Building Inequality: Wage Disparity between Bangladeshi and Thai Guestworkers in Singapore’s Construction Industry

Theoretical underpinnings

This article strives to bring the incipient literature on migration and class into dialogue with that on the segmentation of guestworker labour markets. Van Hear (2014: 100–101; 117) observes that ‘class has lately been eclipsed in much of migration studies by consideration of other forms of social difference’ and to the extent that class is considered, it is mostly with ‘respect to migration outcomes in host or destination countries’ such as ‘social mobility in the second generation’. He argues that the application of the Marxist understanding of ‘“class in itself” . . . [as] a collectivity with certain identifiable attributes and interests in common’ can lead to a deeper understanding of migration as it informs ‘who is able to move and who cannot, and the hierarchy of destinations that different migrants are able to reach’ (Van Hear, 2014: 113; 117).

This article argues that additionally ‘class in itself’ informs migrants’ socialization in their origin countries, which may in turn inform their positions within the labour market of the host country. Class position in the origin country may not be determinant of migrants’ positions in the host society’s overall class hierarchy (both working-class Thais and middle-class Bangladeshis are lowly Singapore guestworkers); however, migrants’ class-specific socialization in their origin countries may contribute to their position in the host country labour market to the extent that it informs workplace performance. Post-war Britain offers a historical example of how origin-country class and caste socialization contributes to immigrants’ positions in the construction labour market. Sikhs, who underwent caste-based training in carpentry in Gujarat, became renown tradesmen in Britain (Wall, 2017), while the Irish, agricultural workers at home, became semi-skilled building workers reputed for their physical strength (Thiel, 2013).

Although the national origin-based segmentation of the Asian construction guestworkforce has attracted little academic scrutiny, the same is not true for guestworkers in agriculture or domestic service. In the former case, growers’ preferences for particular national groups is related to heightened structural vulnerability as evident in weak advocacy by the labour-sending state, inability to speak English and absence of co-ethnics in the local area (Knott, 2016; McLaughlin, 2010; Preibisch and Binford, 2007). The literature on foreign domestic workers reveals the role of labour brokers and sending-state bureaucracies in contributing to labour market segmentation by branding lower-paid Indonesians as ‘docile servants’ and higher-paid Filipinas as ‘compassionate, loving super-workers’ (Constable, 2007; Guevarra, 2014: 131; Lan, 2006: 80). ¹ Social difference that precedes migration is sometimes recognized. For example, the English language ability of Filipina domestic workers and Jamaican farm workers is attributed to colonialism (Lan, 2006; Preibisch and Binford, 2007). However, the literature’s focus is accounting for how attributes ‘discursively . . . [ascribed] to certain social groups based on an understanding that these qualities and traits are inborn’ are socially constructed by employers, brokers and labour bureaucracies to create a compliant workforce with desired subjectivities (Deshingkar, 2018: 2643; also see Collins and Bayliss, 2020). The case of Thai and Bangladeshi workers instead directs our attention to the significance for labour market segmentation of socialization processes that begin before employment and even recruitment.

This article is organized as follows. The research methodology is presented in the first section. The second section offers an overview of Singapore’s low-wage construction industry and considers the role of sending-state governments in positioning their nationals in Singapore’s labour market. The third section demonstrates that Thais and Bangladeshis originate in different classes in their origin countries and that this contributes to their variant positions in Singapore’s construction labour market. The conclusion reviews the contributions and limitations of this research.

Singapore’s low-wage construction industry

Beset by a reputation for poor conditions, Singapore’s construction industry historically struggled to attract local labour (Debrah and Ofori, 1997). From 1965, the government permitted the hiring of Malaysians; beginning in 1980, workers from designated ‘non-traditional source countries’, initially Bangladesh, India, Indonesia, Sri Lanka and Thailand, were recruited (Fong and Lim, 1982). Attributing the industry’s low productivity to a dependence on foreign workers and the liberal use of subcontractors (Debrah and Ofori, 1997; Ofori and Lim, 2009; Pheng and Zheng, 2018), since the mid-1990s, the Singaporean government has required prospective migrants to undergo months of self-funded trade training at Overseas Training Centres in labour-sending states (Channel NewsAsia, 2013; Tan, 1997). Yet the training requirement has not transformed the industry to one in which labour processes always incorporate the expertise of a skilled workforce. Many firms stay afloat not by increasing productivity through the deployment of skilled labour but by hiring the cheapest workers available (Au, 2016) and subjecting them to wage theft (Bal, 2016; Yea and Chok, 2018).

The role of labour-sending states

As Singapore resists bilateral migration agreements, labour-sending states have limited latitude to influence the terms of migrant nationals’ employment (Kaur, 2010). Nevertheless, Thailand and Bangladesh appear to have assumed somewhat different stances on the ‘numbers-vs.-rights trade off’, the dilemma of whether to maximize migration or promote migrant rights (Ruhs and Martin, 2008: 261). Differences first became apparent in the 1970s when Thailand introduced a minimum wage that Middle-Eastern employers were required to offer Thai construction workers (Thanakhan Haeng Prathet Thai, 1978), while Bangladesh only established negotiable wage guidelines (Hossain, 1981). In 1990, Thailand stipulated a minimum wage for Thai construction workers in Singapore (Straits Times, 1990), but Bangladesh never followed suit.

Nevertheless, it seems unlikely that the Thai government is exerting upward pressure on Thai workers’ wages. The minimum wage has remained stagnant at SG$23 per day since 2005 (Samnakngan Borihan Raengngan Thai Pai Tangprathet, 2016), although researchers in Thailand’s Department of Employment contend that low wages are contributing to a reduction in Thai migration to Singapore (Samnakngan Borihan Raengngan Thai Pai Tangprathet, 2014). ³ When wage discrimination (vis-a-vis Chinese workers) was broached with the Thai Labour Attaché in 2017, he stated that wages were established at the discretion of employers and declined to intervene (Simot, 2017).

Migration aspirations and ability

Carling and Schewel (2018: 947) suggest migration can be broken into two discrete processes: aspiration – ‘the evaluation of migration as a potential course of action’, and ability – ‘the realisation of actual mobility or immobility’. The application of this ‘two-step approach’ (Carling and Schewel, 2018: 945) is useful for understanding how class or socioeconomic difference is a determinant of ‘who is able to move and who cannot’ (Van Hear, 2014: 117) from a given labour-sending state. An examination of potential earnings and recruitment expenditures offers insight into why different subsets of the Bangladeshi and Thai populations have both the aspiration and ability to migrate.

Relative to the economies in their origin countries, Bangladeshis stand to earn much more than Thais in Singapore. In 2016, the average monthly salary in Bangladesh was SG$212 (12,915 Taka) compared to SG$567 (13,803 Baht) in Thailand. ⁴ Earning SG$18 per day (SG$468 per month), Bangladeshis would realize more than double their country’s average salary, whereas Thais earning SG$23 per day (SG$598 per month) would earn only nominally more than their country’s average salary. We therefore might imagine that the aspiration to migrate to Singapore is more limited in Thailand than in Bangladesh.

Prospective migrants must pay for their recruitment, training and travel in the form of a fee payable upfront to a sending-state employment agency or training centre (TWC2, 2012). Partially because of economic difference between the two sending states, the fees – in 2013, first-time Thai migrants were charged the equivalent of SG$4000 in employment agency fees; Bangladeshis paid about twice this much, an average of nearly SG$7256 (TWC2, 2012) ⁵ – present a more formidable barrier for Bangladeshis (about three years of average earnings) than for Thais (about seven months of average earnings). We can therefore surmise that the ability to migrate to Singapore is restricted to a smaller subset of Bangladesh’s population than that of Thailand. ⁶

While these numbers are instructive, more definitive proof of migrants’ class backgrounds comes from ethnographic research in labour-sending communities. Observing that migration is limited to ‘families who are able to afford the recruitment fees’, Bal (2016: 51) argued that his ‘interview sample of 45 confirmed that Singapuri migrant workers tend to be overwhelmingly rural middle-class folk’ as delimited by their family’s landholdings and engagement in business and/or white-collar employment (also see Panaech, 2017; Rahman, 2017). In contrast, Kitiarsa (2006: 27; 32; also see Kitiarsa, 2014; Peth et al., 2018) describes Thai migrants as ‘working-class’ persons ‘from agricultural backgrounds’ as signalled by their village origins and limited capital resources. In rural Bangladesh, where strongly demarcated agrarian class divisions are rooted in the highly uneven distribution of land, the middle-class self-identifies as non-labourers relative to the functionally landless community members they hire to work their land (Bal, 2016; Rahman, 2017). Consequent to a comparatively egalitarian distribution of agricultural land in rural Thailand (Rigg, 2019), Thais originate in communities where class differentiation is relatively unpronounced and where land is historically worked by its owner with intermittent input from other community members through labour exchange. Rural Thais self-identify as working-class or as farmers against a non-labouring urban elite (Keyes, 2012; Kitiarsa, 2014; Rigg, 2019).

Differently classed socialization

The variant social classes of the two groups of migrants – and their consequences for migrants’ socialization in the origin country – may be illustrated by relating the biographies of two informants: Som, a Thai, and Harun, a Bangladeshi. Som – who was 33 years old in 2013 and had been working in Singapore for 11 years – grew up on a rice farm in Thailand’s Northeastern Region. Although expected to help on his family’s farm, Som managed to complete nine years of schooling before his ambition to become an electrician was thwarted by a brother’s failed migration attempt. Citing his decision to leave technical school as an act of sympathy for his parents, he found employment in a Bangkok factory and then as a cleaner at a hotel. Finally, when he was 22 years old, he decided to migrate to Singapore because he was ‘tired of not having money’.

Harun, 28 years old and working in Singapore for six years in 2013, grew up in Southwestern Bangladesh. Harun’s grandfather had extensive landholdings, and the family hired up to a dozen labourers to grow rice, sugar cane and vegetables. Before his death, when Harun was 10, Harun’s father had a small business. Harun’s construction job in Singapore was his first ever. In Bangladesh, Harun’s time was engaged in study and socializing. However, two years into his study for a BA, Harun left for Singapore. He decided to migrate; first, to relieve tension with his mother who worried that Harun’s peer group was a bad influence, and second, because of a sense of the futility of education: ‘If you study, you have to pay a lot of money [for tuition] and then you still can’t get a job. Because to work the office job in Bangladesh you have to pay money [for the position].’

These biographies reveal certain similarities between migrant construction workers from Thailand and Bangladesh. Both groups of workers originate in rural areas and seek work in Singapore because of a lack of satisfying employment opportunities in their home communities. Nevertheless, there are also significant differences between the migrant groups captured in the life stories. First, both 22 when they migrated to Singapore, Som quit studying at age 15, whereas Harun continued to study until just before his departure for Singapore. This pattern with respect to educational differences is borne out across the interview data. Most Thai informants had only a primary school (6th grade) education, although three had studied through lower-secondary school (9th grade) and one had completed 12th grade. In contrast, over half of Bangladeshi informants stated that their highest level of education was completion of secondary school (10th grade), while the others had completed their bachelor’s degrees (n = 4) or had some college (n = 2). While college-educated Bangladeshis are likely overrepresented in the sample, ⁷ the differential educational attainment of Thai and Bangladeshi migrants is attested to in other researchers’ data. A survey of 205 Bangladeshi migrants found that 84% had at least a secondary school certificate (Yeoh et al., 2017). In contrast, Kitiarsa (2014: 15), who met hundreds of Thai construction workers while conducting more than three years of ethnographic research, observes that ‘most Thai workers have attained only four- or six-year compulsory education’.

Second, like every Thai informant, Som had worked on his family’s farm and had held down various blue-collar jobs. Harun, like most other Bangladeshi informants, had never had a job and his farm was worked by hired labourers. Thais’ accounts of their lives before migration to Singapore were structured by the struggle to supplement farm income with off-farm employment in rural and urban Thailand – in jobs including tractor driving, factory work, breeding dogs – and by seeking employment in countries such as Taiwan, Israel and Saudi Arabia. For example, Jerun related:

I’ve done almost every kind of work . . . I graduated from primary school when I was 12. Then I worked with my mother and father, with my relatives. We grew rice, we had a farm, we grew sugar, and taro. And I cut down wood . . . illegally in some forests . . . I worked in a mirror factory in Taiwan for three years . . . I tapped rubber trees in Malaysia for two months.

Whereas all Thai informants held jobs in Thailand before migrating and nearly half had done some form of construction work, only three Bangladeshi informants had worked in Bangladesh (one as a part-time video clip editor, one as a quality inspector in a factory and one as a motorcycle mechanic) and none had worked in construction. Although all Bangladeshi informants came from families with agricultural landholdings, only two reported helping on the farm. As Mikhail explained: ‘My father gets poor men to help. He pays them a little. Poor men want jobs. If we have money, we give jobs. My father doesn’t want [his children] to work . . . . We study.’

Third, Som moved to Singapore in pursuit of more money for a job similar in prestige to the ones he had worked on in Thailand. Harun, on the other hand, accepted a job in Singapore’s construction industry, although he once imagined he would work in an office. This difference is reflected throughout the data – Thais leave to earn a better wage for the blue-collar employment to which they are accustomed, whereas Bangladeshis submit to doing blue-collar work for the high salary. Pan, a Thai who first migrated to Singapore at 21, explained that he quit working as a carpenter for the Sino-Thai construction firm in Bangkok, a job he held for two years, after meeting a friend home from Singapore on vacation and noting he had money. Thiradit also worked in construction in Thailand and compared the industries as follows:

Thiradit: Here there is overtime. But in Thailand, . . . when you’re working with a contractor building small houses, there will never be overtime. If you’re building an apartment complex [in Bangkok] there will be overtime.

Researcher: So why not build an apartment complex?

Thiradit: Here [with overtime] I can earn 20,000 to 30,000 baht per month. I can earn 20,000 baht a month [with overtime] in Bangkok, but there I would have to ask my supervisor for an advance to buy food and there would be nothing left at the end of the month.

Hasin, like Thiradit, also related how he would have no savings if he worked as a labourer in his home country:

My country not like Singapore with big industry. You know garments? So dirty . . . . Only manager, they take very high salary. Worker salary, not enough . . . . First time in Dhaka salary [is] 2000. You rent bed 1000, food 1000 . . . Cannot live life like that. Money come and then go. Cannot live life. Very difficult. So I think . . . I have to go other country.

Yet Hasin supplied this example for illustrative purposes as he had never seriously considered becoming a garment industry labourer. He was on a pathway to a respectable job, like that held by his civil servant father and brothers, before his plans were thwarted:

I no finish my studies. I cannot get a good job in my country, cannot. My father want me to be a doctor but I no study good. I go other city, so far from my village . . . I study two years, but no good . . . I love one girl. Then the girl give me love. I go crazy. I cannot attention my study.

Hira also experienced a disruption in his life trajectory when his father, a university-educated manager at a garments factory, died while Hira was in high school, which led him to seek work in Singapore to alleviate the family’s ‘money problem’.

While Harun, Hasin and Hira reportedly veered off their pathways to middle-class jobs because of personal problems and unlucky circumstances, the narrative of all Bangladeshi informants was not so dramatic. Many simply related that Singapore jobs paid much more than jobs in Bangladesh. Arshad explained: ‘Bangladesh job not very good. Singapore, I thinking I get more money. I develop my life more with job in Singapore. Many income.’

‘One Thai is equivalent to two Bangladeshis’

According to employment agents, construction firms account for workers’ differently classed backgrounds when hiring. According to one employment agent, Singaporean firms find the labour of ‘hardened Thai farmers’ more conducive to realizing value than that of ‘Bangladeshi schoolboys’. He explained that employers ‘love’ Thai workers because they ‘take initiative’ and are ‘hard-working’. He cited one of his clients, a European multinational contractor with multiple large-scale projects in Singapore:

The manager says, ‘Give me quality not quantity’. He said one Thai is equivalent to two Bangla workers . . . I don’t know. I don’t go to the work site, but this is what the employers tell us. They say, if they leave and say, ‘Do A, B and C’, if it is a Thai worker, it will be done. If it is a Bangla worker, A will be done, but B and C will be left undone.

A second employment agent described Bangladeshi workers as follows: ‘If [the supervisor is] not there, [they are] sitting. If [the] boss comes, [they are] running to do everything.’ In contrast, ‘Thais work hard, [they are] sincere’. Then he goes on to explain: ‘Thais are farmers. They work under the sun. Banglas are very smart. Heavy jobs, hard jobs – they don’t want. They want to clean [the worksite].’

Bangladeshis’ purported lagging commitment to their work vis-a-vis Thais also emerged as a theme in interviews with Thai workers, as is illustrated by the following two examples:

Researcher: Why do Singaporean employers hire Thai labour?

Jerun: Well, it is because Thais can do everything.

Researcher: Why can they do everything?

Jerun: I also don’t know. Employers order us to do something and we can do it . . . Thais can do everything that the foreman orders. Thai people, they’re good at work . . . . Tell a Thai person to pour cement at 5 p.m. At 5 p.m. they pour the cement and then they’re done. A South Asian, you order them to do something at 5 p.m., they’ll stand by.

Researcher: Why did the company receive Thais if they were more expensive than Bangladeshi workers?

Bwee: Thais are smart, they know the work, they are hard-working. They learn fast, work fast . . . But Bangladeshis don’t work well. The manager tells them to work. And they don’t work.

Researcher: Why don’t they work?

Bwee: They don’t try. They make a little, so they try a little. They tell the manager, my wage is just this, I won’t work, let the person making a lot of money work.

Whereas Thais described Bangladeshi workers with attributes such as slow-working and working according to the clock, they described fellow Thai workers as hard-working, skilled, capable and quick-to-learn. A couple of the Bangladeshi informants expressed an outlook on the quality of labour performed by their own national group consistent with the narratives related above. For example, Harun estimated that only 20% of Bangladeshis work hard on the jobsite. The other 80% slack off, reasoning, ‘working can earn money, not working can also earn money’. Most of the Bangladeshi informants, however, neither disparaged the work habits of their own national group nor that of the Thais, whom they described as knowledgeable (‘working understand’), capable (‘can do the hard job’) and ‘good’. The Bangladeshi informants were more likely to attribute the wage differential strictly to Thailand’s stronger economy (also cited by several Thais) or to Thai government policy.

Despite the consensus of the Bangladeshi informants, it seems probable that first-time Thai migrants have a greater aptitude for construction labour than do first-time Bangladeshis. For one, as described above, in contrast to Bangladeshis, Thais often arrive in Singapore with prior construction experience and some construction skills. In some cases, Thai informants were not placed in jobs where they could draw substantially upon previously acquired experience – Harn who arrived with six years of welding experience, constructed roads and curbs during his first Singapore contract; Som with electrical experience from technical school was tasked with building bus stops. ⁸ Yet, in other cases, the Thai informants were placed in a position akin or somewhat akin to a previous job. Pan, who was a welder in his first Singapore contract, had worked as a welder for six months in Thailand (in addition to his two years as a carpenter); Yien, who did electrical wiring in elevators in Singapore, had been a general maintenance worker in Thailand.

Not only are Thais more likely to have more prior experience with construction labour and manual labour more generally, but ethnographic research conducted in labour-sending states reveals that Thai and Bangladeshi workers are socialized to hold manual labour in variant levels of regard. Kitiarsa (2009) reveals that the performance of hard, physical labour is crucial to the realization of a mature masculine identity in the labour-sending regions of Northeastern Thailand. Meanwhile, Bal (2016: 54) found that ‘from the [migrants’] class background, respectable professions are either reputable business enterprises or white-collar employment’. Rahman (2017: 75) likewise found that Bangladeshi migrants in Singapore came from a social world where historically ‘the wealthiest landowners appeared not to work at all. They hired others to perform agricultural and domestic tasks. As a result, [they] . . . saw work as a sign of low status.’

Differences in job type and firm types

While it should be kept in mind that this article draws from a small sample of workers in an industry with considerable diversity in firm type and occupation, the data suggest that first-time Bangladeshi workers are more likely than first-time Thai workers to be placed in positions where skill or capability is considered less essential. Of the seven Bangladeshi workers interviewed who were placed with a main contractor or subcontractor, four identified themselves as working primarily as ‘helpers’ or ‘cleaners’. For example, Arshad described his job with a subcontractor as: ‘I was just like a general worker, every day the same, like painting, install the structure, just helping’. Sazedur worked for a subcontractor that installed cables and explained his role as follows: ‘I only just help. But supervisor have. We just helping.’ Sohal worked for a main-contractor that did structural formwork and rebar on a high rise. However, he was given the task of cleaning up the jobsite because ‘I didn’t understanding anything’.

Of the 16 Thai workers interviewed who were placed with a contractor or subcontractor, most described themselves as doing a specific type of work; for example, welding, carpentry, laying cement, or setting up gondolas. No Thais identified themselves as helping and only one identified himself as cleaning, a job for which he had considerable disdain:

I complained to the foreman: ‘I request not clean now. Let me do something else.’ He said, ‘Okay’. Then he told me to collect leaves. I withstood working for him for six more months . . . I told him: ‘Thai people can do everything! And you’re going to have me collect leaves?’

Wee described being placed in an informal supervisor position vis-a-vis South Asian labour for his first year in Singapore:

The foreman, a Chinese-Malaysian, showed me how to do many things. The other Thais also helped me to learn. I was almost like a foreman. Sometimes the foreman wouldn’t come in until 10 a.m. and I would have to be responsible for the work before he arrived. The Bangladeshis mostly unloaded materials and picked up trash off the sites.

According to Boonpuk, Wee’s situation is not uncommon: ‘Most contractors hire at least a few Thais. It is their job to go around and correct the work of other workers.’

The interview data also suggest that Bangladeshi workers are more likely to be placed in labour-only supply companies than Thais. Seven of the Bangladeshi informants were placed in labour-only supply companies for their first contract, and six of these workers described the workforce of their supply company as all South Asian. Only one Thai informant, Boonluan, was placed in a supply company where he was one of only a handful of Thais.

Labour-only supply companies are presumably less concerned with worker productivity because profits depend on supplying workers (and realizing money from their contract extension fees) rather than ensuring the timely completion of a building project (TWC2, 2012). According to Boonluan:

If [Bangladeshi workers] extend [their one-year contract], they have to pay [the labour-only supply company where he worked] SG$2000. They say, those in Bangladesh who are waiting to come, there are 100,000 to 200,000 of them. So, if you aren’t interested, you don’t have to stay. It’s a business. They are selling people.

Hira explained that labour-only supply companies sought to make money from workers: ‘They cut your salary SG$50, if you one day no working, main-con, I know, one day, no working, he cannot cut one dollar’.