Refiguring the nature of undeclared work: Some evidence from England

In the European Union, the issue of tackling undeclared work is now at the top of the policy agenda. In 2002 at the Lisbon Summit, the European Council listed tackling undeclared work as one its top ten priorities for action (European Commission 2003). Throughout the European Union and beyond, therefore, public policy-makers are busily engaged in considering what can be done to eradicate such work. In this paper, we wish to contribute to this process by using some evidence from England to re-read the nature of undeclared work and then considering its implications for public policy. Until now, that is, most studies of undeclared work have focused upon assessing the volume of undeclared work and how this varies socio-spatially (e.g. Dallago 1991; Williams and Windebank 1998). The nature of such work, meanwhile, has been seldom subject to detailed empirical investigation. Instead, the widespread and recurring assumption has been that undeclared work is market-like activity conducted for the purpose of monetary gain. As Travers (2002: 2) so succinctly puts it, ‘most research on [undeclared work] gives short shrift to the motivations of people to do this work. It is usually said that people do the work to earn extra money and left at that’.

Yet such a ‘thin’ reading of the economic relations and motives underpinning undeclared monetary transactions lies in stark contrast to the study of monetary exchange more generally. In recent years, the narrative that monetary exchange is always market-like and motivated by monetary gain that ran deep in both conventional economistic discourses ranging from the neo-classical to the Marxian variety (see Jessop 2002) and the ‘formalist’ anthropological tradition that viewed exchange mechanisms in contemporary societies as less ‘embedded’, thinner, less loaded with social meaning and less symbolic than those in pre-industrial societies (see Mauss 1966), has come under increasing critical scrutiny (e.g. Gibson-Graham 1996; Carrier 1997; Crang 1997; Byrne et al.1998; Crewe and Gregson 1998; Crewe 2000; Lee 2000; Gibson-Graham and Ruccio 2001; Kovel 2002). Recognising that ‘the major defect of such market-based models of exchange is simply that they do not convey the richness and messiness of the exchange experience’ (Crewe and Gregson 1998: 41), ‘thicker’ descriptions have been sought in which monetary transactions are read in terms of social norms and values and their social, cultural and geographical embeddedness (e.g., Davis 1992; Zelizer 1994; Crang 1996; Lee 1996, 2000; Crewe and Gregson 1998; Bourdieu 2001; Community Economies Collective 2001; Slater and Tonkiss 2001; Comelieau 2002).

Until now, however, undeclared work has remained relatively untouched by this wider refiguring of monetary exchange. The aim of this paper, therefore, is to start to rethink the nature of undeclared work so as to open it up to re-signification. To do this, firstly, this paper will review the literature on undeclared work to show how a ‘thin’ reading of such work as market-like and motivated by monetary gain widely prevails. Reporting evidence from 861 face-to-face interviews conducted in English localities between 1997 and 2001, an alternative reading of undeclared work is then provided more appreciative of the heterogeneous economic relations and motivations involved. Unpacking the culturally, socially, and geographically embedded meanings of undeclared work, this will reveal that to view undeclared work as profit-motivated and market-like is to read it more from the standpoint of the employed, men and those living in affluent areas than through the lens of the unemployed, women and people living in deprived areas. Until now, based on the ‘thin’ reading that undeclared work is universally market-like and conducted for the purpose of economic gain, and thus that participants make a rational economic decision to engage in such work, the predominant public policy response has been to pursue its deterrence by changing the cost–benefit calculation. The final section of this paper, however, argues that this ‘thicker’ description of undeclared work means that the current policy approach towards undeclared work needs to be reconsidered.

At the outset, however, it is first necessary to define what is meant by ‘undeclared work’ in this paper, or what is sometimes referred to as ‘cash-in-hand work’, ‘the underground sector’ and ‘informal employment’ (Thomas 1988; Williams and Windebank 1998). It is important to state that the definition adopted here does not differ to the widely accepted interpretation of such work. Following the vast majority of the literature, such work is here defined as the paid production and sale of goods and services that are unregistered by, or hidden from the state for tax and/or welfare purposes but which are legal in all other respects (Thomas 1992; Portes 1994; European Commission 1998; Williams and Windebank 1998, 2001a,b; Feige 1999). It thus includes only paid work that is illegal because of its non-declaration to the state for tax and/or social security purposes. Unpaid work is not included and nor is paid work in which the good and/or service itself is illegal (e.g., drug trafficking). In other words, undeclared work includes only paid work where the means, rather than the ends, are illegitimate.

Casting an eye over the literature on undeclared work, it quickly becomes apparent that the vast majority focuses upon assessing the magnitude of undeclared work and how this varies socially (e.g., Leonard 1994, 1998; Pahl 1984) and spatially (e.g., Renooy 1990; Dallago 1991; Williams and Windebank 1998; International Labour Office 2002). To greater or lesser degrees, the starting point of much of this work is the ‘marginality thesis’ that views undeclared work as concentrated in marginalised populations (e.g., Castells and Portes 1989; De Soto 1989) in that most studies ultimately do little more than seek to either corroborate or falsify this thesis. With the exception of the cross-national comparative studies of economists using indirect measurement methods that seek statistical traces of undeclared work in data collected for other purposes (e.g., Schneider 2001; International Labour Office 2002), the predominant finding has been that the marginality thesis is invalid. Undeclared work has been usually found to chiefly benefit those already in employment (e.g., Van Geuns et al.1987; Barthe 1988; Renooy, 1990; Mingione 1991; Mingione and Morlicchio 1993; Jensen et al.1995; Fortin et al.1996), affluent rather than deprived localities and regions (e.g., Renooy 1990; Mingione 1991; Williams and Windebank 2002) and men rather than women (e.g., Van Eck and Kazemeier 1989; Renooy 1990; Fortin et al 1996; Williams and Windebank 2003). In recent years, nevertheless, a growing number of studies have corroborated the marginality thesis (e.g., Leonard 1998; Kesteloot and Meert 1999). The outcome has been a more nuanced understanding that explains the variations in different populations in terms of the broader social, cultural, economic and geographical context (e.g., Williams and Windebank 1995, 1998).

Such attempts to embed understandings of the variable size of undeclared work into the broader cultural, social and geographical context in which such work takes place, however, have seldom been evident when considering the nature of undeclared work. Instead, most literature simply accepts that such monetary transactions are near enough universally conducted under market-like work relations for the purpose of monetary gain. This supposition predominates wherever undeclared work is studied in the world (e.g., Castells and Portes 1989; De Soto 1989; International Labour Office 2002) and whether the analysts write from a neo-liberal perspective, viewing such work as a form of self-employment that people pursue as rational economic actors confronted by rules and regulations that are inherently unfair (Sauvy 1984; De Soto 1989), or a political economy perspective viewing it as a form of exploitative employment that a weak and unprotected workforce is obliged to undertake for unscrupulous employers (e.g., Castells and Portes 1989; Portes 1994; Sassen 1997).

Such a reading is also prevalent whether such work is seen as in long-term terminal decline or one views economic globalization (a dangerous cocktail of de-regulation and increasing global competition) as causing its emergence and growth as a new facet of contemporary capitalism (e.g., Castells and Portes 1989; Sassen 1997; International Labour Office 2002), especially in global cities and amongst immigrant/ethnic minority populations (e.g., Sassen 1997). This discourse also applies whether it is economists studying the magnitude of undeclared work such as its cross-national variations (e.g., Dallago 1991; European Commission 1998; Feige 1999; Friedman et al2000; Schneider 2001; International Labour Office 2002) or sociologists and geographers examining its social and/or spatial variations (e.g., Renooy 1990; Jensen et al 1996; Williams and Windebank 1998). It also pertains both amongst those who adhere to the ‘marginality thesis’ (e.g., Button 1984; Castells and Portes 1989; Portes 1994; Kesteloot and Meert 1999), those viewing it as more concentrated in affluent areas (e.g., Pahl 1984; Van Geuns et al.1987; Renooy 1990; Mingione 1991; Mingione and Morlicchio 1993; Jensen et al.1995; Fortin et al.1996) and even those adopting a more nuanced understanding of its variable size (e.g., Williams and Windebank 1995, 1998).

This ‘thin’ reading of undeclared work as market-like and conducted for monetary gain even predominates when multiple types of undeclared work are recognized ranging from its ‘organized’ forms where such work is conducted under an employee/employer relationship for formal or informal businesses to its ‘autonomous’ form where it is conducted on a self-employed basis (e.g., Pahl 1984; Renooy 1990; Leonard 1994, 1998; MacDonald 1994; Jensen et al.1995; Fortin et al.1996; Williams and Windebank 1998). Even if undeclared work is no longer viewed as an exploitative form of low-paid employment sitting at the bottom of a hierarchy of types of formal employment, but as a heterogeneous labour market with a hierarchy of its own, all such work remains widely viewed as market-like and pursued for the purpose of monetary gain, as displayed by the fact that it is referred to as a ‘segmented informal labour market’ (Williams and Windebank 1998).

Yet exceptions exist (e.g., Cornuel and Duriez 1985; Jensen et al.1995; Nelson and Smith 1999). Cornuel and Duriez (1985), in their study of relatively affluent households in French new towns for example, highlighted how favours were exchanged on a paid basis between neighbours primarily in order to forge fledgling networks of social support rather than to make extra money in small town America, meanwhile, Nelson and Smith (1999) identify that ‘ersatz entrepreneurial activity’ (e.g., vehicle repairs) by men displaced from traditional male occupations appeared on the surface to be undertaken for monetary gain but turned out on closer examination to be driven by the need to maintain masculine identities in a changed industrial environment – and often to cost rather than generate money. In a similar vein, Jensen et al. (1995) in non-metropolitan Pennsylvania find that the predominant motive underpinning such work in some 61% of all reported cases of undeclared work was to help out neighbours.

Here, in consequence, and drawing inspiration from both the wider literature discussed above that has pursued ‘thicker’ descriptions of monetised exchange in general, and these few studies that have sought to unravel the nature of undeclared work, the intention is to subject this dominant view to critical scrutiny.

In order to evaluate critically this dominant reading of undeclared work, we here report the results of 861 face-to-face interviews conducted between 1998 and 2001 in eleven English localities (see Table 1) that sought to understand ‘household work practices’ (see Pahl 1984; Warde 1990; Nelson and Smith 1999; Smith 2002; Wallace 2002; Williams and Windebank 2002). Previous papers have reported the broader findings on household coping capabilities and work practices (see Williams and Windebank 2002, 2003). Here, for the first time, the full data set is analysed with regard to its findings on the nature of undeclared work.

To collect evidence on household coping capabilities and work practices in general, and undeclared work in particular, a preliminary pilot study attempted to use a relatively unstructured interview schedule. The finding, however, that interviewees found it difficult to recall instances of undeclared work and that comparative data was not being collected, resulted in a relatively structured interview schedule being designed. This was centred on a list of 44 common services.¹ The starting point in generating this list was the tasks used by Pahl (1984) in his seminal study of household work practices on the Isle of Sheppey. Unlike this earlier study, however, the interviews also sought to explore the motivations of suppliers and consumers.

To identify evidence of households employing undeclared work to get tasks completed, for each of the 44 tasks, the sources of labour used by the household to complete these activities were investigated. The interviewee was asked whether each task had been undertaken in the household during the previous five years/year/month/week (depending on the activity). If not, they were asked why not. If it had been undertaken, however, they were asked: who had conducted the task (e.g., a household member, kin living outside the household, a friend, neighbour, firm, landlord, etc.), whether the person had been unpaid or paid, and if paid, whether it was ‘cash-in-hand’ or not as well as how much they had been given. For each task completed, moreover, the respondent was asked in an open-ended manner why they had decided to get the work completed using that source of labour (so as to enable their motives to be understood) and how they would have got the task done if they had not employed this labour.

Following this, the supply of both undeclared work and unpaid community exchange by household members was examined. The interviewee was asked whether a household member had conducted each of the 44 tasks for another household and if so, who had done it, for whom, whether they had received money, how much they had received and why they had decided to do the task. To collect data on other types of undeclared work conducted and received, meanwhile, a series of open-ended questions with probes were used to elicit additional information.

Similar to previous studies (e.g., Pahl 1984; Leonard 1994; MacDonald 1994), interviewees were found to display little reticence in openly talking about this work. Just because it is hidden from government authorities does not mean that people hide it from each other or even academic researchers. This perception of a lack of reticence by interviewees, moreover, was confirmed by the data. The total amount customers reported spending on undeclared work in deprived urban neighbourhoods for example was £23,354 and this was near enough exactly the same amount that suppliers of undeclared work asserted that they had received (£22,986). So too was the mean price customers paid (£90.24) broadly equitable with the mean price suppliers asserted that they received (£84.48). It was similarly the case in affluent suburbs and rural areas. There is thus little evidence either of suppliers greatly under-reporting such work or their income from this activity, or of customers falsely allocating economic activity to undeclared work when this was not the case. Given that the order of magnitude of customer and supplier responses are approximately the same, few grounds exist for believing that these results under-report either the level of this work or the income from such activity.

As shown in Table 1, this interview schedule was implemented in a variety of affluent and deprived rural and urban localities. The rationale was that previous research had identified the urban/rural and level of affluence variables as important influences on both household coping capabilities and practices in general, and undeclared work more particularly (e.g., Pahl 1984; Renooy 1990; Jensen et al. 1996; Kesteloot and Meert 1999). As such, a maximum variation sampling technique was employed. Firstly, two cities were chosen that starkly varied in terms of their level of affluence. Southampton is a successful service-oriented economy with low unemployment rates in the affluent South East of England whilst Sheffield is a poorer northern English city, once famous for steel making but now suffering relatively high rates of unemployment. Continuing this desire for maximum variation sampling, three neighbourhood-types in each city were then chosen using the Index of Multiple Deprivation of the DLTR that ranks all wards in England and Wales according to the level of their multiple deprivation (DLTR 2000). In each city, two of the most deprived wards and the most affluent ward were selected. Following this, and to provide a stark contrast with the urban areas, a range of rural localities was chosen with a broad regional spread. Again, affluent and deprived rural localities were chosen for investigation, continuing the theme of maximum variation sampling.

To select households for interview in each locality, meanwhile, a spatially stratified sampling procedure was used (Kitchin and Tate 2001). The researcher called at every nth dwelling in each street, depending on the size of the neighbourhood and the number of interviews sought. In consequence, if there were 1000 households in the ward and 100 interviews were sought, then every 10th household was visited. If there was no response, then the researcher called back once. If there was still no response and/or they were refused an interview, then the 11th house was surveyed (again with one call back), then the 9th dwelling, 12th and so on. Here, therefore, the findings of this survey with regard to the nature of undeclared work are reported.

Until now, a market-centred reading of the economic relations and motives underpinning undeclared work has been often accepted without investigation. Take, for example, the common finding that much undeclared work is low-paid. When identified, it is too often interpreted that this means that such work must be an exploitative form of employment (e.g., Sassen 1997; International Labour Office 2002). The economic relations within which low-paid undeclared work is embedded and the motives of the suppliers and/or employers are seldom, if ever, unpacked to see whether this is always the case.

Indeed, even in those few instances where it has been identified that undeclared work is conducted under non-market relations and for motives other than making money (e.g., Cornuel and Duriez 1985; Jenson et al. 1996; Nelson and Smith 1999), such findings have not been used to mount a challenge to either the hegemony of market-centred readings of undeclared work or the dominant discourse that monetary exchange is always market-like and motivated by monetary gain. If anything, such findings have been instead treated as mildly interesting exceptions to the general rule. Here, however, and using the data set from these English localities, we wish to argue that these previous findings of the existence of non-market relations and motives other than monetary gain need to be treated far more seriously than has so far been the case.

Starting with the economic relations within which undeclared work is embedded, the finding of this study was that just 30% of the undeclared work received by these 861 households was conducted by formal firms and/or self-employed people previously unknown to the household (see Table 2). Friends, neighbours and kin thus supplied some 70% of all undeclared tasks. As such, most undeclared work is provided by fairly close social relations rather than by anonymous firms and/or individuals. One of the most fascinating findings, moreover, and reinforcing previous studies (e.g., Van Eck and Kazemeier 1985), is that this undeclared work provided by close social relations is paid at wage rates 35% lower than the undeclared work carried out by firms and/or individuals previously unknown to the supplier.

Does this mean, therefore, that rational self-interest and monetary gain has penetrated so deeply that we now exploit our friends, neighbours and kin by paying them low wages to do work for us? Or are different motives prevalent when such work is conducted meaning that the low wages cannot be taken as indicator of exploitative market-like work relations?

It is normally assumed that both employers and suppliers of undeclared work engage in such activity in order to save/make money. However, Tables 3 and 4 display that this is not always the case. Starting with the employers of undeclared labour, Table 3 displays that the notion of saving/making money is the primary motive in just 31% of instances where undeclared work was used and in nearly every case, it was either firms and/or self-employed people previously unknown to the employer who were conducting this work. There is thus a strong correlation between the desire to get jobs done cheaper than would be the case if somebody were formally employed and the use of unknown firms and people to do the undeclared work. For example, the motive whenever building firms, plumbers or electricians previously unknown to the employer were paid on an undeclared basis was always to save money. This was similarly the case whenever previously unknown domestic cleaners, gardeners and private tuition were paid for on an undeclared basis. When there is social distance between the purchaser and supplier, therefore, it appears that monetary gain prevails in employers’ rationales.

It is similarly the case when suppliers of undeclared work are considered (see Table 4). In only half of all reported cases (50%) of undeclared work did suppliers assert that monetary gain was their primary rationale and even then, it was frequently qualified with other statements especially when conducted for closer social relations. Representative of such statements were the following: ‘I did it for the money but it was only pocket money really and it was a chance to help them out at the same time’ (employed woman), ‘I did it for the money I suppose but it was also a chance to get out of the house’ (unemployed man) and ‘It was to make some money but I was also doing it [mending a leaking tap] for them because they were desperate to get it repaired’ (employed man).

Although monetary gain was the principal motive when working for firms ‘on the side’ (5% of all undeclared work) and engaging in self-employed activity for people they do not know well (25% of all such work), it was not when doing work for closer social relations, which covers 70% of undeclared work. Contrary to the conventional view that undeclared work involves low-paid employment conducted for and by formal or informal businesses, therefore, it was found that just 5% was of this type. This ‘organised’ undeclared work included such jobs as cleaning a truck for £10 which took 3 hours, doing bar work for 2 weeks for £50, working in a restaurant for £2.00 per hour, working early mornings in a small bakery for £20 per month, working for three weeks in a canteen on a building site for £50, staffing an ice-cream stall for £2.80 per hour and refurbishing a pub for one week for £100.

The 25% of work conducted on a self-employed basis for people previously unknown to the supplier, meanwhile, again covered a diverse array of activities. It includes activities ranging from the provision of various professional advisory services in spheres such as architecture, landscaping and financial planning through gardening services to household cleaning for people previously unknown to the self-employed person. This, moreover, was sometimes very well paid work and in 90% of cases was undertaken purely to make money although there were again often caveats.

Examining purchasers’ and suppliers’ primary reasons when closer social relations are involved, however, some very different motivations emerge. When purchasers employed friends, neighbours and kin to conduct undeclared work, or when suppliers conducted undeclared work for such close social relations, their primary motivation was found to revolve around either ‘community-building’ or ‘redistribution’.

When ‘community-building’ rationales predominate, employers pay on an undeclared basis, and suppliers conduct undeclared work, so as to forge or cement a social relationship, and this tends to prevail amongst friends and neighbours (rather than kin). This is perhaps unsurprising. After all, exchange is a principal nexus through which social relations are forged and maintained in contemporary society. Why, however, should monetary payment be involved? Payment was often preferred due to both wariness about engaging in unpaid exchange and the embedding of payment in norms of reciprocity. Firstly, people did not want to feel that they owed others a favour. Secondly, there was a feeling that you could no longer rely on people to return favours, indicating the demise of trust in at least some of these populations. Therefore, cash was seen as a necessary medium when maintaining or building community networks, especially when neighbours or friends were involved. Rather than allow such relations to turn sour if and when they reneged on their commitments, the exchange of cash prevented such a situation arising. Norms of reciprocity, it seems, often involve cash that acts as a substitute for trust in the formation and maintenance of networks of reciprocity, at least in some populations.

‘Redistributive’ rationales, meanwhile, usually apply when kin conduct the work. Here, employers view paying for jobs undertaken as a way of giving much needed spending money to kin, such as when they are unemployed. Indeed, asking kin to do a task in order to give them some needed money was the principal rationale behind 22% of all undeclared work. It was thus a way of giving money to kin that avoided all connotations of charity, even if this was the intention. As Kempson (1996) has previously revealed, people avoid accepting charity at all costs. This was well understood by the populations in the English localities surveyed. As a result, anybody wishing to give aid paid them for doing a task to ‘help them out’.

For suppliers, meanwhile, these redistributive motives reflected a desire to help out a customer that they already knew who might otherwise not be able to get the work completed. In these instances, which constituted 22% of all undeclared work supplied, the undeclared worker was usually either mainly a person with professional craft skills (e.g., plumbers, electricians) and often employed who supplied these skills to close social relations for a fee well under the market price, or else they were unemployed or early retired people who saw themselves as helping out those who had less free time than themselves by supplying their time.

Therefore, even if there has been the penetration of monetary relations into ever more spheres of social life, this study shows that these exchanges are by no means everywhere characterised by monetary gain. To read it through the lens of making money is to portray the supply of such work much more through the eyes of men, the employed and affluent areas (see Tables 3 and 4). If such a partial and narrow experiential reading of undeclared work is to be transcended, and the meanings of such work for women, the unemployed and deprived areas better portrayed, then there will be a need to not only recognise the heterogeneous work relations and motives involved but also greater emphasis will have to be put on non-market relations and rationales other than monetary gain.

Given this re-reading of the nature of undeclared work, it is here necessary to reconsider what should be done about this form of work. Until now, a view that such work is market-like and conducted primarily to make extra money, and thus that employers and suppliers make a rational economic decision to engage in such work, has led public policy to seek to deter such work by ensuring that the expected cost of being caught and punished is greater than the economic benefit of participating. To achieve this, stringent regulations and punitive measures have been adopted to change the cost–benefit calculation of participants (e.g., Allingham and Sandmo 1972; Falkinger 1988; Milliron and Toy 1988; International Labour Office 1996, 2002; European Commission 1998; Hasseldine and Zhuhong 1999; Sandford 1999; OECD 2000).

The above re-representation of undeclared work, however, raises grave doubts over such a deterrence approach. If many people are engaged in undeclared work for moral reasons rather than monetary gain, then there will be little societal consensus over the need to stamp out such work and thus grave difficulties in achieving this end. Indeed, this is displayed in Quebec (Fortin et al. 1996). Here, 15% of the population believe that undeclared work is morally acceptable and 43% perceive it as neither moral nor immoral, whilst just 31% see it as immoral. Deterring undeclared work is also undesirable. Take, for example, deprived neighbourhoods. One of the most fascinating findings of the English localities study reported above is that although some residents do indeed engage in market-like undeclared work to make money, the vast majority is a type of mutual aid. Indeed, of all reported instances where favours were provided by a neighbour or friend, payment was involved in 84% of such instances. Paying for favours, therefore, is heavily embedded in the norms of reciprocity in such populations. To pursue a policy of deterrence is thus likely to result in governments pursuing the eradication with one hand of precisely those networks of social support that their ‘active citizenship’ agendas are seeking to nurture with another hand. Consequently, although deterrence appears to be largely applicable for tackling undeclared work amongst affluent populations, men and those in employment, it seems likely that such a policy approach will have deleterious impacts on the development of social capital and redistribution amongst deprived populations, women and the unemployed. The resulting challenge for public policy is perhaps to consider how some ‘joined-up’ thinking can occur with regard to policy towards undeclared work and its agenda to facilitate active citizenship or community engagement.

What alternative policy options, therefore, are available and what are the implications of pursuing these approaches? One possibility is to pursue a laissez-faire approach towards undeclared work. Although this would enable people to freely engage in the moral economy of paid favours and thus help cement and forge networks of material support, it would do little to legitimise what remains essentially fraudulent work and neither would it do anything about the exploitative forms of informal employment that would continue unabated.

Another possibility, therefore, is to adopt some combination of deterrence and laissez-faire. A laissez-faire approach towards informal employment and a deterrence approach towards the moral economy of favours is perhaps the worst possible option that could be chosen. This would leave intact informal employment with all of its negative impacts whilst seeking to deter undeclared work conducted as a form of paid mutual aid.

What, however, about the option of seeking to deter informal employment and adopting a laissez-faire approach towards the moral economy of favours? Although more focused on retaining the moral economy of favours and preventing those forms of fraudulent undeclared employment deliberately undertaken primarily for monetary gain by both suppliers and employers, several key problems remain. Firstly, it would be very difficult for state actors to distinguish between undeclared work conducted for monetary and non-monetary reasons. Although some forms of undeclared work such as informal or formal businesses working on an undeclared basis are easily categorised as informal employment, it becomes much more difficult to distinguish between informal employment and the moral economy of favours once one starts examining individuals conducting undeclared work on an autonomous basis. Whether they are acting as self-employed people engaged in informal employment for monetary gain or as individuals engaged in mutual aid to help others would be difficult to decipher.

A second problem with this approach is that adopting a laissez-faire approach towards the hidden economy of favours leaves intact the inequalities that presently exist in this sphere. A significant minority of the population, amounting to some 25% in deprived neighbourhoods according to the above English localities survey, are currently excluded from drawing upon the resources of this moral economy of favours as a coping practice in order to meet their material and social needs, and these are often amongst the lowest income households in these localities. To adopt a laissez-faire approach is thus to leave intact these disparities. Third and finally, such an approach that continues to seek to deter informal employment fails to recognise some of the more positive aspects of informal employment such as its function as a school for entrepreneurialism and a route into formal labour market participation (e.g., International Labour Office 2002; ODPM 2004). For these three reasons alone, it is here considered that the policy option of deterring informal employment and adopting a laissez-faire approach towards the moral economy of favours is inappropriate.

Besides, even if undeclared work in the moral economy of paid favours is conducted for redistributive or sociability reasons, it remains fraudulent activity that deprives the state of revenue for social cohesion purposes. As such, it needs to be eradicated. The issue in need of exploration, however, is whether deterrence is the appropriate policy response. Given how the norms of reciprocity, especially in deprived neighbourhoods, are based on cash payments, it is perhaps positive rather than negative reinforcement that is required, or put another way, initiatives to marshal such activity into more legitimate forms of mutual aid so as to retain and even develop these networks of material support rather than deterrence measures alone. Similarly, even if informal employment needs to be eradicated, not least due to its fraudulent and exploitative nature, it is perhaps again the case that more positive initiatives might be considered to help direct such work into the formal realm rather than simply seek to deter it. How, therefore, can one facilitate the legitimisation of these different forms of undeclared work be achieved?

In recent years, those recognising some of the more positive features of undeclared work (e.g., as a breeding ground for the micro-enterprise system, a test-bed for fledgling businesses) have begun to view such work not as an obstacle to modernisation but, rather, as a potential asset to be harnessed and even as a driver of economic development (e.g., Tabak 2000; Global Employment Forum 2001; International Labour Office 2002). Until now, however, this emergent literature that discusses the notion of ‘cultivating’ undeclared work has viewed such activity purely as something to be transformed into formal employment. Although the ideas propounded here resonate with this emergent stream of thought, the argument is that recognition of the heterogeneity of this sphere necessitates some significant modifications to existing views regarding how to harness such work. In the realm of informal employment, that is, ‘harnessing’ undeclared work means transforming such work into formal employment. So far as the moral economy of paid favours is concerned, however, ‘harnessing’ undeclared work means transferring paid mutual aid out of the illegal undeclared realm into a sphere of legitimate paid mutual aid.

How, therefore, can such a twin-track approach be implemented? To aid the transfer into formal employment of undeclared work, on the one hand, state regulations could be changed (e.g., shifting the income tax threshold and changing the earnings disregard levels) so that the economic activity is no longer classified as undeclared work, and on the other hand, initiatives could be implemented to ease their transition to the formal sphere ranging from amnesties, through formal job creation initiatives using subsidies and incentives such as is the case in France and The Netherlands in the realm of domestic services (see Cancedda 2002), to initiatives to help those currently working on an undeclared basis to formalise their businesses such as Street UK (ODPM 2004). What is certain, however, is that understanding what works so far as providing positive incentives to facilitate the legitimisation of undeclared work is in its infancy.

To transfer undeclared work in the moral economy of paid favours into legitimate forms of paid mutual aid, meanwhile, mutual exchange systems are required that use some form of tally system or payment when participants conduct favours for each other so as to reflect the culture of paying for favours. Two relevant systems in this regard are local exchange and trading schemes (e.g., Williams et al. 2001) and time banks (e.g., Seyfang and Smith 2002). Although both are relatively ineffective at creating jobs, evaluations have shown that they are highly effective in providing a substitute for undeclared work and facilitating mutual aid. Indeed, some 35% of activity conducted on LETS would have been undertaken through undeclared work if the LETS did not exist (Williams et al.2001; Seyfang and Smith 2002). There are doubtless many more similar initiatives.

Grounded in the view that suppliers and employers of undeclared work undertake market-like work in pursuit of monetary gain, and that their engagement is a rational economic decision, the widespread public policy response has been to seek to deter such work by ensuring that the expected cost of being caught and punished is greater than the economic benefit of participating. In this paper, however, research from the UK has been reported which reveals how undeclared work is composed of not only such informal employment but also a moral economy of paid favours. If deterrence is pursued, therefore, then public policy will with each new initiative to clamp down on undeclared work further destroy the active citizenship that other arenas of public policy seek to cultivate. Here, therefore, a number of alternative policy options have been considered.

A laissez-faire approach towards undeclared work, it has been shown, would leave intact informal employment whilst doing little or nothing to facilitate the engagement of that significant minority of people currently excluded from participating in the hidden economy of favours who are perhaps most in need of the benefits that come from such engagement. Combining laissez-faire and deterrence in a variety of ways, meanwhile, by applying one to informal employment and the other to the moral economy of favours is neither possible nor desirable. Here, therefore, an approach that seeks to facilitate the legitimisation of both forms of undeclared work has been advocated. Building upon the emerging view that undeclared work is an asset to be cultivated rather than a hindrance to development (e.g., Tabak 2000; Global Employment Forum 2001), but recognising the heterogeneous varieties of undeclared work, the argument has been that a twin-track ‘facilitating’ approach should be pursued that transfers informal into formal employment and activity presently undertaken in the moral economy of paid favours into legitimate paid mutual aid. To show how this might be implemented, a number of possible policy initiatives have been very briefly highlighted. Obviously, however, the initiatives reviewed here are just indicative of the types that could be employed. There are doubtless many more. What is certain, however, is that the current deterrence approach should not continue to dominate thought and action since this is perhaps the least satisfactory of all possible approaches that could be adopted.

Colin C Williams is Professor of Work Organisation at the University of Leicester Management Centre and Director of the Collective for Alternative Organisation Studies (CAOS). His recent books include Informal Employment in the Advanced Economies (1998), Routledge)with Jan Windebank, and Cash-in-Hand Work (2004, Palgrave Macmillan)

Dr Jan Windebank is Senior Lecturer in French Studies and Associate Fellow of the Political Economy Research Centre (PERC) at the University of Sheffield. Recent books include Women and Work in France and Britain (2001), Palgrave-Macmillan) and Community Self-Help (2004), Palgrave-Macmillan).