New insights into informal migrant employment: Hand car washes in a mid-sized English city

This article provides new analytical insight into migrant labour by examining a newly emergent low-margin sector, hand car washes (HCWs). The sector is co-created by pressures from above in the form of economic restructuring and from below by employers and migrants who diffuse fluid and flexible low-wage employment. The diffusion of HCWs demonstrates how exploitative privatized employment generates autonomous economic growth in the unregulated economy. The formal and informal economies are however interlinked and overlapping within and beyond the labour process. Locally, HCWs have the potential to become the established car wash sector, putting regulated outlets in a state of uncertainty as informalization in employment if not business practice becomes the norm.


Introduction
The diffusion of hand car washes (HCWs) and the construction of an informal HCW sector in Leicester, a mid-sized English city, are informed by a range of factors. As a construction of the surrounding political economy, the growth of HCWs is stimulated by recent economic restructuring in petrol retailing which witnesses supermarkets now dominating the market. One consequence of this is the closure of road-side petrol filling stations. A second element of economic restructuring relates to deregulation in retailing and associated changes in drinking habits where consumers can buy alcohol in bulk and at discounted price in supermarkets. This combines with a third element in the impact of the smoking ban in bars, clubs and public houses. One consequence of these developments is that many road-side public houses and some restaurants too have closed. Theoretically then it is possible to link the role of 'from above' macro-economic restructuring and the process of informalization in employment to 'from below' micro developments in the informal economy, the labour process and other forms of exploitation.
The following section outlines our theoretical contribution in relation to the established literature on precarious and vulnerable workers. Specifically, our contribution fuses macro-economic restructuring, the detail of micro practices in HCWs and the association between these developments and the wider context of vulnerable migrant workers and precarious work. Precarious work and vulnerable workers within and beyond the 'employment relationship' are central concerns of developing labour process theory: they relate to the conditions in which work is performed in a low-margin sector; who controls the work of (HCW) workers; and the skills these workers possess and how they are paid. We then outline the characteristics of the informal economy and those of the wider social relations of informal employment. This is followed by a detailed discussion of the development and workings of and distinctions within HCWs both nationally and in Leicester. We then outline our research methods, further detail our associated methodological contribution and the challenges of undertaking research in this sector. This is followed by our findings, and the final section contains a discussion and conclusions on our contribution.

Analytical and theoretical contribution
Our analysis provides new analytical insight to the evaluation of low-margin employers such as HCWs in the informal economy. First, our examination of HCWs sees these outlets operate alongside formally established mechanized car wash providers. Much of the literature examines the contention that employers use migrant labour because they cannot find suitable domestic labour and that therefore 'labour shortages' and indeed migrant labour markets are culturally, economically and socially constructed. Critical evaluation of these constructions focuses on migrant labour labelled by employers as 'good workers' who are 'better motivated' and more likely than indigenous labour to accept poor pay and precarious working conditions (Findlay and McCollum, 2013;McCollum and Findlay, 2015;Ruhs and Anderson, 2010). That is, employers prefer to engage precarious workers in the formal or informal economy to enhance workforce flexibility and promote competitiveness which reduces wages to levels which only migrants are likely to accept (Standing, 2014: 176).
We provide new analytical insights to these findings by illustrating how employers and migrant workers co-create HCWs and employment therein as a fluid, dynamic informal sector. In addition to this we show how the growth and fluidity of HCWs have the potential to threaten the viability of the mechanized car wash sector. Building on Sassen (1997Sassen ( , 2009, we argue that the diffusion of HCWs in the informal economy is part of advanced capitalism where deregulation and flexibility in the formal economy have restructured the retail sector broadly defined. In turn, in the UK's post-industrial postcrisis finance-led economic regime this has stimulated employment in low-wage, lowskill informal labour-intensive sectors -the dynamics of informalization from above and below. The co-creation, growth and diffusion of HCWs demonstrate how informal employment generates economic growth in the unregulated economy (Moser, 1978(Moser, : 1052. In contemporary language we see this as autonomous and in effect privatized employment where no labour shortage previously existed. Hence in Leicester at least the specifics of our contribution relate to how HCWs have the potential to become the established car wash sector. This potential places established outlets in an uncertain future as informality of business practice and employment practice in car washing becomes the norm. This gives witness to the readiness of all sectors to accept and embrace elements of informalization as 'old' economic frameworks associated with Fordism give way to new economic trends of deregulation and privatization (Slavnić, 2010). Here an unintended consequence of socio-economic restructuring sees marginalized actors such as migrant labour become the objects of economic exploitation by actors who are already established in the economic system (Sassan, 2009). Migrant employers and equipment suppliers to the car wash sector, for example, benefit from informalization as a proxy for flexibility and innovation but largely on the backs of labour.
Our second contribution focuses on the role of intermediaries. Because UK regulatory standards are light-touch, regulations are frequently disregarded and informal employers are able to cut costs at labour's expense. We argue that migration channels and the role of intermediaries within these channels are partially responsible for the cultural, economic and social vulnerability of precarious migrant workers in HCWs but informally so. Labour market intermediaries in formally regulated employment have changed attitudes and behaviours on both sides of the employment relationship. Employees and employers may be brought together by 'information providers' who broker information on candidates and jobs, for example, search and selection consultants. Similarly, 'matchmakers' provide recruitment and placement services whereas 'administrators' represent professional employment organizations (see Bonet et al., 2013). An approximate application of these categories to HCW networks and related migration channels enables us to build further on studies of marginal and migrant workers that focus on the methods these workers utilize to access employment opportunities.
The presence of intermediaries in the informal labour market for HCW workers illustrates the diffusion of 'sub-contract' capitalism where workers may be unsure who their real employer is (Wills et al., 2009). This uncertainty may provide a buffer between employer and employee. As Ram et al. (2000: 505) first established and Alberti (2016) recently reaffirmed, exclusion from citizenship rights may increase the dependence of informal labour on their employer or intermediaries. However, formalization of employment via a contract of employment may not necessarily mean that all aspects of a job become formalized: for example, recently arrived migrants who have been matched to a job through migratory networks may experience multiple dependency pressures which increase vulnerability. A dependence on their employer, intermediaries too and on the relationship they have with their family in their country of origin may be evident (see Cam, 2013). So connectivity between country of origin home-life and networks can limit labour market progression for migrants (MacKenzie and Forde, 2009). Intermediaries and wider social networks can contribute to the (exclusion and marginalization) difficulties precarious workers face in securing better jobs in either formal or informal employment (McCollum and Findlay, 2015: 430;Williams, 2009). However, the authority of this view is contested and precarious migrants may not be merely passive victims of their position in the market for bad jobs; rather they can use informal networks to promote job mobility and build skills to escape such jobs (Alberti, 2015;Hagan et al., 2011). The new analytical insight we provide in this contribution focuses on the scope of strategic choices intermediaries impose on HCW employers and workers. These choices combine with wider social relations and the reproduction of key labour process characteristics such as de-skilling and managerial control evident in more formally regulated car wash employment.
Our third contribution is methodological: specifically, our use of 'participatory observation', explorative mapping and the innovation we term 'dead-end' interviews and 'dead-ended' access. This method goes beyond studies of migrant workers that choose to not engage directly with migrants but their employers and intermediary recruiters instead (McCollum and Findlay, 2015: 439).

The informal economy and hand car washes
Contemporary capitalism witnesses the presence of unregulated labour markets which are very often populated by vulnerable migrant labour (Likic-Brboric et al., 2013). The informal economy incorporates 'economic' activities outside the arena of the regulated economy which therefore escape official record keeping (Castells and Portes, 1989: 12). Informalization carries particular consequences for the employment relationship and accounts for a significant share of workers who remain outside the world of regulated economic activities and protected employment relationships. Williams (2014) provides a classification of European Union economies by the size and character of informal employment therein. Williams finds that while the UK economy is largely formal (where up to 20% of GDP occurs in the informal economy), 12% of economic activity was generated in the informal economy in 2007. Similarly, Williams (2014) divides informality into informal waged work (23%), envelope wages, that is an undeclared percentage of total remuneration (21%), and informal own-account self-employment (9%) or working for paid favours (47%). HCWs may include informal waged work and elements of own-account selfemployment, particularly work for paid favours, yet also appear to reproduce the labour process associated with more formally regulated car washes. Restructuring the formal economy stimulates and reshapes the informal economy and the social reproduction of capital and precarious labour in both sectors (Slavnić, 2010: 4;Standing, 2014: 155-165).

Precarious workers and the social relations of informal employment
The capitalist mode of production incorporates both formal and informal employment: hence the forces of production, labour and labour power supplemented by the technical rules of labour engagement combine with the social relations of production. These centre on patterns of ownership and control over the means of production, which in the case of HCWs are capitalist because the dynamics of service provision are derived from the accumulation of capital. In turn this accumulation is derived from the creation of surplus value. So the economic and social behaviour of the actors in HCWs may derive from the social relations of production but it is necessary too to recognize that these relations of production may also extend to social relations in households and communities of workers. Therefore we cannot argue that all HCWs are capitalist. Rather some may represent simple commodity production within a broader capitalist dynamic and an associated social formation; that is, we do not fully analyse the potential differences that may arise from self-employment and employment. It follows from this that precarious remuneration standards for informal workers and the associated vulnerable working conditions and denial of entitlements cannot be effectively measured and understood in the light of statutory interventions such as employment protection legislation and the national minimum wage or formal stakeholders such as trade unions and the Low Pay Commission. Rather, the presence of wider social relations of production and embeddedness in social networks in particular informal sectors are more likely to inform the acceptance of low pay and poor working conditions. (See also Ram et al. [2007] on informal work in small firms generally and Hammer et al. [2015] on the re-shoring of garment manufacturing to the East Midlands of England.) We argue it is necessary too to recognize that, notwithstanding the theoretical and empirical importance of social relations beyond the means of production, the balance of evaluation must shift from the internal properties of groups towards wider economic and social contexts within which informal employers and migrants are located (see Kloosterman et al., 1999: 257;Ram et al., 2011: 339). Migrant workers provide hyperflexible labour, working under many types of arrangements but not always in 'employment' (Anderson, 2010). So while intermediaries may disseminate business support information for employers and workers they also contribute to the reproduction of precarious employment because intermediary networks are embedded within the wider social relations of production and regulation which surround the labour market for HCWs. As we indicate above, these wider social relations can be capitalist or focus on simple commodity production.
A focus on the broader political economy, the social relations of production and associated HCW labour process avoids the simplicity of proximate explanations -explanations which themselves need explaining. These approximations abstract and separate low-cost, informal recruitment-associated low pay and flexibility in marginal business survival from more universal capitalist processes which create economic and social marginalization. In so doing they present ethnically inspired paternalism as a local solution to conflict-bound exploitation in the labour process. For further discussion on this point of theoretical and methodological detail, see Jones and Ram (2010: 165-167) and Bloch and McKay (2015: 40).

The hand car wash 'sector'
This section provides the background context on the sector and its growth and diffusion. It explains our research methods and methodological contribution and the extent to which we draw on 'stakeholder voices'. We also summarize emerging themes from the extant American literature.

The context, growth and diffusion of the HCW sector
The only national survey of the car wash sector was recently produced by the Car Wash Association (CWA, 2014). The study reported that in 2013 there were approximately 18,500 car wash operators in the UK, of which 7000 (38%) were mechanized conveyor washers, rollovers and jet washes. The CWA argues that the latter constitute the formally regulated sector. So according to the CWA, 60% of the market or over 10,000 operators constitute informal HCW enterprises, which their study divides into four sub-groups. First, there are approximately 280 trolley wash businesses operating in supermarket car parks and city centre car parks. Approximately 600 HCWs operate in redundant rollover bays, whereas at least 2000 HCWs operate on former petrol station forecourts. Lastly, a further 7500 HCWs operate out of former or still open public house car parks, former tyre provider outlets or on waste ground. So the presence of HCWs has grown significantly from virtually none in 2004. In 2013-2014 the estimated income of broadly defined informal HCWs was £449,550,000, nearly three times that of the regulated sector at £158,929,750. On the basis of our research on Leicester, and more recently Nottingham, these figures while useful are likely to be overestimates provided as they are by a key stakeholder in the sector. The UK has 69 cities. Scotland has its own environmental regulations for car washing (SEPA/EA, 2007); discounting the seven Scottish cities, 62 cities in England, Wales and Northern Ireland would mean that on average each contained 161 HCWs. This is unlikely; when we extended HCW mapping into Leicestershire and Nottinghamshire we found only another 60 HCW sites. It follows too therefore that the CWA revenue figures may also be overestimations.
Economic restructuring in retailing combined with the financial crisis and the presence of precarious workers, many of whom are migrants, creates a potential niche for HCWs. Prospective HCW operators need only acquire an open property lease to set up a car wash on a former petrol station forecourt, public house car park or former supermarket parking area. Internet guides for HCW start-ups stress the low set-up costs and low operating costs compared to automated car washes but emphasize the labour-intensive nature of the business. Such labour intensity means paying lawful wages is likely to be prohibitive to profitability. Internet guides, for example, the Business Index (www.thebusinessindex.com/guides-car-wash-businesses), Start-Ups UK (www.start-ups.co.uk/ forum/topics) and the UK Business Forum (www.ukbusinessforum.co.uk), however, state that customers will pay a premium for a hand car wash even though they may be trading down from automated car washes in the context of recession since 2007 (see Business Index, 2014). Lastly, the regulatory environment is permissive: electrical, plumbing, water and ground works regulations are minimal and often not enforced. While useful and easily accessible, internet sources such as these and HCW websites are often unreliable: for example on investigation we found the links do not always work and often only contain discussion bulletin boards from prospective business owners. One feature, however, which most sites report on are the swathes of land available for HCW sites and the relative ease of acquiring a lease on such sites.
The American literature examines the more extensive diffusion of HCWs in three large cities, Chicago, Los Angeles and New York. Theodore et al. (2012: 210-214) track labour standards in low-wage US industries in the three cities and report significant wage theft violations from car wash workers. A third of workers in the sector are paid less than the legal minimum wage for tipped workers, and many suffer underpayment of more than $1 per hour, where wage theft by employers translates into 15% of earnings. Workers Aligned for a Sustainable and Healthy New York (WASH, 2012) reports that migrant car wash workers are subject to wage and working time theft and suffer unprotected exposure to hazardous chemicals. The Office of the Attorney General in California (OAG) estimated that in 2010 there were 1500 HCWs in the state employing 28,000 workers, where a third of the units were unlicensed (OAG, 2012). In 2010, the AFL-CIO reported 10,000 car wash workers in LA county and estimates suggest that 27% of these were undocumented (AFL-CIO, 2010). In 2008 the Community Labor Environment Action Network (CLEAN) was established by the car wash organizing committee of the US Steel Workers aiming to secure the right to organize the car wash sector (Tapia et al., 2014). CLEAN operates within and beyond collective bargaining to enforce labour standards and highlight examples of non-compliance in wages, terms and conditions and health and safety (Narro, 2007(Narro, , 2009. This commitment has however secured little success but demonstrates an effort to deter the exploitation of vulnerable migrant labour and the division of working people between the formal and informal sectors. It follows from this that in the United States union strategy reaches further than the organizing model towards that of a 'countermovement' based on social justice and fairness beyond formal industrial relations (Tapia et al., 2014: 14-18). We found little of this in the UK; at present there is no evidence of a countermovement of this type. In England, Wales and Northern Ireland we found only two campaigns (in Coventry and Luton) designed to secure effective regulation of HCWs, which we report on in the findings section.
A finding common to the American and British literature on precarious vulnerable workers is, however, the ability of employers to continue to violate employment regulations and labour laws. In the USA many workers who are forced to accept informal employment feel unable to seek recourse to regulatory agencies. Indeed, the absence of regulatory recourse is so extensive in nail bars, HCWs and restaurants that in New York City violations of employment and labour laws constitute the dominant business strategy, putting pressure on responsible employers to follow suit in order to compete (Bernhardt et al., 2008). The lessons for this study from the American literature are that in the presence of a sector dominated by informalization employers are effectively insulated from regulatory compliance because worker grievances are unlikely to be mobilized into activism in the absence of a countermovement which operates beyond collective bargaining.

Research methods and emerging themes
Research in this area poses ethical and practical difficulties: employers who choose not to comply with regulations are unlikely to be predisposed towards participation in research, while workers are in particularly vulnerable positions. Flexibility and some opportunism in methodology is an established approach in fieldwork of this kind and our empirical focus required that we adopt similar techniques (for a discussion of methodological issues, see Bernhardt et al., 2008). We began by 'mapping' the location of car wash sites within the city boundaries, using direct observation supported by content analysis of business directories, internet and social media sites.
Our mapping exercise contained two components: first, customer participation in and observation of HCW sites and second, an effort to position the labour process more directly within the social relations which surround HCWs and trolley washes. Here we utilized what Alberti (2015: 865, 870) in her study of precarious migrants in hospitality jobs terms 'participatory research'. In our study participatory research combines using HCWs, direct observation, confronting and engaging participants and semi-structured interviews to map, observe and then describe and identify the labour process in 20 HCWs in Leicester. Semi-structured interviews allowed us to follow our participants in terms of their responses and then return to the structure. We found that two of the HCWs were formally regulated and long-established firms offering mechanized car washing with some hand finishing and drying. Both outlets were members of national or regional chains. Of the 18 remaining sites we secured good interview access with workers in six outlets.
Our initial contact with these HCW workers led to the use of another established method in researching difficult-to-access workers, namely 'snowball' referral, which is a form of non-probability sampling where existing subjects recruit further potential subjects from their friendship and associated work groups (see Hagan et al. [2011] for similar application to migrant workers). Our use of snowballing led to a series of direct conversational interviews with workers other than our initial contacts on six further HCW sites, five on redundant business premises and one trolley car wash at a large supermarket. Application of participatory research combined with snowball sampling enabled us to appreciate what Però (2014Però ( : 1160 terms 'the lived experiences', 'points of view' and context of migrant workers. In total, we secured 20 complete interviews on 12 sites, 17 with workers and three with business owner-workers.
It was impractical to record the interviews as they were held in outdoor/noisy locations such as bus stops, cafes, car parks, public houses, petrol forecourts, or walking with interviewees on their way home, but in each case simultaneous notes were made and then re-copied. These interviews were up to an hour in length. All interviews were conducted in English with some translation by other HCW workers. Because of the ethical sensitivity of our research the 18 car washes are not identified and workers' testimony is not connected to particular HCWs. We are conscious of the potential of exposing fieldwork contacts or drawing the attention of taxation authorities to places which may not be paying taxes appropriately.
More formal semi-structured interviews were held with three sector stakeholders: the commercial manager of the Petrol Retailers Association (PRA), the deputy chairman of the Car Wash Association (CWA) and the CEO of a nationally recognized trade supplier of products and equipment to mechanized car washes and the HCW sector. A further three interviews were conducted with local authority regulators in Leicester, Coventry and Luton. Telephone interviews were held with government and private sector enforcement agencies, for example the Border Agency, Environment Agency and enforcement officers in Coventry and Luton and Severn Trent Water. Interviews were recorded and transcribed, or if they were telephone interviews notes were made then transcribed.
In total we secured 40 interviews: 20 interviews at HCWs, nine full interviews with sector stakeholders and 11 partial 'dead-end' interviews. We started mapping in spring 2014 and this and all interviews were completed by March 2015. The 18 HCWs were in business throughout this period, some however have changed their name whereas others do not have names listed on their signage or utilize generic names, e.g. Name of District Car Wash. From our mapping we estimate that at least another 10 HCW sites have come and gone during the research period; further still, more detailed mapping which involved driving up and down all streets in a particular district revealed the presence of part-time pop-up HCWs, some of which too have been abandoned. To theorize this in terms of Williams (2014), our 18 more permanent HCWs combine informal waged work and informal own-account self-employment whereas pop-up HCWs appear as more opportunistic own-account self-employment. So while our locations and related findings are representative of HCWs across the city, our research does not present a statistically representative sample of HCWs but rather a purposive sample which aims to examine social processes (McDowell et al., 2014: 860). Given the presence of pop-up outlets and the widespread use of hidden HCWs, for example, those sometimes present in pub car parks, or those which appear on match days near football and rugby grounds, it is not feasible to generate a statistically representative sample of HCWs.

Findings
We report our findings from interviews with sector stakeholders, regulatory bodies and HCW owners and workers.

Car wash sites
We checked the legitimacy of business ownership via lease ownership or property owners and by direct questions on site and found that the majority of HCW sites were rented but three were owned by operators. We found that all 18 HCWs were single-site establishments and not part of a branded chain; however, one of the three owner-workers revealed that he was involved in the ownership of three sites in Leicester: The sites including those of this owner-worker each employed fewer than 20 workers and all exhibited price lists, some more detailed than others; however, the services available were in the main of two types, outside and inside cleans or outside cleans only for standard cars. Similarly we found that all sites displayed the availability of credit card payment yet in four cases on presentation of a credit card we were asked to pay by cash. We observed that the presence of appropriate work wear, high-visibility jackets and waterproof footwear was variable, with some operatives wearing high-visibility jackets and others not. The same was so for waterproof footwear and trousers. We found HCW operatives were highly gendered with only one site employing women; we were able to interview one of these two female workers. It was clear that many workers could speak little English other than scripted speech, for example: 'Forward please' [beckoning driver to move so underside of alloy wheels can be sprayed] and 'outside or inside with outside' [i.e. inside and outside or only outside clean]. (encounter with HCW operative on-site) Only one HCW displayed a website address, which on enquiry routed through to a social network site for car washes in Leicester. None of the washes we used offered a receipt, even though two had tills, and asking for a receipt aroused immediate suspicion.
Our participatory method combined with interviews revealed that these sites were now in effect embedded with a regular customer base (we found queues on Saturday and Sunday afternoons). HCWs offered similar prices: £6 for 'outside service' and £12 for an inside and outside service. In contrast to the claims of business start-up guides, we saw no evidence that consumers would pay a premium for an HCW, that is, they are less expensive than mechanized washes. Across the 18 HCWs all the on-site workers identified themselves as Bulgarian or Romanian. So in terms of EU freedom of movement rules for employment in 2015 these workers have full employment status in the UK. In interviews, however, eight HCW workers revealed that they had been in the UK for more than two years, potentially therefore as far back as February 2012.
Trolley washes on supermarket car parks are more professional in appearance if not operation. Trolley wash workers had professionally manufactured base equipment trolleys and they wore liveried uniforms, high-visibility jackets and waterproofs. But professional looking trolley washing is weather dependent, whereas HCWs under forecourt or rollover cover are able to work regularly in terms of hours. Trolley wash 'professionalism' was also exposed by team members evidently struggling with 5 gallon/22 litre water barrels, that is, they carried their own water with them. The supermarket manager, on whose car park the trolley wash operated, told us: It is the responsibility of the operators to conform to environmental and employment regulations -we assume that all our contractors are compliant. (Manager, national supermarket, March 2015)

Stakeholders
Interview findings from sector stakeholders are forthright, as one might expect where the backdrop to their views is that HCWs operating informally are undermining a legitimate business sector. For the PRA, who represent independent petrol retailers, this is a serious issue. A mechanized car wash costs around £40,000 to install and requires regular maintenance. Many PRA members are losing significant amounts of money on these units (revenue figures for such units have halved since 2007), effectively finding themselves in a situation of 'negative equity' and therefore unable to sell a unit other than at a loss or close a unit and bear the loss.
A study commissioned by the CWA reports that mechanized car washes have up to three HCWs within two miles of any site (CWA, 2013). Our research confirmed this was the case in Leicester. Therefore the CWA is similarly concerned, but while they recognize the presence of migrant labour, which might operate informally, the main concern of the deputy chairman was the failure of HCWs to comply with environmental regulations over the disposal of water: Failure to comply with these regulations puts legitimate operators at a significant cost disadvantage. (Deputy chairman CWA, January 2015) The CWA consequently lobbies for better regulation of water use by HCWs and appropriate charging for water by water companies. The deputy chairman of the CWA and the commercial manager of the PRA both indicated that HCWs have been a competitive threat since 2007, with the CWA deputy chairman arguing that they (HCWs) will effectively … put mechanized car wash chains such as 'IMO' and 'Shiny' out of business in the next few years. Many of our operators have already been forced to reduce prices and seek to add value to their services by the provision of 'deep clean' car valeting on a contract basis. (Deputy chairman CWA, January 2015) Peter, the CEO of a leading Midlands trade supplier to the car wash sector, confirmed that 2007 was when he began to see a change in the nature of the sector. HCW operators in his experience are: … in the main Romanian entrepreneurs who own chains of HCWs with whom I deal with on a regular basis. The terms of business they prefer are 'cash only, off invoice'. As a legitimate business I have to complete invoices after dealing with these operators to keep my own accounts straight. (CEO trade supplier, February 2015) Peter went on to tell us that while HCW operators always pay they are effectively changing the rules of the sector by the manner in which they use the equipment. Pumps and spray equipment are bought without service agreements or training use guides and 'Totalled', they run them into the ground and return them for repair even when the repair will last only a short time because compressors and rubber housings are damaged and worn effectively beyond repair.
Although a stakeholder in the sector, Peter expressed a different view to the PRA and CWA on the trajectory of the sector: I have seen several informal HCWs 'go legitimate', this is a likely way forward because they operate as family and networked businesses which don't employ people in the manner I understand employment.
Peter then speculated that: … a tipping point may soon be reached where Romanians will try to buy out legitimate operators and then run them through their own networks on a local basis -a kind of informal franchise.

Regulators, 'dead-ended' access and interviews
Our interview with an enforcement officer in Coventry revealed that in contrast to Leicester, which is only 20 miles away, all visible HCW sites in Coventry were investigated in 2014, with non-compliant informal HCWs shut down on a regular basis and owners appropriately fined for breaches of chemical handling regulations, failure to manage wash water correctly and breaches of employment regulations. This finding was corroborated by material passed to us by our interviewee in the Environment Agency (EA, 2014). Coventry City Council now prohibits HCWs in supermarket car parks, but the enforcement officer conceded that prohibition did not necessarily terminate HCWs so enforcement remains a contested issue.
Our interview contacts in Coventry 'snowballed' us to another enforcement campaign which provided us with more comparative regulatory material. Luton Borough Council's hand car wash project has more weight. Local authority inspectors and the Health and Safety Executive inspect HCW sites and are authorized to take on-the-spot enforcement action. Our interview with an enforcement officer in Luton revealed that the Council aims as a matter of public policy to enforce health and safety laws, certificated evidence of compulsory employer liability insurance, planning permission and correct and appropriate discharge of contaminated water. After initial success locally in 2013-2014, the officer reported the project had spread across 23 local authorities in Bedfordshire, Hertfordshire, Cambridgeshire, Suffolk and Essex. These state-sponsored enforcement campaigns do not mean that informal HCWs are not present in Coventry, Luton and elsewhere, rather that it is possible to mount and sustain enforcement campaigns. (We found three informal HCWs in Coventry which appeared similarly embedded like those we found in Leicester.) We also conducted interviews with the Border Agency, the enforcement team at Leicester City Council, Severn Trent Water and nationally known supermarkets on whose car parks trolley washes have operated for some years. Nationwide, of the large national supermarket chains only Tesco told us that it prohibited all sub-contracted trolley washing in their car parks, replacing these with Tesco HCWs; our interview with the PRA subsequently confirmed this finding (PRA, 2014). We also visited two Tesco sites in Leicester (Wigston and New Parks) and found Tesco HCWs present on site. We were unable to secure a complete interview with the Border Agency without providing full details of what we were undertaking and the intended location of the study on the Agency website contact page. The enforcement officer in Coventry suggested to us that revealing such material might lead to one or two pre-emptive raids. Ethically we were unable to do this for fear of facilitating exposure of our fieldwork respondents, so we desisted from following this line of enquiry. Significantly, private sector firms with legally enforceable regulatory capture such as Severn Trent Water, supermarket managers and some leaseholders were candid in their responses. The following was typical of these responses: … we don't want to get into this. (Severn Trent Water, November 2014) Our conversation was then dead-ended. Regulatory non-compliance such as this legitimizes the exposure of workers to dangerous working conditions both physically and in terms of exposure to chemicals, fumes and vapour. Ordinarily these attempts would not be cited as findings but interviews which were dead-ended are methodologically significant factors in the political economy of informal work and employment. Dead-ended interviews are not just denials of access from which researchers move on but rather they are a source of contextual information in themselves. That is, it is more likely to be the case that dead-ends relate less to how busy a hoped for interview contact is but that these contacts prefer not to have probing questions put to them.

HCW workers and the HCW labour process
We report findings on how HCW workers perform work, who controls the HCW labour process, what if any is the skill base and how HCW workers are paid. Road-side HCWs have up to 12 workers operating on-site at any time. Crew members circulate around tasks from spraying on wheel cleaner to remove engine oil, soaping and hand buffing cars, rinsing and shining windows, to vacuuming and wheel and trim coating. Supermarket trolley washes have two or three operatives who work as competitive teams. Team members approach shoppers as they park and hawk for work. On one large supermarket car park we found four teams of three workers operating on a car park with 725 spaces, approximately 181 spaces per team. The most favoured pitch was the parking aisles in front of the store as footfall was greatest there. This pitch was secured competitively and involved … getting here and setting-up as early as possible. (Trolley wash worker, August 2014) Trolley wash workers are much faster than HCW workers in former petrol stations and they explained this to us in two ways; on the one hand it is necessary to: Get the 'wash complete' before it rains or before the driver returns.
Other trolley wash workers, on the other hand, emphasized their skill and time served, where they had invariably learnt how to wash cars on petrol station sites: I am now part way up the hill [hierarchy] as self-employed.
To us, though, the model trolley wash workers operate within would be more appropriately compared to the sharecropping model in American agriculture in the 1930s where a land owner -here the supermarket -effectively appropriates a portion of their income. In terms of the HCW labour process and the conditions in which work was performed we found relative formality evident in many sites, where signage was clear and often presented professionally; workers wore protective gear which was often liveried (though this was not universally the case); and aspects of the labour process were very clearly demarcated and organized where direct and indirect controls were in evidence.
In road-side HCWs we found team leaders who controlled the pace of work and the pressurized water jet and directed customers on the basis of either a full (inside and outside) or outside service. In addition an overseer manages the money where individual operatives ask for change for customers if necessary, handing over the money to the overseer. Any tips were also given to the overseer. The presence of multi-role teamworking established that the labour process in HCWs was not amateur or disorganized but highly structured and standardized both within and across sites, punctuated by scripted speech (inside or outside service) and simple and technical controls, for example centralized control of money and tips. When charges for outside washes increased from £5 to £6 we noticed that if presented with a £10 note the operative would always return with two £2 coins, leaving the customer with the dilemma of no tip or a £2 tip (33%). Less direct controls over labour resulted from family and network intermediaries. Several interviewees stated that they chose to do HCW work, and typical of these responses was: … my family and friends in X [country of origin] guaranteed me this work I came as a student and stayed on, I am now full EU and do this work when not at X [name of local college].
In summary, HCW operatives worked in teams under the direction of a designated team leader. We found operatives who had previously worked professionally as hairdressers, front of house hotel staff and teachers in their country of origin but who chose HCW work, which they understood as de-skilled and degrading, to adjust to a new environment and improve their English. In contrast to this, rather like Janta (2011), we found that some new arrivals with little or no English and few or no qualifications and no vocational skills relied on what Bonet et al. (2013) term ethnic niche experts, either as information providers or matchmakers, to secure jobs. Here a particularly interesting finding was revealed to us by one of the few female workers we encountered: I was lucky [country of origin entrepreneurs] used X [social media] and on-line matchmakers to recruit me. I now know that I am (what you call) a 'no language no skill no personal network' because of this I have to stay longer to learn and pay passage for board and living. I could move faster if I am willing to do other work.
So the skill of HCW operatives is measured by the ability of team members to multi-task and by the speed with which they can work. This sounds obvious but operatives need to avoid getting in the way of each other and we observed the law of diminishing marginal returns come into play if more than four operatives work on a car at once. We observed that during periods of high-traffic all workers worked on cars at the front of the queue. This further speeded up washing times, the fastest we recorded was 5 minutes where 8 minutes was an average time. How HCW workers are paid varies.
Discussion of wages and terms and conditions proved difficult in all interviews and resulted in the termination of two interviews where the subjects were convinced we were Border Agency inspectors. Separately eight interviewees agreed that their wages were well below the per hour national minimum wage which we quoted to them but several argued that this did not matter to them. Typical of this response is the following: Work here cannot be understood in terms of your [UK] standards as I told you I work in an extended [country of origin] family. In these networks my family and friends help the family of [business owner] in [country of origin] in return for assistance with entry into English labour market.
For up to a nine-hour day HCW workers earn £50, which for more than half our interviewees is supplemented by housing benefit. The trolley wash teams who operated in supermarket car parks earned much more than this and we observed them over an hour securing five wash deals at £6 per wash. Due to 'dead-ending' we were unable to secure information on what rental trolley wash teams paid the HCW owner or intermediary who had secured the sub-contract from the supermarket. The supermarket manager too declined to discuss this issue with us. So operatives at road-side HCWs experience wage theft per week of approximately £42.50 per week, in comparison to the national minimum wage for 2014 (9 hours @ £50 per day @ five days = £250, however, 9 hours at adult minimum wage in 2014 = £58.50 @ five days = £292.50; this minimum wage theft of £42.50 equates to 14% of the equivalent national minimum wage). We discerned that operatives were less concerned about this but did contest the levels of work intensity they experienced. Trolley wash workers were less exploited because they consider themselves to be self-employed but £30 per hour for a nine-hour day is £270 which is only £90 for three-person teams or £67.50 per day for four-person teams. This is, though, a significant mark-up on road-side HCW remuneration.
There are limitations to our research. First, our aim is confined to providing qualitative information on the presence and size of the HCW sector. We have established it is fluid and dynamic and because of this and for reasons of participant vulnerability we were sometimes unable to secure and at other times chose not to present quantitative information on the sector and its participants. A second limitation relates to our notes from interviews. Here we made every attempt to rule out any possible bias by presenting our view of our findings to fieldwork participants and rechecking specific points raised by participants through repeat observation of research sites.

Discussion and conclusions
What emerges from our research is that the economic and social behaviour of actors in informal HCWs is embedded in the social relations of informalization and possibly within these relations a series of embedded dependency relationships in the labour process. We present our discussion and conclusion around the three analytical insights which our contribution makes to the study of informal migrant employment. First is the cocreation of the HCW sector and pressures for economic restructuring from above and below. Informal employment in this sector is fluid but the technical rules of production, the land resources and the instruments of labour and labour power which represent the forces of production in HCWs are formally constituted but informally regulated. HCWs perform a lawful activity and we found that many acted lawfully in that they possessed leases which enabled them to run a business on the premises where they operate. Environmental and employment regulations, however, were not applied. The extent to which HCWs constitute a formal sector is one subject to very rapid and deeply ingrained pressures towards informalization. This is manifest both in terms of the structure of the sector and businesses within it and the implications this carries for work and workers. In terms of work, employment is expanding as a consequence of de-mechanization but in highly contingent ways. The overall consequence in labour process terms is undoubtedly to increase the rate of exploitation. However, the experience of exploitation is complex and fluid. Workers may endure it so long as they believe their employers have limited options to behave otherwise and that the work provides a stepping stone into more solid employment. In terms of businesses, the structure of the sector is fragmenting with a greater degree of business activity channelled through sub-contracting where sites are effectively run on a franchise model.
Technical rules combine with the social relations of production in HCWs which are principally defined by the ownership and control of the means of production. Kinship is an integral component within the labour process and the broader social relations of production in HCWs, where enclave migrant labour provides a competitive advantage but at the cost of pronounced labour exploitation and long hours. Therein networks create dependence and also are part of the broader political economy of Leicester's HCWs and associated social relations of production. The 'from above' and 'from below' drivers of economic restructuring inform the diffusion of HCWs, where HCW owners, some of whom are also migrants from the same country of origin as the migrant workers they employ, operate low-road, low-cost business models. As such they make a strategic choice to take advantage of exploitative labour practices to gain competitive advantage over mechanized car washes where a low-cost business model by HCWs dictates the competition to usurp higher productivity mechanized car washing.
Our second contribution is on the role of intermediaries. The employment relationship in HCWs in Leicester is subject to degrees of informalization where the broader social relations of production are controlled by intermediaries. Intermediaries operate within and without personal networks to create stratification in the exploitation of migrants. We found that 'documented status' for migrant workers is less important than the network status of workers. Employment of what we term 'no-network' migrants in their country of origin can lock migrants into particular sectors such as HCWs. Here our findings are consistent with the arguments of Bonet et al. (2013) and Janta (2011) that migrant entry to the labour market is regulated less on the basis of an employment relationship and more on the presence or not of extended networks.
Here we provide an analytical insight which moves the established stock of research further forward, albeit in a small study where there are some previously undocumented migrants who are now documented. This status reduces the dependence of these workers on their employer where their work in HCWs represents a potential entry point into the UK labour market; however we found only a few of these workers, in large measure because of the newness of HCWs. The documented status of migrants now encourages HCW owners to continue to employ migrants but those who are not part of their established family and social networks. Employers aim to raise the dependence of these migrants on them. In turn this acts as a retention device where we found the presence of recently arrived migrants who were more, rather than less, dependent on their employer. Relatively settled migrants and more recent migrants expressed disdain for UK institutions and patterns of regulatory intervention and had no interest in trade unions and are unlikely to seek regulatory recourse because they are embedded in what they feel are positive dependency networks confined to or beyond employers. These findings build on MacKenzie and Forde's (2009) work on the inter-connectivity of home-life, work and ethnic networks and how they inhibit the progress of migrant labour by locking in workers to a sector and associated forms of exploitation. Therefore in HCWs migrant labour, 'documented' or 'undocumented', networked or otherwise, comply with imposed work discipline due to a lack of choice of immediately available alternative employment. The relatively recent advent of 'documented status' for Bulgarian and Romanian migrants results in more choice for settled workers, but may reinforce the acceptance of long working hours, dirty and unsafe working conditions and underpayment for more recent arrivals with fewer network contacts and skills. Our findings unpack marginal status to reveal direct and indirect labour control strategies and associated social relations that organize informal labour in HCWs where employees and employers co-produce a marginal collusive business model and associated labour process.
In terms of our third contribution, on methods, we found HCWs hard to research but not impossible. Rather this research requires the deployment of a variety of methods, some more established than others. More interestingly, we found that HCW workers and some HCW owners were willing to participate in this research and that both showed disdain for campaigns or regulatory compliance and none had any interest in them. We suggest too that HCW sites will be difficult to organize because of the small-scale singlesite nature of HCW employment, which contrasts strongly with the more successful UK-based campaigns to organize migrant workers within more formal patterns of institutional regulation. In addition to this, the presence of established patterns of 'complicit' exploitation is unlikely to mobilize into campaigns of resistance.
In our city-wide study we found a political economy where economic and public policy shifts, have, over the past 10 years, acted as drivers and enablers for the spread of HCWs. Neoliberal economic management, apparently light-touch regulation, the financial crisis, austerity and the diffusion of migrant labour combine to lay the ground for the emergence of lawful businesses employing precarious and vulnerable informal migrant labour.
Essentially these enablers and drivers encourage market creation and informal employment which in turn, drives, grounds and legitimizes deregulation that extends beyond the formal political economy. Here capital, labour, product and in this case service markets emerge and interact to intensify price competition and the UK's low-cost, low-wage economic trajectory, and lead us to three conclusions.
First, on restructuring and the fluidity of HCWs. The labour intensity in HCWs has the potential to make redundant technologies designed to replace labour in the form of mechanized car washes -a regression where capital reduction rather than capital increase makes sense because labour is so cheap. The diffusion of HCWs has undermined the utility and viability of mechanized rollover car washes and it is not unusual to see such units redundant on petrol station forecourts alongside a hand car wash facility or indeed housing a HCW facility. Here exploitation in wage levels and the labour process is reproduced on the basis of (low-cost) migrant labour power. Hence newish small-scale lowvalue-added service providers who operate informally have the potential to be a transitional mode which not only mimics the formal sector but may gradually evolve to become the formal sector. So HCWs appear to play a role in changing the occupational structure of the car wash sector from formal to informal employment, while they 'go legitimate' in terms of business formation. This demonstrates that the informal economy is not a distinct proximate sector, rather it is more appropriately understood in terms of capitalist dynamics which make informalization possible. 'From above' dynamics are evident in the economic restructuring of petrol retailing and the socio-cultural restructuring of alcohol consumption. These dynamics combine to create a 'from below' stimulus to informalization as an unintended consequence of 'from above' economic restructuring, where marginalized actors use land resources such as petrol filling stations and public house car parks which are located in the formal economy in a flexible and innovative manner. Informal low-cost, low-profit HCWs substitute labour for capital and satisfy a growing demand for low-wage work.
Our second conclusion relates to the role of intermediaries. Common nationality of owners and workers does not inhibit control of the labour process and work organization, which mimics that of formally regulated car washes. That is, HCW employers and HCW workers co-construct the sector because workers accept their exploitation while employers accept the low margins that informal employment sustains. By association, family or social networks where intermediaries operate are a focus for exploitation and are intimately connected with the social relations of production rather than a replacement for them.
Third, with regard to methodology, our analytical insight poses questions for regulators because the visibility of HCW sites is very high yet regulatory capture appears very poor. Our dead-ended access and interviews -the 'we don't want to get into this' type response -are not just 'so what' responses. Rather, they reveal something about the institutional framework of the UK's political economy where regulators and those with enforcement powers turn a blind eye to informal privatized unregulated business and employment practice.
Our conclusions on Leicester find that many well-established HCWs retain informal labour but edge towards legitimacy in terms of business formalization. This argument resonates with some of Peter's observations that HCW operators go legitimate if they can. The embedded nature of the UK's broader political economy suggests that formal firms are likely to import and sustain informal employment practices. Our argument here echoes further points raised by Peter, where in his words 'go legitimate' excludes formalized employment practice.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: the Centre for Sustainable Work and Employment Futures is funded by the ESRC.