximately one fifth of the out-migrants still assist their originating households with agricultural work at peak times. The drop in the level of assistance in household works for the female migrants is higher. About three fifths of the out-migrants regularly visits their home village, with the higher share being single migrants. It should be emphasized that the major structural shift in the migrants’ occupations are from the agricultural sector in rural areas to the construction and service sectors in urban areas.  6.2 Characteristics of the Rural Migrants’ Households  The majority of the sample rural migrants’ households (61.8 percent) have two to four out-migrants and more than one fourth (26.5 percent) have five to seven out-migrants (see Table 6.13). The average is 3.5 migrants per remaining household at the sample villages. A noteworthy point is the existence of at least one son remaining at the rural migrant household(i.e., the traditional caretaker of the household) for 66.2 percent of the sample households (see Appendix C, table C.4).  6.2.1 Size and Composition  In the five sample villages, the average size of the 68 sample remaining households of rural out-migrants were 4.69 (see Table 6.14), composed of 2.32 male and 2.37 female members (see Appendix C, Tables C.3 & C.4).  The Provincial average for the rural  households in 1986 was 5.55 persons, including 2.88 men and 2.67 women (HPPBO 1992, 29). As a result, the sex ratio in the rural migrants’ households was 97.9 compared 141  to 108 in the Province. This indicates a smaller size and fewer male members in the rural migrants’ households than the average provincial statistics for rural households. This is consistent with our findings about out-migrants’ characteristics (i.e., male selectivity) in the preceding section. Table 6.13: Distribution of Rural Households by the No. of Out-Migrant Members  No. of Migrants in Household  Frequency  One Migrant Two Migrants Three Migrants Four Migrants Five Migrants Six Migrants Seven Migrants Nine Migrants  7 15 13 14 10 5 3 1  Total Mean: 3.559  68 Median: 3.0  Valid Percent  Cum. Percent  10.3 22.1 19.1 20.6 14.7 7.4 4.4 1.5  10.3 32.4 51.5 72.1 86.8 94.1 98.5 100.0  100.0 Source: Author’s Survey in Hamadan’s Rural Areas  Table 6.14: Distribution of Rural Migrants Households by Size No. of Rural Household Members  One Person Two Persons Three Persons Four Persons Five Persons Six Persons Seven Persons Eight Persons Nine Persons Ten Persons Eleven Persons Mean: 4.69  Total Median: 4.00  Frequency  Percent  Cum. Percent  8.8 6 8.8 23.5 10 14.7 11 16.2 39.7 13.2 9 52.9 8 11.8 64.7 10.3 75.0 7 3 79.4 4.4 94.1 10 14.7 95.6 1 1.5 97.1 1 1.5 2 100.0 2.9 100.0 68 Source: Author’s Survey in Harnadan’s Rural Areas  142  6.2.2 Literacy  It was found that out of 289 household members of 6 years of age and over, 179 were literate or were going to formal schools (Table 6.15 and see Appendix C, Tables C.5 & C.6 for details). households,  The rate of literacy was 61.94 percent among the rural migrants’  Compared to the 49.6 percent literacy rate of the rural population of  Hamadan Province in 1986 (SCI 1990a, 115), the rural migrants’ households were more literate than the average rural households.  However, they were less literate than the  migrants’ households itself (79.2 percent literate as described in section 6.1.2).  The  remaining households lose better educated members while on the whole they are still more literate than other households. Table 6. 15: Literacy among Members of Rural Migrants’ Households  Description  Frequency  Mean  Mode  Literate Members of 6 Yrs. ofAge & More Total Members of 6 Yrs. ofAge & More  179 289  2.63 4.25  3 4  Source: Author’s Survey in Hamadan’s Rural Areas  Table 6. 16: Distribution of Rural Migrants’ Household-Heads by Education Education  Frequency  Percent  Cum. Percent  No Education (Illiterate) Grade Twoto Grade Five Grade Six (End ofPrimaiy School) Grade Eight to Grade Nine* Grade Twelve (End qfHigh School) First/Second Year ofAdult School Old Traditional School Education Total  40 8 3 2 1 3  58.8 11.8 4.4 2.9 1.5 4.4  58.8 70.6 75.0 77.9 79.4 83.8  11 68  16.2 100.0  100.0  No cases were observed for Grade Seven, Grade Ten, and Grade Eleven Categories. Source: Author’s Survey in Hamadan’s Rural Areas *  With respect to the head of rural migrants’ households, 58.8 percent of them were illiterate and only 4.5 percent had attended high school (Table 6.16). The small portion of those  143  who have attended adult schools, as in the case of out-migrants, indicates that postrevolutionary efforts for literacy have failed to broadly address rural illiterates.  6.2.3 Occupation of Head of Households The occupation of head of households is overwhelmingly that of farmer (Table 6.17). Adding half of the number of shop owner/farmer and housekeeper/farmer categories, the share of farmers rises to 70 percent. The same Provincial figure stood at 57.7 percent in 1986 (Sarrafi et al. 1991, 70).  Thus, rural migrants households rely more than the  average rural households do on the agricultural jobs. Table 6. 17: Distribution of Rural Migrants’ Household-Heads by Occupation Occupational/Ability Category Farmer/Land Owner Housekeeper/Land Owner carpet Weaver/Land Owner Driver/Self-Employed Disabled/Land Owner Aged/Land Owner Shop Owner/Farmer Shop Owner/Vendor Artisan (Self-Employed) Public Employee  Frequency  Percentage  43 4 1  63.2 5.9 1.5  3 2 2 6  4.4 2.9 2.9 7.4 8.8  1 1 68  1.5 1.5 100.0  5  Total Source: Authors Survey in Hamadan’s Rural Areas  Of special importance from the social welfare viewpoint is the existence of 5.8 percent aged or disabled heads of household. Obviously, the number of aged or disabled persons is quite higher among the other members of households which has not been recorded. The extended family is the common shelter for these persons in rural areas. However, when  144  the head of household rests in this category, it implies a desperate situation for the whole 64 household who mainly rely on the head.’  The distribution of the occupational categories of the household-heads by their educational achievements shows that, both in absolute and relative terms, there are more illiterate among the farmer-headed households than among the others (27 persons or 62.8 percent of the farmer/land owner category in Table C.7 of Appendix C).  6.2.4 Land and Livestock Ownership On average, the sample rural migrants’ households, excluding those who have no land ownership in each category, have 2.2 and 20.2 hectares irrigated and rain-fed arable land respectively, and 0.44 hectare of Orchard (Table 6.18 and see Appendix C, Tables C.8, C.9 & C. 10 for details). In the rural areas of the Hamadan Province, the averages for producers of each category are 2.9, 7.9, and 0.19 hectares, respectively (Ebn-e-Ali 1992, 207).  Thus, our sample farmers own some less irrigated land and more of other  agricultural lands.’ 65 This indicates that the majority of the migrants’ households fall into 166 the middle and the upper-middle stratum of land owners.  In terms of livestock ownership, it was found that near two third of the sample households had on average 2.1 heads of cow and less than half had an average of 23 heads of sheep and goats (see Appendix D). The rest of the households in each category had no livestock at all. The inclusion of livestock breeding with cultivation was an important source of 164  The almost non-existence of social supports (e.g., welfare/retirement payments) in the rural areas of Hamadan Province, creates a stark disparity with the urban areas where some kind of social supports may be reached. As will be pointed out in the next chapters, this disparity is a major cause of out-migration. 165  On the basis of my knowledge of the Province’s rural areas, I believe that the sum of the sample villages’ irrigated lands are Less than the commensurate provincial average. Therefore, the aforementioned lesser possession of irrigated lands may be ignored. 166  This seems to be true, especially considering the predominance of small holdings and a fairly non polarized land distribution in the Province.  145  income for the pertinent rural owners. In general, these units of production were small, using traditional methods for a semi-subsistence/semi-commercial production. There is no accurate provincial statistics to compare the livestock ownership of the sample households with the rest of the province. Table 6.18: Average Land Ownership among the Sample Households by Agricultural Type Type of Agriculture  No. of Owners  Total Area (Hec.)  28 Rain-Fed Land 50 Irrigated Land 62 Orchard Source: Author’s Survey in Hamadan’s Rural Areas  567 111.2 27.3  Average Ownership 20.25 Hec. 2.22 Hec. 0.44 Hec.  6.2.5 Sources of Income  95.6 percent of respondents have stated selling surplus agricultural products as the most Selling surplus husbandry products (e.g.,  67 important source of household income.’  dairy, sheep) is the first source for 2.9 percent and the second most important source of income for 48.5 percent of households.  The next important sources of income, and  usually supplementary ones, are wage labour activities, shopkeeping/vending, and migrants remittances (Appendix C, Table C. 15). In general, the heavy reliance of rural migrants’ households on the sale of agricultural and husbandry products is observed.  6.2.6 Remittances and Spending Patterns According to the interviewees, 60.3 percent of the rural out-migrants of the five sample villages do not send money home (Table 6.19). For the remaining 39.7 percent, however, the remittances were spoken of as being crucial. As various studies in the South countries 167  The staple food products in Iran are wheat and then rice. The predominant production pattern in Hamadan Province is one of wheat and barley. The common exercise of rural households is to set aside their yearly consumption at first and then to sell the rest mainly to the governmental organization which is the only formal buyer (i.e., monopsony).  146  suggest, there is an obvious need to take into account the out-flow of capital (e.g., for students) to ascertain the net income transfer (Standing I 984b). This was not available for the sample villages. Table 6.19: Distribution of Sample Households by Recipient/Non-Recipient of Remittance Receiving Remittance  Frequency  Percentage  41 27 68  60.3 39.7 100.0  No Yes Total : Source: Author’s Survey in Hamadan’s Rural Areas  Despite the complexity of household budgets for ascertaining the allocations, the responses to the remittance spending prove that the major part has been spent on purchasing everyday household needs, and only next, for agricultural production (Appendix C, Table C.16). It may be postulated that ‘the spending of remittances reflects the poverty and lack of investment opportunities from which the migrant came” (Connel et al. 1976). In the field, not much evidence of spending for conspicuous consumption was observed.  6.2.7 Migration Intention As part of probing the would-be migrants’ characteristics, the intention to migrate in the near future was probed both for the respondent and the whole household. shows that the majority have no short-term plan for migration.  Table 6.20  The number of those  individuals who want to migrate is greater than that for whole households, which is consistent with our previous observation about the pattern of rural out-migrants (section 6. 1. 6). 168 168 As mentioned before, part of the did not express categoiy can be considered as intends to migrate for respondents. This is so because of the unwillingness of individuals to directly express their risky and not appreciated decisions (in the eyes of co-villagers).  147  Table 6.20: Rural Migrants’ Household and Respondents’ Migration Intention  Migration Intention  Respondent (Individually) Frequency Percentage 42 No intention 61.8 20 Intends to migrate 29.4 Did not express/Missing 6 8.8 68 100.0 Total Source: Author’s Survey in Hamadan’s Rural Areas  Household (Collectively) Frequency Percentage 49 72.1 17 25.0 2.9 2 68 100.0  Cross-tabulation of respondents’ relationship to the household-heads with their migration intention reveals that 75 percent of the heads of remaining households do not want to migrate. Inversely, almost the same ratio among his/her children want to migrate from the village (see Appendix C, Table C. 11).  Moreover, crossing the occupation of heads of households with whole households’ intentions indicate that 81.4 percent of farmers (i.e., 35 out of 43 persons) do not want to migrate while the majority of self-employed people in the service sector want to migrate (Appendix C, Table C.12). This leads us to tentatively conclude that the place-related capital (e.g., land) are a hindrance to movement of owners, whereas better opportunities for businesspersons in the urban areas and the higher degree of capital mobility in the service sector make the private service prone to migration.  6.2.8 Housing and Economic Status  To peruse the original socio-economic status of rural out-migrants, the housing quality and the income level of remaining households at the villages were investigated. The quality of households’ houses were observed and classified loosely into three general classes (i.e.,  148  poor, average, and good) by the interviewers themselves.’ 69 The housing quality of rural migrants’ households is reported as follows: -  -  10.3 percent (7 cases)  Poor quality:  Average quality: 51.5 percent (35 cases) Good quality:  -  38.2 percent (26 cases)  In addition, the number of rooms utilized by the household, were asked about and it was found that on average 3.1 rooms are available (see Appendix C, Table C. 13 for details). Owing to the average size of the sample households (i.e., 4.69), this means that there are 1.5 persons per room, which is less than the provincial average of 2.2 (Arjomandnia 1991). This fortifies the above findings that generally the remaining households’ houses are located in the upper half of rural standards.  Because of the confidentiality of household income in a somewhat close rural community (the so-called threatening question), respondents were reluctant to provide income-related information.  As acknowledged in other Iranian studies, it is not reliable to obtain  responses as regard to amount of income, though it could be inferred from the amount of land ownership and other sources of income, as well as from the quantity and the quality of the household labour force. This was the case for assessing the approximate income level of the sample households. Therefore, the interviewers themselves chose one of the three broad categories (i.e., low-, middle-, and upper- income level) and it was checked with other data by the principal researcher. The ensuing results were found: -  -  -  Lower-income: 25.0 percent (17 cases) Middle-income: 39.7 percent (27 cases) High-Income:  35.3 percent (24 cases)  Ahead of fieldwork in each sample village, the interviewers were instructed by the author on how to select the class of each house. The rule of thumb was to typify the house appearance (constructional materials, design and size) in each class according to common sense and the local standards. 169  149  Again, like the housing situation, it implies that generally the remaining households’ income levels are slightly above the village average.  This was implied also by the  households’ ownership of certain house-wares (see Appendix C, Table C. 14 for details).  6.2.9 Summary  On the basis of the foregoing analysis, the characteristics of typical rural migrants’ households in the sample villages are summarized as follows: •  The majority of households have two to four out-migrants.  •  Average size of households are 4.7 persons with slightly more female members, contrary to the general rural situation in the Province with more male members.  •  One third of the remaining households has two male members, another third has more than two male members, and the remaining third has one or none (Appendix C, Table C.4).  •  The remaining households are better-educated than the rest of rural households in Harnadan Province, even though they have lost their own better-educated members.  •  The sample household-heads were mostly illiterate and farmers.  •  On average, the farmer-headed sample households own slightly less irrigated land and quite more rain-fed land and orchards than do corresponding households in the Province.  •  The main source of sample households’ income was the sale of surplus agricultural products. Other important ones were migrants’ remittances, shopkeeping, and waged labouring.  •  Only two fifths of out-migrants remit; nonetheless, remittance is crucial for receiving households. A major part of these remittances is spent on purchasing everyday needs and some on agricultural production by the households.  •  The majority of respondents in the sample villages, being household-heads and farmers, do not want to migrate in the near future, neither individually and nor with the 150  household.  However, adult children of these households and non-farmer headed  households are more prone to migration. •  Considering the socio-economic status of rural migrants’ households, It appears that the upper half of the spectrum (according to their pertinent village standards) are more present than the lower half.  6.3 Stated Causes of Migration As discussed in Chapter V, investigating the cause of migration is more complex than simply asking migrants for the cause. Owing the research objective, the perceptions of respondents from remaining households were used as proxy for the statement of migration cause by migrated members. During the survey, interviewers strove to make respondents understand that they should convey the causes according to the out-migrant’s perception. Nevertheless, unavoidably, these responses should be considered as mixed with the respondents’ own perceptions as well.  In addition, two more questioned were pursued.  Those respondents who intended to  leave the village were questioned as to their reasons. Those rural households who also intended to migrate, were also asked to state their reasons. After listing the causes, for all three questions, the respondents were asked once more to prioritize their reasons. Therefore, sets of first, second, third, and all of the causes were generated.  Being asked “why did your household member(s) migrate?”, rural respondents have stated an array of causes. The interviewers were instructed to ask this simple and open question. There were a detailed list of probable causes to be marked, as well as enough space to write the non-fitted answers in the questionnaire.  The latter were categorized by the  principal researcher later. 151  However, for an effective analysis of varied causes, it is found that the following categorization is comprehensive for comprehending the driving causes of out-migration: Type 1- Survival: This includes stated causes such as unemployment, underemployment and insufficient income, which indicate a desperate decision to leave the village for adequate income and job. Type 2- Promotion: Seeking a higher-paying job, expanding a business and employment  transfer are motivations for migrants in this category.  This shows an act of  preference on the basis of perceived better opportunities somewhere else (i.e., in urban areas). Type 3- Education: Continuing education for oneself or one’s children which shows the  absence of higher levels of educational institutions in the village and/or aspiring a better quality of such services. Type 4- Enhancement: Pursuing better quality of services, social welfare arrangements  and an urban life-style which implies the desire for life enhancement and secure livelihood. This cause is beyond a simple economic cause as in Type 1. Type 5- Resentment: Inter-/intra village skirmishes or household disputes which show a  resentment toward the village milieu or household authority as the motive to migration. Type 6- Subordination: Marriage and joining/accompanying the family which indicate a  subordinate decision-making and in most cases a compelling one to move. This cause should not be considered as the primary motivation for migration, and is stimulated by the head of household’s decision. The foregoing categorization was based on identifying the root causes as to: •  whether the migrant had a job or not in the originating community (promotion & enhancement vs. survival).  This differentiation is important to substantiate if out  migration was excessive and detrimental to the village viability or not. 152  •  whether migration was motivated by mostly personal economic reasons vs. non economic ones (survival & promotion vs. enhancement & resentment).  Here the  potential of infrastructure roles is examined as was mentioned in section 4.5.1. •  whether migration was initiated by the migrant or not (all types vs. subordination). This distinction demonstrates the weight of dependent migrants for household migrations.  Besides, essentially the first, third and fifth types (i.e., survival, education, and resentment) reveal the dominance of push factors in the village (equal to subsistence motivation posed in section 4.4), while the second and fourth (i.e., promotion and enhancement) can be related to pull factors of the cities (similar to enhancement motivation mentioned in section 4.4). The sixth type cannot be interpreted in such distinction as push or pull and it depends on the principal migrants’ causes.’ ° 7  6.3.1 Stated,Perceived Causes for Out-Migrants Table 6.21 provides the results of the above-mentioned questions. The significance of marriage and accompanying the household (i.e., type six: subordination) are evident in migration decisions. However, setting aside both the under six years of age and the female migrants (for whom type six has often been stated as the predominant cause of migration) the predominant cause remains only an economic one (i.e., type one and type two) with 42.1 percent of all the paramount causes and 41.9 percent of all the three main causes.’ 71 Nonetheless, educational cause (14 percent) comprises a notable force, particularly in 170 It bears repeating that the push and pull distinction is not clear a cut distinction and not mutually exclusive. However, it is useful for identii’ing the chief cause of rural migration as being internal or external (also see footnote 122). 171  As will be discussed later, it should be mentioned that ascribing the causes to economic motivation is not mutually exclusive. There is no doubt that the expectations of migration are higher than those of mere economic gains. They have to do with security, hope, preference, and human dignity. Economic motivation is undeniably a major cause of migration but not the exclusive or necessarily the decisive one.  153  driving out the rural youths.  The family disputes or village skirmishes were not  underscored as an important cause, barring two cases.’ 72 Table 6.21: Distribution of Rural Out-Migrants’ Causes of Migration Typical Cause of Migration Type One: Survival Type Two: Promotion Type Three: Education Type Four: Enhancement Type Five: Resentment Type Six: Subordination Missing Total:  Paramount Cause Frequency Valid Percent 66 29.9 27 12.2 31 14.0 3 1.4 2 0.9 92 41.6 17 238 100.0 ---  Three Main Causes Frequency Valid Percent 87 50 46 48 2 94 387* 714  26.6 15.3 14.1 14.7 0.6 28.7 100.0  The majority of respondents did not state the second and third overriding causes of out-migration. Source: Author’s Survey in Hamadan’s Rural Areas *  As expected, the frequency of stated causes of migration in relation to the quality of life (type four and to some extent type two), dramatically increases in the “three main causes” category as compared to their weight in the “paramount cause” category.  In the  aggregate, as shown in Table 6.21, the push related causes (sum of type one, three, and five) with 44.8 percent and 41.3 percent, are greater than the pull related causes (sum of type two and four) with 26.2 percent and 30.0 percent. This summation should be crosschecked with the key-informants’ data in Ch.VII.  Cross-tabulating the typical causes of migration by occupation of the male out-migrants (see Appendix C, Table C. 17) shows that 70.7 percent of previously family workers have left the village for type one cause (i.e., survival), and the rest for type two cause (i.e.,  promotion). Moreover, of those male out-migrants who used to be students at the village, 172  This might be understated because of reluctance on the rural household side to admit the domestic problem for an out-sider (i.e., interviewer). However, on the basis of our field observation and the key informants’ interviews, type five cause of migration is not rampant in the sample villages.  154  43.9 percent have left for survival, 36.4 percent for education, and 18.2 percent for promotion and enhancement. Thus, unemployment and meager income (the overriding force for survival cause) are the main repulsive forces for male family workers.  This is also true for the male students, and is augmented by the lack of higher-order educational facilities at home, for those who would like to continue their education. The educational migration mostly takes place after finishing primary school or, to a lesser extent, after the end of junior high-school for both males and females (see Appendix C, Tables C. 19 & C.20). It is seen that the survival cause is relatively higher for those male out-migrants with less education than others. 173  Cross-tabulation for female out-migrants reinforces the preponderance of the type six cause (i.e., subordination).  This includes 98.4 percent of previous homemakers, 93.3  percent of previous carpet weavers, 73.7 percent of previous students and all the previous  public employees (see Appendix C, Table C.18). Subordination remains the chief cause of out-migration for rural females irrespective of their level of education (see Appendix C, Tables C.19 & C.20).  Once more, it was found that for the male rural out-migrants of sample households, being overwhelmingly the sons of the remaining household heads 174 (see Appendix C, Table C.21), the typical first causes were: survival (52.5 percent), promotion (22.5 percent), and continuing education (20.8 percent).  The same first cause for females, being  overwhelmingly the daughters of the remaining household heads were again subordination 173  Not that a causal relation can necessarily be elicited from this, as the higher the education, the higher the aspirations. There could be a root-cause of the previous socio-economic status of the household in the rural community. In other words, it may be said that higher status leads to higher education which in turn leads to higher aspirations. Therefore, the education factor may be an intervening variable. 174  Interesting to know that no son-in-low has been stated as a member of an extended family (see Tables C. 19 & C.20). Actually, it is not common for sons-in-law to live with the wife’s family in rural Iran.  155  and to a lesser extent continuing education. Considering the second and third causes of out-migration (see Appendix C, Table C.22), enhancement comes up as a major cause for both sons and daughters of the remaining household heads.  6.3.2 Stated Causes for Would-be Individual and Household Migrants  The rural respondents were asked if they individually intend to migrate in the near future, and also, they were questioned about the migration intention of their entire household. The stated paramount and second causes of migration intention have been presented in Table 6.22 (for the rural respondents) and Table 6.23 (for the sample households). Whether those who intend to migrate will in fact do it or not, “the expression of intentions reveals the normative acceptance of migration as a reasonable option” (Lee 1985, xviii). This provides insights into the potential roles of infrastructure in our later discussion. Table 6.22: Distribution of Rural Respondents’ Causes of Migration Intention  Typical Cause of Migration*  Paramount Cause Frequency  4 Type One: Survival 9 Type Two: Promotion 0 Type Three: Education 7 Type Four: Enhancement 0 Type Five: Resentment 0 Type Six: Subordination 42 No Migration Intention 6 Missing/Not Stated Total: 68 * For description of each type see section 6.3 of text. Source: Author’s Survey in Hamadan’s Rural Areas  Valid Percent  6.5 14.5 0.0  11.3 0.0 0.0  67.7 ---  100.0  Second Cause Frequency  Valid Percent  2 3  3.2 4.8  1  1.6  8 3 3 42 6 68  12.9 4.8 4.8 67.7 100.0  According to the rural respondents, the paramount cause of their tendency to migrate is type two (promotion), type four (enhancement), and then type one (survival).  As the  second cause, type four (enhancement) comes first. In the aggregate, enhancement and  156  promotion (with 27 out of 40 or 67.5 percent of the stated causes of those individuals who intend to migrate) cause the most incentive to leave the village.  From the household point of view, the paramount cause of migration is foremost type six (subordination) and then equally type one (survival), two (promotion), three (education), and four (enhancement). Like the respondents’ case, as the second causes of migration intention, type four (enhancement) and type two (promotion) have been stated most frequently.  In the aggregate, enhancement and promotion, similar to the respondents’  case, cause the most incentive to leave the village (with 20 out of 34 or 59 percent of the stated causes of those households who intend to migrate). The following scrutiny of the survey data with the help of cross-tabulation provides more details on the stated causes. Table 6.23: Distribution of Sample Households Causes of Migration Intention  Typical Cause of Migration*  Second Cause  Paramount Cause Frequency  3 Type One: Survival 3 Type Two: Promotion 3 Type Three: Education 3 Four: Enhancement Type 0 Type Five: Resentment 5 Type Six: Subordination 48 No Migration Intention 3 Missing/Not Stated Total: 68 * For description of each type see section 6.3 of text. Source: Author’s Survey in Hamadan’s Rural Areas  Valid Percent  Frequency  Valid Percent  4.6  1  4.6 4.6 4.6  4 1 10  1.5 6.2  0.0 7.7 73.8  0 1 48 3  0.0 1.5 73.8  68  100.0  ---  100.0  1.5 15.4  The migration intention of most household heads and their children is caused by types two and four (promotion and enhancement appeals- see Appendix C, Table C.23). The same , though education 75 chief causes are observed for the sample agricultural households’  ‘  Those households whose heads work in the agricultural sector.  157  (type three) is also added (see Appendix C, Tables C.24 & C.25).  Regarding the  households whose heads are engaged in self-employed service occupations (e.g., artisan, shop-owner, driver), again, promotion and enhancement causes are primary.’ 76 In conjunction with literacy, the survey data reinforce enhancement and promotion causes for the greatest number of literate respondents (see Appendix C, Tables C.26 & C.27). While the corresponding data for sample households who intend to migrate reveal that in addition to enhancement cause, education and promotion causes are significant, for illiterate sample respondents who intend to migrate, survival and enhancement are maj or causes.  The preceding cross-tabulation of stated causes of migration intention and household characteristics further supports the importance of the aforementioned enhancement and promotion causes in general.  This in turn, contrary to the out-migrants’ causes of  migration (refer to section 6.3), implies the dominance of pull factors for the migrationprone households of the sample villages.  6.4 Conclusion  The data presented in this chapter have shown that better-educated children of agricultural households (mostly sons) have left the village mainly for survival and promotion types of causes toward the urban areas.  The dominant cause for female migrants (mostly  daughters) is subordination type.  The continuation of education, particularly, is an  important cause for youths after finishing primary school.  The out-migrants usually  176 Type six cause of migration intention (subordination) is often a significant cause when the household migration cause is asked. However, it is not a generic cause of migration and should be regarded as a consequence of other causes (which usually are found in household heads’ responses). Thus, I do not emphasize type six cause in this summary.  158  maintain their contact with home, while a minority assist their original rural households in terms of financial aid and physical labour.  The survey reveals that the sample households with migrant members are of a smaller size than the other rural households, and also have more educated members. The majority of these rural migrants’ households own proportionally more agricultural lands and are often considered to be among the upper half, in terms of socio-economic status. 177 In general, it seems the principal farmers (often the head of household) have not abandoned their land, 78 according to these findings.’  In approximately two thirds of rural households, there  have remained one or more sons.  Logically, in wake of youth flight, village viability  largely depends on the rural ability to retain the last remaining sons.  On probing the migration intention of the rural migrants’ households, it was found that the majority of them do not want to migrate. However, the young adult educated children are quite prone to leave the village.  The tendency to out-migration is higher among non  agricultural sample households.  Generally, the so-called pull factors seem to be more  effective on the would-be out-migrants than the push ones. Inversely, the push factors were reported greater for most of the already rural out-migrants.  In view of the foregoing observations, the major conclusion of this chapter with respect to potential rural migrants is as follows:  It bears recalling that uprooted migrant households (wholly absent in the village) were not covered in this survey. It can be postulated that the poor and landless households mainly left the villages after the Shah’s land reform, when there were lots of menial work opportunities in the large cities (refer to section 3.3). 17  178  Once again, the result could have been different if the uprooted migrant households had been taken into account. Albeit, one study of the impact of migration on cultivated land use has suggested that the farmlands of migrant households are cultivated either by their rural relatives (share-cropping) or by informal purchasers (Farrokhian et al. 1977, 88).  159  •  the typical illiterate would-be rural migrant is seeking a job and trying to make life secure (tantamount to survival and enhancement causes) by moving to the city.  •  the typical literate would-be rural migrant is pursuing a better job and a higher quality of life (tantamount to promotion and enhancement causes) by going to the city. All-in-all, they come to cities to secure life, though with different aspirations.  160  CHAPTER VII: GROUP INTERVIEWS t FINDINGS  161  7.0 Introduction  This chapter, unlike the preceding one, relies on qualitative data and employs group interviews with key-informants in the rural communities to cross-check the foregoing conclusions and to gain further insights into the out-migration mechanism and context, and into its impact on the originating villages. Following this, in order to find remedies to the incessant migration of rural youths, the participants’ views as to rural development and the infrastructural services are sought.  Starting with field observation and a survey in the five sample villages of Hamadan Province, I identified some key questions at the community level. Indeed, survey method has weaknesses in examining intricate patterns of social interaction (Marshal & Rossman 1989, 85).  As discussed earlier, in-depth interviewing as a qualitative data collection  technique was selected in order to acquire the answers to context-related questions. However, due to interviewing scope and pitfalls (e.g., data distortion and personal biases of interviewees or interviewers- Yin 1989, 89), the supplemental data collection by survey 79 In short, the use of and existing data at the other levels was still found to be essential.’ the group interview is not meant to replace individual interviewing, but to provide another level of data gathering and a perspective not available through individual interviews (Fontana & Frey 1994, 364).  In order to respect how the participants (i.e., interviewees) frame and structure the responses, and at the same time not losing sight of the research focus, a semi-structured  179  Interviewing is based on: “...an assumption fundamental to qualitative research the participant’s perspective on the social phenomenon of interest should unfold as the participant views it, not as the researcher views it”(Marshal & Rossman 1989, 82). According to Fontana and Frey, “recently, postmodernist etknographers have concerned themselves with some of the assumptions and moral -  problems present in interviewing and with the controlling role of the interviewer. These concerns have led to new directions in qualitative interviewing, focusing on increased attention to the voices and feelings of the respondents and the interviewer-respondent relation.” (Fontana & Frey 1994, 363)  162  format was used.  Semi-structured group interviewing refers to a situation in which an  interviewer asks the respondents a series of open-ended questions and participates in follow-up questions and clarification discussions. The interviewer somewhat directs and moderates the discussion among the participants, trying to create an authentic communication environment and minimizing the misinterpretation due to cultural differences, observers’ effects, etc. (Marshal & Rossman 1989, 104).  The following text summarizes the responses to a list of questions at five full-session group interviews in the sample villages (see: Appendix B, for the list of questions and participants, and Appendix E, for an abridged record of interviews).  s:’ 8 Out-Migrant 7.1 On the Village ° Two categories of out-migrants were discussed with the key-informants from the sample villages: first, those households who have wholly migrated from the village and second, those villagers who individually have migrated and left their households behind. Regarding the wholly migrated families, it was said that the majority of them (except in ’ 8 Se-miran where it was the minority) had transferred their lands to other villagers.’ Some of them (majority in  Se-miran) have kept their land and are supervising the  labourers’ works by periodic visits in the work season (particularly in the case of  vineyards/orchards) or taking advantage of share-cropping with some rural peasants. It is rare for the wholly migrated families to abandon their lands, except those of very poor quality (e.g., in terms of available water, soil, steepness). The inquiry about the characteristics of rural out-migrants at this level was quite general and could  not be as precise as our survey results. However, the general pattern, perceived by the respondents, assists in our generalization of the survey data. Some new data about wholly migrated households are also complementary.  The buying and selling of agricultural lands in Iran is prohibited by law. However, quite often the lands are informally transferred [bought and sold] with mutual agreements. A unique survey of the consequences of landholders’ migration in the Hainadan rural areas before the Revolution demonstrated 181  cases of this kind of transfer with almost no cases of abandonment of lands (Farrokhian et al. 1977, 59).  163  The economic status of wholly migrated households (Boneh-kan) were perceived by the majority of interviews’ participants as being mainly among the two tails of the wealth 82 distribution spectrum (i.e., poor/landless and richllandlord strata).’ poor/landless households outnumbers the others in migration streams.  However, the Also, it was  mentioned that there were instances in which only the head of household had migrated. The economic status of this group was described as being that of landless or very small landholders. The remaining members are usually living in their extended families and are visited frequently by the out-migrant head.  The ultimate objective of this group is  overwhelmingly to permanently settle the whole household in the city.  For individual out-migrants, the situation is different. Many of them, as described by the interviews’ participants, are offspring of small- or medium-size landowner families. Unanimously, it was accepted that single young, educated male children are most 83 Traditionally, they will inherit part of their family susceptible to city-ward migration.’ lands. The divided land, though, is often inadequate for survival of remaining children (i.e., multiplication of users due to formation of new families vs. fractionated land). This group of migrants, which consists in bulk of decisive out-migrants in our survey (Chapter IV.), may not be considered as poor/landless. In certain times, they benefit from their original household resources (e.g., during studying, wedding, and unemployment) and have the prospect of cashing their share of land in the village.  Hence, their original  84 economic status is often middle-class in the rural scale.’ This finding Singh 1980). 182  is consistent with the extensive studies on the South Asia (Connell et al 1976; Oberai &  In fact, the prevailing pattern is that an unmarried male Leaves the village (in many cases after finishing the compulsory military service) and comes back to the community later to marry a village girl (i.e., arranged marriage) and eventually, when getting established, takes his wife to the city. In the prevalent Islamic tradition, single females rarely may migrate by themselves. This finding is at odds with Connell et al.s generalization about migrant backgrounds in the South rural areas (Connell et al. 1976). 184  164  Gradual out-migration streams have been going on for decades in all the sample villages. The key-informants of four villages (excluding Jarban-lou) indicated that there was a sudden out-migration intensity during the first or the second year after the establishment of the Islamic Republic (due to rising expectation), but that it was reduced to the same 85 chronic occurrence as before.’  The common description by interviewees about other aspects of out-migration patterns were as follows: •  Rural-to-rural migration is very small (even the exchange of brides is not 86 common).’  •  Destination of out-migrants is mostly Tehran and then, the township center (Markaz  e-Shahrestan). •  A few villagers (from remaining Khosh-neshins and/or very small landholders) seasonally out-migrate (increased in high-yielding and well-accessed villages, i.e.,  Mian-roudan and Maanizan in our samples). •  Seasonal migrants go to national poles of constructional works (e.g., Tehran, Isfehan, Arak) and to the warm climate ports (i.e., Bandar Abbas, Boushehr, Abadan).  •  Return migration is non-existent.  The national census data of 1986 indicate the climax of in-migration to the urban areas circa 1980 for the whole country (SCI 1993a, 11) and Hamadan Province as well (SCI 1992b, 11). The contracting urban economy in the following year reduced the influx. 186 However, following the discussions I found out that there were some temporary rural in-migrants for 185  contractual labour during the harvest season.  165  As might be already noted, that information which has been pursued for out-migrants characteristics and out-migration patterns in our survey (Section 6.1.6) is not contradictory to the pertinent information in this section.  7.2 On the Causes of Out-Migration: The overwhelming first response to the question of the cause of out-migration in all the group interviews was lack of job and/or insufficient income for the out-migrants (i.e., the survival type of causes in section 6.3) who were largely rural youths (refer to Appendix E, for summary of views). The general logic implied the population pressure on agricultural resources (water and land) and dwindling share of the next generation of farmers as the most effective cause for the young generation of small and medium land-holding households.  Upon scrutinizing the responses, it is evident that the disguised unemployment is also quite important. This is so because of two issues: first, the intermittent and precarious nature of agriculture, being mostly a seasonal activity and depending on appropriate weather (some villagers spoke of the long winters, or slack season, of agriculture, and of the frustration of being idle at home for up to eight months-a-year in some areas); second, the reduction of the per-capita work load of the labour force. In other words, since the prevalent unit of rural production in Hamadan Province is the household (like family-run farm), there is always the possibility of a new division of labour among the members. A new working member of the household can be added, while the marginal productivity may be quite smaller than the additional needs and the result is impoverishment (i.e., smaller piece of pie for the members) and underemployment (i.e., more free-time for the working force). Thus, many villagers tend to complain about insufficient income rather than a lack ofjobs. In their view, they can share one of the household responsibilities which is somehow a  166  part-time job but without enough reward (insufficient income). Here we note that the spreading mechanization also reduces the work load of the household labour force.  Those rural out-migrants who come from a non-agricultural or landless households, in addition to those who are from land-owner households (whether with sufficient land or not) but are not willing to continue at an agricultural job, all out-migrate for other causes, as was explained by the participants regarding various issues.  One of the issues cutting across the interviews was schooling and the educational system for encouragement of out-migration.  As a good example of causal sequence in our  dialogue with the participants, and the inter-connectivity of causes, I summarize the views below.  According to a Se-miran councilor, even the aspiration of junior high school students in his village cannot be fulfilled by present agricultural work, which is perceived as menial work. At the same time, the prospects for getting non-agricultural jobs in rural areas are quite limited. Thus, the majority of students flock to the cities. Despite the fact that the school textbooks were wholly revised after the Revolution to remove “Westernlurban values”, two rural teachers in our interviews in Hassan-taimur and Jarban-lou attributed the above mentality to the content of textbooks (i.e., biased in favor of urbanization).  However, most of the other village key-informants agreed that children’s schooling is a distraction from agricultural work. Rural children as early as seven years old are involved in domestic and agricultural family work in these villages (Aghajanian 1988, 95). Among this work, sheep-herding for boys (Mir-Hosseini 1987, 460), carpet weaving for girls, and bringing fresh water for both, were observed as being most helpful to household work. Gradually, as the children grow up (especially in the case of boys) they get involved in 167  cultivation hardship. The skills of traditional agriculture in Hamadan Province, as in other parts of Iran are learned through supervised and continuous practice, a kind of masterapprenticeship relationship within the household unit of production. Attending the formal education detracts from the learning process of the young villager by interrupting regular work along with the household long-time farmer(s) and by not offering alternative agricultural knowledge and experience.  Consequently, according to the old farmer in  Maanizan, the rural children are raised ‘without callous hands” (i.e., unable to tolerate backbreaking work) and “without loving soil” (i.e., with no sense of affinity to the farm  -  see: Appendix E).  From a different angle, another participant from Se-miran, a community activist, suggested that the logic is vise-versa.  He elaborated: “Education is instrumental for  youths to get rid of village. It is a bridge to promotion in cities.” Indeed, he connoted that the blame should not be placed on the education, but on the village stagnation. He went on to warn that the village cannot compromise on education if it has to be 87 The challenge, as the Jarban-lou teacher put it, is to expand the education developed.’ while being able to attract the necessary educated labour force in rural areas: “Why not being able to have educated and prosperous farmers?”  When I raised a question about the future of village agriculture in the wake of youth exodus, the rural teacher in Jarban-lou sarcastically answered: “Don’t worry, there are always many lazy pupils who fail to continue their education even at the primary level. Given the increasing unemployment in the cities, most of them are doomed to work on the land for their daily bread.”  In Hassan-taimur, somewhat the same response was  encountered. A young villager stated that he and some of his friends have no option other 187 Interestingly, these farmers are often illiterate but they do recognize the value of education; even though, the life becomes harder for them, when sending their children to school.  168  than that of staying in the village. Educational failure (in some cases by voluntary drop out due to poverty), the harsh, discriminating and dwindling urban jobs for unskilled workers, as well as family obligations and social bonds were given as the reasons for their staying in the village. The idea that “stupid people remain in the village” was repeated 188 many times in different discussions.  Another issue that was brought out at a different stage of interviews was uncertainty about agricultural works and about farmers’ futures. Farmers underscored the calamities they were facing in their work (particularly: draught, flood, crop infestation, and reduction in crops prices) and the uncertainty associated with these for one’s whole life. As an old farmer in Jarban-lou complained: “You never know if you can survive another year or not. Looking back, if everything goes fine, one can only survive. Otherwise, like many, have to leave the village after one or two years of crop failure.” Living with such anxiety prompts the successful farmer to leave the village as soon as accumulating enough money for a new start in the city and as long as having the ability to do so.  Regarding the farmer’s future, strong concerns were raised about the lack of old-age pensions and social welfare in the villages. The disabled villagers (i.e., too old to work, disabled, or retarded) have almost no support, unless their children provide support for them. Access to social welfare in the cities is remarkably better than in the rural areas. A far-sighted villager takes this into account when considering migration and even considers some present losses in return for future gains (i.e., the enhancement type of causes).  In Maanizan a middle-aged councilor described another limitation of rural life causing the out-migration of well-to-do villagers. He elucidated that the rural economy has a limited 188  One of the most famous poets in Persian literature, Mo’iavi, has a well-known poem exhorting: “Do not go to the village. Village makes the man idiot.” (verbatim translation)  169  capacity of growth for each business. Having a contracting economy on the basis of semisubsistence and semi-commercial cropping, the non-farm business of the village cannot expand itself. Augmenting other cultural barriers, like narrow-minded and parochial views with respect to successful business, the local investors (entrepreneurs) prefer to go to cities, if possible.  Another councilor in Jarban-lou revealed that it was not socially  acceptable to be rich while most of the community was striving to meet ends. Thus, the big capitals and the entrepreneurial individuals tend to leave the almost closed circle of the village to the urban anonymity (i.e., the promotion type of cause).  In general, there was unspoken agreement regarding the domination of subordination 89 Also, in all the sample villages, the precedence of cause for rural female out-migrants.’ out-migration streams underscored as a facilitating factor, not a generating one. It was mentioned that exaggerated or exceptionally successful ex-villagers motivated the already 90 migration-prone villagers.’  I asked the village key-informants about their ideas as to how significant are the cities’ amenities and urban excitements in the youths’ cause of out-migration (i.e., the bright lights hypothesis). Except for one participant, others strongly rejected the cause. 191  The  unanimous approach was that a trade-off was seen between the friendly and congenial milieu of the home village on one hand, and the better income and job opportunities in the target city on the other hand. In their view, the cities’ amenities are not considered in a  However, in Maanizan (with a higher level of school-girl enrollment comparing to the others) it was said that some brides set the condition of migration to the city before getting married. 189  190  The same observation has been made in the south of India (refer to; Singh 1989, 153).  191  A rare empirical study on rural out-migration in Garm-sar (a township 120 kilometers to the west of Tehran) indicated the insignificance of bright lights in villagers decisions to migrate (Hamidi et al. 1983, 193). Nevertheless, it signifies the presence of basic services (e.g., electricity, potable water, road, public bath, school) in the villages with less out-migration. The cardinal cause of out-migration was seen to be a shortage of agricultural water (Ibid. 191) which implies that there is population pressure on resources.  170  mostly survival type of decision-making. For instance, a father of three out-migrants in Se-miran objected to my suggestion by saying: ‘No, it is not the bright lights of city; rather it is the dimness of village which drives the youth.” He later complained of the mass media propaganda about “the