Remittance Micro Determinants and Socioeconomic Impacts: A Household Unit Analysis of Gender Behaviour of Rural Household Head in Bangladesh

This paper analysed the effect of gender differences of household head amongst 10 villages on the remittance determinants and their socioeconomic impacts of remittance receiving households in rural Bangladesh. Using micro-economic data from a survey conducted in 2013, multivariate analysis was carried out on 300 rural households. The empirical findings provided that the remittance determinants (migrant age, marital status and number of visit; household head age, marital status, level of education and employment status) and impacts (saving and investment patterns; determinants: age of migrant, number of visit by the migrant, marital status of the migrant, age and education of household head) vary from male to female household head at the same community level of the households. The study suggest that the strong gender differences in the remittance behaviour of Bangladeshi households and highlight the importance of differentiating with respect to gender background when analysing the determinants of remittances.


Introduction
According to the International Organization for Migration (IOM), there are an accounted 191 million global migrants in 2005, up from 176 million in 2000. Migrants include 3.0 per cent of the worldwide population. For the period 2000-10, the world migrant stock increased double as fast than during the last decade. In 1990s, the global migrant stock increased at an average of about 2 million migrants per year. During the period 2000-10, the outgrowth in the migrant stock accelerated to about 4.6 million migrants annually. There are 232 million international migrants are staying in the world today. Since 1990, the number of international migrants in the global North grew by about 53 million (65%), on the other hand the migrant population in the global South increased by about 24 million (34%). Nowadays, around six out of every ten international migrants stay in the developed nations (UN, 2013).
In 2006, remittance flows are accounted to have go beyond USD 276 billion globally, USD 206 billion of which sent to developing countries. According to World Bank database (2014), the global remittance flow, which has touched $550 billion last year, is expected to grow by 8 per cent per annum in the next few of years. Of the total remittance fund, $414 billion were received by developing countries, especially Bangladesh, China, India, Mexico, the Philippines, and Pakistan.
Bangladeshi migrants in Italy are predominantly single and male migrants who are living under 'transnationally split' (Yeoh, Graham, and Boyle, 2002) conditions and obligated to maintain economic and social relations with their family members back home (Rahman and Kabir, 2012). The obligation of maintaining sustained economic and social ties with home stems from the dominance of the household in the social and economic affairs of the Bangladeshi society and their transnational household members. Individual migrant is deeply enmeshed in a complex web of household relations and dependencies: He/she moves internationally for work as an envoy of the extended household that places the well-being of the extended family above the individual migrant's interests (Rahman, 2011). Whether it is temporary labour migration such as migration to the Middle East or more permanent form of migration such as migration to Italy, maintaining sustained economic relations with left behind households remain one of the key priorities for migrant members (Ullah, 2010, Rahman 2009). This is comprehensive evidenced in the annual inflow of remittances to Bangladesh, which has increased from around $4.2 billion in 2005 to nearly $10.9 billion in 2013 (BMET, 2014).
These facts and figures indicate that international migration and remittance is an intricate phenomenon, the dynamics of which are increasingly turning a drastic policy topic global economic, social, legal and cultural topic.

Problem definition
What is gender difference of household head of factors influences remittance inflows and the socioeconomic impact of remittance at the left behind rural household member in Bangladesh those who have been receiving remittances from their household migrant member in Italy?
Research questions of the study 1. Does remittance determinants influencing due to gender difference the amount of remittance sent by household members working in Italy to their homes in Bangladesh?
2. What is the socioeconomic impact of such remittances for the gender issue of household head of the rural households in Bangladesh?

Objective of the study
Based on the research questions this research has three objectives.
1. To explore the factors influencing the amount of remittance sent by household members working in Italy to their homes in Bangladesh.
2. To investigate the socioeconomic impact of such remittances for the gender issue of household head of the rural households in Bangladesh.

Theoretical framework
Research and empirical findings on the determinants of international remittances in Bangladesh is limited by the remittance processes and data discrepancies. Different theories explicate different outcomes of the remittances. Among the factors, household structure (e.g., household size, male-female ratio), income sources, gender of household head, time abroad, etc. play influential roles in determining the amount and use of the remittances in the household.
The New Economics of Labour Migration (NELM) characterises migration as a household decision determined by the specific household characteristics (Stark and Bloom, 1985). Therefore, the issue of remittances and their determinants has become a key consideration in a number of theoretical and empirical studies. A distinguishing attribute of the NELM model is its simultaneous consideration of migration determinants, remittance behaviour and impacts making NELM models relatively demanding in estimation methods and data requirements. Another key insight of NELM is that households allocate members to improve not only absolute but relative income and alleviate their deprivation against a reference group like the village community (Stark and Bloom 1985;Stark, Taylor and Yitzhaki, 1986;Stark and Taylor, 1989;Stark, 1991).
Micro-level studies of remittances are either based on household surveys that include remittance-receiving households (Gubert, 2002) or specific surveys of the migrants in the originating or destination country (Amuedo-Dorantes and Pozo, 2006;Holst and Schrooten, 2006). Against such a theoretical framework, a broad stream of studies has pioneered the determinants of remittances and their socio-economic impact on countries, communities and households that have sent workers abroad (Borja 2012). Two broad perspectives have been adopted: macro determinants and impact of foreign remittances on originating countries and their micro determinants and consequences. The approach adopted in this study will be on the latter as the study area is the Shariatpur District of Bangladesh which has traditionally sent migrant workers to Italy for a considerable time.

Literature review
The literatures on remittances and gender have so far explored mostly on the impact of gender on the remitting patterns of migrants at the destination. Early 1980s, a few studies, using mostly internal remittances data, indicate that gender does influence remitting behaviour (Lucas and Stark, 1985;Kaufmann and Lindauer, 1986). In the 1990s also supported this idea, but these studies have also been able to construct a story of why we may observe these differences between males and females. As for, Osaki (1999) explains that in Thailand, children are expected to repay their parents for time and money invested in raising them. According to Buddhist traditions, males can earn religious merit for their parents if they become ordained and spend a period of their lives in monkhood. However this option is not available for females, female migrants are more likely to contribute financially to the household via remittances. During the 2000s have also supported differences in remitting patterns across gender in different countries (Semyonov and Gordozeisky, 2005).
Using data from the Dominican Sierra (de la Brière, Sadoulet, de Janvry and Lambert 2002), find that there are structural differences in remittance behaviour across gender, reflected in the significance of the interaction terms of gender with other determinants of remittances. However, Agarwal and Horowitz (2002), using data from Guyana, find that gender matters for the amount remitted (males remit a higher amount), but not for the likelihood of remitting. Moreover, Sana and Massey (2005) find that in the Dominican Republic, daughters living abroad are stronger predictors of remittances than sons living abroad. The inverse is true for Mexico, while sons living abroad are stronger predictors of remittances than daughters living abroad. The authors feel that Mexican sons are expected to be trustworthy remitters, while Dominican men are less reliable remitters. Moreover, Blue (2004) found that female Cuban migrants are more reliable remitters than male Cuban migrants. Furthermore, Naufal (2008) finds that female migrants behave more altruistically than their male counterparts. Therefore, the nexus between migrant and household members are also play important role in the remittance behaviour.
The literature has also contributed insights on the impact of remittances on household consumption. For instance, Airola (2007) finds that remittances induce greater spending (as a share of income) on healthcare and housing in Mexico. Moreover, Adams and Cuecuechea (2010), using data for Guatemala, find that international remittance-receiving households spend more on education, health, and housing, and less on food than do other households. However these and other studies have explored the differences in household expenditures across remittance receiving and non-receiving households, these studies have not disclosed in detail the role of gender.
However, one of the study that does address this issue is Rahman and Fee (2009) for the case of Indonesian domestic workers in Hong Kong, Singapore and Malaysia. Their results suggest that the majority of migrants (about 67%) had control over the use of remittances and that female recipients of remittances tend to use the money for human capital acquisition, while males tend to use the money for physical capital acquisition. The results of Rahman and Fee (2009) have an important implication for this study. The expectations of the migrant about the use of remittances seem to matter; thus, the gender of the migrant could be of importance. There are, however, three important differences between the study of Rahman and Fee (2009) and the present study. Firstly, they just explore descriptive statistics of the survey but do not conduct any type of econometric analysis. Secondly, the authors emphasize the use of remittances and the gender of the receiver, but not the gender of the migrant. Thirdly, they focus exclusively on the case of domestic contact workers that, while interesting, limits the implications of their results for other types of migrants.
This paper also argue that it is important to look at the overall expenditure distribution of the households. If the households take remittance money and put that money towards food, but then re-direct other sources of income from food to housing, the overall impact of remittances may not be an increase in the demand for food products. Therefore, the pivotal point is to delve out how remittances affect the overall household expenditure distribution. Hence, this study take part different statistically analysis remittance from male migrant to the different level of rural household This investigation is strongly related to the literature on the role of gender in household consumption patterns, remittance determinants and also their socioeconomic impact.
The study on intra-household expenditure patterns rejects the unitary household model and suggests that there are differences in preferences among household members (Guzmán, Morrison, and Söjblom, 2008). In point of fact, changes in gender control of income could translate into changes in expenditures and budget shares (Doss, 2006). For example, Hoddinott and Haddad (1995) find that a larger share of income given to the wife increases the budget share of food and decrease the budget share of alcohol and cigarettes in Côte D'Ivoire. Further, recent evidence suggest that an increase in education spending is the most consistent effect across countries of having a larger share of resources in the hands of females (Guzmán, Morrison, and Söjblom, 2008).
The research on the role of gender in the household has mainly focused on the spending impact of increasing the share of income controlled by women. Certain significant event that can cause such an increase in the monetary role of women is migration (de Haas and Van Rooij, 2010). While men leave the household to work abroad, women may inherit additional tasks and responsibilities, which often involve increased access to income (mainly through remittances received) and the decision on how to spend it. To understand the outcomes of such intrahousehold dynamics, Guzmán, Morrison, and Söjblom (2008) examine the effects of the gender of the remitter and the head of the remittance-receiving household on household budget allocations. In this study use cross sectional data from Bangladesh to find that female-headed households have different expenditure patterns than male-headed households.
The investigation of the nexus between gender of household head and the use of remittances is not only a contribution to the studies on male migration, but also to the literature on migration, remittance and development in general. There are some macro level studies suggest that remittances may affect economic growth positively (Giuliano and Ruiz-Arranz, 2009;Mundaca, 2009), other studies suggest otherwise that remittances have only minor impacts on economic growth (Catrinescu, Leon-Ledesma, Piracha and Quillin, 2009;Ruiz, Shukralla and Vargas-Silva, 2009;Vargas-Silva, Jha and Sugiyarto, 2009). However, these studies differ in their conclusions, they agree on the fact that the actual use of remittances on the part of households is one of the key factors determining the growth impact of these transfers. For example, the impact of remittances should be different if remittances are used for household consumption and housing than if remittances are directed towards luxuries. However, at the micro level, remittances may also affect several variables. Within other variables, the impact of remittances on the human capital acquisition of both children and adults has received special attention in the literature (Amuedo-Dorantes, Georges and Pozo, 2008;Bansak and Chezum, 2009;Borraz, 2005;Edwards and Ureta, 2003). Moreover, remittances could also be used for housing purposes, either to purchase a home, improve the current one or rent a better one (Obeng-Odoom, 2010;Osili, 2004). The growing investment in housing in developing countries could potentially increase household assets and improve the quality of life.

Methodology
This study chose a quantitative method approach as its methodology to accommodate method for an extensive solution of the research problem and answer the research questions. Below Table 1 has specified the analytical tools and method pertinent to the research objectives and questions of this study.

Selection of survey village and course of the survey
In line with the study focus, the selection of the study area in Bangladesh was based on the high incidence of household members migrating to Italy at the sub-district level (Upazila) and the prevalence of remittance-receiving households at the sub-sub-district level (Union Parisad). Shariatpur is located in the Dhaka division and in the greater Faridpur District. Among the households, a significant number of migrants are from Naria Upazila, Shariatpur District. Naria sub-district has 14 sub-sub-districts and Vogeshore union one of the sub-sub-districts, has been selected randomly for census data because there is no available published data on Bangladeshi migrant workers in Italy. Emigration from Bangladesh to Italy is predominantly a rural phenomenon. Therefore, the fieldwork undertaken for this research consists of an ethnographic village study in Bangladesh with particular reference to remittance sending migrant worker in Italy to bridge the micro and macro paradigms of migration and remittance, and offer analytical insights into the determinants and impacts of such remittance.

Study Design
The primary data was collected from households in the Naria Upazila of Shariatpur District in Bangladesh as the researcher is from this area and is familiar with its geography and people. In the second phase, first-hand knowledge was obtained through ask a single question (whether the household has members who have worked in Italy or not) to each of the 4013 households in the 10 study villages. Thereafter a structured questionnaire in which several open ended and closed ended questions were asked to exactly identify different factors playing a pivotal role for migrants' families. The responses were collected in a quantitative way, i.e. through an appropriate questionnaire, and through a qualitative method, through conducting direct interviews. The respondents were the heads of households or senior members of families which had a member. Their responses were analysed and summarised to derive conclusions about the migration impacts, by post-and pre migration data.

Sample Size
In selecting a representative sample of the population, Krejcie and Morgan's (1970) recommendation was accepted in this study. After categorising the household migrant members in Italy a random sample of 300 households was selected, the share in each village corresponding to their proportion in the whole population (the remittance received household). Then, the remittance received households in each village were picked randomly. In the process, every household was coded during the first stage census survey and recorded pn a separate identical size of piece of paper. Thereafter, all folded papers were thoroughly mixed up to assure the same probability of selection of each household and to overcome systematic sampling error. One folded paper was picked up each time by the researcher himself. After each selection, the pile of folder papers was mixed up again and another person was chosen only to pick up another folded paper and the process continued until the sample remittance received household total was attained. Finally the interviews of selected households were administered with structured and semi-structured questionnaires.

Ethical Issues
This research was conducted in compliance with the National Statement on Ethical Conduct in Human Research (2007) and was approved by the Human Research Ethics Committee of Southern Cross University (Approval Number .

Econometric model building and multivariate analysis
The long and short regression allows the assessment of the direction and strength of causality existing between the dependent and independent variables. The best fit model has statistically been developed, both (short and long) regression models are applied in the data analysis because whole sample has been broken into sub-samples with specific attributes (

Results
The cross tabulation of the 300 household participants survey response about the remittance receiver gender and remittance indicates in the table 3 and the results shows that the majority (53.0%) household head were male while female (47.0%). Both of the household heads were received various ranges of remittances. The 1.9% of male household head were received the highest range of remittances BDT 9,00,001 to 10,00,000 while the female household head BDT 14,00,001 to 15,00,000 which also 2.1%.

Table 3: Cross Tabulation Household yearly remittance received and gender of household head
However, as depicted in the figure 1 indicates that the male household highest (2%) remittances range BDT 9,00.001 to 10, 00,000 and maximum percentage of (24%) household were received BDT 500,001 to 600,000 yearly remittances from their migrant household members at the destination.

Figure 1: Distribution of male household head remittance
Remittance frequency table 4 show that the male household were received remittance bimonthly (64.15%), at necessary (22.64), monthly (11.32%), and festivals (1.88%) while female household bimonthly (48.93%), monthly (29.78%) and at necessary (21.27%).  Uses of remittance by the male and female Table 6 explores that both 100% participants households use for consumption. Both genders almost nearest percentage of uses of remittance were repayment of loan, education, health telecom, renovation of house, gift and donation, social ceremonies, durable purchase, jewellery, fishing and others. The male household were invested more on the business sector while less percentages of female. The male household head were bought (73.6%) motor cycle and spent money on tobacco (86.8%) while female very less in these sectors.

Discussion
The results revel that the male household head receive remittance during the festivals while none of female household head receive to their festivals. This might happen because female household head receive remittance more on bimonthly and monthly basis. On the other hand, migrant might visit on the festival who have left female household head at the origin. In this case, they bring lots of remittance as a kind, such as gift for relatives and friends.
Migrant marital status and residency pattern of household members, including spouse and children are significant determinants of remittance motivation (Johnson and Whitelaw 1974;Menjivar et al. 1998;Vanwey 2004;Casale and Posel, 2006;Luke, 2007;Alba and Sugui 2009), it also vary single migrants and married heads living status (Sahu and Das, 2009), with gender discrimination (Sorenson, 2004a(Sorenson, , 2004b(Sorenson, , 2005Atekmangoh, 2011), with the number of nuclear household members living outside the household (DeVoretz and Vadean, 2007). The present study finds the marital status of migrant also significant determinants of remittance but the significance depends on the gender of household head. The study revel that the study area, migrant marital status is only small significance difference between male and female household head.
The age of household head factor vary from country to country (DeVoretz and Vadean, 2008), gender behaviour in the remittance motive, for example male household less like to receive remittance rather than female (McDonald and Valenzuela, 2012), the older household head receive more remittance than the younger household head which reveal that the adult children care for their old parents as well their grant parents (Germenji et al 2001). The current study also finds that the age of household head is one of the key determinants of remittance. But there is also little significance difference between the two genders of household head.
The households head employment status play vital role in remittance behaviour (Quartey, 2006), there are few sectorial differences, for instance, the household head employment in public sector, private formal sector, export farmer, crop farmer, private, informal and unemployment play different impacts. This study finds the difference between the gender differences of household head remittance behaviour with the employment status of household head. The male household head are statistically significant while female household head are insignificant.
Usually remittances are used by the left behind household members are various sectors like luxurious life style, welfare sector, loan repayment, housing, financial investment, business sector, healthcare, consumption, children education, agro based sector, land buying and also for others. This study also finds almost similar outcomes with other studies. However, the study revel new finding that there is discrimination in remittance use due to gender behaviour of the household at the origin. The male household head were invested more on business, buying motor cycle, tobacco while female household head on the fixed deposit.
Socioeconomic impact of remittance for the male household head is highly correlates with number of visit by the migrant, marital status of the migrant and age of household head. While the female household head is highly correlates with age of migrant, number of visit by the migrant and educational level of the household head.

Contributions
This research has contributed to three fields of knowledge as follows:

Methodological implication
From the economic perspective, the analysis helps in the examination of what factors influence to enhancing household remittance and what can lead to its deterioration. This implies that the impacts of diverse remittances determinants are crucial in designing the policy for specific categories of left behind village household.

Theoretical implications
This study has reviewed the major theories and literature on international migration and remittances to explore the dynamics of the socioeconomic phenomenon as it relates to its causes, consequences, and impacts in the left behind household as well as community level.

Policy implications
The importance of steady economic growth has been well recognised in a poor country like Bangladesh. It is anticipated that the findings from this study will contribute to the understanding of this phenomenon, its theoretical underpinnings and the formulation of relevant policies to improve the use of such remittances on the livelihoods of the originating households in emerging countries. The results then will be interpreted and conclusions drawn. However, it is very obvious that foreign remittances play a vital role both at micro and macro level of the economy therefore it is very significant to identify the factors that can cause the inflow and growth of foreign remittances in Bangladesh. In this case, successful identification also helps to formulate appropriate policy decision from the part of Bangladesh Government so that the country can reap the best earnings from the inflow of foreign remittances.

Limitation and future research direction
The questionnaire survey participants were located mainly rural sub-sub-district of Shariatpur. This was a limitation as the findings might not be transferable and generalised for the whole of Bangladesh. More research is required adopting two-way surveys -Bangladesh and Italy -and ethnographic fieldwork for better understanding of the migration and integration process in Italy and their implications for the left-behind household in Bangladesh.

Conclusion
The present study investigated the gender difference of remittance determinants and their socioeconomic impact in rural area at the origin. The findings provided valuable discrimination between the gender discrimination of the household head. The household unit analysis show that remittance determinants vary from male to female household head at the same community level of households. Moreover, the remittance consumptions and used also differ between the genders. Furthermore, relationship between the socioeconomic impact from remittance and demographic determinants also vary due to the genders discrimination among the same rural area at the origin country.