PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE Fifi % at} 7.009 TfTBUF 6/01 c:/ClRC/DateDue.p65—p. 15 AN EXPLORATORY STUDY OF GENDER PATTERNS IN URBANWARD MIGRATION IN VIETNAM By Dung Thi Kieu Vu A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Sociology 2002 ABSTRACT AN EXPLORATORY STUDY OF GENDER PATTERNS IN URBANWARD MIGRATION IN VIETNAM By Dung Thi Kieu Vu Migration is not a new phenomenon for the Vietnamese people. This phenomenon has tended to be more frequent during Vietnam's transition period from a centrally commanded economy to a market economy. This study compares patterns and characteristics of female and male migration in Vietnam in order to see whether women and men’s movement is similar or different. Moreover, this study explores the effects of demographic attributes and human, financial, and social capital on men and women in the migration process, that is. the decision-making process, the moving process, and adaptation outcomes. This study’s model of gendered patterns of urbanward migration suggests that the migration process is strongly shaped by personal attributes (including human and financial capital), and the social relations and assistance (i.e., social capital) that male and female migrants obtain. To test this model, I use a 1996 data set on internal migration into Hanoi produced by the Ministry of Labor, Invalids and Social Affairs (MOLISA), in addition to other sources from the 1989 census, the 1992-93 Vietnam living standard survey and the Vietnamese statistical yearbooks of 1986-97. Copyright by Dung Thi Kieu Vu 2002 ACKNOWLEDGMENTS This work would not have been possible without the help of many people. First, I would like to thank Dr. Rita Gallin, my Committee Chair. Since the fall of 1997, when I began my doctoral program in this department, Dr. Gallin has provided intensive. efficient. and never-ending support during my demanding program. Without her, this study would have been impossible to complete. I must next give special thanks to Dr. Steve Gold. Dr. Harry Perlstadt, Dr. Ruben Rumbaut, and Dr. Chris Vanderpool for being constant sources of constructive thoughts and for guiding me towards my degree. I would like to offer my special appreciation to Dr. Tom Conner, Dr. Fred Robert (Department of Anthropology), and Dr. Kyung-Soo Chun (Seoul National University). for their help and support in my coming here. I must recognize Harvard-Yenching Institute, for their generous scholarship. I especially want to thank Dr. Edward Baker, for his confidence in my abilities and emotional support, and Mrs. Laura Epperson, for her excellent paperwork regarding my case. I also want to thank the Department of Sociology at Michigan State University (MSU), for its financial and scholarship support; the Graduate School of MSU, for the Dissertation completion fellowship. I would also like to acknowledge: Dr. Do Van Hoa, Head of Devision of Population Relocation at Ministry of Agriculture and Dr. Trinh Khac Tham, Vice- president of College of Management of the Ministry of Labor, Invalids, and Social Affairs, for their allowance to use the 1996 MOLISA data set; Dr. Randall Fotiu, for his excellent help and feedback in statistical analyses; Andrew McCullough. Patricia Walters, Cathy Fleck and Nevin Leder for their wonderful friendships and efficient support in correcting my papers; Mrs. Tammy Dennany, from the Sociology Department, for her constant cares in making sure I met all the necessary deadlines during my doctoral program. Finally, and most of all, I am eternally grateful to my family and all my friends for their lifelong encouragement. their love and endless support. TABLE OF CONTENTS LIST OF TABLES ................................................................................................. LIST OF MAP ....................................................................... LIST OF FIGURES ................................................................................... INTRODUCTION ..................................................................................... CHAPTER 1: VIETNAMESE SOCIO-ECONOMIC AND HISTORICAL ............ CONTEXT OF MIGRATION Introduction ................................................................................... Vietnamese migration in the feudal regime (before 1858) . . . Confucianism and Vietnamese tradition in the feudal regime .......... French colonialization and migration characteristics (1859-1954) ............. The war against America under the socialist system and migration patterns (1955-1975) ................................................................................ Women and the socialist system. Migration and the market economy (after m197..5). Migration characteristics and patterns... Women and migration in the transition period... .. Summary ....................................................................................................... CHAPTER 2: MIGRATION THEORIES AND LITERATURE REVIEW Introduction ....................................................................................... Migration theories ............................................................................. Human and social capital theories ofmigration Human capital theories... Social capital theories... Integrative approach to gendered urbanward migration Explanation of the model ................................................................. Literature review and guiding hypotheses... Decision-making process of male and female mlgration Reasons for migration. Making the decision to move............... Moving process........ Type of move. Distance of move" Adaptation outcomes of male and female mlgration Economic improvement... Degree ofsatisfaction..... Summary ..................................................................................................... vi xii xii 14 l4 l7 19 24 27 31 33 2O 40 43 45 45 47 47 47 50 57 60 62 62 62 66 69 69 73 76 77 82 86 CHAPTER 3: METHODOLOGY Study site ...................................................................................................... Data source: research design. sampling. and data collection........... . . . . . . . . . . .. The MOLISA data set ..................................................................... Definition of migrants .......................................................... Sample size and sample design of the MOLISA survey ........ Questionnaire... Data selection for this study. . Conceptualization and measurement 0th main variables” Independent variables ..................................................................... Dependent variables ........................................................................ Analysis strategy ...................................................................................... . CHAPTER 4: PORTRAYAL OF URBANWARD MIGRANTS IN THE TRANSITION PERIOD IN VIETNAM .. . Introduction ............................................................................................... General features of migrants’ places of origin” . . General picture of the Vietnamese situation in the I98Ds and 19905.. Places of origin .............................................................................. . Migrants’ attributes ..................................................................................... Sex .................................................................................................. Age ................................................................................................. . Educational level ............................................................................. Marital status .................................................................................. . Employment status ........................................................................... Occupations ................................................................................... . Income ............................................................................................ . Summary and discussion ............................................................................ . CHAPTER 5: DECISION-MAKING PROCESS Introduction ............................................................................................... . Results and discussion ................................................................................. Reasons for migration ...................................................................... Chi- -square results of reasons for migration by six nominal variable... . ANOVA resiiits of reasons "for migration by three ordinal variables... Results of logistic regressions for reasons ”for migration controlling for gender by each of eight independent variables and tests of hypotheses ................................ Results of logistic regressions of reasons for migration by all significant independent variables ............................... Decision- makers in the migratory process. .. Chi- -square results of decision- makers of migration by” six nominal variables” vii 90 90 90 91 91 91 93 94 95 95 110 118 121 121 122 122 124 132 132 133 137 139 141 144 145 148 152 152 155 155 155 160 163 I72 174 175 ANOVA results of decision-makers of migration by three ordinal variables... Results of logistic regressions for deCISion makers of migration controlling for gender, by each of eight independent variables and tests of hypotheses . . . . . . . . Results of logistic regressions of decision-makers of migration by all independent variables Summary .................................................................................................... . CHAPTER 6: MOVING PROCESS ..................................................................... . Introduction ................................................................................................. Results and discussion ............................................................................... . Type of migration ............................................................................ Chi— —square results of type of migration by six nominal variables... ... ANOVA results of ty pe of migration by three ordinal variables ............................................................................ Results of logistic regressions for type of migration. controlling for gender, by each of eight independent variables and tests of hypotheses... . Results of logistic regressions of type of migration by amll independent variables ....................................................... Distance of migration ...................................................................... ANOVA results of distance of migration by six nominal variables... Correlation results of distance of migration by three ordinal variables... . Results of linear regressions for distance of migration, controlling for gender, by each of eight independent variables and tests of hypotheses... . Results of linear regressions of distance of migration by all” independent variables” . . Summary .................................................................................................. . CHAPTER 7: ADAPTATION OUTCOMES OF MIGRATION Introduction .............................................................................................. . Results and discussion ................................................................................. Economic improvement ................................................................... ANOVA results of economic improvement by six nominal variables. . ... Correlation results of economic improvement by three ordinal variables... . . . .. Results of linear regressions form economic improvement controlling for gender. by each of eight independent variables and tests of hypotheses ................................ Results of linear regressions of economic improvement by all viii 183 186 194 197 202 202 204 204 204 208 210 216 218 218 222 223 229 230 235 235 238 238 239 243 245 independentvariable........................... Degree of satisfaction“ Summary ...... CHAPTER 8: SUMMARY AND CONCLUSIONS .................................................... Findings ........ Implications .. ANOVA results of degree of satisfaction by six nominal variables. Correlation results of degree of satisfaction by three ordinal variables... . . Results of linear regressions for degree of satisfaction controlling for gender. by each of eight independent variables and tests of hypotheses. Results of linear regressions of degree of satisfaction by all independent variables. ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Limitations and future research .................................................................... APPENDICES .......... Appendix 3.]: Appendix 3.2: Appendix 5.1: Appendix 5.2: Appendix 5.3: Appendix 6.1: Appendix 6.2: Appendix 6.3: Appendix 6.4: Appendix 6.5: Appendix 7.]: Appendix 7.2: Appendix 7.3: Appendix 7.4: Appendix 7.5: Appendix 7.6: Appendix 7.7: Appendix 7.8: Appendix 7.9: BIBLIOGRAPHY ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Means of occupation by education before moving ............... Means of occupation by income before moving .................. Frequency distribution of marital status by reasons for migration of those working in the informal sector .............. Mean of age by reasons for migration ............................. Mean of type of relation by education before moving .......... Means of income before moving by type of relations ........... Mean of distance of move ........................................... Frequency distribution of education by age ....................... Means of employment status by education before moving. Correlation of income by education before moving ............. Mean of current education and current income by Gender. . . .. Frequency distribution of employment status before and after moving by marital status for men .................................. Mean of income change by marital status for men ............... Mean of current occupation by education in Hanoi .............. Means of occupation change by gender ........................... Frequency distribution of occupation status before and after moving by sex ......................................................... Correlation of current income by current education ............. Frequency of employment status before moving by reasons for migration .......................................................... Mean of employment status in Hanoi by past education ........ oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo I 253 254 254 257 259 265 266 271 271 286 294 298 299 299 299 299 299 299 300 300 300 300 300 301 301 301 301 301 302 302 302 303 Table 1.1 Table 1.2 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 4.7 Table 4.8 Table 4.9 Table 4.10 Table 5.1.1 Table 5.1.2 Table 5.1.3 Table 5.1.4 Table 5.1.5 Table 5.1.6 Table 5.1.7 Table 5.1.8 Table 5.1.9 Table 5.1.10 Table 5.1.1 1 Table 5.2.1 Table 5.2.2 Table 5.2.3 Table 5.2.4 Table 5.2.5 Table 5.2.6 Table 5.2.7 Table 5.2.8 LIST OF TABLES Urban - Rural Population of Vietnam, 1976-96 .......................... Ha Noi Population Structure in 1986-96 ................................... Population and Employment Structure of Places of Origin of Migrants in Hanoi in 1986 and 1996 ................................... Distribution of Types of Work of Vietnamese Population in 1992-93 Distribution of University and Colleges by Regions of Vietnam in 1994 ....................................................................... Age Distribution of Vietnamese Population in 1989 and of Migrants of This Study .............................................................. Distribution of Age and Gender of Migrants ............................... Distribution of Education of Migrants and Vietnamese Population in 1989 by Gender and Areas .............................................. Distribution of Marital Status by Gender of Migrants and Vietnamese Population in 1989 ........................................................ Distribution of Employment Status by Gender of Migrants and Vietnamese Population in 1989 ........................................ Distribution of Occupation by Gender of Migrants and Vietnamese Population in 1989 ...................................................... Distribution of Income by Gender of Migrants and Vietnamese Population in 1992-93 ................................................... Cross-tabulations of Reasons for Migration by Gender .................. Cross-tabulations of Reasons for Migration by Marital Status ......... Cross-tabulations of Reasons for Migration by Occupations ........... Cross-tabulations of Reasons for Migration by Employment Status. . .. Cross-tabulations of Reasons for Migration by Social Connections. . .. Cross-tabulations of Reasons for Migration by Type of Relations. . . ANOVA Results of Reasons for Migration by Age ....................... ANOVA Results of Reasons for Migration by Income ................... ANOVA Results of Reasons for Migration by Education ............... Coefficients of Logistic Regressions for Reasons for Migration by Each of Eight Independent Variables, Controlling for Gender. . Coefficients of Logistic Regression for Reasons for Migration by All Independent Variables ............................................... Cross-tabulations of Decision-makers of Migration by Gender ........ Cross-tabulations of Decision-makers of Migration by Marital Status. Cross-tabulations of Decision-makers of Migration by Occupations Cross-tabulations of Decision-makers of Migration by Employment Status ....................................................................... Cross-tabulations of Decision-makers of Migration by Social Connections .............................................................. Cross-tabulations of Decision-makers of Migration by Type of Relations .................................................................. ANOVA Results of Decision-makers of Migration by Age ............. ANOVA Results of Decision-makers of Migration by Income ........ 34 39 127 131 I32 134 136 138 140 142 145 146 155 156 157 158 159 159 160 161 162 164 173 175 176 178 179 180 182 183 184 Table 5.2.9 Table 5.2.10 Table 5.2.11 Table 6.1.1 Table 6.1.2 Table 6.1.3 Table 6.1.4 Table 6.1.5 Table 6.1.6 Table 6.1.7 Table 6.1.8 Table 6.1.9 Table 6.1.10 Table 6.1.11 Table 6.2.1 Table 6.2.2 Table 6.2.3 Table 6.2.4 Table 6.2.5 Table 6.2.6 Table 6.2.7 Table 6.2.8 Table 6.2.9 Table 7.1.1 Table 7.1.2 Table 7.1.3 Table 7.1.4 Table 7.1.5 Table 7.1.6 Table 7.1.7 Table 7.1.8 Table 7.1.9 ANOVA Results of Decision-makers of Migration by Education Coefficients of Logistic Regressions for Decision-makers of Migration by Each of Eight Other Independent Variables, Controlling for Gender .................................................. Coefficients of Logistic Regression for Decision-makers of Migration by All Independent Variables ............................. Cross-tabulations of Type of Migration by Gender ....................... Cross-tabulations of Type of Migration by Marital Status .............. Cross-tabulations of Type of Migration by Occupations ................ Cross-tabulations of Type of Migration by Employment Status ........ Cross-tabulations of Type of Migration by Social Connections ........ Cross-tabulations of Type of Migration by Type of Relations .......... ANOVA Results of Type of Migration by Age ........................... ANOVA Results of Type of Migration by Income ........................ ANOVA Results of Type of Migration by Education .................... Coefficients of Logistic Regressions of Type of Migration by Each of Eight Other Independent Variables, Controlling for Gender ..................................................................... Coefficients of Logistic Regression of Type of Migration by All Independent Variables ................................................... ANOVA Results of Distance of Migration by Gender ................... ANOVA Results of Distance of Migration by Marital Status ........... ANOVA Results of Distance of Migration by Occupation .............. ANOVA Results of Distance of Migration by Employment Status ANOVA Results of Distance of Migration by Social Connections .. . ANOVA Results of Distance of Migration by Type of Relations ...... Correlation coefficients of Distance of Migration by Age, Income, Education, and Work Experience ....................................... Coefficients of Linear Regressions of Distance of Migration by Each of Eight Other Independent Variables, Controlling for Gender ..................................................................... Coefficients of Linear Regression of Distance of Migration by All Independent Variables ................................................... ANOVA Results of Economic Improvement by Gender ................. ANOVA Results of Economic Improvement by Marital Status ......... ANOVA Results of Economic Improvement by Occupation ............ ANOVA Results of Economic Improvement by Employment Status. ANOVA Results of Economic Improvement by Social Connections. ANOVA Results of Economic Improvement by Type of Relations. Correlation coefficients of Economic Improvement by Age, Income, and Education ............................................................. Coefficients of Linear Regressions of Economic Improvement by Each of Eight Other Independent Variables, Controlling for Gender ..................................................................... Coefficients of Linear Regression of Economic Improvement by All Independent Variables ........................................ xi 185 I87 195 204 205 205 206 207 207 208 209 209 211 217 218 219 219 220 221 221 222 224 230 239 240 240 241 241 242 243 246 253 Table 7.2.1 ANOVA Results of Degree of Satisfaction by Gender .................. 255 Table 7.2.2 ANOVA Results of Degree of Satisfaction by Marital Status ........... 255 Table 7.2.3 ANOVA Results of Degree of Satisfaction by Occupation .............. 256 Table 7.2.4 ANOVA Results of Degree of Satisfaction by Employment Status. . . .. 256 Table 7.2.5 ANOVA Results of Degree of Satisfaction by Social Connections. 257 Table 7.2.6 ANOVA Results of Degree of Satisfaction by Type of Relations ...... 257 Table 7.2.7 Correlation Coefficients of Degree of Satisfaction by Age, Income. and Education ............................................................ 258 Table 7.2.8 Coefficients of Linear Regressions of Degree of Satisfaction by Each of Eight Other Independent Variables, Controlling for Gender ..................................................................... 260 Table 7.2.9 Coefficients of Linear Regression of Degree of Satisfaction by All Independent Variables .................................................. 266 LIST OF MAP Map I: Administrative map of Vietnam ...................................................... 15 LIST OF FIGURES Figure la: Gender model of urbanward migration in Vietnam ........................... 59 Figure 1.b: Empirical model of urbanward female and male migration in Vietnam... 96 Figure 2: Empirical models of gender patterns of urbanward migration in Vietnam ............................................................................ 280 Figure 3: New theoretical framework for gender patterns of urbanward migration in Vietnam ....................................................................... 289 xii INTRODUCTION AN EXPLORATORY STUDY OF GENDER PATTERNS IN URBANWARD MIGRATION IN VIETNAM This study addresses the issue of rural migration into cities in a country experiencing economic restructuring. More specifically. this study examines differences in the internal migration of women in contrast to that of men, focusing on urbanward migration to Hanoi in the northern part of Vietnam. Vietnam is a good case to test the relationship between gender and the migration process for four reasons. First, Vietnam has been strongly influenced by different foreign cultures. especially Confucianism with its patriarchal ideology, which shape social structure as well as the social and geographical mobility of the population. Second, Vietnam has experienced continuous foreign invasions (Chinese, French, and American) that, to some extent, lessened rigid Confucian norms for women during struggles for national independence. This lessening also results from specific Vietnamese traditional customs and norms that assess women’s roles and contributions highly. Third, Vietnam experienced a Socialist system for over 50 years, a system which legislated equal rights for women in the household and society and encouraged their mobility. Fourth, Vietnam is now experiencing a period of innovation that abruptly impacts the roles and status of “modern" (socialist) women with respect to their ability for social and geographical mobility. That is. women’s opportunities and choices seem to be problematic lately due to the occurrence of market mechanisms (Moghadam 1995a.1995b. Berry 1996, Frenier et al. 1995, Bui Thi Kim Quy 1996). These four factors have affected and continue to affect women's life chances in Vietnam. But they also affect those of men and their relations with women, i.e.. gender relations in Vietnam. The purpose of this introduction is to provide general information about gender and the transition period in Vietnam and to discuss how these concepts are used in this dissertation. Then, the structure ofdissertation is introduced. Gender and gender ideology in I-"ietnam Gender, like culture. is a human production. There are many different definitions of gender. First, gender is a process, in which social differences that define “women’ and “men” are created. In social interaction, individuals learn what is expected and act and react in expected ways, thus simultaneously constructing and maintaining the gender order. Indeed, the gender order depends on everyone constantly “doing gender” (West et al. 1995, Lorber 1994). Second, as a social institution. gender creates distinguishable social statuses for the assignment of roles and responsibilities (e.g., women’s roles and responsibilities as daughters, mothers. or wives) (West et al. 1987. Spelman 1988). Third, as a part of a stratification system that ranks these statuses unequally, gender often treats women as a marginalized or subordinated minority group (Connell 1989, Glenn 1999). Specifically, in most societies, women and men are perceived as different and are evaluated differently. These supposed differences in female/male characteristics and capabilities justify the power differences between them. Men are assumed to be more capable and, thus, have more access to resources and high positions in society than do women. In all societies, the belief that women and men are inherently different provides the justification for allocating certain kinds of work to women and to men. Obviously, social institutions. particularly work and family, and the positions within them are id gendered: occupations are segregated by gender; main kin statuses are gendered, and signified by names (mother and father, daughter and son, sister and brother) (Berk 1985). Different names demonstrate different positions and a ranking system in the household that are based on male superiority and old age privilege. These rankings and positioning. in general, consequently affect social relations between women and men. In other words, gender is used to denote socially and culturally determined differences between women and men. Gender is a binary and hierarchical system that opposes male to female and masculinity to femininity, not on an equal basis but in hierarchical order; gender systems rank the traits and activities associated with men higher than those associated with women. As a result, it is clear that gender is a socially- constructed perception about social relations between women and men within the family and in the major institutions of society such as the hierarchies of large-scale organizations and the occupational structure (Acker 1989, Reskin 1989). In this dissertation, then, gender is referred to as a socially-constructed system of male-female power relations that structures the life chances of women and men by assigning them different roles and positions which bear unequal value. This concept of gender will help explain how women and men were involved in moving and resettling and why their patterns of migration may be different or similar. A concept closely related to gender is patriarchy. There are many different definitions of patriarchy, but in this discussion, I adopt the following: “Patriarchy is a social system in which family systems or entire societies are organized around the idea of father / man rule. This system is based on the principle of male dominance and of giving primary attention and importance to men and what men do” (Fisher 19802127). According to this definition. society provides men privilege in their choices and opportunities. Male privilege affords advantages to men at the expense of women. For example, men are given greater access than women to high education and well-paid jobs. thereby leading to men’s greater socio-economic and political power. Men’s privilege also can be seen through the different evaluations given to the tasks done by women or men. The same tasks might be highly valued if they are done by men. but lowly valued if they are done by women. The term patriarchal is used to describe any society where men monopolize authority, and where women's access to it is restricted. As a result, it is clear that the outcome of patriarchy is similar to that of gender. Because of these similar outcomes, gender and patriarchy are used interchangeably in this study. So too are the terms gender and patriarchal ideology. I define ideology as a system of beliefs that is created by those with power (usually men) and that serves their interests (Hardwick 1998). Gender/patriarchal ideology, then, is a socially-constructed system of ideas about women‘s and men’s roles and positions, rooted in beliefs that men should dominate households (as family heads) and society (as power holders and decision-makers) while women should be relegated to subordinate positions (as housewives and low-paid workers) (Herbert 1993). According to patriarchal ideology, men are superior and women are inferior. Thus. gender or patriarchal ideology produces, reproduces, and legitimates the opportunities people have or do not have, on the basis of their biological sex. As a result. gender/patriarchal ideology reinforces male privilege, and this ideology is widespread throughout the world. This gender/patriarchal ideology exists regardless of women’s actual contributions and roles at home and at work because it protects the privilege of men. Obviously. gender/patriarchal ideology plays an important role in explaining the gender division of labor as well as social relations at the societal and household levels in a society such as Vietnam. Similar to many cultures, in Vietnam. gender is associated with roles and positions that are assigned either to women or men. For example, being a woman is associated with caring for the household and its members. while being a man is associated with leading and economically supporting the household. As in any other patriarchal societies, women in Vietnam are expected to take care of housework, children, sick family members and the elderly while men are expected to deal with “more important” matters inside and outside the household (e.g., housing matters, career decisions, strategies for survival of the household, interaction with communities, and participation in social activities). Women are generally held to higher standards of family responsibility than are men. Women also are rewarded for being feminine, physically weak, emotional, passive, and housewives. In contrast. men are rewarded for being masculine, strong. unemotional, active, and heads of households or decision-makers. As might be expected then, in Vietnam, as in the world in general, gender segregation in occupations exists. Women’s occupations are different from those of men. For example, women tend to occupy jobs such as those of teacher, nurse, nun, and secretary or work in light industries. services, or low-skilled jobs. Men, in contrast, tend to work as supervisor, doctor, priest, president. or in heavy industries, high technology, or high-skilled jobs. These differences are rooted in gender ideology that discriminates against women and privileges men. During the socialist regime, the status of Vietnamese women actually improved, even though their status was still lower than that of men. The government attempted to eliminate gender inequalities. Women were legally provided equal rights to men in education and employment as well as in participation in different social activities. Such an effort can be seen in the continual propaganda disseminated to recruit women into occupations and social activities previously dominated by men. The public media widely propagated new ideas about women’s roles and capability in working and learning, and men’s responsibility for sharing housework with their wives. These ideas were intended to create a new image of socialist women and men (i.e.. women can do men’s jobs and vice versa). This effort helped lessen gender occupational segregation, although this segregation was not completely removed. Moreover. to lessen women’s domestic labor. the government provided different services to women such as community kindergartens and common kitchens. With these efforts, women had more chance to obtain higher education, better and well-paid jobs, and more voice in their households than in the past. This campaign, however, did not cover all aspects of life. For example, traditional customs were not targeted and they remained in place. Customary laws still ordered gender relations in the way that favored men; for example, a woman moved to live in her husband’s house upon marriage, and land was divided among sons rather than among daughters and sons. As a result, women continue to have a lower status and position than men; they still are more closely associated with childcare and housework than are men. In addition, because of the mixed influence of traditional gender ideology and the new image of gender roles in the socialist system, women tend to have a double workload: they work both as employees and as housewives. Transition to a market economyfrom a centrally-commanded economy Prior to colonialism (before the end of the 19th century), Vietnam was primarily an agricultural society with a weakly developed manufacturing sector. This economy was essentially a self-sufficient system in which the household was the core economic unit. The family, rather than the individual. was the unit of taxation and it was responsible for recruiting labor for public service (e.g., construction of roads, bridges, or king’s palaces and military service) by the feudal system. During French colonialism (the end of the 19th century and the early 20th century), heavy industrial production was introduced into the economy. The French explored natural resources (mining) to serve French development through industry. Thus, Vietnamese industry developed slowly, and it was biased toward natural resource extraction. Upon the arrival of communism, Vietnam was primarily a rice-production economy with a weak industrial sector. Then, from 1954 to 1975, the country was divided into two parts: the north was ruled by communism and the south was ruled by America or capitalism. This period was characterized by the development of both light and heavy industries. At the time of reunification (1975), Vietnam was one country with two economic systems. A form of comprador capitalism had emerged in the south that was heavily distorted by the need to service a large foreign military force, while in the north a system of central planning had emerged whose primary purpose was to mobilize resources for the war effort. The task after reunification was to transform these two incompatible economic systems into one and then redirect it so that it would serve developmental purposes (Tran Hoang Kim 1992.) In the pre-reform period (before the 19803), Vietnam was characterized by a centrally-commanded or planned economy. This socialist system was characterized by subsidies,l collectivism,2 and a household registration system.3 The household registration system limited population mobility within and between regions because it included complicated registration procedures and strict requirements (e.g., only those who were employed could be registered). Under the command system, private ownership was eliminated and state ownership was ubiquitous. Production served the war effort, and was used in trade with other countries of the socialist block. The population was mainly employed in the state sector, where people received full welfare benefits (e.g., free education, free healthcare, fully paid sick leave, annual vacation, and pensions) and opportunities for social mobility (e.g., promotion in economic and political status based on their revolutionary contributions and work experience). During wartime. revolutionary contributions (e.g., length of military service and number of medals) were used as the main measurement of people’s capability to fulfill their roles as members or leaders in productive and non-productive organizations or activities. The transformation of the Vietnamese economy began in the early 19805 with the commercialization of agriculture and industry. Socialist collectives were collapsed, and the household again became the core unit of the agricultural sector. Part of the state sector ' Subsidies involved a system in which people received equal portions of food, clothing. and other necessities such as housing for their existence. 2 Under collectivism, land and all production were owned by the state. Production was organized collectively, i.e., farmers worked according to their assigned duties and hours to earn work “points.“ A portion of the products of the harvest in the form of tax went to the government to finance its administration. The remainder of the harvest was divided between farmers based on their work “points." 3 Under the household registration system. people had to be registered at their living place in order to have the right to healthcare, subsidized necessities, education. and employment. (e.g., food processing, transportation, and auto assembly) was privatized, while some core industrial sectors (e.g., electricity, fuel, mines, and oil) remained within the state sector. This prepared the ground for the later market economic reforms in which the state sector was narrowed and the informal sector grew (Wolff 1999). In accord with the different socio-economic conditions, the Vietnamese govemment changed its focus from a war economy to economic development and initiated new norms and principles that valued social, human. and financial capital differently, These shifts had several consequences. First, this transitional period strongly impacted people’s way of thinking and acting. Privatization and liberation of prices replaced the subsidy system, collective ownership. and the distribution of social welfare and economic rewards. Socialist values and norms (e. g., communalism versus individualism) were gradually replaced by market ideology and mechanisms. For example, the revolutionary contributions of workers as the main criterion for employment and promotion was replaced by professional work skills and education; production based on plans for exchange within the socialist block was replaced by production based on incorporation into the world market. Human capital became more valued than revolutionary contributions. In fact, human capital gradually gained its primary position in the labor market after national unification, when the main goal of the government was economic development rather than the mobilization of resources for the war effort. Second, because, under the socialist system, the elimination of inequality was the aim of all governmental efforts, the state tried to minimize the income gap between rich and poor. Financial capital, thus, did not play an important role in the ranking of social classes. For example, there were no rich or poor people. but intelligentsia, farmers, or workers, who were assumed to be in different occupations. but not at different income levels. Nevertheless, with the introduction of market mechanisms, financial capital became one of main criteria to classify social classes and statuses. Third, traditionally, the Vietnamese strongly relied on their social networks to survive or to achieve their goal because of the shortage of economic opportunities in an agrarian economy. Under the socialist system. people’ economic situation was equally safe because of the subsidized system. Theoretically. people also had equal access to employment and other economic opportunities because of strict and rigid employment procedures. Thus, the effects of social connections tended to be weaken than they had been in the past, pre-socialist period. In other words. the importance of social capital as a means to maximize economic opportunities was submerged. During the transition into market economy, however, the importance of social networks for an individual’s development reemerged and increased. This is because market mechanisms are characterized by flexible employment contracts and terms that enable social connections to channel their members into the labor market. The other reason for the increasing importance of social capital may be that people have experienced economic uncertainty because of their lack of experience in dealing with market mechanisms. Therefore, they use their social networks to explore different economic opportunities in accord with their human and financial capital levels. While the collectivism of the socialist system failed to encourage people’s attention to productivity and efforts to improve their skills, privatization with its stress on individualism does because the household. as an economic unit, is directly linked to each member’s effort. Human, financial, and social capital, thus, helps people to reduce risk and maximize economic opportunity in a changing context. The private sector initially built on the household economy. The household is the core institution responsible for the economic survival of its members and it is responsible for labor allocation, based on its members’ market attributes (levels of education and skills, or human capital, and physical strength). The rationale for the success of the household economy is directly tied to the efficient utilization of men and women’s work. This means that in this transition period, the Vietnamese people tend to experience the effect of the market economy in ways similar to that of citizens in other countries. Those with extensive social networks and high levels of human and financial capital tend to have greater opportunities to succeed economically than do their less-endowed counterparts. In sum, the application of market mechanisms in Vietnamese society has led to crucial changes in all aspects of life at the national, regional, local, household, and individual levels. Competition in marketing and trading has stimulated improvements in production and the rational use of the labor force, causing a labor surplus, especially in rural areas. Innovation with flexible employment contracts (full time or part time) and the elimination of the household registration system have tended to deepen differentiation in terms of the unequal distribution of economic opportunities between regions. In general, cities or urban areas provide more opportunity for people to be employed than rural areas because the market economy has primarily stimulated changes in urban economies in which public and personal services have burgeoned. Rural areas differ greatly from one another in the way they provide more or less income and employment opportunities because of factors such as population density. location. resources, and geographical conditions. Under these new conditions, voluntary migration has rapidly expanded as a means of redistributing the labor force and generating wealth. However, what the nature of female migration is, how women are involved in the migration process, and what the consequences are for them under the transition to the market economy are unknown. Answering these questions is the aim of this study. It is known that economic restructuring does not affect the labor force participation rate per se of women because they have always been active in the labor force. Nevertheless, it affects the structure of employment, providing flexible employment contracts (full-time, part-time, or seasonal contracts) that deepen gender segregation in occupations. For example, women are mainly employed in light industries or the service sector and tend to cluster in the urban informal sector and home-based work. Whether this is true or not for female migrants will be examined in this work. The details of male and female migration will be uncovered by testing hypotheses generated from social capital and human capital theories of the migration process. Social capital theories emphasize the role of networks (social relations and assistance) in the decision-making, moving, and adaptation processes of migration. Human capital theories emphasize the importance of personal attributes such as education, work skills, and demographic indicators (age, sex, marital status) for male and female migrants in moving and resettling. Financial capital is assumed to stimulate people to migrate. This study contributes to the existing literature by testing theories and generating new insights into the determinants of migration and the resettlement process of Vietnamese migrants. By testing hypotheses derived from these theories, this study seeks to explain the nature and consequences of the male and female migration process in a restructuring economy. This dissertation also seeks to explain the impact of development on women’s status in a market economy. Structure oft/re dissertation The dissertation is organized as follows. Chapter 1 describes the Vietnamese historical context that shapes the male and female migration process and gender relations. Chapter 2 includes the problem statement, reviews the literature on theories of social, human, and financial capital in migration and discusses issues of urbanward migration in Vietnam. Chapter 3 reviews methodology including research design, sampling and data selection for the study and conceptualization and operationalization of the main variables. Here, I provide general information about the data set used in this study, i.e, the Ministry of Labor, Invalids and Social Affairs (MOLISA) data set. Chapter 4 describes the characteristics of migrants in Hanoi in comparison to the national population. Chapter 5 explores the effects of social, financial, and human capital on the decision-making process. Chapter 6 presents the influence of social, financial, and human capital on the moving process. Chapter 7 explores adaptation outcomes in relation to the social, financial, and human capital of male and female migrants. In the conclusion, I summarize the results of the study, relate the findings to the theoretical perspectives used, and offer suggestions for future research. CHAPTER 1: VIETNAMESE SOCIO-ECONOMIC AND HISTORICAL CONTEXT OF MIGRATION Introduction Vietnam is located in Southeast Asia, bordered by China in the north, Laos and Cambodia on the west, and by the East (Tonkin) Sea on the east. Vietnam consists of seven regions: Northern Upland, Red River Delta, Northern Central, Central Coast, Central Highland, Southeast, and Mekong Delta. Hanoi is located in the Red River Delta in the northern part of Vietnam, while Ho Chi Minh City is located in the Southeast in the southern part of Vietnam (see Map I). The country covers an area of 329,600 square kilometers, of which about 25 percent is forested and about 21 percent cultivated. The remaining area (54 %) is un—arable for agricultural production. Because of this, the forests have been badly overexploited for wood and agricultural cultivation. The majority of Vietnam's population is concentrated in the limited arable area (21 %). The total population of Vietnam is estimated at 76.7 million, (GSO 1997), and the average increase in population is more than 1 million per year. The population is basically rural (85 % in 1960, 80 % in 1987, 79 % in 1994, and 78 % in 1997) (GSO 1996:28), and is concentrated in the two main rice-growing deltas: the Red River Delta in the north and the Mekong River Delta in the south. In the Red River Delta the population density averages 989 per square kilometer, excluding Hanoi (2.685 per square kilometer), and Thai Binh (1.141 per square kilometer). These densities are among the highest in the world for a rural area. The Mekong River Delta has a far lower population density than the Red River Delta, 385 per square kilometer (GSO 1998z7). Map 1: Administrative map of Vietnam l ‘, Cl” 0' ‘ ‘9 Mount-d 5‘5.“ THAILAND Vim "MW * War 0 MW“ ”hula-nu.- ~M~~fl~mu ...-Mm O O arm-- h .m‘ A Worm CAMBODIA Sam: region In Vim-u 1. Northern UpIIId region 2. Red River Delta region 3- Northern Central region a. Central Coast ream 5. Central Highl‘md region 6 Saunas! region 7. W The country‘s delta population is almost entirely ethnic Vietnamese (kinh). According to the 1989 Census, ethnic Vietnamese made up four-fifths of the total population in 1979 (87 %) and in 1989 (86.8 %). The remainder are minorities, consisting of 54 different groups with different cultures and traditions, who mainly live in the upland areas of the country, except for the Hoa (Chinese) minority, which mainly lives in cities and towns (Banister 1993243). The minorities primarily are Nung, Muong, Khome, Hoa, Thai, and Tay. The rate of population growth has slowed from 3.1 percent per year in the 19605- 705 to 2.1 percent in the 19805; this rate is expected to fall further, to 2 percent by the year 2000. The most important reasons for this decline from the 19605 to the 19805 are the effects of wars (particularly the American war and the 1978-79 conflicts with China) and large scale emigration (EIU 1986-1997). The sex ratio of the Vietnamese population in the 1979 Census was 94.2 percent (percentage of men per 100 women), one of the lowest ratios in the world compared to the average ratio (106.0%) of some countries in Africa, America, Europe, and Asia (UN 1984: 178-239); it was 94.7 percent in 1989. Vietnam has a young age structure; the median age of the population in 1979 was about 18 to 20, but it grew to 23.5 in 1989 (VPC 1991a216). Historically, the Vietnamese people have always moved around in search of cultivable lands, to avoid wars, or to absorb the labor surplus during agricultural slack seasons. In this chapter, I describe the socio-economic and historical context of migration in Vietnam. Specifically, this chapter focuses on four main stages of the history of Vietnam in terms of migration. i.e., the feudal regime, the French colonial regime, the war against America, and the national unification and innovation period. In each period. I aim to delineate the socio-economic and historical characteristics that affect the volume, trends and patterns of internal population movement. 1. Vietnamese migration in the feudal regime (before 1858) Migration is a common phenomenon in Vietnamese history; three-fifths of the territory of present day Vietnam was mainly obtained by the migration of the Vietnamese people from the north to the south and toward the East Sea over the centuries. This migration aimed to open new land and was encouraged by the feudal regimes which issued incentives such as exemptions from duty services and taxes, and by providing migrants’ titles to unoccupied lands (Dang et al. 1997. Tana 1996, Mackerras 1988). Before 1858, Vietnam was a feudal society with a self-sufficient lifestyle supported by wet-rice production. The feudal society was a patriarchal system, which was govemed by the king. However. since China invaded Vietnam over ten centuries (till 10 AD), Vietnamese were strongly influenced by Buddhism and Confucianism with its norms of male domination and privilege (discussed below). The economy of this society was mainly agrarian and with some manufacturing, and it sustained the operation of the government apparatus through a tax system. The citizens had to deduct a portion of their harvest to pay taxes in the form of raw products or an equivalent amount of money. If they were not able to pay their taxes in this manner. they were required to pay with their labor, as indentured workers. In this society, people exchanged their products within local markets, and communicated mainly within and between their villages of residence. Connections with outsiders were considered an indication of unfaithfulness to their communities and traditions (Phan Dai Doan 1995). Under the feudal regime, Vietnamese society was mainly made up of four classes: Confucian scholars, including rulers (sy), famiers (nong), handicrafters (cong), and traders (thuong); the last two classes had the lowest status in society, while the first two were highly valued and respected. Class rankings resulted from Confucian norms, which highly valued knowledge (education) and farming while discriminating against trading. considering it an un-solid way to earn a living (Nguyen Tu Chi 1993, Tran Dinh Huu 1989, Nguyen Huu Minh 1998). This discrimination toward trading and manufacturing limited the Vietnamese people’s geographical mobility. They tended to be strongly attached to their land. They had no need to enlarge their circle of knowledge and communication. Thus, the structure of villages and feudal society with closed cultural borders shaped the geographical and social mobility of people of different classes and gender. The gap between rich and poor or among high and low status groups was always great because of the concentration of land and wealth in the hands of a small number of upper-class landlords. However, this class stratification was not so rigid, and people could move from the lower to higher class through educational attainment. The system of government in Vietnam was similar to that of the Chinese, which consisted of a complex bureaucracy, in which officials were selected through examinations in Confucian moral and political philosophy (Mackerras 1988, Tran Dinh Huou 1989). Rulers or officials served at different levels of management based on their educational attainment. Thus, high status people had more flexible geographic mobility than did people of the low status. Confucian scholars could move to different places in order to perform their social responsibilities as rulers, while others tended to remain attached to their farmland. In this mainly rural country, there were only a few cities, such as Hoi An and Thang Long (1 1‘h century AD). These cities were administrative and trading centers of the agrarian society. Accordingly. they did not play a leading role either in attracting peasants to migrate from rural areas. or in encouraging the development of non- agricultural occupations. Therefore. population movement seemed to occur between and within rural areas, i.e., it took only a rural-rural form. Because the feudal regime was closely related with Confucianism. in the next section. I characterize Confucianism in Vietnam. Confucianism and Vietnamese traditions in the feudal regime Vietnam was invaded by China in 1 l 1 BC: this invasion lasted for over ten centuries. During this period, Vietnam was a tributary of China. Annually, the Vietnamese government had to pay tribute to China in the form of rare natural resources (precious stones and metals, animals), selected man-made products, and beautiful girls. There were continuous attempts by the Vietnamese people to liberate the nation from the Chinese rulers. Finally, Vietnam achieved its independence from China in 939 AD. However, the Chinese influence was much stronger in what is today the northern part of Vietnam than in the southern part because the north was closer to China than the south (Do Thai Dong 1991). China strongly influenced Vietnam through the spread of, first, Buddhism and then, Confucianism.l Buddhism came to Vietnam in the first century BC and was the dominant religion until the 15th century AD, when Confucianism replaced it as the ' Confucianism appeared early in North China before the 2nd century BC, but it became the formal ideology in the 3rd century AD. Only in the 12th and 13th centuries AD did it become the national dominant “religion” in China. national dominant belief system. Confucianism constituted a guide for behavior and the governing of society and the household, and for structuring the social order (Tran Dinh Huou 1996, Do Thai Dong 1993, Nguyen Binh Yen 1997). This means that superiority. domination, and power belonged to men. Age provided high privilege, position and authority to both men and women in household and society. but women always were in a lower position compared to men. Confucian principles, centered on the belief in men’s superiority. shaping the structure of society and the family with this emphasis. Confucianism highly valued collective interests, particularly family interests, over individual interests. In the family. all members were integrated into a tight relationship of responsibility. Gratitude and responsibility were highly valued, even higher than individual feelings and needs (Do Thai Dong 1993, VHT 1993, Tran Dinh Huou 1996, Nguyen Binh Yen 1997, Phan Dai Doan 1995). Social relationships were based on age and sex differences. The oldest men had the highest authority, while women’s status increased with the birth of sons and age. Confucian ideas also underscored one's roots and hometown. Thus, these perceptions tied people to their birthplace with an endless loyalty. The people’s attachment to their hometown as well as to their family shaped their closed lifestyle. Confucian principles in patriarchal society focused on encouraging women’s silence. reproductive and domestic functions, and female identities through the “three submissions” and “four virtues.” According to the “three submissions”, women had to obey their father prior to marriage; their husband after marriage and, upon a husband’s death, their sons. The “four virtues” included “being a good housewife, having an appropriate appearance, speaking well and having a good personality”.2 The former emphasized women’s subordinate roles, which were completely dependent upon the will of a father, husband, and sons. The latter highly valued women’s submissiveness and compliance to serve men (Nguyen Binh Yen 1997). In this way, women were subjects of men’s domination. They did not have the same rights as men such as the right to participate in the decision-making process at the household and community levels, in social affairs, or schooling. Women were confined to their roles in their family and housework. Nevertheless, the influence of Confucianism was not the same in all parts of Vietnam and was not the same for different groups of people due to their location, economic condition, and social status. Noticeably, the degree of Confucian influence was negatively associated with geographical distance relative to its origin (Frenier et al. 1995). The northern Vietnamese people were more strongly impacted by Confucianism than the southern ones because the northern part of Vietnam bordered on China while the southern did not. Confucian influence also was felt differently by various groups of people (Tran Dinh Huou 1996). People of the higher class tended to be more influenced by Confucian principles than those of the lower class because Confucian principle were beneficial to rulers rather than to citizens. These principles emphasized the strict social order, social relations between king and citizens, and the power of the ruling class. 2 “A good wife” means women need to be good at taking care of everything related to housework and taking care of their own family members and paternal family members. “Appropriate appearance” means women always are physically kempt, and display appropriate manners in different contexts. “Speak well” means women need to know what to say in every situation. “Good personality” means women need to be obedient, submissive, gentle, and stoic (i.e., endure tremendous strain). 21 The difference in the influence of Confucianism also resulted from the fact that the upper class could afford to follow certain Confucian principles and life style. For example. according to Confucianism, good women were never to be seen by outsiders. This criterion tended to be supported by the upper class household since women in these households could stay at home and avoid contact with outsiders, while women of lower classes had to work outside to earn a living. The upper classes. thus, followed Confucianism more strictly than did the lower classes. In short. Confucianism shaped social relations and the social structure of Vietnamese feudal society, in which women and men had different roles and positions. Men had more important roles and a higher position than women in both the household and society. Men had more opportunities than women in social and geographical mobility. Besides the strong influence of Confucianism, the Vietnamese people were shaped by their own traditional norms and values, which, to some extent, lessened the rigid Confucian principles that governed women. Even though Buddhism was supplanted by Confucianism, its influence remained strong in Vietnamese communities, especially in the low classes. Buddhism highly valued virtues such as kindness, charity, and behavior that demonstrated human endurance to overcome difficulties In addition, the Vietnamese people tended to follow their own traditional customs. and values that highly valued national patriotism, communal and neighborhood relationships, and family harmony. Vietnamese people highly valued relationships with neighbors and considered them as important as those of kinship. People could rely on their close neighbors more readily in times of need than on distant relatives, which can be seen through the Vietnamese saying “ban anh em xa. mua lang gieng gan” [neighbors IQ Ix) can replace siblings/relatives living far away]. Due to a history of continuous wars, the Vietnamese tradition also highly valued national patriotism that unified all people into one group with the same aim, a struggle for national independence. This tradition also highly valued women's roles and contributions, and gave women a degree of power in relations with their husbands. For example they could participate in making decisions, or they could replace their husbands as a head of household or breadwinner when their husbands were at the front. Moreover, women were considered and continue to be the “interior chief ‘ (Noi tuong) of the household. Vietnamese women could actively participate in the production process, economic activities, and household matters (Le Thi Nham Tuyet 1975). Women (wives) could interfere in public affairs through men (husbands) and “a man’s commands were a result of his wife‘s advice” (lenh ong khong bang cong ba). Nevertheless, women were better recognized in the household than in society since a wife‘s successes were socially known as her husband’s; men benefited socially from women’s work. A powerful Vietnamese saying is “A husband’s property is his wife’s work.” (cua Chang cong vo). These features of Vietnamese tradition helped soften the rigid Confucian principles and men’s judgment of women. The conflicting influence of Confucianism and Vietnamese cultural traditions has affected the specific condition for gender relations. On the one hand, male dominance at home and in society still prevail. Generally, men are perceived as the breadwinners and the heads of household, and the decisive voice in making decisions belongs to men, not women, even though in many cases women are the actual breadwinners of their households. On the other hand, Vietnamese women are encouraged to share economic responsibility with their men for their household’s betterment. Both have actively IQ b) participated in the labor process as well as in geographic movement; women have gained a degree of power in their households because of their undeniable value and contributions. Nevertheless. women tend to be less mobile than men and they experience gender discrimination; women might contribute more than men to a household’s survival, but they suffer more than men due to their lower position in the household power hierarchy. Their roles are devalued due to a widespread gender ideology in the household and society. In summary, all these cultural features shaped Vietnamese society and people’s behavior and geographical mobility. These features continue to exist and affected people throughout different historical stages of Vietnam such as during French colonialization, the socialist period, and the market transition period. 2. Frer_1ch colonialization and migration characteristics (1859-1954). The thousand-year Chinese invasions stopped in the 19th century. Shortly before French colonialization, trading and exporting activities were not permitted in Vietnam. There were only a few ports with very few international traders visiting. Therefore, internal exchange markets tended to develop at different levels: commune, village, district, province and region (Maskerras 1988). Agrarian and handicraft products had been the main source of self-sufficiency in the household. Small trading and handicrafts became the most important form of additional earnings since farming land was never enough to support a household’s needs in terms of labor surplus and food (Nguyen Tu Chi 1996). Some features of French colonialization could be regarded as positive. Highways. railways and bridges were constructed, resulting in better communication between 24 regions. Public health and sanitation improved and hospitals were built. The French began the country’s industrialization, but not on a full scale. Mainly they developed light industry in Sai Gon, Ha Noi and Hai Phong, and exploited coal and bauxite deposits in Tonkin (Trinh Duy Luan 1996. Tana 1996). Nevertheless. the gap between the rich and the poor widened, while land and wealth became concentrated in fewer hands. The old economy broke down and was replaced by large F rench-owned estates. Basically, Vietnamese society became more segmented and a more diverse social class structure emerged, adding industry workers, traders, capitalists, specialists. or professionals. Vietnamese were second-class citizens in their own country. The ideals of “liberty, equality and fraternity” were for the French, not the Vietnamese. For this reason, resistance to the French was persistent in different ways with traditionalists,3 nationalists,4 modernists,5 and communistsf’ holding different ideas about how to restructure society (VHT 1993, Nguyen Huu Minh 1998). The volume and trends of Vietnamese migration were strongly affected by continuous wars. The severe wars to subdue the Vietnamese by the French and its policies of invasion and extraction of the country’s natural resources strongly affected 3 This movement occurred during the latter part of the 19th century in Vietnam. Its leaders were national patriots. In mid-I665 Kim Ham Nghi made a proclamation calling on the whole country to support the monarchy in struggle (Can Vuong Movement). The main idea of this movement was to struggle against French colonization and to reestablish the feudal regime in Vietnam. 4 The nationalist tendency was clearly expressed by the founding of the Vietnam Quoc Dan Dang (Vietnamese Nationalist Party) in 1927 under Nguyen Thai Hoc’s leadership. The Quoc Dan Dang aimed at gaining national independence and establishing a democratic government. It showed anti-Communist tendencies with a goal to overthrow French colonization by Vietnamese forces, and construct the nation, following the capitalist model. 5 This movement occurred in the first years of the 20th century aiming to free Vietnam from French colonialization by two means, either armed struggle with the help of other countries (led by Phan Chu Trinh), or reform ofthe colonial regime (led by Phan Boi Chau). 6 Its leader was Nguyen Ai Quoc (Ho Chi Minh), and it emerged in the early 20th century in Vietnam. Its program was to overthrow French colonization through the revolution for national independence and to construct Socialism under the management ofthe Communist party ofthe working class. 10 L11 population movement within Vietnam (Trinh Duy Luan 1996). For example, to extract natural resources in Vietnam, the French regime opened new mining zones. where they forced native people to move in order to work. Basically, migration took three forms: seasonal migration between agricultural areas during slack seasons, rural-urban migration of landless people, and the movement of waged laborers to plantation/mining zones during the French colonial period (1859- 1954) (Dang et al. 1997, Tana 1996). The first form (seasonal migration) had the largest volume (two-thirds of residents in the Red River Delta in the 19305). This resulted from the high rate of unsuccessful harvests due to natural disasters (drought, floods, storms) and an insufficient labor force during harvest times in different regions (Ha Dong, Ha Nam, Nam Dinh and Ninh Binh), or in the highlands that extended along the rim of the Red River Delta. The second form (rural-urban) became more permanent during the French period because of the concentration of at least 85 industrial enterprises in the Hanoi and Hai Phong regions by 1905 (Tana 1996) and the establishment of a number of international trading ports in Hai Phong, Sai Gon, and Da Nang. Migrants were mainly landless people from rural areas in search of a livelihood in these industrial enterprises. The third form was closely associated with the extraction of natural resources and the rubber plantations of the French government in the I ndochinese area. Migration became more frequent with the development of mineral and mining extraction industries in Quang Ninh, Hong Gai, and Cam Pha. Women were involved in all three forms of migration. Female migration resulted from the great uncertainty involved with agricultural production and the strategies of the 26 French to recruit women and children. because they were cheaper labor than men, as workers in order to reduce the cost of extraction and production. They moved to the industrial complexes, where they worked under unsafe working conditions and were treated badly by employers. Women were usually employed as porters or carriers, and they were paid much less than men. Their employers rationalized less payment for women by ideologically arguing that they were physically weaker than men. Female migration was also impacted by the development of the service sector in urban areas, which attracted rural women to low-skilled jobs (e.g., carriers, cleaners. petty traders) and domestic services (Le Thi Nham Tuyet 1975). Despite the significant p0pulation movement within regions, urbanization in Vietnam did not reach a high level during the century of French colonialization. For example, the urban population comprised 7.5 percent of the total population in 1931, 7.9 percent in 1936, and 11 percent in 1955. In comparative terms, the urban population increased 3.1 percent while the urban population of the world increased 17 percent (from 31% to 48%) during the same period (1936-55) (Trinh Duy Luan 1996). In short, pOpulation movement during this period was limited mainly because agricultural production dominated the economy, and industrialization moved at a slow pace. 3. The war against America under the socialist system and migration patterns (1954-75) After the defeat of the French in 1954, immediately prior to the partition of the country as agreed in the Geneva Accords, more than 850,000 people, mainly Catholics and small landowners, moved from the north to the south. About 80,000 of those who were loyal to the new Communist government moved from the south to the north (Hitchcox 1994). Vietnam was divided into two: North Vietnam with a socialist form of government, and South Vietnam with a capitalist system under the control of the US. In the north, the socialist regime was based on collective ownership with a centrally-commanded economy. In this society, a system of subsidies provided food stamps and other free social services. such as healthcare, education, housing, or social insurance. Along with this subsidy, a household registration system was implemented in order to identify who could get what and in which place. Basically, people had to be registered in order to get all necessary subsidies, without which they either could not survive or were faced with great uncertainty in their lives (unemployment, no source of income and services). As a result of this subsidized system, the socialist government provided female citizens equal rights in different fields (marriage, education, social activities, and opportunities). Women could get equal portions of food, goods, and work and they could get equal opportunities in pursuing higher education as well as well-paid and high prestige jobs as did men (Le Ngoc Hung 1996, Nguyen Huu Minh 1998). Vietnamese women had more opportunities than in the past to participate in all socio- economic activities; they had more opportunities to move geographically and socially. In the south, the Americans came in the 19505 with the aim to replace French colonialization and to exploit natural resources to serve American capitalist development after World War 11. They also aimed to defeat the evolutionary Vietnamese government and to separate Vietnam from the influence of Communism. Despite American control, a great number of southern people were strongly influenced by Communism. Moreover, women were as actively involved in the struggle against the Americans as were men. This period (1954-75) was characterized by ruralward migration in the north to escape American air attacks, and urbanward migration in the south to avoid the conflicts in rural areas. Therefore, in the north. this was a period in which the urban system developed with an increase in small cities and urban centers. Moreover, a million people from the Red River Delta resettled in the New Economic Zones located on the borders of the Democratic Republic of Vietnam in the 19605 (CPLS 1996). The urbanward trend of migration, particularly to Hanoi. was mainly organized by the government, which recruited labor for the state sector. This trend also was a result of the voluntary movement of people from rural areas who pursued higher education in urban areas, and later were recruited by the government. For the south, urbanization was referred to as "compulsory" by the southern government (ruled by Nguyen Van Thieu). And it resulted in a number of crowded cities (Sai Gon, Da Nang. Nha Trang, Buon Me Thuat). This increasing urban population was not a result of industrial or economic development. Rather, it was a product of the political strategy of America to force the rural population to move to cities in order to cut off the influence of communism in the southern rural areas. As a result, the urban population increased in the southern part from 20 percent in 1960 to 43 percent in 1975 (Desbarats 1987:46). Sai Gon became the most crowded city in the world with a population density of 34,000 persons per square kilometer (Trinh Duy Luan 1996. Desbarats 1987). Beginning in the 19605, the communist government initiated an organized migration process to develop New Economic Zones in the North. The major objectives of the organized migration programs were to reduce population pressure in the Red River Delta and to redistribute population and to claim natural resources (land) as well as to strengthen national defense and security for the mountainous regions (CPLS 1996, Trinh Khac Tham 1996, Desbarats 1987). Women and men were both subject to organized migration. Even though there are no specific data on female migrants, the total number of household migrants can help us to understand women‘s participation in organized migration during this period. Between 1960-75. an estimated. 210.000 people were moved annually to New Economic Zones. More specifically. between 1960-64, more than 600,000 people were moved from the Red River Delta to the Northern Upland regions (about 120,000 households) (CPLS 1996222). During 1965-75, the volume of organized migration was reduced due to the war against the US. The pattern of population redistribution aimed at developing New Economic Zones in Vietnam. and in other socialist countries such as the USSR and China, were implemented as a means of redistributing natural and human resources (Habitat 1996. MOLISA 1997a, 1997b, CPLS 1996). On the one hand, this organized migration helped the government rationally distribute the labor force within different regions. On the other hand, this population distribution tended to affect people‘s lives negatively in terms of their mental and material well-being. Organized migrants experienced a great challenge in a new place because although the government provided subsidies (which at face value may seem generous), in truth were inadequate because there was not enough infrastructure necessary for resettling, such as transportation, roads, or electricity. For example, each household received a food subsidy for the first 12 months of resettlement. In addition, they received certain amounts of money for resettlement: VN$ 3,110,000 (US$ 300) for a household that moved from the north to the south; VN$ 1.490.000 (US$ 150) for a household that moved within a province. At the same time, a buffalo cost VN$ 500,000 (USS 50), one simply-constructed house cost more than one million dong (US$ 100) (Hitchcox 19942209). In addition, migrants were able to get free healthcare services, free schooling, a partial subsidy for housing construction, or some cash to buy the means of production. They also were freed from the production tax (for the first five years) and the production duty7 (for the first three years) (C PLS 1996). Because the subsidy system co-existed with the household registration system, people had to stay in the place where they registered 50 that they could get food stamps and other social services such as healthcare, schooling, or retirement benefits (Habitat 1996, CPLS 1996). Therefore, this system limited people’s voluntary migration before innovation period (before 19805). People could only follow the govemment’s arrangement, otherwise, they could not survive. Women and the socialist system Since 1945, the socialist system in Vietnam has provided a series of legal reforms which gave women equal rights with men, e. g., the constitution 1946 (women given equal rights in pursuing education, in employment, earnings, and all social, political and economic activities), land reform policy in 1953 (women had the same rights to land as did men), marriage and family law in 1959 (custom of child marriage, absolute rights of parents and elderly in marriage, and bride payment practices were discouraged) (Mai Thi Tu et a1. 1978, Nguyen Huu Minh 1998, Nguyen The Giai 1988). Women’s active Participation in different socio-economic activities was encouraged by different policies that offered a 6-month-maternal leave and child-sick leave. Kindergartens and childcare 7 . , , This was a “voluntary” portion ofa harvest that farmers contributed to the government. 31 centers were freely provided until the 19905 in order to maximize women’s participation in and contribution to social production and management activities. With the implementation of these equal rights and the subsidy system, Vietnamese women gained a great improvement in their status at home and at work. After 1945, the Vietnamese socialist government placed great emphasis on education, completely subsidizing it from the lowest to the highest levels. With its great effort, the educational level of the Vietnamese population grew from 2 percent with elementary education in 1945 to 93 percent of the northern population aged 12-50 becoming literate in 1959 (Nguyen Quang Kinh 1994). The gap between women and men’s education was gradually narrowed. A5 Nguyen Huu Minh (1998) noticed, by 1990, the gender difference in the literacy rate was about 40 percent for people aged 60 and over, but it was about 18 percent for people ages 50-54, 5 percent for people ages 35-39, and almost no different for younger age groups. Women were encouraged to engage actively in different agricultural and non- agricultural occupations, especially during war-time. Their greater opportunities in obtaining non-agricultural occupations in the state sector helped strengthen the power and influence of women within their families and communities, and hence, helped loosen gender prejudice about women’s ability. In 1960, women comprised 20 percent of state employees in contrast to one-sixth in 1955, and 52 percent in 1989 (Nguyen Huu Minh 1998, Le Thi Nham Tuyet 1975, VPC 199lazl45). In social and political activities, women increasingly participated in different levels of management. For example, women deputies in the National Assembly cOmprised 2.5 percent in the First Legislature in 1946, 11.6 percent in the Second Legislature in 1961, and 21.8 percent in the Eighth Legislature in the late 19805 (Ngo Ba Thanh 1989, Nguyen Huu Minh 1998, Tran Thi Van Anh et a1. 1996:199). So, we can see that Vietnamese women have gained high educational levels. high prestige occupations and positions. and equal rights with men at work. However, this equality varied between urban and rural areas and among different groups of people with different educational levels or occupations. Mai Thi Tu et a1. (1978) and Nguyen Huu Minh (1998) maintain that people with a lower educational level, living in rural areas tended to be less affected by the gender equality ideal of the socialist system than those in urban areas, or those with a high educational level working in the state sector. The prevalence of gender ideology or patriarchal hierarchy within these groups was seen as an indication of the resistance of men to the gender equality that threatened their power and control over their households and women. These contradictory gender relations, in turn, strongly affected women’s ability and choices in the migration process. 4. Migration and the market economy (after 1975) 1. Migration characteristics and patterns Migration in this period had a close relationship to the urbanization pattern of Vietnam, which was marked firstly by a sharp reduction in the urban population (after 1975), and then, a gradual increase in the urban population in the whole country (since the 19805). These changes in the urban population can be seen in Table 1.1. For instance, shortly after national reunification (1975), population movement reversed, changing to ruralward in the south and urbanward in the north. Within five years (1975- 80), the urban population of the nation decreased from 21.5 percent in 1975 (GSO 1978215) to 18.6 percent in 1981 (see Table 1.1). This decrease resulted not only from 33 ‘\ ruralward population movement (both voluntary [seasonal] and organized migration) but also from a rapid growth of the rural population with a much higher rate of fertility (mean = 4.3 children) compared to that of the urban population (mean = 2.3 children) (VPC 1991aziv). This decrease also was a result of the govemment’s urban planning process, in which the urban system was rearranged, e. g.. many urban centers (industrial centers) were reclassified and became rural areas. Table 1.1: Urban - Rural Population of Vietnam, 1976-96 Year Total Urban Urban Rural Rural population population population population population (thousand (thousand (%) (thousand (%) people) people) people) 1976 49,160 10,127 20.6 39,033 79.4 1977 50,413 10,108 20.1 40.305 79.9 1978 51.421 10,130 19.7 41,291 80.3 1979 52,462 10,094 19.5 42,386 80.5 1980 53,722 10,301 19.1 43,421 80.9 1981 54,927 10,223 18.6 44,704 81.4 1982 56,170 10,626 18.9 45,544 81.1 1983 57,373 10,981 19.1 46,392 80.9 1984 58,653 1 1,203 19.1 47.450 80.9 1985 59,872 1 1,495 19.2 48,377 80.8 1986 61,109 11,817 19.3 49,292 80.7 1987 62,452 12,303 19.7 50,149 80.3 1988 63,727 12,682 19.9 51,045 80.1 1989 64,774 12,919 20.3 51,855 79.7 1990 66,233 13,281 20.4 52,952 79.6 1991 67,774 13,619 20.4 54,155 79.6 1992 69.4052 13,285 19.4 56,1205 80.6 1993 71,0256 13,663 19.5 57.3626 80.5 1994 72,509.5 14,139.3 19.9 58,3705 80.1 1995 73,962.4 14,575.4 20.0 59,387 80.0 1996 75,355.2 15,231.5 20.5 60,141.7 79.5 Source: GSO 198717, 1997az7. Moreover, this decrease resulted from a large-scale external movement of Vietnamese people in response to political and economic policies issued by the centrally-commanded socialist government. This outmigration included the movement of a number of students to obtain education and training in other socialist countries, particularly the USSR 34 (starting in the 19505 and becoming more frequent since the 19705).8 In addition, labor migration to socialist countries was organized based on the Labor Cooperation Agreement9 (signed in 1981 by the Vietnamese government and other socialist counties such as the USSR, East Germany. Bulgaria, Hungary, Poland. and Czechoslovakia). Many of these laborers are still working and living in these countries. In addition to these legal forms of migration, illegal outmigration also flourished. Wealth seekers and refugees fled to Western countries such as the US. Canada. and Australia, and there was a massive emigration of Vietnamese-Chinese due to the conflict between Vietnam and China in 1978-79. An estimated 35-37,000 people emigrated, and of this number, 26,000 were Vietnamese-Chinese (CPLS 1996:19, Hitchcox 1994). Nevertheless, during the transition period to the market economy, especially since 1986, changes in the socio-economic sphere have tended to increase population movement. Privatization of production and ownership and the removal of the subsidy and household registration system spurred population movement. Migration developed new features. Voluntary migration increasingly expanded compared to organized migration. Ruralward migration took place from less developed to more developed areas, and from lower to higher income areas. Urban-rural migration tended to immediately expand when personnel restructuring took place in the state sector of the economy. This restructuring aimed to increase productivity and to reduce labor costs by the rational use " 1n the 19605, the relationship between the USSR and Vietnam or China was in a crisis because the new political orientation of the USSR (ruled by Khorutsop) aimed to cooperate with the West and criticize Stalin (Revisionism). Therefore, Vietnam recalled its students back to the country until the 19705 when the USSR changed their orientation and the relationship between Vietnam and USSR was normalized. The two countries signed the agreement of the all-field-cooperation in 1978. Vietnam sent more students to the USSR to improve its population’s human capital. 9 This Agreement was based on the international corporation of SEV (Soviet Economitreskovo Sotrunhitrestva, i.e., committee of economic mutual aid, established in 1949). which aimed to redistribute labor force and technology to reduce the gaps of economic development within the socialist countries. 35 of labor, i.e., cut off unnecessary employees. Many low skilled workers were laid off and many people had no way to survive. A5 a result. they returned to their places of origin to engage in agricultural production. Among state employees. women were more likely than men to lose their jobs during the restructuring of the state companies because of their lower levels of education and work skills compared to men. Among those workers who were fired in 1992. 60.0 percent to 65.0 percent were women (Tran Thi Que 1995). As a result, many women who stayed in Hanoi were mainly involved in the informal (especially service) sector. Between 1976-95, migration organized by the government developed on a nation- wide scale. Approximately 4.8 million people were moved from the Red River Delta, North Central, and Central Coast, and Ho Chi Minh city. and were resettled in the Central Highland (Tay Nguyen). the Southeast, and the Mekong Delta. However, due to inadequate support for resettling, a sizable percentage (about 20 %) of organized migrants left the New Economic Zones to return to their places of origin or moved to other urban areas, such as Ho Chi Minh city, Vung Tau. Da Nang. and Hanoi (CPLS 1996, Hitchcox 19942210). With respect to migration patterns, besides organized migration aimed at establishing New Economic Zones, the govemment organized migration by recruiting workers, or transferring personnel working in the state sector, in order to redistribute human resources between regions. Between 1975-79, 31.7 percent of migrants (162,000/ 570,000 of the total migrants) to the South were transferred employees and highly educated employees, who had to satisfy a number of criteria defined by the government ——-—‘ (specified professional / technical qualifications) (MOLISA 1997a, CPLS 1996161, Trinh Khac Tham 1996). In 1993 a migration of young people was organized by the government to establish the mobile Youth Organization for specific economic tasks. such as to build the fundamental infrastructure for New Economic Zones. or to construct highways, roads or public service houses. It was made up of more than 80,000 young migrants (soldiers, workers) who were involved in construction, forestry, new land opening, mining, and irrigation for 32 provinces and cities (C PLS 1996. Trinh Khac Tham 1996). Information about voluntary migration was not systematically collected because the government did not focus on this type of migration. However. with the transition to the market economy, this kind of migration expanded as an individual or household strategy to rationally use labor. Voluntary population movement has taken a rural trend, reallocating labor within rural areas in the Deltas and mountains. This type of migration mainly resulted from regional differentials in population density and economic opportunities ( more land to do diverse agricultural activities). For example, people moved from the Red River Delta with a population density of 773 persons per square kilometer to the Central Highland (36 persons per square kilometer) (GSO 198725-6). The urban-urban trend had the lowest rate. while urbanward trends (rural-urban migration) rapidly expanded with the direction from North to Hanoi and to South, e.g.. from Nam Dinh, Hai Phong, Thai Binh, or other provinces in the north to Ha Noi, Ho Chi Minh City, and Vung Tau. The other direction was from southern provinces to Ho Chi Minh City, Vung Tau, Can Tho, and Bien Hoa (MOLISA 1997b, Trinh Khac Tham et al. 1998). In addition, highly educated people in rural areas tended to move to urban areas in the Red River Delta (which contains 50 % of the country's universities andicolleges, 25 % of the educational and professional training and 38 % of the professional skills training centers). and in the Southeast (which contains 20 0/0 of the country's universities. and 18 % of the educational and 15 % of the training centers) (C PLS 1996217). Between 1980—93. particularly from 1989-93. voluntary ruralward migration from 15 northern provinces (Northern Upland. Northern Central) had a direction to Dak Lak. Lam Dong and Gia Lai-Kon Tum (Central Highland), Be River and Dong Nai (Southeast) and Thuan Hai (Central Coast) (see Map I). This movement included approximately 1 13,000 households with 543,000 persons (C PLS 1996:50). Province Cao Bang had the highest rate of emigrants (19,715 households with 1 18,295 persons); the province Dak Lak had the highest rate of immigrants ( 1 9,5 78 households with 97,890 persons) (C PLS 1996253). The receiving areas were famous for their good land, low population density, and mild climate. Voluntary migration also includes the illegal international movement of Vietnamese-Chinese between 1978-80, and the movement of ethnic minorities to the west bordered by Lao, where primitive forests and land still exist. in order to exploit arable land. However, there is very limited information about these movements (C PLS 1996219, Trinh Khac Tham 1996). The voluntary urbanward migration mainly centered in Ho Chi Minh City, Ha Noi and Vung Tau. Since 1986, the Ha Noi population increased by 55,000 people per year, of which, 22,000 were a result of net migration. Three-fourths of the net migration was concentrated in the inner city of Ha Noi. The volume of iii-migration was five times 38 greater than the volume of out-migration. In addition, there were a number of temporary or seasonal migrants during the slack agricultural seasons. It is estimated that there were about 20 to 25.000 temporary migrants from Thanh Hoa, Ha Bac, Nam Ha, Hai Hung, Ha Tay. or Vinh Phu moving to Ha Noi (CPLS 1996, Trinh Khac Tham 199723). They played a remarkable role in the population dynamic of the city. In addition. Hanoi’s population has experienced uneven growth since the 19805. As can be seen in Table 1.2. it is clear that the Hanoi population decreased between 1986- 90, but increased in the 19905. Specifically. in 1986, the population of Hanoi was 2.987 thousand people (GSO 1987210). but in 1996, the Hanoi population included 2,395.9 thousand people. less than those in 1986 (HSO 1998:17). Table 1.2: Ha Noi Population Structure in 1986-96 Ha Noi 1986 1990 1995 1996 thousand "/0 thousand % thousand % thousand % Male 1,487.7 49.8 1,002.2 48.8 1,146.9 49.1 1,176.6 49.1 Female 1,499.8 50.2 1,049.7 51.2 1,188.5 50.9 1,219.3 50.9 Rural population 1,834.1 61.4 994.0 48.4 1.1 14.2 47.8 1.104 46.1 Urban population 1,153.4 38.6 1,057.9 51.6 1,221.2 52.2 1.291 .6 53.9 Total population 2,987 100 2,051.9 100 2,335.4 100 2,395.9 100 Source: GSO 198728.10 and H80 1998:17. The decline of the Hanoi population was a result of geographic mapping changes. That is, before the 19905, Hanoi was extended with the inclusion of some rural towns. which later were removed. Therefore, Hanoi’s population was larger in the 19805 than in the 19905. Even though these figures show the decline of the Hanoi population. this does not means that population movement into Hanoi was reduced. Hanoi’s natural population growth rates were reduced at 0.09 percent between 1990-96 (1.51 % in 1990 and 1.42 % in 1996) (HSO 1998:22) while its population growth rate was 2.3 percent between 1990- 96 (51.6 % in 1990 and 53.9 % in 1996) (see Table 1.2). This growth rate reflects net migration into Hanoi rather than urban birth rates. 2. Women and migration in the transition period In studies of internal migration patterns. female migrants were not the subject (Trinh Khac Tham 1998, Le Thi 1998. Dang Nguyen Anh 1998a). Rather, migration in general was the focus. However, general information on women in this period helps explain women's roles in the migration process, as well as patterns and characteristics of female migration. Vietnamese women comprised about 51 percent of the total population in 1987 (GSO 1987). Hanoi’s population had a similar sex ratio as the national one. From the data in Table 1.2, it is clear that women in Hanoi comprised 50.2 percent of Hanoi population in 1986, and about 51 percent in the 19905. The 1989 Census shows that the employment rate for women age 13 and above was 67 percent, and it was 72 percent for men; of Vietnam‘s employed population, 51.9 percent were women and 48.1 percent were men. Moreover, rural women participated in the labor force at a high rate, constituting 53 percent of total rural employment in 1989 (VPC 1991a). The fact that women actively participated in the labor force may imply that they also actively participate in labor migration for employment and earning, and this might be because the returns of migration were greater than non-migration. Despite the great effort of the government to eliminate gender inequality, gender differences still existed in earnings and occupations. and these differences tended to widen during the transition to a market economy. Women experienced traditional socialization patterns, and occupational sex segregation (Tran Thi Van Anh et al. 1996. 40 Hoang Thi Lich 1993). Their occupations were usually low-paid and less prestigious than were those of men. Gender occupational segregation was enforced through a predominant perception of suitable jobs for women and men (Le Thi Nham Tuyet 1998, Le Ngoc Hung 1996, Tran Thi Que 1995). For example, housework (childcare, preparing food), occupations in the service sector (nurses, schoolteachers, tailors) and light industry (producing consumer goods) were widely regarded as suitable jobs for women. Men's suitable jobs were in heavy industry. e.g.. electrical engineering, mechanical manufacturing. electronics and construction (producing the means of production such as equipment and materials). For example. in 1989 more than 80 percent of employees in mining, extraction and construction were men, whereas women comprised 86 percent of the workers in the textile industry, 72 percent in the garment industry, and 79 percent in trading (VPC 1991a, 1991b). In rural areas. female occupations included housework and many agricultural tasks such as planting and harvesting. Men were expected to deal with ploughing and watering, and to take care of labor exchange or work for others to earn additional income (Vu Manh Loi 1991). This division was based on the perception that these tasks were professional, while other agricultural tasks were not, and that some tasks showed visible value (measured by products or money) while others were too small to be valued. As a result, male occupations had a higher prestige and higher income compared to those of women. More specifically. women always earned less than men (mean = 118.7 thousand VN$ for women, 174.8 thousand VN$ for men); women in urban areas had a higher income (mean = 174.5 thousand VNS) than women in rural areas (mean = 67.8 thousand VNS) (5pc 1994). 41 Gender differences in occupation and earning might be based on the differences between women and men in education. Men had higher educational levels (mean = 9.8 years) than women (mean = 9.3 years) (VPC253). Further, more men were literacy (92.5 %) than were women (83.4 %) (VPC 1991a:55). Moreover, the educational differences existed not only between men and women but also between women in rural and urban areas. More women in urban areas were literate (91.0 %) than in rural areas (81.2 %) (VPC 1991a:55). Noticeably, even though women had gained a great degree of power in their households from the socialist system, gender differences still existed in both rural and urban areas, at home and at work due to the persistence of gender ideology in Vietnam as a patriarchal society. For example, men might share housework with their wives, or do the same housework as did women, but the decisive voice in making decisions belonged to men, not women. Women had equal opportunities in different fields as did men. For example. they might obtain as high education or high prestige occupations as did men, but they tended to be in charge of housework more than men. Housework roles limited women’s opportunities in improving their work skills because housework occupied women’s energy and time more than men’s (Le Ngoc Van 1997, Yarr 1995). Because of women’s closeness to housework, women were the ones. not men, who had to take off time from work to do housework if necessary (e.g., childcare, care for sick family members). Therefore, women experienced slower promotion at work than did men. Thus, women were always subject to discrimination. Nevertheless, the inferiority of women varied by age, marital status, and education (e. g., the degree of participation in decision making of the household). Older, married women tended to have greater power in their household than did younger single women; highly educated women tended to be more independent in making decisions than less educated women (Tran Thi Van Anh et al. 1996). Gender discrimination tended to be increasingly restored by market mechanisms (Le Thi Quy 1992) because the social security of gender equality collapsed, and free competition in the market economy tended to favor men. As a result, women gradually have returned to their housework, self-employment, or sex-segregated occupations since the 19805 (Le Thi 1995, 1998, Le Thi Quy 1998). Women contributed greatly to the household because they actively participated in the migration process in order to improve their households’ economic status (Le Thi 1998, Bui Viet Bao 1994, Doan Kim Thang 1997). Female migration definitely had different features from male migration (Doan Kim Thang 1997, Dang Nguyen Anh 1998c). Particularly, women tended to cluster in the informal (service) sector with low- paid and low-killed jobs (Bui Viet Bao 1994, Phan Quynh Nga 1998, Luu Phuong Thao 1996). They were more vulnerable than male migrants. They experienced the adaptation process differently from men (Le Thi 1995, 1998). Generally, gender differences in the migration process were results of gender ideology in the Vietnamese society. However, the nature and degree of this difference is unknown. Hence. this difference will be the focus of this study. SUMMARY Vietnam's migration has been shaped by the history of endless wars against different foreign invasions. Migration was impacted by the characteristics of feudal society, Chinese invasions, French colonialization, the American invasion, the socialist regime and the market economy. The most significant feature of Vietnamese migration is that until the French came, it assumed a rural-rural pattern throughout much of its history. Over the past decades, the collapse of the socialist system strongly affected current urbanization patterns and trends in population movement in contemporary Vietnam. Urban-rural migration became more frequent, the rate of the urban population increased, and the ruralward trend to some well-defined destinations such as Dak Lak, or the Mekong River Delta increasingly spread without government interference. Today, like men, women migrate more frequently, but they migrate primarily to urban areas while men tend to migrate to both rural and urban areas. Women contribute significantly to the urban labor market and economy as well as to their households’ economic improvement. However, Vietnamese women still experience gender discrimination at home and at work despite the fact that they have gained equal rights to men. and higher education or more prestige occupations than in the past. They always have a lower status than men. 44 CHAPTER 2: MIGRATION THEORIES AND LITERATURE REVIEW Introduction This study focuses on labor migration, which is associated with employment, income, and livelihood, rather than migration for political and environmental reasons (e.g., civil wars, natural disasters. famine). Migration in this study is referred to as a gendered process because women and men are involved differently in the migration process (Bilsborrow et al. 1992. 1993, Zlotnik 1993, Fawcett et al. 1984, Chant 1992a). This difference results from socially constructed gender relations, which determine who does what and shape men's and women's behaviors, choices, and opportunities. Migration is a process that contains the sub-processes of decision making, moving, and adaptation. The framework used to explain migration for this study heavily relies on theoretical approaches to labor migration in developing countries. Specifically, I use human and social capital theories of migration to explore how human and social capital differentially shapes these sub-processes for women and men under the influence of their demographic attributes (age, marital status) and financial capital (income). As Pugliese et al. (1998) and Douglas et a1. (1992) remind us. patterns of migration are related to the personal characteristics of migrants as well as to the availability of social networks and levels of human capital. The integration of migrants’ attributes with the human capital approach and the social capital perspective of migration will help clarify the characteristics, patterns and determinants of the male and female migration process. Capital consists of tangible and intangible assets that facilitate productive activity (Machlup 1982). Tangible or physical capital is embodied in an observable material form such as tools, machines, and other productive equipment. Intangible capital is embodied 45 in human capital (skills and knowledge) and in social capital (relations among persons) (Coleman 1988). Human capital is understood as all personal capabilities such as education, occupation. and other professional marketable attributes. Individuals have varying amounts of human capital that are embodied in their various abilities (Johnson 1995;). Social capital is understood as social relations / connections / networks that people develop and maintain in their daily life. These networks are a collection of relationships that connect people. social positions. or other units such as group and organizations. Within these relations/networks, individuals’ experiences, behaviors. and outcomes are shaped (Johnson 1995). In the migration process, people with different demographic characteristics (e.g., gender. age, marital status) use their human resources (human and social capital) differently to generate other resources such as income, wealth, or financial capital. Patterns of male and female migration will be uncovered by testing hypotheses derived from existing human and social capital theories of the migration process. Human capital theories emphasize the importance of personal attributes such as education, work skills, and occupations in relation to individual's characteristics (age and marital status) that are implicated in decision-making, moving, and resettling (Sjaastad 1962, Becker 1962, 1975, Farber 1983, Cole et al. 1985, Trzcinski et al. 1991, Robinson 1993, Sanders et al. 1996, Zlotnik 1995a, 1995b). Social capital theories emphasize the role of social relations (social networks and assistance), which impose their influence and help differently on different groups of people in the decision-making process, moving. and adaptation of migration (Coleman 1988. Boyd 1989, Douglas et al. 1992, Massey et al. 1993, Greeley 1997, Laurian et al. 1998, Portes et al. 2001). This chapter will explore 46 findings in the migration literature of developing countries and those of Vietnam in order to derive hypotheses for this study. The main focus of this study is gender differences in the migration process. specifically, the effects of demographic attributes (gender, age, marital status) and capital (social. human. and financial) of migrants on patterns of decision-making, moving. and resettling. Thus. the first part of this chapter includes a discussion of migration theories such as social and human capital theories and presents the integrative approach to gendered urbanward migration, and the model of this study. The second part includes a review of literature, from which the hypotheses of this study are drawn. I conclude by summarizing the hypotheses of this study. 1. Migration theories 1.1. Human and social capital theories of migration. 1.1.1. The economic theory of migration (Human capital model) focuses on regional differentials in economic opportunities as a source of migration. It assumes that people migrate to maximize their expected income and employment opportunities and to minimize the risks and constraints associated with labor market uncertainty (Todaro 1976, Lewis 1982, Lee 1966, Du Toit 1990, Massey et al. 1993). The human capital model emphasizes the importance of education and personal ability (e.g., work skills) as a means to gain a great return from migration. This model shows that ability and education are positively correlated with earnings. Migrants with more education and skills earn greater income than those with less education and fewer skills (Becker 1975, 1962). Human capital is referred to as skills and knowledge such as education and working skills or other marketable professional attributes (Sanders et al. 1996, Robinson 1993, Light et al. 2000). Migrants invest their human capital by seeking work elsewhere. 47 Migrants are those who have the greatest ability and who best meet labor market demands (Becker 1962, 1975, Sjaastad 1962). This selectivity tends to be applied to women as well as to men depending on labor market demands; these demands, indicators of preferred labor force create a selectivity based not only on human capital levels, but also on age (young) and marital status (single) (Massey et al. 1993). Younger, single migrants with higher educational levels will earn more, whereas older. married migrants with lower educational levels will earn less. Thus, potential migrants go to where the expected returns of migration are the greatest. i.e., they try to locate in a place where they can optimally use their human capital. The human capital approach helps explain the motivations and selectivity of migrants as well as the outcomes of the migration process, which are the most important features in understanding the nature of migration at the individual (micro) level. This model emphasizes the importance of education and work skills in the decision-making process (Laurian et al. 1998) and emphasizes economic reasons for migration over non- economic ones (Massey et al. 1993). Human capital theories reveal that a domestic unit’s socio-economic background facilitates or limits family members’ educational and occupational attainment (Cooke et al. 1996, Trzcinski et a1. 1991, Light et al. 2000, Gold 1992), i.e., it implies that the economic status of a household affects the human capital levels of its members. These educational and occupational attainments are argued to contribute to the degree of a persons’ participation in the decision-making process of migration. Family members with a high educational attainment are argued to affect the decision-making process more than those with low educational levels because of their 48 greater advantages and higher possibility of economic success by migrating (Pugliesi et al. 1998, Douglas et al. 1992). In the moving process, human capital theories reveal a positive relation between migrants’ educational levels and work skills and (1) the distance they move and (2) the degree of independence they exhibit in moving in search of the best match for their acquired human capital. People with high educational levels and work skills appear to be responding to a broad set of factors. implying that they are more capable of considering alternatives than are those with low educational levels and work skills (Shihadeh 1991, Shah et al. 1984, Khoo et a1. 1994, Hou et al. 1994, Chattopadhyay 1998). Consequently, a positive relation between human capital level and economic success in the adaptation process of migration is explained by human capital theories by emphasizing the opportunities and possibilities of economic upward mobility of people with high educational levels and work skills (Gold 1992, Portes et a1. 1996, Foner et al. 2000). These people are able to prevail over adversity and attain higher levels of well- being than are their less advantaged counterparts (Massey et al. 1993, Sander et al. 1996, Hunt et a1. 1985, Robinson 1993, Vecoli 1998, Sowell 1996, Light et al. 2000, Portes et al. 1996). Moreover. the human capital approach indicates the influence of economic success on the satisfaction of migrants with the outcomes of their movement (William et al. 1986, Laurian et al. 1998. Huq-Hussian 1995, Trzcinski et al. 1991, Rodenburg 1993). Economic achievement determines satisfaction because the main motivation for labor migration is economic. In other words, migrants with greater economic achievements are more satisfied with migration than are those with less economic achievement, and there is 49 no gender difference in this model. i.e., men and women with the same human capital can equally achieve economic improvement and are equally satisfied with this achievement. Since this approach pays attention only to the economic aspect of migration and individual attributes (human capital levels. age, marital status, and economic status). ignoring gender differences (e.g., female / male differences in education and opportunities). it cannot explain migration as a gendered process. Moreover, the human capital approach does not explain the fact that migrants are not always well-selected but may still succeed in the migration process. and that destinations with the best economic opportunities are not always chosen by migrants; this approach also cannot explain the heterogeneity of migrants involved in moving and the outcomes of their migration. That is, people with higher human capital levels might not always achieve greater improvement by moving than those with lower human capital levels; and people with higher human capital levels might not be as satisfied with migration as those with lower human capital levels. Hence, human capital is very important but not the single decisive factor in the study of migration. To uncover the nature of male and female migration it is necessary to integrate the human capital theory with other theories of migration such as social capital, in addition to demographic and financial indicators. 1.1.2. Social capital (network) theory emphasizes the importance of social networks to gain access to employment and high income in the migration process (Gallin 1973, Repak 1994, O’Connor 1990, Sander et al.1996. Portes 1993, Massey et al. 1993, 1994, Portes et al. 1996, 2001). Social capital has been described in a variety of ways including access to jobs, housing, and financial assistance, and job information provided by network members (Massey et al. 1988, Taylor 1986. Mullan et al. 1998, Portes et al. 50 2001). It is also considered a process by which established human relations (community- based. friendship-based, family-based, ethnicity-based) act as a resource for economic cooperation (Gallin 1973. Hugo 1981. Coleman 1988. Rumbaut 1997, Light et al. 2000). Social capital is embedded in social relations at the macro and micro levels. Social relations form the basis of social networks. in which roles. responsibilities. expectations. choices. and opportunities of the social networks’ members are shaped. Thus. social capital is referred to as the capacity of individuals to command scarce resources (employment opportunities) through their membership in networks or broader social structures (Portes 1998, Portes et al. 1996, Hugo 1981). Networks represent individual and family interests and concerns differently depending on their members’ demographic characteristics (e.g., gender, age, marital status), human capital (educational levels or work skills), and economic status (e.g., income levels). That is, young single women and men with higher educational levels and social networks at destination might have a different voice in making the decision to move or they may have a different moving pattern than do married people and those with limited education and connections. Social capital theories, which focus on social networks, help us understand the arena in which social life occurs; they. therefore, provide a model of society and daily life (Greeley 1997). The other important contribution of social capital theories is embedded in the gendered aspects of social relations, i.e., the social capital of migrants cannot be fully understood without considering their personal characteristics (e.g., gender. age. marital status, education, income). Gender ideology defines different roles, duties and positions for women and men in the household and society; hence. women and men are involved in social relations differently in terms of a power hierarchy (subordinate or dominant). Accordingly. there is a clear perception of suitable male/female work, behavior, ability or position. based on which social networks channel their members into their “suitable” environment. Social networks / relations shape migrant's choices and opportunities based on this socially-constructed gender perception and the obligations of kinship, friendship and membership in a shared-community (Douglas et al. 1992, Chant 1992a, Cook et al. 1996, Robinson 1993). The nature of social networks also shapes the direction of male and female migration in particular contexts. That is, people tend to migrate to where they have kinship networks rather than non-kinship networks. This notion again shows that human capital alone is not enough to explain the complicated migration process. People simply cannot go to wherever economic opportunities are greatest; they need to consider social factors that might limit their capacity by relying on their social networks. which are different for people with different backgrounds. Specifically, network size and nature are influenced by individual characteristics, family background and composition, and socio-economic conditions at the community and society levels. As noted in the migration literature. migrants with higher educational levels tend to have more non-kin relations than those with lower educational levels; older people are likely to have broader social relations than younger people; people with more non-kinship relations tend to have more economic opportunities than those with kinship networks at destination (Pugliesi et al. 1998, Moore 1990). The expansion of social networks is directed to reduce the costs and risks of migration. Each act of migration creates the social structure needed to sustain it. Hence, migration becomes a self-perpetuating process with a continuous expansion of social 'JI IQ networks (Massey et al. 1993. 1994, Repak 1994, Fawcett 1989, Holmes-Eber 1997, Wilson 1998). Noticeably, social networks play a major role in the decision-making, moving and adaptation processes by providing migrants information and assistance which reduces the uncertainty of migration. In the decision-making process, social networks, along with gender ideology embedded in social relations, impose a powerful influence over the selection of who migrates and why. Members of these networks provide information about economic or non-economic opportunities that help potential migrants evaluate the situation at the micro level and assign priorities (Douglas et al. 1992. Wilson 1994). Social networks carry the potential to increase the economic benefits of wage labor and hence, raise the incentives for migration. Consequently. social networks strongly predict the likelihood of migration. Thus, the nature of networks shapes the direction of moving, reasons for migration, and degree of participation in the decision-making process as well as the selection criteria for migration in particular contexts. As mentioned earlier. social capital is embedded in social relations that shape people’s responsibilities, obligations, choices, opportunities as well as roles; hence, these relations help us to understand heterogeneity in terms of power relations involved in the decision-making process. Kinship networks are hierarchical, and members have different roles and positions within them, depending on the gender division of labor and on kin obligations and responsibilities toward each other (De Jong et a1. 1998, Tyner 1994, Chant 1992a, Khoo et al. 1984). In this hierarchy, kinship networks, age, stage in the life cycle (marital status), and gender are the most important factors determining the degree of participation in the decision-making process (Riley et al. 1993, Hugo 1995). These 53 determinants have their roots in gender and patriarchal ideology, which assumes that men are superior, women are subordinate, and that women only gain power when they become mothers or grandmothers. Relationships between family members are shaped according to their personal characteristics (gender. age. marital status); thus, women’s participation in making decisions is limited in accord with their housekeeper role. In this hierarchy. single women have the lowest status and the weakest voice in decision making compared to men and married women (Thadani et al. 1984). The decision-makers are usually male heads of households or older men (Hugo 1995). Put another way, the social capital of migrants shapes their choices and opportunities. which are differentiated according to migrants’ characteristics (age, marital status, gender). In the moving process, social capital theories emphasize the necessity of social connections or relations in shaping patterns of movement (individual, associational) and distance of move. The nature of these relations will determine who moves alone or with others and how far she or he moves. These decisions are based on stereotypes of women and men’s roles, responsibilities, and position and, thus, migration patterns are different for women and men (Douglas et al. 1992, Prothero et al. 1985, Chapman et al. 1985, Lee 1966). Type of migration has a close relationship with the different roles women and men play. i.e., subordinate or breadwinners in the household (Thadani et al. 1984, Lim 1993). For instance, due to their subordinate roles and limited human capital, women are usually involved in associational migration (they migrate with family members or relatives) while men move alone with their social networks channeling them into more advantaged spaces in the labor market than they do women (Kofman 1999, Zlotnik 1993, 1995a. 1995b, Fawcett et al. 1984, UN Secretariat 1993. Bilsborrow et al. 1992). 54 Moreover, migration distance is closely related with the availability of social connections at destination (Kontuly et al. 1995, Guest 1993, Collins 1983), and the availability of social networks at destination helps reduce the selectivity of migrants. As noted by the human capital approach, people with higher educational levels and work skills migrate a longer distance than those with lower levels but. with the assistance of social networks. these people also might be able to move as far as those with higher educational levels and work skills. In addition. women move shorter distances than men due to their subordinate positions, their reproductive roles, and their lower human capital levels. Clear gender stereotypes, in addition to age and marital status differences, determine the pattern of movement for women and men. Obviously, social capital theories tend to complement human capital theories in explaining the migration of people with different characteristics and backgrounds. In the adaptation process, social networks provide assistance (housing, information, money and emotional supports) in order to help migrants successfully resettle (Gallin 1973, Gold 1992, Portes et al. 1996, 2001). Resettlement is a gendered process since male and female migrants enter new labor markets and urban life within culturally specific socially-constructed gender relations. i.e., gender-based social networks. These relations or networks determine women and men's jobs and opportunities, or women and men's duties and roles at the household and societal levels (Repak 1994, Zlotnik 1995a, 1995b, Lichter 1983, Bilsborrow et al. 1992). This means that social capital channels people into their presumed roles and positions based on their gender. This process also has an ethnic dimension since social networks tend to cluster migrants based on kinship, friendship or shared-community origins by providing limited 55 employment information about certain occupations (Boyd 1989, Tyner 1994, Portes 1993. Gold 1992). Thus. migrants tend to be centered selectively in some occupations. F urthemiore. social networks directed to reduce costs and risks for migrants will increase the likelihood of migration by providing community members employment sources. for instance (Gold 1992. Foner et a1 2000. Portes et al. 2001. Rumbaut 1997). Hence. migration expands over time while connections between origin and destination diffuse widely so that migration becomes much easier for later movers. The magnitude of migration depends more on network size at destination than on wage differentials or employment rates. Migration becomes less selective with the assistance of social networks (Massey et al. 1993. 1994. Mullan et al. 1998). In sum. social networks shape adaptation outcomes of migration. Not all social networks are equally effective, however. Human capital affects the number and composition of social relations and social networks increase the level of human capital by assisting in the learning of skills (Techman et al. 1997, Wilson 1998, Pugliesi et a1. 1998, Light et al. 2000). People with higher educational levels tend to have a greater number of and more diverse social relations than do those with lesser education. This means educational levels provide opportunities for people to enlarge their non-kin relations with people of high social positions while people with low educational levels or who are illiterate tend to be limited within kin or non-kin relations with people who occupy low social positions. To summarize, social capital theory emphasizes the importance of social networks, which help to reduce the costs and risks of moving by way of “maintaining their links to origin, influencing the selection of migrants. destination and origin, 56 conditioning the integration of migrants in destination. serving as channels for information, other resources and normative structures and shaping the size and momentum of migration" (Douglas et al. 1992:153). Moreover. social capital theories are not as gender-blind as is the human capital perspective. They emphasize the gender aspects of the nature of social networks as these connections relate to personal ties, and cultural and normative principles that are embedded in socio-economic structures at different levels of the nation, region and household. From the foregoing discussion. it is clear that the migration patterns vary according to migrants’ demographic indicators, economic status, and human and social capital. And thus, the integration of these approaches will help explain migration as a complicated and gendered process. 1.2.Integrative approach to gendered urbanward migration The model of this study draws on the central tenet of human capital theory that the purpose of migration is income maximization and that migration takes place when regional differentials are present in labor supply and demand and in economic opportunities. In addition, the model draws from the human capital approach the notion of the importance of personal marketable abilities (education, experience, age. and marital status) and the positive correlation of levels of education with economic improvement in the migration process. Social capital theories also have contributed to this model. This perspective emphasizes social networks as a means to support the migration process; social networks provide access to economic opportunities for their members. Social relations direct migration based on the nature of obligations of kinship, friendship, and membership. Social networks carry gender dimensions in migration. 57 Social networks are especially important for female migrants to succeed at destination, given the persistence of socially-constructed gender relations at home and at work. Moreover. the human capital and social capital of different people are not the same, as discussed above; they vary depending on people’s economic status (e.g., income levels or financial capital) or demographic characteristics (i.e., marital status, age, gender). Therefore. this model also includes demographic attributes and financial capital to explain migration as a gendered process. The integration of these different insights will be used to explore the nature of male and female migration processes in terms of decision making. moving and adaptation. The human capital approach will help us understand the motivation and selectivity as well as outcomes of male and female migration; the social capital approach will help explain the way people are involved in the migration process as well as the way they overcome the risks and uncertainty of migration; the inclusion of demographic attributes and financial capital in the study model will enhance understanding about the diversity of the migration patterns of different groups of people. An integrative view is best presented in a gender model (see Figure la). This mOdel is based on the assumption that male and female migration is usually shaped by SOCio- economic conditions at both origin and destination (Brown et al. 1977, Agostinelli 1991, Kobayashi 1995, Nabi 1984, Harris et at. 1970, Robert 1989, De Jong et al. 1981. Gfeendwood 1985, UN 1993, F awcett et al. 1984). Under these conditions women and men migrants have different choices and opportunities in the migration process (decision- making, moving and adaptation processes) relative to their human capital. social capital, -..;— Bfiocm :o @338.» 25 mo Soto cm 886E AAl v $583 £03353 2: 50.33 20:22 30% A - - - V 8:: @826 ”802 4 3305 5:932 6:98 Boom l 3E3 RESET.— . \ 3:93 SE53 8:538 0Ea2w0800 $888 28288 -oCom Ace—53> E .833me Eat—Eng: he .25:— Bucow é; 95w?— 59 financial capital, and personal characteristics. This model emphasizes that female migration differs from male migration in terms of its selectivity, patterns of migration, and adaptation outcomes. 1.3.Exp1anation of the model This framework explains the individual migration process within the broad socio- economic context of Vietnam during the late 19805 and 19905. At the national level, there have been important policy changes that facilitate market mechanisms and population movement, which, in turn, affect household survival strategies and individual economic opportunities, i.e., changes at the macro level lead to changes at the micro level. These macro changes include abolition of subsidy systems in the economy, the abolition of the household registration system in rural and urban areas, and the shift from a command economic system to a system based on market principles. In addition, privatization of production or enterprises, decentralization of decision making, liberalization of trade, prices and the exchange rate strongly impact people, institutions, and management. In rural areas, decollectivization, privatization of production, redistribution of farmland, new loan policies,' the household contract system, and new land law positively affect productivity while increasing the surplus of labor. As a result, peOple are encouraged to search for employment in other places. In urban areas, many of the above listed changes open up a diversity of urban industries and the development of urban subsistence economies with primary emphasis on the service sector. This sector is fueled by low-cost ' These policies allow farmers to borrow a certain amount of money to invest in production during a certain period of time with their properties used as a guarantee for the loan. These policies did not exist before the innovation period. 60 labor and low-skilled workers from rural areas. At the regional level, a great differentiation in earnings, labor supply and demand, and economic opportunities results from regional geographical features and access to markets and major transportation lines. This differentiation also results from differing population densities, regional production strategies. and regional development programs. The differences in economic opportunities enlarge interregional connections (long distance trade, seasonal population movement), and. thereby, encourage the household to rationally use its labor. Due to the removal of restrictions on household registration and on private trade and transportation, social networks are rapidly expanding as people search for economic opportunities in other areas such as urban areas. Therefore, migration is increasing within Vietnam. Under the influence of socio-economic conditions and changes at the macro level, motivations and decisions to migrate are fostered within a household context. Depending on regional differences in economic opportunities, household economic status and strategies, individual characteristics (age, gender, marital status). and the availability of human capital, financial capital, and social capital, Vietnamese people participate in population movement differently. People are involved in migration for different reasons, engage in different types of movement (individual or associational), and, they make different types of decisions to move (individual or household), which. in turn. affect adaptation outcomes at destination. In sum, this model demonstrates the mutual relationship between the macro and micro levels and male and female migration through the interaction of human. financial, and social capital and the demographic attributes of the individual. 61 2. Literature review and guiding hypotheses Below I provide the general hypotheses of the study as they are derived from the literature. The operationalization of variables is described in detail in Chapter 3. 2.1. The decision-making process of male and female migration The decision-making process is referred to as a process in which people make decisions about who will migrate and why to migrate. It is widely accepted that migration is a selective process in which certain people are involved in spatial movement for economic betterment (Lee 1966, Todaro 1976. Brown et al. 1977. Lewis 1982. Du Toit 1990, Massey et al. 1993. Cohen 1996). 2.1 .1 . Reasons for migration In general. people have different reasons to be involved in migration, but most often they migrate for economic reasons, as the migration literature reveals. Migration is a way to maximize economic opportunities. People migrate to pursue better possibilities of employment and earnings than are available at origin (Todaro 1976, Cohen 1996, Du Toit 1990. Massey et al. 1993, Lee 1966, Lewis 1982. UN 1994. Yang 1993). The evidence shows that people usually migrate to improve their economic status directly rather than indirectly (Bilsborrow et al. 1992. Trzcinski et al. 1991). This means that people are involved in migration in order to maximize their expected income and employment opportunities rather than to invest in improving their educational level. Many different studies found that male and female migrants share some similarities in terms of reasons to migrate. Employment and economic factors tend to be the main determinants of male and female movement in different countries in Asia and Africa (Chant 1992a, 1992b, Pryer 1992. Hugo 1991, 1993, Brydon 1992, Thadani et al. 1984, UN 1994, Winchie et al. 1998. Singh 1998, Drachman et al. 1996, Wallimann 1994, Adamchak 1987). Many issues affect women and men similarly. For example, household strategies of survival or mobility tend to impact both women and men in a similar way (Morrison et al. 1988) by shaping their motivations for migration. The motivations of labor migration for both women and men are usually related to employment and earning opportunities. a result of the fact that not only men, but women as well. have economic responsibilities -- even if women’s are unacknowledged by society (Bilsborrow et al. 1993, De Jong et al. 1996). This household strategy directs men or women into a migration stream based on their productive and reproductive labor in the household and the demand of labor markets (Chant 1992a, Hugo 1993, Cooke et a1. 1996). As a result, even though economic motivations of migration are dominant in both female and male migration, men tend to migrate for economic reasons more than do their female counterparts while women tend to migrate for family reasons more than do men. The findings of those few migration studies in Vietnam confirm the common pattern of internal migration found in other developing countries. People most often migrate for employment and higher earnings, or for economic betterment (Dang et al. 1997. Trinh Khac Tham 1996. Doan et al. 1996, 1995, Tana 1996. Nguyen Van Tai 1998, Bach Van Bay 1996, Phan Quynh Nga 1998, Bui Viet Bao 1994, Nguyen Van Chinh 1996). Based on the similarity of Vietnam’s socio-economic context with that of other developing countries (e.g., patriarchal ideology, overpopulation, high levels of poverty, and a labor surplus), even though both women and men, move for economic reasons, I hypothesize that men move for economic reasons more than women while women move for family reasons more than men (Hypothesis 5.1.a). 63 Thadani et al. (1984). studying urbanward migration of women in Asia, found that regardless of migrants’ marital status, women and men tend to move to seek the possibility of better employment. Economic motive is the principle force in both female and male migration. In contrast, many studies on migration in developing countries in Asia such as Thailand, Indonesia, Korea. Bangladesh. and China, or in Africa such as Kenya, and in Latin America such as Costa Rica, Peru, and Caribbean. found that there is a difference between migrants with respect to their marital status in reasons to migrate (Fawcett et al. 1984, Chant 1992a). It is reported that, for example, single male and female migrants manifest almost the same pattern of migration. They tend to move for economic reasons (employment) or for further education rather than for family reasons (Rhoda 1980, Brydon 1992). Married women, in contrast, as found in other studies, mainly migrate for family reasons, family reunion, or their husbands’ careers (Chant 1992a, Cooke et al. 1996, Morrison et al. 1988, Boney et al. 1991, Robinson 1993). Married women voluntarily sacrifice their careers for that of their husbands because they expect that a husband’s income and status are highly valued and, hence, the returns from male migration will be higher than those from female migration. This expectation results from the perception that men are superior in the household and they, rather than women, deserve to be given priority to fulfill their roles in society. This close connection between married women and family reasons for migrating reflect their greater constraints in labor markets compared to men (Boyd 1989, Morrison et a1. 1988), their attachment to housework that is assigned to them based on gender ideology (Cooke et al. 1996, Payner et al. 1992). or both. Married women have greater responsibilities toward their family and more 64 involvement in household management than do single women due to their roles as mothers or wives. Implicitly. we can see that social relations affect reasons to move. There are only a few studies on Vietnamese female migration (Le Thi 1998, Phan Quynh Nga 1998) and they provide no information about the main motivations for female compared to male migration in Vietnam. Moreover. there are not many analyses addressing the differences among migrants in terms oftheir marital status (Li 1996. Nguyen Van Tai 1998). We can infer, however, that married and single Vietnamese migrants experience migration differently because Vietnam shares similarities with other patriarchal societies and is experiencing the globalization ofits economy as are other developing countries. Economic restructuring tends to impose a similar pattern on the Vietnamese labor market that provides more economic opportunities for single over married women. Like other developing countries under the influence of patriarchy, age and marital status are avenues via which Vietnamese women gain power and position in their households and society. Thus. married women have more power in the household than do single ones. They are more closely engaged in the household’s management and existence than are single women. Therefore, they move for family rather than for economic reasons while single women tend to move mainly for economic reasons. Single women move for economic reasons because they are a preferred labor force compared to married women in a restructuring economy. Based upon the foregoing discussion, I hypothesize that in Vietnam, the marital status of migrants has an influence on reasons for migration. Specifically. single women and men (both single and married) migrate for economic reasons while married women migrate for family reasons (Hypothesis 5.1.b). 65 2.1.2 Making the decision to move Many studies reveal that the decision of migrants to leave their place of origin is determined by an overall household strategy to improve the unit’s economic status (Thadani et al. 1984, Chapin 1989, Spitze 1986). The decision-making process of migration involves a different degree of participation by women and men who play different roles in the household. These differences result from an individual’s position in the family in terms of gender. stage in the life cycle. age. the household division of labor. educational attainment, and their social relations at destination (Chapin 1989, Erman 1998, Sandu et al. 1996. De Jong et al. 1996, Hugo 1995, Riley et al. 1993). Women. in general. are less involved in making decisions than are men. They have less voice in the process and their movement tends to be determined by family members while men are independent in making the decision to migrate (De Jong et al. 1981, Erman 1998). Due to traditional roles and the gender division of labor in the household, women’s choices are more likely than those of men to be shaped by a family’s strategy to maximize the economic returns of migration. Thus, making decisions in the migration process is shaped by factors at both the macro and micro levels, in which women are marginalized and directed by their households according to the requirements of labor markets, while men tend to be self-determining because of their traditionally presumed advantages over women as well as their dominant position in the household. a result ofa patriarchal hierarchy (Boyd 1989, De Jong et al. 1983, Lim 1993. Arnold 1987). According to Sandell (1977), migration decisions are based on labor-market- related personal characteristics of laborers, most usually education and work experience. 66 and employers’ preference for certain age groups (young) as employees. These characteristics tend to advantage men over women in the decision-making process. Generally, reports on Vietnamese population resettlement by the MOLISA or the Department of Population Relocation, Ministry of Agriculture, mention population in general. and state that organized migration is directed by the government in combination with household decisions (CPLS 1996). Tana (1996), focusing on temporary migration into Hanoi. also notes that labor movement in Vietnam tends to be directed by a household’s strategy in response to rapid changes during the transition period. Faced with a great labor surplus, rural women and men migrate to Hanoi and they are directed to do so by family members. Other studies found that women and men are directed to migrate in order to support family members (their own children or siblings and parents) (Le Thi 1998. Trinh Khac Tham et al. 1998). Individual migration, therefore, is directed by the household. and the degree of this direction is much stronger for women migrants than for men migrants. Based on similarities among patriarchal societies and the findings from the migration literature in other developing countries. as mentioned above, I expect that there is a gender difference in making decisions. That is, men are self-determined movers while women are directed to migrate by family members (Hypothesis 5.2.a). Decision-making models usually emphasize the importance of age and stage in the life cycle of people in making decisions (De Jong et al. 1983, William et al. 1986. Donato 1993, Hugo 1995, Robinson 1993. Riley et al. 1993, Lingam 1998). This difference results from the social relations between and obligations of family members toward each other. Many studies reveal that wives most often defer to their husbands in the decision to move, especially when the husband’s moving is related to employment 67 reasons (Shihadeh 1991, Boney et al. 1991, McDevitt et al. 1986). Other studies reveal that single women (daughters) are usually directed to migrate by their parents as a result of a household strategy of survival (to support family or siblings’ schooling) (Trager 1994, Harbison 1981. Lauby et al. 1988, Khoo et al. 1994). Yet other studies indicate that young people are usually directed to migrate by family members because of their lesser responsibilities and power in the household (De Jong et al. 1981, Harbison 1981). As for the Vietnamese case, there is no study on this issue. Based on the similarities among patriarchal societies. I expect that young single people tend to be directed to move by family members. Nevertheless, based on the socialist system in which women gain more rights with men than women in other countries. I also expect that married women in Vietnam tend to have more voice in making decisions than do their counterparts in other countries. i.e., they tend to join in making the decision to migrate with their husbands, but not to be directed passively in this process. Thus, young women and men are directed to migrate by family members (Hypothesis 5.2.b); and I expect that single women and men are directed to migrate by family members while married men are self-determined movers and married women join with their husbands in making the decision to migrate (Hypothesis 5.2.c). As noted in the migration literature, the gender roles of migrants in the household and the community are closely tied to their decision-making power. In most societies, men are considered breadwinners and women housewives. This means that men have a greater voice in making decisions while women’s activities are directed by family members (Boyd 1989, De Jong et al. 1983, Williams et al. 1986, Lim 1993. Donato 1993. Hugo 1995, Robinson 1993, Riley et al.1993). It is found. moreover. that women and 68 men’s degree of participation in making decisions is closely tied to their specific social roles in the community (Wilk 1989, Riley et al. 1993). Implicitly. these roles might be a result of the human capital levels of people, in addition to their gender roles. That is, those with higher educational levels tend to have more power in making decisions than people with lower educational levels. Both women and men with “safer” characteristics (well-educated and skilled) can be independent in making the decision to migrate while the movement of those with lower levels of human capital are determined by a household. Vietnamese people might also experience a similar effect of their social roles in the community on their degree of participation in making decisions in the household, and this might be because women’s active participation in social activities in the socialist system has been increased as they gained more equal right to men than in the past. Furthermore, since most often Vietnamese people migrate for economic betterment, calculating based on a positive relation between human capital and migration outcomes, human capital levels might play an important role in giving people a voice in making the decision to move. Based on this discussion. I expect that both women and men with higher educational levels are self-determined movers while those with lower educational levels are directed to move by family members (Hypothesis 5.2.d). 2.2. Moving process The moving process is referred to as the way people migrate. including who moves alone or with others and how far they move. 2.2.a. Type of move The migration literature indicates that in developing countries type of migration shares common features (Chant 1992a, Khoo et al. 1984, Thadani et a1. 1984). More 69 specifically, female patterns of migrating generally differ from those of males. Women are involved in migrating as tied movers rather than as independents as are men. Women usually move with family members. children or relatives, while men move solitarily or with friends (Hugo 1995. Morrison et a1. 1988. Guest 1993). On the one hand, women’s reproductive and housework roles make them tied movers. On the other hand, according to gender ideology or patriarchal perception as presented in Chant (1992a), Khoo et al. (1984), or Thadani et al. (1984). women are considered to be inferior to men; they are seen to have no capability to move alone but rather have to rely on men’s support. Within Vietnam, the most common pattern of migration is the male pioneer (Dang et al. 1997, Tana 1996, Trinh Khac Tham 1996). since migration is closely related to economic reasons or labor migration. Men move first because they have more advantages than women in the labor market. They have more chance to be employed in the state sector of well-paid occupations than women due to the persistence of gender ideology at work (Le Thi 1998, Nguyen Quoi 1996). This male pioneer movement is considered a household’s strategy to reduce the risk of moving and maximize the returns ofmigration (Dang Nguyen Anh 1998c, Trinh Khac Tham et al.1998). From this discussion, I hypothesize that men move alone while women move with others or associationally (Hypothesis 6.1.a). The migration literature reveals that marital status plays an important role in type of female migration. Single women tend to migrate alone while married women are associational movers (Hugo 1993, Cooke et al. 1996. Robinson 1993). This difference is a result of the availability of employment for single but not for married women with the expansion of the industrial sector in cities (Sexton 1996). Single women tend to have 70 higher possibilities for employment in urban labor markets than married women for several reasons. Because women earn less than men do. employers tend to use women’s labor. especially that of young single women. to increase the profits of production. Moreover, female labor is particularly attractive to employers because of their presumed vulnerability and feminine qualities. such as docility. manual dexterity, deference to authority. and willingness to withstand rigid discipline (UN 1986. Lim 1993, Guest 1993. Orig 1987). There are some reports on the number of single people who moved to Hanoi or other cities in Vietnam to pursue higher education and on the number of highly educated people who were hired in cities in Vietnam (CPLS 1996, Trinh Khac Tham 1996). These reports implicitly discuss the relationship between marital status and type of migration. Based on the historical and socio-economic context of Vietnamese society that is described in these reports, marital status affects type of migration in Vietnam. Single people might migrate alone if they have a job waiting for them at destination, if they seek occupations suitable to their human capital, or if they want to pursue higher education, which is only available in industrial areas or cities. This was the case for Vietnamese rural people with higher education and work skills under the socialist system. In the 19905, during the transition period to a market economy, this feature still remains, i.e.. single people migrate alone to pursue employment in the state sector and higher education. However, even though Vietnamese women have gained higher educational levels than in the past as a result of socialism and their economic contributions are noticeable, they are still reported to be dependent in the moving process like women in other developing countries (e.g., China, Thailand, Malaysia); they move with family. This is especially true for married women 71 (Trinh Khac Tham et al.1998. Nguyen Quang Hue 1998). This phenomenon is explained by the persistence of traditional gender ideology that shapes women’s roles and status at home and at work as well as the traditional perception about women’s opportunities and abilities to achieve a successful move. Given this contradictory situation, I expect that men and single women move solitarily while married women move with others (non- solitarily) (Hypothesis 6.1.b). As noted in the migration literature, social networks shape type of move; men, regardless of social relations. rarely move associationally, while women are usually involved in associational migration, especially married women with social networks (UN Secretariat 1993). Nevertheless, the studies of Piampiti (1984) and Arnold et al. (1984) found that women tend to move alone if they have social networks at destination as a secure source of assistance. The study of Nguyen Van Tai (1998) on migration to Ho Chi Minh City or that of Than Van Lien (1997) to Hanoi and Hue found that people tend to move both alone or with their families when they have social connections at destination. These studies confirm the importance of social networks in moving patterns, however, it is not clear about the pattern of movement for those without social connections. Therefore, based on findings from the migration literature, I expect that the Vietnamese migration pattern also shares the same feature. That is, both men and women with social connections move alone (Hypothesis 6.1.c). Furthermore, findings in the migration literature reveal that older people with higher educational levels have more non-kin relations that younger people with lower educational levels, and that men tend to have more non-kin relations than women (UN Secretariat 1993, Fawcett et al. 1984, Harbison 1981). which allow men to move 72 independently. In addition. traditionally, the source of ultimate assistance for Vietnamese people in dealing with an important event such as changing their living place mainly came from kin who are imbued with a high sense of obligation and responsibility toward relatives (Tran Dinh Huu 1996, Phan Ke Binh 1996): these kin relations are very important for women to overcome their disadvantages: hence. they tend to cluster with kin-relations and to be involved in associational movement. Based on these discussions, I expect that women’s migration is more strongly shaped by kin-relations than that of men. That is. men, regardless of type of relations at destination, move alone while women with kin relations move alone and those with non-kin relations move with others (Hypothesis 6.1.d). 2.2.b. Distance ofmove Studies of migration in developing countries such as Thailand, Malaysia, Costa Rica, and Indonesia reveal that women move shorter distances than men (Hugo 1993, 1995, UN Secretariat 1993, Lee. 1966. Williams et al. 1986. Hoffman et al. 1992). The fact that men move a farther distance than women is explained by men’s advantages over women; men’s higher educational levels encourage men to move farther to seek the best match for their human capital, a higher possibility of earnings, and more opportunities to get well paid jobs. Short distance moves can reduce the risk of moving of people. especially of women who are more vulnerable in the labor market than men. because short distance helps them reduce the monetary cost of moving and provides them easy access to information about economic opportunities. As for Vietnamese migration, distance of move seems to be related to gender. Like women in other countries. Vietnamese women are reported to migrate shorter distances than men. This fact might be explained by women’s traditional roles and responsibilities as well as their low educational levels. which tend to attach women to their housework roles and highlight their greater risk in the labor markets. Thus, the limited distance they migrate may be a way to reduce the cost of moving or a way their families protect and control them (Dang Nguyen Anh 1997, VPC 1991a, Le Thi Nham Tuyet et al. 1995). Drawing upon this discussion, I hypothesize that men move a farther distance while women move a shorter distance (Hypothesis 6.2.a). In addition, the findings of migration studies show that there are not only gender differences in distance of migration. but also differences in the social status of potential migrants in relation to distance of migration (Lewis 1982, Ringdal 1993, Hoffman et al. 1993, Landry et al. 1988). The relation between social status and migration distance is understood as a reflection of the social and economic capabilities of people in seeking the best place to reside, no matter how far it is. People with high social status (income levels) migrate a farther distance than those with low social status. They can cover the cost of a long distance move while those who are poorer often cannot. Similar to other transitional economies, Vietnam is experiencing uneven development that affects the movement of different social classes, especially those with low income and low prestige occupations such as in agriculture (Vu Dung 1996, Trinh Duy Luan 1992, Pham Bich San 1992, Luu Dai Thuyet et a1. 1995, Nguyen Dinh Tan 1995). The better-off classes tend to be able to search out good economic opportunities to improve their social status and, therefore. they are involved in long-distance migration. In contrast, people of low income groups are unable to afford a long distance move because of the great financial investment such a process involves (Nguyen Luong Trao 74 1998). Hence, poor people have less chance to move as far as the rich because they cannot afford the cost of moving. I thus hypothesize that people with higher income levels move a farther distance than those with lower income levels (Hypothesis 6.2.b). With respect to the role human capital plays in distance ofmigration, many studies reveal that human capital has a positive relationship with distance of migration. That is. people with higher levels of human capital (higher educational levels) move a farther distance than those with lower levels because they seek to optimally use their existing human capital and increase the economic returns of migration (Hugo 1993, Becker 1975. Trzcinski et al. 1991). This datum is true for male migrants as migrants in general. but it is not clear if women with higher levels of human capital move a farther distance than those with lower levels, because these studies mainly focus on the economic aspect of migration and ignore the gender dimension. Under the socialist command economy, Vietnamese population movement from rural to urban areas or cities was closely related to the labor recruitment of highly educated people for work in the non-agricultural sector (CPLS 1996, Nguyen Luong Trao 1998, Nguyen Thanh Tuan 1998. Nguyen Xuan Thao 1998). Rural people with high educational levels could move long distances to urban locations where they were hired regardless of their origin. Consequently, similar to the findings in the migration literature, I hypothesize that people with higher educational levels move a farther distance than those with lower educational levels (Hypothesis 6.2.c). Further, as findings in the migration literature mentioned above reveal, there is a positive relation between distance of move and people’s social status, i.e., income levels 75 and occupational status. Even though no studies in Vietnam have explored this relationship. based on findings in the migration literature and some reports about the association between population movement and the labor recruitment of highly educated people for non-agricultural occupations (see above discussion), I hypothesize that people working in agricultural occupations move a shorter distance than those working in other occupations (Hypothesis 6.2.d). Nevertheless. other studies found that. regardless of gender. social connections at destination play an important role in selection of migration destination. People tend to move to places where they have social networks (Wilson 1998, Repak 1994, Straugh 1984). Implicitly. these networks determine distance of migration. People with social connections tend to move farther than those without social connections because of their greater guarantee of success in moving. The Vietnamese population tends to rely very much on their social networks in migration. The well-defined destination is the place where potential migrants have social networks. regardless of distance (Dang Nguyen Anh 1997, 1998c, Trinh Khac Tham et al.1998). Social connections provide needed assistance (economic and non-economic) for people’s resettlement at destination. Therefore, the existence of social connections at destination rather than distance. is a prime consideration for potential migrants (Le Thi 1998, Dang Nguyen Anh 1997. 1998a). From this discussion, I hypothesize that people with social connections at destination move a farther distance than those without social connections (Hypothesis 6.2.c). 2.3.Adaptation outcomes of male and female migration 76 The outcome of labor migration is frequently viewed as the economic improvement of migrants (Massey et al. 1993). Economic improvement is seen as an increase in the level of income and employment or occupational status at destination. In addition, in this adaptation process, labor migrants might experience different degrees of satisfaction with their move depending on what they achieve by moving. 2.3.a. Economic Improvement Migration studies of the adaptation process in different countries found that the outcomes of migration depend greatly on migrants’ background, such as their gender, age, marital status, and educational level. Specifically, it is found that migration outcomes differ between women and men (Cooke et al. 1996, Chant 1992a). Many studies found that men gain more by moving than women because they have greater economic opportunities at destination than women (Hugo 1993, Piampiti 1984, Ariffin 1984, Hong 1984). Women are tied movers; they move for their husbands’ career or for family reunion, and thus, they gain less economically than men. Moreover, women experience gender discrimination in the labor market and their gender roles in the household limit their choices and opportunities at destination (Robinson 1993, Cooke et al. 1996, Chant 1992a). These gender differences in outcomes of migration also result from a household strategy, which expects greater economic returns from male than female migrants because women are assumed to be disadvantaged in the labor market. Hence, male migrants are provided more opportunities than women to fulfill their roles in society (Robinson 1993, Cook et al. 1996). Consequently, men’s economic achievements are greater than those of women. 77 Vietnamese migration studies usually focus on the effects of migration at both the macro and micro levels of social structures such as socio-economic conditions at destination and origin, and socio-economic status of migrants’ households (Bui Viet Bao 1994, Le Thi 1998, Trinh Khac Tham et al. 1998). They mainly aim to investigate the impact of voluntary migration on urban life and environment. Specifically. they explore changes in urban labor markets in terms of labor cost. employment criteria. social perceptions about the service sector. or certain forms of labor (e.g., service workers, maids) (Nguyen Van Chinh 1996, Trinh Duy Luan 1996). At the micro level, many studies reveal that voluntary migration strongly impacts household economic status. Most migrants reported that their economic status improved due to migration. They had higher employment rates, and higher income levels compared to those they had before moving (Trinh Khac Tham 1996, CPLS 1996). Most studies report that male migrants are more successful than female migrants (Than Van Lien 1997. Phan Quynh Nga 1998. Le Thi 1998, Luu Minh Tri 1998, Trinh Khac Tham et al.1998. Tana 1996, Nguyen Luong Trao 1998). And this is explained in ways similar to those in studies about migration in other developing countries (Dang Nguyen Anh 1998a). For example, even though Vietnamese women’s educational levels have increased. their educational level is still lower than that of men. Migrants’ economic successes in cities depend greatly on their educational levels. areas in which men are advantaged over women. Women have lower educational levels than men. In addition. women tend to be clustered in low- skilled and low-paid jobs (occupations in the service sector and in mass production). while men have more and varied opportunities to obtain greater returns from migration (both economic and social status) (Khoo et al. 1984, 1994, Piampiti 1984. Ariffin 1984. 78 Hong 1984. Eviota et al. 1984. Le Thi 1998, Bui Viet Bao 1994, Trinh Khac Tham et al.1998. Nguyen Quoi 1996). Obviously. women achieve less with migration than men and this is a result of relations of patriarchy, not women’s abilities. Thus, I hypothesize that men achieve greater economic improvement with migration than women (Hypothesis 7.1.a). Some studies in developing countries reveal that migration outcomes are associated with age and marital status. Such differences result from the fact that young single people have greater possibilities for employment in the labor market than do older ones (UN Secretariat 1993). This age differential is explained by the aim of employers to maximize their profits by reducing the cost of labor. hiring young single people, especially young single women (Jelin 1997. Lim 1993, Aslanbeigui et al. 1994, Summerfield 1994, Pyle 1994, Fuess et al. 1994). Marital status differences among women and between women and men in the outcomes of migration are explained by the negative effect of associational migration on married women: tied movement disrupts women’s employment. Moreover. single women have more freedom to move than married women who. as wives or mothers, are burdened with a myriad of domestic duties (Payner et al.1992, C hattopadhyay 1997, Lichter 1983, Portes 1993). Therefore. the economic achievements of single women are implicitly greater than those of married women. F urther. married women experience a less successful adaptation process (difficulty in finding jobs and obtaining high incomes) than single ones because of employers’ perception of their labor force commitment, assuming that they are attached to their housewife role rather than to the worker role 79 (Morrison et al. 1988, Cooke et al. 1996). Their labor and human capital are devalued in the labor market. There is no report on the relationship between age and marital status and economic achievements with migration in Vietnam. Nevertheless, based on the fact that most Vietnamese migrants are young and single. and almost all migrants report that their economic status improved with migration (Phan Quynh Nga 1998, Than Van Lien 1997. Bui Viet Bao 1994. Nguyen Van Chinh 1996, Nguyen Quoi 1996, Tana 1994), we can infer that the economic achievements of migration are affected by age and marital status (youth and singleness). These attributes help migrants succeed in resettlement, providing them with a high possibility of being employed at high earnings. Therefore, I hypothesize that young people gain more economically by moving than older people (Hypothesis 7.1.b) and that single people achieve greater economic improvement with migration than married ones (Hypothesis 7.1.c). Findings in the migration literature reveal that adaptation outcomes are positively associated with educational levels (Massey et al. 1993, Sander et al. 1996. Hunt et al. 1985, Vecoli 1998. Sowell 1997, Wilson 1994, Khoo et al. 1984). That is, the higher the educational levels possessed, the more likely migrants will be employed and hold well- paid jobs. This is not always the case for women. however. Female migrants tend to be overeducated for some of the jobs they find due to gender discrimination in the labor market, and their associational type of movement. More specifically. although they may be highly educated, their economic achievements are lower than those of men in terms of income level and occupational status (Tyner 1994, Lichter 1983, Zlotnik 1995b. 1993, Robinson 1993). 80 Similar to findings in the migration literature, studies in Vietnam have found a positive relation between education and economic achievement. That is, migrants are reported to achieve greater economic improvement if they have high rather than low educational levels (Dang et al. 1997, Trinh Khac Tham 1996, CPLS 1996). Based on the assumption in human capital theories about the positive relation between human capital and outcomes of migration as well as findings in the migration literature, it is expected that the same positive relation will hold in Vietnam. 1 thus hypothesize that people with higher educational levels achieve greater economic improvement by moving than those with lower educational level (Hypothesis 7.1.d) Moreover, the studies of Lingam (1998), Guest (1993), and Rodenburg (1993) showed that the adaptation outcomes of migration also are determined by a household’s socio-economic status. This means that people of higher status gain greater returns from migration than those of lower status. As noticed in migration studies of developing countries (e.g., Lim 1993. Guest 1993. Chattopadhyay 1997, and Hugo 1993), migrants of higher status, i.e.. with higher income levels. have greater possibilities of being employed and obtaining well-paid jobs at destination than those of lower status. These are achieved via their social connections to people of the same strata. So, in addition to economic status, implicitly, social connections strongly shape the adaptation outcomes of migration (i.e., economic improvement). With respect to social connections, only a few studies on Vietnamese migration indicate the importance of social networks, particularly kin assistance in resettlement (employment) (Trinh Khac Tham et al.1998. Le Thi 1998, 1995, Dang Nguyen Anh 1998a, 1998b. Than Van Lien 1997, Bui Viet Bao 1994). They confirm that assistance 81 from kin helps migrants improve their economic status at destination. Drawing upon this result. I hypothesize that people with social connections at destination achieve greater economic improvement with migration than those without such social connections (Hypothesis 7.1.e) and I expect that people with kin relations at destination achieve greater economic improvement with migration than those with non-kin relations (Hypothesis 7.1.1). 2.3. Degree of satisfaction It is clear that degree of satisfaction with the outcomes of migration is closely related to the returns, both economic and non-economic. of migration. People who experience greater economic improvement are more satisfied with migration than those who experience less economic improvement (Poot 1993, Marshall 1984, Williams et al.1986, Lu 1998). With respect to gender differences in degree of satisfaction with migration, findings in the migration literature reveal that men are more satisfied with migration than women (Lu 1998, Williams et al. 1986). This is explained by the fact that men are afforded greater economic opportunities and achieve more by moving than women. Moreover, women are less satisfied with migration because they face discrimination in the urban labor market where their acquired education is devalued. and employers consider them less committed to work than men (Morrison et al. 1988, Cooke et al. 1996). As tied movers. women experience greater constraints than men in the migration process, and their tight connections with social networks reflect their strategy to overcome disadvantages (Boyd 1989). Studies of migration in Vietnam deal with degree of satisfaction with migration. Specifically, it is reported that Vietnamese migrants are highly satisfied with migration. 82 especially with the economic improvement they achieve by moving (Bui Viet Bao 1994. Nguyen Van Chinh 1996, Luu Minh Tri 1998, Nguyen Quoi 1996). Nevertheless, women tend to be less satisfied with migration than men due to the constraints they face as a result of the persistence of gender discrimination at home and at work (Le Thi 1995, 1998, Dang Nguyen Anh 1997, 1991). Thus. similar to findings in the migration literature, a similar gender difference in degree of satisfaction with migration is found in Vietnam. Given the forgoing discussion. I hypothesize that men are more satisfied with migration than are women (Hypothesis 7.2.a). Some studies show that marital status also strongly impacts the degree of satisfaction with migration (Williams et al. 1986. Lu 1998). Married migrants are not as satisfied as singles because they have more conflicts with spouses or family members due to their position in their households as well as their responsibilities compared to singles. Married women in particular may be less satisfied with migration because of its negative effects on their employment, which is interrupted due to their tied movers’ role (they sacrifice their career for that of their husbands), or due to their domestic roles that require them to take care of children, or sick family members instead of devoting their time to work as men do (Morrison et al. 1988, Payner et al.1992, Chattopadhyay 1997, Lichter 1983, Portes 1993). Marital status also plays an important role in shaping the degree of satisfaction Vietnamese people have with migration. Similar to findings in the migration literature, Vietnamese single people tend to be more satisfied with migration than married ones. This is especially true for women. Despite the fact that Vietnamese women gained equal rights with men under the subsidized system. they experience difficulties because of their 83 marital status and educational levels. Single women are more satisfied with migration than married women, because they are able to adjust in the urban environment more easily. i.e.. they have a broader range of economic opportunities than married women (Dang Nguyen Anh 1998a, Phan Quynh Nga 1998). Thus, I hypothesize that single people are more satisfied with migration than are their married counterparts (Hypothesis 7.2.b). Other studies reveal that degree of satisfaction with migration depends greatly on migrants’ social capital. There is a significant relationship between degree of satisfaction and social capital (Marshall 1984, Martin et al. 1983, Speare et al. 1982, Sidorenko 1995, Douglas et al. 1992, Massey et al. 1993, Boyd 1989, Huang 1983). As for human capital. it is found that degree of satisfaction depends greatly on migrants’ achievements at destination, particularly. economic success. which is shaped by individual levels of human capital. The higher the educational level, the more satisfied with migration people are (Martin et al. 1983, Speare et al. 1982, Sidorenko 1995). However, women with higher educational levels are less satisfied with migration than those with lower educational levels because they are overeducated for the jobs available for them at destination and discriminated against in the urban labor market. They are paid less than men and their economic success is not as great as that of men (Tyner 1994, Wilson 1998, Zlotnik 1993. Morrison et al. 1988, Lichter 1983). As for male migrants, it is reported that they also experience discrimination in the labor market due to their migration status. cultural and ethnic differences, and lack of social networks; degree of discrimination tends to be increased for those with high educational levels because of the keen competition with urban workers for well-paidjobs (Du Toit 1990. Lewis 1982). Thus, 84 men with higher educational levels are less satisfied with migration than those with lower educational levels as well, even though they might be more satisfied with migration than their female counterparts because of their greater economic achievement. Vietnamese women with higher educational levels are reported to be more able to find the best match for their human capital in urban areas than in rural areas (Le Thi 1995. Hoang Thi Lich 1993. Do Thi Binh et al. 1996). It is not clear. however, ifthey are more satisfied with migration than those with lower educational levels. Based on findings from the migration literature and the globalization of Vietnamese economy. it is expected that Vietnamese people also experience a similar discrimination and exploitation in the labor market due to the burgeoning of the service and informal sectors. Drawing from this discussion, 1 hypothesize that people with higher educational levels are less satisfied with migration than those with lower educational levels (Hypothesis 7.2.c). Migrants with social capital tend to be more satisfied with migration than those without social capital (Douglas et al. 1992, Massey et al. 1993. Marshall 1984, Speare et al. 1982, Boyd 1989) because the members of their social networks provide assistance for their needs in resettlement. This is true for both male and female migrants. Even though there is no study on the impact of social connections on degree of satisfaction in Vietnam, from findings in the migration literature discussed above, we can infer that social capital plays an important role in degree of satisfaction with migration. This inference is based on the fact that most migrants are satisfied with migration (Bui Viet Bao 1994), and the fact that most migrants rely on social connections in the adaptation process (Dang Nguyen Anh 1998c). Traditionally, Vietnamese people have very close relationships and 85 are enmeshed in webs of great obligations to those in their social networks; they tend to cluster in the same or nearby areas with their kin and friends as a way to obtain or provide help in dealing with difficulties (Van Tao 1998, VHT 1993). Thus, satisfaction with migration can be achieved more easily with social networks than without such connections. 1 thus hypothesize that people with social connections at destination are more satisfied with migration than those without such social connections (Hypothesis 7.2.d). SUMMARY This study explores three sub-processes of migration -- decision making, moving. and adaptation -- in relation to the demographic attributes and human, financial, and social capital of a sample of Vietnamese migrants. To clarify gender differences in the migration process, I drew on findings in the migration literature to propose the following hypotheses: - Decision-making process: 1. Reasons for migration: Hypothesis 5.1 .a. Men move for economic reasons more than women while women move for family reasons more than men. Hypothesis 5.1.b. Single women and men (both single and married) migrate for economic reasons while married women migrate for family reasons. 2. Decision-makers of migration: Hypothesis 5.2.a. Men are self-determined movers while women are directed to migrate by family members. 86 Hypothesis 5.2.b. Young women and men are directed to move by family members. Hypothesis 5.2.c. Single women and men are directed to migrate by family members while married men are self-determined movers and married women join with their husbands in making the decision to migrate. Hypothesis 5.2.d. Both women and men with higher educational levels are self- determined movers while those with lower educational levels are directed to move by family members. 0 Moving process: 1. Type of migration Hypothesis 6.1.a. Men move alone while women move with others or associationally. Hypothesis 6.1.b. Men and single women move solitarily while married women move with others. Hypothesis 6.1.c. Both men and women with social connections at destination move alone. Hypothesis 6.1.d. Men, regardless of type of relations, move alone while women with kin relations move alone and those with non-kin relations move with others. 2. Distance of migration: Hypothesis 6.2.a. Men move a farther distance while women move a shorter distance. 87 Hypothesis 6.2.b. People with higher income levels move a farther distance than those with lower income levels. Hypothesis 6.2.c. People with higher educational levels move a farther distance than those with lower educational levels. Hypothesis 6.2.d. People working in agricultural occupations move a shorter distance than those working in other occupations. Hypothesis 6.2.e. People with social connections at destination move a farther distance than those without such social connections. - Adaptation outcomes: 1. Economic Improvement: Hypothesis 7.1 .a. Men achieve greater economic improvement with migration than women. Hypothesis 7.1.b Younger people gain more economically by moving than older people. Hypothesis 7.1.c. Single people achieve greater economic improvement with migration than married ones. Hypothesis 7.1.d. People with high educational levels achieve greater economic improvement with migration than those with low educational levels. Hypothesis 7.1.e. People with social connections at destination achieve greater economic improvement with migration than those without such social connections. Hypothesis 7.1.f. People with kin relations at destination achieve greater economic improvement with migration than those with non-kin relations. 88 2. Degree of satisfaction: Hypothesis 7.2.a. Men are more satisfied with migration than are women. Hypothesis 7.2.b. Single people are more satisfied with migration than are their married counterparts. Hypothesis 7.2.c. People, especially women, with higher educational levels are less satisfied with migration than those with lower educational levels. Hypothesis 7.2.d. People with social connections at destination are more satisfied with migration than those without such social connections. 89 CHAPTER 3: METHODOLOGY This study uses the MOLISA data set on internal migration into Hanoi in 1996, which focuses on the characteristics of population movement in the period 1986-96. The purpose of this chapter is to describe the MOLISA data set in terms of its research design, sampling. and data collection. The chapter also provides a discussion of the conceptualization and operationalization of the main study variables as well as the analysis strategy adopted for this study. 1. Study site Vietnam consists of seven regions: Northern Upland. Red River Delta, Northern Central, Central Coast, Central Highland, Southeast, and Mekong Delta. The Red River Delta and Southeast are the most crowded areas of the nation due to advantages associated with their quality of land, location, and access to market and transportation; they thus have become the main receiving areas of internal migrants. Hanoi is located in the Red River Delta in the northern part of Vietnam. This study of migration into Hanoi will contribute to understanding about urbanward migration patterns during Vietnam’s transition period (19905), particularly enriching knowledge about gendered migration in Vietnam. 2. Data source: research design, sampling and data collection The MOLISA data set has a sample size of 1.300 migrants, aged 13 and over. In the following discussion, I describe the content of the original data set and its features. Then, I clarify the data selected for this study. 90 _a. The MOLISAMa set 0 Definition of migrants Within the MOLISA data set, the identification of a migrant was based on the difference of individuals’ place of residence before April 1. 1986 and their present place of residence. This means that migrants were people whose place of residence in Hanoi on April 1, 1986 was different from their previous residence. This category constitutes the selected sample for my study. Seasonal migrants were people who moved to Hanoi after October 1, 1995 and intended to go back to their place of origin within the three months following the date of interview. This category will not be used in my study. Non-migrants were those who never had moved out of Hanoi or had moved somewhere else for a time and then returned to Hanoi before April 1. 1986. This category is also not included in my study. 0 Sample size and design of the MOLISA survey Hanoi consists of five inner districts named Hoan Kiem, Ba Dinh, Dong Da, Hai Ba Trung, and Tay Ho, and five suburban districts named Gia Lam, Thanh Tri, Soc Son, Tu Liem and Dong Anh. The five inner districts and some quarters of Gia Lam, Thanh Tri, and Tu Liem were chosen for the MOLISA study. + Sample size Due to time and budget constraints, a sample size of 1,800 was drawn for the MOLISA survey, of which 1.300 were migrants and 500 were non-migrants. + Sample design 91 First, staff of the Center for Population and Human Resources Studies organized meetings with agencies of Hanoi to gather information on migration into the city. Then they worked with all sections at the district level to identify the areas in which migrants clustered. This information consisted of the total population and number of migrants in each quarter, sub-quarter, and cluster of households. Second. areas in which migrants were most clustered were selected. The criteria for selection were (1) high rate of migration. (2) each district had sub-quarters represented in the sample. and (3) total number of households was about 10,000. All sub- quarters that had a high rate of migrants were included in the sample frame. Then, sub- quarters were randomly selected to ensure that the total number of households included in the survey was about 10,000. The total sub-quarters selected were 24 in 14 quarters. Third, all persons in all household units in the selected sub-quarters were listed using a household slip that included information on each person’s name, age, place of residence on April 1.1986 and the time she/he expected to stay in Hanoi. The data were classified into two lists. The first list consisted of permanent migrants aged 13 and older and those who intended to stay in Hanoi for at least 15 days. The second list consisted of non-migrants aged 13 and over. In this step, the total number of household units included was 10,158 with 42,811 persons, among whom 35,001 persons were 13 years of age and over (81.76%). Among those 35,001 persons, 6,299 persons were considered migrants. Subtracting those who intended to stay less than 15 days. 5,813 persons were included in the sample frame of migrants. Fourth, respondents were selected according to the proportion of migrants in each district. Then the list of respondents of each district was randomly selected from 92 the first list mentioned above (about 30 % of the total migrants in each district were selected). Thus. 1.800 respondents were selected. + Information was collected in face-to face interviews using questionnaires. I do not know, however, if interviewers arranged an interview with the heads of households only or with any adult in a selected household. The person interviewed supplied information for her/himself and demographic information for all other residents in the household. 0 Questionnaire The MOLISA data set was developed based on six sets of questions. I. Background and characteristics of the households: This section included questions about the number of household members. and each person’s relationship to the head of household, gender, data of birth. place of birth, educational attainment, marital status, economic activities, usual place of residence on April 1, 1986, date of arrival in Hanoi. migration status, and the conditions of housing and its facilities. 2. Migration experience: This section included questions on the migration history (of the person responding to the questionnaire) such as place of residence prior to April 1,1986. date of the first move to Hanoi, date of the last move to Hanoi, place of residence prior to the last move to Hanoi. educational attainment prior to the last move, main reasons for leaving previous place, main reasons for choosing Hanoi as place of destination. relations with any one who lived in Hanoi before moving, secondary migrants to Hanoi (followers or associational movers), type and sources of information 93 about Hanoi before moving, sources and kinds of assistance after moving to Hanoi, and employment status before moving. 3. Employment and income: This section asked respondents about their economic activities and income with questions on main economic activities during the last 30 days. difficulties while looking for a job. means of job search, status at work. procedure for recruitment. economic sector. main and secondary jobs. income of main and secondary jobs. main source of support for the unemployed, and evaluation by female migrants of their working place. 4. Current conditions of migrants: This part consisted of questions on the respondents’ assessment of their situation in Hanoi compared with that at origin, their registration status. place of residence of spouse and children, remittances, and financial support for migrants. 5. Seasonal migrants: This section included questions about the demographic characteristics of respondents’ family members at the last place of residence, housing conditions and facilities. land size, availability of public services, entertainment, and frequency of visiting their hometown. 6. Registration: This section included questions about the registration status of respondents in Hanoi. the difficulties facing those without a permanent registration card and the attitude of local authorities toward their presence in Hanoi. 2b_.Data selection for this stud;I Given the aim of this study, migrants in Hanoi. I selected only permanent migrants aged 13 and over from the MOLISA data set. My sample includes 805 persons. among whom 426 are women (52.9%) and 379 are men (47.1%). 94 Since this data set does not include information about non-migrants at origin, I use documentary material to describe the socioeconomic context of migrants at origin and to compare the characteristics of migrants to those of non-migrants. These sources include published materials, the 1989 national census. the 1992-93 Vietnam Living Standard Survey (VLSS) and the Vietnamese statistical yearbooks for 1986-97. 3. Conceptualization and measurement of main variables Due to the limitation of using secondary data. the 1996 MOLISA survey, 1 was only able to measure a few aspects of human, financial. and social capital, in addition to some demographic attributes of migrants (gender, marital status. and age). This study focuses on the relation of these aspects to the three stages of the migration process: the decision-making, moving, and resettling (adaptation) processes (see Figure 1.b). Human capital is understood as all personal capabilities. However, I only measured aspects of human capital related to the economy, i.e., education, employment status, and occupation; work skills cannot be explored due to the ambiguity of the question included in the MOLISA survey. Social capital usually means the social relations of people, but this study measured only the effect of the availability of social networks and type of relations (kin or non-kin) on the decision-making, moving and resettlement processes. Financial capital is defined as the income level of migrants. 3.1. Independent variables The migration process is impacted by various factors, but in this study, given the limitation of the MOLISA data set and the goal of this study, I mainly focus on four sets of indicators. They are migrant's demographic attributes (AGE, GENDER, MARITAL STATUS), human capital (EDUCATION. OCCUPATION, EMPLOYMENT 95 $528 so oEflE> 25 Lo Soho cm 2365 Tlv @588 $0538,» 2: 5953 20:22 30%.. A- - -V men: 32% ”Bo Z Queue—2 mo womb .mfoéo: 368 do bzfingmv 3:98 Boom Amummuooca 839%? AmESm REE: m can .wE>oE 388:3 , .wEme 556qu @8005 5:832 awe. .5953 ~2me 365:5 / . moazntxm . oEaEoneQ \ \ $3me EuEtAoEEo . £28380 #538463 6:30 52:32 Hxvucoo UMEOEOOO IOMUOW ...-«52> E :23..me 2:88 1:5 2.2: East—St: he .35:— EUEEEH A: 0.5%..— 96 STATUS), financial capital (INCOME), and social capital (TYPE OF RELATIONS and AVAILABILITY OF SOCIAL NETWORKS IN HANOI or SOCIAL CONNECTIONS). 3.1.1. Mim'ant’s demographic attributes are measured by GENDER, AGE, and MARITAL STATUS. 3.1 .1 .a. Respondents were asked to report their GENDER, which was coded O for male and l for female. Female migrants comprised 52.9 percent of the sample and 47.] percent were male migrants. 3.1.1.b. AGE is measured in years. Respondents were asked to report their age in actual years in the question “How old are you?” The age of migrants ranged from 13 to 90. The frequency distribution revealed that migrants under 15 comprised a small proportion (3.6 %) of the sample and this was similar for those over 60 (4.8 %). Migrants in this study mainly were 15-34 years of age (64.4 °/o of the total). within which people of ages 20-29 made up 42.3 percent of the total. Migrants aged 35-60 comprised 27.] percent. I will use raw data in all statistical analyses. 3.1.1.c. The information about the MARITAL STATUS of migrants was derived from the question “Are you married?.” Four categories were offered to migrants and coded as: l = never married, 2 = married, 3 = widowed. 4 = separated. Half of the migrants (50.9 %) were never married while 44.3 percent were married. The proportions of widowed (3.9 %) and separated (0.9 %) were small. Therefore, I combined these two categories and the category married into one new category. ever married. As a result, I have two dichotomous values, 1 = never married and 0 = ever married, for the marital status variable. A higher percentage of male migrants were single, i.e., never married, 97 (56.2%) than female migrants while more female migrants than male migrants fell into the ever married category (63.7%). 3.1;. Financial CM is measured by INCOME. Respondents were asked to report their income before and after moving with the actual numbers in two questions “How much do you estimate you earned per month at origin?" and “How much do you estimate you earn per month in Hanoi?” The respondents’ past incomes reported ranged from VN$ 7,000 to VN$ 900,000 (about US$1 to US$90) while their incomes in Hanoi ranged from VN$ 10,000 to VN$ 2,000,000 (about US$ 1 to USS 200). I grouped migrants’ income into five categories based on the national scale so that I am able to compare the migrants” income levels to those of the nation and regions within the country (SPC 1994) 1= less than VN$ 75,000 (about US$ 7.5) (poverty line was below VN$ 91 ,000)' 2: VNS 75.000 — 150,000 (about USS 7.5 - 15) 3= VN$ 151,000 — 300,000 (about US$ 15 - 30) 4= VN$ 301,000 — 600,000 (about US$ 30 - 60) 5: more than VN$ 600,000 (above US$ 60) The proportion of non-responses to the question about income before moving, was high (63.9 %). Of those responding, 23.4 percent were at the 1St income level (63.1 % women and 36.8 % men) ' In 1993 the national average poverty line was calculated at 1,090 thousand dong per person per year (90.830 dong per month), taking into account local differences in prices and regional differences. On average, the poverty line is higher for Vietnamese living in urban areas: 1,293 thousand dong per person per year (107.750 dong per month) compared to 1,040 thousand dong per person per year (86.667 dong per month) in rural areas. With this poverty line, about 51 % of the Vietnamese population is classified as poor (World Bank 1995:6-7) 98 22.0 percent were at the 2nd income level (46.9 0/o women and 53.1 % men) 33.7 percent were at the 3rd income level (54.1 % women and 45.9 °/o men) 17.5 percent were at the 4‘h income level (29.4 0/o women and 70.6 % men) 3.4 percent were at the 5'h income level (40 % women and 60 % men) As for migrants’ income in Hanoi, the proportion of non-response was lower than that before moving, but also high, 45.6 percent. Among the responses, the frequency distribution was as follows: 45.8 percent were at the 1St income level (46 % were women and 58.9 % were men) 1.1 percent were at the 2'1d level (88.9 % were women and 1 1.1 % were men) 17.9 percent were at the 3rd level (60.4 % were women and 39.6 % were men) 25.0 percent were at the 4‘h level (42.8 "/0 were women and 57.2 °/o were men) 10.1 percent were at the 5th level (33.1 % were women and 66.7 % were men) These income groups are used in Chapter 4 only for comparative purposes. The actual incomes of respondents before migration are used in the statistical analyses included in Chapters 5 and 6, while that of incomes of migrants in Hanoi are used for Chapter 7. 3.1.3. Human capital is measured by EDUCATION, OCCUPATION, and EMPLOYMENT STATUS. 3.1.3.a. To investigate the EDUCATION of migrants, respondents were asked two questions. One question asked about education before migration: “Which grade did you finish and in which educational system?” The other question asked respondents to report their current education by filling in information about the grade completed, higher education, i.e., university or college education, and the educational systems 99 attended. Information about educational levels was detailed because Vietnam has experienced several education reforms over the past five decades. The first reform was taken to change the education system from 9 grades to 10 grades (in the 19505); the second reform was to change from the 10-grade-system to the 12-grade-system in 1981, which has remained in place. As a result, responses were listed from first grade to twelfth grade, but they belonged to different education systems (9-grade-system, lO-grade- system, 12-grade-system). These incomparable grades needed to be rescaled to get one consistent educational system. I rescaled these different educational levels based on the models of these educational systems. For instance, the 9-grade-system had a model of 4+3, i.e., 4 = elementary = lst, 2nd, 3rd, 4th grades; 3 = middle = 5th, 6th, 7th grades; the lO-grade-system has a model of 1+4+3+3. i.e., 1 = preschool; 4 = elementary = lst, 2nd, 3rd, and 4th grades; 3 = middle = 5th, 6th, 7th grades; 3 = high school = 8th, 9th, 10th. and the l2-grade-system has a model of 5+4+3. i.e.. 5 = elementary = lst, 2nd, 3rd, 4th and 5th grades; 4 = middle = 6th, 7th 8th, and 9th grades; 3 = high school = 10th, 11th, and 12th grades. As a result, school grades from l“t to 12‘" were coded as l-12, and higher education (technical training, college, university, and post-university level) was coded 13. 100 F urthermore, in Vietnam we do not usually use grade to measure educational levels, but three degrees at school, i.e., elementary, middle and high school, and technical training, university or post-university. Therefore. I recodcd educational levels based on the model of a 12-grade-system such as: 1 = elementary. 2 = middle school, 3 = high school, 4 = technical training. university. and post- university. 1 group technical training, university and post-university into one category because the proportions of the sample achieving post-university and technical training were small (0.4 % and 1 1.0 %, respectively). Almost half of migrants (47.2 %. within which 44.2 0/o were women and 55.8 % were men) had a high education level (university, college) before moving. The percentages for the remaining migrants before moving having the middle and high school levels were 26.8 percent (59.4 % women and 40.6 ”/0 men) and 14 percent (53.1 % women and 46.8 % men), respectively. The proportion of migrants having elementary levels of education was 12.1 percent (66.3 0/o women and 33.7 % men). This 4-levels-education-variable is used in Chapter 4 for comparative purposes. In Chapter 5 and 6, education in its 13 levels (from 1St to 12th grade and 13 as higher education) is used for statistical analyses. As to current educational levels, migrants in Hanoi who had the middle educational level made up 34.5 percent (62.4 °/o women and 37.6 % men), while migrants with high school and higher educational levels comprised 37.4 percent (52.2 0/o women and 47.8 % men) and 24.8 percent (36 % women and 64 % men). respectively. This current education variable, coded as 13 levels is used in Chapter 7. 3.2.3.b. Respondents were asked to report their OCCUPATION in two different questions “What was your occupation before moving?” and “What is your main occupation in Hanoi?” The responses were coded as: 101 0= no answer. 1: managers of3 levels 2: managers of enterprises 3= managers of finance 4: technical managers 5 = technical managers of forestry or agriculture 6 = managers of education and training 7: managers of culture and art 8 = managers of medical and pharmacy 9 = managers of law 10 = secretarial/mental labor 1 l= heavy-duty electric equipment 12 = mineral exploration 13: metal train 14 = metal, electronic & electrical industries 15: chemical industries 16: paper products 17: construction, ceramic. glass products 18: forestry industries 19: publishing 20= textile 21= garment 22= fur and woolen 102 24=construction 25=agriculture 26: fishery 27= transportation 28: postal 29: trading. material supply 30: public services 31: domestic work 32= other 301 = goods collection 302 = public trading 303 = petty trade 304 = food services in hotels 305 = other trading and food services 31 l = three-cycle 312 = public cleaning 313 = cosmetic services 314 =jewelry's repair 315 = electronic maintenance 316 = shoes maintenance 317 = motor and bike maintenance 318 = housing decoration 319 = other. 23= food processing The categories of occupations could not be classified according to a scale of occupational prestige because they did not describe exactly what migrants did, i.e., whether they were blue- or white-collar workers or technicians. They did not connote the prestige which was accorded one or another job. Therefore. I could not use an international occupational prestige scale to recode these variables. I decided to use a scheme that (I) is commonly used to differentiate occupations in developing countries. and (2) implied notions of superiority / inferiority in terms of prestige. More specifically, the economic sector in Vietnam, like that in other countries, can be classified as formal and informal sectorsz, which differ from one another by the institutional framework of economic activity (Portes et al. 1989). This means that the formal sector stipulates terms of work in contracts and offers workers insurance and benefits; the conditions of work in this sector (e.g., health and safety) are covered by government rules and regulations while they are not in the informal sector. In contrast, workers in the informal sector are not protected and benefited institutionally; they are more vulnerable in terms of wages, health safety, and economic security than are those working in the formal sector. Moreover, the formal sector in Vietnam differs greatly 3 The economic sector in Vietnam before the 19805 included mainly a state sector, which was identical to the formal sector. In the transition period to the market economy, the economic sector included five elements: the state sector, agricultural sector or cooperatives, state capitalist sector, private capitalist sector, and self-employment (Dang Tho Xuong 1991, Nguyen Huu Tien et al. 1996, GSO 1996). Based on the definition of the formal and informal sector in Portes et al. (1989), the economic sector in Vietnam can be classified as the formal sector (including the state sector and state capitalist sector) and informal sector (including cooperatives, private capitalist sector, and self-employment). Noticeably, based on the official regulations, the informal sector is required to meet the minimum work safety, wage, social security or health insurance (NPPH 1994, DJP 2000), but it is not the case in reality. 103 from the informal sector in terms of the opportunity it provides to be promoted up the power ladder. Until the 19803, people working in the agricultural sector in Vietnam were organized into collectives (Dang Tho Xuong 1991, Do Thi Binh et al. 1996, Do Thi Binh 1997, GSO 1996). While they shared some characteristics with those working in the formal sector, they also were subject to some of the same conditions as those working in the informal sector. For example, they received the same insurance and benefits as those working in the formal sector. but their wages (as farmers) were not paid in cash by the government, but paid in kind by collectives based on their work points and the productivity of a season. Farmers received wages in the form of agrarian products which later were transformed into cash via sale in the market. Thus, their income heavily depends on weather, land, seeds, and demand. Consequently, due to the nature of agrarian production, people in this sector are more vulnerable than those working in the formal sector, and this feature is similar to that of the informal sector. The above discussion helps to acknowledge the superiority of the formal sector over the other two sectors (informal and agricultural sectors) in Vietnam. But it does not help rank these two sectors. To rank these economic sectors, 1 need to consider the educational (an aspect of power) and income (an aspect of privilege) levels of people working in these sectors based on the model of occupational hierarchy (Treiman 1977). As can be seen in Appendix 3.1 and 3.2, migrants in this study who worked in the agricultural sector prior to moving to Hanoi had the lowest educational (mean = 8.2) and income levels (mean = 105K) compared to those in the formal (mean = 1 1.8 for education and 308K for income) and informal sectors (mean = 10.2 for education and 356K for 104 income). Although migrants who worked in the informal sector had higher income levels than those who worked in the formal sector, the former had lower educational levels than the latter, and they also worked without protections and benefits. Therefore, implicitly, the formal sector is in the highest ranking while the informal sector is in the second and the agricultural sector is in the lowest ranking. Noticeably, respondents who reported their occupations as work in the informal sector mainly were employed in the service sector, which can be seen as a result of Vietnam’s transition into a market economy with the burgeoning of the service sector. Based on this argument, I recodcd the responses into four categories: 0 = agricultural, 3 previously coded as 25-26 1 = informal sector, previously coded as 31, 33-34, 301, 303, 311. 313-318 2 = formal sector, previously coded as 1-30. except 25-26, 302. 304-305,312 3 = other, previously coded as 32 and 319 This new coding is a ranking of occupational prestige with 0 (agriculture) the lowest rank and 2 (formal sector) the highest rank. Category 3 = other cannot be ranked because it is unclear to what kind of job it refers. Moreover, no respondents chose this category (0.0 %) as either their past or present occupation. This category is excluded from all statistical analyses. The proportion of people who did not report their occupation before moving made up 63.2 percent of the total migrants of this study (49.5 % women and 50.5 % men). 3 The categories of technical managers in forestry or agriculture (5) and forestry industries (18) were not included in the agricultural sector (or collectives) because these people were classified as workers in the state sector receiving wages from the government. Fishery was considered a part of the agricultural sector (GSO 1987:33) and classified accordingly. 105 Among the remainder who were employed (36.9 %), about half (45.1 %) of the migrants worked in the agricultural sector (59.7 % women and 40.3 % men), 12.1 percent held jobs in the informal sector (41.7 % women and 58.3 men), and 42.8 percent were employed in the formal sector (40.9 % women and 59.1 °/o men). This recoded past occupation variable is used in the analyses included in Chapters 5 and 6. As for present occupation, the proportion of people who did not report their occupation in Hanoi (non-response) made up 45.6 percent of the total migrants (58.6 % women and 41.4 % men). Of the remainder (54.4 %), the proportion of migrants in Hanoi who worked in the informal sector was 28.9 percent (46 % women and 54 % men) while that working in the formal sector was 78.1 percent (48.7 % women and 51.3 % men). No one worked in agriculture in Hanoi. This recoded present occupation variable is used in the analyses in Chapter 7. 3.1.3.c. EMPLOYMENT STATUS indicates the working status of migrants who had a job at least one week before the date of interview or one month before migration. Respondents were asked to report their employment status in two different questions “What was your employment status in the last month before moving?” and “What is the main current employment status you have in the last month.” Answers to these two questions were coded into seven categories: = employed 3 = schooling 5 = retired 7 = unemployed/ 2 = unemployed 4 = housework 6 = disabled/invalid don’t want to work. In my study, I mainly aim to uncover the economic status of migrants through their employment status (employed or unemployed). However, these two categories tended not to reflect the actual working status of women. who often worked hard without 106 formal payment (housework and/or in a family business). Therefore, housework was recoded as unpaid employment. The category 7 “unemployed, don’t want to work” was included in the unemployment category. The schooling category cannot be considered employment or unemployment because students devote more time to studying than to earning money even though they may work to pay for tuition. In addition, the retired and disabled/invalid categories cannot be considered as employment or unemployment because they include people who don‘t work, but receive pensions or financial aid. Therefore schooling as well as other categories such as retired and disabled/invalid were included in one new category “other employment status.” As a result, employment status was reclassified as paid employment, unpaid employment, unemployment, and other, and recoded as 0 = unemployment (previously coded as 2 and 7) 1 = unpaid employment (previously coded as 4) 2 = other (previously coded as 3,5, and 6) 3 = paid employment (previously coded as 1) This coding is also a ranking, which is based on the different possibilities people in these employment statuses had to earn money, as mentioned above. Specifically, people who were unemployed, obviously, had no earnings while those working without pay were creating income even though they were not paid for their labor. People included in the “other employment status“ (i.e., those in school and the retired and disabled) might be able to get financial aid or pensions to support their living and this might be considered a clearer financial source than that of those working without pay and 107 unemployed people. Finally, people working for pay have the most obvious earning source compared to people in the other categories. With respect to employment status before moving, originally, migrants who were employed made up 36.9 percent, while those who were unemployed comprised 10.6 percent. and those who were housekeepers, in school, unemployed but not seeking ajob, or retired/disabled comprised 4.6 percent, 32.8 percent. 3.5 percent, and 11.6 percent, respectively. As a result of recoding. the unemployment proportion was 14.0 percent (48.7 % women and 51.3 % men) while that for paid employment was 36.9 percent (49.5 % women and 50.5 % men). Workers without pay (unpaid employment) made up 4.6 percent (86.5 % women and 13.5 % men) while the “other” status of migrants was 44.5 percent (53.6 % women and 46.4 % men). This recoded variable is used in the statistical analyses of Chapters 5 and 6. Regarding employment status in Hanoi in the original categories, 54.4 percent of the migrants were employed, 24.3 percent were in school, 8.6 percent were doing housework, 7.6 percent were retired, 3.4 percent were unemployed, and 1.7 percent were disabled/invalid. As a result of recoding, the distribution of the responses for these categories was as follows: the proportion of unemployed was 3.4 percent (55.6 % women and 44.4 % men); the proportion of paid employment was 54.4 percent (48.2 % women and 51.8 % men); the proportion of unpaid employment was 8.6 percent (95.7 % women and 4.3 % men); and migrants of other status comprised 33.7 percent (49.4 % women and 50.6 % men). This recoded variable for current employment status is used for the analyses in Chapter 7. 108 3.1.4. Social capital is measured by the AVAILABILITY OF SOCIAL NETWORKS. and TYPE OF RELATIONS. 3.1.4.a. To investigate whether migrants had any connection with people in Hanoi before moving (labeled AVAILABILITY OF SOCIAL NETWORKS or SOCIAL CONNECTIONS), respondents were asked the question “Did you have any relative or friend who lived in Hanoi before your move?” Respondents were offered the choices of “yes.” “no.” and “don‘t know,” coded l. 2 and 3 respectively. The “yes” responses totaled 88.8 percent while the “no” responses were 10.6 percent and the “don't know” were 0.7 percent. Due to the very small missing information (i.e., “don’t know”). I combined this category with “no” and coded them as 0 = yes (previously coded as 1); l = no (previously coded as 2.3). The “yes” responses totaled 88.8 percent (53.6 % women and 46.4 % men) while the “no” responses made up I 1.2 percent (47.8 % women and 52.2 % men). 3.1.4.b. To investigate to whom migrants were connected before moving to Hanoi (labeled TYPE OF RELATIONS), respondents were asked the question “Who living in Hanoi before your moved were you connected to?” There were five possible responses, coded as: 1 = family members, 2 = relatives, 3 = friends, 4 = employers / mediators, 5 = other. Respondents were allowed to make three choices; the rate of non- responses on the first choice was small (11.3 %) while those on the second and third choices were great (68.0 % and 92.5 %. respectively). Thus, I excluded the last two choices from my analysis. As a result, I used only the first choice of respondents, within which family members represented 63.7 percent of the responses, relatives 29.8 percent. friends 5.9 percent, and employers/mediator 0.4 percent. Since the percentage of employers/mediators was small, and due to the close kinship between family members 109 and relatives within Vietnamese traditional customs and the nature of these relations, I combined the first two categories (family members and relatives) into one new category. kin relations, and the other two (friends and employers/mediators) into another new category. non-kin relations. They were coded as: l = kin (previously coded as 1.2); 0 = non-kin (previously coded as 3.4.5). As a result, the responses of kin were 71.5 percent (55 % women and 45 % men) while those of non—kin were 28.5 percent (50 °/o women and 50 % men). 3.2. Dependent variables: Migration is understood as a movement of population from one area to another area, which includes a number of sub-processes, such as preparation prior to move, involvement in moving. and resettlement at destination. The migration process in this study is examined as three sub-processes: decision making, moving, and resettlement (adaptation) processes. 3.2.1. Decision-making process is a process in which people make the decision about who migrates and why they migrate. This process is operationalized through two variables: REASONS TO MIGRATE, and DECISION MAKERS. 3.2.1 .a. To investigate REASONS TO MIGRATE, respondents were asked the question “Could you choose the most important two reasons that made you migrate?” There were 17 possible responses, coded as: 1 == unemployment/lack of job 10 = family reunion 2 = low income 1 l = married, separated. divorced 3 = being employed or change in workplace 12 = family conflict 4 = further education or training 13 = neighborhood conflict 110 5 = lack of school facilities for children 14 = returned from business, schooling, 6 = lack of health service military service 7 = housing problem 15 = visiting 8 = lack ofliving facilities 16 = professional visits 9 = urban attractiveness 17 = other The distribution of the first of the two most important reasons indicates that people most often migrated because of unemployment (24.6 %), family reunion (23.2 %), and educational improvement (18.3 %). The responses for the second reason people migrated were clustered in the category of family reunion (19.8 %) and low income (10.8 %). However. the proportion of non-responses for the second reason was great (37.3 %) compared to that for the first reason (0.2 %). Hence, I use the results only for the first reason given for migration, thereby helping to ensure the reliability of the information. Furthermore, due to the small number of responses for other reasons of migration (1.9%) and the similar nature of different responses (e.g., income and employment share an economic aspect). I have recoded the first reasons into the following four categories: economic, family, education, and other reasons. Economic reason was recoded as 1 (previously coded as 1,2,3) Education reason was recoded as 2 (previously coded as 4,5) Family reason was recoded as 3 (previously coded as 10,11.12) Other reasons were recoded as 4 (previously coded as 6-9, 13-17). The distribution of the frequency of the responses for these four categories was as follows: 38.6 percent for economic reasons (44.5 % women and 55.5 % men) 21.9 percent for educational reasons ((47.7 % women and 52.3 % men) 111 24.8 percent for family reasons (69.8 % women and 30.2 0/o men) 14.7 percent for other reasons (53.4 % women and 46.6 % men) 3.2.1.b. To explore DECISION MAKERS of migration. respondents were asked “Who decided that you migrate to Hanoi?” They were offered nine possible responses. but they had to choose only one. The responses were coded as: 1 = self 4 = parent 7 = relatives/friends 2 = wife & husband 5 = children 8 = employers 3 = wife or husband 6 = sibling 9 = other Most of the responses clustered in self-determining (36.1 %. within which 41.6 % were women and 58.4 % were men), parent (25.7 %, within which 51.7 % were women and 48.3 % were men). and wife & husband (16.9 %, within which 71.3 % were women and 28.7 % were men). The rest had a small number of responses (4.2 % for wife or husband. 3.4 % for children, 4 % for sibling, 2.6 % for relative or friends, 4.8 % for employers and 2.1 % for other). I kept the categories self and wife & husband (renamed as couple) the same, but grouped the categories wife or husband, parent, children, and sibling in one category, family members, and the rest in the category other. As a result, I had four new categories and the distributions of responses within the new categories were as follows: 1 = self-determining (36.1 %, within which 41.6 0/o were women and 58.4 % were men) 2 = couple (16.9 %, within which 71.3 % were women and 28.7 0/o were men) 3 = family members (37.4 % within which 59.8 % were women and 40.2 % were men) 4 = other (9.6 %. within which 36.4 % were women and 63.6 % were men). 3_. .2. Movingprocess refers to the way people migrate, including whether they move alone or with others and how far they move. This process is measured by TYPE OF MOVE and DISTANCE OF MOVE. 3.2.2.a. The information about TYPE OF MOVE came from the question “How many people moved with you?” which asked respondents to list the number of migrants who accompanied them to destination. The proportion of responses for “nobody” was 57.6 percent while the proportion of responses for accompanied migrants was 12.4 percent for one person. 10.7 percent for two people, 9.6 percent for three, and 5.5 percent for four people; the other responses (4.2 %) ranged from 4 to 18 people with very few cases. Since I am interested in whether an individual moved alone or with others, it is not necessary to take into account the number of people accompanying migrants. Thus, I recoded these categories as: 0 = solitarily (nobody); I = non-solitarily or associationally (from 1 to 18 people). The solitary category included 57.5 percent of responses (47.1 % for women and 52.9 % for men). while the non-solitary category included 42.5 percent of responses (60.8 % for women and 39.2 % for men). 3.2.2.b. DISTANCE OF MOVE was measured by the proportion of kilometers from the migrants’ origin to their destination. The information for this variable was given in response to the question “Where was your last place of residence before moving to Hanoi?” Respondents had to write in their province or city of origin. The total of migrants’ place of origin was 37 provinces. The proportion of migrants who had their origins in the Red River Delta was the highest (64.3 %). In this region, Ha Tay, Hai Hung, and Nam Ha were the most frequently reported (13.8 0/o. 17.1 % and 15.9 % respectively); Ha Bac, Hai Phong. Thai Binh, and Ninh Binh were reported in fewer 113 instances (7 %. 2.5 %. 6 0/o and 2.0 % respectively). Other provinces were reported infrequently (a total of20.9 "/0 for the other 18 provinces), except for provinces such as Thanh Hoa (7.2 %), Nghe An (3.5 %), and Ha Tinh (1.5 %), which belonged to the Northern Central region; and the provinces in the Northern Upland region (Vinh Phu 6.1 0/o. Hoa Binh 2.0 %, Bac Thai 1.7 %). Based on official geographical figures, I calculated the distances of move to Hanoi from other provinces: Ha Tay = l 1 km Hai Hung = 64 km Ninh Binh = 93 km Thanh Hoa = 153 km Ha Bac = 31 km Hoa Binh = 76 km Hai Phong = 103 km Bac Thai = 166 km Vinh Phu = 63 km Nam Ha = 90 km Thai Binh = 11 1 km Nghe An = 291 km Ha Tinh = 340 km Because of the expense involved in moving (i.e.. it is more costly to move a long distance than a short distance), I used a 50-km-scale to investigate the distance of move and recoded them as follows: 1 = Ha Tay, Ha Bac (S 50 km); 2 = Vinh Phu, Hai Hung, Hoa Binh, Nam Ha, Ninh Binh (>50 & S 100); 3 = Hai Phong, Thai Binh (>100 & S 150 km); 4 = Thanh Hoa, Nghe An, Ha Tinh, Bac Thai (>150); 5 = other provinces. The distribution of the responses of these categories were as follows: category 1 was 20.7 percent (52.1 % women and 47.9 % men) category 2 was 43.2 percent (50.4 % women and 49.6 % men) category 3 was 8.4 percent (63.2 % women and 36.8 % men) 114 category 4 was] 3.9 percent (57.1 % women and 42.9 "/0 men) category 5 was 13.8 percent (51.4 % women and 48.6 % men) This coding system is used for comparative purposes in Chapter 4. I use the actual kilometers ofa province from Hanoi to run statistical analyses for Chapter 6. 3.2.3. Adaptation is the resettlement process of migrants at destination, in which migrants experience success or failure. Adaptation outcomes are measured by ECONOMIC IMPROVEMENT and DEGREE OF SATISFACTION. 3.2.3.a. ECONOMIC IMPROVEMENT was measured by the difference between migrants‘ past and present occupations, employment statuses. and incomes, and was coded as WORK STATUS CHANGE and INCOME CHANGE“. 3.2.3.a. 1. WORK STATUS CHANGE is a combination of the OCCUPATIONAL and EMPLOYMENT STATUS CHANGES of migrants. OOCCUPATIONAL CHANGE is the difference between the prestige of the current and past occupations of migrants, which was ranked based on the economic sector in which they worked. Using the new coding system for occupations of migrants before and after moving, explained in Section 3.1.3.c, I subtracted the value of a migrant’s past occupation from that of his/her present one to obtain this new measure, OCCUPATIONAL CHANGE. As a result, OCCUPATIONAL CHANGE ranges from -3 (the least) to 3 (the most). Slightly over two-fifths of the respondents (42.6 %) had no information about their occupational change as a result of non-response. Among those having occupations before or after moving. the proportion of migrants who did not “ These two variables cannot be created as one index of economic improvement because they have different measurement scales. 115 achieve any occupational improvement was small (14.9 %). while those having the same occupational status were nearly twice as great (24.2 %). Most migrants who had occupations prior to and after moving achieved occupational improvement (60.8 %), within which those who achieved the greatest improvement made up more than half of the total (i.e.. 31.6 % compared to 60.8 %). OEMPLOYMENT STATUS CHANGE is the difference between the present and past employment statuses of migrants. which were discussed in Section 3.1.3.d. From this new ranking of employment status, to get a change score (i.e., the difference between employment statuses after and before moving) I subtracted the past from the current employment status. As a result, EMPLOYMENT STATUS CHANGE was created with scores ranging from -3 (the least) to 3 (the most). A great proportion of the migrants (65.0 %) had no change in employment status (50.9 % women and 49.1 % men). Very few migrants (8.7 %) had their employment status worsen (70.4 % women and 29.6 % men). Those who achieved employment status improvement made up about one quarter (26.3 %, within which 51.9 % were women and 49.1 % were men). OWORK STATUS CHANGE is the combined value of OCCUPATIONAL and EMPLOYMENT STATUS CHANGES. Adding the two values created WORK STATUS CHANGE, which ranged from -6 (the least) to 6 (the most). Migrants without responses made up 42.6 percent. About one-quarter of migrants (24.2 %) experienced no change in their work status with migration (39.3 % women and 60.7 % men) while those whose work status worsened comprised a small proportion (14.8 %, within which 59.4 % were women and 40.6 % were men). Most migrants achieved work 116 status improvement (58.9 "/0, within which 52.7 % were women and 47.3 0/o were men). Among people whose work status improved. 13.4 percent of migrants (43.5 % women and 56.5 0/o men) achieved the greatest improvement in work status. This newly created variable is used for analyses in Chapter 7. 3.2.3.a.2. INCOME CHANGE is derived by deducting the past income from the present income. These incomes were generated by the actual amount of money earned (see Section 3.1.2). As a result, INCOME CHANGE is the actual difference between the two amounts of money migrants earned before and after moving. Very few migrants (3.9 %) had no change in their income levels (21.1 % women and 78.9 % men), and only a small percentage (12.0 %) had their income levels reduced (69.0 % women and 31.0 % men). Most migrants achieved an increased income by migrating. Specifically, migrants who gained up to VN$ 300,000 (US$ 30) made up nearly half of the responses (46.3 0/o. within which 55.8 % were women and 44.2 0/o were men). Those who gained more than VN$ 300,000 to 500,000 (US$ 30-50) comprised 20.5 percent (44.4 0/o women and 55.6 % men) of the sample while those who gained more than VN$ 500.000 (US$ 50) made up nearly one-fifth of the sample (17.4 %, within which 33.3 % were women and 66.7 % were men). This variable (INCOME CHANGE) is used in analyses in Chapter 7. 3.2.3.b. Respondents were asked about their DEGREE OF SATISFACTION in the question “Are you satisfied with the decision to migrate to Hanoi?” which offered six categories for respondents to choose from. They were coded as: 1 = very satisfied, 2 = satisfied, 3 = not satisfied. 4 = don’t know, 5 = disappointed, 6 = very disappointed. The responses for the category “very satisfied” comprised 30.9 117 percent of the total sample. and, those for the category “satisfied” was 64.7 percent. The other categories comprised a very small proportion (2.4 %, within which 2.0 % were for the “not satisfied”. .0 “/0 for “don’t know”, and .4 % for the “disappointed” and “very disappointed”). which I combined into one new category, not satisfied. Hence, I recoded them as: l = very satisfied (previously coded as 1); 2 = satisfied (previously coded as 2); 3 = not satisfied (previously coded as 3.5. and 6). The responses for the very satisfied category were 31.6 percent (55.4 % women and 44.6 % men); the responses for the satisfied category were 66.0 percent (51.5 % women and 48.4 % men); and the responses for the not satisfied category were 2.4 percent (42.1 % women and 57.9 % men). This recoded variable is used for analyses in Chapter 7. 4. Analysis strategy In Chapter 4, I use descriptive methods to portray the contemporary socio- economic context of mi gration, using data from the national census and statistical yearbooks of 1986-96. Then, to assess the representativeness of my sample, I use frequency distributions and cross tabulations derived from the MOLISA data set to compare the sample to Vietnam’s national population. My strategy of analyses is threefold. First, Chi-Square, ANOVA or Correlation analyses are used to describe and evaluate the relationships between the dependent variables and four sets of nine independent variables: demographic attributes and financial, human. and social capital (depending on types of variables, ordinal or nominal). These tests are run separately for female and male migrants in order to compare and contrast results. Second, Linear or Logistic Regression analyses are used (depending on whether the dependent variable is ordinal or nominal) to investigate whether or not these dependent variables have 118 relationships with any of eight independent variables individually, controlling for gender; these analyses also measure the strength of the relationships between dependent and independent variables. That is, these regressions are run separately with three variables: Gender. one dependent variable, and one independent variable. Finally, to examine which of the independent variables are the best predictors for the dependent variables, I use Linear and Logistic Regressions with a stepwise method. Specifically, in Chapter 5, first, Chi—Square tests are used to examine the relations between Reasons for Migration and Decision-makers of Migration and six nominal variables (Gender, Marital Status, Occupation, Employment Status, Social Connections, and Type of Relations); ANOVA tests are used to examine the relations between Reasons for Migration and Decision-makers of Migration and three ordinal (continuous) variables (Age, Income, and Education). Then, second, Logistic Regression analyses are used to measure the strength of the relations between Reasons for Migration and Decision- makers of Migration and each of eight independent variable, controlling for Gender. Finally, third, Logistic Regression analysis is used to determine the best predictors of Reasons for Migration and Decision-makers of Migration. In Chapter 6, first, Chi-Square Tests are used to examine the relations between Type of Migration and six nominal variables (Gender, Marital Status, Occupation, Employment Status, Social Connections, and Type of Relations); ANOVA tests are used to examine the relations between Type of Migration and three ordinal (continuous) variables (Age, Income, and Education). For Distance of Migration, ANOVA tests are used to examine the relations between Distance of Migration and six nominal variables (Gender, Marital Status, Occupation, Employment Status, Social Connections, and Type 119 of Relations). Correlation tests are used to examine the relations between Distance of Migration and three ordinal (continuous) variables (Age. Income, and Education). Second, Logistic Regression analyses are used to measure the strength of the relations between Type of Migration and each of eight independent variable, controlling for Gender. while Linear Regression is used for Distance of Migration. Finally, third, Logistic Regression analysis is used to find out the best predictors of Type of Migration while Linear Regression is used for Distance of Migration. In Chapter 7, ANOVA tests are used to examine the relations between Economic Improvement and Degree of Satisfaction and six nominal variables (Gender, Marital Status. Occupation, Employment Status, Social Connections, and Type of Relations). Correlation tests are used to examine the relations between Economic Improvement and Degree of Satisfaction and three ordinal (continuos) variables (Age, Income, and Education). Next, Linear Regression analyses are used to measure the strength of the relations between Economic Improvement and each of eight independent variables, controlling for Gender. Finally, Linear Regression analysis is used to find out the best predictors of Economic Improvement and Degree of Satisfaction. 120 CHAPTER 4: PORTRAYAL OF URBANWARD MIGRANTS IN THE TRANSITION PERIOD IN VIETNAM Introduction Migration plays an important role in stimulating economic changes in a nation. Economic changes also strongly influence migration by shaping its characteristics, patterns. and volume. C learly, migration balances and regulates the needs and demands of the labor force as well as the standard of living between regions (Morrison 1977, ‘ Lewis 1982, Massey 1988, Cohen 1996). Empirically and theoretically, an urbanward trend tends to be prevalent in developing countries because urban areas are usually more advanced than rural ones; the more prosperous economic opportunities in earning and employment in urban areas attract rural people (Du Toit 1990, Weeks 1996, Dang et al. 1997). Earning differentials are the primary factor that pulls people to urban areas despite the fact that the proportion of unemployment in urban areas is high (Sjaastad 1962, Todaro 1976, Morrison 1977, Cohen 1996). The migration literature on developing countries shows that migrants there have tended to be young, single men in their twenties with higher education levels and work skills than stayers (Lee 1966, Lloyd 1979, Todaro 1976). Nevertheless, over the past two decades, female migration has tended to increase, especially in an urbanward trend. Female migrants also have tended to be young, in their twenties, single, highly educated and highly skilled as compared to the female national population (Khoo et al. 1984, Chant 1992a, UN 1993, Bilsborrow et al. 1993, Hugo 1993). Only a few studies on voluntary migration have been done in Vietnam (e.g., CPLS 1996, Tana 1996, Phan Quynh Nga 1998, Doan Mau Diep et al. 1996, 1995. Than Van Lien 1997. IS-MHS97 1998, Bach Van Bay 1996). Hence, the characteristics of urbanward migration there have not been fully explored. What is known, however, is that urbanward Vietnamese migrants tend to be young, single men and women who have higher education levels than those of the national population (Tana 1996, CPLS 1996). The goal of this chapter is to describe the characteristics of Vietnamese urbanward migrants in the transition period. This chapter will examine the differences between the national population in general and movers included in this study’s sample. To investigate migrants" characteristics. it is necessary to understand the socio-economic context that fostered their decision to move. Thus, features of migrants’ places of origin will be discussed in the first part of this chapter. Second, only certain people migrated, so the migrants’ characteristics before their move to Hanoi will be clarified in comparison with those of the national population in the second part of this chapter. Finally, a summary of the migrants’ characteristics will be given at the end of this chapter. A. General features of migrants’ places of origin 1. General nictare of the Vietnamese situation in the 19808 and 1990s At its independence in 1954. North Vietnam was one of the poorest and least developed countries in Asia. Less than 6 percent of the population lived in urban areas. (Vietnam is as urbanized as Bangladesh and about one third as urbanized as China [Bryant 1998]). In 1976, one year after national reunion, the Vietnamese urban population had grown to 20.6 percent (see Table 1.1 in Chapter 1). In the late 19703, the Vietnamese economy was marked by an unstable combination of central planning and 122 semi-legal marketsI (Bryant 1998). During the 19805, the economic institutions of central planning collapsed while the country’s political institutions underwent little change. Specifically, since 1981, the government authorized state sector enterprises to trade directly with each other and permitted collectives to subcontract parts of production processes to individual households (Fforde et al. 1996). Resources continued to shift to the private sector as people took advantage of legal and illegal opportunities to make money. The income state-sector workers earned from sideline activities grew from one- fifth of household income in 1979 to about one-half in the mid-19803 (Fford et al. 1996192). In the innovation period. i.e.. transition to a market economy, since the 19805, a number ofchanges in institutional policies encouraged an expansion of free markets, diverse exchanges of commodities within and between regions, and population movement. First. in 1988, land was de-collectivized under Contract 10 (khoan 10) or the household contract (khoan ho). This policy treated the family as the basic unit of the rural economy and distributed land to individual households (usually heads of households) based on the number of family members. Implementation of this policy encouraged the rural population to innovate new forms of agrarian production and diverse production relationships to maximize their households’ income from their plots of land. For instance, the rural population could develop enterprises such as fishery, handicraft, brick manufacturing. food processing. and petty trades. or engage in inter-provincial ' Semi-legal markets were places where people could sell. buy, or exchange food and groceries or goods that were provided by the government or self-produced. They were semi-legal because private trade generally was not allowed under the socialist system, but due to the diverse needs of people the equal distribution from the government could not satisfy, people were allowed to trade some certain products. trade. Due to the rational use of labor, the labor surplus in rural areas increased, and, consequently. the labor force sought employment and earning opportunities in other places such as urban areas. Beginning in the late 19803, due to the privatization of production, economic activities, and prices, entrepreneurship rapidly spread from the cities to the countryside; different economic activities were developed with the introduction of market reform. Rural population movement in search of economic Opportunities tended to occur in different directions: within regions and between regions, from north to south or central areas. from rural to urban areas, or within rural areas. Second, the compulsory residential registration of the central economy, which was vital to all people in both rural and urban areas to obtain subsidized food, goods or other welfare provisions, became unimportant with the removal of the subsidy system in the late 19803. In the innovation period, the massive decentralization of power resulted in a relaxation of the restrictions on population movement. which, in turn, encouraged population movement between and within regions. Third, the removal of restrictions on the development of the private sector led to the development of private transportation facilities that facilitated population movement. Thus, people were provided the political and physical means to move. Due to these changes, rural people could migrate voluntarily to improve their lives. However, it was one of the most difficult and risky decisions due to the great uncertainty of migration during the transition period. 2. Places of origia Vietnam’s population densities in 1986, except in Hanoi and Ho Chi Minh City, varied from 23 persons per square kilometer in Lai Chau province in the Northern 124 Uplands to 1,1 12 persons per square kilometer in Thai Binh province in the Red River Delta (see Table 4.1. column 1). The population ofVietnam was mainly concentrated in the river deltas in the north and the south because of their fertile land. The highlands and coastal areas were less densely populated (36 persons per square kilometer for the Central Highland, 161 for the Central Coast) (GSO 1987z6). The Red River (where Hanoi is located with 1,397 persons per square kilometer) and Mekong Deltas, and the Southeast region (where Ho Chi Minh City is located with 1.847 persons per square kilometer) were the most crowded areas of the nation. In 1989, these regions comprised only 24.3 percent of the national area but contained 56.2 percent of the population, while the Northern Uplands and Central Highlands constituted 46.5 percent of the national area with only 19.8 percent ofthe population (VPC 199laz31). Migrants in this study mainly came from the densely populated provinces in the Red River Delta, the Northern Upland, and the Northern Central Coast regions. That is, migrants in Hanoi mainly came from the northern part of a country. Nearly 80 percent of migrants came from the provinces of Ha Bac, Hoa Binh, Vinh Phu (the Northern Uplands region), and Hai Hung, Ha Nam Ninh, Thai Binh, and Hai Phong port (the Red River Delta region). within a distance of 100 kilometers of Hanoi (see Chapter 3). The population of these provinces mainly engages in rice production, which provides work for only three-fourths of a year. The rest of the year (one fourth) is the agricultural slack season. Due to lack of land and underemployment, farmers in these provinces, while producing rice, participated in other economic activities such as cereal production, petty trading, wine production, poultry husbandry, and fishery. People in Hoa Binh and Vinh Phu tend to work on green tea plantations and engage in its production, while people in 125 Ha Bac specialize in producing a silk textile. People in Ha Nam Ninh and Hai Hung specialize in the cultivation of litchis. People in Hai Hung also specialize in the production of pottery and porcelain. From the data in Table 4.1, it is clear that in 1986 the pOpulation of the places of origin of migrants had features different from those of Hanoi and the nation. More specifically. the proportions of urban population in these places of origin were rather low (all had less than 10 0/o. except Hai Phong with 31.7 %) compared to the national urban population in 1986 (19.3 %) and that of Hanoi (38.6 %) (see Table 4.1, column 2). This also means that these sending areas of migration had lower levels of urbanization, i.e., they had higher proportions of agricultural population (all had more than 90 % rural population) than that of nation (80.7 %) and Hanoi (61.4 %) (see Table 4.1, column 3). A comparison of the proportion of labor in different occupations will help to highlight the degree of urbanization or development of these places of origin. Those places which have a high proportion of industrial laborers demonstrate a high level of urbanization (Goldscheider 1977, Weeks 1996). i.e., they are more developed than those with a lower proportion of industrial laborers. As can be seen in Table 4.1, column 5, the proportion of industrial laborers in the migrants’ places of origin was lower (all of them had less than 5.0 %, except Hai Phong with 6.8 %) than that of Hanoi (7.5 %). The fact that most of these places of origin (except Hai Phong) had a higher proportion of agricultural laborers (ranging from 24.9 % for Ha Nam Ninh to 34.1 % for Ha Bac) than did Hanoi (18.5 %) (see Table 4.1, column 4) indicates that people in these areas tended to be more vulnerable in terms of economic status than those in areas with a higher level of industrialization. This was because of the nature of agricultural 126 .02: E 52o 05w 2 8mm sausages :o 03.3 a-» £5.28 .50 3088 .0w2 5 Sap—22> mo 5:233 2:5 2: :o 083 mm o\° - .3 2528 E t 2 332 $302 2:: 30.0 n Ex - .cocflsaoa _99 05 00 ..\o 3% a: 002: 52o Ba 2 wows a 803 98 08.83 on? cots—smog .0w2 E - ”802 $0 3828 08 N22 .v 5:28 no... 3 _d .m-m 3828 no.“ 2d ._ 5:28 .8 0d .53. Omo Hoocsom me 9Q. o._ 0.2 Wm mam how md— _a. 09295.. 0.5 VS. ho mdm 9m m.0~ v.3 0.0 mm _ 55H 2&2 + Nd ”R Yo odm 2V Q3. 0.3 06 0mm mom flash + _0_ 6.80 _a.—:50 595.52 mém Own 2 0.3. mm oém Ndo wd «mm :52 :52 mm + 04cm «.05 0.0 a.0m m6 5mm 58 m0 m. 2 £5 22:. + 0.3 <05 m._ a. _ m Wm flow 0:8 vs o0o was: _a.: + 0.3 0.2. We mam EN m._m odo 3: 0%.. :25 com a: + Nam wit ad w.0m w.0 odm m.w0 h. _ m $9 98.3 BE + N3. w.m0 wd o.0~ m.» 3: v.20 0.3 3m; 62 a: + m2. axon ..ozm com mdm 5%. Yo _.0m o.m :1.” ”mo N0 wmv cum «I + msm ”Q. 0.o Ndm mgr. 50.x” _Na as _wm BE 55> + 3: 3 85...: ...—2:32 5 a: E 80 5 3 5 E :0 £0 £0 £0 £0 £0 £0 $0 @5333 Eu gov a £2002 £2002 £8002 £8002 c2538 coca—smog 55:00 30565 onanocD SEQ—9:...“ 0835 83m 3520:. EBEutw< _Esm SEED :ocflaom \ Wigwam .092 0:0 0mm: 5 _ocm: E 8:822 .00 EwtO 00 385 no BEoEHm EofionEm can cocflsmom “:V 203. 127 production and the introduction of market mechanisms. For example, farmers could freely produce and sell their products in the markets, but if their harvests were not good or their products did not satisfy consumers, they could not get help from the government to cover their loss. Moreover, the government encouraged farmers to increase their productivity or to specialize in the production of certain fruits, cereals, or industrial crops by providing them with new methods of planting, fertilizers, and newly improved seeds, but farmers might not be able to sell their products due to a lack of transportation or changes in market demand. Therefore, the fact that people from places with a high proportion of agricultural laborers and relatively vulnerable economic conditions tended to migrate to more developed areas in order to improve their economic status is understandable. This is consistent with the assumption in the migration literature that people move from less developed to more developed areas (Weeks 1996, Torado 1976, Lewis 1982, Du Toit 1990). With respect to the proportion of employment, from Table 4.1, it is clear that in 1996, these sending provinces of migration tended to have a higher proportion of employment (more than 70%) and a lower proportion of unemployment (less than 20 %) than those of Hanoi (65.8 % employed people and 34.2 % unemployed). Despite the fact that more people in rural areas were employed than in urban areas, rural people continued to move to urban areas. This can be explained by the “pull” factor of urban areas, which were seen to offer economic opportunities (higher earnings). Until the middle of the 19803, people within the agricultural sector were mainly involved in collectives, which 128 was a part of the state sectorl that provided subsidy assistance for its employees such as goods, food stamps. free education. social insurance. health care, and transportation. Therefore. workers in the state sector had more advantages than those working in the infomial sector. Nevertheless. within the state sector, people working in the agricultural sector tended to be less advantaged than those working in other branches. Their less advantaged situation resulted from features of the collectives of the socialist system. Farmers could receive the same benefits as did workers (health care, food stamps, goods, free education, social insurance). but they did not have stable salaries and pensions paid by government. Their earnings depended greatly on harvests and their actual working hours (as compared to workers who automatically receive a fixed wage for an eight-hour working day). People in the informal sector did not have any social insurance or pensions; except free education, they had less access to free public services (health care, transportation, electricity. water), unless they paid for the same services; further, they did not have any opportunity in being promoted up the power ladder. So, compared to farmers and those working in the state sector, people in the informal sector were disadvantaged. Under the socialist system, people primarily worked in the state sector and this can be seen in the employees’ structure of Vietnam in the 19803. Specifically, in 1986 the proportion of employees in the state sector was 31.6 percent of the total Vietnamese ' As explained in Chapter 3. in Vietnam. the formal sector was identical to the state sector before the renovation period. population (see Table 4.1, column 6), while that in the informal sector was 1.0 percent (see Table 4.1. column 7). In the transition period, collectives are shifting into the informal sector.2 With the loss of subsidies from the government, the agricultural sector’s advantages over the informal sector were eliminated. The advantage of the formal sector still remains. however. People who work in the state sector are paid by the government; they are still provided some benefits (health insurance, pension, sick leave, maternal leave) while farmers like people in the informal sector do not receive any above listed benefits from the government. There was a great difference in the availability of high prestige occupations as well as education facilities between rural and urban regions, which can be seen through the great difference in types of employment available, and through the number of educational institutions located in cities and in provinces. It is clear that within types of employment (wage earners, farm or non-farm self- employment), wage earners had the highest prestige since these people were characterized as workers in the state sector, who had more advantages in terms of socio-economic privileges and possibility of upward mobility than those in the private sector (self- employment). From the data in Table 4.2, we can see that in1992-93, urban areas tended to have a much higher proportion of wage earners (32.7 %) than did rural areas (4.2 %). 3 Land is not considered private property that is owned by people. but they can sell, buy, or exchange their using rights to their piece of land. Farmers treat the land as if it were their own, they pay for the fee of using that land for the government and behave like private entrepreneurs. 130 Furthermore, more than half (55.1 %) of the population in rural areas was involved in self-employment. while only about one-third (36.1 %) in urban areas was involved in self-employment. Thus. it is clear that people in rural areas were more likely to be involved in low prestige occupations while people in urban areas were more likely to be involved in high prestige occupations. These differences also highlight the degree of development of urban areas over that of rural areas. Again. this helps confirm the assumption of migration theories about population movement from less developed to more developed areas (Todaro 1976. Lewis 1982, Du Toit 1990, Bilsborrow et al. 1977. Cohen 1996). Table 4.2: Distribution of Types of Work ofVietnamese Population in 1992-93 Type of work Rural Urban Average (%) (%) (%) Wage earners 4.23 32.73 19.48 Farm self-employment 55.1 12.34 33.72 Non-farm self-employment 4.95 36.1 20.53 Farm self-employment and wage earners 16.27 6.27 1 1.27 Non-farm self-employment and wage 0.95 3.81 2.38 earners Farm and non-farm self-employment 15.27 7.18 1 1.27 All three types 3.23 1.59 2.41 Total 81.31 18.69 100* Source: VLSS 19951131. Note * : The total may not equal 100 due to rounding of numbers. Hanoi (25.7 %) and Ho Chi Minh City (15.6 %). as the largest cities in Vietnam, contained about 41.3 percent of all universities and colleges in 1994 (see Table 4.3). From the data in Table 4.3. we can see that Hanoi was home to 41.2 percent3 Of the total universities and colleges within the northern parts of the country and 25.7 percent of universities and colleges for the whole nation, while Ho Chi Minh City contained 42.2 3 The result of25.7 % over the total of(l 1.0 % + 41.28 % + 101%) 131 percent“1 of the southern universities and colleges and 15.6 percent of those of the nation. Compared to these two cities, other provinces had a very modest number of universities and colleges. That is, the sending areas had very few facilities in which their population could improve their education. and, therefore, rural people who wanted to pursue higher education tended to be involved in urbanward migration. In sum. the places of origin of migrants were not the poorest nor did they have the least advantages compared to other areas of Vietnam. But they were less developed areas compared to Hanoi in terms of their urbanization levels. economic opportunities, types of production. and educational opportunities. Table 4.3: Distribution of University and Colleges by Regions of Vietnam in 1994 Region or cities Number of universities and colleges % Northern Uplands 12 11.0 Red River Delta 45 41.28 + Hanoi 28 25.7 Northern Central 1 1 10.1 Central Coast 8 7.0 Central Highlands 5 4.6 Southeast 20 18.35 + Ho Chi Minh city 17 15.6 Mekong River Delta 8 7.0 Total 109 100 Source: GSO 1995:352. B. Migrants’ attributes _l_._§g Migrants in this study had a slightly higher sex ratio (52.9 % women to 47.1 % men) than that of the nation: 51.1 percent women to men in 1989 (VPC 1991a), and higher than that of Hanoi, 51.2 percent women to men in 1990 (HSO 1998). This higher ‘ The result of 15.6 % over the total of(7.0 % + 4.6 % + 18.35 % + 7.0 %) proportion of female than male migrants in urbanward migration in Hanoi might be explained by the relative short distance of the move from their origins to Hanoi, which reduced the risk they face in the migration process due to their lower educational levels compared to that of men (see Chapter 3). This fact also might be rooted in the higher percentage of female migrants having social connections in Hanoi before moving than that of men (53.6 % and 46.4 %, respectively) (see Chapter 3). which helped ensure women’s success in moving to Hanoi. This high proportion of female migration might also result from the fact that Vietnamese women participated in economic activities to support their households as actively as did men. They were specifically likely to participate if they were poor, their households lacked men, the men in their households lacked employment, or if their households faced great uncertainty such as unsuccessful harvests. Women’s active participation in migration was clearly seen during the transition period with many changes in economic opportunities (employment, earnings) resulting from policy adjustments. For example, due to the privatization of production, many state enterprises restructured their personnel and many workers were laid off; women were more likely (60.0 % to 65.0 %) than men to be fired (see Chapter 1) (Tran Thi Que 1995). Ma: In 1989, Vietnam had a young age structure with a mean age of 24.5, a median age of 20.2, of which the mean for women was 21.4 while that for men was 19.1. That is. most Vietnamese population were clustered in age 20. Migrants in this study had a mean age of 31 and median age of 25. i.e., most migrants were in age 25. The difference between the median ages of the people in this study and in the nation was a result of the large percentage of the Vietnamese population aged 14 and under (39.5 %) (VPC l991a:l 1). while the sample ofthis study included people only 13 years and older. From the data in Table 4.4, we see that in 1989, 11.7 percent of the total Vietnamese population was under age 15 (10-14 age group), while people over 60 years of age made up 7.0 percent only. The proportion of migrants under 15 (13-14 age group) comprised 3.6 percent while 4.8 percent of migrants were over 60. This age structure tended to be consistent with that of the nation, even though the proportion of the population under 15 was different because of the absence of data in this study for the 10- 12 age group. Table 4. 4: Age Distribution of Vietnamese Population in 1989 and of Migrants of This Study 10-14 * 15-34** 35-60 61+ (%) (%) (%) (%) 1996 female 5.6 61.1 27.4 6.3 MOLISA male 1.9 68.3 26.6 3.2 survey total 3.6 64.4 27.1 4.8 Vietnam female 11.0 36.3 16.8 8.2 male 12.4 35.8 18.8 6.2 total 11.7 36.05 17.0 7.0 Source :VPC 1991a:l 1. Note *: This age group is applied for Vietnamese population, for the 1996 MOLISA survey it is 13-14 only. **: In this group, those who were in between 20-29 in the 1996 MOLISA survey comprised 42.3 % (45.5 % male and 41.3 0/o female), while in Vietnam they comprised 18.1 % (17.8 % male and 18.7 % female) Nevertheless, there were differences between the proportion of migrants and that of the general Vietnamese population who were between 15 to 60 years of age. As can be seen in Table 4.4, the proportion of migrants ages 15-34 was 64.4 percent while that of the nation was slightly more than half of that, 36.05 percent in 1989. In this study, then, most of migrants were much younger than their proportion in the national population suggests. They mainly migrated when they were ages 20-29 (42.3 %), and of these migrants, 41.3 percent were women and 45.5 percent were men, while this age group within the Vietnamese population was 18.1 percent only. In Vietnam, people 20 to 29 years of age were traditionally considered the most mentally acute and physically strong in comparison to all others. People of this age were considered mature enough to be independent and to build their careers. The advantages of young age were especially important for people who sought a better life in a new place (i.e.. through migration), which presented a great challenge for people attempting to fit into a new environment. Migration for most people was seen as a way to start a new life. Moreover. the fact that more people in the age groups 20-29 than in other age groups migrated might result from the Vietnamese young age structure mentioned above (median age = 20.2). In comparison with the national population distribution of ages, it is clear from the data in Table 4.4 that migrants in this study mainly clustered in the ages of 15-34 (64.4 %) while only 36.0 percent of the national population made up these ages. Implicitly, migrants had a much younger age structure than that of the nation. In other words, they were younger than the Vietnamese population in general. This result tends to be consistent with findings in the migration literature which show that migrants are mainly young or in their twenties (Massey et al. 1993, Cohen 1996, Du Toit 1990, Todaro 1 976). The mean age for female migrants was 32 years with a standard deviation of 16 While that of men was 30 years with a standard deviation of 13 (see Table 4.5). This means that women mainly migrated between the ages of 16 to 48, while men mainly migrated between the ages of 17 to 43. Thus, women migrated at earlier, as well as later ages. than did men. This is illustrated in the large difference between the proportions of 135 male and female migrants at younger and older ages (under 15 and over 60). There were more female migrants ages 13-14 (5.6 %) than male migrants (1.9 %). Similarly, 6.3 percent of women migrated at age 61 and older, while only 3.2 percent of male migrants were these ages. Women were involved in migration longer than men. This result tends to be inconsistent with that of the Vietnamese population census survey of 1989. Women were reported to be involved in different economic activities for a shorter duration (mean = 43.5 months) than were men (mean = 44.9) (VPC 1991az64). Although women started working one year earlier than did men (age 15.1 for women and Table 4.5: Distribution of Age and Gender of Migrants Age group Female Male Total (%) (%) (%) 13-14 5.6 1.9 3.6 15-18 9.4 12.4 11.0 19-55 75.6 80.5 77.9 56-60 3.3 2.1 2.7 61+ 6.3 3.2 4.8 Mean 32 30 3] Std. D 16 13 14.5 Median 24.8 25.2 25 Source: 1996 MOLISA survey. 16.1 for men), they stopped working 2.4 years earlier than did men (58.6 for women and 61.0 for men). This difference can be explained by the fact that women were reported to leave School earlier (mean age = 15.4) than men (mean age = 16.1) and hence, started working before men (VPC 1991A253). Women’s active participation in the labor force reached the highest proportion at ages 20-29 (89 %), and then slowly declined to 69 percent at ages 50-54. According to the Vietnamese Labor Code, women have to leave the work force at age 55 while men can work until age 60. Men had the highest proportion 136 participating in labor at ages 25-29 (97.4 %). and their participation remained high at ages 55-59 (71 %). Women also tended to work for a shorter time span than men. This result might be understood as the inclusion in the census of women involved only in paid employment rather than in both paid employment and unpaid employment or self- employment. The results of the national census might have been similar to this study if it had included this type of employment. In sum. migrants were generally younger than the Vietnamese population. Both men and women mainly migrated when they were 20-29 years of age; however. women migrated at younger and older ages than men. 3.Ed1_ication level As mentioned in Chapter 1, since the establishment of the socialist system, the proportion of the Vietnamese population that is literate is high despite the great proportion of people living in poverty. According to the results of the Vietnam Living Standard Survey in 1992-93. about 86.6 percent of the population over 10 years old was literate. Of those who were literate, 91.4 percent were men and 82.31 percent were women. Those living in urban areas were more likely to be literate (93.33 %) than those in rural areas (84.76 %) (SPC 1994). The fact that the rural population had lower educational levels than the urban population might be explained by the existence of agriculture in which traditional ways of farming were practiced. Formal education is not necessary in traditional agriculture because farming methods and knowledge are readily passed down from parents to children. Thus, people could do farming without obtaining any degree of education, or in 137 other word. their low educational levels were not a matter for farming as ways of earning a living. The low level of education of the rural population also resulted from the scarcity of educational institutions in these areas (see part A in this chapter). These low educational levels are closely linked to the great expense of pursuing higher education in urban areas. and. consequently. the limited opportunities for the rural population to obtain such education. Ironically. as can be seen in Table 4.6, migrants had a higher education level (mean = 10.5) than did the national population (mean = 9.5). This difference was particularly striking at the university and college levels (47.2 % of migrants compared to 1.6 % ofthe general population). Table 4.6: Distribution of Education of Migrants and Vietnamese Population in 1989 by Gender and Areas Education Rural areas Urban areas Female Male Total levels (%) (%) (%) (%) (%) 1996 Elementary 13.0 10.2 15.3 8.5 12.0 MOLISA Secondary 33.9 I 1.4 30.5 22.8 26.8 survey High sch. 17.0 7.7 14.3 13.8 14.0 Univ. + Colle 36.1 70.7 40.0 55.0 47.2 Mean (years) 9. 7 11.29 IO II [0.5 Vietnam Elementary 79.4 71.2 74.2 79.5 76.8 Secondary 4.2 12.59 8.2 8.3 6.9 High sch. 1.4 5.4 2.9 2.5 2.7 Univ. + Colle 0.7 4.5 1.1 2.1 1.6 Mean (years) 9.2 I 1.6 9.3 9. 8 9.5 Source: VPC 1991a:53, 55. The high educational level of migrants compared to that of the national population might result from the characteristics of migration as a selective process that help reduce the risk and maximize the success of moving. because, as noted in the migration literature, education is positively associated with economic success (Massey et a1. 1993. Vecoli 1998, Sander et al. 1996, Hunt et al. 1985, Robinson 1993). This advantage could 138 help migrants compete with city dwellers in urban labor markets and overcome the disadvantages created by regional differences. The result again supports the finding from the migration literature that migrants usually are highly educated (Massey et al. 1993, Todaro 1976, Cohen 1996. Lewis 1982). 4.Marital status Among the Vietnamese population in 1989, 29.55 percent (32.4 % men and 26.7 % women) was single, and 61.8 percent (64.7 % men and 58.9 % women) was married (see Table 4.7). The proportion of widowed persons was 6.9 percent (2.2 % for men and 1 1.6 % for women), that of divorced persons was 0.6 percent (0.3 % for men and 0.9 % for women), and 0.7 percent was separated (0.4 % for men and 1 % for women) (VPC 1991az78). A clear difference between men and women was found in the proportion of the widowed (2.2 % for men compared to 1 1.6 % for women). The fact that a lower proportion of men were widowed can be explained by the following reasons. First, men had a higher mortality rate and died at earlier ages than women. Women lived longer and thus. they were more likely to be widowed than men. Second, the possibility for men to re-marry was higher than that of women due to age differences in marriage. Men were usually older than their female partners and it was easier to find a partner of a younger age than to find a partner of an older age. Third, the higher proportion of widowed women (over 50) was a consequence of war, in which there was a high death rate among men. The data in Table 4.7 clearly show that both married and single people were involved in moving to Hanoi. One half of the migrants (50.9%) were single (46.2 % women and 56.2 % men) and slightly fewer (44.3 %) were married (45.5 % women and 43 % men). In comparison with the proportion of single or married people in Vietnam, it is clear that migrants were almost two times as likely to be single (50.9 %) compared to the total Vietnamese population (29.5 %). while married migrants made up about two fifths (44.3 %). much less than the married population of Vietnam (61.8 %). This great difference might be seen as a reflection of the active participation of single people in the migration process as a way to improve their economic status and that of their family or to start new lives. Singleness may help ease resettlement, making it easier to deal with uncertainty because of lesser responsibilities within the family. Furthermore, labor markets tend to prefer single people in order to reduce the cost spent on social welfare for employees’ family members. Regarding gender differences in the marital status of migrants, as can be seen in Table 4.7. married female and male migrants made up equal proportions (45.5 % and 43.1 %. respectively. i.e.. ratio = 1.1) while married women in the national population made up 58.9 percent. less than married men (64.7%) (i.e., ratio = 0.9). Table 4.7: Distribution of Marital Status by Gender of Migrants and Vietnamese Population in 1989 Gender Single Married Widowed Separated (%) (%) (%) (%) Vietnam female 26.7 58.9 1 1.6 1.9 male 32.4 64.7 2.2 0.7 total 29.55 61.8 6.9 1.3 1996 MOLISA female 46.2 45.5 6.8 1.4 survey male 56.2 43.1 0.5 0.3 total 50.9 44.3 3.9 0.9 Source: VPC 199laz78. This means that married women tended to be more active in migration than did married men. This difference might be understood as the increasing active participation of married women in migration in order to support their households’ betterment. Moreover. 140 there was a consistency between the proportion of single female and male migrants (46.2 % and 56.2 %, respectively). and the single female and male population of Vietnam (26.7 % and 32.4 %, respectively). i.e., there were always more single males than single females in both data sets. Therefore, both single female and male actively participated in migration. The proportion of separated migrants was 0.9 percent (1.4 % women and 0.3 % men) and that of the widowed was 3.9 percent (6.8 % women and 0.5 % men) (see Table 4.7). The proportions of widowed and separated female migrants (6.8 % and 1.4 %, respectively) were higher than those of male migrants (0.5 % and 0.3 %, respectively), and were similar to those of the Vietnamese population. The proportion of female widowed migrants was especially greater than that of men because of the high death rates of men during the war. This difference also can be explained by the reasons mentioned above. To summarize, migrants were more likely to be single than married; they made up a higher proportion of singles than that of the Vietnamese population. Married women tended to participate in migration more than married men. The results, then, only partly support findings from the migration literature, i.e., that migrants mainly are single (Du Toit 1990. Todaro 1976. Lewis 1982, Bilsborrow et al. 1993). Both single and married people in Vietnam tended to be involved in migration. especially married women. 5. Employment staLaa From the data in Table 4.8, we can see that the proportion of unemployed in Vietnam in 1989 was 5.8 percent of the total population aged 15 and older (5.4 % for women and 6.2 % for men). while the proportion of those 15 years and older who were 141 employed was 74.2 percent. These proportions differed from those for migrants in this study. Migrants were almost two times as likely to be unemployed prior to moving (10.6 %) than their national counterparts (5.8 %), and half as many (36.9 %) were employed compared to the nation (74.2 %). This high proportion of migrants who were unemployed can be explained by the much higher proportion of those in school (32.8%) than nationally (4.2 %) (see Table 4.8). This means that people in this study migrated at young ages, and, at that time, they may have just finished school. did not have a job yet, or planned to seek higher education, which was available only in the cities. Therefore, the proportion of migrants who were unemployed was high. This high proportion might also result from the restructuring of personnel when the market economy was introduced in Vietnam in the 19803 and 19903. The number of employees in the state sector decreased annually (e.g., 3859.2 thousand people in 1985, 3421.4 thousand people in 1990, and 2933 thousand people in 1994) (GSO 1995:32), while the population increased, on average, about one million people per year during the same time period. Table 4.8: Distribution of Employment Status by Gender of Migrants and Vietnamese Population in 1989 Employment Unemployment Housework Schooling Other Total (%) (%) (%) (%) (%) (%) Vietnam female 71.3 5.4 10.92 3.3 6.06 100* male 77.5 6.2 1.44 5.1 13.45 100* total 74.2 5.8 6.18 4.2 9.74 100* 1996 female 34.5 9.4 7.5 31.7 16.9 100* MOLISA male 39.6 11.9 1.3 34.0 13.2 100* survey total 36.9 10.6 4.6 32.8 15.1 100* Source: VPC 1991a:143. * The total may not equal 100 due to rounding of numbers. Noticeably, as was seen in Table 4.1, people usually moved from provinces that had higher proportions of employment than that of Hanoi. However, Hanoi was one of the areas that had the highest population density (1,397 pers/km2) and the highest proportion of unemployed (34.2 %). Therefore, it was not necessary for migrants to consider the proportion of unemployment ofHanoi in making decisions to move, but to consider economic opportunities. as noticed in the migration literature (Todaro 1976, Cohen 1996). Furthermore, in 1989. the proportion of unemployment in urban areas was 13.2 percent (13.2 % for men and 13.1 % for women) while that of rural areas was four percent (4.4 % for men and 3.7 % for women) (VPC 1991a:143). Again, as explained above. the proportion of the unemployed in urban areas was always higher than that in rural areas. but the unemployed population still moved to urban areas because of the greater economic opportunities in urban areas as compared to rural areas. The data in Table 4.8 show that 4.6 percent of migrants identified their employment status as housework prior to moving to Hanoi. For the nation as a whole. slightly more people (6.18 %) identified their job as housework. It is clear that fewer migrants identified themselves as such. This means that migrants who did housework were less likely to migrate than those with other employment statuses because they had very few economic opportunities in the labor market due to their low human capital levels. In other words, migrants had better employment status compared to that of the national population. They were selected people; they were better-off and this can be understood as a migration strategy adopted to encourage success in moving. As for the “other” category of employment status (i.e., invalid, disabled or pensioned people). migrants in this study made up a slightly higher proportion than found 143 in the nation (15.1 % compared to 9.74 %). The fact that these people were involved in migration might be understood as their motivation to migrate for family reasons or rather their movement with others, i.e., they could not be left behind. To summarize, migrants had diverse employment statuses. but were mainly employed and/or attending schooling. This diversity might reflect the attempts of migrants to adapt to the changing period. Their high proportion of employment and schooling demonstrated their attempt to succeed in dealing with new market mechanisms. 6.0ccuaations Vietnam is primarily an agricultural country and, in 1989, the proportion of the population engaged in farming or farm-related activities (agricultural sector) was high, 54.7 percent (52.0 % for men and 57.3 % for women) compared to occupations in other sectors such as the state or formal and informal (see Table 4.9). Accordingly, the greatest proportion of migrants in this study worked within the agricultural sector prior to migration, 45.1 percent (59.7 % women and 40.3 % men). However, this proportion which constituted nearly half of the sample was slightly less than that of the nation (45.1 % compared to 54.7 %) while migrants working in the formal sector prior to moving made up a much greater percentage (almost triple) than that of the nation (42.8 % compared to 15.2 %). These differences between migrants and national population working in the agricultural and formal sectors showed that migrants had much better occupational status than did the national population. They made up a lower percentage of those working in the agricultural and informal sector, but much more in the fomial sector. 144 This better occupational status ofmigrants compared to that of the whole population might be seen as a way to maximize the economic returns from migration; migrants who worked in the formal sector before moving may have more opportunity to get jobs in the city than those who worked in agriculture or in the informal sector because of their higher educational levels (means = 1 1.8 for formal sector, 10.2 for informal sector and 8.2 for agriculture) (see Appendix 3.1). Alternatively, the fact that people working in the formal sector moved more than those working in other sectors might also be explained by the consequence of the socialist system which emphasizes the privilege of the formal sector over the informal sector that limited the motivation of these people to move (see Part A, Section 2 of this chapter). Table 4.9: Distribution of Occupation by Gender of Migrants and Vietnamese Population in 1989 Agricultural sector Formal sector Informal sector (%) (%) (%) Vietnam 96 MS Vietnam 96 MS Vietnam 96 MS female 57.3 59.7 14.2 40.9 28.5 41.7 male 52.0 40.3 16.2 59.1 31.9 58.3 total 54.7 45.1 15.2 42.8 30.1 12.1 Source: VPC 1991a:145. Note : 96 MS means the 1996 MOLISA survey. In sum, migrants had much better occupations compared to the national population. They were more likely to work in the state or in the agricultural sector rather than in the informal sector compared to the general population. 7. Income Since the 19903, five different income levels in Vietnam’ have been used for 5 Income levels are . Less than 75 thousand VN$: 4. 301-600 thousand VNS; . 75-150 thousand VNS; 5. More than 600 thousand VN$. . 151-300 thousand VNS; DJ h)— 145 statistical purposes. Because more than 50 percent of the Vietnamese population had incomes below the poverty line“ in 1992. it is not surprising to see the great pr0portion of the population at the lowest level of income. The urban population had a much better economic status than did the rural population. In 1992-93. the monthly income mean was 151.5 thousand dong in urban areas while that in rural areas was only 77.391 thousand dong (see Table 4.10). The large gap between the areas can also be clearly seen in Table 4.10. The urban population had a larger proportion of people at the highest level of income (5.1 %) and the smallest proportion at the lowest income level (22.4 %) compared to those of the rural population (0.7 % and 54.5 %. respectively). Table 4.10 : Distribution of Income by Gender of Migrants and Vietnamese Population inl992-93 Studies Sex <751< 76-150K 151-300K 301-600K >600K mean (%) (%) (%) (%) (%) (K) female 29.7 20.7 36.6 10.3 2.8 186.5 MOLISA male 17.1 23.3 30.8 24.7 4.1 254.3 total 23.4 22 33.7 17.5 3.4 220.4 Vietnam urban 22.43 32.03 30.87 9.44 5.12 151.5 rural 54.46 30.82 10.66 2.36 0.7 77.391 total 48.11 21.86 14.69 3.77 1.58 92.092 Source : SPC 1994:228-229. Note: K : thousand Vietnamese dong However, migrants in this study tended to have a higher income level prior to moving compared to the national average. The data in Table 4.10 show that the proportion of migrants at the lowest level was less than half that of the nation (23.4 % compared to 48.1 %). They also had slightly more than two times the proportion at the highest level of income (3.4 % compared to 1.58 %). The better economic condition of “ See Footnote 1, Section 3.1.2. Chapter 3. 146 migrants also can be seen in the fact that one-third of migrants (33.7 %) were clustered at the third level of income while the national population mainly clustered at the lowest level (48.1 %). In comparison with income levels in urban areas, migrants tended to cluster at the third level, while the general urban population tended to cluster at the first three levels. Migrants had higher income levels prior to moving than did urban citizens in general. This result shows that migrants were much better off than the national population in general. This better off status also can be seen in the income means. Migrants had a mean income of 220.4 thousand dong/month, while that of the country in 1992 was 92.092 thousand dang. Again, this result shows that migrants were not the poorest in their places of origin. a result that confirms findings in the migration literature (Lewis 1982. Cohen 1996). With respect to gender differences in income prior to migration, female migrants were poorer than male migrants with a mean income of 186.5 thousand dong while that of men was 254.3 thousand dong. A smaller proportion of women also were at high income levels (4th & 51h) than their male counterparts. Their poverty relative to men probably reflects the fact that a higher proportion of women was also at the low income levels (lst & 2nd) compared to their male counterparts. No national data for women and men’s income are available; thus. the gender difference in the earnings of migrants with that of the general population cannot be compared. However. we can infer that a similar picture prevails because of the lower income of female-headed households in the general population (435.4 thousand dong) compared to that of male-headed households (465.4 thousand dang) (SPC 19942221). 147 In short, migrants were much better off than the national population. They had higher income levels compared to the average national income level and to the average for urban dwellers. Even though female migrants had much higher income levels compared to the national population, they were poorer than male migrants. C. Summary and discussion Vietnamese urbanward migrants basically came from provinces nearby their destination, Hanoi, rather than from areas a long-distance away, supporting the finding from the migration literature that the place of origin of migrants usually is located near a boundary with an urban area; the closer a place of origin is to an urban area, the more likely it is for people to migrate (Lewis 1982). In addition, provinces nearby Hanoi had higher proportions of people employed with lower income levels than did that city. This result is also consistent with findings from the internal migration literature on developing countries. i.e.. internal migration occurred due to regional differentials in income rather than employment rates (Sjaastad 1962, Brueckner 1990). Put another way, migration occurred more frequently in areas that had greater unequal income distribution than elsewhere (Stark 1984). This finding can be explained according to the neo-economic approach of migration. Specifically, migrants usually aim to maximize their earnings, and regional income differentials will be the greatest stimulus for people to choose a destination. Even though a higher proportion of migrants was unemployed than in the general population, they had more human capital than did the Vietnamese in general. That is, they had higher educational levels, better occupations, and higher income levels than did others in Vietnam. Migrants were mainly young, single and married people who 148 migrated at ages 20-29. Women migrated at both younger and older ages than men. Migrants were more likely to have high prestige occupations compared to the national population. but women. compared to men. tended to cluster in the lower prestige occupations. e.g.. in agriculture. Migrants had especially high income levels compared to those of the national population but. again, women migrants were poorer and earned less than men. As with the findings in the migration literature (Todaro 1976. Du Toit 1990, Cohen 1996'), the result ofthis analysis generally shows that migrants are young, they have high rates of unemployment, and high educational levels, and are not the poorest compared to the national population. These demographic attributes and the human capital of migrants were found to be closely related to income. Education is positively associated with income, and young ages are in a curvilinear relation with income, i.e.. people at ages 20-29 tend to have the highest income level compared to those in other age groups (Wilson 1994, Tana 1997). The results of the analysis also shows that migrants were found to have preferable labor market characteristics (highly-educated, young, single people) in comparison with the national population and urban dwellers, which is consistent with findings in the migration literature (Massey et al. 1993, Du Toit 1990). These characteristics have the potential to help migrants overcome their vulnerability due to a change in their living places. limitation of social networks. or policies related to migrants (Dijk et al. 1986). As with migration in other developing countries, that in Vietnam was marked by different sex ratios depending on the urbanward or ruralward migration (e. g., ruralward migration prevailed among men while women tended to move urbanward). Even though there are no data on sex ratios in migration trends, from the sex ratio of the migrants in 149 this study in comparison with the sex ratio of the general population, we can infer that women tended to be increasingly involved in the urbanward trend, as the migration literature notices (Richter 1994, Dang et a1. 1997). Nevertheless, the results showed some differences from the findings in the migration literature (Chant 1992a, Khoo et al. 1984, UN 1993). Vietnamese migrants tended to move regardless of marital status; this is especially distinct for women migrants. F irst. this difference can be explained by the nature of the transition economy ofVietnam. The country. in the 19803, did not have many transnational corporations that recruited a great number of young, single women. Second, highly educated women, following the traditional recruitment model of the socialist system, migrated to obtain high prestige occupations (in the state sector). which required education, experience and work skills. regardless of their marital status. Consequently, both young single and old married women had a similar chance in the labor force if they had similar educational levels and work skills as did their male counterparts. Thus, both single and married women migrated, not primarily single women as found in the migration literature (UN 1993). In addition, female migrants with high educational levels, a result of the guarantee of equal rights with men in the socialist system, tended to be more active and creative in seeking household betterment compared to women in other developing counties. The fact that Vietnamese women always actively participated in the labor force and in different economic activities was evident. Vietnamese women tended to have a higher status than women in other countries (Frenier et a1. 1995, Barry 1996, Hoang Thi Lich 1993) because female migrants working in the formal sector made up nearly half of the total migrants of 150 this study, slightly less than their male counterparts (40.8 % compared to 59.2 %) (see Chapter 3). Both single and married women actively participated in economic activities in order to improve their household’s economic status, and migration was one of the ways to do so. This result confirms that the characteristics of Vietnamese migrants generally are consistent with the findings of the migration literature (Massey et al. 1993, Du Toit 1990, Cohen 1996. UN 1993, Khoo et al. 1984. Chant 1992a). 151 CHAPTER 5: DECISION-MAKING PROCESS Introduction The decision-making process of migration is referred to as a process in which one or more persons decide who will migrate and why. This process is seen as an operation of social relations and a manifestation of a power hierarchy, in which men and women have different roles and voices (Riley et al. 1993, Thadani et al. 1984, De Jong et al. 1981). This chapter focuses on this process with respect to who makes decisions about mi gration, people’s participation in the decision-making process, and the reasons people migrate. More specifically, 1 demonstrate the nature of gender differences in the decision-making process in relation to migrants’ demographic characteristics and their human, financial, and social capital. Vietnamese women experience contradiction in their lives within both their households and the wider society; this contradiction is a result of the clash between their traditional cultural roles and the influence of socialism. On the one hand, women were, and continue to be, ideologically defined as housekeepers, inferior to men who are considered the breadwinners in the household. On the other hand, women's “actual” roles require them to help their households and the men in them overcome difficulties, often by becoming breadwinners. Under the socialist system, because they were provided more opportunities than previously in schooling and in employment, women developed the capability to fulfill their roles as breadwinners. Fulfillment of this role was closely related to their human capital. This situation increasingly enhanced women’s investment in their human capital. Under the market economy. in contrast, women tend to enlarge their social capital in 152 order to handle their roles. This is so because in a free market characterized by competition. it is reported that women suffer greatly due to the persistence of gender ideology (Le Thi Quy 1998). Their human capital is devalued, they earn less than men, and thus. they cannot help their households as much as they might want. Social capital, however. might help channel them into certain occupations or self-employment in order to maximize their economic success. These contradictions in women’s lives may be reflected in the migration of women compared to that of men. Past research shows that with regard to the decision-making process involved in migration, the decisive voice belongs to men, not women (De Jong et al. 1981). Although the degree of women’s participation in this process is argued to be very low, their participation might differ. depending, for example, on whether they are younger or older, single or married. Given that men are considered to be superior to women in patriarchal societies, the question of whether or not these two indicators play a role in Vietnamese women’s participation in the decision-making process will be addressed. In addition, the questions of whether or not both men and women migrate for economic reasons. or of whether or not men are more independent than women in the decision- making process of migration will be discussed. The theoretical background and literature review of the decision-making process as well as methodology were presented in Chapters 2 and 3. In this chapter, I focus on interpreting the results of my analyses and on examining whether or not the data support my hypotheses. My strategy of analysis is threefold. First, to investigate whether or not Reasons for Migration and Decision-makers of Migration (dependent variables) have relationships 153 with any ofeight independent variables with respect to gender, I use the Chi-Square test to examine the existence of associations between Reasons for Migration and Decision- makers of Migration and the six nominal independent variables (i.e., Gender, Marital Status. Occupation. Employment Status, Social Connections, and Type of Relations). 1 use the ANOVA technique to examine the existence of relationships between Reasons for Migration and Decision-makers of Migration and the three ordinal variables (Age, Income, and Education). These tests are run separately for female and male migrants in order to compare results. The results are presented in Tables 5.1.] through 5.1.9 for Reasons for Migration and Tables 5.2.1 through 5.2.9 for Decision-makers of Migration. Second. to examine the relations and the strength of the relations between each dependent variable and individual independent variables by gender as well as to test Hypotheses 5.1.a. 5.1.b, and 5.2.a through 5.2.d, I separately run logistic regressions of three variables: Reasons for Migration and Decision-makers by Gender and each of the independent variables. The results are presented in Tables 5.1.10 and 5.2.10. Finally, to determine which of the independent variables are the best predictors for Reasons for and Decision-makers of Migration, logistic regression is used for each dependent variable (Reasons for Migration and Decision-makers of Migration) and all significant independent variables. The results are presented in Tables 5.1.11 and 5.2.1 1. 154 1.Reasons for Migration RESULTS AND DISCUSSION Chi-Square Results of Reasons for Migration by Six Nominal Variables 5.1.1. Gender The Reasons for Migration of the studied migrants included economic (38.6 %), family (24.8 %). educational (21.9 %). and other reasons such as health services, housing, or urban attraction (14.7 %) (see Table 5.1.1). This means that migrants generally moved for economic reasons. Family and educational reasons were next in importance to economic ones. The reasons for moving were different for women and men, however. Women were as likely to move for family (32.8 %) as economic (32.5 %) reasons while men moved primarily for economic reasons (45.4 %) (see Table 5.1.1). Gender and Reasons for Migration were associated with each other (x = 35.1, p < .001). Table 5.1.1: Cross-tabulations of Reasons for Migration by Gender Reasons for Migration Female Male All % N % N % N Economic 32.5% 45.4% 38.6% 138 172 310 Educational 19.8% 24.3% 21.9% 84 92 176 Family 32.8% 15.8% 24.8% 141 60 199 Other 14.9% 14.5% 14.7% 63 55 1 18 Totals 100% 100% 100% 424 379 803 x=35.1.p=.000 5.1.2. Marital Status Female and male migrants who were single moved for similar reasons (see Table 5.1.2). Both tended to migrate for economic (36.7 % for women and 34.3 % for men) and 155 educational reasons (29.1 % for women and 35.7 % for men), although men were more likely to move for educational reasons than were women. In contrast, married men mainly migrated for economic reasons (59.6 %) while married women were likely to migrate for family reasons (44.3 %). In addition, there was a clear difference between the proportion of single male migrants moving for educational reasons (35.7 %) and the proportion of married ones (9.6 %) moving for this reason; single men tended to migrate for educational reasons while their married counterparts migrated for economic reasons. The Chi-Square test shows that Marital Status was associated with Reasons for Migration (x = 62.2, p < .001). Table 5.1.2: Cross-tabulations of Reasons for Migration by Marital Status Reasons for Never married Ever married migration Female Male Female Male % N % N % N % N Economic 36.7% 34.3% 28.9% 59.6% 72 73 66 99 Educational 29.1% 35.7% I 1.0% 9.6% 57 76 25 16 Family 20.4% 16.9% 44.3% 14.5% 40 36 101 24 Other 13.8% 13.1% 15.8% 16.3% 27 28 36 27 Totals 1 00% 1 00% 1 00% 1 00% 196 213 231 166 5.1.3. Occupation From the data in Table 5.1.3. it is clear that regardless of occupationI prior to migration, both men and women tended to move for economic rather than for educational and family reasons. ' Only 36.9 % of the sample reported their occupation before moving and among them, 45.1 % worked in the agricultural sector (see Chapter 3). 156 Table 5.1.3: C ross-tabulations of Reasons for Migration by Occupations Reasons for Agriculture Informal sector Formal sector Migration Female Male Female Male Female Male % N % N % N % N % N % N Economic 60.5% 89.1% 45.5% 80.0% 45.1% 52.7% 49 49 5 12 23 39 Educational 3.7% 3.6% 9.1% 6.7% 5.9% 8.1% 3 2 1 1 3 6 Family 22.2% 1.8% 18.2% .0% 33.3% 21.6% 18 l 2 0 17 16 Other 13.6% 5.5% 27.3% 13.2% 15.7% 17.6% 1 1 3 3 3 8 13 Totals 100% 100% 100% 100% 100% 100% 81 55 1 1 15 51 74 7. = 16.852, p = .010 This is very clear for people working in the agricultural sector, especially for men (89.1 %) (compared to 60.5 % of women). Nevertheless, regardless of occupation, women tended to migrate for family reasons more than men, especially women who held jobs in the agricultural sector (22.2 % of women compared to 1.8 % of men). Only men who worked in the formal sector before moving tended to migrate for family reasons (21.6 % compared to 0.0 % working in the informal sector and 1.8 % engaged in agriculture). Occupation was associated with Reasons for Migration (x = 16.852, p < .01). 5. l .4. Employment Status Those who were unemployed (50.9 % for women and 70.7 % for men) and paid employees (53.1 % for women and 68.7 % for men) prior to migration primarily migrated for economic reasons, while people who were workers without wages mainly migrated either for family reasons (56.7 % for women) or for educational reasons (40.0 % for men) (see Table 5.1.4). In addition, gender differences existed across employment status. For instance, regardless of employment status, women were more likely to migrate for family 157 reasons than were men (see Table 5.1.4. row 3). Employment Status was associated with Reasons for Migration (x = 219.805. p < .001 ). Table 5.1.4: C ross-tabulations of Reasons for Migration by Employment Status Reasons for Unemployment Unpaid employment Other Paid employment Migration Female Male Female Male Female Male Female Male 0‘0 N % N % N % N % N % N % N % N Economic 50.9% 70.7% 16.7% 20.0% 14.1% 16.3% 53.1% 68.7% 28 41 5 l 27 27 78 103 Educational 5.5% 17.2% 16.7% 40.0% 35.9% 42.8% 4.8% 6.0% 3 10 5 2 69 71 7 9 Family 27.3% 5.2% 56.7% .0% 35.9% 22.9% 25.9% 12.7% 15 3 17 0 69 38 38 19 Other 16.4% 6.9% 10.0% 40.0% 14.1% 18.1% 16.3% 12.7% 9 4 3 2 27 30 24 19 Totals 100% 100% 100% 100% 100% 100% 100% 10 % SS 58 30 5 192 166 147 150 x = 219.805. p = .000 5.1 .5. Social Connections The data in Table 5.1.5 show that migrants. regardless of social connections in Hanoi prior to migration. tended to move for economic reasons. This trend toward moving for economic reasons was much clearer for migrants without social connections (57.4 %)2 than for those with social connections (36.6 %).3 In addition. despite either the presence or absence of social connections in the city. women were more likely to migrate for family reasons (34.1 % for those with and 25.6 % for those without social connections) than were men who moved primarily for economic ones (42.5 % and 66.0 %, respectively). Men with social connections in the city, however, were more likely to migrate for educational reasons (25.3 %) than were their female counterparts (19.7 %). 2 This percentage is a product of adding the percentages of economic reasons for men (66.0 %) and women (48.8 %) without social connections and then dividing by 2. " This percentage is a product of adding the percentages of economic reasons for men (42.5 %) and women (30.7 %) with social connections and then dividing by 2. 158 Reasons for Migration and Social Connections were related to each other, as the result of the Chi-Square test shows (7. = 16.0. p < .001). Table 5.1.5: Cross-tabulations of Reasons for Migration by Social Connections Reasons for Social Connections Yes Social Connections No Migration Female Male Female Male % N % N % N % N Economic 30.7% 42.5% 48.8% 66.0% 1 17 141 21 31 Educational 19.7% 25.3% 16.3% 17.0% 75 84 7 8 Family 34.1 /o 17.2% 25.6% 6.4% 130 57 1 1 3 Other 15.5% 15.1% 9.3% 10.6% 59 50 4 5 Totals 100% 10 % 100% 100% 381 332 43 47 7. 2 16.0. p : .001 5.1.6. Type afRelations Almost three-quarters (71.4 %) of the migrants had kin relations in Hanoi and slightly over one quarter (28.6 %) had non-kin relations there before moving (see Chapter 3). People with non-kin relations in the city tended to migrate mainly for economic reasons. and this was true for both men (66.0 %) and women (48.0 %) (see Table 5.1.6). Table 5.1.6: Cross-tabulations of Reasons for Migration by Type of Relations. Reasons for Kin relations Non-kin relations migration Female Male Female Male % N % N % N % N Economic 24.4% 37.8% 48.0% 66.0% 68 87 49 31 Educational 19.0% 27.0% 21.6% 21.6% 53 62 22 22 Family 39.8% 20.0% 18.6% 10.8% 1 1 1 46 19 1 1 Other 16.8% 15.2% 1 1.8% 14.7% 47 35 12 15 Totals 100% 100% 10 % 100% 279 230 102 79 x=31.5.p=.000 159 Men with kin relations in Hanoi tended to migrate for economic reasons (37.8 %), while women with relatives in Hanoi tended to migrate mainly for family reasons (39.8 %). Moreover. men with kin relations in Hanoi were slightly more likely to migrate for educational reasons (27.0 %) than were women (19.0 %), but both men and women with non-kin relations in Hanoi were equally likely to move for educational reasons (21.6 % and 21.6 %. respectively). The Chi-Square test revealed that Type of Relations had an association with Reasons for Migration (x = 31.5. p < .001). In short, Reasons for Migration were associated with all six nominal variables, Gender and Marital Status (i. e.. demographic attributes), Occupation and Employment Status (i.e., human capital). and Social Connections and Type of Relations (i. e., social capital). ANOVA Results of Reasons for Migration by Three Ordinal Variables 5. 1 .7. Age As can be seen in Table 5.1.7, those who migrated for educational reasons tended to be the youngest in the sample (mean = 24.9 for women and 24.0 for men). Table 5.1.7: ANOVA Results of Reasons for Migration by Age Reasons for Migration Female Male Total (mean) (mean) (mean) Economic 28.5 31.4 30.1 Educational 24.9 24.0 24.4 Family 37.0 31.2 35.3 Other 36.1 34.2 35.2 Totals 31.6 30.2 30.9 F value significance F = 22.4, p = .000 Women who moved for family reasons were the oldest (mean = 37.0), while men who moved for other reasons were the oldest (mean = 34.2). The mean age of those who 160 moved for economic reasons fell in between (mean = 28.5 for women and 31.4 for men). Women who migrated for economic reasons were slightly younger (mean = 28.5) than men who moved for this reason (mean = 31.4), but they were almost the same age as men who moved for educational reasons (mean = 24.9 and 24.0, respectively). Women who migrated for family reasons were older (mean = 37.0) than men moving for this reason (mean = 31.2). There was a relation between Age and Reasons for Migration (F = 22.4, p < .001). 5.1.8. Income It is clear that people who migrated for economic reasons tended to be the poorest in the sample (mean = 181K for women and 242K for men) (see Table 5.1.8). In contrast, people who migrated for family reasons tended to be the richest; this was true for both men (mean = 323K) and women (mean 2‘ 209K). Table 5.1.8: ANOVA Results of Reasons for Migration by Income Reasons for Migration Female Male Total (mean) (mean) (mean) Economic 181K 242K 215K Educational 203K 260K 235K Family 209K 323K 247K Other 162K 250K 200K Totals 189K 269K 220K F value significance F = .7, p = .559 5.1.9. Education Regardless of their reasons for moving, women’s mean incomes were much lower than those of men, especially those who migrated for economic (181K compared to 242K) and family reasons (209K compared to 323K). The mean income differences 161 between reasons for moving were not significant. In other words, Income had no association with Reasons for Migration (F = .7, p > .1). Among all migrants. those who moved for educational reasons had the highest educational levels (mean = l 1.0 for women and 12.2 for men) (see Table 5.1.9). People who migrated for “other” reasons had the lowest educational levels (mean :87 for women and 9.9 for men). The educational levels of migrants who moved for economic (9.7 for women and 10.4 for men) and family (9.4 for women and 10.6 for men) reasons were relatively similar. In addition, men always had higher educational levels than women for all reasons for migrating (mean = 10.8 for men compared to mean = 9.7 for women). As the ANOVA results show, Education was associated with Reasons for Migration (F = 15.3. p < .001). Table 5.1.9: ANOVA Results of Reasons for Migration by Education Reasons for Migration Female Male Total (mean) (mean) (mean) Economic 9.7 10.4 10.1 Educational 11.0 12.2 11.6 Family 9.4 10.6 9.7 Other 8.7 9.9 9.3 Totals 9.7 10.8 10.2 F value significance F = 15.3, p = .000 In short, the results of the ANOVA show that Reasons for Migration were associated with two ordinal independent variables. i. e.. Age and Education. Income had no association with Reasons/for Migration. Therefore, combining the results of Chi- Square and ANOVA tests. it is clear that, with the exception of Income, Reasons/Or Migration were associated with the independent variables. 162 Results of Logistic Regressions for Reasons for Migration. Controlling for Gender. bv Each of Eight Independent Variables,_and Tests of Hypotheses As can be seen in Table 5.1.10. in general, Reasons for Migration was affected by all independent variables. Income being the exception. Specifically, moving for economic reasons were significantly affected by Gender, Marital Status, Occupation (agriculture), Employment Status (unpaid and other employment status), Social Connections, and Type of Relations. Moving for educational reasons were affected by Age, Marital Status, Education. and Employment Status. Moving for family reasons were affected by almost all independent variables, except Income, Education, and Unemployment Status. Noticeably. Gender affected economic and family reasons for migration, but not educational reasons. In fact, in only one case — that of Employment Status — did Gender affect educational reasons. The explanations for gender’s effects are offered in Model 5.1.1. These explanations are applicable for all those instances in which gender had an effect on economic and family reasons but no effect on educational reasons. 1. Model 5. I. 1. Gender affected moving for family reasons positively (B = .974. p < .001) and affected migration for economic reasons negatively ([3 = -.543, p < .001). This means that men moved for economic reasons while women moved for family reasons. This result supports Hypothesis 5.1.a which states that men move for economic reasons while women move for family reasons. The similarity between this finding and those in other countries (Cook et al. 1996, Morrison et al. 1988, Lichter 1983, Bonney et al. 1991, UN 1993, Nguyen Van Tai 1998) might be seen as a reflection of the fundamental goal of labor migration, economic betterment. and the worldwide prevalence 163 Table 5.1.10: Coefficients of Logistic Regressions for Reasons for Migration by Each of Eight Independent Variables, Controlling for Gender Model Variable Economic (0) Educational (B) Family ([3) 1 Gender -.543*** -.209 974*" Gender -.535*** -.270 .945*** 2 Age -.005 -.057*** .023*** Gender -.580*** -.174 .927*** 3 Marital status -.335* 1.407*** -.646*** Gender -.712** -.235 .991 ** 4 Income -.000 .000 .002 Gender -.554*** -.114 .915*** 5 Education -.035 . 199* * * -.032 Gender -.924*** -.137 1.168*** 6 Occupation - Agriculture 1.187*** -.684 -1.048** - lnforrnal sector .700 -.O73 -1.543* Gender -.530** -.386* .889*** 7 Employment status - Unemployment .001 .823* -.224 - Unpaid employment -1.847*** 1.634” 1.113" - Other -2.183*** 2.451 *** .568" Gender -.533*** -.299 967*" 8 Social connections .866*** -.379 -.626* Gender -.494** -.327 .907 *** 9 Type of relations -.833*** .076 .939*** * :p<0.05 ** :p<0.01 ***:p<0.001 Note: In the logistic regressions, the referent group is the last category of a variable. For example, the formal sector becomes the referent to which other occupations are compared. 164 of gender ideology about men’s dominance (breadwinners) and women’s subordination (housekeepers). Gender did not affect moving for educational reasons ([3 = -.209, p > .05), however. This insignificant relation might be explained by the fact that rural Vietnamese people, regardless of gender, need to move to cities if they want to pursue higher education (see Chapters 1 and 4). 2. Model 5. I .2. Age affected moving for educational reasons negatively ([3 = - .057. p < .001) and migration for family reasons positively ([3 = .023, p < .001). It did not affect migration for economic reasons ([3 = -.005, p > .05). In combination with gender’s effects. as explained in Model 5.1.1, both younger women and men tended to migrate for educational reasons. while their older counterparts, especially older women, migrated for family reasons. In general, however, men, regardless of age, were likely to move for economic reasons. The fact that younger people moved for educational reasons might reflect a household strategy to invest in their human capital levels, enabling them to pursue well-paid jobs and thus improve the household’s economic status. The fact that older women were more likely than older men to migrate for family reasons may reflect different patterns of movement; men moved first and, if they could find a job and become relatively settled, their wives would follow, moving to reunite the family. Age had no effect on migration for economic reasons and this might reflect the gender roles of men and women. Regardless of age, men moved for economic reasons because they were assumed to be breadwinners while women moved for family reasons due to their housekeeper and subordinate roles. 165 3. Ill/lode] 5. I .3. Marital Status affected migrating for economic (B = -.335, p < .05) and family reasons negatively ([3 = —.646, p < .001) while it affected moving for educational reasons positively ([3 = 1.407, p < .001). This means that married people moved for both economic and family reasons while single people tended to move for educational reasons. Nevertheless. when seen in combination with gender’s effect in this model, it is clear that men. regardless of marital status, and single women moved for economic reasons while married women moved for family reasons. This result supports Hypothesis 5.1.b which states that men (both single and married) and single women move for economic reasons while married women move for family reasons. The result is similar to findings from migration studies in other developing countries about marital differences in reasons for migration (Cook et al. 1996, Morrison et al. 1988. Lichter 1983, Bonney et al. 1991, UN 1993, Nguyen Van Tai 1998), and it may be a product of women’s age and stage in the life cycle. Married women are more closely engaged in managing the household than are single women and thus, they move for family reasons, maintaining the home at destination. In contrast, single women move for economic reasons because, being the preferred labor force, they have better opportunities to find work in cities than their married counterparts. 4. Model 5.1.4. Income, i.e., financial capital, did not affect Reasons for Migration. Despite this non-significant relation, there was a tendency that, regardless of income levels, men moved for economic reasons and women moved for family reasons. This non-significant relation might be explained by the fact that, first, about 51 percent of the Vietnamese population lives below the poverty line (see Chapter 4). Thus, there was 166 not enough variance in the variable to show how income might motivate migration. Second. this finding may result from sample size of the data set or the inadequacy of the income measure in the MOLISA study: income was collected for individuals. not for households as an income unit, and it was based on the amount of agrarian products harvested rather than cash. 5. .lelodel 5.1.5. Education had an effect only on migration for educational reasons and this effect was positive (B = .199. p < .001). When seen in combination with the insignificant effect of gender. this means that both female and male migrants with higher educational levels tended to move for educational reasons while those with lower educational levels moved for other reasons. The finding is not surprising because those with high educational levels would be more likely to move in order to seek higher education than those with low educational levels. Moreover, educational facilities in cities are better than those in rural areas. Vietnamese rural people, thus, have to move to cities if they want to pursue higher education. The fact that education did not affect moving for economic and family reasons may be the result of the relatively high educational levels of migrants in the sample compared to that of people in the nation (see Chapter 4). 6. Model 5. I. 6. Work in the agricultural sector affected migration for economic reasons positively (B = 1.187. p < .001) and moving for family reasons negatively (B = - 1.048. p < .01 ). More specifically, when seen in combination with gender’s effects. although both women and men who worked within the agricultural sector prior to migration primarily moved for economic reasons. men tended to migrate for economic 167 reasons while their female counterparts were likely to move for family reasons. This difference may result from the prevalence among agriculturists of traditional ideology which views men as breadwinners and women as homemakers. Thus, women may have reported that they were migrating for family reasons (e.g., to manage the household) even though they might eventually seek work in the city. The fact that work in the agricultural sector did not affect moving for educational reasons may be a result of the low educational levels of migrants who worked in this sector (mean = 8.2) compared to those in other sectors (mean = 10.2 for the informal sector and 1 1.8 for the formal sector) (see Appendix 3.1). People with low educational levels may not choose to invest in their human capital. Work in the informal sector affected migration for family reasons negatively (B = -1.543. p < .05). When seen in combination with gender’s effect, this means that women working in this sector were more likely than men to move for family reasons. This result might be explained, as mentioned above. by women’s traditional gender roles as housekeepers and men’s as breadwinners. This result also might be explained by women’s double disadvantage (work in the informal sector and discrimination as a female). Men who worked in this sector experienced only occupational disadvantage (see Chapter 3). not gender discrimination. Women’s double disadvantage might limit their economic opportunities. therefore encouraging them to move for family rather than economic reasons. In the absence of a potential job, they could keep house at destination and eliminate the cost of maintaining two households. one in the country and one in the city. 168 Work in the informal sector did not affect moving for economic and educational reasons. This finding reflects the fact that migrants who worked in this sector were mainly married (65.7 %) and the oldest in the sample (mean = 50.5) (see Appendices 5.1 and 5.2) compared to those moving for other reasons (mean = 26 for economic and 30.3 for educational reasons). Therefore, those who worked in this sector were less likely to move for economic or educational reasons, but rather moved for family reasons as a way to maximize the returns of migration (maintaining one household instead of two). Alternatively. they may have been too old to find work and thus moved with others family members. who could provide for them financially. 7. Model 5.1. 7. Employment Status significantly affected all three reasons for moving. With the exception of those who were unemployed, when seen in combination with gender’s effect, men tended to move for economic reasons while women moved for family reasons. Regardless of employment status, however, men were more likely than women to move for educational reasons (B = .823, p < .05 for unemployment, 1.634 for unpaid employment and 2.451 for other, p < .01). Women who worked without pay (B = -l .847. p < .001) or had an “other” employment status (B = -2.183, p < .001) were less likely than men to move for economic reasons but more likely than men to move for family reasons (B = 1.1 13 for unpaid and .568 for other employment status, p < .01). This gender difference might be explained, as above, by the ideological definition of gender roles. Although women have gained equal rights with men in employment, i.e., they can work for pay as do men, their roles still are considered bound within the household. The fact that men were more likely to move for educational reasons than 169 women might be explained by the fact that (1) the men in the sample were younger than the women (see Table 4.5), and (2) they had higher educational levels than the women (see Chapter 4). Men thus, moved to seek higher education more than did women. Furthermore, the fact that men moved for educational reasons more than women might be rooted in gender ideology about male ability and superiority. Given this belief, people may have probably thought it was more worthwhile to invest in their education for the betterment of the household rather than in that of women. 8. Model 5.1.8. Social Connections affected migration for economic reasons positively (B = .866. p < .001) while they affected moving for family reasons negatively (B = -.626, p < .05). When seen in combination with gender’s effect, this means that, even though both female and male migrants without social connections in Hanoi tended to move for economic reasons. men were more likely than women to move for economic reasons. In contrast. men with social connections moved for economic reasons while their female counterparts moved for family and economic reasons. Women, however, regardless of social connections in Hanoi, were more likely to move for family reasons than men, who were more likely to move for economic reasons. The similarity of women and men without social connections in Hanoi might reflect a high degree of economic need on the part of such people. That is, despite the lack of potential support in the city, they took the risk of moving in order to meet their immediate economic needs. Moreover, the reasons for moving of those with social connections differed by gender. This gender difference may reflect the obduracy of gender ideology about women’s and men’s roles, as noted above. This gender difference 170 may also reflect the fact that women moved to reunite the family, thereby saving money by maintaining one household only. Social Connections did not affect moving for educational reasons and this may be explained by (1) the migrants’ relatively high educational levels that made moving for educational reasons not a major goal of migration4 and (2) the great proportion of migrants (88.8 %) having social connections before moving. This finding also may be rooted in the fact that there were more facilities for higher education in the city than in the rural areas and thus. people could gain entry into institutions of high education without help from their social networks. 9. Model 5. I . 9. Type of Relations affected moving for economic reasons negatively (B = -.833, p < .001) while it affected migration for family reasons positively (B = .939. p < .001). When seen in combination with gender’s effect, it is clear that men, regardless of type of relations, and women with non-kin relations in Hanoi tended to migrate for economic reasons while women with kin relations in Hanoi migrated for family reasons. The lack of gender difference among those with non-kin relations in Hanoi may reflect the advantage well selected migrants (highly educated) had in the labor market. That is, migrants with non-kin relations in the city had higher educational levels (mean = 1 1.2) than did those with kin relations there (mean = 10.03) (see Appendix 5.3). The gender difference among those with kin relations in Hanoi also might be explained by the fact that relatives and family members in Vietnam are enmeshed in close networks " Migrants in this study had an average of 10.5 years of education while the Vietnamese population had an average of only 9.5 years of education. The difference in educational levels was striking at the university and college level: 47.2% ofthe migrants had attained this high educational level compared to 1.6% of the nation’s population (see Chapter 4). I71 of responsibility and obligation: women tend to be bound within their family more strongly than are men, and such ties may influence women’s motivation for moving more than they do that of men. Type of Relations did not affect moving for educational reasons. This result might be explained, as it was in Model 5.1.8: the relatively high educational levels of migrants and easy access to higher education in the city allowed migrants to obtain higher education without help from kin or non-kin in their social networks. In sum, reasons for migration were affected by eight out of nine independent variables, that is, demographic attributes and human and social capital, but notfinancial capital. Hypothesis 5. I .a and Hypothesis 5. 1. b were supported. Results of Logistic Regression of Reasons for Migration byAll Significant Independent Variables The results of the regression analysis presented in Table 5.1.1] show that among the significant variables, Gender (B = -.796, p < .01), Marital Status (B = 1.163, p < .001), Occupation (agriculture) (B = 1.156, p < .001), and Type of Relations (B = -.970, p < .01) were good predictors of moving for economic reasons: Age (B = -.033, p < .01), Education (B = .218, p < .001), and Employment Status (B = 1.761, p < .001 for unpaid employment, B = 2.020. p < .001 for other employment status) were good predictors of moving for educational reasons; Gender (B = 1.167, p < .01), Age (B = .039. p < .05), Occupation (agriculture) (B = -1.071, p < .001), and Type of Relations (B = 1.303. p < .01) were good predictors of moving for family reasons. These results mean that men and single women who worked in the agricultural sector before moving and who had non-kin 172 Table 5.1.1 1: Coefficients of Logistic Regression for Reasons for Migration by All Independent Variables Variables Economic (B) Educational (13) Family (B) Demographic attributes: Gender -.796** . 1 . 167* Age . -.033** .039* Marital status 1163*“ Financial capital: Income Human capital: Education . 218* * * Occupation , - Agriculture 1.156*** . -1.071** - Informal sector Employment - Unemployment . . - Unpaid employment . 1.761 *** - Other employment . 2020*“ Social capital: Type of relations -.970** . 1.303“ * : p < 0.05 ** : p < 0.01 ***: p < 0.001 . : insignificant relation. Note : - In the logistic regressions, the referent group is the last category of a variable. - Social connections variable cannot be computed because there was no variation in responses. relations in the city migrated for economic reasons. Both young women and men with high educational levels, who worked without pay (especially women) or who were students’ prior to migration were likely to move for educational reasons. Older women who worked in the agricultural sector before moving and who had kin relations in Hanoi were likely to migrate for family reasons. Since those who moved for educational reasons made up a smaller percentage (21.9 %) of the sample than those moving for economic (38.6 %) and family reasons (24.8 %), we can infer that education was not the main reason people moved. Rather, economic reasons were their major motivation for migrating. ’ Students comprised 73.7 % of people in the “other” employment status (based on calculation from the MOLISA data set by the author). 173 The fact that human and social capital are good predictors of reasons for migration confirms assumptions of (l) the human capital approach that the goal of migration is to maximize its economic return by finding the optimal use for a migrant’s human capital rather than investing in her or his human capital (Trzcinski et al. 1991), and (2) of the social capital perspective that social networks help reduce the risk and ensure the success of moving, and raise the incentives for migration (Douglas et al. 1992, Hugo 1995, Fawcett 1989. Massey et al. 1993). Financial capital was not a good predictor of reasons for migration, and this might be explained by the migrants’ relatively high income levels compared to those of the national population (see Chapter 4) or the way in which the variable was operationalized (see Model 5.1.4). In sum, these results tend to be consistent with findings from the migration literature. People migrate mainly for economic reasons, and human capital as well as social capital affect their motivation to migrate. Moreover, the great proportion (about 90%) of people who had social connections in Hanoi before moving and the high educational levels of migrants compared to those of the national population (see Footnote 4) demonstrate the importance of social capital and human capital for Vietnamese migrants. Social and human capital play extremely important roles in the decision- making process of Vietnamese urbanward migration. 2. Decision-makers in the migratory process. Below, I explore the decision-makers of migration, i.e., self, couples, family members, and others (such as employers and friends). As noted in Chapter 3. the category of family members does not include a married couple. A couple means that a wife as well as a husband had a voice in the decision-making process. In other words, the 174 decision to move was jointly made. Much of the migration literature suggests that women have little voice in the migration decision-making process (Chant 1992a. De Jong et al. 1981). The categories used in the MOLISA study, however, suggest that participation in the process may range across a continuum from those with a large voice (i.e., self-determining) to those with no voice (i.e., deployed by family members). Joint decision making represents the mid-point between these two ends. The MOLISA data thus provide an opportunity to approach women’s role in the decision-making process in a more nuanced way than has heretofore been done. Chi-Sjuare Results of Decision-Makers of Migration by Six Nominal Variables 5.2.1. Gender As can be seen in Table 5.2.1. the decision to move by migrants in this study was self-determined (36.1 %). directed by family members (37.3 %), and determined by joint decision-making (17.0 %). Men tended to be more self-determining (44.9%) than were women (28.4 %). Table 5.2.1: Cross-tabulations of Decision-makers of Migration by Gender Decision-makers of Female Male All Migration % N % N % N Self 28.4% 44.9% 36.1% 121 170 291 Couples 23.0% 10.3% 17.0% 98 39 137 Family members 42.0% 31.9% 37.3% 179 121 300 Other 6.6% 12.9% 9.6% 28 49 77 Totals 100% 10 % 10 % 426 379 805 x = 48.020. p = .000. I75 The decision-making process for female migrants was not as straightforward as that of men. however. Women’s movement was decided on by family members (42.0 %), couples (23.0 %). and themselves (28.4 %). If we combine the percentages of self- deterrnining and joint decision making (couples), however. it is clear that female migrants had influence on decision making in migration. Slightly more than half of them (51.4 %)“ had a voice in making the decision. This proportion was similar to that of male migrants (55.2 %)7. even though the proportion of male migrants who made the decision to move by themselves (44.9 %) was much greater than that of female migrants (28.4 %). There was a relation between Gender and Decision-makers (x = 48.020, p < .001). 5.2.2. Marital Status Single men (47.9 %) and women (58.1 %) tended to be directed by family members to migrate rather than to make the decision themselves (34.2 % for women and 40.8 % for men) (see Table 5.2.2). Table 5.2.2: Cross-tabulations of Decision-makers of Migration by Marital Status Decision-makers Never married Ever married for Migration Female Male Female Male % N % N % N °/o N Self 34.3% 40.8% 23.2% 50.0% 68 87 53 83 Couples .0% .5% 42.5% 22.9% 0 l 97 43 Family members 58.1% 47.9% 28.5% 1 1.4% 1 15 102 65 19 Other 7.6% 10.8% 5.7% 15.7% 15 23 13 26 Totals 100% 100% 100% 100% 198 213 230 166 x = 192.150. p = .000. " This is the sum of self(28.4 %) and couples (23.0 %). 7 This is the sum of self(44.9 %) and couples (10.3 %). 176 Married women tended to make the decision to move jointly with their husbands (42.5 %) while married men tended to be self-determining (50.0 %). The proportion of married women (23.2 %) who were self-determining was similar to the proportion of married men whose decision to move was made jointly with their wives (22.9 %). Both married women (28.5 %) and men (1 1.4 %) were unlikely to be directed by family members to migrate in contrast to single migrants (58.1 % for women and 47.9 % for men). Moreover. in relation to influence in decision making. it is clear that most married people (about 70 %) had a voice in decision making (65.7 % for women8 and 72.9 % for men”) while single people were less likely to participate in the decision-making process (34.3 % for women and 40.8 % for men). Not surprisingly, Marital Status was associated with Decision-makers ofMigration (x = 192.150, p < .001). 5 .23. Occupation Male migrants tended to be self-determining, regardless of their occupation prior to migration (63.6 % for men within the agricultural sector, 40.0 % in the informal sector, and 47.3 % in the formal sector) (see Table 5.2.3). Women, in contrast, were more likely to decide to migrate jointly with their spouses (54.5 % in the informal sector, and 33.3 % in the formal sector). Women who worked in the agricultural sector, however, tended to be self-determining (38.3 %), although about one fourth (25.9 %) made the decision jointly with their spouses. There were some differences between men and women for whom the decision to migrate was made by family members. Although family members tended to direct only women working in the informal sector (9.1 %), they directed both 8 This is the sum of self(23.2 %) and couples (42.5 %). 0 This is the sum of self(50.0 %) and couples (22.9 %). 177 Table 5.2.3: Cross-tabulations of Decision-makers of Migration by Occupations Decision- Agriculture Informal sector Formal sector makers for Female Male Female Male Female Male Migration % N 0/o N 0/o N 0/o N 0/o N 0/o N Self 38.3% 63.6% 36.4% 40.0% 29.4% 47.3% 31 35 4 6 15 35 Couples 25.9% 10.9% 54.5% 20.0% 33.3% 16.2% 21 6 6 3 17 12 Family 29.6% 14.5% 9.1% .0% 19.6% 13.5% members 24 8 1 0 10 10 Other 6.2% 10.9% .0% 40.0% 17.6% 23.0% 5 6 0 6 9 17 Totals 100% 100% 100% 100% 100% 100% 81 55 1 1 15 51 74 x = 16.986, p = .009 men (14.5 %) and women (29.6 %) who worked in agriculture and in the formal sector (13.5 % for men and 19.6 for women). Even in these cases. however, family influence was minimal for men. Across occupations, both female and male migrants tended to be self-determining and involved in joint decision-making, i.e., they were likely to wield influence in decision making, regardless of occupation. For example, 64.2 percent of women in agriculture.'0 90.9 percent in the informal sector,ll and 62.7 percent in the formal sector’2 had a voice in the decision to migrate. The respective figures for men were 74.5 percent, 60.0 percent, and 63.5 percent.13 Occupation was associated with Decision-makers ofMigration (x = 16.986, p < .01). 5.2.4. Employment Status The data in Table 5.2.4 indicate that men who were unemployed (62.1 %) and working for pay (51.3 %) prior to migration tended to be self-determining decision- '0 This is the sum of self(38.3 %) and couples (25.9 %). ” This is the sum of self(36.4 %) and couples (54.5 %). '2 This is the sum of self(29.4 %) and couples (33.3 %). '3 These are sums of self(63.6 %, 40.0 %. 47.3 %) and couples (10.9 %, 20.0 %, 16.2 %). respectively. 178 Table 5.2.4: C ross-tabulations of Decision—makers of Migration by Employment Status Decision- Unemployment Unpaid employment Other Paid employment makers for Female Male Female Male Female Male Female Male Migration % N % N % N % N % N % N % N % N Self 34.5% 62.1% 18.8% 40.0% 23.4% 33.1% 34.7% 51.3% 19 36 6 2 45 55 51 77 Couples 21.8% 8.6% 40.6% .0% 14.1% 6.0% 30.6% 16.0% 12 5 13 0 27 10 45 24 Family 40.0% 24.1% 40.6% 60.0% 56.3% 51.2% 25.2% 12.7% members 22 14 13 3 108 85 37 19 Other 3.6% 5.2% .0% .0% 6.3% 9.6% 9.5% 20.0% 2 3 0 0 12 16 14 30 Totals 100% 100% 100% 100% 100% 100% 100% 100% 55 58 32 5 192 166 147 150 x = 112.448. p = .000. makers. Women who were unemployed (34.5 %) and working for pay (34.7 %) also tended to be self-determining, although less so than men. More than half of men who worked without pay (60.0 %) and half of those who held an “other” employment status (51.2 %) were directed to move by family members. Women who reported “other” as their employment status (56.3 %) also were directed by family members to migrate in contrast to the two fifths (40.6 %) of women involved in unpaid employment who were joint decision-makers. Female migrants who were unemployed (56.3 %)M had less voice than men (70.7 %)’S in decision making. In contrast, women who worked without pay (59.4 %”’) had more voice than their male counterparts (40.0 %’7). Women who worked for pay (65.3 %‘8) and women who had an “other” employment status (37.5 %’9) were as likely as their male counterparts (67.3 %20 and 39.1 %2’, respectively) to participate in '4 This is the sum of self(34.5 %) and couples (21.8 %). "’ This is the sum of self(62.l %) and couples (8.6 %). ”‘ This is the sum of self(18.8 %) and couples (40.6 %). ‘7 This is the sum of self(40.0 %) and couples (.0 %). ‘8 This is the sum of self(34.7 %) and couples (30.6 %). '° This is the sum of self(23.4 %) and couples (14.1 %). 3“ This is the sum of self(5 I .3 %) and couples (16.0 %). I79 decision making. Employment Status and Decision-makers of Migration were associated with each other (x = 1 12.448. p < .001). 5.2.5. Social (.‘onnections From the data in Table 5.2.5 it is clear that the presence or absence of connections in Hanoi bore little relationship to who made the decision to migrate. Male migrants with or without social connections were self-determining (43.4 % and 55.3 %. respectively). While social connections also were unimportant in this instance for women, the decision maker for them varied. That is. regardless of social connections, some women were deployed by family members (42.7 % with connections and 38.1 % without connections), others migrated as a result of a joint decision with their husbands (23.2 % with connections and 19.0 % without connections). and a number made the decision to move themselves (28.4 % for women with connections, and 28.6 % without connections). Table 5.2.5: C ross-tabulations of Decision-makers of Migration by Social Connections Decision-makers Social connections Yes Social connections No for Migration Female Male Female Male % N % N % N % N Self 28.4% 43.4% 28.6% 55.3% 109 144 12 26 Couples 23.2% 10.2% 19.0% 10.6% 89 34 8 5 Family members 42.7% 33.7% 38.1% 19.1% 164 1 12 16 9 Other 5.7% 12.7% 14.3% 14.9% 22 42 6 7 Totals 100% 100% 1 00% 100% 384 332 42 47 Regardless of social connections. women were more likely to be joint decision- makers than were men (23.2 % of women with social connections compared to 10.2 % of 3' This is the sum ofself(33.l %) and couples (6.0 %). 180 men. and 19.0 % of women without social connections compared to 10.6 % of men). Further. regardless of social connections. men were less likely to be directed to migrate by family members than were women (33.7 % for men and 42.7 % for women with social connections. and 19.1 % for men and 38.1 % for women without social connections). In addition. it is clear that regardless of the presence or absence of social connections, about half ofthe women and half ofthe men had a voice in decision making (51.6 % for women and 53.6 % for men who had social connections in Hanoi”; 47.6 % for women and 65.9 % for men who had no social connections in Hanoi”). Social connections were not associated with Decision-makers of Migration, however, as the Chi-square test shows (x = 6.132. p > .1). 5.2.6. Type ofRe/ations As can be seen in Table 5.2.6, women (52.1 %) and men (44.3 %) with kin relations in Hanoi were likely to be directed to move by family members, although a third (36.5%) of the men with kin relations were likely to make the decision themselves. In contrast, women (40.2 %) and men (58.8%) who had non-kin relations in Hanoi were self-determining, although women who had non-kin relations also were involved in joint decision-making (35.3 %). Regardless of type of relations in Hanoi, women were involved injoint decision-making (35.3 % and 18.8 %) more than were men (14.7 % and 8.3 %), while men were self-determining (58.8 % and 36.5 %) more than were women (40.2 % and 24.1 %). Both women and men who had non-kin relations in the city were 3: These are sums of self(28.4 "/o for women and 43.4 % for men) and couples (23.2 % for women and 10.2 % for men), respectively. 3" These are sums of self(28.6 % for women and 55.3 % for men) and couples (19.0 % for women and 10.6 % for men). respectively. 181 Table 5.2.6: Cross-tabulations of Decision-makers of Migration by Type of relations Decision-makers Kin relations Non-kin relations for Migration Female Male Female Male 0/0 N "/0 N "/0 N "/0 N Self 24.1% 36.5% 40.2% 58.8% 68 84 41 60 Couples 18.8% 8.3% 35.3% 14.7% 53 19 36 15 Family members 52.1% 44.3% 16.7% 9.8% 147 102 17 10 Other 5.0% 10.9% 7.8% 16.7% 14 25 8 17 Totals 100% 100% 100% 100% 281 230 102 102 x = 76.891. p = .000. m ore likely to have a voice in decision making (75.5 % for women and 73.5 % for men)24 than were those with kin relations (42.9 % for women and 44.8 % for men).25 Nevertheless, women as well as men who had kin relations in Hanoi were directed to move by family members (52.1 % for women and 44.3 % for men) more than those with no n-kin relations in Hanoi (16.7 % and 9.8 %, respectively). The Chi-Square test showed that Type of Relations was associated with Decision-makers (x = 76.891, p < .001). In short. Decision-makers of Migration was associated withfive out of six nominal variables, that is, Gender and Marital Status (i. e., demographic attributes), ()ccupation and Employment Status (i. e. human capital), and Type of Relations (i. e.. so cial capital). \ 2” These are sums of self(40.2 % for women and 58.8 % for men) and couples (35.3 % for women and L47 % for men), respectively. " These are sums of self(24.1 % for women and 36.5 % for men) and couples (18.8 % for women and 8.3 % for men). respectively. 182 ANOVA Results of Decision-makers of Migration by Three Ordinal Variables 5.2.7. Age The youngest migrants (mean = 29.0 for women and 23.6 for men) (see Table 5.2.7) tended to be directed to move by family members, while the oldest migrants (mean = 40.3 for women and 42.7 for men) tended to make the decision to move jointly with their spouses. Table 5.2.7: ANOVA Results of Decision-makers of Migration by Age Decision-makers of Female Male Total Migration (mean) (mean) (mean) Self 29.3 30.9 30.2 Couples 40.3 42.7 40.9 Family members 29.0 23.6 26.8 Other 31.1 32.3 31.9 Totals 31.8 30.0 30.9 F and p value F = 32.092. p = .000 Migrants who were self-determining were of moderate ages (mean = 29.3 for women and 30.9 for men). Clearly. age does play a role in decision making. Those who were older Were more likely to participate in decision making than their younger counterparts, and this was true for women as well as men. More specifically, women who made the decision themselves orjointly with their husbands were on average older (mean = 34.8)26 than were women who were deployed by family members (mean = 29.0). The respective aVerages for men were 36.83"7 and 23.6. Age was associated with Decision-makers of Migration as the ANOVA test shows (F = 32.092, p < .001). i” This is an average ofthe means of self(29.3) and couples (40.3). ’7 This is an average ofthe means of self(30.9) and couples (42.7). 5 .28. Income The mean income of male migrants who were directed to move by family members was the lowest (mean = 192K) while women who were self-determining were the poorest (mean = 166K) (see Table 5.2.8). Table 5.2.8: ANOVA Results of Decision-makers of Migration by Income Decision—makers of Female Male Total Migration (mean) (mean) (mean) Self 166K 224K 201K Couples 200K 322K 243K Family members 182K 192K 185K Other 228K 322K 289K Totals 186K 254K 220K F and p value F = 3.734, p = .012 Those who made the decision jointly with their spouses had a much higher mean income, although not the highest (200K for women and 322K for men). Clearly, across decision— makers. women always had lower incomes than those of men (mean = 228K for women compared to 322K for men in the “other” category and mean = 200K for women Compared to 322K for men in the category of joint decision making). Furthermore, those W’ho had a voice in making the decision tended to have higher income levels (183K for Women and 273K for men)28 than those who were directed by family members (182K for W'Omen and 192 K for men). Income was associated with Decision-makers (F = 3.734, p < . OS). 5.2.9. Education From the data in Table 5.2.9, we can see that female and male migrants who were directed to move by family members tended to have the lowest educational levels (mean ‘2 2” These figures are averages ofthe means of self(166K) and couples (200K) for women. and of self (224K) and couples (322K) for men. 184 = 8.9 for women and 9.8 for men). Among female migrants, those who were self- detennining movers or were influenced to move by “other” decision-makers had the highest educational levels (mean = 10.5 and 10.4, respectively). Male migrants who were joint decision-makers had high educational levels (mean = 1 1.6), although those whose move was directed by “other” decision-makers had the highest educational level (mean = 1 2.0). Table 5.2.9: ANOVA Results of Decision-makers of Migration by Education Decision-makers of Female Male Total M i gration (mean) (mean) (mean) Self 10.5 11.0 10.8 Couples 9.8 11.6 10.3 Family members 8.9 9.8 9.3 Other 10.4 12.0 11.5 Totals 9.7 10.8 10.2 F and p value F = 14.977, p = .000 Among male migrants, men who were self-determining did not have the highest educational level (mean = l 1.0) as was the case for women (mean = 10.5), even though men’s educational levels were higher (mean = 10.8) than those of women (mean = 9.7). Education was associated with Decision-makers (F = 14.977. p < .001). In sum. the results of the ANOVA tests demonstrate that Decision-makers of 1141' gration was associated with all three ordinal variables: Age, Income, and Education. In combination with the results of Chi-Square tests, it is clear that, with the exception of SOctal connections, Decision-makers of Migration was associated with eight independent Variables, i.e., demographic attributes, and human, social, andfinancial capital. 185 Results of Logistic Regressions of Decision-makers of Migration, Controlling for Gender, bv Eight Independent Variables, and Tests of Hypotheses From the results presented in Table 5.2.10, it is clear that, in general, decision- makers were affected by all four sets of independent variables. Specifically, self- determination in decision making was affected by Gender, Income, Education, Occupation (agriculture). Employment Status (other), and Type of Relations. Joint decision making was affected by Gender, Age, Marital Status, Employment Status (other). and Type of Relations. Family members as decision-makers were affected by Gender, Age. Marital Status, Education, Employment Status, and Type of Relations. Noticeably. Gender had significant effects on Decision-makers of Migration. In fact, in only one case -— that of Education — did it not affect the decision-makers of migration (e. g., family members). The explanations for gender’s effects are offered in Model 5.2.1. These explanations are applicable for all those instances in which gender had a significant effect on the dependent variable. 1. Model 5. 2. l . As can be seen in Table 5.2.10, gender affected self-determination Of migration negatively ([5 = -.718. p < .001), but affectedjoint decision making and family members as decision-makers positively (B = .944, p < .001 and .445, p < .01 , respectively). That is. male migrants were self-determining while female migrants moved as a result of both joint decision-making and the direction of family members. Hypothesis 5.2.a, which states that women are directed to move by family members While men are self-determining in making the decision to migrate, was thus rejected. Vietnamese female migrants were not simply passively deployed into migration streams 186 Table 5.2.10: Coefficients of Logistic Regressions for Decision-makers of Migration by Each of Eight Other Independent Variables, Controlling for Gender formal sector becomes the referent to which other occupations are compared. 187 Model Variable Self(l3) Couples (0) Family members (0) I 1 Gender -.718*** 944*" .445W Gender -.712*** .891 *** 515*" 2 Age -.004 .046*** -.036*** Gender -.71 1*“ .878*** .660M 3 Marital status .072 —5.339*** 1515*“ Gender -.846*** .941*** .748* 4 Income -.002* .001 -.00| Gender -.633*** 1017*“ .267 5 Education .071 ** .032 -.124*** Gender -.843*** 1065*“ .740* 6 Occupation - Agriculture 524* -.406 .35l - Informal sector -.054 .578 -.l.583 Gender -.679*** .974*** .430“ 7 Employment status - Unemployment .226 -.546 .708*** - Unpaid employment -.765 .273 1.041 *** __ - Other employment -.659*** -l 725*“ l.6l4*** Gender -.7l3*** 941*” .536“ _ 8 Social connections .255 -.l39 -.450 Gender -.642*** 1029*” .357* 9 Type of relations -.828*** -.789*** l.8l8*** * : p < 0.05 * * : p < 0.0] * * *2 p < 0.00] Note: In the logistic regressions, the referent group is the last category of a variable. For example, the by family members. Rather, they were active participants in the decision-making process. They tended to have a much greater voice in decision-making than has been reported for women in other developing societies (Boyd 1989, De Jong et a1. 1983, Lim 1993. Amold 1987). This result may reflect the fact that they have gained more rights vis-a-vis men under the socialist system than have women in capitalist societies. 2. Model 5. 2.2. Age affected joint decision-making positively (B = .046, p < .001) while it affected family members as decision-makers negatively (B = -.036, p < .001). Age did not affect self-determination (B = -.004, p > .05). When seen in combination with gender’s effects. it is clear that regardless of age, men were more likely than women to be self-determining; older migrants -- women as well as men -- tended to be joint decision-makers while younger people were directed to move by family members. This result supports Hypothesis 5.2.b which states that young women and men are directed to move by family members. This finding is consistent with that in the migration literature about the advantage of old age within the household (De Jong et al. 1981, Riley et al.1993, Lim 1993, Hugo 1995), i.e., a woman’s status increases with age, thereby giving her a greater voice in decision making than that accorded to younger women. In contrast, the fact that men tended to be more self-determining than women, regardless of age, reflects the persistence of a patriarchal hierarchy that provides more autonomy and power to men than to women. 3. Model 5. 2.3. Marital Status affected joint decision-making negatively ([3 = - 5.339. p < .001) while it affected family members as decision-makers positively ([3 = 1.515, p < .001). More specifically, in combination with gender’s effects, married 188 women tended to be joint-makers while single women and men were directed to migrate by family members. Hypothesis 5.2.c, which states that single women and men are directed to move by family members while married men are self-determining and married women join with their husbands in making decisions, was supported. This result is consistent with findings in the migration literature about the effect of age and marital status on the power hierarchy in a household (Riley et al. 1993, De Jong et al. 1981, Ilarbison 1981, Hugo 1995). Stage in the life cycle (i.e., marital status) give women an advantage in the decision-making process, providing them more voice and influence in their households than their single counterparts. Marital Status did not affect self-determination (B = .072, p > .05). Regardless of marital status, men tended to be more self-determining than were women. This finding might be explained by the persistence of a power imbalance in the household. Although Vietnamese women gain influence in the household with increasing age and marriage, and although the socialist government encouraged women's equality with men, women still do not have as great a voice in the decision-making process as do men. 4. Model 5. 2. 4. Income affected self-determination in decision making negatively (B = —.002. p < .05), but it did not affect couples (B = .001, p > .05) and family members (I = -.001, p > .05) as decision-makers of migration. When seen in combination with gender’s effects. it is clear that among those whose move was self-determined, the women were poorer than the men. This result might be explained by the impact of poverty on degree of participation in making decisions. If a person was poor, migration for economic reasons became an imperative that probably did not need discussion. Poor 189 women, in particular, decided by themselves to migrate because their dire economic situation (they were twice as poor as men) (see Table 5.2.8) pushed them to try their luck by moving without the intervention of family members. 5. Model 5. 2. 5. Education affected self-determination positively (B = .071, p < .01) and affected family members as decision-makers negatively (B = -.124, p < .001). When seen in combination with gender’s effect, even though men tended to be self- determining more than women, female and male migrants with high educational levels were likely to be self-determining while those with low educational levels were directed to move by family members. This result provides support for Hypothesis 5.2.d which states that both women and men with higher educational levels are self-determining in making the decision to move while those with lower educational levels are directed to move by family members. This result is consistent with findings in the migration literature about the effect of education on degree of participation in making the decision to move (Wilk 1989, Riley et al. 1993). Obviously, educational level has the potential to give people a voice in the decision-making process since it reflects migrants’ ability to compete in the urban labor market and, thus, their potential for economic success in the migration process, as noted by the human capital approach and findings from the migration literature (Massey et al. 1993, Sanders et al. 1996, Hunt et al. 1985, Robinson 1993). Consequently, those with higher educational levels are more likely to exert influence in the decision-making process than are their less educated counterparts. Education did not affect joint decision-making. In combination with gender’s effect, women, regardless of educational levels, were more likely than men to be joint 190 decision-makers. This insignificant effect might be rooted in the fact that migrants had relatively high educational levels compared to that of the nation’s population, and the fact that women gained equal rights to men under the socialist system, potentially increasing their voice in their households. In this instance, gender had no effect on family members as decision-makers of migration. Both women and men with low educational levels were directed to move by family members. This result might reflect the control a family had over members disadvantaged in the labor market or over members who were entry-level laborers. 6. Model 5. 2. 6. Occupation (agriculture) affected self-determination positively (B = .524. p < .05). but it did not affect couples or family members as decision-makers. When seen in combination with gender’s effects, we learn that male migrants who worked in the agricultural sector tended to be self-determining while, regardless of occupation, women were likely to be directed to move by family members or decided to move jointly with their husbands. The fact that only work in the agricultural sector had a significant effect on decision-makers may be a result of the high proportion of migrants (45.1 %) who worked in agriculture prior to migration (see Chapter 3). In general, agriculturists are imbued with traditional and patriarchal ways of thinking. The fact that male migrants who worked in the agricultural sector were mainly self-determining while their female counterparts were not might be explained by the peasantry’s adherence to traditional gender ideology about women’s inferiority and vulnerability. Although women were active participants in the agricultural labor force, they continued to be 191 viewed as subordinate and disadvantaged. Thus, the family exerted control over them, attempting to reduce the presumed risks associated with their gender. Work in the informal sector did not affect Decision-makers of Migration. In combination with gender’s effects, men tended to be self-determining while women were likely to join with their husband in making the decision to move or to be directed to move by family members. This finding can be explained, as mentioned previously, by the persistence of traditional gender ideology about women’s inferiority and vulnerability. It may also result from the small proportion of those holding occupations in this sector (12.1 %) (see Chapter 3). which led to its insignificant relation with decision-makers of migration. ,7. Model 5. 2. 7. Employment Status significantly affected who made decisions about migration. More specifically, the “other” employment status affected both self- deterrnination (B = -.659, p < .001) and joint decision-making negatively (B = -1.725, p < .001), but affected family members as decision-makers positively (B = 1.614, p < .001); unemployment and unpaid employment statuses affected family members as decision- makers positively (B = .708, p < .01, and B = 1.041, p < .001, respectively). When seen in combination with gender’s effects, both women and men with an other employment status were more likely to be directed to move by family members than to be self-determining or to join in making decisions, although men were more likely to be self-determining and women were more likely to join with their husbands in making the decision. Both women and men who were unemployed or who worked without pay before moving were more likely to be directed to move by family members than those 192 who worked for pay. The fact that people who had an other employment status were directed to move by family members rather than being self determining or joint decision- makers undoubtedly is a reflection of the great percentage (32.8 %) of migrants who were in school before moving (see Chapter 3). Implicitly, they were young, and thus the family exerted control over them in order to avoid any risks and failures that might be associated with their move. The statuses of unemployment and unpaid employment did not affect self-determination or couples as joint decision-makers, and this may reflect a household strategy of survival. in which control is exerted over disadvantaged members (with no access to earnings) in order to maximize the returns of migration. 9. Model 5. 2. 8. Social connections did not affect decision-makers of migration. Despite this insignificant relation, there was a tendency, regardless of social connections in Hanoi, for men to be self-determining and women to be directed to move by family members. This gender difference might be understood as the prevalence of ideas about female inferiority and male superiority, which shaped women’s and men’s degree of participation in decision making, regardless of social connections. 1 0. Model 5. 2. 9. Type of Relations affected self-determination (B = -.828, p < .001) and joint decision making negatively (B = -.789, p < .001), while it affected family members as decision-makers positively (B = 1.818, p < .001) (see Table 5.2.10). When seen in combination with gender’s effects, it is clear that male migrants with non-kin relations in Hanoi tended to be self-determining while female migrants with non-kin relations in Hanoi tended to be involved in joint decision-making. This result might be explained by the positive relationship between education, non-kin relations, and self- 193 determination. That is. people with high educational levels were likely to have non-kin relations (Pugliesi et al. 1998) and a high possibility of employment and well-paidjobs in Ilanoi. Hence. they had a greater voice in making the decision to move than those with low educational levels. Both women and men with kin relations in the city tended to be directed to migrate by family members. This result. as noted above, might be explained by the importance of social networks in ensuring the success of migration and of a household strategy adopted to mitigate the risks involved in moving. In sum. decision-makers of migration were affected by eight out of nine independent variables, Social Connections being the exception, i. e., almost all capital (human, social. and/inancial) and demographic attributes. Hypotheses 5. 2. b, 5. 2. c, and 5. 2.d were supported while Hypothesis 5. 2.a was rejected Results of Logistic Regression of Decision-makers of Migration by All Significant Independent Variables. The results of the regression in Table 5.2.1 1 show that of the eight significant independent variables, Gender and Marital Status (i.e., demographic attributes), Education and Employment Status (i.e., human capital), and Type of Relations (i.e., social capital) were good predictors for who made the decision to migrate. Gender (B = 1.054, p < .001) was a good predictor of self-determination. This means that men were self-determining in making the decision to migrate more than were women. Gender (B = -.967, p < .001) and Marital Status (B = -5.l976, p < .001) were good predictors of joint decision-making (couples). Married women were joint decision- 194 Table 5.2.1 1: Coefficients of Logistic Regressions for Decision-makers of Migration by All Independent Variables Variables Self(B) Couples (B) Family members (13) Demographic attributes: Gender -l.054*** -.967*** Age . . . Marital status . -5.1976*** .971 ** Financial capital: Income Human capital: Education . . -.l4l*** Occupation - Agriculture - lnfonnal sector . Employment Status — Unemployment . . .615* - Unpaid employment . . 985* - Other employment . . 1235*” Social capital: Type of relations . . -1.414* * : p < 0.05 ** : p < 0.01 ***: p < 0.001 . : insignificant relation Note : - 1n the logistic regressions, the referent group is the last category ofa variable. - The social connections variable cannot be computed because there was no variation in responses. makers while married men were self-determining. In contrast, single women and men were directed to move by family members. Marital Status (B = .971, p < .001), Education (B = -.141, p < .001), Employment Status (B = .615 for unemployment, .985 for unpaid employment. p < .05, and 1.235, p < .001 for other), and Type of Relations (B = -1 .414, p < .001) were good predictors of family members as decision-makers. Both single women and men with low educational levels and kin-relations in Hanoi, who were unemployed, worked without pay. or were students, pensioners, or disabled (“other” employment status) were mainly directed by family members to move. Married women joined their spouses in making the decision while married men decided to move by themselves. The results show that participation in decision making by women as well as men was afleeted by social capital and human capital. These results support the findings from the migration literature that emphasize the importance of human and social capital in 195 making the decision to move (De Jong et al. 1981, Douglas et al. 1992). This congruence also can be seen in the fact that migrants had high educational levels and most (88.8 %) had social connections in Hanoi before moving. The finding also is consistent with that previously assumed in theories of social and human capital about the important contribution of human and social capital to access to employment and well-paid jobs at destination (Donato 1993, Strauch 1984, De Jong et al. 1983, Herzog et al. 1983, Lim 1993). The existence of social capital among people with low education as well as high education increased their potential to find employment in the city. According to the results, gender and marital status were important factors in decision making. Even though the married women in the sample had more voice in the process than was expected, and Vietnamese women have theoretically gained rights equal to those of men, they tended to have less influence in the decision-making process than men. This might be explained by the close relation between the traditional gender division of labor and the power hierarchy within Vietnamese households. Within this hierarchy, roles and authority are assigned on the basis of gender and men have the highest status. Stage in the life cycle (marital status) intersects with gender, however Thus, married men because of their knowledge and experience, have a higher status than single men and women. Married women also are assumed to be more experienced, mature, and responsible than are single ones. Thus, married women have more voice in making decisions than single women. Financial capital was not a good predictor of who made the decision to migrate. This can be explained by the close relation between income and occupation or employment status that might overshadow the meaning of income. Further, this also 196 might be rooted in the great missing values in responses for income that blurred the impact of income on the decision-making process. SUMMARY Similar to earlier findings in the migration literature (Massey et al. 1993, Du Toit 1990, Lewis 1982. Brown et al. 1977), this study found that people in Vietnam migrated for economic reasons rather than for educational reasons, and that men and single women tended to migrate for economic reasons while married women tended to migrate for family reasons. In addition to providing support for the work of others, this study contributes to the migration literature by providing some insights into the relationships between demographic attributes of migrants, their human and social capital, and reasons for moving. Specifically, men and single women with low prestige occupations, such as in agriculture, and with non-kin relations in the city, migrated for economic reasons. Young women and men with high educational levels who worked without pay or were students (other employment status) migrated for educational reasons. Older women who were employed in the agricultural sector and who had kin relations in Hanoi, migrated for family reasons. Human capital and social capital had a strong effect on economic and family reasons for migrating while migrating for educational reasons was affected by human capital. Even though women were more likely than were men to move for family reasons, both Vietnamese women and men mainly moved for economic reasons. This result might be explained by the high poverty level and labor surplus at the migrants’ origin areas, the active participation of Vietnamese women in the labor force, and the limited economic opportunities in Vietnam’s rural areas. These circumstances pushed both men and 197 women to migrate to improve their economic situations (see Chapter 4). The findings of this study. thus. confirm the general pattern of migration found in the literature: people migrate for economic opportunities. The human capital and social capital literature on migration in developing countries reveals that people, across Third World nations with different economic, political, and cultural conditions, migrate for economic reasons (Massey et al. 1993, Du Toit 1990, Staajad 1962, Todaro 1976, Bilsborrow et al.1993). This similarity might be explained by the common structural characteristics of developing countries - great labor surplus. few rural economic opportunities, and high levels of poverty - that push people to migrate for economic rather than non-economic reasons. Economic success is highly dependent on human capital and the existence of social capital because these resources enable people to compete in labor markets and succeed in resettlement. As for decision-makers, Vietnamese migration differs from that of other countries, in which different patterns of decision making for female and male migrants were found. In contrast to other countries (Boyd 1989, Lim 1993, Chant 1992a, F awcett et al. 1984, UN 1993), Vietnamese female migrants tended to have a voice in decision-making rather than primarily being deployed by others. More specifically, different from findings in the migration literature that note that both single and married women’s choices are determined by the household (De Jong et al. 1983, William et al. 1986, Donato 1993, Hugo 1995, Robinson 1993, Riley et al.1993, Lingam 1998), married women in Vietnam were likely to make the decision to move jointly with their husbands, a testament to their influence in the process compared to that of women in other countries. This difference from the findings in the migration literature may be a result of 198 the socialist system which encouraged equal rights and opportunities for women both at home and at work. Women could empower themselves with high educational levels and money earned working in the state sector. Consequently, contemporary women have a greater voice in making decisions than did their counterparts in the past. Nevertheless, women do not have as many opportunities as do men due to (1) the obduracy of a gender ideology that shapes people’s choices and opportunities over time, and (2) their reproductive roles. Thus, even though women have gained opportunities and rights, they tend to have less voice and power in making decisions than do men. This study also found that both women and men were affected by financial, social, and human capital in making the decision to move. That is, financial capital, which has the potential to strengthen women’s position in their household, can give them a voice in making decisions. Non-kin relations in the city helped enlarge migrants’ possibility to gain employment and well-paying jobs in the city while kin relations provided them security in moving, thereby helping to ensure the success and reduce the risk of migration. Human capital demonstrated migrants’ ability to achieve economic improvement, thereby helping them improve their position within the household and participate in decision making. This combined effect of financial, human, and social capital on the Vietnamese decision-making process highlights the applicability of a theoretically integrated model in explaining migration decision making as a gendered and complex process. In addition. except for the effect of gender (discussed above), this study found effects similar to those in other developing countries in terms of demographic attributes on the decision-making process of migration (Chapin 1989, Erman 1998, Sandu et al. 199 1996, Hugo 1995, Riley et al. 1993, Boney et al. 1991, McDevitt et al. 1986, Lingam 1998). That is, regardless of age and marital status, Vietnamese men were more likely to be self-determining movers than were women. Married people tended to be self- determining while single people tended to be directed to move by family members. The decision-making process of migration in Vietnam shares some similarities with patterns found in other developing countries, as mentioned above. These similarities might result from the Vietnamese social context that has been shaped by the combined influence of Confucianism and traditional values. Vietnam, like many other Asian countries. is strongly influenced by Confucianism, which values men more than women. ldeologically, women tend to be viewed as dependent members in their households. They have a smaller voice and less power in the authoritarian hierarchy of the household than do men. The similarity in the decision-making process between Vietnam and other developing countries in Asia and Africa demonstrates patriarchal gender ideology’s resistance to change worldwide. Furthermore, Vietnamese people, both men and women, tend to rely strongly on their social networks (type of relations), especially in the migration process, to ensure the success of moving. This is so because they experience great difficulties in dealing with market mechanisms, which challenge their ways of thinking and doing, and their traditional and socialist values and norms. For example, people can no longer depend on the subsidized system to survive. Rather, they are in charge of their own lives under the market economy. They must be involved in different economic activities to support their households. Prior to the transition to a market economy, agriculture under the centrally- commanded economy. was aimed at either self-subsistence or the collective. This system 200 has changed to one in which commercial agricultural is the primary form of production; concomitantly the labor use system has changed from one of labor exchange to one of wage labor. The criteria for employment, which had been based on “revolutionary roots” or “contributions in wartime”,” have been replaced by the human capital of employees, e.g.. education and work skills. In this changing period, social networks (kin or non-kin relations) play an especially important role in helping people to update their information and knowledge of the requirements of the labor market. Therefore, Vietnamese women and men who are involved in the migration process rely very much on the assistance of their social networks. i.e.. social capital, in order to obtain the best economic opportunities in the great uncertain transitional economy of this period. 3" These are direct translations from Vietnamese expressions. “Revolutionary roots" indicates faithful service to the Party. by any family member; “Contribution in wartime” refers to military service; those who have served are given preference in economic opportunities and social positions. 201 CHAPTER 6: MOVING PROCESS Introduction This chapter aims to explore the moving process and the nature of gender differences in the migration to Hanoi of Vietnamese people. This process includes different aspects of moving such as how people move and how far they move. These aspects of moving potentially reflect socially constructed gender relations. Thus, in this chapter. I ask what factors shape female and male patterns of moving (alone or associationally) and the potential mobility of women and men. In the first section of the chapter, I focus on how people move to see if women’s and men’s migration is similar or different. More specifically, I explore if women and men move alone or with others. The way people move (alone or associationally) can be a product of a household strategy and power relations. On the one hand, it may reflect a household strategy to reduce the cost of moving by selecting the most capable family members for migration. On the other hand, embedded in this household strategy are gender relations or a power hierarchy that assign roles for women and men and, therefore, establish an order of who should move and how. To clarify these features, I examine if demographic attributes and human, financial, and social capital affect patterns of moving. In the second part of this chapter, I explore whether or not there are differences in the distance women and men move. This exploration focuses on indicators that affect the ability of women and men to move to places farther or nearer to their home areas. This ability might reflect migrants’ education (human capital) and / or their social networks (social capital) that provide assistance in moving. Moreover, this ability may reflect the position of women and men as subordinate or dominant, which shapes the distance 202 women and men may move. This focus, thus, helps uncover the nature of the geographical mobility of women and men. Specifically, I focus on the interrelation between distance of move and migrants’ human, social, and financial capital, and individual characteristics. In the conclusion, I summarize the results of the analyses of male and female moving patterns and confirm if female migrants experience different patterns of moving from male migrants in Vietnam, i.e., if they move with others or a shorter distance compared to men. The theoretical framework, literature review of the moving process, and methodology were presented in Chapters 2 and 3. In this chapter, I interpret the results of the analyses and test hypotheses. Similar to the analytical procedure used in Chapter 5, I analyze the data in three steps. First, in bivariate analyses, I use Chi-Square and ANOVA or Correlation tests to determine whether there is an association between Type of Migration and Distance of Migration and the independent variables. The results can be seen in Tables 6.1.1 to 6.1.9 (for Type of Migration) and Tables 6.2.1. to 6.2.7 (for Distance of Migration). Second, to investigate how gender influences the relation between Type of Migration and Distance of Migration, and eight other independent variables, I use logistic and linear regression analyses. In these multivariate analyses, I examine the strength of relationships between Type of Migration and Distance of Migration and the individual independent variables, controlling for gender, and test Hypotheses 6.1.a though 6.1.d and 6.2.a through 6.2.c. The results are shown in Tables 6.1.10 and 6.2.8. Finally, to determine the best predictors of Type of Migration and Distance of Migration, I run logistic and linear regressions for these dependent variables with all significant independent variables. The results are presented in Tables 6.1.11 and 6.2.9. RESULTS AND DISCUSSION 1. Type of Migration Chi-Square Results of Type of Migration bv Six Nominal Variables 6.]. 1. Gender Migrants in the sample (both women and men) tended to migrate alone (57.6 %), rather than with other people (42.2 %) (see Table 6.1.1). Nevertheless, men tended to migrate alone more than women (64.6 % for men and 51.4 % for women). Gender and Type of Migration were associated with each other (x = 14.389, p < .001). Table 6.1.1: Cross-tabulations of Type of Migration by Gender Type of Migration Female Male All % N % N % N Non-solitary 48.6% 36.2% 42.2% 207 134 341 Solitary 51.4% 64.6% 57.6% 219 245 464 Totals 100% 100% 100% 426 379 805 7. 2 14.389, p :000 6.1.2. Marital Status As can be seen in Table 6.1.2, both single women and men moved alone (56.3 % for women and 68.1 % for men) rather than with other people (43.4 % for women and 31.9 % for men). Married migrants, however, had a different type of migration. While married men migrated alone more than with others (60.2 % compared to 39.8%), married women were involved slightly more in associational than in solitary movement (53.0 % compared to 47.0 %). Regardless of marital status, men moved alone (68.1 % for single and 60.2 % for married) more than did women (56.3 % for single and 47.0 % for 204 married). Marital Status was associated with Type of Migration (x = 8.350, p < .05). Table 6.1.2: Cross-tabulations of Type of Migration by Marital Status Type of Never married Ever married migration Female Male Female Male % N "/0 N 0/o N % N Non-solitary 43.4% 31.9% 53.0% 39.8% 85 68 122 66 Solitary 56.3% 68.1% 47.0% 60.2% 1 1 1 145 108 100 Totals 100% 100% 100% 100% 197 213 230 166 x = 8.350, p =.()04 6. l . 3. Occupation From the data in Table 6.1.3, it is clear that both women and men working in the agricultural sector prior to migration moved alone (53.1 % for women and 61.8 % for men) rather than with others (46.9 % for women and 38.2 % for men). Table 6.1.3: Cross-tabulations of Type of Migration by Occupations Type of Agriculture lnforrnal sector Formal sector migration Female Male Female Male Female Male % N % N % N % N % N % N Non-solitary 46.9% 38.2% 54.5% 80.0% 66.7% 24.3% 38 21 6 12 34 18 Solitary 53.1% 61.8% 45.5% 20.0% 33.3% 75.7% 43 34 5 3 17 56 Totals 1 00% 100% 100% 100% 100% 100% 81 55 1 1 15 51 74 x = 6.897, p =.032 The picture, however. was different for women and men with jobs in the informal and formal sectors. Women working in the informal sector, like those working in the formal sector, tended to move associationally rather than alone (54.5 % compared to 45.5 % for women in the informal sector; and 66.7 % compared to 33.3 % for women in the formal sector). Men working in the informal sector, in contrast, moved associationally rather than alone (80.0 % compared to 20.0 %) while those working in the formal sector, like those working in the agricultural sector, moved alone more than associationally (75.7 % 205 compared to 24.3 %). Occupation was associated with Type of Migration (x = 6.897, p < .05). 6. [.4. Employment Status The data in Table 6.1.4 show that, regardless of employment status, men moved alone more than did women (over 60.0 % for men and less than 60.0 % for women, except those who were unemployed). While men always made up a higher proportion of those moving alone (69.0 % for unemployment, 80.0 % for unpaid employment, 62.7 % for other. and 64.7 % for paid employment), only Women who worked for pay moved alone (46.1 %) less than associationally (53.7 %). Women who were unemployed (61.8 %), working without pay (56.2 %), and held an other employment status (51.6 %) before migrating moved alone rather than associationally. The relation between Employment Status and Type of Migration was not significant (x = 3.557, p > .1). Table 6.1.4: Cross-tabulations of Type of Migration by Employment Status Type of Unemployment Unpaid employment Other Paid employment migration Female Male Female Male Female Male Female Male % N % N % N % N % N % N % N % N Non-solitary 38.2% 31.0% 43.8% 20.0% 48.4% 37.3% 53.7% 35.3% 21 18 14 1 93 62 79 53 Solitary 61.8% 69.0% 56.2% 80.0% 51.6% 62.7% 46.1% 64.7% 34 40 18 4 99 104 68 97 Totals 100% 10 % 100% 100% 100% 100% 100% 100% 55 58 32 5 192 166 147 150 x = 3.557, p =.313 6. l . 5. Social Connections As can be seen in Table 6.1.5, regardless of the existence of social connections in Hanoi. men moved alone rather than with others (66.1 % for those with social connections and 53.2 % without social connections). While women also tended to move alone, the difference between those with and without social connections was less clear cut. Women without social connections in Hanoi migrated alone (62.8 %) rather than 206 associationally (37.2 %), in fact, a higher proportion of women lacking ties in Hanoi moved alone (62.8 %) than did their male counterparts (53.2 %). Women with social connections. in contrast, migrated both alone (50.1 %) and with others (49.1 %). Social Connections were not associated with Type of Migration (x = .001, p > .1). Table 6.1.5: Cross-tabulations of Type of Migration by Social Connections Type of Social Connections Yes Social Connections No migration Female Male Female Male % N % N % N % N Non-solitary 49.9% 33.7% 37.2% 46.8% 191 1 12 16 22 Solitary 50.1% 66.1% 62.8% 53.2% 192 220 27 25 Totals 100% 1 00% 1 00% 100% 383 332 43 47 x = .001, p =.978 6. l . 6. Type ofRelations From the data in Table 6.1.6, it is clear that both women (52.7 %) and men (68.3 %) with kin relations in Hanoi migrated alone rather than with others (47.3 % for women and 31.7 % for men). Table 6.1.6: Cross-tabulations of Type of Migration by Type of Relations Type of Kin relations Non-kin relations migration Female Male Female Male % N % N % N % N Non-solitary 47.3% 31.7% 56.9% 38.2% 133 73 58 39 Solitary 52.7% 68.3% 43.1% 61.8% 148 157 41 63 Totals 100% 100% 100% 100% 281 230 102 102 x = 3.126, p :077 For those with non-kin relations in the city, women moved with others (56.9 %) more than alone (43.1 %) and they were more likely to follow this pattern than were their male counterparts (38.2 % compared to 61.8 %). Regardless of type of relations, men were more likely to move alone (68.3 % for those with kin relations and 61.8 % for those with non-kin relations) than with others (31.7 % with kin relations and 38.2 % with non-kin relations). Type of Relations was not associated with Type of Migration (x = 3.126, p > .05). In short. Type of Migration was associated with three out of six nominal variables: Gender, Marital Status, and Occupation. ANOVA Results of Type of Migration by Three Ordinal Variables 6.1. 7. Age From the data in Table 6.1.7, we can see that male migrants who moved alone tended to be youngest in the sample (mean = 29.3) while women who moved with others tended to be oldest (mean = 32.4). Both women and men who moved with others were slightly older (mean = 32.4 for women and 31.2 for men) than those who moved alone (mean = 31.2 for women and 29.3 for men). Nevertheless, the difference was not great and Age was not associated with Type of Migration (F = 2.807, p > .05). Table 6.1.7: ANOVA Results of Type of Migration by Age Type of Migration Female (mean) Male (mean) Total (mean) Non-solitary 32.4 31.2 31.8 Solitary 31.2 29.3 30.2 Totals 31.8 30.0 30.9 F and p value F = 2.807, p = .094 6.1.8. Income The data in Table 6.1.8 show that men who moved with others tended to be the richest (mean = 280K) in the sample while women who moved alone tended to be the poorest (mean = 151K). In fact, men who moved with others were nearly twice as wealthy as women who moved alone (mean = 280K for men and 151K for women). while women who moved with others were nearly as poor as men who moved alone 208 (mean = 217K for women and 241K for men). Regardless of type of migration. men were always richer (mean = 254K) than were women (mean = 186K). Income was not associated with Type of Migration (F = 3.123, p > .05). Table 6.1.8: ANOVA Results of Type of Migration by Income Type of Migration Female (mean) Male (mean) Total (mean) Non-solitary 217K 280K 249K Solitary 151K 241K 196K Totals 186K 254K 220K F and p value F = 3.123, p = .078 6.1.9. Education As can be seen in Table 6.1.9, both women and men who moved alone had higher educational levels (mean = 10.1 for women and 1 1.3 for men) than did those who moved with others (mean = 9.3 for women and 9.9 for men). Moreover, men tended to have higher educational levels (mean = 10.8) than women (mean = 9.7). Education was associated with Type of Migration (F = 25.437, p < .001). Table 6.1.9: ANOVA Results of Type of Migration by Education T, 0 Migration F (mean M (mean) Tota (mean) Non-soli . 9.9 9.6 So Itary . 1 T s . 10.2 Fand value F=25.437, =.000 In short, Type of Migration was associated with only one out of three ordinal variables, Education. Therefore, combining the results of the Chi-Square and ANOVA tests, it is clear that Type of Migration was associated with four out of nine independent variables.“ Gender and Marital Status (i.e., demographic attributes), and Occupation and Education (i. e.. human capital). Results of Logistic Regressions for Type of Migration by Each of Eight Independent Variables, Controlling for Gender, and Tests of Hypotheses As can be seen in Table 6.1.10, Type of Migration was significantly affected by Gender and Marital Status (demographic attributes), Income (financial capital), Education and Occupation (informal sector) (human capital). and Type of Relations (social capital). Noticeably. Gender affected Type of Migration negatively. The explanation for gender’s significant effect is offered in Model 6.1.1. This explanation is applicable for all those instances in which gender had a negative effect on the dependent variable. 1. Model 6.1.]. It is clear that Gender had a significant influence on how people moved. More specifically. gender affected type of migration negatively (B = -.547, p < .001). i.e.. men were more likely to move alone while women were more likely to move with others. This result supports Hypothesis 6.1.a which states that men move alone and women move with others. Supporting past research in the migration literature about an associational pattern of female migration, this finding might be rooted in women’s reproductive and subordinate roles and their presumed disadvantage in the labor market (Chant 1992. Shah et al. 1984. Thadani et a1. 1984, Lim 1993). 2. Model 6.1.2. Age did not affect Type of Migration (B = -.007. p > .05). Despite this insignificant relation, there was a tendency that, regardless of age, men tended to move alone while women moved with others. This non-significant relation might be explained by the fact that the migrants in this study mainly were clustered in the 20 to 30 age group. The sample’s homogeneity tended to be reflected in people’s moving pattern, i.e., solitarily. As noticed in the migration literature, children and people of very old ages Table 6.1.10: Coefficients of Logistic Regressions of Type of Migration by Each of Eight Other Independent Variables, Controlling for Gender Model Variable Type of Migration (B) 1 Gender -.547*** 2 Gender -.535*** Age -.007 3 Gender -.514*** Marital Status .368“ 4 Gender -.945*** Income -.002* 5 Gender -.460*** Education 096*” 6 Gender -.855*** Occupation - Agriculture .089 - lnfonnal sector -1.186* 7 Gender -.563*** Employment Status - Unemployment .420 - Unpaid employment .368 - Other employment .071 8 Gender -.548*** Social Connections -.026 9 Gender -.688"* Type of relations 338* * :p<0.05 ** :p<0.01 ***:p<0.001 Note: In the logistic regressions, the referent group is the last category of a variable. For example. the formal sector becomes the referent to which other occupations are compared. tend to be tied movers (Morrison et a1. 1988. Guest 1993) and, in this study, the proportion of those under 15 and over 60 was too small (3.6 % and 4.8 %, respectively) to offset the predominant pattern of people in the 20 to 29 age group (more than 40 %) (see Table 4.4). In addition, the fact that, regardless of age, people were involved in solitary migration might result from the presence of social connections in Hanoi (88.8 % of migrants had ties in the city prior to moving) who could provide needed assistance to ensure the success of moving. Thus, both younger and older people tended to move alone, assured that help would be forthcoming upon their arrival in Hanoi. 3. Model 6.1.3. Marital Status affected Type of Migration positively (B = .368, p < .05). In combination with gender’s effect. male and single female migrants moved alone, while married female migrants moved with others. This result supports Hypothesis 6.1.b which states that single women and men move alone while married women move with others. The similarity of the Vietnamese migration pattern with that of other developing countries might be seen as a household strategy and the gender roles of single and married people. especially women (Hugo 1993, Cooke et a1. 1996, Lim 1993. Thadani et al. 1984). On the one hand, single women may have moved alone because their parents were aware of urban employers’ perception that such women’s commitment to work is greater than that of married women. Relatively secure that their daughters would find employment in the city, parents deployed them to Hanoi by themselves to seek work. On the other hand, single women might have migrated to the city alone because, unlike married women, who were burdened with a myriad of tasks associated with their roles as wives or mothers, single women’s labor was less critical to the maintenance of the household. They thus were able to leave the household while married women were not. Alternatively, differences between married women and single women and men (both single and married) might be rooted in gender ideology, which tends to emphasize that married women’s main roles are non-economic (care for housework and the internal affairs of their households) rather than economic. Even though women might also assume economic roles after moving, they moved with others so that they could maintain the household at destination. Another explanation for the differences between married and single women and both married and single men may be their different goals in the migration process. Single women and men moved alone to cities or industrial centers in order to pursue higher education or well-paid occupations in accord with their high educational levels, while married women who were considered homemakers moved for family reasons and married men who were thought to be breadwinners moved for economic reasons (see Table 5.1.2). 4. .I-l/lodel 6.1.4. Income had a negative effect on Type of Migration (B = -.002, p < .05). In combination with gender’s effect. this result means that both women and men with lower income levels migrated alone while those with higher income levels migrated with others. This result might be explained by the fact that poor people, in order to reduce the cost of moving, tended to migrate alone rather than with family. Moreover, even though maintaining one unified household at destination instead of two in different places could reduce the cost of migration, those with low incomes could not afford an associational move that required the immediate expenditure of a large amount of money. Conversely. people with higher income levels could afford the cost involved in moving as a group. 5. Model 6.1.5. Education had a positive effect on Type of Migration (B = .096, p < .001). More specifically. in combination with gender’s effect, both women and men with higher educational levels tended to move alone while those with lower educational levels tended to migrate with others. This result might be explained by the greater opportunities in employment and earnings available to people with high educational levels. It might be easier for them to start a new career or to resettle alone rather than if their family members moved with them to the city. 6. Model 6. 1. 6. Occupation affected Type of Migration negatively (B = -1.186 for informal sector. p < .05). When seen in combination with gender’s effect, both women and men working in the informal sector moved associationally rather than alone. This N b) result might be explained by the fact that women and men who worked in the informal sector could afford to move associationally because they had high income levels (mean = 356K) compared to those working in the agricultural and formal sectors (means = 105K and 308K. respectively) (see Appendix 3.2). Working in the agricultural sector did not affect how people moved and this can be explained by the fact that even though these migrants had lower educational and income levels compared to those working in other sectors (see Appendices 3.1 and 3.2). with the assistance of social connections at destination (88.8 % of the migrants had social connections before moving). these people could move both alone and associationally. 7. Model 6.1. 7. Employment Status had no effect on Type of Migration. This insignificant effect might be rooted in the fact that the migrants in the sample mainly moved for economic reasons. i.e., in search of employment in the city. Therefore, their employment status prior to migration did not affect their pattern of move. Moreover, moving alone might be a way to increase opportunities to obtain employment and to reduce the risks and costs of moving. People could move back to their rural areas if they did not succeed economically with migration. 8. Model 6.1.8. Social Connections had no effect on Type of Migration. Despite this insignificant relation, there was a tendency that, regardless of social connections, women and men moved alone rather than associationally. Hypothesis 6.1.c, which states both men and women with social connections at destination move alone, was thus rejected. This insignificant result might be rooted in the fact that 88.8 percent of migrants had social connections in Hanoi before moving; the variation was small. 9. Model 6.1.9. Type of Relations affected Type of Migration positively (B = .338, p < .05.). When seen in combination with gender’s effect, women with kin relations and men with kin and non-kin relations in Hanoi tended to move alone while women with non-kin relations in the city tended to move with others. This result supports Hypothesis 6.1.d which states that men, regardless of type of relations, move alone while women with kin relations move alone and those with non-kin relations move with others. The similarity between this result and findings in the migration literature might be seen as a product of a household strategy to ensure the success of moving and to mitigate the presumed vulnerability of migrants, especially women, in the migration process (Huang 1983, Findley et al. 1993). Both kin and non-kin were important in providing assistance in the moving process. Nevertheless, the nature of this assistance was different: the help of kin included obligations and responsibilities based on familial ties and, thus. was more certain and secure than that of non-kin, which was voluntary in nature, and, thus, more uncertain and insecure. With assistance from kin available, both women and men moved alone with an expectation of success in the city. In contrast, people with non-kin relations in Hanoi, especially women, tended to move associationally and this may be seen as a household strategy to deal with the insecure nature of assistance from non-kin relations as well as a protection for women due to their presumed vulnerability. Moreover, as can be seen in Appendix 6.1, those with non-kin relations were richer than those with kin relations in the city and, as we saw in Model 6.1.4, richer people tended to move associationally. Therefore, the fact that people with kin relations tended to move alone and those with non-kin moved associationally is not a surprising result. In sum, the results of the logistic regression supported Hypotheses 6.1.a, 6.1.b, and 6.1.d. but rejected Hypotheses 6.1.c. That is. regardless of social connections, men and single women with high educational levels and kin relations in Hanoi moved alone while married women and those who had worked in the informal sector prior to migration, had low educational levels, and had non-kin relations in Hanoi moved with others. The finding that men moved alone was not unexpected. Nor was the difference in the moving pattern of married and single women unexpected since this finding supports those in the migration literature (Hugo 1993, Boyd 1989, Piampiti 1984). The difference may be seen as product of women’s different roles and power compared to men, that are located in their gender, stage of the life cycle, and age in a patriarchal society. Results of Logistic Regssion for Type of Migration by All Significant Independent Variables The data in Table 6.1.11 clearly show the significant relationships between Type of Migration and Gender, Marital Status, and Type of Relations. That is, of the six independent variables that had a significant effect on type of migration, these three variables were the best predictors of type of migration. Gender affected type of migration negatively (B = -1.490, p < .001), i.e., men were more likely to move alone while women tended to move with others. Marital status affected type of migration positively (B = 1.142. p < .01). This means that single people were more likely to move alone while married people tended to move with others. Type of relation affected type of migration positively (B = .687, p < .05). People with kin relations in Hanoi tended to move alone while those with non-kin relations in the city Table 6.1.1 1: Coefficients of Logistic Regression of Type of Migration by All Independent Variables Variables Type of Migration (B) Demographic attributes: Gender -1 490*" Age . Marital Status 1.142" Financial capital: Income Human capital: Education Occupation - Agriculture - Informal sector Social capital: Type of Relations .687* * : p < 0.05 ** : p < 0.01 ***: p < 0.001 . : insignificant relation Note: The Employment Status and Social Connections variables cannot be computed because there was no variation in the responses. were involved in associational movement. Demographic attributes and social capital played an important role in determining type of migration for women and men while human and financial capital were not good predictors of the way people moved. The fact that gender and marital status were good predictors of type of migration, as discussed previously. might be explained by the gender roles of married and single people as well as a socially constructed perception about women’s and men’s labor commitment. Gender ideology tended to influence evaluations of the presumed abilities of women and men strongly and, thus, shape the way they moved. Women are assumed to be less capable than men, despite the fact that gender ideology does not measure their actual capability. Moreover, social capital. not human or financial capital, was a good predictor of type of migration and this may be because moving alone or associationally is a household strategy that attempts to control the vulnerability of migrants in the moving process. The results once again help confirm 217 the influence of gender relations and a power hierarchy on people’s movement as well as the importance of social networks in the migration process. 2. Distance of migration As described in Chapter 3, migrants in the sample mainly came from nearby provinces within the Red River Delta region. The average distance of their move was about 148 kilometers (see Appendix 6.2), a rather short distance compared to that of other Vietnamese interregional migrations with an average of 600 kilometers (Dang Nguyen Anh 1998a). Therefore. it is expected that the differences in distance of move will not be great in the results of the analyses. ANOVA Results of Distance of Migration by Six Nominal Independent Variables 6. 2.1. Gender The data in Table 6.2.1 show that women migrated shorter distances (mean = 136.0 km) than did men (mean = 162.4 km). Nevertheless, Gender was not associated with Distance of Migration, as the ANOVA test shows (F = 1.1 17, p > .1). Table 6.2.1: ANOVA Results of Distance of Migration by Gender Female (mean) Male (mean) All (mean) Distance of Migration 136.0 162.4 150.0 419 370 789 F and p value F = 2.205, p = .138 6. 2. 2. Marital Status As can be seen in Table 6.2.2, single migrants moved slightly shorter distances (mean = 140.1 km) than married ones (mean = 159.9 km). Both single and married women moved approximately the same distance (mean = 133.7 km and 136.1 km, respectively), and the distances they moved were slightly shorter than those of men (mean = 136 km compared to 162.4 km). Married men moved a farther distance than single men (mean = 183.6 km compared to 146.2 km). Marital Status was not associated with Distance of Migration, as the result of the ANOVA test shows (F = .714, p > .1). Table 6.2.2: ANOVA Results of Distance of Migration by Marital Status Marital Status Female (mean) Male (mean) All (mean) Never married 133.7 146.2 140.1 Ever married 136.1 183.6 159.9 Totals 136.0 162.4 150.0 419 370 789 F and p value F=.714,p=.398 6. 2. 3. Occupation From the data in Table 6.2.3. we can see that people with agricultural occupations prior to migration moved the shortest distance (mean = 97.8 km), while those who worked in the formal sector moved the farthest distance (mean = 276.3 km). Female and male migrants with jobs in the agricultural and informal sectors tended to move the same distance (means = 98.3 km for women and 97.0 km for men in agriculture; means = 120.5 km for women and 102.9 km for men in the informal sector). Nevertheless, male migrants who had worked in the formal sector prior to migration moved much farther than their female counterparts (means = 356.4 km for men and 166.8 km for women). Occupation was associated with Distance of Migration (F = 10.370, p < .001). Table 6.2.3: ANOVA Results of Distance of Migration by Occupation Occupation Female (mean) Male (mean) All (mean) Agriculture 98.3 97.0 97.8 Informal sector 120.5 102.9 110.0 Formal sector 166.8 356.4 276.3 Totals 123 .9 224.5 173.6 144 286 F and p value F = 10.370. p = .000 6. 2. 4. Employment Status As can be seen in Table 6.2.4, both women and men working for pay before migrating moved the farthest distance (mean = 171.0 km) while those working without pay moved the shortest distance (mean = 87.9 km). Except for those having an “other” employment status, men moved a slightly farther distance than did women (means = 162.4 km for men and 136.0 km for women). The difference between the distance of women and men’s moves was greatest for unemployed people (means = 108.9 km for women and 124.4 km for men) and for those who worked for pay (122.5 km for women and 220.2 km for men). Employment Status was not associated with Distance of Migration (F = 1.955, p > .1). Table 6.2.4: ANOVA Results of Distance of Migration by Employment Status Employment Status Female (mean) Male (mean) All (mean) Unemployment 108.9 124.4 117.1 Unpaid employment 87.6 89.8 87.9 Other 161.] 128.2 145.8 Paid employment 122.5 220.2 171.0 Totals 136.0 162.4 148.2 419 370 789 Fand pvalue F=1.955.p=.119 6. 2. 5. Social Connections The data in Table 6.2.5 show that migrants who had social connections in Hanoi moved slightly farther than those without social connections (means = 150.0 km and 136.0 km, respectively). Men, regardless of social connections, moved a slightly farther distance than did women (means = 162.4 km for men and 136 km for women). In addition, men with and without social connections moved a similar distance (means = 161.3 km and 169.6 km, respectively) while women with social connections moved slightly farther than those without social connections (means = 138.7 km compared to ix.) Is.) 0 102.4 km). Social Connections had no association with Distance of Migration (F = .102. p>.l‘). Table 6.2.5: ANOVA Results of Distance of Migration by Social Connections Social Connections Female (mean) Male (mean) All (mean) Yes 138.7 161.3 150.0 No 102.4 169.6 136.0 Totals 136.0 162.4 143.0 419 370 789 F and p value F = .162, p = .688 6. 2. 6. Type ofRelations From the data in Table 6.2.6. we can see that, regardless of type of relations in the city. women and men migrated almost the same distance (means = 149.6 km for those with kin relations and 154.4 km for those with non-kin relations). Moreover, regardless of type of relations. men moved farther than did women (means = 161.3 km for men and 138.7 km for women). Women with kin relations tended to move a shorter distance than those with non-kin relations in Hanoi (means = 130.5 km compared to 164.4 km). Men who had non-kin relations in Hanoi tended to move a shorter distance than those with kin relations there (means = 144.3 km compared to 168.7 km). Table 6.2.6: ANOVA Results of Distance of Migration by Type of Relations Type of Relations Female (mean) Male (mean) All (mean) Kin relations 130.5 168.7 149.6 Non-kin relations 164.4 144.3 154.4 Totals 138.7 161.3 150.0 376 323 699 F and p value F = .066. p = .797 Men with kin relations in the city moved approximately the same distance as women with non-kin relations there (means = 168.7 km and 164.4 km, respectively) while men with non-kin relations moved almost a similar distance as did women with kin relations (means =144.3 km and 130.5 km, respectively). Type of Relations was not associated with Distance of Move (F = .066. p > .1). To summarize, Distance of Migration was associated only with Occupation. i. e., human capital. All other nominal variables were not associated with Distance of Migration. Correlation Results of Distance of Migration by Three Ordinal Variables. The results of the correlations are presented in Table 6.2.7 and show that all three ordinal variables were significantly associated with Distance of Migration. Table 6.2.7: Correlation coefficients of Distance of Migration by Age, Income, Education, and Work Experience Variable Female (r) Male (r) Total (r) Age .069 .140W .097W Income .263M .124 .189" Education .086 .128* .| 12'” * :p<.05 ** :p<.01 6. 2. 7. Age Age was associated with Distance of Migration positively (r = .097, p < .01). Older people tended to move farther distances than younger people, and this was especially true for male migrants (r = .140. p < .01). 6. 2. 8. Income Income was associated with Distance of Migration positively (r = .189, p < .01). This means that people with higher income levels tended to move a farther distance than those with lower income levels, and this was especially true for women (r = .263, p < .01). Ix) Ix.) Ix) 6. 2. 9. Education Education was associated with Distance of Migration positively (r = .112, p < .01). This was especially true for men (r = .128, p < .01). This means that people. particularly men, with high educational levels tended to move a farther distance than those with low educational levels. In sum, distance of move was significantly correlated with all three ordinal variables. In combination with the results from the ANOVA tests, it is clear that Distance of Migration was associated with four out of nine independent variables: Age (demographic attributes), Income (financial capital), and Education and Occupation (human capital). Results of Linear Regressions for Distance of Migration by Each of Eight Independent Variables, Controlling for Gender, and Tests of Hypotheses The data in Table 6.2.8 show that Distance of Migration was significantly affected by Age, Income, Education, and Occupation (agricultural and formal sectors). That is, distance of move was affected by demographic attributes, and financial and human capital. but not by social capital. Noticeably. Gender had no effect on Distance of Migration. The explanation for gender’s non-significant effect is offered in Model 6.2.1. This explanation is applicable for all those instances in which gender had no effect on the dependent variable. 1. Model 6. 2. l . Gender had no effect on Distance of Migration (B = 26.65, p > .05) (see Table 6.2.8). Both women and men tended to migrate the same distance. This result does not support Hypothesis 6.2.a which states that men move a farther distance than do women. The rejection of this hypothesis might be explained by the 223 Table 6.2.8: Coefficients of Linear Regressions of Distance of Migration by Each of Eight Other Independent Variables. Controlling for Gender IQ Id 1:; Model Variable Distance of Migration (B) 1 Gender 26.65 2 Gender 30.19 Age 1.78" 3 Gender 28.38 Marital Status 17.7 4 Gender 72.07 Income .302" 5 Gender 17.74 Education 8.27" 6 Gender 24.01 Occupation - Agriculture -59.07* Gender 27.37 - Informal sector -43.25 Gender 17.18 - Formal sector 147.96*** 7 Gender 27.98 Employment Status - Unemployment -38.06 Gender 23.18 - Unpaid employment -55.2O Gender 26.61 - Other -4.09 Gender 25.26 - Paid employment 34.60 8 Gender 27.04 Social Connections 14.47 9 Gender 21.50 Type of relations 3.79 * :p'~“ 0.05 ** :p<0.01 ***:p<‘0.001 fact that most migrants (63.9 %) came from provinces which were within 100 kilometers of Hanoi (see Chapter 3), and the mean distance of move was 148 kilometer (see Appendix 6.2). Because of this short distance, a gender difference in distance of move did not emerge. 2. Model 6. 2. 2. Age affected Distance of Migration positively (B = 1.78, p < .01). Regardless of gender, older people tended to move a farther distance than younger people. This result might be explained by the curvilinear relation between age and education. Older people (age 36 and over) tended to include the highest proportion (35.7 %) of those who had a university education before moving compared to those of other age groups; the relative percentages for those ages 21-25 and 26-35 were 27.1 percent and 17.4 percent (see Appendix 6.3). Consequently, people of older ages tended to have higher educational levels compared to younger people and thus, they tended to move a farther distance. The similarity between younger and older women might be explained by the fact that, even though older women were more likely to participate in decision making than younger women (see Chapter 5, Model 5.2.2), the distance women moved was still strongly influenced by family members because of women’s presumed vulnerability. 3. Model 6. 2. 3. Marital Status did not affect Distance of Migration (B = 17.7, p > .1). This result may be, as mentioned above, a product of the short distance of the migrants’ rural homes from Hanoi. The result may also have roots in the high prevalence of social networks among migrants (88.8 %). Both married and single women and men moved, regardless of distance, because they knew people at their destination who potentially could help them. 225 4. Model 6. 2. 4. Income affected Distance of Migration positively (B = .302, p < .01). Both women and men with higher income levels tended to move a farther distance than those with lower income levels. Hypothesis 6.2.b, which states that people with higher income levels move a farther distance than those with lower income levels, was thus supported. The effect of income might be explained by the fact that people with higher income levels were better able to cover the cost of a long distance move than were those with lower incomes. Certainly, this result is consistent with the finding from the migration literature that emphasizes the important role of financial capital in covering the cost of a long distance move (Lewis 1982. Ringdal 1993, Hoffman et al. 1992, Landry et al. 1988). 5. Model 6. 2. 5. Education affected Distance of Migration positively (B = 8.27, p < .01). That is, both women and men with higher educational levels tended to move a farther distance than those with lower educational levels. This result supports Hypothesis 6.2.c which states that people with high educational levels move a farther distance than those with low educational levels. It is also consistent with the findings from the migration literature that reveal the positive relation between educational levels, distance of move, and economic returns of migration (Hugo 1993, Baker 1975, Trzcinski et a1. 1991). Compared to those with low educational levels, people with high educational levels tended to move longer distances in search of the best match for their educational levels. They have more opportunities for employment and a better change of obtaining well-paid jobs than those with low educational levels. 6. Model 6. 2. 6. With the exception of those who worked in the informal sector, the occupation of migrants prior to their move affected the distance they moved. h.) Ix) O Specifically. work in the agricultural sector affected distance of move negatively (B = - 59.07. p < .05) while employment in the formal sector affected distance of move positively (B = 147.96. p < .001). This means that both women and men who had worked in the agricultural sector before migration tended to move a shorter distance while those who had had jobs in the formal sector tended to move a farther distance. Hypothesis 6.2.d, which states that people working in agriculture tended to move a shorter distance than those working in other occupations, was thus supported. The similarity between the finding of this study and those found in the migration literature might be seen as a reflection of the close relation between low prestige occupations such as those in agriculture and low educational levels (see Appendix 5.2) 1 that limit people’s ability to consider alternatives (Hou et al. 1994, Chattopadhyay 1998). Consequently, people holding low prestige occupation do not move as far as those with high prestige occupations. The same rationale might be used to explain the insignificant relationship between work in the informal sector and distance of move. People working in the informal sector had higher educational and income levels than those working in the agricultural sector; they thus tended to move a farther distance. Although their income levels were higher, however. their educational levels were lower than those of people working in the formal sector. Therefore. it is understandable that people in the informal sector moved both short and long distances. 7. Model 6. 2. 7. Employment Status had no effect on Distance of Migration. Despite this insignificant effect, there was a tendency that both women and men tended to ' Prestige of occupations is measured based on the income and educational levels of those who hold that particular occupation (Blau et a1. 1967). In this study, those who held agricultural occupations before moving had the lowest income (mean = 104K for those working in agriculture, 331K in the informal sector 227 move the same distance, regardless of employment status. This finding may be explained by the migrants’ short distance of move that tended to blur the effect of education on distance of move (see Model 6.2.4). As can be seen in Appendix 6.4, there is a close relationship between employment status and education, which, in turn, is related to migrants’ ability to move short or long distances. Both those who worked for pay and those who were unemployed had similarly high educational levels. but those working without pay had the lowest educational levels. 2 Because of the short distance involved in the movement of the sample. people with different employment statuses or different educational levels could move the same distance. 8. Model 6. 2. 8. Social Connections had no effect on Distance of Migration. Despite this insignificant relation. there was a tendency that both women and men with or without social connections in Hanoi tended to move the same distance. Hypothesis 6.2.e, which states that people with social networks at destination move a farther distance than those without social networks, was thus rejected. This result might be explained by the fact that Vietnamese rural people who moved into Hanoi tended to move regardless of social connections due to the great difference in earnings and educational opportunities between those available in their rural homes and the city. Almost two-fifths (38.6 %) of migrants in the sample moved for economic reasons and slightly less than one quarter (21.7 %) moved for educational reasons (see Chapter 5). Alternatively, almost 90 percent of migrants had social connections in Hanoi and this might be a reason for this insignificant relation. and 306K in the formal sector) and educational levels compared to those in the informal and formal sectors (means = 8.2. 10.4, and I 1.7, respectively). 2 Education means = 10.6 for the unemployed, 10.4 for those working for pay, 8.8 for those working without pay, and 10.48 for other employment status. I») 10 DO 9. lllodel 6. 2. 9. Type of Relations had no effect on Distance of Migration. Despite this insignificant relation. there was a tendency that people with both kin and non-kin relations in the city tended to move the same distance. This result might be explained by the fact that migrants in the sample mainly came from provinces close to Ilanoi. Thus. the importance of type of relations tended to be less salient than they might have been for those moving longer distances. In sum, the results of the linear regressions support Hypotheses 6.2.b, 6.2.c, 6.2.d, but do not support Hypotheses 6.2.3 and 6.2.c. That is. regardless of social connections, both women and men moved the same distance; people with higher incomes and educational levels and those working in the formal sector migrated a farther distance than those with lower incomes and educational levels and those working in other sectors. Results of Linear Regression for Distance of Migration by All Significant Independent Variables The results of the regression analysis presented in Table 6.2.9 show that distance of move was affected only by occupation (formal sector) and that this efect was significant (B = 162.71 . p < 05). This means that occupation was the best predictor of distance ofmigration. Put another word, human capital had some effect on distance of move. The fact that occupation shaped distance of migration is understandable in the Vietnamese context because people with occupations in the formal sector had higher educational levels (mean = 1 1.8) than those working in the agricultural and informal sectors (means = 8.2 and 10.2, respectively) (see Appendix 3.1). Moreover, because of their higher educational levels, they received higher incomes than those with lower ls) Ix) \O Table 6.2.9: Coefficients of Linear Regression of Distance of Migration by All Independent Variables Variables Distance of Migration (B) Demographic attributes: Gender Age Marital Status Financial capital: Income Human capital: Education Occupation - Agriculture - Informal sector . - Formal sector 16271” Social capital: Type of Relations * : p < 0.05 ** : p < 0.01 ***: p < 0.001 . : insignificant relation Note: The Employment Status and Social Connections variables cannot be computed because there was no variation in responses. educational levels (r = .383, p < .001) (see Appendix 6.5), which enabled them to subsidize a long distance move. Their higher educational levels (human capital) helped such migrants easily find jobs in different areas, especially in cities such as Hanoi, before the 19903. As a result, people with occupations in the formal sector were able to move farther than those in other sectors. Nevertheless, during the late 19803 and early 19903, with the introduction of a market economy in Vietnam, the service sector burgeoned. Along with this there was a reduction of the state (formal) sector, and consequently, a reduction in the number of state employees. Therefore, population movement during this transition period included a relative large number of people had been dismissed by state enterprises. So. the fact that occupation was the best predictor of distance of move indicates a special feature of labor migration into cities in Vietnam during this period. SUMMARY The Vietnamese migration process, as shown in this chapter, shares some similarities with and differences from the findings in the migration literature on 230 developing countries. Type of Migration was affected by five out of nine independent variables (Gender, Marital Status. Income, Education. Occupation, and Type of Relations). Nevertheless, the best predictors of Type of Migration were Gender and Marital Status (i.e.. demographic attributes) and Type of Relations (i.e., social capital). Similar to the findings from the migration literature (Hugo 1995, Morrison et al. 1988. Guest 1993, Shah et al. 1984. Thadani et al. 1984, Lim 1993, Bonney et al. 1991, Lichter 1983. Cook et al. 1996), type of migration in Vietnam was significantly affected by gender and marital status. While there were some similarities between single women and men in terms of type of migration, married women had a distinct pattern of mi gration. which might be seen as a product of their diffuse roles (economic and family) compared to the specific role (economic) of men. Women’s roles were likely to be defined as homemakers even when they were involved in economic activity while men’s roles were mainly perceived as economic. These roles of women, especially married ones. closely tied them to a household’s maintenance, and thus, they were more likely to move associationally rather than solitarily. Specifically, although both Vietnamese women and men tended to migrate alone rather than with others, men migrated alone more than women. Male and singlefemale migrants moved alone, while married female migrants moved with others. In addition, the results of this study provided some insights into the relationships between financial and human capital and type of migration, i.e., income, education, and occupation with type of migration. which have not been mentioned in previous studies in the migration literature. Specifically, both women and men with low incomes and IN) L») high educational levels tended to move alone. That is, migrants’ financial and human capital affected type of migration. Type of migration in Vietnam was also significantly affected by social capital, e.g.. type of relations. Men with either kin or non-kin relations and women with kin relations in Hanoi migrated alone, while women with non-kin relations there migrated with others. Social capital more strongly impacted women’s type of migration than men’s. With the assistance of kin relations, Vietnamese women moved alone; those with non-kin relations moved associationally. The similarity of the finding of this study with that of the migration literature (Huang 1983, Findley et al. 1993, Boyd 1989, Guest 1993. Piampiti 1984) confirms that social capital (kin relations) provides irreplaceable support for both women and men, especially women, helping them to maximize economic returns in the moving process. This result also might reflect a household strategy to ensure success in moving. Like those in many other Asian countries, Vietnamese household strategies, more or less, were shaped by a gender hierarchy and ideology about women’s abilities (Tran Thi Que 1995). Even though women were empowered by education and earning opportunities under socialism, due to the obduracy of gender ideology, women with non-kin relations moved associationally, thereby offering them protection against their presumed vulnerability. With respect to distance of migration, the results of this study indicate that Distance ofMigration was affected by Age (demographic attributes), Income (financial capital), and Education and Occupation (human capital). The Vietnamese moving pattern shares some similarities with and differences from that of other developing countries. The similarity includes the effects of income (financial capital) and education Is) b) t\) and occupation (human capital) on the distance of a move (Becker 1975, Trzcinski et al. 1991. Lewis 1982, Ringdal 1993. Hoffman et al. 1993, Landry et al. 1988). That is, both women and men with low income and educational levels and those working in the agricultural sector moved a shorter distance than their more advantaged counterparts. This similarity might be seen as a relatively universal relation between human capital and economic success. Moreover, while previous studies mainly focused on the effects of migrants’ income (financial capital or economic status), occupation and education (human capital), and social networks on distance of migration, this study examined the effects of other factors such as type of relations (one aspect of social capital), marital status and age (demographic attributes), and employment status (human capital). In this way. this study contributes to the migration literature by providing some insights into the relationship between distance of migration and other indicators. Specifically, both single and married women and men, regardless of employment status and type of relations, moved the same distance. Younger people moved a shorter distance than older people. Nevertheless, the best predictorfor Distance ofMigration was Occupation (formal sector). This means that people who worked within theformal sector of the economy moved afarther distance than those working in other sectors. Put another way, human capital had an effect on distance of move. This result truly reflects the superiority of occupations in the formal sector compared to those in other sectors, in terms of the human capital of those employed in this sector. This advantage increased people’s chance to find a suitable occupation in different places. They were able to move farther and to be more secure about finding a job than others. On the other hand, these results also may be explained by the burgeoning service sector and personnel restructuring in the state sector in Vietnam; people moved to Hanoi in search of employment because of the great differences in economic opportunities between urban and rural areas and, perhaps, because many of them were fired from the state sector or underemployed in the agricultural sector. Interestingly, gender had no effect on distance of migration. Women and men moved almost the same distance. This might be understood as the product of a common pattern of migration to Hanoi, which mainly occurred within the region (Red River Delta). i.e., within short distances. Because of this short distance, migrants probably experienced few costs and limited risks in moving and, consequently, women were able to move as far as were men. CHAPTER 7: ADAPTATION OUTCOMES OF THE MIGRATION PROCESS Introduction The adaptation process is referred to as a process in which migrants experience changes in economic and social status by moving. These changes may be positive or negative compared to conditions before moving, and they may be different for male and female migrants. This potential difference might reflect the fact that Vietnamese women experience contradictory gender relations at home and at work due to the mixed influence of the persistent gender ideology of the feudal system and the gender equality ideal of socialism. The consequences of the migration process and gender differences in the adaptation process of Vietnamese migrants in Hanoi will be examined in this chapter within the context of these contradictory relations. The outcomes of migration will be explored with respect to the migrants’ economic improvement and degree of satisfaction at destination, Hanoi. The analyses of this study must be placed within the complex context of Vietnamese society, which is changing into a market economy, but under the management of the communist party. These macro changes strongly influence a household’s strategy for survival in rural and urban areas. As noticed by others (Truong Van Phuc 1996, Le Dang Giang 1996), different economic sectors offer differential benefits (e.g., more or less pay, security, and the like). For instance, the state (formal) sector offers stable employment, safe work conditions, health insurance and pensions as well as other benefits (sick leave, maternity leave, annual vacation) while the informal sector does not. In the context of reductions in personnel in the state (formal) sector, people tend to search for employment in other economic sectors such as the informal one. 235 Most people, however, do not want to work in this sector due to its uncertain and insecure employment conditions. Nevertheless, they may have no other choice, as competition for employment in the state sector tends increasingly to be keen. Personnel restructuring in the state sector may place great pressure on migrants who are already in a vulnerable situation because of their migration status and residency change and / or regional prejudice or ethnic discrimination. Among migrants, women might experience a greater disadvantage than men due to the obduracy of gender ideology in particular and the disadvantages (noted above) that apply to migrants in general. Thus, the focus on the adaptation process will help to clarify what factors shape migration outcomes and whether female and male migrants in Hanoi experience the same or different consequences by moving. In this chapter, I pay special attention to the effects of migrants’ demographic attributes and capital (human, social and financial) on the outcomes of migration in Hanoi in order to determine what role these play in the adaptation process. More specifically, adaptation outcomes will be explored through the economic improvement of migrants in Hanoi to see if women and men experience resettlement in the city differently. In addition, adaptation outcomes will be explored through the analysis of migrants’ degree of satisfaction with their move, thus, highlighting migrants’ perception of their success in Hanoi in terms of economic improvement. The economic improvement of migrants in Hanoi will be explored by examining the differences in migrants’ current and pre-migration occupations, employment statuses, and incomes under the effects of different types of capital and demographic attributes. Degree of satisfaction with migration also is examined in relation to these factors. The 236 unclear situation of migrants in Hanoi in the transition period will be explored in terms of their economic betterment and their perceived success. The theoretical background, literature review, and measurement and analysis methods were presented in Chapters 2 and 3. Noticeably, different from other chapters, the information about migrants in this chapter mainly draws on their situation at the time they were interviewed in Hanoi, i.e., their current economic status. This chapter includes the results of the analyses. a discussion of the findings about Economic Improvement and Degree of Satisfaction, and a summary of the findings. As in Chapters 5 and 6, the strategy of analysis for this chapter is threefold. First, I provide general information about Economic Improvement and Degree of Satisfaction as well as identify if these dependent variables (ordinal scale) have an association with six nominal independent variables (Gender, Marital Status, Occupation, Employment Status, Social Connections, and Type of Relations) by using ANOVA tests. I also examine if these ordinal dependent variables have an association with three ordinal variables (Age, Income. and Education) by using Correlation tests. These tests are run separately for female and male migrants in order to compare results, which can be seen in Tables 7.1.1 to 7.1.7 (for Economic Improvement) and Tables 7.2.]. to 7.2.7 (for Degree of Satisfaction). Second, to examine the relations and the strength of the relations between each dependent variable and individual independent variables, controlling for gender, I separately run linear regression of three variables: Economic Improvement and Degree of Satisfaction by Gender and each of eight independent variables. The results are shown in Tables 7.1.8 and 7.2.8 and are used to test Hypotheses 7.1.a through 7.1.g and 7.2.a 237 through 7.2.d. Finally, to determine which of the independent variables are the best predictors of Economic Improvement and Degree of Satisfaction, linear regression is used for each dependent variable (Economic Improvement and Degree of Satisfaction) and all significant independent variables. The results are presented in Tables 7.1.9 and 7.2.9. RESULTS AND DISCUSSION 1. Economic Improvement Economic Improvement includes two change variables: Work Status Change (which is the combination of occupational and employment status changes) and Income Change. The two variables are the results of the differences between present and past occupations and employment statuses in the first case, and incomes in the second. Work Status Change is a function of change in (a) the occupational sectors in which migrants worked prior to and after migration and (b) their employment statuses pre and post migration; scores range from -6 (indicator of the least change) to 6 (indicator of the most change). Income Change is a function of money earned prior to and after migration, and it ranges from -550 thousand to 2 million dong. Two variables (current occupation and employment status and income) will be excluded in the analyses of Work Status Change and Income Change, respectively, and they thus are not included in the results of work status change or income change in Table 7.1.8. Work status change will not be included in Tables 7.1.3 and 7.1.4 and income change will not be examined in Table 7.1.6. 238 ANOVA Results of Economic Improvement by Six Nominal Variables 7. I. I . Gender The data in Table 7.1.1 show that women achieved less economic improvement by migrating than did men. While their work status change was fairly similar to that of men (mean = 1.68 for women and 1.75 for men). they continued to earn less than men (mean = 240K and 359K, respectively). Put another way. women may have been as likely as men to work for pay in the state (formal) sector after moving, but the difference between their incomes prior to and after migration was less than those of men. Gender was associated with Income Change (F = 15.852. p < .001), but was not associated with Work Status Change (F = .066, p >.1). Table 7.1.1: ANOVA Results of Economic Improvement by Gender. Female (mean) Male (mean) All (mean) Work status change 1.68 1.75 1.71 Total (N) 426 379 805 F value F = .066, p = .798 Income Change 240K 359K 300K Total (N) 241 243 484 F value F = 15.852. p = .000 7. I . 2. Marital Status From the data in Table 7.1.2. we can see that married people, both women and men, achieved less economic improvement (mean = .96 for work status change and 266K for income change) than did their single counterparts (mean = 2.80 for work status change and 350K for income change). Specifically. married women and men were less likely than single ones to work for pay in the state sector, and their incomes did not change as much as that of single people. Nevertheless. regardless of marital status, while women’s work status improved as much (mean = 1.68) as did that of men (mean = 1.75). IQ their incomes changed less by moving (mean = 240K) than those of men (mean = 359K). Marital Status was associated with Economic Improvement (F = 50.617, p < .000 for work status change; F = 7.615, p < .01 for income change). Table 7.1.2: ANOVA Results of Economic Improvement by Marital Status. Marital Status Female (mean) Male (mean) All (mean) Work status Never married 2.85 2.76 2.80 change Ever married .91 1.02 .96 Totals 1.68 1.75 1.71 F value F = 50.617, p = .000 Income change Never married 282K 415K 350K Ever married 213K 320K 266K Totals 240K 359K 300K F value F = 7.615, p = .006 .7. l . 3. Occupation Regardless of occupation, women’s incomes changed less by moving than did men’s (mean = 305K for women and 404K for men) (see Table 7.1.3). Both women and men working in the informal sector gained slightly less income change (mean = 326K) than did those working in the formal sector (mean = 364K). The result of the ANOVA test shows that Occupation was not associated with Income Change (F = 1.060, p > .1). Table 7.1.3: ANOVA Results of Economic Improvement by Occupation Occupation Female (mean) Male (mean) All (mean) Income change Agriculture . Informal sector 248K 367K 326K Formal sector 31 1K 414K 364K Totals 305K 404K 356K F value F = 1.060. p = .304 7. [.4. Employment Status From the data in Table 7.1.4, we can see that, regardless of employment status, women’s incomes changed less (mean = 240K) by migrating than did those of men (mean 240 = 359K). Women and men who had an “other” employment status in Hanoi (i.e., pensioners, disabled people, students) lost the most economically by migrating (income change mean = -332K) while both women and men working for pay in Hanoi gained the most (income change mean = 356K). Employment Status was associated with Income Change (F = 57.142. p < .001). Table 7.1.4: ANOVA Results of Economic Improvement by Employment Status Employment Status Female (mean) Male (mean) All (mean) Income change Unemployment -160K -213K -192K Unpaid employment -150K -150K -150K Other -293K -345K -332K Paid employment 301K 406K 356K Totals 240K 359K 300K F value F = 57.142, p = .000 7. I. 5. Social Connections As the data in Table 7.1.5 indicate. regardless of social connections in Hanoi, women were as likely as men to work for pay in the state sector at destination (mean = 1.68 and 1.75. respectively). but their incomes changed less (mean = 240K) than those of their male counterparts (mean = 359K). Table 7.1.5: ANOVA Results of Economic Improvement by Social Connections Social Female (mean) Male (mean) All (mean) Connections Work status Yes 1.69 1.84 1.77 change No 1.22 1.22 1.37 Totals 1.68 1.75 1.71 F value F = 1.009, p = .316 Income change Yes 243K 363K 302K No 218K 340K 290K Totals 240K 359K 300K F value F = .065, p = .799 Further, regardless of gender, migrants with social connections were more likely to work for pay in the state sector and to earn more income in Hanoi (mean = 1.77 for 241 work status change and 302K for income change ) than those without social connections there (mean = 1.37 and 290K, respectively). Social Connections, however, was not associated with Economic Improvement (F = 1.009 p > .1 for work status change, F = .065. p > .1 for income change). 7.1.6. Type of Relations The data in Table 7.1.6 show that female and male migrants with non-kin relations in Hanoi were more likely to work for pay in the state sector and to increase their incomes (mean = 2.05 for work status change and 312K for income change) than did those with kin relations there (mean = 1.59 and 296K, respectively). Moreover, women with non-kin relations were more likely to work for pay in the state sector (mean = 2.24) than were women with kin relations in the city (mean = 1.38) while the work status of men with non-kin or kin relations was similarly improved (mean = 1.86 and 1.83, respectively). Nevertheless, Economic Improvement was not associated with Type of Relations (F = 2.300. p > .1 for work status change, F = .218, p > .1 for income change). Table 7.1.6: ANOVA Results of Economic Improvement by Type of Relations Type of Relations Female Male (mean) All (mean) (mean) Work status Kin relations 1.38 1.83 1.59 change Non-kin relations 2.24 1.86 2.05 Totals 1.69 1.84 1.77 F value F = 2.300, p = .130 Income change Kin relations 229K 367K 296K Non-kin relations 268K 356K 312K Totals 243K 363K 302K Fvalue F=.218,p=.641 In short. Economic Improvement was associated with two variables: Marital Status and Employment Status. Even though women were as likely as men to work for pay in the state sector at destination. their income changed less than that of their male counterparts. Single women and men achieved more work status and income change than their married counterparts. Both female and male migrants who were disabled, pensioners. and students (e. g., “other”) lost the most economically while those who worked for pay gained the most by moving. Correlation Results of Economic Improvement by Three Ordinal Variables 7.1. 7. Age The data in Table 7.1.7 indicate that Age was associated with Economic Improvement negatively (r = -.404 for work status change and -. 196 for income change, p < .001). This means that younger women and men were more likely to work for pay in the state sector and gain more income change with migration than their older counterparts. Table 7.1.7: Correlation coefficients of Economic Improvement by Age, Income, and Education Variable Female (r) Male (r) Total (r) Work status Age -.430*** -.384*** -.404*** change Income -.019 .032 -.002 Education -.216*** -.056 -.090 Income change Age -.259*** -.195** -.196*** Education .162* .039 .1 14* * :p<’0.05 ** :p<0.01 ***:p<0.001 7. I . 8. Income As can be seen in Table 7.1.7, Income was not correlated with Economic Improvement (r = -.002, p > .1 for work status change). That is, regardless of their income levels in Hanoi. both women and men were more likely to work for pay in the state sector than in the informal sector after migrating to the city. 243 7.1.9. Education The data in Table 7.1.7 show that Education was not correlated with work status change (r = -.090. p > .1), even though women with lower educational levels were more likely to work for pay in the state sector than those with higher educational levels (r = - .216. p < .001). Education, however, was associated with income change positively (r = .1 14. p < .05). i.e., migrants with higher educational levels received more pay in the city than those with lower educational levels, and this was especially true for women (r = .162. p < .05). In sum. Economic Improvement, both work status change and income change, was negatively associated with Age (demographic attributes), while Income Change was positively associated with Education (human capital). Younger women and men were more likely to work for pay in the state sector at destination than were older people; both women and men with higher educational levels earned more by migrating than those with lower educational levels. In combination with the results of the ANOVA tests, Economic Improvement was significantly affected by Marital Status, Age, and Employment Status. This means that economic improvement was affected by demographic attributes and human capital, but not byfinancial and social capital. More specifically, work status change was associated with Marital Status and Age while income change was associated with Gender, Marital Status, Age, Employment Status, and Education. Younger and more highly educated single women and men held better jobs and received more income after moving than their older married counterparts. Regardless of marital status and employment status, however, women earned less than their male counterparts with migration. 244 Results of Linear Regressions of Economic Improvement by Each of Eight Independent Variables, Controlling for Gender, and Testing Hypotheses The results of the linear regression presented in Table 7.1.8 indicate that Work Status Change was significantly affected by Age and Marital Status while Income Change was affected by Gender. A ge, Marital Status, Education, Occupation (formal sector), and Employment Status. Noticeably. Gender had no effect on Work Status Change, but affected Income Change positively. The explanations for gender’s effects are offered in Model 7.1.]. These explanations are applicable for all those instances in which gender had a non-significant effect on work status change and a significant effect on income change. 1. Model 7. I. 1. Gender affected Income Change positively (B = 1 19.19, p < .001) but it had no effect on Work Status Change (B = -.012, p > .1). More specifically, men’s incomes changed with migration more than those of women, although both women and men were more likely to work for pay in the state sector in the city than they were in the rural area. When economic improvement is referred to as income change. Hypothesis 7.1.a, which states that men achieve more economic improvement by migrating than women, was supported. But, when economic improvement is measured by work status change. this hypothesis was rejected. In the case of the first result reported, the positive effect of gender on income change probably reflects the fact that the male migrants were more educated than the female migrants (mean = 10.95 for men and 10.09 for women) (see Appendix 7.1), enabling them to obtain more lucrative paying jobs in the city. This result is consistent with findings in the migration literature about the higher educational levels of male migrants compared to those of women and the positive relation between Table 7.1.8: Coefficients of Linear Regressions of Economic Improvement by Each of Eight Other Independent Variables, Controlling for Gender Model Variable Economic Improvement (B) Work status change Income change 1 Gender -.012 119.19*** 2 Gender .311 133.63*** Age -.116*** -6.94*** 3 Gender .002 1 17.42*** Marital Status -1.839*** -82.24** 4 Gender -.263 Income -.000 . 5 Gender -.001 109.09*** Education .01 1 12.71 * 6 Gender 119.11*" Occupation - Informal sector 31.03 Gender 114.35*** - Formal sector 174.I4*** 7 Gender 121.25*“‘ Employment Status - Unemployment -509.16** Gender 75.50" - Unpaid employment -438.59*** Gender l41.79*** - Other -689.54*** Gender 84.23“” - Paid employment . 556.75*** 8 Gender -.008 120.22*** Social Connections .403 23.32 9 Gender .144 l I9.73*** Type of relations .457 14.11 * :p 0.05 ** : p <1 0.01 ***: p < 0.001 246 education and economic success (Massey et a1. 1993, Du Toit 1990, Todaro 1976, Brown et al 1977. Lewis 1982, Vecoli 1998, Sowell 1998, Wilson 1994). This result also is consistent with findings in the migration literature about the way gender ideology shapes women’s roles and defines their presumed abilities, thus decreasing their earnings compared to those of men (Jelin 1997, Zlotnik 1995b. UN 1993). In the case of the second result reported, the insignificant relation between gender and work status change might be explained by the fact that despite the difference in women’s and men’s educational achievements, both had relatively high educational levels compared to the population at large. They thus were able to find employment and occupations in the formal sector in the city. 2. Model 7.1.2. Age had a negative effect on Economic Improvement (B = -.1 16 for work status change and —6.94 for income change, p < .001). In combination with gender’s effect. it is clear that both younger women and men obtained better jobs and earned more money with migration than did older people. Hypothesis 7.1.b, which states that younger women and men gain greater economic improvement with migration than do older people, was thus supported. This result is consistent with findings from other developing countries which indicate that migrants accrue greater economic returns when they are younger rather than older because of employers’ preference for a youthful labor force (UN Secretariat 1993, Jelin 1997. Lim 1993, Pyle 1994. Fuess et al. 1994. Aslanbeigui et al. 1994, Summerfield 1994). 3. Model 7.1.3. Marital Status affected Economic Improvement negatively (B = - 1.839. p < .001 for work status change and -82.24, p < .01 for income change). When seen in combination with gender’s effects. we learn that single women and men migrants 247 were more likely to work for pay in the state sector and to increase their incomes with migration than were married ones. This result supports Hypothesis 7.1.c which states that single people achieve greater economic improvement by migrating than do married people. This result is consistent with findings in the migration literature about the advantages of singleness in the labor market (single women and men have more opportunities for employment and earnings than married ones) and about the effect of gender ideology and roles on the economic achievement of married women; their employment and earnings are limited because of their presumed lack of labor commitment and the demands of their housework roles (Payner 1992, Chattopadhyay 1997, Lichter 1983, Portes et al.1993). Single men achieved greater economic improvement than did married ones. This finding might reflect the fact the percentage of single men who were students, pensioners, or invalids (“other” employment status) prior to migration (60.1 %) declined to 50.7 percent after moving. This decline is accounted for by the increase in the percentage (45.5 %) of single men who worked for pay after moving; only 20.7 percent of these paid workers held jobs for remuneration before migrating (see Appendix 7.2). Married men, in contrast, worked for pay both before and after moving (63.9 % compared to 78.3 %) and, therefore, their employment status did not change or improve as much as that of single men. Similarly, as single men’s employment status changed from “other” into working for pay, their income change was greater (mean = 415K) than that of their married counterparts (mean = 320K) (see Appendix 7.3). 4. Model 7.1.4. Income in Hanoi had no effect on Economic Improvement (work status change) (B = -.000, p > .1). Despite this insignificant relation, there was a 248 tendency. regardless of their income levels in Hanoi, for both women and men to work for pay in the state sector. This insignificant result might be explained by a causal relation between employment / occupation and income levels; employment and occupation generated income levels, but not vice versa. 5. 7 ode] 7.1.5. Educational levels in the city affected Income Change positively (B = 12.71. p < .05) but did not affect Work Status Change (B = -.011, p > .1). More specifically, in combination with gender’s effect, both women and men with higher educational levels achieved greater income change than those with lower educational levels, and this was especially true for women. Despite the non-significant effect of educational levels on Work Status Change, there was a tendency for both women and men to work for pay in the state sector after than before migration. When economic improvement is referred to as income change, Hypothesis 7.1.d, which states that people with high educational levels achieve greater economic improvement by migrating than those with low educational levels, was supported. But when economic irmarovement is measured by income chflge. this hypothesis was reiected. The first result reported is consistent with findings in the migration literature about the positive relation between education and economic success (Vecoli 1998, Sowell 1998. Wilson 1994, Khoo et al. 1984). In the case of the second result, the insignificant effect may be explained by the great percentage (about 45.0 %) of migrants who had high educational levels (university level), compared to that of the national population (see Chapter 4), which enabled them to obtain employment in the state sector after moving; more than 70.0 % of migrants in this study worked in the formal sector in Hanoi (see Chapter 3). 249 6. Model 7. 1. 6. Occupation at destination affected Income Change positively (B = 174.14. p < .001 for formal sector). More specifically, when seen in combination with gender’s effect, both women and men who worked in the formal sector in Hanoi achieved greater income change than those working in the informal sector. but women’s income changed less than that of men. This result might be explained by the fact that (1) migrants working in the formal sector had higher educational levels (mean = 10.82) than those who worked in the informal sector (mean = 9.35) (see Appendix 7.4). (2) And, as shown above. those with higher educational levels achieved greater income change than those less endowed. Further, and (3), the fact that work in the informal sector did not affect income change may be a product of the large amount of missing data; only a small proportion of migrants working in this sector reported their occupations before (12.1 %) and after (28.9 %) moving (see Chapter 3). Nevertheless, regardless of occupation, women’s incomes changed less than those of men and this may be explained by the fact that women had lower educational levels (mean = 10.09) than did men (mean =10.95) (see Appendix 7.1). The effect of occupation on work status change was absent because, as explained earlier in this chapter, work status change is a function of occupation. Hence, the statistical analysis cannot be run. 7. Model 7.1. 7. Employment Status in Hanoi affected Economic Improvement (income change). More specifically, paid employment affected income change positively (B = 556.75. p < .001) while unemployment, unpaid employment, and an “other” status affected income change negatively (B = -509.16, -438.59, and —689.54, p < .001, respectively). This means that migrants who worked for pay in Hanoi achieved greater income change than migrants who were unemployed, working without pay or had an 250 "other” employment status in Hanoi, who experienced a decrease in earnings. This result is not surprising given the advantages in earnings of those who worked for pay in Hanoi compared to those who had an other employment status or were unemployed or working without pay. who had no access to earnings. Despite the non-significant relation between employment status and work status change, however. in combination with gender’s effect. there was a tendency for men to achieve greater income change than women, and this is explained by men’s higher educational levels which afforded them greater opportunities for well-paid jobs compared to women, as discussed in Model 7.1.1. The relation between Employment Status in Hanoi and Work Status Change was absent because, as explained previously, work status change is also a function of employment status. 8. Model 7.1.8. Social Connections did not affect Economic Improvement (B = .403 for work status change and 23.33 for income change, p > .1). Despite this non- significant relation, regardless of social connections, in combination with gender’s effect, both women and men tended to work for pay in the state sector. and men’s income improved more than that of women. Nevertheless, Hypothesis 7.1.f, which states that people with social connections achieve greater economic improvement than those without social connections, was rejected. This insignificant result might be rooted in the high prevalence (88.8 %) of social connections among migrants in the sample before moving. The small number of those who had no social connections may not truly reflect the importance of social networks in the adaptation process (lack of representativeness). Perhaps. the migrants’ great economic improvement was a result of the existence of social networks. But. we do not know if their improvement was greater than that of those without social connections in Hanoi. Furthermore, Gender affected income change positively (B = 120.22. p < .001). i.e.. regardless of social connections, men gained greater income change than did women. and this is explained by the positive relation between education and earnings as explained in Model 7.1.1. Implicitly, we might infer that human capital tends to be more important than social capital in achieving income improvement in Hanoi. 9. .Ilodel 7. I .9. Type of Relations did not affect Economic Improvement (B = .457 for work status change and 14.1 1 for income change, p > .1). Despite this insignificant relation. when seen in combination with gender’s effect, both women and men, regardless of type of relations in Hanoi. tended to work for pay in the state sector, although women’s income changed less with moving than that of men. Nevertheless, the result did not support Hypothesis 7.1.g which states that people with kin relations achieve greater economic improvement by migrating than those with non-kin relations. This finding might be rooted in the fact that social networks did not affect economic improvement (as can be seen in Model 7.1.8) possibly because almost all migrants in the sample had social connections in Hanoi. Therefore, the type of relations migrants had at destination also had no effect on economic improvement. In sum, Economic Improvement was affected by six out of nine independent variables, that is, demographic attributes and human capital. Income, Social Connections, and Type of Relations (i. e.. financial and social capital) did not affect Economic Improvement. Hypotheses 7.1.a, 7.1.b, 7.1.c, and 7.1.d were supported while Hypotheses 7. I .e and 7. I .f were rejected 252 Results of Linear Regression of Economic Improvement by All Significant Independent Variables The data in Table 7.1.9 show that of the six significant independent variables, Age and Marital Status (B = -.009. p < 001 and -.749. p < .01. respectively) were good predictors of Work Status Change while those of Income Change were Gender (B = - 98.57. p < .001). Education (B = -12.20, p < .01), and Employment Status (B = -276.64, p < .01 for other employment status, 415.69, p < .001 for paid employment). That is, younger single women and men were more likely to work for pay in the state sector than Table 7.1.9: Coefficients of Linear Regression of Economic Improvement by All Independent Variables Variables Economic Improvement Work status change Income change Demographic attributes: Gender . 38.57"" Age -.009*** Marital Status -.749* Financial capital: Income Human capital: Education . -12.20* "‘ Occupation - Informal sector - Formal sector Employment status - Unemployment - Unpaid employment . . - Other employment . -276.64** - Paid employment . 415.69 *** Social capital: Type of Relations * : p < 0.05 ** : p < 0.01 *** : p < 0.001 . : insignificant relation Note: The Social Connections variable cannot be computed because there was no variation in the responses. were their older married migrants. Women with higher educational levels working for pay achieved more income change than those with lower educational levels and an “other” employment status. Nevertheless, women as a group achieved less income change than did their male counterparts. This means that demographic attributes and 253 human capital play an important role in determining migrants’ economic success in the adaptation process. The fact that demographic attributes and human capital were better predictors of economic improvement than was social capital might be explained by the nature of market mechanisms that emphasize the advantages of education and age. This preference in the labor market tended to affect the strategy of a household to maximize economic success by selecting the best matched laborers to meet the demands of the labor market (a youthful and highly educated labor force). Social capital tended to be less salient than demographic attributes and human capital because of the nature of the assistance members of migrants’ social networks are able to provide; they cannot change the demographic attributes and human capital of migrants, but they can help channel migrants into jobs with different wages in accord with the preference of the labor market. 2. Degree of Satisfaction Migrants’ perceptions of their satisfaction with migration ranged from very satisfied (1) to not satisfied (3). Although most migrants (95.6 %) felt satisfied with migration, a smaller percentage felt very satisfied (30.9 %) compared to those who were merely satisfied (64.7 %) with migration. Most cases (95.6 %) fell between very satisfied (1) and satisfied (2). ANOVA Results of Degree of Satisfaction by Six Nomipal Variables 7. 2.] . Gender As can be seen in Table 7.2.1, women were slightly more satisfied (mean = 1.69) with migration than were men (mean = 1.73). Gender was not associated with Degree of Satisfaction, however (F: 1.625, p > .1). 254 Table 7.2.1: ANOVA Results of Degree of Satisfaction by Gender Female (mean) Male (mean) All (mean) Degree of Satisfaction 1.69 1.73 1.71 Totals 415 374 789 F 2 1.625, p z .203 7. 2. 2. Marital Status The data in Table 7.2.2 show that both single women (mean = 1.62) and men (mean = 1.72) were more satisfied with migration than were their married counterparts (mean = 1.75). Although single women (mean = 1.62) were more satisfied with migration than were married women (mean = 1.75), single and married men were satisfied with migration at almost the same degree (mean = 1.72 compared to 1.75). Marital Status was associated with Degree of Satisfaction (F = 4.715, p < .05). Table 7.2.2: ANOVA Results of Degree of Satisfaction by Marital Status Marital Status Female (mean) Male (mean) All (mean) Never married 1.62 1.72 1.67 Ever married 1.75 1.75 1.75 Totals 1.69 1.73 1.71 415 379 794 F=4.715,p= .030 7. 2. 3. Occupation As can be seen in Table 7.2.3, both women and men were equally satisfied with mi gration. regardless of their occupation in Hanoi (mean = 1.74 and 1.76, respectively). Nevertheless, men working in the formal sector (mean = 1.72) were slightly more satisfied with migration than were those working in the informal sector (mean = 1.90); women, regardless of occupation, were similarly satisfied with migration (mean = 1.73 for formal sector and 1.76 for informal sector). Occupation was associated with Degree of Satisfaction (F = 5.176, p < .01). 255 Table 7.2.3: ANOVA Results of Degree of Satisfaction by Occupation Occupation Female (mean) Male (mean) All (mean) Agriculture 1.76 1.90 1.83 Informal sector Formal sector 1.73 1.72 1.73 Totals 1.74 1.76 1.75 201 Is) I\) b) 424 F: 5.176.p=.002 7. 2. 4. Employment Status From the data in Table 7.2.4, it can be seen that women, regardless of employment status in Hanoi, were slightly more satisfied with their move (mean = 1.69) than were men (mean = 1.73). Men who were unemployed (mean = 2.08) or working without pay (mean = 2.0) were the least satisfied among the migrants in the sample. Women who were unemployed and had an “other” employment status (i.e., were students. pensioners, or disabled) were the most satisfied with their move (mean = 1.57 and 1.56, respectively). Employment Status was associated with Degree of Satisfaction (F = 4.574. p < .01). Table 7.2.4: ANOVA Results of Degree of Satisfaction by Employment Status Employment Status Female (mean) Male (mean) A11 (mean) Unemployment 1.57 2.08 1.81 Unpaid employment 1.75 2.00 1.76 Other 1.56 1.68 1.62 Paid employment 1.77 1.76 1.77 Totals 1.69 1.73 1.71 415 374 789 F = 4.574. p = .003 7. 2. 5. Social Connections The data in Table 7.2.5 show that, regardless of social connections, women were slightly more satisfied with the outcomes of their migration than were men (mean = 1.69 for women and 1.73 for men). However, both women and men with social connections (mean = 1.69) were more satisfied with migration than those without social connections 256 (mean = 1.81). Social Connections was associated with Degree of Satisfaction (F = 4.227. p < .05). Table 7.2.5: ANOVA Results of Degree of Satisfaction by Social Connections Social Connections Female (mean) Male (mean) All (mean) Yes 1.68 1.71 1.69 No 1.76 1.87 1.81 Totals 1.69 1.73 1.71 374 379 753 F :4227. p = .040 7. 2. 6. Type ofRelations The data in Table 7.2.6 indicate that regardless of type of relations in Hanoi, women were almost as satisfied with migration as were men (mean = 1.68 for women and 1.71 for men). Women with kin relations were as satisfied (mean = 1.68) with the outcomes of migration as were those with non-kin relations (mean = 1.68), and this was the same for men (mean = 1.72 for men with kin and 1.70 with non-kin). Type of Relations was not associated with Degree of Satisfaction (F =.007, p > .1). Table 7.2.6: ANOVA Results of Degree of Satisfaction by Type of Relations Type of Relations Female (mean) Male (mean) All (mean) Kin relations 1.68 1.72 1.70 Non-kin relations 1.68 1.70 1.69 Totals 1.68 1.71 1.69 374 379 753 F : 007.. p = .934 In short. Degree ofSatisfaction was associated with four out of six nominal variables, i. t Marital Status, Occupation, Employment Status, and Social Connections. Correlation Results of Degree of Satisfaction by Three Ordinal Variables 7. 2. 7. Age As can be seen in Table 7.2.7, Age was not associated with Degree of Satisfaction (r = .045, p > .1). although it was in the case of women (r = .098. p < .05). This means 257 that both younger and older migrants had a similar degree of satisfaction, but younger women tended to be less satisfied with migration than older women. Table 7.2.7: Correlation Coefficients of Degree of Satisfaction by Age, Income, and Education Variable Female (r) Male (r) Total (r) Age 098* -.019 .045 Income -.228*** -.049 -.l 14* Education -. 125* -.127* -.I 14*” "‘ :p<0.05 ** :p<0.01 ***:p<0.001 7. 2. 8. Income The data in Table 7.2.7 indicate that Income was associated with Degree of Satisfaction negatively (r = -.1 14, p < .05). This means that people, especially women, with lower income levels were more satisfied with the outcomes of their migration than those with higher income levels. 7. 2. 9. Education It is clear from the data in Table 7.2.7 that Education was associated with Degree of Satisfaction, and that the association was negative (r = -.114, p < .001). That is, both women and men with lower educational levels were more satisfied with migration than those with higher educational levels. In sum, Degree of Satisfaction was associated with two out ofthree ordinal variables: Income and Education. Combining these results with those of the ANOVA tests. we learn that Degree of Satisfaction was associated with six variables out of nine: Marital Status, Occupation, Employment Status, Social Connections, Income, and Education. That is, it was associated with demographic attributes and human, financial, and social capital. 258 Results of Linear Regressions of Degree of Satisfaction by Each of Eight Independent Variables, Controlling for Gender, and Testing Hypotheses The results in Table 7.2.8 show that Degree of Satisfaction was affected by Marital Status. Income, Education. Occupation (informal sector), and Employment Status (other and paid employment). That is, Degree of Satisfaction was shaped by demographic attributes and human and financial capital. Noticeably, Gender had no effect on Degree of Satisfaction. The explanation for gender’s non-significant effect is offered in Model 7.2.1. This explanation is applicable for all those instances in which gender had no effect on the dependent variable. 1. Model 7. 2. 1 . Gender did not affect Degree of Satisfaction (B = .005, p > .1). Both women and men were similarly satisfied with the outcomes of their migration. Hypothesis 7.2.a, which states that men are more satisfied with migration than women, was thus rejected. This result may be explained by the fact that, even though Vietnamese female migrants experienced less income improvement (see Table 7.1.1) than did their male counterparts. they gained slightly greater occupational change than did men (mean = 1.21 for women and 1.13 for men) (see Appendix 7.5). More specifically, as can be seen in Appendix 7.6, a smaller percentage of women (35.7 %) worked in the formal sector before moving than did men (51.4 %), but in the city they were as likely to work in this sector as were men (77.8 % compared to 78.3 %). Movement into this sector thus provided women with the same stable employment and benefits enjoyed by men. Moreover. women made more money after migration than before moving (see Table 7.1.1). Because money is a resource that can be translated into power in the household. women might have gained greater voice in making decisions with their entry into the Table 7.2.8: Coefficients of Linear Regressions of Degree of Satisfaction by Each of Eight Other Independent Variables, Controlling for Gender Model Variable Degree of Satisfaction (B) 1 Gender .005 2 Gender .005 Age .000 3 Gender .006 Marital Status .008“ 4 Gender .002 Income -.000* 5 Gender .008 Education -.003*** 6 Gender Occupation - Agriculture . Gender .005 — Informal sector .126** Gender .005 - Formal sector .004 7 Gender .005 Employment Status - Unemployment .009 Gender .006 - Unpaid employment .008 Gender .006 - Other -.153*** Gender .004 - Paid employment .105M 8 Gender .005 Social Connections -.103 9 Gender .004 Type of relations -.000 :p'r::0.05 ** :p<0.01 ***:p<0.001 260 formal sector than they had previously had. Consequently. they were as satisfied with the outcomes of migration as were men. 2. Model 7. 2. 2. Age did not affect Degree of Satisfaction (B = .000, p > .1). Despite this insignificant relation. there was a tendency for both younger and older female and male migrants to be satisfied with their move to Hanoi. Perhaps, this is because almost two-fifths (38.6 %) of migrants in this study, both younger and older, moved for economic reasons; the remainder moved for family (24.8 %), educational (21.9 %) or other (14.7 %) reasons (see Chapter 3). And all migrants reported that their economic status improved with migration; they obtained better jobs with better pay by moving (mean = 1.02 for work status change and 300K for income change) (see Table 7.1.1) than they had at their origins. Implicitly, then, all migrants in this study succeeded economically by moving to Hanoi, even though some migrants gained more economically than others. Consequently, both younger and older people were satisfied with migration and their degree of satisfaction did not range greatly (between 1 and 2), as mentioned above. 3. Model 7. 2.3. Marital Status affected Degree of Satisfaction positively (B = .008, p < .01). Single people, regardless of gender, were more satisfied with migration than married ones. Hypothesis 7.2.b, which states that single people are more satisfied with migration than are their married counterparts, was supported. This finding is consistent with those in the migration literature that emphasize the advantages of singleness and power relations in the migration process (Williams et al. 1986, Lu 1998, Payner 1992, Lichter 1983, Dumon 1989, Neto 1995, Fawcett et al. 1984). On the one hand, single people, especially women. are the preferred labor force due to employers’ 261 perception about their labor commitment compared to their married counterparts. Single employees are also economically beneficial to employers, helping them to reduce the cost they spend providing social benefits for workers’ family members if they are married (e.g., housing). Therefore. single people tended to have more economic opportunities than their married counterparts and hence, they were more satisfied with migration. On the other hand, single people tend to have fewer responsibilities and obligations than do married ones as mothers or fathers or wives and husbands. They also experience fewer constraints than married ones who must consider their spouses’ opinions in making decisions. Further, single people in the sample tended to be involved in solitary movement more than their married counterparts (see Table 6.1.2). In the absence of the older generation, they had freedom to make decisions about their own lives while married people tended to move associationally and continued to be shackled with responsibilities. Therefore, the result that single people were more satisfied with migration than their married counterparts is understandable. 4. Model 7. 2. 4. Income affected Degree of Satisfaction negatively (B = -.000, p < .05). Both women and men who earned lower incomes in Hanoi were more satisfied with their move than those with higher incomes. While this result might seem surprising, it is important to note that those with low incomes tended to have lower educational levels than those with higher incomes (see Appendix 7.7). Given this fact, they might have had lower expectations than those with higher educational levels and any economic improvement they achieved would be likely to increase their satisfaction level. 5. Model 7. 2. 5. Education affected Degree of Satisfaction negatively (B = -.003, p < .001). People with lower educational levels were more satisfied with migration than 262 those with higher educational levels. This result provides support for Hypothesis 7.2.c which states that people with high educational levels are less satisfied with migration than those with low educational levels. This finding is consistent with that in the migration literature about the negative relation between education and degree of satisfaction due to migrant’s different expectations in accord with their educational levels. As explained in Model 7.2.4, migrants with higher educational levels are less satisfied with migration than those with lower educational levels (Tyner 1994, Wilson 1998. Zlotnik 1995a, 1995b). 6. Model 7. 2. 6. Occupation (informal sector) affected Degree of Satisfaction positively (B = .126. p < .05). Both women and men working in the informal sector were less satisfied with their move to Hanoi than were those working in the formal sector. Even though the result in Table 7.2.4 shows that migrants with lower educational levels tended to be more satisfied with their income improvement than those with higher educational levels, and those who worked in the informal sector had lower educational levels (mean = 9.35) than those who worked in the formal sector (mean = 10.82) (see Appendix 7.4), migrants who worked in the informal sector tended to be less satisfied with migration than those working in the formal sector. This result may reflect the disadvantages of the informal sector compared to the formal sector. First, the income change of migrants working in the informal sector was slightly less (mean = 326K) than that of those working in the formal sector (mean = 364K) (see Table 7.1.3). Second, the conditions of employment were more insecure in the informal sector than they were in the formal sector. Moreover, the informal sector offers fewer opportunities for social 263 mobility than does the formal sector. Consequently. those working in the informal sector were less satisfied with migration than those working in the formal sector. 7. Model 7. 2. 7. Employment Status (paid employment) affected Degree of Satisfaction positively (B = .105. p < .01) while the “other” employment status affected it negatively (B = -.153. p < .001). Both women and men working for pay were less satisfied with migration than those in school, on pensions. or disabled (“other” employment status). This result might be explained by the great proportion of migrants in the other employment status category (65.0 %)I who were in school in the city, i.e., they moved to the city to satisfy their educational goal: 39.1 percent of migrants in this category moved for this reason compared to 29.9 percent who moved for family and 15.9 percent who moved for other reasons (see Appendix 7.8). Further, this finding might be rooted in the high expectation of those who worked for pay; because of their high educational levels (i.e., 10.64 years of schooling) (see Appendix 7.8) they migrated probably expecting to receive more pay than they actually did. Hence, they were less satisfied with migration than might be supposed (as can be seen in Models 7.2.4 and 7.2.5). 8. Model 7. 2. 8. Social Connections did not affect Degree of Satisfaction (B = - .072. p < .05). Despite this insignificant relation, there was a tendency that both women and men, regardless of social connections in Hanoi, to be satisfied with migration. Nevertheless, Hypothesis 7.2.d, which states that people with social connections are more satisfied with migration than those without social connections, was rejected. ’ People who were in school in the city made up more than three-fifths (65.0 %), ofthose reporting an “other” employment status (33.7 %). People who were invalids or retired comprised only 35.0 percent of the people in this category (calculated from the MOLISA data set by the author). 264 This insignificant result might be rooted in the great number of migrants having social connections in the city before moving. 9. Model 7. 2. 9. Type of Relations did not affect Degree of Satisfaction (B = - .000. p > .1 ). Despite this insignificant relation, both women and men, regardless of type of relations in Ilanoi. tend to be similarly satisfied with migration. This result, undoubtedly. reflects the importance of the assistance of social networks in the adaptation process. Regardless of the sources of assistance, whether it came from kin or non-kin relations. help at destination was likely to improve people’s chances of success and thus. their degree of satisfaction with their move. In short. Degree of Satisfaction was affected by marital status and human, financial. and social capital, but not by gender and age. Hypotheses 7. 2.b and 7. 2.c were supported while Hypotheses 7. 2.a and 7. 2.d were rejected. Results of Linear Regression of Degree of Satisfaction by All Significant Independent Variables As can be seen in Table 7.2.9, among five significant variables, Education was the best predictor of Degree of Satisfaction (B = -.004, p < .05). People with lower educational levels were more satisfied with migration than were those with higher educational levels. This result might be rooted in the different expectations of people, based on their human capital levels, about their economic improvement. Those with higher educational levels may have anticipated achieving greater success than they realized and hence. were not as satisfied with migration as those who had lower expectations based on their lower educational levels. Human capital played an important role in migrants ’ degree of satisfaction at destination. Table 7.2.9: Coefficients of Linear Regression of Degree of Satisfaction by All Independent Variables Variables Degree of Satisfaction (B) Demographic attributes: Gender Age Marital Status Financial capital: Income Human capital: Education -.004* Occupation - Agriculture - Informal sector - Formal sector Social capital: Type of Relations * : p < 0.05 ** : p < 0.01 ***: p < 0.001 . : insignificant relation Note: The Employment Status and Social Connections variables cannot be computed because there was no variation in responses. SUMMARY It is clear that the migrants in the sample experienced positive changes by moving to Hanoi. Most gained economically in terms of both work status and income. Subjectively, these positive changes were evidenced by the high proportion of migrants who were satisfied (65%), and very satisfied (31%) with their move. Objectively, they were demonstrated by the migrants’ economic improvement which was affected by demographic attributes (Gender, Age, Marital Status) and human capital (Education, Occupation. and Employment Status), but not financial (Income) or social capital (Social Connections and Type of Relations). Among these variables, Gender, Age, Marital Status. Education. and Employment Status were good predictors of Economic Improvement. These results are both similar to and different from findings in the migration literature. Similar to the situation in other countries, the economic improvement of migrants was affected by demographic attributes and human capital. That is, single and younger women and men with higher educational levels working for pay in the state sector tended to earn more money than their older married counterparts with lower educational levels working in the informal sector. This similarity might be rooted in the prevalence ofthe positive relation between human capital and economic success in accord with migrants’ single status and youth (based on the preference of the labor market) (Massey et a1. 1993. Sanders et al. 1996, Hunt et a1. 1985, Robinson 1993). Vietnamese migration differs from that of other countries by the fact that no relationship was found between economic improvement and social connections and type ofrelations as reported elsewhere (Wilson 1994, Becker 1975, Khoo et al. 1984, Douglas et al. 1992). Regardless of social connections and type of relations (social capital), Vietnamese migrants tended to achieve economic improvement by moving. This was especially true for women. This result might be explained by the prevalence of pre- existing social connections among the migrants in this study. This result also may be a product of the changing conditions associated with Vietnam’s transition period into a market economy, where structural changes (privatization of ownership and production, the growth of business and trade, and the elimination of the household registration system) have affected household strategies rooted in ideas about maximization of economic opportunities and people’s possibilities for employment and social mobility. The changing economy tended to affect population movement by providing more diverse options for people in different areas to find jobs matched with their human capital in the labor market. Thus, regardless of their social capital at destination, people had more economic opportunities than they did prior to the transition. Moreover, people might have been pulled to migrate, regardless of social capital. because of the great income difference between rural and urban areas. This study also found that the income change (i.e., economic status) of migrants in the city was associated with occupation and employment status at destination. Obviously. economic success was positively associated with occupation and employment status: being employed or working for pay in the state sector definitely helped improve incomes and stabilize the earnings of migrants. With regard to Degree of Satisfaction, Vietnamese migrants also shared some similarities with and differences from migrants in other countries (Williams et al. 1986, Lu 1998, Payner 1992, Dumon 1989, Neto 1995, Tyner 1994, Wilson 1998, Zlotnik I995a. 1995b, Marshall 1984. Speare et al. 1982, Poot 1993). Like that of migrants in other countries. Vietnamese migrants ' satisfaction with migration was aflected by marital status. income. and education. Single migrants with lower income and educational levels tended to be more satisfied with migration than married ones with higher income and educational levels. This result again confirms the importance of human capital in the adaptation process. and the implications of marriage. especially for women. It was found in this study that single Vietnamese migrants are more likely to obtain prestigious occupations, to work for pay, and to earn high incomes with migration than (1) they had had before moving and (2) their married counterparts. This can be understood as the advantage of singleness in the urban labor market. in which employers prefer single over married workers due to their perception about single and married people’s labor commitment. 268 Education was a good predictor of satisfaction with migration, but surprisingly, it affected degree of satisfaction negatively. This finding can be explained by the fact that people with higher educational levels might have had higher expectations about the outcomes of migration than those with lower educational levels. Moreover, migrants with hi gh educational levels might have been overeducated for the jobs they found in Hanoi due to their migrant status in the city. Therefore. they were less satisfied with migration than those with lower educational levels. Nevertheless, different from findings in the migration literature, both women and men, regardless of age, social connections, and type of relations in the city, were similarly satisfied with migration. This result suggests that the socialist system has an advantage over the capitalist one in terms of gender equality in the migration process. Under socialism. women were empowered by educational and equal employment opportunities. They achieved high educational levels (but slightly lower than that of men), participated in paid employment, and, as the results showed, achieved greater occupational improvement than did men. Consequently. they were as satisfied with the outcomes of migration as were men. In addition, migrants with higher income levels were less satisfied with their move to Hanoi than those with lower income levels. This finding seems to contradict the common understanding of the positive relation between income level and degree of satisfaction. Nevertheless, if we consider the problems migrants face in cities (residency change. limited social networks compared to those of urban dwellers, ethnic and cultural differences). we can see that this finding is understandable. In addition to potential problems in locating employment, their change in residence presented them with logistic 269 problems ofdaily living. Further. although most (88.8 %) had social connections in the city. they left many useful contacts in the rural areas. They thus had fewer social networks. compared to urban dwellers. available to help them cope with difficulties. Alternatively. migrants with higher income levels might experience more difficulty in adapting because they had higher expectations about the outcomes of their migration than did those with lower income levels. This study also found that people working in the informal sector were less satisfied with migration than were those working in the formal sector. This result clearly reveals the advantage of the formal sector over the informal sector in terms of economic security and social mobility. In addition, those who worked for pay in the city achieved more economic improvement (better jobs and higher pay) than others. So, the fact that they were more satisfied with migration is understandable. In short, the adaptation outcomes of Vietnamese migration were significantly affected by four sets of factors: demographic attributes and financial, human. and social capital. 270 CHAPTER 8: SUMMARY AND CONCLUSIONS The goals of this chapter are threefold. First, I summarize the results of the study about urbanward migration of women and men in Vietnam in the transition period from a centrally-planned economy to a market economy. This summary focuses on the similarities among and differences between Vietnamese urbanward migration and that of other developing countries in the decision-making, moving, and resettling processes. Following this summary, I discuss the implications of this study. Finally, I consider the weakness of the study data and suggest further research to enrich our understanding of gendered migration in developing countries. 1. Findings This study examined patterns of gendered urbanward migration in Vietnam. According to the proposed model of gendered migration, the study focused on the determinants of three sub-processes of migration: decision-making. moving, and adapting. The results from linear and logistic regressions tend to support the suggested model and hypotheses related to the impact of demographic attributes and human, financial, and social capital on reasons for and decision-makers of migration, type and distance of move, and economic improvement and degree of satisfaction with migration. 1.1. Reasons for migration The findings of this study are similar to those in the migration literature via their support of my hypotheses about gender and marital status differences in reasons for migration. The results appear to support the explanation of the human capital perspective that (a) people move for economic reasons because of the differences in employment opportunities between regions and that (b) married men, single men, and single women 271 move for economic reasons while married women move for family reasons. This study also contributes to the existing literature on labor migration by providing some new insights into the decision-making process. The results show that reasons for migration are also affected by age, education. occupation, employment status, social connections. and type of relations. i.e., the intersection of demographic attributes and human and social capital. Those who moved for economic reasons were young. were unemployed or worked for pay in the agricultural sector prior to migration. and had high educational levels and no social connections in Hanoi; among those with social connections, these tended to be non-kin. Those who moved for educational reasons were young, single, and unemployed. unpaid workers. or students prior to migration. Those who moved for family reasons were older, married women who worked in the non-agricultural sector, worked without pay. or were disabled or retired before migration, and had urban social connections who tended to be kin. 1.2. Decision-makers of migration The study’s results show that gender, age, marital status, and education (human capital) were implicated in the decision-making process. Similar to findings in the migration literature, young, single women and men with low educational levels were directed to move by family members while older, married men with low incomes but high educational levels were self-determined movers. Different from previous findings, older married women were likely to join with their husbands in making the decision to move rather than being deployed by others. The study also contributes to the migration literature by providing findings about the effects of income. occupation. employment status. social connections, and type of relations on the decision-makers of migration. People with kin relations who were unemployed, worked without pay or were disabled, pensioners or students (i.e., “other” employment status) before migration were directed to move by family members. In contrast. those who had non-kin relations in the city and worked in the agricultural sector before migration were self-determining movers. Single women with non-kin relations in the city who had an “other” employment status before moving were likely to be directed to move by family members. Thus. this study supports the explanation of the human capital perspective which posits that people participate in the decision-making process differently because they have different human capital levels. In other words, people with higher human capital levels will be more independent in making decisions than their less endowed counterparts. In this instance, women had lower educational levels than men and thus had less voice in the decision-making process. The results also support the explanation of the social capital perspective that people with non-kin relations in the city have a greater voice in making decisions than those with kin relations there because of the positive relation between education, non-kin relations, economic opportunities. and self- determination. The results of the study do not support the contention that female migration is primarily determined by family members because of women’s subordinate position. Rather. Vietnamese married women join with their husbands in making the decision to migrate, i.e., they have more voice in decision making about migration than their counterparts in other developing countries. And this may reflect the ideological commitment to gender equality under the socialist system. to \1 Lu 1.3. Type of migration This study supports past findings about the relationships between gender, marital status. and type of relations (social capital) and the way people move, i.e., either alone or with others. As in previous studies, the results showed that single women and men with kin relations in the city moved alone while married women with non-kin relations in Hanoi moved with others. In contrast to other studies (UN Secretariat 1993, Piampiti 1984. Arnold et a1. 1984), however. this study found that the existence of social connections does not affect type of move. Regardless of social connections, women moved associationally while men moved alone. This insignificant relation might be explained by the great proportion of migrants in the sample (88.8 %) having social connections in Hanoi before moving. Finally, this study contributes to the migration literature by providing some insights into the relations between type of migration and income (financial capital) and education and occupation (human capital). People with low income levels but high educational levels who worked in the formal sector prior to migration moved alone while those with high income levels but low educational levels who worked in the informal sector before migration moved with others. 1.4. Distance of migration This study’s results support my hypotheses about income (financial capital) and educational and occupational (human capital) differences in distance of migration. As found in other studies, people with high income and educational levels moved a farther distance than their less endowed counterparts. As never reported before in the migration literature (Wilson 1998, Repak 1994, Straugh 1984, UN Secretariat 1993), this study focused on other measures of human capital such as employment status and occupation in 274 addition to education. This study found that people who worked in the formal sector prior to migration moved a farther distance than those who worked in other sectors. In addition. in Vietnam, distance of migration was not affected by gender and the existence of social connections at destination; regardless of social capital, both women and men moved the same distance. Thus. the results do not support the presumption that men move a farther distance than women because of their dominant role and ideologically- based superiority over women. This result might be explained by the relatively short distance migrants in this study moved and by the active participation of Vietnamese women in migration as a household survival strategy. Furthermore, this study contributes to the migration literature by providing information about the relations between distance of move and age and occupation (human capital). Older people who worked in the formal sector prior to migration moved a farther distance than their younger counterparts, especially those who worked in the agricultural sector. As shown above (in both sections 1.4 and 1.5), this study supports the explanation offered by the human capital perspective that people participate in the moving process differently because they have different human capital levels. In other words, people with higher human capital levels are more independent in moving and move a farther distance than their counterparts with lower human capital levels. They move a farther distance because they are more capable of considering alternatives and meeting the demands of the labor market than those with lower human capital levels. Moreover, they move farther in a search to make optimal use of their human capital. The results also support the explanation of the social capital perspective that people with kin relations at destination move alone because they are assured of assistance and believe this assistance will help ensure success in moving. 1.5. Economic improvement The findings also support my hypotheses about the effect of gender, age, and marital status (demographic attributes), and education (human capital) on outcomes of migration. Similar to the findings of previous studies, this study found that young, single women and men with higher educational levels achieved greater economic improvement by migrating than did their older married counterparts. The study thus supports the explanation of the human capital perspective that people with higher human capital levels achieve greater economic success than those with lower human capital levels because they have greater opportunities to obtain employment and well-paid jobs. Nevertheless, it was found that, in some ways, Vietnamese migration differs from that of other countries. The results showed no relation between economic improvement and the presence of social connections and type of relations (social capital) at destination. This result may be a function of the great percentage of the sample having social connections in the city before moving, thereby blurring the differentiating effect of the presence of kin and non-kin on outcomes of migration. This study also provides some insights into the relations between economic improvement and occupation and employment status at destination. Financial capital (income) after migration had no effect on economic improvement, and this is understood as a function of the close relation between income and occupational and employment statuses which might have overshadowed the actual importance of income. In fact, in Vietnam. occupational and employment statuses (associated with employment in the 276 formal/state sector) provide migrants more social status than income itself because of the social welfare benefits and upward mobility such jobs provide. 1.6. Degree of satisfaction with migration This study supports my hypotheses about the effects of marital status (demographic attributes). education (human capital) and social connections (social capital) on degree of satisfaction. Like the findings in the migration literature, the study shows that single people with lower educational levels and social connections in the city were more satisfied with migration than their married counterparts. In contrast to findings in the migration literature (Lu 1998, William et al. 1986), the results of this study show that gender does not affect degree of satisfaction with migration; both women and men were equally satisfied with the outcomes of migration. This result might be rooted in the fact that both female and male migrants had high educational levels and mainly worked in the state (formal) sector, in which. to some extent, gender discrimination was less than in the informal sector. Furthermore. this study contributes to the migration literature by providing some insights into the relations between degree of satisfaction and income, occupation, and employment status. Specifically, people with lower income levels working for pay in the formal sector were more satisfied with migration than their counterparts with higher income levels. This result may reflect the negative relation between the income levels and expectations of migrants or the fact that so many of the migrants (more than 70.0 %) worked in the formal sector after moving; the advantages of employment there, compared to the disadvantages associated with work in the informal sector, are striking in Vietnam. While this study did not support the explanation of the human capital perspective that people with higher human capital levels are more satisfied with migration than are those with lower levels. it did support the proposition that people’s satisfaction with migration is closely associated with their expectations, which are determined by their human capital levels. This study found that women were as satisfied as men with migration. even though they achieves less economic improvement than did men. This result may be explained by the fact that women’s satisfaction is closely related to (a) their subordinate gender roles in the household, which lower their expectations in comparison to men. and (b) the different reasons women and men have for migration (family versus economic). The results support the explanation of the social capital perspective that the existence of social networks increases satisfaction with migration because of members’ assistance in achieving economic betterment. 1.7. Discussion The similarities in patterns of migration in Vietnam with those of other countries indicate that, across borders, labor migration shares commonalities. Female and male migration is a selective process. and degree of participation in making the decision to migrate varies based on age, marital status, and human capital levels; reasons for migration. type of move, and outcomes of migration vary in accord with human, social, and financial capital levels and demographic attributes. Nevertheless, female migration. in general. is not homogeneous with that of men; women and men with different demographic attributes and levels of capital are involved in migration differently. These similarities and differences can be seen clearly through the different models in Figure 2 for women and men in the migration process. The findings of this study illustrated in this figure are summarized as follows: In addition to age and marital status, human and social capital, not financial capital. play an important role in determining reasons to move for both women and men (see Models 1.1, 1.2. and 1.3). With regard to decision-makers of migration, women’s voice in making decisions tends to be affected by age, marital status, and social and human capital. while that of men is affected by all three forms of capital (social, human. and financial). regardless of demographic attributes (see Models 2.1, 2.2. and 2.3). With respect to type of migration. even though both women’s and men’s type of migration is affected by marital status and all three types of capital, capital affects women’s and men’s moving patterns differently (see Models 3.1 and 3.2). As for distance of migration and degree of satisfaction with migration, both women and men are affected by age or marital status, and financial and human capital, but not by social capital (see Models 4 and 5). With respect to economic improvement, both women and men share the same change (see Model 6.1); their work status change is affected by age and marital status, but not by any type of capital. As for income change, even though women’s and men’s income change is affected by age, marital status. and human capital, these effects are different for women and men (see Models 6.2.1 and 6.2.2). Figure 2: Empirical Models of Gender Patterns of Urbanward Migration in 1. Reasons for Migration ......... HC Older ....................... married SC , HC Single Young HC ................................ Vietnam Model 1.1 for women - formal sector - unpaid and other empl family reasons - with social networks - kin relations Model 1.2 for men - agricultural occup. - paid empl. economic reasons - without social networks - non-kin relations Model 1.3 for women and men - high education educational reasons —_——_> - unempl., un-paid empl, and other empl. Note: HC means Human capital; FC means Financial Capital; SC means Social Capital. 280 2. Decision-makers of Migration older married younger_ single .. ." HC SC HC SC FC HC SC Model 2.1 for women - unempl., unpaid, and paid empl. \ . joint decision maker - non-kin relations Model 2.2 for women - low education - unempl.. unpaid and other empl. family members - kin i'elations Model 2. 3 for men - low 1ncome - high education - agricultural occup. self-determined - unempl., unpaid, and paid empl. - non-kin relations 281 3. Type of Migration Model 3.1 for women FC - high income married eff-7 .......................... HC - low education associationally - informal sector occup. SC - non-kin relations Model 3.2 for men FC - low income single ----------------------- HC - high education solitarily - formal sector occup. SC - kin relations 4. Distance of Migration Model 4 for women and men .. u .0’ FC - high income single farther distance HC - high education " - formal sector occup. Is) 00 l\) 5. Degree of Satisfaction with Migration Model 5 for women and men F C - low income ,n' .v" ." single more satisfied HC - low education - informal sector occup. 6. Economic Improvement 6.1. Work Status Change Model 6.1 for women and men young single $ greater gain 6.2. Income Change Model 6.2.1 for women young HC - high education single ...................................... - formal sector occup. __> little gain - paid empl. Model 6.2.2 for men young ...................................... HC - formal sector occup. —__> greater gain single - paid empl. 283 The similarities found in gendered migration patterns might be explained by the common needs of people who want to improve their economic status via geographic movement and by the prevalence of a household strategy of survival in the migration process. These similarities also might be rooted in common phenomena in developing countries: a high labor surplus, limited land, and different economic opportunities between urban and rural areas. Additionally. the similarities may reflect a worldwide resistance to change of gender ideology, and the effect of the globalization of the world economy. Given these characteristics, in general, human and social capital theories of migration can be useful to explain migration in developing countries. But, taken alone, human or social capital theories cannot fully explain migration as a complex gendered process. They need to be integrated to explain this process at both the macro and micro levels. Specifically, gender ideology, as a socially-constructed system of ideas about women’s and men’s roles and positions, plays an important part in explaining social relations at the household and societal levels in a patriarchal society such as Vietnam. Household strategies at the micro level are also critical to understanding the who, how, and why of geographic movement within a country. The inclusion of such variables in the theoretically-integrated model used in this study have facilitated the exploration of gendered migration in a society undergoing transition to a market economy. Migration patterns in Vietnam are also different from those of other developing countries. That is, the selectivity of migrants is not affected by marital status; type of move is not affected by migrants’ social capital; distance of move and economic improvement are not affected by gender and social capital; satisfaction with migration is not affected by gender. These differences might be a result of the contradiction between the specific political conditions of Vietnam that provide women opportunities to be equal to men and the obduracy to change of gender ideology. Or they may result from the political economy of Vietnam. which is changing from a centrally-commanded system into a market economy, and which is marked by a backward economy, but a good social welfare system. Alternatively, these differences might result from the data set used by this study. The results of this study might have provided support for more of the assumptions of extant migration theories than they did if the migrants in the study had come from distances farther from Hanoi. or if they had not clustered mainly in the formal sector after moving. or if they had not almost universally had social networks before moving. Given these characteristics, some assumptions of theories of human and social capital are not applicable to explain gendered migration in Vietnam. These failures might suggest further studies to examine migration patterns of other countries with similar situations in order to enrich theorization about migration. In summary, the results presented above show that, while the regression models can well explain the gendered migration process under the influence of demographic attributes and human, social, and financial capital, not all indicators included are statistically significant. The findings of this study tend to be consistent with the findings in the migration literature but, on the other hand, some findings are different. This indicates that, in general, theories of human capital and social capital of migration can be applicable in the gendered migration process of developing countries. Nevertheless, these theories alone cannot explain migration phenomenon fully in a restructuring economy such as Vietnam. The use of gender ideology as well as the use of the notion of a household strategy of survival to explain migration sub-processes at the macro and 285 micro levels are especially meaningful in understanding the prevalent pattern of urbanward migration in developing countries such as Vietnam. 2. Implications 2.1. Theoretical implications: The notion of human capital theories of migration that migrants are selective because of the positive relations between human capital, age, and marital status and the possibility of a move cannot be hypothesized universally, as Massey and his colleagues (1993) have discussed. This is because this selection depends not only on human capital and demographic attributes but also on the socio-economic and historical context specific to the destination involved. The results of this study confirm this argument by showing that in Vietnam, given high levels of poverty and the existence of a socialist system which emphasized equal opportunities for women and men, marital status did not affect the selectivity of migrants as usually happened in other countries. Both single and married Vietnamese women moved. This finding is important because it contests a long- accepted truism and challenges scholars to interrogate the nexus of context and geographic movement. The most important contribution of this study, however, is its integration of human and social capital perspectives to explain migration as a gendered process. The human capital perspective primarily focuses on economic aspects of migration, emphasizing the importance of human capital levels and ignoring gender differences, neglecting the non-economic, and underestimating the importance of household strategies in migration (Dang et al. 1997). The social capital perspective focuses on social relations and assistance. highlighting the importance of the personal networks that shape migrants’ 286 choices and opportunities and that help promote successful outcomes of migration (Wilson 1994). The theoretically integrated model used in this study helps explain migration as a complex and gendered process that emerges within a broad socio- economic. cultural. and historical context. Specifically, this theoretically integrated model helped successfully explore gender differences and the diversity of different groups of migrants in terms of their human and social capital levels, and their demographic attributes, an explication which cannot be accomplished by one or another perspective singly. Financial capital tended to be less salient in this model, and this might be a result of the overlapped effects or close relationships among variables such as occupation, employment status, and income. This non-salient effect also might be rooted in the sample size or the inadequate measurement of income by the MOLISA survey. While this study challenges the explanatory efficacy of using theoretical models in isolation, it nonetheless confirms some assumptions of human and social capital theories. This study’s findings support assumptions about the positive relationship between (1) human and social capital (type of relations) and economic opportunities and outcomes of migration, and (2) human capital levels and degree of participation in making the decision to migrate. By integrating them, this study was able to explain migration as a gendered process in a developing country such as Vietnam. That is, this theoretically integrated model helps explain who moves, why they move, how they move, what they obtain from migration or, put another way, what is different about female and male migrants and why their patterns of movement are different. From Model 1.1 through 6.2 in Figure 2, we can see that the theoretically proposed framework (Figure 1a in Chapter 2) has changed its components’ order and 287 location. As the results showed, financial capital does not have a strong effect on these gender patterns of migration. Among these three types of capital, human capital tends to have the clearest effects on the dependent variables (i.e., more significant effects). However. as noted elsewhere. the insignificant effects of social capital may not truly reflect the importance of social capital due to the great number of migrants in the study who had social connections before moving. By considering the findings of this study, a new theoretical framework for the study of gender patterns of urbanward migration is proposed in Figure 3. In this framework. human capital plays a central role in determining gendered patterns of migration. This might be explained by the context within which migration occurs. i.e.. the beginning of the transition period to the market economy. Within this context, human capital, as a favorable attribute in the labor market, is acknowledged and increasingly becomes an important resource to maximize economic opportunities and to ensure migration success. This acknowledgement tends to reflect the change that the transition to the market economy has brought, i.e., an image of workers as gendered beings in a market economy. More specifically, workers are evaluated and compensated based on their human capital levels, not primarily on their revolutionary contributions as was the case in the command economy. However, men, not women, dominate in well- paid or high prestigious jobs because of their higher human capital levels compared to that of women. Women workers are more likely than men to cluster in low-paid jobs because of their lower human capital levels and the cost employers must spend for welfare benefits related to their reproductive roles and childcare. This means that even though the data of this study were collected at the beginning of the 288 < £805 558?: 13:30 328:5 3E8 5:52 8:98 _Eoom HXOHCOU U_EOCOUoiomoom magnets“ oEQSonuQ Esp—BS E 55¢..sz FEB—Sf: ..o 2:823 .5?va .5,. zuctefiaum 13:950.; >62 mm e....wE 289 transition period. the results reflect people’s actions to meet the demands of market mechanisms and to maximize their economic improvement via physical and occupational mobility. Another contribution of this integrated model is that it furthers exploration of migration as a process that is propelled by both macro and micro forces. While the human capital perspective helps explain patterns and outcomes of migration as a product of individual attempts to maximize migration returns (i.e., micro level), the social capital perspective spurs examination of the structured social relationships that shape migrants’ roles. choices. and opportunities (i.e., macro level). By integrating these insights, the study fosters a model of understanding of migration at both the micro and macro levels. Nevertheless. the macro level of society involves more than social and economic phenomena. It also has to do with the political. The gendered migration process in Vietnam has been strongly affected by the political system in which geographic movement occurred, and in which gender relations were improved. The socialist system as a political regime helped improve women’s status, as the results of this study suggest. Thus, this study contributes another important insight to the migration literature: gender relations can be improved with the intervention of political institutions that provide women rights and opportunities equal to men, thereby helping them to empower themselves. Nevertheless, even though Vietnamese women have been provided rights and opportunities equal to men, and have gained high educational levels and good occupations with equal pay under socialism, their status remains lower than that of men. This situation might be seen as a result of the obduracy of gender ideology in a patriarchal society such as Vietnam. 290 Le Thi Quy (1998) argues that the status of Vietnamese women has declined with the introduction of market mechanisms. The results of this study, however, did not provide evidence in support of this argument. The female migrants in the sample earned more and had a more stable economic status after migration than they had before moving. As the results show. they mainly worked for pay in the formal (state) sector in Hanoi, thereby holding economic positions that were more secure and well-paying than those they had in rural areas. Working for pay also gave women access to cash, a resource that has the potential to give them a greater voice in the power hierarchy of the household than they have had in the past. It is obvious then that the development process was advantageous, not disadvantageous, for Vietnamese female migrants. This assertion, which calls into question Quy’s thesis, might reflect the fact that the effect of market mechanisms on female migrants’ status was not clear at the time this study’s data were collected; the empirical information used in this study was collected at the very beginning of the transition period and, therefore, its consequences had only a short period of time in which to tally effects. Further, the migrants in this study may have been highly selected. Thus, if data had been collected after the process was long underway or if the sample had been different. its results might have confirmed the decline of women’s status with marketization. 2.2. Substantive implications: The transition from a centrally planned economy to a market economy and the removal of the household registration system has opened up a voluntary population movement in search of employment and earnings in different regions and areas within Vietnam. This trend tends to create and intensify urban problems such as population 291 concentration in cities. underemployment of the urban population, crime, and degradation of urban infrastructure (Doan Minh Thang 1997). Thus. study of this population mobility. uncovering the cause of this influx and the characteristics of this labor force, can help policy makers to predict the potential volume of labor movement and the quality of labor mi grants. thereby. facilitating plans for the management or redistribution of the labor force between regions. Study of this population mobility also can help policy makers understand the vulnerability of migrants at destination, promoting them to implement strategies that might improve their situation there. Voluntary population movement, on the one hand, tends to balance regional differences in labor supply and demand and in living standards. As Agostinelli (1991) has noted. such movement contributes low-cost labor to the urban economy and remittances for the rural economy. Thus, it has a positive effect on development in general. On the other hand, this movement tends to deepen urban problems, intensifying population concentration in cities where the government has been trying to deconcentrate overpopulated areas. Thus, systematic study of voluntary migration can encourage the design of plans to eliminate its negative consequences and to increase its positive effects. Moreover. systematic study of this movement might also help promote the creation of projects that maximize the utility of this labor force so that it furthers household, regional, and national development. Incorporating gender into the study of voluntary migration highlights the similarities of and difference between women and men who move. Understanding of such phenomena may encourage policy-makers to develop policies that compensate the female labor force more equitably than currently is the case, 292 i.e.. to create jobs that compensate highly educated women appropriately, thereby improving their status relative to men. This study found that most migrants move for economic reasons and that they are highly educated and economically well-off. In this sense, then, current voluntary migration in Vietnam may result in a loss of a valuable labor force in rural areas. Such movement can lead to a deepening of regional differences in development rather than a balancing of them. Given this situation, development planners may need to devise schemes to promote economic opportunities in rural areas that encourage people to stay. This study offers a theoretical framework of gender patterns of migration, in which human and social capital play more important roles than financial capital in the migration process. This finding might suggest that policy makers should diversify the system of employment contracts and the occupational structure available to people with different human capital levels in order to mobilize and utilize the workforce at place of origin effectively. This finding also suggests a need to restructure educational and training programs that focus on developing a skilled workforce for future labor markets in rural areas. As the country moves into the economic transition, which is based on the notion of individualism and privatization, social capital becomes an important resource to channel individuals into different paths of social and physical mobility. Policy makers should be aware of the influence of social networks on these types of mobility in order to attract people to stay in their place of origin instead of moving out. While the findings of this study did not demonstrate directly the effect of voluntary migration on Hanoi, they do suggest that the massive entry of migrants into cities may lead to or may be a product of an urban bias in government planning (Dang et 293 al. 1997). The government must acknowledge the potential hazards of such skewed planning and create a comprehensive development program that integrates rural and urban development. Ways to accomplish this include shifting industrial production from the inner city to suburban areas or to nearby provinces and establishing manufacturing opportunities in rural areas. Regardless of how this is done, the government needs to devise plans to redistribute economic opportunities between rural and urban areas. 3. Limitations and future research 3.1. Limitations ofthis study As with other studies, this study has limitations that primarily stem from the data used for this study. They were secondary and, thus, this study encountered problems similar to those associated with other studies using such data. First, this study could ask only certain questions and could focus only on certain issues based on the information available. Specifically, the data this study used had no information about stayers, compelling it to use the national population in general as a point of comparison. This comparison may not reflect true differences between stayers and movers, and it does not reveal why some people stay but others move. Future research could avoid this problem by exploring both stayers and movers. Second, there are some methodological problems related with the MOLISA data set. Some questions were based on recall or misconceptions in the definition and operationalization of concepts, thereby leading to incorrect information and to the ultimate results of this study. For example, information about income levels was asked for individuals rather than the household. It is most usual in Vietnam, however, to count total household income in rural areas because its members cooperate in farming and the 294 products of their labor cannot be counted separately, i.e.. it is not clear who earns more or less. Hence this problem might lead to incorrect income levels, and the weak or non- existent associations between financial capital and the migration process. Moreover, due to monetary inflation during the period the data were collected, respondents’ calculations oftheir income prior to migration may have been incorrect. Information about the occupations of migrants also was ambiguous because respondents only reported the industries in which they worked (e.g., agriculture. forestry, garment industry), not their actual occupations. In addition to these problems with the occupational and income variables. the data set had a great number of missing cases, thus. probably undermining ,1 the representativeness of the sample. Future research could avoid these problems by investing time and effort in the conceptual definition and operationalization of variables and developing data more rigorously. Third, the analysis focused on only certain indicators of demographic attributes and human, social. and financial capital. There might be other important factors that help explain migration as a gendered process and that help illuminate the effect of human and social capital on female or male migrants in the migration process. For example, to clarify the effect of human capital, researchers may need to include variables such as work skills. To examine the effect of family structure on household strategies of survival, they might want to include household type (e.g., complex versus simple) and size as demographic indicators. To explore the effect of migration on places of origin or on the adaptation process. researchers may need to include social capital indicators such as connections to origin. Further, to explore in-depth the effect of gender on the migration process. researchers may want to include variables such as relations between women and 295 men. time devoted to household labor and childcare, and the characteristics of those who are consulted on important matters. Similarly, to explore the role of financial capital in the migration process. researchers also need to include not only an income variable but also other variables such as wealth or property people owned. In conclusion, despite the limitations mentioned above, this study can contribute to the existing literature on the gendered migration process in relation to human, financial. and social capital. Specifically, the findings test the validity of the hypotheses generated from human and social capital theories of migration in a particular socio- economic and political context and provide insights for fiirther study. 3.2. Future research There are many other possibilities for future research on the gendered migration process in relation to capital (human. social, financial) in Vietnam. First, as reported, since the 19805. Vietnam has been experiencing a transition to a market economy; the migrants in the sample tended to be involved in urbanward movement during this time (the average time in Hanoi of the migrants of this study was three to five years) and they mainly worked in the formal sector after moving. The results of this study might be replicated and extended, focusing on migrants working in the informal sector (which has expanded lately due to the introduction of market mechanisms), in which gender differences tend to be deepened (Le Thi 1998). and on self-employed migrants whose situations tend to be more vulnerable than those of people working in the formal sector. Second. in-depth qualitative studies would enrich the interpretations of this study’s quantitative analyses. Specifically, the positive relations between human capital levels and influence in making decisions and the outcomes of migration may suggest the 296 benefits of future research that examines the degree of accordance between the actual roles and contributions of female migrants at home and at work and their position in the male-female power hierarchy. In addition, this kind of research might help explore how gender roles and work roles are reconciled and the process via which development advantages or disadvantages women. Third. the strong effect of the political system (socialist) on the improvement of women’s status in Vietnam suggests that a comparative study of gendered migration in Vietnam and in capitalist countries would be useful. By examining the decision-making, moving. and adaptation processes across political systems, such a study might reveal how the emphasis on or commitment to gender equality is implicated in changing women’s status. Moreover, the close relations between human capital and the outcomes of migration suggest the benefits of future comparative research of urbanward and ruralward migration that investigates differences in directions or patterns of move, controlling for genden Fourth, this study supports the assumption about the importance of social capital and the positive relationship between social capital and economic improvement. Social capital provides material and immaterial supports to its members that help to reduce the risk of moving. especially for those with less favorable market attributes. This means that social capital helps lessen the selectivity of migrants and maximize economic success at destination. However. social capital channel gendered migrants into gender segregated occupations. This finding suggests the benefits of a longitudinal study that investigates the establishment and development of social networks in well-defined origins or destinations that perpetuate migration as a gendered process and reduce its selectivity. 297 APPENDICES 298 Appendix 3.1: Means of Occupation by Education Before Moving APPENDICES Occupation Mean Standard Deviation N Agriculture 8.2 2.57 133 Informal sector 10.2 3.06 25 Formal sector 1 1.8 2.20 125 Other . . Total 9.95 3.01 285 F : 71.697, p = .000 Appendix 3.2: Means of Occupation by Income Before Moving Occupation Mean Standard Deviation N Agriculture 105K 92K 130 Informal sector 356K 202K 26 Formal sector 308K 166K 125 Other . . . Total 219K I77K 281 Appendix 5.1: Frequency Distribution of Marital Status by Reasons for Migration of those working in the informal sector Marital Economic (%) Educational (%) Family (%) Other (%) Total (%) status Single 11.4 11.4 2.9 8.6 34.3 Married 5.7 8.6 45.7 5.7 65.7 Total 17.1 20.0 48.6 14.3 100.0 Appendix 5.2: Mean of Age by Reasons for Migration Mean Standard. Deviation N Economic 26.7 8.0 6 Educational 30.3 12.3 7 Family 50.5 19.5 17 Other 39.4 25.6 6 Total 40.7 20.0 35 Appendix 5.3: Means of Type of Relation by Education Before Moving Type of Relation Mean Standard Deviation N Kin 10.03 3.55 500 Non-kin l 1.2 2.70 202 Total 10.36 3.36 702 F = 16.288. p = .000 299 Appendix 6.1: Means of Income Before Moving by Type of Relations Type of relations Mean Standard Deviation N Kin 212K 176K 155 Non-kin 249K 183K 90 Total 225K 179K 245 F:2.473.p—— 117. Appendix 6.2: Mean of Distance of Move. Mean = 148 km N = 789 Std. Deviation = 259 km Appendix 6.3: Frequency Distribution of Education by Age Age elementary (%) secondary (%) high school (%) higher education (%) total (%) 13-20 54.7 28.8 15.3 19.8 25.8 21-25 5.3 18.9 48.6 27. 25.3 2635 4.2 24.1 25.2 35.7 18.7 36+ 35.8 28.3 10.8 17.4 30.2 Total 100.0 100.0 100.0 100. 0 100.0 Appendix 6.4: Mean of Employment Status by Education Before Moving Past Employment Status Mean Standard Deviation N Unemployment 10.40 3.14 113 Unpaid employment 8.82 3.36 34 Other 10.48 3.71 349 Paid employment 9.98 3.00 295 Total 10.21 3.38 791 F = 3.287. p : .020 Appendix 6.5: Correlation ofIncome by Education Before Moving Past Income Past Education Person Correlation .383 Sig. (2-tailed) .000 N 289 Appendix 7.1: Mean of Current Education and Current Income by Gender Gender Current Education Current Income Female Mean 10.09 404K N 416 210 Standard Deviation 2.52 217K Male Mean 10.95 548K N 379 227 Standard Deviation 2.23 309K Appendix 7.2: Frequency Distribution of Employment Status Before and After Moving by Marital Status for Men Unemployment Unpaid Other Paid Total employment employment (%) (%) (%) (%) (%) before after before after before after before after Single 16.9 3.3 2.3 .5 60.1 50.7 20.7 45.5 100.0 Married 13.3 3.0 .0 1.2 22.9 17.5 63.9 78.3 100.0 Total 14.0 1.3 43.8 39.6 100.0 3.2 8 36.1 59.9 100.0 Appendix 7.3: Mean of Income Change by Marital Status Group for Men Marital status group Mean N Std. Deviation never married 415K 101 370K ever married 320K 142 385K Total 359K 243 381K Appendix 7.4: Mean of Current Occupation by Education in Hanoi Current Occupation Mean N Standard Deviation Informal sector 9.35 86 2.40 Formal sector 10.82 310 2.26 Total 10.47 396 2.37 Appendix 7.5: Means of Occupational Change by Gender Female (mean) Male (mean) All (mean) Occupational Change 1.21 1.13 1.17 N 233 229 462 F = .244. p = .621 01 Appendix 7.6: Frequency Distribution of Occupation Status Before and After Moving by Sex Agriculture Informal sector Formal sector Total (%) (%) (%) before after before after before after (%) Female 56.6 7.7 22.2 35.7 77.8 100.0 Male 38.2 10.4 21.7 51.4 78.3 100.0 Total 47.4 9.1 43.6 100.0 21.9 78.1 100.0 Appendix 7.7: Correlation ofCurrent Income by Current Education Current 1ncome Current Education Person Correlation .205 Sig. (2-tailed) .000 N 436 Appendix 7.8: Frequency of Employment Status Before Moving by Reasons for Migration Economic (%) Educational (%) Family (%) Other (%) Total (%) Unemployment 22.3 7.4 9.0 1 1.0 14.1 Unpaid employment 1.9 4.0 8.5 4.2 4.4 Other 17.4 79.5 53.8 48.3 44.6 Paid employment 58.4 9.1 28.6 36.4 37.0 Total 100.0 100.0 100.0 100.0 100.0 Appendix 7.9: Mean of Current Employment Status by Past Education Employment status Mean Standard Deviation N Unemployment 11.58 2.63 26 Unpaid employment 8.56 3.15 66 Other 9.78 4.07 262 Paid employment 10.64 2.84 437 Total 10.21 3.38 791 BIBLIOGRAPHY 303 BIBLIOGRAPHY Adamchak. Donald. 1987. “Further evidence on economic and non-economic reasons for turnaround migration.” Rural Sociology, 52(1):]08-18. Agostinelli. Granni. 1991. “The impact of developmental pattern and pace on migration.” Pp. 1 3-43 in “Migration - Development interrelationships: The case of Philippine.” New York: Center for Migration Studies. Ariffin. Jamilah. 1984. “Migration of women workers in Peninsular Malaysia: impact and implications.” Pp 213-226 in F awcett, James, Siew-Ean Khoo, and Peter Smith (eds). Women in the cities of Asia. Colorado: Westview. Arker. .Ioan. 1989. Doing comparable worth: Gender, class, and pay equality. Philadelphia: Temple University Press. Arnold. Fred and Suwanlee Piampiti. 1984. “Female migration in Thailand.” Pp.143-64 in Fawcett, James. Siew-Ean Khoo, and Peter Smith (eds), Women in the cities of Asia. Colorado: Westview. Arnold. Fred. 1987. “Birds of passage no more: migration decision-making among Filipino immigrants in Hawaii.” International Migration, 25(1):41-61. Aslanbeigui. Nahid. Steven Pressman and Gale Summerfield (eds.) 1996. Women in the age of economic transformation: Gender impact of reforms in post-socialist and development countries. New York: Routledge. Bach. Van Bay. 1996. (Institute for Economic Research of Ho Chi Minh City 1996). Migration. human resources, employment and urbanization in Ho Chi Minh City. Ilanoi: National Political Publishing House. Banister. Judith. 1993. Vietnam population dynamics and prospects. Berkeley: Institute of East Asian Studies, University of California at Berkeley. Barry. Kathleen (ed.). 1996. Vietnam ’s women in transition. London: Macmillan Press. Becker. Gary. 1962. “Investment in Human capital: A theoretical analysis.” Political Economy. 7029-49. Becker, Gary. 1975. Human capital: A theoretical and empirical analysis, with special reference to education. New York: Columbia University Press. Berk. Sarah F enstermaker. 1995. The gender factory: The appointment ofwork in American households. New YorkIPlenum. Bilsborrow. Richard and Hania Zlotnik. 1992. “Preliminary report of the United Nations expert group meeting on the feminization of internal migration.” International .«lligration Review, 26(1 )2138-1 72. Bilsborrow, Richard E. and United Nations Secretariat. 1993. “Internal female migration and development: An overview.” Pp: 1-17 in Internal migration of women in developing countries. New York: United Nations. Blau. Peter and Otis Duncan. 1967. The American occupational structure. New York: .Iohn Wiley and Sons. Bonney. Norman and John Love. 1991. “Gender and migration: geographical mobility and the wife’s sacrifice.” Sociological Review. 39(1-2): 335-348. Boyd. Monica. 1989. “Family and personal networks in international migration: recent development and new agendas.” International Migration Review, 23(3):638-670. Brown. Alan & Egon Neuberger (eds.), 1977. Internal migration: A comparative perspective. New York: Academic Press. Brueckner, Jan. 1990. “Analyzing Thirst World urbanization: A model with empirical evidence.” Economic Development and Cultural Change, 38(3):591-612. Bryant. John. 1998. “Communism, poverty and demographic change in North Vietnam.” Population and Development Review, 24 (2):235-269. Brydon. Lynne. 1992. “Ghanainan women in the migration process.” Pp.91-108 in Chant (ed). Gender and Migration in developing countries. London and New York: Balhaven Press. Bui. Thi Kim Quy. 1996. “The Vietnamese women in Vietnam’s process of change.” pp. 159-66 in Barry, Kathleen (ed.) Vietnam ’s women in transition. London: Macmillan Press. Bui. Viet Bao. 1994. “Di dan tu do vao Ha Noi tim viec lam.” [Voluntary migration into Ha Noi to search for employment] Labor and Society, 318-] 1. Center for Population and Labor Studies (CPLS). 1996. Danh gia di dan noi dia 0 Vietnam (bao cao) [Evaluation on internal migration in Vietnam (report) ]. Ha Noi. ' Chant. Sylvia (ed.), 1992a. Gender and migration in developing countries. London and New York: Belhaven Press. Chant. Sylvia. 1992b. “Migration at margins: gender. poverty and population movement on the Costa Rican periphery.” Pp.49-72 in Chant (ed.), Gender and migration in developing countries. London and NewYork: Belhaven. Chapin. Frances. I989. Tides of migration. A study of migration decision-making and social progress in Sao Miguel, Azores. New York: AMS Press. Chapman. Murrey and Mansell Prohero. 1985. “ Themes on circulation in the Third World.” Pp. 1-26 in Murrey Chapman and R Mansell Prohero (eds.), Circulation in the Third World. Routledge and Kegan Paul. Chattopadhyay. Arpita. 1997. “Family migration and the economic status of women in Malaysia”. International Migration Review, 31(2):338-352. C hattopadhyay. Arpita. 1998. “Gender. migration, and career trajectories in Malaysia.” Demography. 35(3):335-44. C ohen. Robin (ed.). 1996. Theories of migration. Brookfield: Elgar Reference Collection. Cole. William and Richard Sanders. 1985. “ Internal migration and urban employment in the Third World.” American Economic Review, 75(1-3): 481-498. C oleman, James S. 1988. “Social Capital in the Creation of Human Capital.” American Journal of Sociology, 94, Supplement: 895-8120. Collins, .Iane. 1983. “Fertility determinants in a high Andes community.” Population and Development Review, 9(1):61-75. C onnell. R. W. 1989. Gender and power. Stanford: Stanford University Press. Cooke. Thomas J. and Adrian Bailey. 1996. “Family migration and the employment of married women and men.” Economic Geography, 72:38-48. Dang. Anh. Sidney Goldstein, and James McNally. 1997. “Internal migration and development in Vietnam.” International Migration Review, 31 :313-337. Dang. Nguyen Anh. 1991. “Vai tro va vi tri cua phu nu nong thon trong gia dinh: Mot nghien cuu cong dong.” [Roles and positions of rural women in the household: a community study.] Pp. 157-164 in Institute of Sociology at Ha noi and Department of Sociology at Gothenburg, Sweden (eds), Nhung nghien cuu xa hoi hoc ve gia dinh Viet nam [Studies of the sociology of Vietnamese families]. Ha Noi: Social Science Publishing House. Dang, Nguyen Anh. 1997. “Ve vai tro cua di cu nong thon-do thi trong su nghiep phat trien nong thon hien nay.” [Roles of rural-urban migration in current rural development] Sociological Review, 60(4): 1 5-1 9. Dang. Nguyen Anh. 1998a. “Cac mo hinh di cu va phat trien kinh te o Viet nam.” [Migration and development patterns in Vietnam] in Institute of Sociology (eds.), Bao cao Hoi thao: Di dan va Suc khoe tai Viet nam [Report on migration and health in Viet nam]. Ha Noi. Dang. Nguyen Anh. 1998b. “Di cu va phat trien trong boi canh doi moi kinh te-xa hoi cua dat nuoc.” [Migration and development in socio-economic reform of the country] Sociological Review. 61 (1 ):3-12. Dang. Nguyen Anh. 1998c. “Vai tro cua mang luoi xa hoi trong qua trinh di cu.” [Roles of social networks in migration process] Pp. 48-57 in Do. Van Hoa et al. (eds.), ('hinh sach di dan o C hau A [Migration policies in Asia]. Ha Noi: Agriculture Press. Dang. Tho Xuong. 1991.”Dien bien. thuc trang va dinh huong hop tac hoa nong nghiep o nuoc ta.” [Situation. condition. and orientation of cooperatives in agriculture in our country] Pp.252-287 in The State Agriculture Department. Kinh te_Xa hoi nong than Vietnam ngay nay [Economy-society in rural areas in contemporary I ’ietnam. ] Hanoi: Political-cultural Publishing House. De .Iong, Gorden; Rex Warland. and Brenda Davis Root. 1998. “Family interaction and migration decision making.” Pp.155-168 in Harry Schwarzweller and Brenda Mullan (eds.), Research in rural sociology and development. Focus on migration. Volum 7. Stamford: Jai Press Inc. De .Iong. Gordon, Andrea Johnson. and Kerry Richter. 1996. “Determinants of migration values and expectations in rural Thailand.” Asian and Pacific Migration, 5(4):399-346. De .Iong. Gordon, Ricardo Abad, Fred Arnold, Benjamin Cartino, James F awcett, and Robert Gardner. 1983. “International and internal migration decision making: A value-expectancy based analytical framework of intentions to move from a rural Philippine province.” International Migration Review, 17(3): 470-484. Delong. Gordon and Robert Gardner. (eds), 1981. Migration decision making. New York: Pergamon. Desbarats. Jacquiline. 1987. “Population redistribution in the Socialist Republic of Vietnam.” Population and Development Review, 13(1):43-76. Diij. .Iouk Van and Hendrik F olmer. 1986. “The consequences of interregional labor migration for the regional labor market: Theory. methodology, and Dutch experience.” The Review of Economics and Statistics, 57(1):28-32. Division of Journalism and Propaganda (DJP). Dept of Political Economy. 2000. “So huu va cac thanh phan kinh te trong thoi ky qua do len chu nghia xa hoi 0 Vietnam.” [“Ownership and economic sectors in the transition to socialism in Vietnam,”] pp] 18-31. Chapter 6 in Kinh te chinh tri hoc Mac-Lenin: giao trinh kinh te chinh tri hoc dai cuong [Marxism- Leninism political economy: General political economy texbook.] Hanoi: National Political Publishing House. Do. Thai Dong. 1991. “Gia dinh truyen thong va nhung bien thai o Nam bo Viet nam.” [Traditional family and its features in southern part of Vietnam] Pp.71-84 in Institute of Sociology at Ha noi and Department of Sociology at Gothenburg, Sweden (eds). Nhung nghien cuu xa hoi hoc ve gia dinh Viet nam [Studies of the sociologv of Vietnamesefiimilies ]. Ha Noi: Social Science Publishing House. Do. Thi Binh et al. 1996. Phu nu ngheo trong dieu kien kinh te thi truong [Poor women in the condition ofmarket mechanisms. ] Hanoi: National Political Publishing House. Do. Thi Binh (ed.) 1997. Nhung van de chinh sach xa hoi voi phu nu nong thon trong giai doan hien nay [The issues of social policiesfor rural women in today period]. Ilanoi: Social Science Publishing House. Doan. Kim Thang. I997. “Phan tich tong quan ve mot so khia canh kinh te xa hoi va hieu qua cua di dian nong thon-do thi trong thoi ky doi moi o nuoc ta” [General analysis of some socio-economic aspects and consequences of rural-urban migration in the renovation period]. Sociological Review, Doan. Mau Diep and Trinh Khac Tham. 1996. “Survey on spontaneous migration to a rural and an urban area in Viet nam.” Asian and Pacific Studies, Series No.142:1- 57. Doan. Mau Diep, Patrick Gubry, Jerry Huguet, and Trinh Khac Tham. 1995. Di dan tu do den Da nang va Vung tau [Spontaneous migration in Dong Nai province and Vung Tau city]. Ha Noi: National Political Publishing House. Donato. Katharine. 1993. “Current trends and patterns of female migration: evidence from Mexico.” International Migration Review 27(4):748-77l. Douglas. Gurak and Fe Caces. 1992. “Migration networks and the shaping of migration systems.” Pp. 150-176 in Mary Kritz, Lin Lean Lim, and Hania Zlotnik (eds), International migration systems. Orford: Oxford University Press. Drachman. Diane. 1996. “Migration and resettlement experiences of Dominican and Korean families.” Families in society, 77(10):625-38. Du Toit, Brian. 1990. “People on the move. Rural-urban migration with special reference to the Third World: theoretical and empirical perspectives.” Human Organization, 49(4):305-3 1 7. 308 Dumon. F. 1989. “Family and migration.” International Migration, 27(2):251-70. EIU 1986-97. (‘ountrv Profile. Vietnam 1986-1997. London: Economic Intelligence Unit. Erman. Tahire. 1998. “The impact of migration on Turkish rural women.” Gender and society. 12(2): 146-67. Eviota. Elizabeth. and Peter Smith. 1984. “The migration of women in the Philippines.” Pp165-190 in Fawcett. James, Siew-Ean Khoo, and Peter Smith (eds). Women in the cities of Asia. Colorado: Westview. 1" arber. Stephen. 1983. “Post-migration earnings profiles: An application of human capital and job search models.” Southern Economic Journal, 49(3):693-705. 1" awcett. James. Siew-Ean Khoo, and Peter Smith. 1984. “Urbanization, migration and the status of women.” Pp3-15 in Fawcett, James. Siew-Ean Khoo, and Peter Smith (eds), Women in the cities of Asia. Colorado: Westview. Fawcett. James. 1989. “Networks. linkages. and migration systems.” International Migration Review 23(3): 671-680. 1" I’orde. Adam and Stefan Vylder. 1996. From plan to market: The transition in Vietnam. Boulder: Westview Press. Findley, Sally and Assitan Diallo. 1993. “Social appearances and economic realities of female migration in rural Mali.” Pp. 244-259 in UN (ed.), Internal migration of women in developing countries. New York: United Nations. Fisher. Elizabeth. 1980. Women creation: Sexual evolution and the shaping ofsociety. New York: McGraw-Hill. Foner. Nancy; Ruben Rumbaut; and Steven Gold (eds.). 2000. Immigration researchfor a new century. New York: Russell Sage Foundation. F rcnier. Mariam and Kimberly Mancini. 1995. “Vietnamese women in Confucian environment: Contribution to the study of initial decline of East Asian women’s status.” Pp.279-3l2 in Family and women’s status in society. A Vietnamese and American view. Hanoi: Social Science Publishing House. F uess. Scott and Bun Song Lee. 1994. “Government reforms, economic restructuring and the employment of women: South Korea, 1982-92.” Pp.145-159 in Aslanbeigui, Nahid. Steven Pressman, and Gale Summerfield (eds), Women in the age of economic transformation. Gender impact of reforms in post-socialist and developing countries, London and New York: Routledge. 309 Gallin. Rita. 1973. Migrant networks: Taipei, Taiwan. Thesis for the degree of MA. Michigan State University. General Statistical Office (GSO). 1978. Nien giam thong ke l 9 77. [Statistical Yearbook l 97 7.] Hanoi: Statistical Publishing House. General Statistical Office (GSO). 1987. Nien giant thong ke l 986 [Statistical Yearbook l 986]. Hanoi: Statistical Publishing House. General Statistical Office (GSO). 1996. Nien giam thong ke I 995 [Statistical Yearbook 1995]. Hanoi: Statistical Publishing House. General Statistical Office (GSO). 1997. Nien giam thong ke I 996 [Statistical Yearbook l 996]. Hanoi: Statistical Publishing House. Glenn. Evelyn N. 1999. “The social construction of gender and race.” pp. 3-43 in Myra Marx. Ferree, Judith Lorber and Beth B. Hess. (eds.) Revisioning gender. Sage Publications Gold. Steven J. 1992. Refugee community: A comparativefield study. Newbury Park: Sage (’ioldscheider. Calvin. 1977. “Migration and socio-economic modernization.” Chapter 7 in Goldscheider, Calvin. Population, modernization and social structure. Brown: Little. Greeley. Andrew. 1997. “Coleman revisited: Religious structures as a source of social capital.” American Behavioral Scientist, 40 (5):587- Greendwood. Michael J. 1985. “Human migration: theory, models and empirical studies.” Journal of Regional Science. 25(4):521-544. Guest, Philip. 1993. “The determinants of female migration from a multilevel perspective.” Pp.223-243 in United Nations. Expert meeting on the feminization of internal migration. Internal migration of women in developing countries. New York: United Nations. Gupta. Manash Ranjan. 1993. “Rural-urban migration , informal sector and development.” Journal of Development Economics, 41 :137-151. Ha Noi Statistical Office (HSO). 1996. Ha Noi nien giam thong ke I 997 [l 997 statistical yearbook]. Ha Noi: Ha Noi Statistical Office. Habitat. United Nations Center for Human Settlements. 1996. An urbanizing world. Gbobal report on human settlements. Oxford: Oxford University Press. 310 Ilarbison, Sarah. 1981. “Family structure and family strategy in migration decision- making.” Pp.225-52 in De Jong et al. (eds.) Migration decision making. New York: Pergamon. IIardwick. Julie. 1998. The practice of patriarchy: Gender and politics of household authority in early modern France. University Park: Pennsylvania State University Press. Ilarris. John and Michael Todaro. 1970. “Migration. unemployment and development: A two-sector analysis.” American Economic Review, 60: 126-142. Ilerbert. Eugena. 1993. Iron, gender. and power. Bloomington and Indianapolis: Indiana University Press. Herzog. Henry and Alan Schlottmann. 1983. “Migration information, job search and the remigration decision.” Southern Economic Journal, 50(1):43-57. Hitchcox, Linda. 1994. “Relocation in Vietnam and outmigration: The ideological and economic context.” Pp.202-220 in Judith Brown and Rosemary Food (eds), Migration: The Asian experience. Oxford: St. Martin Press. Hoang Thi Lich. 1993. “The development of household economies and market systems in improving the gender and poverty situation in Vietnam.” p. 3 1 8-3 34 in Noeleen Heyzer and Gita Sen. Gender economic growth and poverty: Market growth and state planning in Asia and the Pacific. Kuala Lumpur: Asian and Pacific Development Center. Hoffman. Costance and Neal Ritchey. 1992. “Assessing the differences in distance of interstate migration, 1980.” Sociological Focus, 25(3):241-55. Holmes-Eber, Paula. 1997. “Migration, urbanization and women’s kin networks in Tunis.” Journal of Comparative Family Studies, 28:54-72. Hong, Sawon. 1984. “Urban migrant women in the Republic of Korea.” Pp. 191-212 in F awcett, James, Siew-Ean Khoo, and Peter Smith (eds), Women in the cities of Asia. Colorado: Westview. Hou. F eng & Roderic Beaujot. 1994. “The differentiation of determinants among return, onward and primary migrants in Canada.” Canadian Studies of Population, 21(1): 1-19. Huang. Nora. 1983. “Female migration to Taipei - process and adaptation.” Population Geography, 5(1/2):12-33. Hugo. Graeme. 1981. “Village-community ties, village norms, and ethnic and social networks: A review of evidence from the Third World.” Pp. 186-226 in De Jong. 311 Gordon and Robert Gardner (eds). Migration decision making. NY: Pergamon Press. Ilugo. Graeme. 1993. “ Migrant women in developing countries.” Pp.47-76 in United Nations. Expert meeting on the feminization of internal migration. Internal migration of women in developing countries. New York: United Nations. Ilugo. Graeme. 1995. “International labor migration and the family: Some observations from Indonesia.” Asian and Pacific Migration, 4(2-3):273-301. llunt. Janet & James Kau. 1985. “Migration and wage growth: A human capital approach.” Southern Economic Journal, 51(3):697-71 1. Iluq-I Iussain. Shahnaz. 1995. “Fighting poverty: The economic adjustment of female migration in Dhaka.” Environment and Urbanization, 7(2):51-65. Institute of Economics Research at Ho Chi Minh City (IER). 1996. Di dan, do thi hoa, nguon nhan luc va viec lam o thanh pho Ho Chi Minh [ Migration, urbanization, human resource and employment in Ho Chi Minh City]. Ha Noi: National Political Publishing House. IS-MI 1897. Institute of Sociology et al. 1998. Bao cao Hoi thao: Di dan va Suc khoe tai Viet nam [Report on migration and health in Viet nam]. Ha Noi. Jelin. Elizabeth. 1997. “Migration and labor force participation of Latin American women: the domestic servants in the cities.” Journal Signs, 3(1): 129-41. Johnson. Allan. 1995. The Blackwell dictionary of Sociology. Cambridge: Blackwell Khoo. Siew-Ean and Peter Pirie. 1984. “Female rural to urban migration in Peninsular Malaysia.” Pp 125-142 in Fawcett, James. Siew-Ean Khoo, and Peter Smith (eds). Women in the cities of Asia. Colorado: Westview. Khoo. Siew-Ean, Peter Smith, and James Fawcett. 1994. “Female rural to urban migration in the Third World: Comparative perspectives and research notes. Migration of women to cities: The Asian situation in comparative perspective.” International Migration Review, 18(4): 1247-1263. Kobayashi, Hideo. 1995. “Features of industrial development and migration in Asia.” Asian and Pacific Migration, 8(1): 14-19. Kofman. Eleonore. 1999. “Female birds of passage’s decade later: Gender and immigration in the European Union.” International Migration Review, 33(2):269- 99. b) to Kontuly. Thomas, Ken Smith. and Tim Heaton. 1995. “Culture as a determinant of reasons for migration.” The Social Science Journal, 32(2): 1 79-93. Landry. David and Yi Zhao. 1988. “The selectivity and impact of migration on Ohio’s population.” Paper presented in the North Central Sociological Association. Lauby. Jennifer and Oded Stark. 1988. “Individual migration as a family strategy: young women in the Philippines.” Population Studies, 42(3):473-486. Laurian. Lucie. Richard Bilsborrow. and Laura Murphy. 1998. “Migration decision among settler families in the Ecuadorian Amazon: the second generation.” Pp. 169-197 in Harry Schwarzweller and Brenda Mullan (eds), Research in rural sociologv and development. Focus on migration. Volum 7. Stamford: JAI Press Inc. Le. Dang Giang. 1996. “C ung vuot cau ve lao dong qua viec tu do di chuyen lao dong tu nong thon vao do thi.” [Labor supply over labor demand in a labor voluntary movement from rural to urban areas] Information of labor market, 1:10-17. Le. Ngoc Hung. 1996. A study of work attitudes in a restructuring economy: job satisfaction and organizational commitment in Vietnam. Dissertation submitted to University of North Carolina at Chapter Hill 1996. Le. N goc Van. 1997. “Phan cong lao dong theo gioi trong gia dinh nong dan.” [Gender division of labor in farmer’s household], Science of women, 3:19-26. Le. Thi Nham Tuyet, Le Van Phung, and La Nham Thin. 1995. Gioi va phat trien o Viet nam [Gender and development in Viet nam]. Ha Noi: Social Science Publishing House. Le. Thi Nham Tuyet. 1975. Phu nu Viet nam qua cac thoi dai [Vietnam’s women in different regimes. ] Hanoi: Social Science Publishing House. Le. Thi Nham Tuyet. 1998. “Dac thu gioi o Viet nam va ban sac dan toc nhin tu goc do nhan hoc xa hoi.” [Gender in Vietnam and traditional national identity from perspective of Social Humanity] Paper presented in lntemational Conference of Vietnamese Studies 14-17 July, 1998 in Hanoi. Le. Thi Quy. 1992. “Mot so nhan xet ve bao luc trong gia dinh.” [Some opinions about domestic violence in families] Social Sciences, 4:81-87. Le. Thi Quy. 1998. “ Phu nu Viet nam trong co che thi truong.” [Vietnamese women in the market economy] Paper presented in lntemational Conference of Vietnamese Studies 14-17 July. 1998 in Hanoi. b) b.) Le. Thi. 1995. “Buoc tien bo cua phu nu Viet nam. muoi nam qua va nhung van de dat ra” [A progress of Vietnamese women in the past ten years and new issues], Science of women, 3(21 ):9-12. Le. Thi. 1998. “May suy nghi ve lao dong nu nhap cu tu do 0 Ha noi.” [Some thinking of voluntary female migrants in Ha noi.] Economic Studies, 241:38-47. Lee. Everett S. 1966. “A theory of migration.” Demography, 3: 47-57. Lewis. G. J. 1982. “Definitions and concepts.” Chapter 2 in Lewis, G J. Human migration. London: Croom Helm. Lichter. Daniel. 1983. “Socioeconomic returns to migration among married women.” Social Forces, 62(2):487-503. Light. Ivan and Steven J Gold. 2000. Ethnic economies. San Diego: Academic Press. Lim. Lin Lean. 1993. “The structural determinants of female migration.” Pp. 207-222 in Internal migration of women in developing countries. New York: United Nations. Lingam. Lakshmi. 1998. “Migrant women, work participation and urban experience.” Indian Journal of Social Network, 59(3):807-23. Lloyd. Peter. 1979. “The migration process.” p. 1 1 1-140 in Lloyd, Peter. Slums of hope. Penguin Books. Lorber, Judith. 1994. Paradoxes of gender. New Heaven: Yale University Press. Lu. M. 1998. “Analyzing migration decision-making: relations between residential satisfaction, mobility intentions. and moving behavior.” Environment and Planning A, 30(8):1473-95. Luu. Dai Thuyet & Nguyen Chi Dung. 1995. “Su bien doi co cau xa hoi-giai cap 0 nuoc ta hien nay” [The change in social-class structure in our country.] Pp.5-21 in Do, Nguyen Phuong (ed.), Situation and development trend Social structure in our country in this period. Hanoi: Agricultural Publishing House. Luu. Minh Tri. 1998. “Van de di dan tu phat ve Ha Noi. Thuc trang, giai phap va dinh huong quan 1y.” [Voluntary migration into Hanoi. Situation, solution and orientation for management]. Paper presented in Conference on Policies for voluntary migration, July 6-7, 1998, Ho Chi Minh City. Luu, Phuong Thao. 1996. Do thi hoa va su bien doi vai tro phu nu trong nong thon ngoai thanh thanh pho Ho Chi Minh [Urbnization and the changing roles of women in rural suburban areas of Ho Chi Minh City]. Ho Chi Minh City: VNRP. Mackerras, C olin, Robert Cribb. and Allan Healy (eds), 1988. Contemporary Vietnam. Perspectivefrom Australia. North Wollingong: University of Wollingong Press. Mai. Thi Tu & Le Thi Nham Tuyet. 1978. Phu nu Vietnam [Women in Vietnam. ] Hanoi: Foreign Languages Publishing House. lvlarshall. Dawn. 1984. “Vincentian contract labour migration to Barbados: the satisfaction of mutual needs?” Social and Economic Studies. 33:63-92. Martin. .Iack and Daniel Lichter. 1983. “Geographic mobility and satisfaction with life and work.” Social Science Quarterly, 64(3):524-35. 1\/Iassey. Douglas, Joaquin Arango, Ali Kouaouci, Adela Pellegrino, and Edward Taylor. 1993. “Theories of international migration: A review and appraisal.” Population and Development Review, 19(3): 431-466. Massey. Douglas. Joaquin Arango, Ali Kouaouci, Adela Pellegrino, and Edward Taylor. 1994. "An evaluation of international migration theory: the North American case." Population and Development Review, 20(4):699-751. Massey. Douglas. 1988. "Economic development and international migration in comparative perspective." Population and Development Review, 14(3):383-413. McDevitt. Thomas and Saad Gadalla. 1986. “Incorporating husband-wife differences in place utility differentials into migration decision models.” Population and Environment, 8(1-2):98-1 19. Ministry of Labor. Invalids and Social Affairs, Center for Scientific Information on Labor and Social Affairs (MOLISA). 1997a. T huc trang lao dong viec lam o Viet nam (Tu ket qua dieu tra lao dong-vice [am I 996)[status of labor-employment in Vietnam (from survey on labor-employment 1996)]. Ha Noi: Statistical Publishing House. Ministry of Labor, Invalids and Social Affairs, Center for Population and Labor Resources Studies (MOLISA). 1997b. Bao cao: Luan cu khoa hoc cho quyet dinh dieu chinh quan he cung [a0 dong nong thon va cau lao dong thanh thi [Report Scientific basesfor the readjustment in the relations between rural labor supply and urban labor demand]. Ha Noi. Moghadam. Valentine (ed.), 1995a. Economic reform, women’s employment, and social policies. World Development Studies 4. UNU/WIDER. Moghadam. Valentine. 1995b. “Market reforms and women’s economic status: Eastern Europe, Russia, Vietnam and China”. Journal of SID, I 995: 61-66. 315 Moore. Gwen. 1990. “Structural determinants of men’s and women’s personal networks.” American Sociological Review, 55:726-35. l\~lorrison. Donna Ruane and Daniel T. Lichter. 1988. “Family migration and female employment: the problem of underemployment among migrant married women”. Journal of Marriage and the Family, 50:161-172. Morrison. Peter. 1977. “The functions and dynamics of the migration process.” P.61-74 in Brown. Alan and Egon Neuberger (eds). Internal migration: A comparative perspective. London: Academic Press. M ullan. Brenda. Chun-Hao Li. Rita Gallin. and Bernard Gallin. 1998. “Family and internal migration in Taiwan.” Asian and Pacific Migration, 7(1): 43-66. Mullan. Brendan. 1989. “The impact of social networks on the occupational status of migrants.” International Migration Review, 27(1):69-86. Nabi. Ijaz. 1984. “Village-end considerations in rural-urban migration.” Development Economics, 14:219-145. National Political Publishing House (N PPH) 1994. “ Thoa thuan ve tien cong. bao hiem y te, bao hiem xa hoi cho nguoi lao dong” [“Negotiation on wages, health insurance. and social insurance for laborers.”] In Bo luat lao dong nam I 994 [ The labor codes 1994.] Hanoi: National Political Publishing House. Neto. Felix. 1995. “Predictors of satisfaction with life among second generation migrants.” Social Indicators Research. 35(1):93-1 17. Ngo. Ba Thanh. 1989. Status of Women: Vietnam. Bangkok: UNESCO principle regional office for Asia and the Pacific. N guyen, Binh Yen. 1997. “Mot so suy nghi ve dac trung co ban cua y thuc he phong kien Vietnam.” [Thinkings of main features of feudal ideology in Vietnam] Information of Social Science, 7:43-48. Nguyen. Dinh Tan. 1995. “Cac dac trung va xu huong bien doi co cau lao dong nghe nghiep o nuoc ta trong giai doan hien nay.” [Characteristics and changing trends of labor and occupational structure in this period] Pp. 47-62 in Do, Nguyen Phuong ( ed.) Thuc trang va xu the phat trien. Co cau xa hoi nuoc ta trong gian doan hien nay [Situation and development orientation. Social structure of the country in this period]. Hanoi: Social Science Publishing House. Nguyen, Huu Minh. 1998. Marriage pattern in Vietnam. Dissertation, submitted to University of Washington in 1998. 316 Nguyen. Huu Tien et a1. 1996. “Ket qua chuyen dich co cau kinh te o nuoc ta sau 10 nam thuc hien cong cuoc doi moi.” [Results of transformation of economic structure in our country after 10 years of innovation] Pp.23-29 in The National Economic University. Doi moi va phat trien nong nghiep nong thon [Innovation and agricultural development in rural areas]. Hanoi: Agriculture Publishing House. Nguyen. Luong Trao. 1998. “Migration in the past decades. Situation and solution” Paper presented in the Conference for Voluntary Migration Policies. held in Ho Chi Minh City, 6-7 July. 1998. N guyen. Quang Hue. 1998. “ Ruralward voluntary migration: situation and solution” Paper presented in the Conference for Voluntary Migration Policies, held in Ho Chi Minh City. 6-7 July, 1998. Nguyen. Quang Kinh. I994. “Complementary education. in-service training and continuing education.” Pp. 1 36-145 in Pham, Minh Hac (ed.) Education in Vietnam 1945-1991. Hanoi: Education Publishing House. Nguyen. Quoi. 1996. “Nguoi nhap cu tu do vao thanh pho Ho Chi Minh” [Voluntary migrants into Ho Chi Minh City.] Sociological Review, 55(3):66-72. N guyen. Thanh Tuan. 1998. “Building the triangle development areas in relation to migration and population distribution” Paper presented in the Conference for Voluntary Migration Policies. held in Ho Chi Minh City, 6-7 July, 1998. Nguyen. The Giai. 1988. Luat hon nhan va gia dinh voi thanh nien [The marriage and family law, and young people. Hanoi: Youth Publishing House. N guyen. Tu Chi. 1993. “The traditional Viet village in Bac Bo: its organizational structure and problems.” Pp.44-142 in Phan, Huy Le et al. The traditional village in Vietnam. Hanoi: The Gioi Publishing House. N guyen. Van C hinh. 1996. “Van de “cho lao dong” o Hanoi” [Problem of “free labor” market in Ha noi]. Sociological review, 54(2): 58-70. Nguyen. Van Tai. 1998. Hien trang, tac nhan thuc day va cac van de phat sinh tu hien turong di dan tu do doi voi su phat trien kinh te xa hoi, moi truong o thanh pho Ho Chi Minh trong qua trinh doi moi nen kinh te dat nuoc. Cac giai phat giai quyet [ Situation, push factors and issuesfrom voluntary migrationfor socio- economic development and environment of Ho Chi Minh City in the innovation. Solutions]. Ho Chi Minh City: VNRP. Nguyen, Xuan Thao. 1998. “Some opinions about policy orientation and solution for migration.” Paper presented in the Conference for Voluntary Migration Policies. held in Ho Chi Minh City, 6-7 July, 1998. O’Connor. Mary. 1990. “Women’s networks and the social needs of Mexican immigrants.” Urban Anthropology, 19(1-2): Ong. Aihwa. 1987. Spirits of'resistance and capitalist discipline: Factory women in .IIIaIaysia. Albany: State University of New York Press Payner. Deborah, John Warner, and Roger Little. 1992. “Tied migration and returns to human capital: the case of military wives.” Social Science Quarterly, 73(2):324- 339. Pham. Bich San. 1992. “Mot so van de nghien cuu Hanoi va nhung nhan xet so khoi qua nghien cuu giai doan 11.” [Some issues of research on Hanoi and evaluation based on research of second stage of university education] Sociological Review, 40(4):29-34. Phan. Dai Doan. 1995. “The state and the society, from the real situation in rural areas nowadays.” Sociological Review, 51(3): 30-40. Phan. Ke Binh. 1996. Vietnam phong tuc [Vietnamese customs] Ho Chi Minh City: Dong Thap Press. Phan. Quynh Nga. 1998. N ghien cuu tac dong qua lai cua hien trang di dan tu do va doi song lao dong nu o cac xa Vinh My, Vinh Hung, Vinh Giang, Vinh Hien,- huyen Phu loc, tinh Thua- Thien-Hue. C ac giai phap de xuat [Interaction impact of voluntary migration and lives of female laborers in communes Vinh My, V inh Hung. Vinh Giang, Vinh Hien,- district Phu loc, province Thua-Thien-Hue]. Hue: VN RP. Piampiti. Suwanlee. 1984. “Female migrants in Bangkok metropolis.” Pp227-246 in Fawcett. James, Siew-Ean Khoo, and Peter Smith (eds), Women in the cities of Asia. Colorado: Westview. Poot. .Iacques. 1993. “Adaptation of migrants in the New Zealand labor market.” International Migration Review. 27:121-39. Portes. Alejandro. Manuel Castells. and Lauren Benton (eds) 1989. The informal economy. Studies in advanced and less developed countries. Baltimore and London: The Johns Hopkins University Press. Portes. Alejandro & Julia Sensenbrenner. 1993. “Embeddedness and immigration: Notes on the social determinants of economic action.” American Journal of Sociology, 98(6): 1320-1350. Portes. Alejandro and Ruben Rumbaut. 1996. Immigrant America: A portrait. Berkeley: University of California Press. 318 Portes. Alejandro. 1998. “Social capital: Its origins and applications in modern sociology.” Annual Review of Sociology, 24: 1-24. Portes. Alejandro and Ruben Rumbaut. 2001. Legacies: the story of the immigrant second generation. Prothero. Mansell and Murray Chapman (eds) 1985. Circulation in Third World countries. London and Boston: Routledge and Kegan Paul. Pryer. .Iane. 1992. “Purdah. patriarchy and population movement: perspectives from Bangladesh” pp. 139-153 in Chant (ed.) Gender and Migration in developing countries. London and New York: Belhaven Press. Pugliesi. Karen and Scott Shook. 1998. “Gender, ethnicity and network characteristics: Variation in social support resources.” Sex Roles, 38(3-4):215-23 8. Pyle. Jean. 1994. “Economic restructuring in Singapore and the changing roles of women, 1957 to present.” pp.129-144 in Aslanbeigui, Nahid. Steven Pressman, and Gale Summerfield (eds) 1994. Women in the age of economic transformation. Gender impact of reforms in post-socialist and developing countries. London and New York: Routledge. Repak. Terry. 1994. “Labor recruitment and the lure of the capital: Central American migrants in Washington, DC.” Gender and Society, 8(4):507-24. Reskin. Barbara (ed.) 1989. Sex segregation in the workplace: Trends, explanations, remedies. Washington, DC: National Academic Press. Rhoda. Richard. 1983. "Rural development and urban migration: Can we keep them down on the farm?" International Migration Review, 17:34-64. Richter. Kerry. 1994. “Internal migration in Thailand.” Asian and Pacific Migration 3(4):531-44 Riley. Nancy E and Robert Gardner W. 1993. “Migration decisions: The role of gender.” Pp. 195-206 in Internal migration of women in developing countries. New York: United Nations. Ringdal. Kristen. 1993. “Migration and status attainment among Norwegian men.” Acta- Sociologica, 36(4):327-42. Roberts. Bryan R. 1989. “Urbanization, migration and development.” Sociological Forum, 4(4):665-9l. Robinson. V. 1993. “Race. gender and internal migration within England and Wales.” Environment and Planning A, 10(25): 1453-1465. Rodenburg. Janet. 1993. “Emancipation or subordination? Consequences of female migration for migrants and their family.” Pp.273-289 in Internal migration of women in developing countries. New York: United Nations Rumbaut. Ruben. 1997. “Introduction: immigration and incorporation.” Sociological Perspectives, 40(3):333-8. Sandell. Steven H. 1977. “Women and the economics of family migration.” Review of Economics and Statistics, 592404-414. Sanders. J imy and Victor Nee. 1996. “Immigrant self-employment: The family as social capital and the value of human capital.” American Sociological Review. 61 :231- 249. Sandu. Dumitru & Gordon DeJong. 1996. “Migration in market and democracy transition: Migration intentions and behavior in Romania.” Population Research and Policy Review. 15(5-6):437-57. Sexton. Sarah. 1996. “A women’s work is never done.” Ecologist, 26:138-9. Shah. Nasra M and Smith. Peter C. 1984. “Migrant women at work in Asia.” Pp. 297-322 in Fawcett, James. Siew-Ean Khoo, and Peter Smith (eds) Women in the cities of Asia. Colorado: Westview. Shihadeh. Edward S. 1991. “ The prevalence of husband-centered migration: employment consequences for married mothers.” Journal of Marriage and the Family, 53:432-444. Sidorenko. Elena. 1995. “Gender, migration and the formation of ethnic niches in the labor market: The case of Ukrainian workers in the Czech Republic.” Mi gracijishe Teme, 1 1(2): 127-49. Singh. DP. 1998. “Female migration in India.” Indian Journal ofSocial Work, 59(3):728-42. Sjaastad. Larry A. 1962. “The cost and returns of human migration.” Political Economy, 70:80-93. Sowell. Thomas. 1997. Migrations and cultures: A world view. New York: Basic Books. Speare. J. Frances Kobbin, and Ward Kingkade. 1982. “The influence of socioeconomic bonds and satisfaction on interstate migration.” Social Forces. 61(2):551-74. Spelman. Elizabeth. 1988. Inessential woman: Problems of exclusion infeminist thought. Boston: Beacon Press. 320 Spitze. Glenna. 1986. “Family migration largely unresponsive to wife’s employment.” Sociologv and Social Research. 70(3):23 1-34. Stark. Oded. I984. “Rural-urban migration in LDC s: A relative deprivation approach.” Economic Development and Cultural Change. 32:475-485. State Planning Committee (SPC). 1994. Khao sat muc song dan cu Viet nam 1 992-1 993 [Vietnam Living Standard Survey 1 992-1 993]. Ha Noi: General Statistical Office. Strauch. Judith. 1984. “Women in rural-urban circulation networks: Implications for social structural change.” Pp63-77 in Fawcett. James. Siew-Ean Khoo. and Peter Smith (eds) Women in the cities of Asia. Colorado: Westview. Summerfield, Gale. 1994. “Chinese women and the post-Mao economic reforms.” Pp.1 13-128 in Aslanbeigui, Nahid. Steven Pressman, and Gale Summerfield (eds) 1994. Women in the age of economic transformation. Gender impact of reforms in post-socialist and developing countries, London and New York: Routledge. Tana. Li. 1996. Peasants on the move. Rural-urban migration in the Hanoi region. Singapore: Institute of Southeast Asian Studies. 'I’aylor. Edward. 1986. “Differential migration, networks, information and risks.” In Research in human capital and development: Migration, human capital and development (ed.) by Oded Stark. Conn.: JAI Press Inc. Teachman. Jay. Kathleen Paasch, and Karen Carver. 1997. “Social capital and the generation of human capital.” Social Forces, 75(4): 1-17. Thadani. Veena and Michael Todaro. 1984. “Female migration: A conceptual framework.” Pp. 36-60 in Fawcett, James, Siew-Ean Khoo, and Peter Smith (eds) Women in the cities of Asia. Colorado: Westview. Than. Van Lien. 1997. Nghien cuu cac yeu to thuc day [an cong di dan tu nong thon ra do thi trong qua trinh chuyen doi kinh te o nuoc ta hien nay va anh huong cua not toi su phat trien kinh te xa hoi cua cac vung do thi va nong thon [Study of push factors of the rural-urban migration under the transition period and their effects on socio-economic development in the rural and urban areas]. Ha Noi: Ministry of Technology and Environment. Todaro. Michael P. 1976. Internal migration in developing countries. International Labour Office Geneva. Trager. Lilian. 1994. “Family strategies and the migration of women: Migrants to Dagupan city. Philippines.” International Migration Review, 18(4):1264-1277. b) I\) Tran. Dinh Huou. 1989. “Ve gia dinh truyen thong Viet nam voi anh huong Nho giao” [About the traditional Vietnamese households under the influence of Confucianism] Sociological Review, 27(2):25-35. Tran. Dinh Huou. 1996. “Gia dinh va giao duc gia dinh” [Family and education in family] pp.49-93 in Tuong Lai et al. (eds) Nhung nghien cuu xa hoi hoc ve gia dinh Vietnam [Sociological research on families] Volume 2. Hanoi: Social Science Publishing House. Tran. IIoang Kim. 1992. Economy of Vietnam: Review and statistics. Hanoi: Statistic Publishing House. Tran. Thi Que. 1995. “Gender issues in Vietnam’s development” pp.187-206 in Norlund et al. (eds) 1995. Vietnam in a changing world. Britain: Curzon Press. Tran. Thi Van Anh & Le Ngoc Hung. 1996. Phu nu, gioi va phat trien [women gender and development]. Hanoi: Women Publishing House. Treiman. Donald. 1977. Occupational prestige in comparative perspective. New York and London: Academic Press. T rinh. Duy Luan. 1992. “Su phan tang xa hoi theo muc song tai thu do Hanoi trong nhung nam dau thuc hien Doi moi” [Social stratification based on standard of living in Hanoi in the first years of Innovation] Sociological Review, 40 (4)216-28. Trinh. Duy Luan. 1996. Tim hieu mon Xa hoi hoe Do thi [Overview of urban Sociology]. Hanoi: Education Press. Trinh, Khac Tham and Doan Mau Diep. 1998. “Di dan den Ha noi - Thuc trang va khuyen nghi. Cac giai phap dieu tiet va quan ly” [Migration to Hanoi - situation and commends. Solution for regulation and management]. Paper presented in Conference on Policies for voluntary migration, July 6-7. 1998, Ho Chi Minh City. Trinh. Khac Tham. 1996. “Di dan noi dia 0 Vietnam - Thuc trang va xu huong” [Intemal migration in Vietnam - situation and orientation] paper presented in lntemational Conference “Some issues on population and development in Vietnam”, May 3-4. 1996. Ho Chi Minh City. 'I‘ruong. van Phuc. 1996. “Chat luong lao dong va van de cung cau lao dong theo co cau dao tao.” [Labor quality and issues of needs and demand of labor based on training structure]. Information of labor market, 1:7-10. Trzcinski. Eileen and Susan Randolph. 1991. “Human capital investments and relative earnings mobility: The role of education, training, migration and job search.” Economic Development and Cultural Change 40(1-2):153-168. 322 Tyner. James. 1994. “The social construction of gendered migration from the Philippines.” Asian and Pacific Migration, 3(4): 589-617. United Nation (UN). 1993. Expert meeting on the Feminization of internal migration. Internal migration of women in developing countries. New York: United Nation. United Nations (UN). 1986. World survey on the role of women in development. Sales No.E.89.lV2 U nited Nations (UN). 1994. The migration of women. Methodological issues in the measurement and analysis of internal and international migration. Santo Domingo: INSTRAW United Nations (UN) Secretariat. 1993. “Types of female migration.” Pp.94-1 15 in Internal migration of women in developing countries. New York: United Nations Van Tao. 1998. “Su chuyen doi cac gia tri truyen thong gia dinh Vietnam hien nay va dinh huong phat trien” [Changes in traditional values of family in Vietnam and development trends] Paper presented in Conference of Vietnamese Studies, held in Hanoi. July 1998. Vecoli. Rudolph. 1998. “Migrations and cultures: A world view.” International Migration Review, 32(2):494-495. VHT 1993. Lich su tu luong Vietnam [History of thoughts in Vietnam]. Hanoi: Publishing House of Social Sciences. Vietnam Population Census-I989. (VPC) 1991a. Phan tich ket qua dieu tra mau [Detailed analysis of sample results]. Ha Noi: General Statistical Office. Vietnam Population Census-1989. (VPC) 1991b. Ket qua dieu tra toan dien [Completed census results]. Volume 1. Ha Noi: General Census Steering Committee. Vu. Dung. 1996. “Su thich nghi cua dan cu voi co che tri truong” [Adaptation of the population with market mechanisms] Economy and Diagnose, 8:35-36. Vu. Manh Loi. 1991. “Khac biet 11am nu trong gia dinh nong thon dong bang Bac bo” [Gender differences in the households of the northern deltas] pp. 141-156 in Institute of Sociology at Ha noi and Department of Sociology at Gothenburg, Sweden (eds) Nhung nghien cuu xa hoi hoc ve gia dinh Vietnam [Studies of the sociology of Vietnamesefamilies]. Ha Noi: Social Science Publishing House. Wallimann. Isidor. 1994. “Can modernity be sustained? Prevention of mass death and genocide.” Population Review, 38(1-2):36-45, Lu to La Weeks. John. 1996. “The six pillars of global population and social change.” in Rumbaut. Ruben (ed.) 1995. Population research group. Research paper 95-01. MSU. West. Candace and Sarah Fenstermaker. 1995. “Doing difference.” Gender and Society, 9( 1 ):3-3 7. West. Candace and Don H. Zimmerman. 1987. “Doing gender.” Gender and Society, 1:125—151. Wilk. Richard. 1989. “Decision making and resource flows within the household: beyond the black box” in Richard Wilk (ed.) The household economy: reconsidering the domestic mode of production. Colorado: Westview Press. Williams. Anne. Patrick Jobes. and Jack Gilchrist. 1986. “Gender roles. marital status and urban-rural migration.” Sex Roles, 15(1 1-12):627-642. Wilson. Tamar. 1994 “ What determines where transnational labor migrants go? Modification in migration theories.” Human Organization, 53: 269-78. Wilson. Tamar. 1998. “Weak ties. strong ties: network principles in Mexican migration.” Iluman Organization, 57(4): 394-403. \Nv’inehie. Diana & David Carment. 1998. “Migration and motivation: the migrant’s perspective.” International Migration Review, 23(1 or 95):96-104. Wolff. Peter. 1999. Vietnam — The incomplete transformation. London: Frank Cass World Bank. (WB)1995. Vietnam poverty assessment and strategy. Washington DC: World Bank. Yang. Xiushi. 1993. “Household registration, economic reform and migration.” International Migration Review, 27(4 or 104):796-818 Yarr. Linda. 1995. “Su han che ve thoi gian va cuoc dau tranh vi binh dang nam nu cua phu nu Vietnam va phu nu My” [Time limitation and struggle for gender equality of Vietnamese and American women] pp. 155-181 in Center for Social Sciences and Humanity (eds) Gia dinh va dia vi nguoi phu nu trong xa hoi. Cach nhin tu Vietnam va Hoa ky [Family and women 's status in society. A Vietnamese and American view]. Hanoi: Social Science Publishing House. Zlotnik. Hania. 1993. “Women as migrants and workers in developing countries”. International Journal of Contemporary Sociology, 30(1):39-62. Zlotnik. Hania. 1995a. “The South-to-North migration of women.” International illigration Review, 29(1):229-254. 324 Zlotnik. Hania. 1995b. “Migration and the family: The female perspective.” Asian and Pacific Migration, 4(2-3): 253-271. b) l\) U! I1111111111111I