Layered vulnerability: Temporary migrants in Australian horticulture

Introduction

On 4 August 2014, the Fair Work Ombudsman (FWO) announced a review of the wages and conditions of young working holiday makers (WHMs) in recognition that ‘overseas workers can be vulnerable and/or require special assistance’ (FWO, 2014a). A substantial literature exists showing the vulnerability of migrant workers in Australia and other countries (Anderson, 2010; Mares, 2012; Preibisch, 2010; Refslund, 2014; Rogaly, 2008). Studies also show temporary migrants to be especially vulnerable in horticultural jobs where they may be disadvantaged by undocumented status (Martin, 2003; Pena, 2010), being bonded to a single employer (Preibisch, 2010), social exclusion (Preibisch, 2004), poor English skills (Sargeant and Tucker, 2009) and precarious employment (Kroon and Paauwe, 2013; Potter and Hamilton, 2014) – factors which generally combine to render unenforceable local (and international) labour rights (Campbell and Tham, 2013; Fudge, 2011).

This article aims to add to this literature focusing upon Australian horticulture. However, it goes further by considering the diversity of the harvest workforce and the ways in which vulnerability is multi-layered rather than homogenous. The concept of layered vulnerability has been associated with a range of factors that cause disparities in migrant workers’ working conditions. Sargeant and Tucker (2009) divide these factors into three groups – migration factors (including migration status and whether bonded to an employer); migrant worker characteristics (especially language, education and skill level); and receiving country conditions (including union and regulatory protection and social exclusion). Because these factors do not necessarily apply evenly to all temporary migrant workers, vulnerability may be layered, producing differential degrees of disadvantage. Previous studies of layered vulnerability have focused on macro- or meso-level analysis (Sargeant and Tucker, 2009) to assist analysis of political, economic and institutional influences. This article focuses instead upon micro-analysis, examining layered vulnerability in the specific context of Australian horticulture to identify the immediate causes of layered vulnerability and the implications for policy.

Horticulture in Australia until the 1990s was distinctive because few temporary migrants were used for harvest work. Since then reliance has grown on two different types of temporary migrants. First are WHMs. These are young travellers, without dependents, who can apply for a second 12-month working visa in Australia once they have completed 88 days’ work in horticulture (or a limited range of other regional industries). Doyle and Howes (2015: 15) found that nearly half of the growers they surveyed reported that backpackers (WHMs) were the main type of worker they employed. The second important group of temporary migrant workers are undocumented workers. While little is certain about their number, the evidence available suggests they have grown recently to form a significant part of the harvest workforce. Drawn largely from Asian countries outside of the WHM program, their exposure to exploitation appears to be very high (Hall & Partners, 2012; Howells, 2010), in part because they have no right to work in Australia and cannot claim legal protection for their employment conditions (Clibborn, 2015).

In contrast, all other horticultural workers, whether Australians or WHMs, nominally enjoy legally enforceable employment conditions under the modern federal award (the Horticulture Award 2010), which fixes minimum hourly wages, provides an additional 25% casual loading, and specifies that piecework rates ‘must enable the average competent employee to earn at least 15% more per hour than the minimum hourly rate’ (Clause 15.2). While the award prescribes a 38-hour week for full-time workers, most harvest workers are hired on a casual basis and are only paid for the hours they work. Piece workers are paid only for what they pick, and their piece rate must be voluntarily agreed between the employer and the harvest worker. Clause 15.9 of the award suggests there is no floor to actual piecework earnings, stating ‘nothing in this award guarantees an employee on a piecework rate will earn at least the minimum ordinary time wage … as the employee’s earnings are contingent on their productivity’ (Fair Work Commission, 2010). Thus, the award clearly provides scope for some variance in piecework earnings and casual hours. Notwithstanding the casual and piecework conditions in the award, it provides a ‘floor of rights’ which are legally enforceable. Whether this floor of rights is observed in practice, and which layers of the workforce are most vulnerable to substandard conditions, are issues explored in the following text.

The remainder of the article is divided into five parts. The first describes the research methods used to collect data. The second section looks at the layered composition of the harvest workforce. The third describes the short-term, casual nature of harvest work and the processes of job search, exploring how uncertain employment exacerbates vulnerability. The fourth section considers evidence upon pay, working hours and work intensity to show which harvest workers experience the most exploitative employment conditions. The final section reviews the article’s findings upon layered vulnerability and then considers policy issues.

Methodology

The article is based upon data collected from empirical fieldwork conducted in three stages during 2013 and 2014. First, an extensive interview program was undertaken at three regional locations in Victoria (Bendigo, Maffra and Mildura), in Tasmania and in the Northern Territory. Harvest workers were interviewed through nine dual-moderator focus groups with a total of 64 participants from the following countries: England (15), France (8), Burundi (7), Ireland (7), Germany (5), Hong Kong (5), Italy (3), Afghanistan (3), Taiwan (3), Malaysia (3), Scotland (2), Estonia (1), Japan (1), South Korea (1). With the exception of those originating from Burundi, Afghanistan and Malaysia, all were WHMs. When interviewed they were harvesting a range of crops including apples, cherries, strawberries, grapes, citrus, salad and mixed vegetables and mangoes. In addition, interviews were conducted with farmers, labour-hire contractors, employment agency staff, backpacker hostel operators, union officials, occupational health and safety (OHS) government agency staff and ethnic community organisations. Second, in late 2013/early 2014 the interview program was supplemented with telephone interviews of backpacker hostel and caravan park operators in all six Australian states and the Northern Territory.

Third, a survey was administered online to harvest workers. The questionnaire was presented in both English and Chinese (10% of respondents used the latter), and was available online from mid-December 2013 to mid-March 2014. Respondents were recruited by placing invitation cards at working hostels used by harvest workers, and through a website used by WHMs to locate harvest work. Data cleaning reduced the initial 417 responses to 303 useable responses. Of these, 58.1% were male and 41.9% female, while the mean age was 24 years. Current crops or tasks included grapes (20.8%), vegetables (18.2%), berries (12.9%), stone fruit (12.5%) and citrus (8.5%). Four out of five were WHM first year visa holders, 72% of whom were seeking a Second Year Visa. Other respondents included Australians, New Zealanders and international students. Table 1 shows the distribution of respondents into demographic categories. No respondents were known to be undocumented workers – a group beyond the reach of a survey distributed through hostels and websites. Also, some degree of response bias should be acknowledged, skewing findings towards dissatisfied workers.

OPEN IN VIEWER

Table 1.

Survey respondents by nationality, state location and visa type.

Nationality/Region (n = 303)	%
European Union (excluding UK/Ireland)	33.3
Asia	23.1
UK/Ireland	21.8
Australia/New Zealand (NZ)	15.5
Other	6.3
Total	100
Current job location (n = 303)	%
Victoria	39.3
Queensland	19.8
New South Wales (NSW)	19.1
Tasmania	6.3
South Australia	5.9
Western Australia	5.9
Northern Territory	3.6
Total	100
Visa (n = 259, excluding Australian/NZ)	%
Subclass 417 year 1	81.5
Subclass 417 year 2	10.8
Student visa	3.9
Other visa	3.9
Total	100

Layered vulnerability: The structure of the harvest workforce

Analysis of the size and composition of the horticultural workforce is complicated by seasonality, the involvement of family members, and the unquantifiable use of undocumented workers. Seasonal workers are not identified separately in official employment statistics, and their movement between harvesting locations impedes an accurate count (Hanson and Bell, 2007). Notwithstanding these problems, estimates have claimed there are about 30,000 growers and 130,000 employees in Australian horticulture (FWO, 2011). The harvest workforce is made up of several groups. In this section, we focus on three – Australian residents, WHMs and undocumented workers. While some international students (Knight, 2011), subclass 457 visa holders (Birrell and Healey, 2013; Deegan, 2008; Toh and Quinlan, 2009), Pacific seasonal workers (Doyle and Howes, 2015) and ‘Workers on Organic Farms’ work in horticulture, they are numerically insignificant and shall be excluded. Harvest workers of all kinds are nominally covered by the Australian Workers’ Union but almost never organised: union presence in horticulture rarely extends beyond packing sheds and processing plants where the National Union of Workers is sometimes active (Union Official Interview, 8 February 2013; Australian Broadcasting Corporation (ABC), 2015).

The composition of the harvesting workforce has changed considerably over the past 20 years. Until the late 1990s, Australian residents were numerically dominant. These included permanent itinerants, local farm labourers, working class families who spent their annual holidays picking, and semi-skilled rural workers who abandoned their regular employment each year for the higher paid picking work (Hanson and Bell, 2007; Mares, 2005). Hanson and Bell (2007) categorised the harvest workforce in 2003 into four somewhat different groups. Most important were ‘permanent itinerants’ (40–50% of the workforce); second were WHMs (20–30% of the workforce); next were Australian retirees (15–25%); and the balance consisted of Australian working holidays makers and students (5–10% of the workforce). Since the 1990s, permanent itinerants have declined in number. A grower in Mundubbera reported that whilst 90% of his workforce used to be permanent itinerants, by 2003 they made up only half (Hansen and Bell, 2007). Hugo (2001) suggests that low wages, poor working conditions, and changing attitudes towards travel to remote locations for uncertain employment all contributed to this decline. The labour force gap left by the exodus of permanent itinerants was initially filled in part by WHMs (Doyle and Howes, 2015).

Introduced in 1975, subclass 417 (WHM) visas offered young travellers from the United Kingdom, Ireland and Canada the right to work during their 12-month stay in Australia. Supported by inter-country reciprocal agreements, WHM visas were intended to provide for employment incidental to travel, and to facilitate cultural exchange and understanding. The visa was available to young people without dependents, and no English language proficiency standard was necessary at that time (Hugo, 2001). By 1995, reciprocal agreements had been reached with three more countries and the number of WHMs had expanded so rapidly that a cap of 33,000 visas per annum (adjusted annually) was introduced (Hugo, 2001). In 1997, a Senate enquiry was held following up concerns that WHMs were limiting labour market opportunities for Australian youth (Parliament of the Commonwealth of Australia, 1997). Evidence presented to that enquiry pointed to the critical role WHMs had come to fulfil in seasonal harvesting, noting a preference by growers for WHMs over local youth. The Federal government accepted the committee’s recommendation that the cap on visa numbers remain. Reflecting community antagonism to guest workers at that time, the committee stated ‘Above all, the Committee wishes to avoid … the working holiday program being used as a form of guest labour’ (Parliament of the Commonwealth of Australia, 1997: xxvi).

By 2005, concerns about WHMs being de facto guest workers were outweighed by labour market needs. Following a period of extreme labour shortage in horticulture, and lobbying by the National Farmers Federation, the Federal government amended WHM visas to grant a Second Year Visa for those who completed 88 days rural work during their first 12 months in Australia. That work had to be performed in three industries: agriculture, mining or construction. The purpose of the changes was expressed clearly by the Minister for Immigration and Multicultural Affairs:

the changes to the visa do provide, I think, a significant incentive to people to get out there and help us bring in the crops. That in turn will provide farmers with a wealth of fit, mobile and enthusiastic workers when they are most needed for farm labour. (Parliament of the Commonwealth of Australia, 2005: 29)

The incentive of a Second Year Visa mostly attracts WHMs into harvest work, DIBP statistics showing that 90% of those who receive a second visa complete the qualifying 88 days work in horticulture (DIBP, 2014a, 2014b). Grower reliance upon WHMs for harvest work remains high. However, the 2015 Doyle and Howes Report (2015) for the World Bank shows a decline from 73% of growers depending on WHMs four years earlier to less than half at the time of their study.

One reason for falling reliance on WHMs may be increased use of a third group – undocumented workers. They are defined by the DIBP as ‘non-Australian citizens who are working in Australia without a visa (mostly “over-stayers”) or who are in Australia lawfully but working in breach of their visa conditions’ (mostly holders of a visitor or ‘tourist’ visa) (Department of Immigration and Citizenship (DIAC), 2009). Over-stayers (predominantly Chinese, American, Malaysian and British) tend to be international students, as well as skilled, high paid, urban workers (DIBP, 2012). Those breaking their visa conditions by doing harvesting or other unskilled work mostly come from low wage, developing countries in Asia. Because undocumented workers are deported for working illegally they are elusive and difficult to count. In 2011, the Department estimated that between 40,000 and 90,000 illegal workers were working in all industries (DIAC, 2011). Checks to identify and deport illegal workers in 2010–2011 located 1788 people, of whom about a third worked in agriculture. This is likely to be the tip of a large iceberg.

While quantification is not possible, these workers may now account for between a quarter and a third of the harvest workforce. This development appears to be recent. The presence of undocumented workers was observed in the past, but was of marginal concern to the Senate enquiry into WHMs of 1997 (Parliament of the Commonwealth of Australia, 1997). Yet, in 1999, the Department of Immigration and Multicultural Affairs found significant numbers of undocumented workers and was examining tougher sanctions on employers – a step opposed by horticulturalists on the grounds that ‘it was not always possible to attract sufficient legal workers during the harvest’ (Department of Immigration and Multicultural Affairs (DIMIA),1999: 27). By 2005, Mares (2005: 5) reported that ‘unless labour-supply problems in the horticultural industry are addressed, there is the potential for a growing number of jobs to be filled by undocumented workers … creating an increasingly antagonistic relationship between primary producers and immigration officials’. His survey of 176 growers in the Murray Valley found that 28% admitted to using undocumented workers at least some of the time (Mares, 2006). Three years later a study in a NSW regional centre also reported widespread but unquantifiable exploitation of undocumented workers (Segrave, 2009). In some Victorian regions, undocumented workers are now believed to make up the majority of horticultural workers (Hostel Manager Interviews, 23 January 2013, 15 January 2014).

Accompanying the growing number of undocumented workers are allegations of other illegal practices and the erosion of employment standards. The most authoritative evidence comes from the Howells Report arising from an inquiry commissioned by the Federal Government in 2010 to explore the effectiveness of those provisions in the Migration Amendment (Employer Sanctions) Act 2007 intended to deter the employment of non-citizens and those without permission to work. The Howells Report criticised growing reliance on this source of labour, saying the presence of these workers ‘is very often organised by intermediaries (contractors) who abuse and exploit these workers … these intermediaries are often involved in tax and welfare fraud and breaches of industrial, health and safety and other laws’ (Howells, 2010: 22). Howells criticised the effectiveness of existing measures to clamp down on undocumented workers by deporting them and prosecuting employers. Until 2011, prosecuting employers was difficult because hiring undocumented workers was a criminal offence. Since the main witnesses (the workers) were deported, the proof necessary for criminal conviction could not be presented and successful prosecutions were rare. Since 2011, more accessible penalties such as infringement notices and civil action have been introduced (DIAC, 2011).

The Howells Report went on to claim undocumented workers are frequently ‘underpaid, misled about what they are doing, undernourished, beaten and threatened’ (Howells, 2010: 56). The nature of these threats is illustrated by a Goulburn Valley hostel operator who described being raided by the Department of Immigration after ‘a contractor “dobbed in” an undocumented worker who complained about being paid AU$7.00 (nearly AU$10 below the award rate) – not AU$12.00 an hour. He told her: ‘if you can’t work for me you can’t work for anyone’ (Hostel Manager Interview, 15 January 2014). Another case involved a Chinese worker in Tasmania who asked his supervisor for back-pay and ‘three from the company physically attacked him causing the police to be brought in’ (Hostel Manager Interview, 14 January 2014). Another case from Mildura revealed both poor living conditions as well as the violence that can be bred in these situations. In 2012, a fruit picking contractor – Pardeep Kumar – was bundled into a car outside the town library and found two weeks later in NSW with his throat cut. The coroner’s enquiry exposed some of his business practices. He was:

paid by farmers to supply Indian nationals and kept a percentage of what he received … He had up to 45 people living in a four bedroom house who he charged $50 a week in rent and $5 a day to be transported to and from farms … In the days before his death he was carrying $18,000 which he had withheld from his workers’ wages because he needed to show the money up front to obtain a new contract. (Murphy, 2012)

The composition of the seasonal harvest workforce has shifted substantially from that identified just over a decade ago by Hanson and Bell (2007). Although statistics are not available, the largest group now appears to be WHMs, followed by undocumented workers; Australians including travelling retirees, and the miscellaneous groups of other temporary migrant workers such as international students and Pacific islanders, comprise a small minority. It is reasonable to conclude that these compositional changes have created layered vulnerability, with undocumented workers most disadvantaged.

Harvest work: Vulnerability in a casual labour market

Harvest jobs are generally short-term and unpredictable, linked to the vagaries of harvest conditions. This creates a challenge for horticulturalists who must recruit casual workers in the correct number and at the right time. The scale of this army of harvest workers is evident in Bowen, where a population of 12,000 swells to 17,000 over the harvesting season (Hanson and Bell, 2007). Job search under these conditions poses challenges for harvest workers who must locate work, pay for travel and accommodation, and minimise idle time between jobs. Hiring is done by growers (performing the normal legal duties of an employer) and by contractors (who relieve growers of most of these duties, charging an all-in rate to cover labour costs and their profit), both of whom depend upon the efficient circulation of job vacancy information. This can occur in several ways.

First, formal assistance for job search is provided by the National Harvest Labour Information Service (NHLIS), created in 2003 when the government contracted the Mildura and District Educational Cooperative to coordinate and distribute information about harvest jobs. This task is accomplished through a booklet (NHLIS, 2014) which provides general information about the harvesting cycle for crops around Australia, the ‘harvest trail’ website, and a free-call telephone service. In larger regional centres job search is also assisted by agencies funded by the government for each placement.

Second job search is aided by communication through the internet and social media. Several websites carry advertisements for harvest jobs. Amongst these are Gumtree, FruitPickingJobs, Harvest Trail, Harvest Bites Labour and Workabout Australia as well as some websites in Asian languages. Also popular with WHMs are backpacker hostel websites which advertise both jobs and accommodation (WHM interviews, 2013). Backpacker websites also exist, some carrying hostile reviews of poor hostels. One reviewer attacked the Lazy Lizard Lodge in Ayr, Queensland, stating ‘I found out that Lazy Lizard hardly ever gets any work because all the farmers in town hate the owner because he is that much of a horrible, rude, arrogant guy’ (The Backpacker, 2013). Reviews of this kind are commonplace in travellers’ blogs and websites. WHMs also establish groups to exchange job information using social media. One group of 30 female Taiwanese harvest workers operated an exclusive Facebook site to pool news about available work (WHM Interview, 29 April 2013). Most often WHMs alert friends informally as their travels take them to places where work is available. Mobile phones and wi-fi access are essential on the harvest trail.

A third source of job information is the network of working hostels and caravan parks (Jarvis and Peel, 2013). Many hostels act as information brokers between horticulturalists and WHMs, sourcing jobs from growers so they can fill their beds. One hostel operator in Victoria described his business model as ‘being about building relationships with growers’ (Hostel Manager Interview, 7 February 2013). He had acquired his hostel complete with a telephone sim card and spreadsheet allegedly plotting farm contacts. Most had lapsed so the network had to be rebuilt. Most hostel income is earned by charging from $120.00 to $180.00 a week for accommodation, and sometimes a daily fee of $5.00 to $8.00 for transport to farms. Driven by the imperative to fill beds, hostels sometimes advertise work when it is not available. One WHM described a Victorian hostel as ‘a .. holes because they have a big website, but only want to fill rooms not jobs’ (WHM Interview, 6 February 2013). More conscientious was another Victorian hostel which guaranteed work and was inundated with about 20 phone calls a day from WHMs. High demand allowed it to screen WHMs to ensure that growers received good workers and so would provide repeat business. Another hostel in Queensland guaranteed worker reliability by restricting on-site alcohol consumption to ensure WHMs were fit for early morning work (Hostel Manager Interview, 25 November 2013).

Some contractors and labour-hire agencies recruit temporary migrants directly from overseas agents (ABC, 2015). Since Australia, unlike some other countries, does not regulate contractors or require them to be registered (Underhill, 2013), little systematic information exists about them. However, recent studies reveal several levels of labour-hire contractors in horticulture. One study observed ‘at the highest level there are legitimate labour hire agents who provide a full labour hire service to their clients, many using backpacker labour … at the other end of the spectrum illegal contractors work with agents/facilitators overseas to recruit workers … and farmers are very willing to abrogate responsibilities to these labour hire contractors including with regard to the extent to which they employ illegal workers’ (Hall and Partners/Open Mind, 2012). Contractors and agencies are attractive to farmers because they remove the problems of workforce recruitment and management. Importantly, growers can delegate to contractors the duty of checking the visa status of their workers, thus abrogating responsibility for the use of undocumented workers (DIBP, 2015).

The three workforce layers outlined in the previous section experience job search differently. First, Australian workers, including permanent itinerants, generally have regular work habits, returning to the same growers every year to perform seasonal work. Local knowledge provides them with the most reliable information about recurrent casual job vacancies.

Second, unlike Australian residents, WHMs rarely obtain repeat work and depend upon the aforementioned job search processes. These processes create risk for WHMs because they are racing against time to record 88 days’ work in one year to gain a visa renewal. One danger of lost time between jobs was highlighted in 2011 when tropical cyclone Yasi stranded two WHMs in Atherton (Queensland) just short of the 88 working days target required for a visa extension, rendering them unable to remain in Australia. As their hostel manager observed, ‘the Department (of Immigration) can be very strict’ (Hostel Manager Interview, 25 November 2013). Most WHMs need multiple jobs to achieve the 88 day target. Our survey found average job tenure was just 20 days, forcing WHMs into constant job search. While interstate travel remained rare (73% worked in only one State), moving between districts was almost universal, creating dislocation and lost work time. Exacerbating job search problems is the lack of useful and reliable job vacancy information. Only 4.7% of survey respondents used the NHLIS website (a poor source of detailed information), relying instead upon working hostels (32.8%), the internet (29.7%) and friends and other travellers (24.6%) for job information. These sources can be unreliable. Indeed false job vacancy information is endemic, reported by half of survey respondents and in focus groups. Those focus group participants who travelled to the Northern Territory for the 2013–2014 mango harvest, for example, made an expensive journey only to discover the harvest had failed (WHM Interview, 17 October 2013). Most vulnerable to untrustworthy job information are Asian WHMs who are further disadvantaged by language barriers. Of the Asian WHMs surveyed, almost 40% relied upon friends and other travellers for job information, and 63% complained about false job information.

Third are undocumented workers. Available reports suggest they are almost entirely reliant upon the contractors who bring them into Australia, place them with growers, and organise their return journeys. The conditions under which they are recruited overseas and hired in Australia resemble bonded servitude rather than informed participation in a free labour market, placing them in the bottom layer of the harvest workforce.

Pay, working time and work intensity: Three dimensions of vulnerability

The FWO, who is responsible for award enforcement, repeatedly reports compliance problems in horticulture. In 2009–2010, a shared industry compliance program was pioneered by the FWO alongside unions and employers to both audit compliance and educate horticulturalists about their obligations (Hardy, 2011). The audit found 36% of farmers in contravention of the award, a quarter of these for underpayment but most often for not keeping pay records or giving payslips (FWO, 2011: 7). In August 2014, inspectors found that strawberry farms around Caboolture (Queensland) had underpaid pickers and packers by $134,000, and were failing to make written agreements with piece workers (FWO, 2014b). Award non-compliance and poor employment record keeping appear widespread in Australian horticulture (Doyle and Howes, 2015; Mares, 2005; Tan and Lester, 2012). This section examines survey and focus group evidence upon earnings, working hours and work intensity to throw light upon whether employment conditions comply with the award and which subgroups are most vulnerable to sub-standard conditions.

Earnings information was provided by 278 survey respondents. Of these, 158 were currently paid by the hour and 120 by piece wages; 198 were paid by farmers/growers and 75 by contractors. Asian WHMs made up almost half of the number of contractor employees. In contrast to the award hourly wage of $16.87 for a level 1 employee (as of August 2014), mean hourly earnings for workers paid by farmers was $14.86 (SD 5.13), while contractor employees were paid a mean of $12.66 (SD 5.74). Table 2 shows hourly earnings for hourly paid workers and piece workers.

OPEN IN VIEWER

Table 2.

Hourly earnings ($) for harvest workers (n = 278).

Payment method	Mean	Median	Minimum	Maximum	Standard deviation
Paid by the hour (time wages) (n = 158)	$16.20	$18.00	$3.00	$28.25	4.833
Paid piece rates (n = 120)	$11.69	$12.00	$2.00	$30.00	5.01

Our earnings data reveals several points. First, farmers pay better than contractors, although average less than the award wage. Second, Table 2 shows that piece-rate hourly earnings average well below the award hourly minimum wage by about $5, and also average $4.51 less an hour than time-wage earnings, despite the award provision that the ‘average competent worker’ on piece rates should be able to earn 15% more. This evidence points strongly towards widespread non-compliance with the award – a finding that resonates with focus group complaints such as ‘long hours for terrible pay’ and ‘awful pay for hard work.’

Table 2 also reveals a third point – high variance in hourly earnings. While this is found amongst all sub-groups, we shall focus on variance in individual piecework earnings. A common explanation is the level of worker experience – an argument illustrated by new fruit pickers who complained that experienced counterparts take the ‘low hanging fruit’ in orchards, leaving them to climb higher and earn less. Ethnic stereotypes about worker capability deserve note. The view persists amongst growers and contractors that Asian workers are quicker and more careful than European or Australian counterparts. One Mildura farmer described Vietnamese orange pickers ‘running up ladders filling 20 kg bags and running down without using their hands – Circe de Soleil stuff … backpackers cannot compete and earn a third as much’ (Farmer Interview, 8 February 2013). Despite their alleged capabilities, we found Asian WHMs averaged the lowest earnings when paid by piece rates ($9.64 compared to the average of $11.69). Tan and Lester’s (2012) analysis of WHM earnings, based on 2008 data, similarly found a lower average hourly rate for WHMs from Japan and Korea. Crop type also affects earnings; some crops (strawberries and peas) being notorious for yielding low pay. A Scottish WHM in Mildura reported earning ‘only $110.00 for a full week picking peas … not enough for the hostel rent of $180’ (WHM Interview, 6 February 2013). At the other extreme an Irish WHM in Tasmania ‘made over $1,000 one week picking apples’ (WHM Interview, 30 April 2013). Clearly a wide range of factors can and do cause variations in the productivity and earnings of individual piece workers. But equally, the low average pay of piece workers suggests that horticulturalists fix piece rates too low on the basis of exaggerated performance expectations of the ‘average competent worker.’

A separate payment problem concerns non-payment of wages – a problem reported by 40 survey respondents (14.8%). Of these, 35 tried to recover their pay but only three were successful. One respondent tried to find the contractor but he had ‘skipped with everyone’s pay.’ Another respondent reported a contractor to the FWO but was then asked to fill in forms which were ‘too long and complicated’. After finding that the contractor had since left the district this picker gave up. Recovering lost wages is clearly difficult, but collective action can help. A German WHM recounted how he had ‘joined 40 backpackers besieging a house in Mildura for five hours until the contractor inside surrendered, came out, and paid everyone their overdue wages’ (WHM Interview, 6 February 2013). While growers are sometimes guilty of not paying wages, the survey indicates that working for a contractor doubles the likelihood of this occurring.

Variations in working hours are another important aspect of vulnerability. Since almost all survey respondents were casual, their earnings (whether time wages or piece rates) depended upon how much work-time they could get. Evidence on working hours was provided by 282 survey respondents. Of these 205 were employed by farmers and 77 by contractors. Figure 1 reveals the average daily hours worked by these two groups. Very short working hours (less than 4) were twice as likely amongst contractor employees while approximately a fifth of both groups reported long hours (more than 8). OPEN IN VIEWER

Working hours in horticulture can be unpredictable for reasons such as rain interruptions and excessive heat. Not surprisingly then the provision of information about expected working hours is important, but can also be a ‘hit and miss’ affair. Just over half the survey respondents were informed about expected hours before starting at the farm (51.4%), a third (33.6%) on their first day, but 15.1% were not told at all. Variable working hours create uncertainty about weekly earnings and can cause dissatisfaction. Table 3 contrasts satisfaction with working hours amongst those paid by farmers and those paid by contractors. Two-thirds of those paid by farmers (62%) found the number of hours about right, in contrast to less than half of those paid by contractors (44.2%), while a further 27.3% of contractor employees complained of too few hours. Similar proportions amongst both groups said they would like more hours but the work is too tiring. This raises the question of whether workers find the intensity of harvest work is too great.

OPEN IN VIEWER

Table 3.

Satisfaction with working hours and paid by farmer or contractor.

Satisfaction with working hours	Paid by farmer		Paid by contractor
	No.	%	No.	%
Too few hours	26	12.7	21	27.3
Too many hours	29	14.1	11	14.3
Number is about right	127	62	34	44.2
Would like more hours but the work is too tiring	23	11.2	11	14.3
Total	205	100	77	100

Focus group evidence clearly showed that WHMs find harvest work is exhausting. Nevertheless, work intensity does differ between groups. Survey evidence on work intensity was constructed from several indicators – carrying excessive loads, working in extreme heat, and not stopping for drink breaks – all indicators that the employment system may be pushing workers too hard. These indicators are shown in Table 4. Piece workers and hourly paid workers were contrasted on two work intensity measures – carrying heavy loads and working in excessive heat. On these measures, piece workers encounter more adverse conditions. Similarly, contractor employees were much more likely than farmer employees to work in extreme heat and miss drink breaks. For many WHMs, the reality of harvest work is somewhat different from the ‘fit and healthy lifestyle’ publicised in harvest trail promotional materials (NHLIS, 2014).

OPEN IN VIEWER

Table 4.

Indicators of work intensity.

Work practice	Piece workers (%)	Hourly paid workers (%)
Often/always carry excessive loads a	31.9	11.2
Often/always work in extreme heat (e.g. More than 35℃) b	48.3	25.0
	Contractor employees (%)	Farmer employees (%)
Often/always work in extreme heat (e.g. More than 35℃) c	46.7	30.1
Often/always missing drink breaks d	24.0	16.1

The evidence presented in this section makes it clear that pay and working conditions vary greatly. Which groups are most vulnerable? While we are unable to make comparisons with undocumented workers, the WHMs who make up four-fifths of our respondents appear themselves to be layered into different groups. Relative vulnerability depends upon several factors, in particular whether they are paid on piece rates (which leads to low pay and intensive work pressures) or engaged by contractors (which leads to low pay, shorter and unsatisfactory hours and more intensive work pressures). WHMs of Asian origin were the most vulnerable, and were more likely than Australians and other nationals to work for a contractor and receive lower hourly earnings. Limited English language skills often leave this group vulnerable to predatory contractors.

Conclusion: Layered vulnerability and policy

Layered vulnerability, as described by Sargeant and Tucker (2009), is a complex concept comprised of three major dimensions and several more subsidiary dimensions – all likely to vary as observed in different contexts. Our findings suggest that all Australian horticultural workers are vulnerable but there are differences in degree which are consistent with various dimensions of layering. First, undocumented workers appear the most vulnerable group in the workforce for several reasons, including their liability to deportation, dependence upon a single contractor, poor language skills, exclusion from unions and regulatory protection, and social isolation. Second, WHMs appear more vulnerable than Australian residents because lack of labour market knowledge exposes them to greater exploitation as they race against the clock to record 88 days’ work for a visa extension. Third, Asian WHMs are more vulnerable than other WHMs, having fewer sources of independent and reliable job information and receiving lower pay once employed. Fourth, harvest workers hired by contractors (often Asian) are more likely than farmer-employed workers to encounter below-award pay, short and unsatisfactory working hours and more intense work. Finally, piece workers are more vulnerable than hourly paid employees to low pay and intense work pressures. Searching for the lowest labour costs, some employers take advantage of these variations in vulnerability to find the weakest workers and the most exploitative contracting arrangements and payment systems. Without enforceable minimum standards of employment, other employers are then forced to follow suit, producing a downward spiral of wages and employment conditions.

Two caveats are needed about these findings. First more research is required about the size and composition of the harvest workforce, and especially the numbers and working and living conditions of undocumented workers. Second, the context of growing industry pressures in horticulture requires greater consideration (Rogaly, 2008). In 2009, the southern states of Australia emerged from an eight-year drought that forced up water prices, restricted water supply, and drove many horticulturalists off the land (Evans, 2014). Also, foreign competition made fresh inroads into Australian markets and the high value of the Australian dollar discouraged exports. At the same time, Australia’s two major supermarket chains forced harsher contracts on produce suppliers and processors. Recent movements to curb the abuse of retailer market power have been met with proposals from Woolworths, Coles and the Australian Food and Grocery Council for a voluntary code of conduct which may not solve the problem (Mitchell, 2014). It is likely that growers transmit these pressures to their workforce.

The multilayered nature of workforce vulnerability makes it unlikely that any single policy will be sufficient to eliminate disadvantage and support compliance with award standards. Thus several policy areas need consideration. First, immigration policies need attention to better manage the supply of documented harvest workers and remove the need for an undocumented workforce – a phenomenon that is growing (in spite of legal reforms) and breeds a host of illegal and unethical practices. Second, there is (as one political leader recently recognised) a strong need to license labour-hire companies (contractors) to drive out those engaged in illegal or unethical practices (Victorian Government, 2015). Almost all countries from which Australia draws WHMs license labour hire agencies to minimise problems of the kind reported in this study (Underhill, 2013). Third, the award requires attention where it prescribes piece rates be fixed by agreement between the farmer and worker so the ‘average competent worker’ can earn a certain amount. This process is unrealistic and allows productivity expectations to be fixed too high. In the United Kingdom, the national minimum wage fixes a floor to piece rate earnings so that exploitative individual bargaining cannot occur (Rogaly, 2008). Such a floor would eliminate many abuses in Australian horticulture. Policy reform is needed. The labour market trends described in this article have made harvest work unattractive to Australian workers, and increasingly to WHMs, and filling the gaps with undocumented workers should not be considered an acceptable solution to labour shortages.