This article was downloaded by: [NUS National University of Singapore]On: 26 June 2012, At: 19:45Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Canadian Journal of DevelopmentStudies/Revue canadienne d'études dudéveloppementPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/rcjd20

Bangladeshi labour migration to theGulf states: patterns of recruitmentand processesMd Mizanur Rahman aa Institute of South Asian Studies, National University ofSingapore, Singapore

Available online: 26 Jun 2012

To cite this article: Md Mizanur Rahman (2012): Bangladeshi labour migration to the Gulfstates: patterns of recruitment and processes, Canadian Journal of Development Studies/Revuecanadienne d'études du développement, 33:2, 214-230

To link to this article: http://dx.doi.org/10.1080/02255189.2012.689612

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae, and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

Bangladeshi labour migration to the Gulf states: patterns of recruitmentand processes

Md Mizanur Rahman∗

Institute of South Asian Studies, National University of Singapore, Singapore

ABSTRACT The Gulf countries in the Middle East are one of the largest regions relying oninternational labour migrants for economic development. Recruitment constitutes animportant part of this migration of labour. This study addresses the complexity andmultiplicity of labour recruitment in the Gulf countries through a case study of Bangladeshilabour recruitment. This study examines the labour recruitment to the Gulf, combiningnetworks and institutions to highlight both the operational and economic aspects of migrantrecruitment. This article reveals how migrant networks and recruitment agencies adapt to thechanging practices of recruitment to funnel migrant workers to the GCC countries and makeprofits out of the migrant workers in the recruitment process.

RESUME Les pays du Golfe constituent une des plus grandes regions reliant sur la migration demain d’œuvre pour le developpement economique. Le recrutement joue un role important danscette migration. Ce travail addresse la complexite et la multiplicite du recrutement dans les paysdu Golfe en analysant un cas d’etude au Bangladesh. Cette etude examine le processus derecrutement, particulierement les reseaux et les institutions qui soulignent les aspectsoperationaux et economiques de la migration. En plus, cette etude revele l’adaptation desreseaux de migration et des agences de recrutement aux pratiques changeants dans lerecrutement ainsi bien que les strategies employees pour assurer leur profit.

Keywords: labour migration; migrant worker recruitment; Gulf migration; temporarymigration; international migration

Introduction

The Gulf Cooperation Council (GCC) member states have emerged as one of the largest regionsrelying on temporary labour migrants since the early 1970s (Arnold and Shah 1986, Eelens andSpeckmann 1990, Shah 1994a). By the 1980s, migrant workers had outstripped the local work-force in the six member countries of the Gulf Cooperation Council (Bahrain, Kuwait, Qatar,Oman, Saudi Arabia and the United Arab Emirates) (Suter 2005, p. 28). The number ofmigrant workers has increased from 9 million in 1990 to an estimated 13 million in 2005 (Dito2008, p. 6). Migrants constitute a majority of the labour force in all the GCC countries, withthe average for 2004 coming to 70 per cent (Kapiszewski 2006, p. 4) and the average for 2008reaching 77 per cent (Winckler 2010, p. 12, Baldwin-Edwards 2011, p. 9). In particular, recruit-ment has made it possible to hire millions of migrants to the Gulf countries for temporary employ-ment (Eelens and Speckmann 1990, Esim and Smith 2004, Kuptsch 2006, Shah 2010).

ISSN 0225-5189 print/ISSN 2158-9100 online

# 2012 Canadian Association for the Study of International Development (CASID)http://dx.doi.org/10.1080/02255189.2012.689612

http://www.tandfonline.com

∗Email: mizan@nus.edu.sg

Canadian Journal of Development StudiesRevue canadienne d’etudes du developpementVol. 33, No. 2, June 2012, 214–230

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

In the migration of labour, several institutions play a role in assisting prospective migrantsmove between regions. Such institutions include: employers who directly recruit prospectivemigrants; public employment services that match local workers with foreign jobs; migrant net-works that facilitate access to foreign labour market; and private and for-profit recruitmentagencies that serve the prospective migrants (Martin 2005, 2006, Kuptsch 2006). Togetherthey form a recruitment structure which facilitates the recruitment of migrant workers in theGulf. Prospective migrants may consult with and use the services of any of these labour interme-diaries. Over time, however, direct employers and public employment services have substantiallydeclined, while migrant networks and private agents have increased worldwide (Martin 2005,p. xii). In the Gulf countries, the role of direct employment and public employment servicesfor low-skilled migrant workers has almost disappeared due to the rising number and the increas-ing diversity of migrants. In addition, the efficiency of networks and private agencies in recruitinglabour migrants has also contributed to the waning influence of direct employers and publicemployment agencies (IOM 2004, Kuptsch 2006, Shah 2010, Ullah 2010).

The existing literature explicitly suggests that recruitment agencies and migrant networks playa critical role in recruiting transient labour from South Asian countries to the GCC countries(Arnold and Shah 1986, Eelens and Speckmann 1990, Eelens et al. 1991, Shah and Menon1999, Gamburd 2000, Zachariah et al. 2001). However, at a theoretical and conceptual level,most of the existing literature focuses either on the role of migrant networks or on the role ofrecruitment agencies in the recruitment process. The explanation of labour recruitment alsotends to concentrate more on the operational aspect of recruitment – that is, how recruitmenttakes place in the migration process as opposed to analysing the prevalence and size of economictransactions. The economic aspect of the Gulf recruitment process is crucial, yet under explored.There is also little attempt to link the operational aspects with the economic side of recruitments.

Focusing on the experiences of Bangladeshi migrants to the Gulf countries, this study exam-ines the labour recruitment process from a holistic perspective, combining both networks andinstitutions to highlight both the operational and the economic aspects of labour recruitment.Linking these two aspects allows the study to unravel some of the complexities which existwithin contemporary migrant recruitment in the Gulf. Bangladeshi migrants have joined theGulf countries since the mid-1970s and Bangladesh since then has emerged as one of themajor labour-sending countries from South Asia. Despite being a major labour-contributingcountry for the Gulf, Bangladeshi labour recruitment has thus far received inadequate attentionin current migration studies. This study attempts to close the gap in the existing knowledge.

This study is broadly divided into two parts: the operational and the economic aspects ofrecruitment. In the operational dimension, the study covers a wide range of issues such as recruit-ment structure, licensed manpower recruiting agencies, their recruitment methods, migrant net-works and their respective methods. Regarding the economics of recruitment, the studyexplains methods of payment for recruitment fees and provides the economic cost of recruitmentto the Gulf counties. Before addressing the broader recruitment issues, the next section offerstheoretical and conceptual issues to migrant recruitment in the Gulf, followed by data sourcesfor the study. The subsequent section provides the trends and patterns of Bangladeshi migrationto the Gulf countries. Part of this section describes the socio-demographic profile of the migrantworkers in the Gulf, followed by a section on the kafala system under which migrants are invitedto work in the Gulf countries.

Theoretical and conceptual issues

Network analyses have generated enormous literature, explaining how networks facilitate andsustains migration across international borders (Massey et al. 1987, Boyd 1989, Faist 2000).

Bangladeshi labour migration to the Gulf states 215

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

Existing studies demonstrate that non-migrants, migrants and returnees are linked to each otherthrough social and symbolic ties that offer the social bedrock for sustained migration flowsacross international borders (MacDonald and MacDonald 1974, Gurak and Caces 1992, Shah1996, Shah and Menon 1999, Rahman 2009). ‘Social ties’ are often defined as a continuingseries of interpersonal transactions to which participants attach shared interests, obligations,expectations and norms, such as family and kinship ties, and ‘symbolic ties’ as a perceivedbond to which participants attach shared meanings, memories and future expectations, such asregional, national, ethnic and religious ties (Faist 2000). The content of social and symbolicties – obligations, reciprocity and solidarity – contributes to the evolution of migrant networksand the sustenance of the migration flows across international borders (Faist 2000, Masseyet al. 1987).

In the migrant network literature, social reward and social debt are usually behind the func-tioning of network-mediated migration (Boyd 1989, Faist 2000). However, there have been somechanges in the modalities of network-assisted migration, especially in low-level migrant workerrecruitment in the Gulf countries. This is partly due to the commercialisation of recruitment in theGulf countries, popularly called ‘visa-trading’ (Shah 2008, De Bel-Air 2011). In the visa-tradingsystem, recruitment agents and key members of migrant networks at the destination are requiredto pay the Gulf employers/sponsors in order to procure work visas for prospective migrantworkers. The financial transaction involved in the visa trading has transformed the modalityand scale of transactions for intermediaries of both migrant networks and recruiting agenciesalike (Shah 2008, De Bel-Air 2011). Forced by market pressures, actors of migrant networksare now required to charge fees to the prospective migrant for Gulf ‘work visas,’ leading tothe commercialisation of network-assisted recruitment.

The other key intermediary of labour recruitment is private recruiting agencies involved inserving prospective migrants for a fee. Recruitment agencies act as private gatekeepers, managingthe modalities that condition access for individuals seeking overseas employment (Goss andLindquist 1995). A recruiting agency is a national-level contact point for a foreign recruitingagency or foreign employers. Recruiting agencies usually have control over information thatallows them to negotiate with prospective migrants and their potential employers. They introducethe prospective migrants to the foreign partners (employers and other recruiters). There has been aremarkable expansion of private recruitment agencies in the last few decades. In fact, there may beas many as several thousand recruiting agencies in each major source country in South Asia andSouth-east Asia (Martin 2006, Baruah 2006, Lian and Rahman 2006). Considering the role in therecruitment process, some scholars refer to networks as the “engines of migration” (Phillips andMassey 2000) and other scholar refers to recruiting agencies as the “other engines of migration”(Hernandez-Leon 2005, p. 2).

To explain the recent changes in recruitment practices, scholars present some mid-level con-cepts such as ‘migration institution,’ ‘migration industry’ and ‘merchants of labour’ (Goss andLindquist 1995, Kuptsch 2006, Castles and Miller 2008). For instance, Jon Goss and Bruce Lind-quist present the concept ‘migrant institution,’ which draws on Anthony Giddens’s structurationtheory (Goss and Lindquist 1995). The migrant institution is conceived as “a complex institutionconsisting of knowledgeable individuals and the agents of organisations (from migrant associ-ations to multinational corporations) and the other institutions from kinship to the state” (Gossand Lindquist 1995, p. 336). The argument is that the institution operates to facilitate overseasmigration because intermediaries profit from the price that potential migrants are willing to pay.

An International Labour Organization (ILO) study on migrant workers and facilitators ofmigration has developed the concept ‘merchants of labour’ to refer to public and privateagents who move workers over national borders for economic rewards (Kuptsch 2006). Thesemerchants of labour include relatives who finance a migrant’s trip and provide housing and

216 M.M. Rahman

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

arrange for a job abroad, as well as public employment services and private recruitment agencies.This group includes virtually all intermediaries who make a profit from migrant workers. Anotherconcept that has been in use for quite a while is that of ‘migration industry;’ it embraces a widerange of people who earn their livelihood by organising migratory movements (Castles and Miller2008, p. 201). Such individuals include travel agents, labour recruiters, brokers, interpreters,housing agents, immigration lawyers, human smugglers and all others who extract fees frompotential migrants for their services. One commonality between merchants of labour and actorsin the migration institution and industry is the service fees that they charge to the prospectivemigrants. These terms particularly connote the presence of a formal and commodified form oflabour recruitment. While I agree that labour recruitment has been formalised and commodifiedover time, I also suggest that informal elements and traditional actors are still crucial to the under-standing of labour recruitment in the Gulf.

The terms ‘recruitment fees’ and ‘economic costs of recruitment’ demand a conceptual clar-ification. Though both recruitment fees and economic costs of recruitment refer to the expensesthat migrants might incur during the recruitment process, they vary depending on the patterns ofpayment within the recruitment and migration process. Since male migrants are usually requiredto pay the expenses for recruitment out of their own pockets prior to arriving at the destinationcountry, the economic costs of recruitment usually refer to the expenses that male migrantsincur in the recruitment process. In some predominantly male migrant-sending countries suchas Bangladesh and Pakistan, the term ‘economic costs of recruitment’ is widely used. On the con-trary, female migrants in Asia often do not need to pay the expenses for recruitment out of theirown pockets prior to migration. This is due to the fact that sponsor-employers often pay forwomen’s migration – especially in domestic worker migration – and deduct the advancepayment from the monthly salary of female workers when they start working in the destinationcountry. In some predominantly female migrant-sending countries (such as the Philippines andIndonesia), the term ‘recruitment fees’ or ‘placement fees’ are usually used as a substitute forthe economic costs of migration (Jones and Pardthaisong 1998, Wee and Sim 2004, Rahmanand Lian 2009, Lindquist 2010, Wilcke 2011).

A key difference between the recruitment of men and women in the Gulf is that men mustgenerally pay a fee to the recruitment agency prior to departure, while women usually do not(Rahman 2011). Recently, Johan Lindquist elaborates the gendered differentiated patterns ofrecruitment by showing that while capital flows “down” in the migration of women, for themigration of men capital flows “up,” from the migrant to the recruitment agency and sponsor(Lindquist 2010). South Asian male migrants usually enjoy the privileges of active expenses inthe recruitment process, by choosing service providers themselves and paying them upfront.This quality of choosing the service providers and paying the incurred expenses upfront justifiescalling the payment for recruitment ‘economic cost of recruitment’ rather than the seeminglybinding ‘recruitment fees.’ However, the critical issue in the recruitment of men from SouthAsia is precisely paying the economic costs of recruitment. This study is advanced in this econ-omic context of recruitment.

Data sources

This research is based principally on the Bangladesh household remittance survey (BHRS) con-ducted in 2009. The BHRS collected information from a nationally representative sample of10,926 migrant households. The BHRS was implemented by the International Organization forMigration (IOM) in Dhaka with financial support from the UK’s Department for InternationalDevelopment (DFID). Unlike the small-scale surveys (e.g. small migrant worker surveys orvillage case studies) that limit the scope of research and delay the generalisability of the study,

Bangladeshi labour migration to the Gulf states 217

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

the strength of the representative survey lies in its ability to generalise the study and to elucidatethe findings with confidence. In addition to the survey data, the study also draws from qualitativeinterviews of prospective migrants, returnees and members of migrant households in Bangladeshand of migrant workers and migrant brokers in the UAE and Qatar conducted between 2008and 2009.

The BHRS covered migrant households across the country through a nationally representativesample from all seven administrative divisions of Bangladesh.2 The districts of the seven div-isions of the country were divided into two strata, with one stratum consisting of ‘more concen-tration of migrant’ (MCM) households and the other stratum consisting of ‘less concentration ofmigrant’ (LCM) households. Following this, clusters were formed with one or more mauzas (nearsynonymous with a village), depending on the cluster’s size as set in terms of number of generalhouseholds. These clusters were selected independently from each stratum using the probabilityproportional to size (PPS) method of selection. The total sample was made up of 457 clusters(i.e. 257 from MCM districts and 200 from LCM districts). All households in every selectedcluster were listed, identifying only the migrant households. Household listings were done bytaking a complete census of the households in each of the clusters. This involved visitingevery household in the designated area.

In the survey, a migrant household was defined as a household that had at least one of itsmembers living or working abroad for at least one year at the time of the survey. The selectionof the migrant household was made independently of their current status (e.g. regular or irregular)in the country of destination. The head or other responsible member of the household was inter-viewed in the survey. The survey covered a wide range of migration issues encompassing avariety of factors: the socio-economic background of the migrants; the process and the economiccost of migration; the working and living conditions migrants faced; the remittances sent to themigrants’ home countries; and, finally, the impact of remittances on the migrant households.The total number of migrants in the survey was 12,893 and approximately three-quarters ofthe surveyed migrants (n ¼ 9,292) were working in the GCC countries. Of these 9,292 migrants,9,154 (98.52%) were male and 138 (1.48%) were female. Given the nature of survey and thenumber of cases, the study is expected to produce much needed insight into the recruitmentexperience of Bangladeshi migrants in the Gulf countries.

Bangladeshi labour migration to the GCC countries

The 1973 oil boom and the subsequent undertaking of an unprecedented number of developmentprojects led to an extremely rapid increase in the demand for foreign labour in the Gulf region(Arnold and Shah 1986, Birks et al. 1988, Winckler 2010). Among Arab countries, there was con-siderable labour mobility in the pre-1973 era (Humphrey 1991, IOM 2010). However, themigration of Asian labour in the post-1973 era marked a shift from predominance of Arablabour migration to Asian labour migration in the Gulf states (Arnold and Shah 1986, Humphrey1991, Winckler 2010). Since 1975, Asian workers have gained an increasing share of the migrantlabour market, and between 1976 and 1981 the annual labour migration flows of Asian workersincreased sevenfold, from 146,000 to over a million (Humphrey 1991, p. 47). Asian labour-sending countries accounted for more than 63 per cent of the migrant worker stock in theGulf states in 1985 (Birks et al. 1988, p. 268). Currently, Asia is the major human resource-contributing region for the GCC states.

The GCC countries are the major destination for Bangladeshi migrants. To seize the opportu-nity for temporary employment in the GCC countries, Bangladesh launched a government agencycalled the Bureau of manpower, employment and training (BMET) in 1976. BMET keeps track ofmigrants who take clearance from this government agency before migrating overseas for work.

218 M.M. Rahman

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

In other words, BMET maintains authorised labour migration statistics. There is no official pro-cedure to record returning migrants to Bangladesh. As a result, the data provided below representauthorised migrants who left for work in the GCC countries. According to BMET, around 5.3million Bangladeshi migrants joined in the GCC states between 1976 and 2010.3 Nearly80,000 migrants went to the GCC countries for work between 1976 and 1980. The totalnumber of emigrants departed for work in the GCC countries reached nearly 1 millionbetween 1981 and 1990, around 1.9 million between 1991 and 2000, and approximately 2.4million between 2001 and 2010 (Figure 1). Almost half of the total migrants joined the labourmarket of Saudi Arabia. The GCC countries are the major source of remittances for Bangladesh.Nearly two-thirds of the US$11 billion remittances in 2010 came from the GCC countries (WorldBank 2010).

Socio-demographic profiles of the migrants in the GCC countries

Of the 9,292 Gulf migrants in the sample, 48 per cent of the migrants were working in SaudiArabia, 34 per cent in the UAE, 8 per cent in Kuwait, 5 per cent in Oman, 3 per cent in Qatarand 2 per cent in Bahrain. As mentioned in the previous section, Bangladeshi migration to theGulf countries is predominantly a male phenomenon: 98.5 per cent in the 2009 survey.Overall, approximately three-fourths of the migrants were in their 20s and 30s, the average agebeing 32 years. Also, around 63 per cent of migrants were married and their wives were livingin Bangladesh. Of the married migrants, 87 per cent had children. Wives left behind were primar-ily homemakers, taking care of children and in-law families. Remittances were often the onlysource of incomes for these migrant households.

Educational attainment of the surveyed migrants was low. Approximately 30 per cent ofmigrants completed only one to five years of schooling and 50 per cent finished six to tenyears of schooling. Of the remaining 20 per cent, half had no formal schooling, while theother half had finished higher secondary and other vocational training courses. In terms ofemployment before their migration to the Gulf countries, 81 per cent of the migrants wereworking in Bangladesh and were engaged in a wide range of occupations from agricultural, indus-trial labour to small business and artisans. After migration, major types of work these migrantswere doing in the GCC countries included construction, janitorial work, driving, factory work,the hospitality industry, general labour gardening and small businesses. The duration of stay is

Figure 1. Bangladeshi labour migration to the GCC states, 1976–2010.Source: Author’s compilation, based on data from BMET (2011a).

Bangladeshi labour migration to the Gulf states 219

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

considered an important indicator of economic success, as the longer the stay the more economi-cally beneficial it is for migrants. Approximately 34 per cent of migrants were living in the Gulfcountries for one to two years, 22 per cent for three to five years, and 23 per cent for six to tenyears. Remarkably, nearly 20 per cent of the surveyed migrants had been working abroad for aperiod of 11 to 21 years or more. On average, the duration of stay of migrants was 6.27 years.

The kafala system – labour recruitment in the Gulf

The Gulf countries have developed a sophisticated migration management system that shares twocommon policy stances of international migration: protectionism and the absence of any projectfor integrating immigrants into society (Fargues 2006, p. 18). Philippe Fargues suggests that theGulf migration policy is protectionist in two ways: by making the entry and stay of foreignworkers more difficult; and by giving priority to the employment of nationals – commonlyknown as Emiratisation, Saudisation, Qatarisation, Kuwaitisation, Bahrainisation and Omanisa-tion initiatives (Fargues 2006). While the former policy has achieved remarkable success, thelatter policy, indigenisation of labour force, has not fully achieved its goals, resulting in sustainedflow of migrant workers to the Gulf countries (Toledo 2006, Dito 2008).

Labour recruitment in the GCC countries is governed by the kafala system (Shah 1994b,Longva 1999, Colton 2010, Baldwin-Edwards 2011, De Bel-Air 2011). In the kafala system, amigrant is sponsored by an employer of GCC citizens who assumes full economic and legalresponsibility for the foreign employee during the contract period. The kafeel (sponsor-employer)may be an individual, a placement agency or a company/institution. The kafeel signs a formwhereby he or she declares that the foreigner works for him or her, undertakes to inform the immi-gration department of any change in the labour contract such as expiry, renewal or cancellation,and lastly pledges to repatriate the employee upon termination of the contract (for details, seeLongva 1999). The migrant worker thereby is tied to his or her kafeel. The system works suchthat the migrant worker can only work for him or her for a specific period.

Recently, Francoise De Bel-Air offers a two-fold classification of kafeels: small or occasionalkafeels; and big kafeels (De Bel-Air 2011, p. 3). ‘Small’ kafeels operate within small-scale,person-to-person frameworks such as employment of domestic workers. However, ‘big’kafeels usually set up national-level networks and possess adequate means and resources tolobby for ‘work visas.’ Many big kafeels are believed to be involved in visa trading. In visatrading, the kafeel sells the work visa to the prospective migrant through intermediary for financialbenefit. Since the visa trading generates kickback fees for kafeels, it has been a multi-milliondollar industry in the Gulf (Shah 2008). In the UAE, for instance, a work visa for an Indian issold for around US$2,000 (or AED7,500) (Shah 2008, p. 9) and for a Pakistani aroundUS$2,700 to 3,000 (AED10,000 to 12,000) (Gulf News 2004). The scale of visa trading ismassive. For instance, the Saudi Minister of Labour reported in the Arab News–Saudi Arabiathat 70 per cent of the visas issued by the government are sold on the black market and the gov-ernment has been determined to crack down on this particular trade (Shah 2008, p. 9).

Kafeels often hold passport and other travel documents and sometimes exploit migrants bydenying proper wages and conditions of employment to them (Colton 2010, Gardner 2010). Inthe kafala system, the change of kafeel is not impossible but entails hurdling bureaucratic pro-cedures. Once the employment relationship is broken, foreign workers become illegal residents.Foreign workers are not allowed to marry or be involved in sexual relationships with locals; dom-estic workers’ bodies are put under medical surveillance to detect sexual activity. The kafalasystem has been met with much criticism over the years. For instance, it is accused of encouragingcorruption, visa trafficking, as well as the import of workers that widely outpaces labour marketneeds (De Bel-Air 2011, p. 10). The system ensures additional lucrative incomes for local

220 M.M. Rahman

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

sponsors and, despite some genuine interests, governments in the Gulf countries have failed tobring about meaningful changes in the visa-trading system largely because of pressure frombusiness elites and clan members (Shah 2008, De Bel-Air 2011).

It is worth mentioning that the sponsorship system is not unique to the GCC countries. Thisapproach to temporary migration management regarding the purpose and the control mechanismsemployed can also be found in East Asia and South-east Asia (Chan and Abdullah 1999, Ruhs2002, Piper 2004). Temporary migration programs are, in essence, a demand-driven systemthat serves the interests of both parties – employers and migrants (Rahman 2008). On the onehand, it is efficient in satisfying the needs of employers and is therefore capable of generatinghigher benefits for the host country. On the other hand, it is economically beneficial for themigrants and their home countries because it spawns remittances that directly go to themigrant families who genuinely require these funds. However, the kafala system has beenmisused to create financial gain from the potential migrants through the help of intermediariesin the recruitment process.

The operational aspect of recruitment

Recruitment structure

In the recruitment process, prospective migrants in Bangladesh use the services of four public andprivate institutions – the BMET (Bureau of Manpower, Employment and Training), BangladeshOverseas Employment Services Limited (BOESL), recruiting agencies and migrant networks –and together they form a semi-coherent system of governance that facilitates smooth flows oflabour overseas. The Bangladeshi government’s most prominent structure regulating migrationis the BMET, established in 1976. The BMET issues and renews the licenses of recruitingagencies, grants permission to agencies to recruit, provides immigration clearances after verifyingvisa papers and employment contracts, looks after the welfare of Bangladeshi workers abroad andmanages many other functions related to training of workers and promotion of migration over-seas. Presently, the BMET is under the administrative control of the newly established Ministryof Expatriates’ Welfare and Overseas Employment.

While the BMET supervises and controls recruitment, actual recruitment takes place throughthree other channels: a government-run organisation called BOESL; private recruiting agencies;and migrant networks. The government-run organisation for overseas employment established in1984, BOESL is mainly involved in migration of highly skilled professionals overseas. The roleof BOESL in skilled and professional migration is not necessarily significant. Out of 179,910professional migrants who took up overseas placement between 1976 and 2008, only 14,811 pro-fessional migrants found jobs through the BOESL (BMET 2011b). The agency coordinates withBangladeshi missions abroad in assessing the needs of labour and puts up advertisements in news-papers for recruitment. BOESL is seen as a model institution in the manpower sector to work inhealthy competition with the private agencies.

Private recruiting agencies emerged in response to the growing demand for labour overseas inthe late 1970s. There are around 700 licensed private recruiting agents in Bangladesh (BAIRA2011). The government regulates recruiting agencies through a combination of economic leverageand bureaucratic requirements. Recruiting agencies assisted nearly 37 per cent (2.33 million) of atotal of 6.26 million overseas job placements between 1976 and 2008 (BMET 2011b). Recruitingagencies have a guild known as the Bangladesh Association International Recruiting Agencies(BAIRA), formed in 1984 with the purpose of catering to the needs of the licensed recruitingagencies. BAIRA is also involved in the welfare of the migrants overseas. They have launchedtwo insurance schemes: one is for the workers before their departure and the other is for theirfamilies left behind, through BAIRA Life Insurance Company Limited. BAIRA is also setting

Bangladeshi labour migration to the Gulf states 221

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

up a modern medical testing center with elaborate facilities for migrants who need to undergomedical tests before leaving the country.

Personal connections form a crucial avenue for overseas placement in Bangladesh. Thesereportedly assisted 62 per cent (3.88 million) of a total of 6.26 million overseas job placementsbetween 1976 and 2008 (BMET 2011b). The growth of migration through personal connectionsis cumulative. More and more migrants have started taking the services of personal networks overthe years. For instance, 524 migrants took the services of personal networks in the year 1976,which escalated to 16,585 in the year 1980, 63,121 in the year 1990, 130,686 in the year2000, and 665,092 in the year 2008. As per the survey data, nearly 58 per cent of surveyedmigrants (9,292) took the services of personal networks to migrate to the GCC countries. Recruit-ing agencies served the remaining 42 per cent.

Recruiting agencies: the work visa

Migration for work though a recruiting agency proceeds as follows (see Figure 2). A recruitingagency in a Gulf country places a ‘demand letter’ with their counterpart in Bangladesh, askingfor certain number of migrants for certain occupations. The recruiting agency in Bangladeshthen approaches BMET for initial clearance. Once the agency gains approval, they activelysearch for prospective migrants and ask them to submit passport, pictures, biographical infor-mation, work experience (including relevant certificates) and a partial payment to begin therecruitment process. At this juncture, the Bangladeshi recruiting agency contacts their counterpartin the Gulf for visa processing. The potential sponsor-employer will then secure the visa from therelevant authority and pass the visa to the recruiting agency in the receiving country. This foreignrecruiting agency then sends the visa to the recruiting agency in Bangladesh. From the time thepotential migrants give their passport to the agency to the time they can commence work abroad,as little as a few weeks or as much as several months may pass.

Although an overview of how recruiting agencies work may seem simple, the actual recruit-ment procedures are much more complex and multilayered. Along with formal recruitingagents, some informal agents located at different points throughout the system have come toassist in the process. Generally, most prospective migrants hail from villages in Bangladesh,but recruiting agencies based in the capital city of Dhaka encounter difficulties locating prospec-tive migrants who might be suitable for particular jobs and who are ready to pay the required fees.As a result, the recruiting agents rely overwhelmingly on a group called subagents who act asmiddlemen between a prospective migrant and a licensed recruiting agent. For an extra fee,these subagents help prospective migrants find jobs and help agencies find prospective workersin a more timely fashion. Although the position of subagent in the official structure is marginal,they play a critical role in matching the demand for specific labour and the supply of such labour.

The subagents approach the prospective migrants and convince them to take up the particularoffer in the GCC countries. The tru,e challenge of this job is the subagents’ need to earn the trustof prospective migrants. Given the fact that migration can incur costs of several thousands ofdollars, a prospective migrant from a village prefers to deal with someone who embodies trustsuch as rural religious, economic and political elites, so that they have a local contact to approachin cases of fraud and exploitation. Subagents are generally based in small cities or villages andhave good contacts with these elites, if they themselves are not already one. Subagents assist pro-spective migrants with a wide range of activities such as paperwork, passports, bank accounts,medical checkups and transportation to the airport. In addition to facilitating the actual migrationprocess, they sometimes vouch to the traditional moneylenders that potential clients have alreadysecured jobs in the Gulf and therefore are eligible for credit. They can even act as guarantor for

222 M.M. Rahman

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

some potential migrants who otherwise could not secure loans for migration, expanding the roleof subagents beyond the simple matching task.

Personal connections: the ‘urro’ or ‘flying’ (work) visa

As migration matures, many potential migrants learn more about the process of migration and theway to seek alternative services for jobs in the GCC countries. Instead of finding jobs withagencies, many potential migrants seek jobs arranged with sponsor-employers in the Gulfcountries through some migrant brokers already connected by social and symbolic ties.Migrant brokers, or what Bangladeshi migrants call ‘dalal’ (broker) or ‘adam babshahi’(human-trader), are usually former migrant workers living in the Gulf countries for a considerableamount of time and having greater access to local recruitment structure and visas.

A working visa arranged through personal networks is called ‘urro’ or a flying (work) visa, as itflies directly from a migrant broker in the Gulf countries to a prospective migrant in Bangladesh,bypassing local recruiting agencies and their subagents. This is different from getting a visathrough a recruiting agency. In the recruiting agencies, a sponsor-employer passes the visa to aforeign recruiting agency, which later goes to the Bangladeshi recruiting agency and finally tothe prospective migrant. Arranging a ‘flying visa’ usually proceeds as follows (see Figure 2). Amigrant broker finds a job for a friend or relative with usually his kafeel (sponsor-employer) orkafeel’s network of friends and relatives. After successfully identifying a potential kafeel, amigrant broker arranges for a photocopy of a passport, a picture and other relevant papers dependingon the profile of employee (skilled or unskilled) to pass to potential kafeel for a work visa. Once awork visa is procured, the migrant broker sends it to the potential migrant in Bangladesh with arequest to finish all local procedures such as medical check-ups, BMET clearance, and fly to theparticular Gulf country within a stipulated period (usually less than three months). The flyingvisa benefits the mediating contact, as he charges a commission to the potential receiver, and thekafeel, as he saves agency fees in his country of origin, in addition to kickback fees in most cases.

More importantly, the strategy of procuring a working visa through personal connectionsdemonstrates the individual migrant’s ability to circumvent bureaucratic procedures and the

Figure 2. Recruitment structure for Bangladeshi labour migration to the GCC states.

Bangladeshi labour migration to the Gulf states 223

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

rising costs of Gulf migration. Since the procurement of a ‘flying visa’ does not involve localmediators in Bangladesh, in particular the subagents and agents, the ‘flying visa’ benefits thepotential migrant who receives it. Depending on the nature of ties (social or symbolic ties)between prospective migrant and migrant broker, the amount of the fees and the timing ofpayment can vary considerably. The direct benefit of a ‘flying visa’ is that the recipientsusually pay part of the set fees out of the first several monthly wages in the destinationcountry, forestalling indebtedness to moneylenders in Bangladesh. Although the ever-increasingcosts of recruitment have made it difficult for many poor potential migrants to secure jobs in theGulf countries, a thriving trade in ‘flying visas’ has opened the door of affordable employment fora growing number of Bangladeshis in the GCC countries.

It is important to note that the Bangladeshi migration to the Gulf countries is deeply rooted inthe social process organised through networks and forged through everyday interpersonal connec-tions (Rahman 2010). It is founded on a social organisation infrastructure that includes commonbonds of kinship and friendship, which are adopted and transformed through the reciprocity ofmass migration. A landless villager in Bangladesh may be poor in financial capital but is oftenrich in social resources (social capital) inherent in such close-knit connections. The beauty ofsocial capital is that it is convertible into other forms of capital such as financial capital oraccess to overseas employment. The actual merit of personal connections is that it increasesaffordability of Gulf migration for those in the social stratum, who may not otherwise envisionundertaking such a costly venture.

From ‘flying visa’ to ‘free visa’

Apart from ‘flying visas,’ there is another type of visa popular among potential Bangladeshimigrants in Gulf countries called a ‘free visa.’ This type of visa is an invention of migrantbrokers in the Gulf countries who are involved in the trade of work visas. Although there is no offi-cial category of visas called ‘free visas’ in the Gulf countries, the term is widely used amongmigrant communities. Pakistanis and some Indian Muslim migrants call it an ‘azad’ (free) visa.This unofficial category of visas allows a potential migrant to enter a GCC country for workunder kafala system, but the sponsor-employer (kafeel) who officially sponsors the migrant doesnot offer paid work. A migrant on a ‘free visa’ is free to find his own job in any sector ofeconomy, although it is illegal to work in another sector or with a sponsor-employer other thanone’s own (Shah 2008, p. 7). In other words, a ‘free visa’ is legal, but paradoxically when afree-visa holder starts working for others, he becomes illegal by law and vulnerable to deportation.

A ‘free visa’ is also obtained through procedures similar to those described above using per-sonal networks and recruiting agencies (Figure 2). Both migrant brokers and recruiting agents areinvolved in the trade of ‘free visas’ in Bangladesh. As there are no legal jobs available for a ‘free-visa’ holder, a ‘free visa’ is cheaper than ‘flying visa.’ Many migrant brokers do not recommend a‘free visa’ to their close relatives for the fear that they may lose social credibility and respect intheir communities of origin should the ‘free-visa’ holder fail to find gainful employment and facedeportation. As a ‘for profit’ organisation, recruitment agencies dominate the ‘free visa’ market.Like many ‘work visas’ and ‘flying visas,’ ‘free visas’ also generate kickback fees for sponsors/kafeels and this has perhaps given birth to such a dubious category of visas in the Gulf. The benefitof a ‘free visa’ is that the holder can bargain their labour and get a good deal. However, the trade-off is that if the free visa holder is caught during working hours by the law enforcementauthorities, then the holder risks deportation. Several Gulf countries have pledged to tacklethe issue of ‘free visas’ and have taken action against people who have abused the system bybringing in thousands of foreigners and enabling them to stay for a monthly monetary sum (Pak-kiasamy 2004).

224 M.M. Rahman

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

The economic aspect of recruitment

In the case of Asian migration to the Gulf countries, wages have fallen and agency recruitmentcosts have risen (Gardner 2010, p. 60). To explain this situation in the Gulf countries,Gamburd (2000, p. 63) has identified changing international economic dynamics as the factorsleading to increasing agency fees and decreasing wages. The benefits of Gulf migration dwindledin the 1980s when employer-sponsors lowered commissions paid to recruiting agencies in theGulf countries (Gamburd 2000, Gardner 2010). With an apparently endless supply of surpluslabour in source countries in several regions, agencies of these countries can often afford tobargain away workers’ benefits in order to secure jobs and to stay in business. Agencies insending countries easily pass the burden of higher fees to potential migrants. Moreover, mostsponsor-employers also refuse to pay for airfares, passports and medical examinations for pro-spective migrants. Therefore, in addition to recruitment fees, most Asian migrants are alsobearing these expenses for migration to the Gulf.

According to the data gathered in the BHMS, the economic cost of recruitment to the GCCstates was on average BDT195,237 or US$2,750 (2010 exchange rate) (Figure 3). The highestaverage cost was BDT221,683 or US$3,132 for Bahrain and the lowest was BDT174,021 orUS$2,458 for Kuwait. This average cost of migration is higher than the cost set by the Banglade-shi government. The government establishes a price of BDT84,000 (US$1,230) as a maximumrecruiting charge for migrants going to Gulf states (Martin 2010, p. 12). Given the reportedaverage costs of migration, it is obvious that recruiting agencies largely ignore this maximum.In recognition of the high cost of migration, this study delves into distribution of recruitmentexpenses. For the convenience of understanding, the study divides the economic cost of recruit-ment into formal fees and intermediary fees. Formal fees include government fees, passport fees,air tickets, wage earners’ contributions and other relevant mandatory service fees in Bangladesh.Intermediary fees include both recruiting agency and migrant broker fees and fees charged byother facilitators such as subagents and network members. Formal fees represent around 23 percent (US$633), while the intermediary fees nearly 77 per cent (US$2,117).

The fee for intermediaries is excessively high as a kickback is inherent in the intermediaryfees. Some Bangladeshi recruiting agents and migrant brokers reported the kickback feebetween US$1,500 and US$1,800 for each work visa. However, the kickback fee varies depend-ing on the sector of economy. The kickback fees for work in construction sector are reported to bethe lowest as it is more demanding and less rewarding, while the fees for the service sector are thehighest as the sector is viewed as less demanding but more rewarding. The kickback fee, inter-mediary fee and subagent fee are charged unofficially, leaving no trace for documentary proof

Figure 3. Distribution of economic cost of recruitment.

Bangladeshi labour migration to the Gulf states 225

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

or legal recourse. When the survey asked about the rising cost of migration, actors at differentlevels often turned their discussion of unethical behavior toward actors elsewhere in the insti-tutional structure. The different perspectives reveal conflicts and contradictions within therecruitment structure in the Bangladesh–Gulf migration corridor.

Conclusion

This study has shown how recruitment of labour takes place across international borders bylinking and explaining both the operational and the economic aspects in the example of Bangla-deshi labour recruitment to the Gulf states. The study illustrates the role of both institutional andnetwork players in the recruitment process. It continues to elaborate the kafala system that formsthe legal basis for labour recruitment in the GCC countries. The policies of entries and exits formigrant labour are mostly dictated by economic imperatives and informed by the principle ofminimal social disruptions. This study has also explained how the recruitment structure ofBangladesh has evolved over time to facilitate the recruitment of millions of migrants to theGulf countries. Four major players are identified: the BMET; BOESL; recruiting agencies; andmigrant networks. These form a semi-coherent system of governance and facilitate Bangladeshimigration to the Gulf countries.

The study has shown that recruiting agencies and migrant networks are currently the actualplayers in labour recruitment in Bangladesh. The recruitment agencies and migrant brokershave different yet complementary roles in the overall recruitment process. Recruiting agencieswork in collaboration with their counterparts in the Gulf countries and use subagents to reachto the potential migrants. They are formal profit-making organisations serving prospectivemigrants who seek jobs in the Gulf labour markets. In network-based recruitment, a migrantbroker in a Gulf country secures a work visa, known as a ‘flying visa,’ for prospective migrantsin Bangladesh. The prospective migrants and migrant brokers are often connected by social andsymbolic ties. The migrant broker charges fees for services, but a full payment is not alwaysrequired depending on the level and strength of ties; the prospective migrant may pay off apart of the set fee after working in the destination countries for the first few months.

The study has documented the ingenuity of migrant brokers in circumventing the convention-al route to labour recruitment. Migrant brokers have invented other types of visas, the ‘urro visa’(flying visa) and the ‘free visa’ to meet the increasing requests from relatives and communitymembers in Bangladesh. When a visa is procured through personal networks, bypassing localrecruiting agencies and their subagents in Bangladesh, it is popularly called an urro visa. Sincean urro visa does not require local mediators’ services, it is relatively more affordable for potentialmigrants. The invention of the urro visa demonstrates the capability of migrant agency to circum-vent bureaucratic procedures and rising costs of Gulf migration. With the other type of visa, calleda ‘free visa’ in Bangladesh and an ‘azad visa’ in Pakistan and India, the kafeel is not liable foroffering paid jobs to the visa holder; the holder is free to find his own job and bargain hislabour costs. Thus, the ‘free visa’ represents the ultimate will of migrants to find their jobs them-selves and exercise bargaining power over their labour by being independent of kafeel’s everydayinterventions in the Gulf.

This study has shown how economic transactions in the recruitment process generate profit forservice providers such as recruiting agents, subagents and migrant brokers in Bangladesh and theGulf. The economic cost of migration for migrants is quantified, as are the financial benefitsmigration generates for different service providers in the recruitment process. Of an averagecost of US$2,750, nearly 23 per cent of the recruitment cost is spent on government fees,agency fees, passport fees, airfares and other related expenses, while the remaining 77 per centgoes to intermediaries such as recruitment agents, migrant brokers, subagents and other

226 M.M. Rahman

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

facilitators. The study has pointed out that intermediary fees also contain kickbacks for theGulf-based recruiting agencies.

While this study has attempted to provide valuable insights into the recruitment of Banglade-shi migrants in the GCC countries, it could not shed light on some other aspects of recruitmentsuch as political economy of recruitment at both ends. This study should be seen as an initialinvestigation of an understudied subject, and building on it can further the analysis of migrantrecruitment in the Gulf.

AcknowledgementsThe author wishes to thank International Organisation for Migration (IOM), Dhaka, for providing access tothe Bangladesh Household Remittance Survey data. Special thanks go to Rabab Fatima, Regional Represen-tative, IOM-Dhaka, Samiha Huda and Disha Sonata Faruque, also of IOM-Dhaka, who all provided supportand were a joy to work with. The author would like to thank two anonymous reviewers of the CanadianJournal of Development Studies for their insightful comments on the draft version of this article. Theauthor also wishes to thank Nasra Shah and Philippe Fargues for their comments at the presentation ofthis paper at the Gulf Research Meeting, University of Cambridge, 6–9 July 2011.

Biographical noteMd Mizanur Rahman is a Research Fellow at the Institute of South Asia Studies, National University ofSingapore. His research interests include labour migration, diaspora, migrant entrepreneurship andremittances. He is currently working on global South Asian diasporas and their engagements with SouthAsia. He has published in International Migration, Journal of Ethnic and Migration Studies, Asian Popu-lation Studies, Journal of International Migration and Integration and Population, Space and Place,among others.

Notes1. Bangladesh is divided into major administrative regions called divisions and each division is further

split into districts. In total, there are 7 divisions and 64 districts in Bangladesh.2. The Bureau of Manpower, Employment and Training, Bangladesh (BMET) is responsible for record-

keeping and granting permission to overseas jobseekers. The BMET posts vital migration statisticsonline at http://www.bmet.org.bd.

ReferencesArnold, F. and Shah, N.M., eds., 1986. Asian labor migration: pipeline to the Middle East. London: West

View.Baldwin-Edwards, M., 2011. Labor immigration and labor markets in the GCC countries: national patterns

and trends, [online]. London, London School of Economics, Global Governance. Available from: http://www2.lse.ac.uk/government/research/resgroups/kuwait/documents/Baldwin-Edwards,%20Martin.pdf[Accessed 27 February 2012].

Bangladeshi Association of International Recruiting Agencies (BAIRA), 2011. Bangladeshi association ofinternational recruiting agencies [online]. Available from: http://www.hrexport-baira.org/index2.htm[Accessed 12 August 2011].

Baruah, N., 2006. The regulation of recruitment agencies: experience and good practices in countries oforigin in Asia. In: C. Kuptsch, ed. Merchants of labour. Geneva: International Organization forMigration, 37–46.

Birks, J.S., Seccombe, I.J., and Sinclair, C.A., 1988. Labour migration in the Arab Gulf states: patterns,trends and prospects. International migration, 26 (3), 267–286.

Boyd, M., 1989. Family and personal networks in international migration: recent developments and newagendas. International migration review, 23 (3), 638–670.

Bureau of Manpower, Employment and Training (BMET), 2011a. Online statistics [online]. Bangladesh,BMET. Available from: http://www.bmet.org.bd/BMET/viewStatReport.action?reportnumber=9[Accessed 23 June 2011].

Bangladeshi labour migration to the Gulf states 227

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

Bureau of Manpower, Employment and Training (BMET), 2011b. Overseas statistics [online]. Bangladesh,BMET. Available from: http://www.bmet.org.bd/Reports/Overseas_Statistics.htm [Accessed 24 July2011].

Castles, S. and Miller, M.J., 2008. The age of migration: international population movements in the modernworld. 4th ed. New York: Guilford Press.

Chan, R.K.H. and Abdullah, M.A., 1999. Foreign labor in Asia: issues and challenges. Commack, NY:Nova Science Publishers.

Colton, N.A., 2010. The international political economy of Gulf migration, migration and the Gulf.Washington, DC: Middle East Institute.

De Bel-Air, F., 2011. Reforming the sponsorship (kafala) in the GCC countries: what socio-political stakes?The case of Saudi Arabia, The 2011 Gulf research meeting: migration in the Gulf, 6–9 July, Cambridge.Cambridge: Gulf Research Centre Cambridge.

Dito, M.E., 2008. GCC labour migration governance, [online]. Bangkok, United Nations Economic andSocial Commission for Asia and the Pacific. Available from: http://www.un.org/esa/population/meetings/EGM_Ittmig_Asia/P07_Dito.pdf [Accessed 26 March 2012].

Eelens, F., Schampers, T., and Speckmann, J.D., 1991. Labor migration to the Middle East: from Sri Lankato the Gulf. London: Kegan Paul.

Eelens, F. and Speckmann, J.D., 1990. Recruitment of labor migrants for the Middle East: the Sri Lankancase. International migration review, 24 (2), 297–322.

Esim, S. and Smith, M., 2004. Gender and migration in Arab states: the case of domestic workers. Geneva:International Labour Organisation.

Faist, T., 2000. The volume and dynamics of international migration and transnational social space. Oxford:Clarendon Press.

Fargues, P., 2006. International migration in the Arab region: trends and policies [online]. Beirut, UnitedNations Secretariat, United Nations Expert Group Meeting On International Migration andDevelopment in the Arab Region. Available from: http://www.un.org/esa/population/migration/turin/Symposium_Turin_files/P09_Fargues.pdf [Accessed 26 March 2012].

Gamburd, M.R., 2000. Transnationalism and Sri Lanka’s migrant households: the kitchen spoon’s handle.New Delhi: Vistaar Publications.

Gardner, A.M., 2010. City of strangers: Gulf migration and the Indian community in Bahrain. New York:Cornell University Press.

Goss, J. and Lindquist, B., 1995. Conceptualizing international labor migration: a structuration perspective.International migration review, 29 (2), 317–351.

Gulf News, 2004. Illegal visa trade still flourishes [online]. Gulf News, 24 April. Available from: http://gulfnews.com/news/gulf/uae/visa/illegal-visa-trade-still-flourishes-1.319255 [Accessed 26 March2012].

Gurak, D.T. and Caces, F., 1992. Migration networks and the shaping of migration system. In: M. Kritz, L.L.Lim and H. Zlotnick, eds. International migration systems: a global approach. Oxford: Clarendon Press,150–176.

Hernandez-Leon, R., 2005. The migration industry in the Mexico–US migratory system [online]. LosAngeles, California Center for Population Research, University of California. Available from: http://escholarship.org/uc/item/3hg44330 [Accessed 26 March 2012].

Humphrey, M., 1991. The changing role of Asian labour migration in the Middle East. Des migrations inter-nationales, 7 (1), 45–63.

International Organization for Migration (IOM), 2004. Arab migration in a globalized world. Geneva: IOM.International Organization for Migration (IOM), 2010. Intra-regional labour mobility in the Arab World.

Geneva: IOM.Jones, H. and Pardthaisong, T., 1998. The commodification of international migration: findings from

Thailand. Tijdschrift voor economische en sociale geografie, 90 (1), 32–46.Kapiszewski, A., 2006. Arab versus Asian migrant workers in the GCC countries [online]. Beirut, United

Nations Secretariat, Department of Economic and Social Affairs, Population Division. Availablefrom: http://www.un.org/esa/population/meetings/EGM_Ittmig_Arab/P02_Kapiszewski.pdf [Accessed26 March 2012].

Kuptsch, C., ed., 2006. Merchants of labour. Geneva: International Organization for Migration.Lian, K.F. and Rahman, M.M., 2006. International labor recruitment: channeling Bangladeshi migrants to

East and South-East Asia. Asia-Pacific population journal, 21 (1), 85–107.Lindquist, J., 2010. Labour recruitment, circuits of capital and gendered mobility: reconceptualising the

Indonesian migration industry. Pacific affairs, 83 (1), 115–127.

228 M.M. Rahman

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

Longva, A.N., 1999. Keeping migrant workers in check: the Kafala system in the Gulf. Middle East report,29 (2), 20–22.

MacDonald, J.S. and MacDonald, L.D., 1974. Chain migration, ethnic neighborhood formation, and socialnetworks. In: Tilly Charles, ed. An urban world. Boston, MA: Little, Brown & Co, 226–236.

Martin, P., 2005. Merchants of labor: agents of the evolving migration infrastructure [online]. Geneva,International Institute for Labour Studies. Available from: http://www.ilo.org/public/english/bureau/inst/publications/discussion/dp15805.pdf [Accessed 26 March 2012].

Martin, P., 2006. Regulating private recruiters: the core issues. In: C. Kuptsch, ed. Merchants of labour.Geneva: International Organization for Migration, 13–25.

Martin, P., 2010. The future of labour migration cost [online]. Geneva, International Organization forMigration. Available from: http://publications.iom.int/bookstore/free/WMR2010_labour_migration_costs.pdf [Accessed 26 March 2012].

Massey, D.S., et al., 1987. Return to Aztlan: the social process of international migration from WesternMexico. Berkeley, CA: University of California Press.

Pakkiasamy, D., 2004. Saudi Arabia’s plan for changing its workforce [online]. Washington, DC, MigrationPolicy Institute. Available from: http://www.migrationinformation.org/USfocus/display.cfmID=264[Accessed 26 March 2012].

Piper, N., 2004. Rights of foreign workers and the politics of migration in South-East and East Asia.International migration, 42 (5), 71–95.

Phillips, J. and Massey, D.S., 2000. Engines of immigration: stocks of human and social capital in Mexico.Social science quarterly, 81 (1), 33–48.

Ruhs, M., 2002. Temporary foreign worker programmes: policies, adverse consequences, and the need tomake them work [online]. Geneva, International Labour Office, International Migration Programme,Social Protection Sector. Available from: http://www.ilo.org/public/english/protection/migrant/download/pom/pom6e.pdf [Accessed 26 March 2012].

Rahman, M.M., 2008. Management of foreign manpower in Singapore. In: L.K. Fee and T.C. Kiong, eds.Social policy in post-industrial Singapore. Leiden: Brill Academic Publishers, 145–172.

Rahman, M.M., 2009. In quest of golden deer: Bangladeshi transient migrants overseas. Saarbrucken: VDMVerlag.

Rahman, M.M., 2010. Remittance as a social process: the Singapore–Bangladesh remittance corridor. Asiaand Pacific migration journal, 19 (2), 265–294.

Rahman, M.M., 2011. Bangladeshi migrant workers in the UAE: gender-differentiated patterns of migrationexperiences. Middle Eastern studies, 47 (2), 395–412.

Rahman, M.M. and Lian, K.F., 2009. Gender and the remittance process: Indonesian domestic workers inHong Kong, Singapore, Malaysia. Asian population studies, 5 (2), 103–127.

Shah, M.N., 1994a. An overview of present and future emigration dynamics in South Asia. Internationalmigration, 32 (2), 217–268.

Shah, M.N., 1994b. Arab labour migration: a review of trends and issues. International migration, 32 (1),3–27.

Shah, M.N., 1996. The role of social networks in migration to Kuwait among South Asian males, IOM/UNFPA workshop on emigration dynamics in developing countries, 2–3 September, Geneva.

Shah, M.N., 2008. Recent labour immigration policies in the oil-rich Gulf: how effective are they likely tobe? [online]. Geneva. International Labour Organization, ILO Asian Regional Programme onGovernance of Labour Migration. Available from: http://www.ilo.org/wcmsp5/groups/public/—asia/—ro-bangkok/documents/publication/wcms_099165.pdf [Accessed 26 March 2012].

Shah, M.N., 2010. Building state capacities for managing contract worker mobility: the Asia-GCC context.Background paper for World Migration Report 2010 [online]. Geneva, International Organization forMigration. Available from: http://publications.iom.int/bookstore/free/WMR2010_contract_worker_asiagcc.pdf [Accessed 26 March 2012].

Shah, M.N. and Menon, I., 1999. Chain migration through the social network: experience of labour migrantsin Kuwait. International migration, 37 (2), 361–382.

Suter, B., 2005. Labour migration in the United Arab Emirates: field study on regular and irregularmigration in Dubai [online]. Malmo, Malmo University, Malmo Institute for Studies of Migration,Diversity and Welfare (MIM), Department of International Migration and Ethnic Relations (IMER).Available from: http://dspace.mah.se:8080/bitstream/handle/2043/3161/Labour%20Migration%20in%20the%20United%20Arab%20Emirates%20-%20A%20Field%20Study%20on%20Regular%20and%20Irregular%20Migration%20in%20Dubai.pdf;jsessionid=F9A57CCC416481D1284233FDC82C30C1sequence=1 [Accessed 26 March 2012].

Bangladeshi labour migration to the Gulf states 229

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012

Toledo, H., 2006. The problems and prospects of Emiratization: immigration in an imperfect labor market.Sharjah: The American University of Sharjah.

Ullah, A.K.M.A., 2010. Rationalizing migration decisions: labor migrants in east and Southeast Asia.London: Ashgate.

Wee, V. and Sim, A., 2004. Transnational networks in female labour migration. In: A. Ananta and E.N.Arifin, eds. International migration in South-East Asia. Singapore: Institute of South-East AsianStudies, 166–198.

Wilcke, C., 2011. Domestic plight: how Jordanian laws, officials, employers and recruiters fail abusedmigrant domestic workers. New York: Human Rights Watch.

Winckler, O., 2010. Labour migration to the GCC states: patterns, scale, and policies. In: Middle EastInstitute, ed. Migration and the Gulf. Washington, DC: Middle East Institute, 9–12.

World Bank, 2010. Migration and development brief [online]. Washington, DC, World Bank. Availablefrom: http://siteresources.worldbank.org/INTPROSPECTS/Resources/334934-1110315015165/MigrationAndDevelopmentBrief12.pdf [Accessed 26 March 2012].

Zachariah, K.C., Mathew, E.T., and Irudaya Rajan, S., 2001. Social, economic and demographic conse-quences of migration on Kerala. International migration, 39 (2), 43–71.

230 M.M. Rahman

Dow

nloa

ded

by [

NU

S N

atio

nal U

nive

rsity

of

Sing

apor

e] a

t 19:

45 2

6 Ju

ne 2

012