The Growth of Urban Informal Economies

Colin C Williams & Jan Windebank. Handbook of Urban Studies. Editor: Ronan Paddison. SAGE Publications. 2001.

In recent years, the emerging orthodoxy in urban studies is that our cities are witnessing the growth of informal employment as it becomes an increasingly important survival strategy for marginalized populations (for example, the unemployed, women, ethnic minorities and immigrants). The aim of this chapter is to evaluate critically this conceptualization concerning both the development and character of urban informal employment. Synthesizing the findings of the vast array of studies of such work, first, whether or not urban informal employment is growing or declining will be analysed and, secondly, the character of this work will be evaluated. On the one hand, this will reveal that rather than a universal process of either formalization or informalization, the trajectory of urban economic development is much more heterogeneous than has been so far considered. On the other hand, it will show that urban informal employment is not solely a survival strategy for the marginalized, but rather, that there is a heterogeneous informal labour market with a hierarchy of its own which reproduces the socio-spatial divisions prevalent in the urban formal labour market. To explain the heterogeneous magnitude, direction and character of different informal labour markets, the third section argues that this is due to the way in which local economic, social, institutional and environmental conditions combine in multifarious ‘cocktails’ in different places to produce specific local outcomes. In the concluding section, the policy opinions and implications are assessed. Evaluating critically the three dominant policy approaches towards urban informal employment, we conclude that the option chosen is inextricably related to broader prescriptions for work and welfare.

Before commencing, however, it is first necessary to define what we mean here by urban informal employment. In this chapter, we are discussing the paid production and sale of goods and services that are unregistered by, or hidden from, the state for tax, social security or labour law purposes, but which are legal in all other respects. As such, unpaid informal work is outside the scope of this chapter, as is criminal activity more widely defined. Instead, we are concerned here with activity which is illicit only because it involves social security fraud, tax evasion and/or the avoidance of labour legislation.

The Development Path of Urban Areas in the Advanced Economies: Formalization, Informalization or Heterogeneous Development?

The fact that there is a substantial amount of paid work in urban areas of the advanced economies which does not appear in the official statistics is without contention. However, the precise amount of such activity and whether it is increasing or decreasing faster than the measured economic activity is a matter of heated debate. Here, we evaluate critically both the popular belief that there is a formalization of urban economies as well as the theory that they are undergoing a process of informalization.

The Theory of Formalization of Urban Economies

One of the most widely held beliefs about urban economic development is that as urban economies become more ‘advanced’, there is a natural and inevitable shift of economic activity from the informal to the formal sphere (that is, the theory of formalization). Indeed, this is often the ‘measuring rod’ which defines Third World cities as ‘backward’ and First World cities as supposedly ‘advanced’. However, there are good reasons for believing that this trajectory of urban economic development is neither natural nor inevitable. In the advanced economies, the natural culmination of formalization – full-employment and a comprehensive welfare state – can no longer be accepted as the end-state of economic development. Between 1965 and 1995 in the European Union (EU), for example, the share of the working age population with a job fell from 65.2 per cent to 60.4 per cent (European Commission, 1996a). When this is coupled with the shift away from permanent full-time jobs towards temporary and part-time work (Nicaise, 1996), full- or even fuller-employment seems to be receding ever further into the distant past. So too is a comprehensive welfare ‘safety net’. Even in those EU nations wishing to retain it, universal welfare provision is in demise, not least due to the pressure from other trading blocs to keep social costs low (Conroy, 1996; Ferrera, 1996; Williams and Windebank, 1995a). As the European Commission (1995) reports, expenditure on social protection benefits per capita fell or remained static in six out of 15 member states during the 1990s compared with the late 1980s.

It is not only in the advanced economies, however, that formalization appears to be neither natural nor inevitable. Across the Third World, formal employment is by no means universally growing. Although employment has risen at a rate well in excess of the increase in the labour force in many East and South East Asian nations, there are whole swathes of the Third World not only at a very low base-level so far as formal jobs are concerned but which are either standing still or are undergoing a process of informalization. In low-income nations, for example, regular waged or salaried employment accounts for only 5-10 per cent of total employment, and in many regions such as Latin America, sub-Saharan Africa, North Africa and the Middle East, employment growth has been either static or decreasing relative to the growth in the labour force (International Labour Office, 1996).

Therefore, the theory of formalization, upon which much of our thinking about both the future of work and welfare in our cities is founded, is cast into doubt by the above findings. In consequence, it might be assumed that the opposite is occurring. If people are so widely and increasingly excluded from both employment and formal welfare provision, then they must be turning towards informal employment as a means of survival. Are we witnessing, therefore, the renewed informalization of urban life?

The Theory of Urban Informalization

Based on the assumption that informal employment is a new form of advanced capitalist exploitation (e.g., Castells and Portes, 1989; Sassen, 1991) or a response to over-regulation by the market (de Soto, 1989), an increasingly popular view is that as the advanced economies fall into economic crisis and the neo-liberal project of deregulation takes hold, we are witnessing the absolute and/or relative growth of informal employment in our urban economies. The principal problem, however, is that there is little solid evidence that this process is taking place. The indirect macro-economic methods which have generated data to support such an assertion suffer from fatal flaws in their design which render the results of dubious quality and validity (see, for example, Smith, 1986; Thomas, 1988). For the same country in the same year, for example, it is normal to find major variations ranging from as little as 2-3 per cent to 30-35 per cent of GNP (Barthelemy, 1991). As Thomas (1988) displays, the size of informal employment has been estimated to range from 1.5 to 27 per cent of GDP for the US and 2 to 22 per cent of GDP for the UK. In consequence, to rely on any of these indirect measures as evidence for the informalization of the advanced economies would be to close one’s eyes to the inherent problems with the methodologies from which they are derived.

The only other source of potential evidence, the micro-social studies of informal employment, although more direct and accurate in their portrayals of such work, tend to be snapshots at one point in time, usually in specific localities. Even if they are an accurate snapshot of informal employment in a particular locality, therefore, they do not show whether there is an informalization of urban economic life taking place. Indeed, the only two longitudinal micro-economic studies of informal employment both refute the informalization thesis. Mogensen et al. (1995) in Denmark find that the proportion of people engaged in informal employment has remained level, at 13-15 per cent, throughout the period from 1980 to 1995, whilst Fortin et al. (1996) in Quebec reveal that between 1985 and 1993, informal employment as a percentage of GNP has stayed constant at 0.65 per cent of GDP. Micro-economic studies, therefore, provide little evidence of informalization.

For some, it might be assumed that the demise of formal work and welfare indicated above is sufficient proof of informalization. This, however, assumes that formal and informal employment are substitutes for each other (e.g. Castells and Portes, 1989; Gutmann, 1978) and that the rise of one leads to the fall of the other. The problem is that an increasing number of micro-social studies suggest that the relationship is not one of substitution but rather one of complementarity (e.g. Cappechi, 1989; Cornuel and Duriez, 1985; Leonard, 1994; Morris, 1995; Pahl, 1984). Given this, it would be misleading to take the de-formalization of urban economic life as a surrogate measure of the growth of informalization. The picture is far more complex. Neither can the supposed recent growth of informal employment simply be viewed as a new form of capitalist exploitation or a contemporary response to over-regulation by the market. Informal employment, after all, is not a recent phenomenon, having been in existence as long as there have been rules and regulations with regard to employment (Henry, 1978; Smithies, 1984) and its recent growth is by no means obvious.

Transcending Informalization/Formalization: The Heterogeneous Development Paths Approach

Here, therefore, precisely because informal employment is not everywhere a substitute for formal jobs, we do not assert that the demise of formal work and welfare is necessarily leading to a relative informalization of the advanced economies. Instead, we introduce the idea that both history and geography matters. Examining the findings of the vast array of direct studies, it is evident that there are different processes in different places at varying times. As such, it is more accurate to describe the development trajectory of the advanced economies as one of heterogeneous development, with different places undergoing either informalization or formalization at varying times and at different rates. The rationale for such a theorization of the trajectory of urban economic development will be more fully explained below. Here, we simply state this as an assertion arising out of our reading of these studies of informal employment. Before more fully theorizing this temporally and spatially refined theorization of the level of urban informal employment in the advanced economies, it is first necessary to explore the nature of such work.

Characterizing the Nature of Urban Informal Employment: Peripheral Labour or Heterogeneous Informal Labour Markets?

Based on the assumption that it is marginalized groups who engage in urban informal employment as a means of survival, the debate concerning how to explain the existence of such work has thus revolved around whether it is a leftover of classical capitalism (that is, the theory of formalization) or a new form of advanced capitalism (that is, the theory of informalization). This section, however, reveals that whichever position is adopted (Castells and Portes, 1989; Sassen 1989), both views are founded upon an erroneous premise.

Urban Informal Employment: Peripheral Labour for Marginalized Groups?

The ‘marginality thesis’ which views informal employment as a form of peripheral labour for marginal groups (for example, women, migrants, the unemployed, poor and children) has a long historical antecedent and can be traced back to the contemporary origins of the study of such activity in the work of Hart (1973) in Ghana. Since then, it has been taken up widely by writers in both the developing and advanced economies (for example, Button, 1984; Lagos, 1995; Maldonado, 1995). In this view, such work is seen as an exploitative peripheral form of employment which is at the bottom of a hierarchy of flexible types of formal employment and is undertaken mostly by marginalized population groups as a survival strategy. However, such a view of informal employment covers but one type of informal employment and one section of the informal workforce.

Informal employment, that is, is not a homogeneous category but is composed of a heterogeneous spectrum of paid economic activities (for example, Fortin et al., 1996; Jensen et al., 1995; Leonard, 1994; Mingione, 1991; Pahl, 1984; Renooy, 1990). It ranges from ‘organized’ informal work undertaken as an employee for a business which conducts some or all of its activity informally, to more ‘individual’ forms of such work. These cover not only self-employed activities which may involve a large proportion, if not all, of their earnings, but also casual one-off jobs undertaken on a cash-in-hand basis, such as for a neighbour, friend or relative. Not all of informal employment, moreover, is low-paid, exploitative and undertaken by marginal groups. Although there are forms of organized informal employment, such as in labour-intensive small firms with low levels of capitalization dependent on larger companies where marginal populations engage in low-paid exploitative activity (for example, Lim, 1995; Sassen 1991), there are also forms of organized informal employment in independent, modern, more capitalized, high technology firms producing higher priced goods and services and whose informal employees are well paid, use higher skills and have more autonomy and control over their work, with relations between employers and employees based more upon cooperation than domination (e.g. Benton, 1990; Cappechi, 1989; Warren, 1994). There are also forms of individual informal employment which are not only deeply socially embedded in urban life, but have more in common with self-employment and/or unpaid community exchange than they do with exploitative forms of informal or formal employment. So, despite the emphasis in much of the literature on informal employment as an exploitative phenomenon, the reality is that this is just one, albeit important, part of the spectrum of activities which constitute this form of work. If informal employment is not solely an exploitative form of peripheral labour for marginalized populations, then how should it be conceptualized?

Retheorizing Urban Informal Employment as a Heterogeneous Informal Labour Market

Based on a detailed review of the numerous one-off locality studies of informal employment in the advanced economies, it is here suggested that existing alongside the urban formal labour market and often segmented along the same lines, is a segmented urban informal labour market. This informal labour market, to adopt a simplistic dual labour market model, is composed of ‘core’ informal employment which is relatively well paid, autonomous and non-routine and where the worker often benefits just as much from the work as the employer, and ‘peripheral’ informal employment which is poorly paid, exploitative and routine and where the employee does not benefit as much as the employer. Just as in the urban formal labour market, moreover, there are also those excluded from even the most exploitative peripheral informal employment. Who, however, receives these high income informal jobs? And who is involved in the lower paid more exploitative informal employment?

To start to answer such questions, we here take a case study of the urban informal labour market in Quebec city (Lemieux et al., 1994). In 1986, 2,134 adults aged 18 years or over were surveyed in the Census Metropolitan Area of Quebec City, with a 63.8 per cent response rate. As Table 20.1 reveals, although 8.5 per cent of the total sample report working in informal employment, participation in this form of work is not evenly distributed across all social groups. Men, students, the unemployed and lower income groups are all more likely to conduct such work. Superficially, therefore, these results appear to reinforce the marginality thesis. However, closer analysis reveals that such a view does not fully describe this urban informal labour market.

Although the unemployed, for example, are more likely to engage in informal employment, they do not constitute the vast bulk of this informal workforce. Indeed, just 12.8 per cent of all informal workers are unemployed. The formally employed, meanwhile, constitute 35.2 per cent of all informal workers and earn more per hour for working informally than their more marginalized counterparts. So too do those who earn a higher formal income. Whilst those earning C$0-10,000 p.a. receive an average of C$4.96 per hour from their informal employment, the hourly wage rate rapidly rises, so that those earning C$30,000-40,000 p.a. from formal employment command informal wage rates of C$24.67 per hour. Informal employment, therefore, is clearly segmented along the lines of formal income and employment status. It is a similar story when gender is examined. Not only do a slightly higher proportion of men than women participate in informal employment (9.9 per cent compared with 7.1 per cent), but men receive higher hourly informal wage rates than women (C$6.93 compared with C$4.16). The informal labour market, therefore, is clearly segmented along similar lines to the formal labour market.

Table 20.1 Participation in informal employment in Quebec city, 1986
Earnings % of total inf. wk by
Characteristics % of sample % doing inf. wk % of all inf. wkrs Hours Total Average/informal worker Av.$/h Hours Value
All 100.0 8.5 100.0 357 2006 171 5.61 100.0 100.0
Gender
Men 48.7 9.9 56.7 331 2294 227 6.93 52.6 64.9
Women 51.3 7.1 43.3 391 1628 116 4.16 47.4 35.1
Employment status
Student 11.4 28.2 38.0 332 1976 557 5.45 35.6 37.4
Retired 5.1 1.9 1.1 120 490 9 4.08 0.4 0.3
Housekeeper 17.6 6.2 12.8 581 2251 140 3.87 21.1 14.4
Unemployed 4.0 27.4 12.8 369 1904 522 5.16 13.4 12.2
Employed 61.9 4.8 35.2 297 2034 98 6.85 29.5 35.7
Official income (C$)
0-10,000 51.4 12.9 74.7 400 1984 256 4.96 84.3 73.9
10,000-20,000 17.2 7.0 13.4 286 2302 160 8.05 10.8 15.4
20,000-30,000 15.5 3.9 7.0 190 1943 75 10.23 3.7 6.8
30,000-40,000 9.9 2.0 2.2 58 1431 29 24.67 0.4 1.5
40,000+ 11.0 2.3 2.7 104 1790 41 17.21 0.8 2.4

Source; Derived from Lemieux et al. (1994: Table 1)

Contrary to popular prejudice, furthermore, the informal labour market does not necessarily have lower wage rates than the formal labour market. Lemieux et al. (1994) find that although the average gross wage per hour is 13 per cent higher in formal than informal employment (C$7.98 compared with C$6.99), the average net wage per hour in formal jobs, which takes account of marginal tax rates and tax-back rates associated with social transfers, is lower than the average informal wage. Informal employment, therefore, is not merely an exploitative low-paid sector of the economy. It has a higher average net wage rate than formal employment. Nevertheless, the standard deviation of the informal wage rate is larger than the standard deviation of the formal wage rate, suggesting that wage differentials are greater in this sector, adding yet further weight to the idea of a heterogeneous informal labour market. Moreover, whilst hours and wages are positively correlated in formal employment, they are negatively correlated in informal employment (Lemieux et al., 1994), suggesting that many more in informal than formal employment work long hours for little pay and short hours for high pay.

Are these findings from the Quebec City informal labour market replicated elsewhere? Do the unemployed, for instance, always constitute a smaller proportion of informal workers than the employed? Do they everywhere earn less than the employed? Does gender always relate in the same way to the informal labour market? And do ethnic minority and immigrant groups, not considered in the Quebec study, relate to the informal labour market in similar ways to these other marginalized groups?

Starting with the relationship between employment status and informal employment, the overwhelming finding of the vast majority of locality studies is that informal employment is primarily a means of accumulating advantage for those already in employment and that the number of regular ‘working claimants’ accounts for a very small proportion of all informal employees. In Spain, for example, the Ministry of Economy estimates that 29 per cent of those in employment also have an informal job (in Hadjimichalis and Vaiou, 1989), whilst Lobo (1990a) finds that just 12 per cent of those claiming benefit perform such work. Benton (1990) thus reveals that 65.7 per cent of all informal employees have a formal job, whilst just 5.2 per cent are working informally and receiving social security benefit at the same time. The remaining 29 per cent of informal workers, we must assume, are those unemployed not entitled to benefit. In Greece similarly, the Ministry of Planning recognizes the fact that it is more likely to be the employed than the unemployed who have informal jobs when it states that 40 per cent of those employed in the private sector also have non-declared jobs, as do 20 per cent of those in public sector jobs (in Hadjimichalis and Vaiou, 1989). This finding that informal employment chiefly benefits those already in employment is also identified in the Netherlands (Koopmans, 1989; Van Eck and Kazeier, 1985; Van Geuns et al., 1987, France (Barthe, 1988; Cornuel and Duriez, 1985; Foudi et al., 1982; Tievant, 1982), Germany (Glatzer and Berger, 1988; Hellberger and Schwarze, 1987), Britain (Economist Intelligence Unit, 1982; Howe, 1990; Morris, 1994; Pahl, 1984; Warde, 1990), Italy (Cappechi, 1989; Mingione, 1991; Mingione and Morlicchio, 1993; Warren, 1994) and North America (Fortin et al., 1996; Jensen et al., 1995; Lemieux et al., 1994; Lozano, 1989).

So far as the character of their work is concerned, the vast majority of studies find that the employed tend to engage in more autonomous, non-routine and rewarding informal jobs than the unemployed, who conduct more routine, lower paid, exploitative and monotonous informal employment (Fortin et al., 1996; Howe, 1990; Lemieux et al., 1994; Lobo, 1990a, 1990b; Pahl, 1987; MacDonald, 1994; Williams and Windebank, 1995a; Windebank and Williams, 1997). A plumber employed by one of the large utilities, for example, may put in central heating for cash-in-hand during the weekend for a customer she/he has met during the course of his/her formal employment, whilst an unemployed person may only have access to low-paid homeworking activity such as assembling Christmas crackers which she/he has learned about through an advertisement in a shop window or newspaper or through a friend who already does such activity. The result, therefore, is a segmented informal labour market in which many of those who already have a formal occupation find relatively well-paid informal employment, often conducted on a self-employed basis, whilst the unemployed generally engage in relatively low-paid organized informal employment which tends to be more exploitative in nature. As Van Eck and Kazeier (1985) identify, informal employment found through employers or colleagues is by far the best paid, namely two or three times higher than informal employment found through other channels. Given that the unemployed have neither employers nor colleagues, the outcome is limited access to these better paid informal employment opportunities. Instead, their networks confine them to seeking work via family or acquaintances. The result is lower wage levels since the average wage for work conducted for one’s family is Dfl10, for acquaintances it is Dfl14, for colleagues Dfl20 and for employers Dfl33 (Van Eck and Kazeier, 1985). Indeed, it is the variation in informal wage rates which perhaps highlights better than any other variable the segmented nature of the informal labour market by employment status.

It is not only the Quebec findings on the relationship between employment status and informal employment which are replicated in many other urban areas. When the gendered configuration of the informal labour market is analysed, the overarching finding of the vast majority of studies is that the extent of women’s participation is less than that of men and that men constitute the majority of the informal labour force. This is identitied to be the case in the Netherlands (Van Eck and Kazeier, 1985; Renooy, 1990), the UK (Macdonald, 1994; Pahl, 1984), Italy (Mingione, 1991; Vinay, 1987), Denmark (Mogensen, 1985) and the United States (McInnis-Dittrich, 1995). It is important to state, nevertheless, that the relative gap in the participation rates of men and women is not great and is perhaps not so surprising as it first appears when it is realized that men more frequently engage in paid employment than women in nearly every context.

Examining the types of informal employment men and women perform, it is striking that women undertake a very different set of tasks than men. Just as there is a clear gender segmentation by sector in the formal labour market, with women heavily concentrated in service sector jobs (e.g. Townsend, 1997), the same is true of the informal labour market. As Hellberger and Schwarze (1986) identify in Germany, whilst 12.3 per cent of informal workers are in the primary sector, 35.8 per cent in manufacturing and 51.9 per cent in services, these figures are 6.2 per cent, 11.8 percent and 82.0 per cent respectively for women (and 15.7 per cent, 49.3 per cent and 35.0 per cent for men). It is similar elsewhere. Women tend to engage in service activities such as commercial cleaning, domestic help, childcare and cooking when they undertake informal employment. Men, on the other hand, tend to undertake what are conventionally seen as ‘masculine tasks’ such as building and repair work (Fortin et al., 1996; Jensen et al., 1995; Leonard, 1994; Mingione, 1991; Pahl, 1984). There is also tentative evidence that in some particular contexts men’s participation in informal employment is more infrequent but full-time than women’s, which although continuous, tends to be part-time (Leonard, 1994; McInnis-Dittrich, 1995). So, not only is informal employment sectorally divided in similar ways to formal employment, but so does the part-time/full-time dichotomy which is often prevalent in formal employment appear to be replicated in the informal labour market. Women, moreover, generally earn lower informal wages than men (Fortin et al., 1996; Hellberger and Schwarze, 1986; Lemieux et al., 1994; McInnis-Dittrich, 1995). This is because women have to engage in informal employment which reflects their domestic roles (resulting in sectoral division) or which fits in with their domestic duties (resulting in the tendency for regular but part-time informal employment so that family needs can be met).

Finally, there is the issue of whether ethnic minorities and immigrants relate to the informal labour market in the same way as these other marginalized groups. Do ethnic minority and immigrant groups more frequently participate in informal employment than their white or indigenous counterparts? Is the informal labour market heavily composed of these groups? To answer these questions, two different groups need to be distinguished, each likely to have contrasting experiences with regard to the informal labour market. On the one hand, there are ethnic minority populations, and on the other hand, there are illegal immigrants. Given that different types of legal immigrant ultimately witness the same experiences as one of these two groups, they are here dealt with under these two headings (see Williams and Windebank, 1998).

Most of the research on this topic has been conducted in the USA and focuses upon sectors and/or occupations in urban neighbourhoods with high concentrations of ethnic minority populations, mostly in global cities. These identify that informal employment is closely associated with such groups (Fernandez-Kelly and Garcia, 1989; Lim, 1995; Portes, 1994; Sassen, 1989; Stepick, 1989). Analysing ethnic minorities, the problem is that even if they are more likely to work informally in some parts of the global cities, this does not mean that the vast bulk of the informal labour force in the advanced economies as a whole are ethnic minorities, as reading the US literature outside its context might imply. Indeed, where ethnicity has been considered as a variable alongside others in alternative geographical contexts, such as in rural Pennsylvania (Jensen et al., 1995), no strong correlation has been identified between ethnic minorities and participation in informal employment. The informal labour market, therefore, appears to be configured differently in different areas so far as ethnicity is concerned.

Similarly, it might be assumed that the participation of illegal immigrants in informal employment is very high, possibly the highest of any socio-economic group in the advanced economies. The reason is supposedly simple: excluded from formal employment as well as social security benefits due to their illegal status, such immigrants have little choice but to engage in informal employment as a means of survival. However, in reality, they do have other options. They can use falsified or other people’s documents (for example, national insurance or social security numbers) to gain access to either welfare benefits or employment without their employers’ collusion. Alternatively, they can work as an employee with the consent of the employer who pays taxes for the worker but the worker remains undetected as an illegal immigrant due to the imperfections in, and lack of coordination between, the various regulatory institutions of the state. Indeed, the Internal Revenue Service (IRS) in the USA has estimated that as many as 88 per cent of illegal immigrants may be paying taxes (cited in Mattera, 1985). It is therefore not the case that all illegal immigrants who work are engaged in informal employment. Although they obviously engage to a greater extent in informal activity than other groups, the picture is much more complex than is sometimes thought. The important point, however, is that even if the vast majority of illegal immigrants do engage in informal employment, this does not mean that all informal workers are illegal immigrants. As Wuddalamy (1991: 91) argues in France: ‘Immigrant clandestinity … only covers fragments of the underground economy’

However, when ethnic minorities and immigrants (whether legal or illegal) do engage in informal employment, the overwhelming finding is that they engage in peripheral exploitative forms of organized informal employment rather than in core, more rewarding autonomous types of work (Lim, 1995; Mingione, 1991; Moulier-Boutang, 1991; Portes, 1994; Portes and Stepick, 1993; Sassen, 1989; Stepick, 1989). Therefore, the informal labour market mirrors many of the racial inequalities prevalent in the formal labour market. As Moulier-Boutang (1991: 119) puts it, ‘all large world labour markets are hierarchical along ethnic lines: foreign workers and different nationalities in different legal positions occupy the posts at the bottom of the scale’.

Indeed, this is the principal finding so far as all marginalized groups are concerned. Nevertheless, there are localities where important exceptions to this rule can be identified. Other locality studies reveal that the unemployed engage in more informal employment than the employed. This is identified to be the case in Belfast (Howe, 1988; Leonard, 1994) and Belgium (Kesteloot and Meert, 1994; Pestieau, 1984). Pestieau (1984), for example, in the Liège area of Belgium, identifies that suppliers of informal labour were especially likely to be found among manual workers, unemployed or unoccupied persons, low income and uneducated groups and younger people. A similar configuration is also identified in Spain (Lobo, 1990a) and Greece (Hadjimichalis and Vaiou, 1980), albeit only so far as organized informal employment is concerned. Ethnic minorities and immigrants (whether legal or illegal), moreover, are not always found to occupy the most peripheral positions in the informal labour market. Rather, different ethnic minority and immigrant groups are to be found in different positions in the hierarchy of particular local informal labour markets and these positions can change over time as some groups progress up the hierarchy of paid informal activities and even formalize them (e.g. Sassen, 1989; Stepick, 1989). Informal employment, therefore, has a hierarchy of its own which mirrors the formal labour market so far as ethnicity and immigrant populations are concerned.

However, such exceptions should not hide the fact that the dominant way in which the informal labour market is segmented is very similar to the formal labour market. If marginal groups find themselves in exploitative informal employment, it is because they find themselves in this kind of work in formal employment, not because informal employment is comprised of this sort of work alone and these sorts of workers only.

Explaining the Magnitude and Character of Urban Informal Employment

To explain why the growth, magnitude and character of urban informal employment varies temporally and spatially, it is here argued that its configuration in any place is the result of the way in which economic, social, institutional and environmental conditions combine in multifarious ‘cocktails’ in different urban economies to produce specific local outcomes. The consequence, therefore, is that at any time, there is neither universal formalization nor informalization but, rather, different processes in varying urban areas. Neither is there one unique composition of the urban informal labour market which is homogenous across all places. Rather, there are different configurations in different areas. Here, therefore, we outline the various conditions and combination factors which regulate the configuration of informal employment in any urban economy through an analysis of the general ways in which each condition normally influences the nature and level of informal employment.

Economic Regulators

Many explanations of the magnitude and character of urban informal employment heavily rely on economic explanations (e.g., Benton, 1990; Portes and Sassen-Koob, 1987). Here, we outline the key regulatory variables considered to be influential in configuring such work. On their own, however, such explanations are insufficient. They fail to address the social, institutional or environmental context in which economic processes are embedded.

Level of Affluence and Employment

For many, there is a strong correlation between the magnitude of informal employment and the level of affluence and employment in a locality. Some, however, have found that the higher the level of affluence and employment in a place, the greater the amount of such work and the more autonomous and well paid is the informal employment (e.g., Leonard, 1994; Pahl, 1984; Renooy, 1990), whilst others have discovered the opposite (e.g., Frey and Weck, 1983; Matthews, 1983). On its own, therefore, this condition is insufficient to produce a particular configuration of informal employment (see Williams and Windebank, 1995b; Windebank and Williams, 1997). It is the way in which it combines with a range of additional economic, social, institutional and environmental regulators in a place to produce a particular configuration of informal employment which is important.

Industrial Structure

The vast majority of studies find that relatively little informal employment takes place in areas dominated by a few large companies because the skills acquired in such industries are less transferable to individual informal work and they less often use organized informal work than smaller firms (Barthelemy, 1991; Howe, 1988; Van Geuns et al., 1987). Informal employment usually flourishes, meanwhile, in local economies with a plethora of small firms (Barthelemy, 1991; Blair and Endres, 1994; Capecchi, 1989; Pahl, 1987; Sassen, 1996; Van Geuns et al., 1987) since trade unions are less active and organized, there is greater scope for employing organized informal work either directly or through sub-contracting arrangements and skills acquired are more likely to be useful in individual informal employment. Similarly, many have drawn a strong positive correlation between the level of self-employment and the extent of informal employment (e.g. O’Higgins, 1981; Pahl, 1990). As Mingione (1991) reveals, however, although tax evasion may be higher amongst the self-employed and small firms than the employed and large firms, not only is this just one part of all informal employment but the magnitude of informality amongst small firms and the self-employed varies considerably between areas due to a host of other factors, including the structure of taxation, level of affluence and employment. Thus, one cannot automatically read-off from the level of self-employment or small firms in a locality that informal employment is necessarily higher.

Sub-Contracting

In some localities, a link has been established between sub-contracting and the growth of informal employment (Benton, 1990; Thomas and Thomas, 1994). As workers previously employed in large firms have been forced to seek jobs in smaller marginal firms or to do piecework in their own homes, the result in economically vulnerable areas has been a weakening of the workers’ bargaining position, decreased wages, reduced working conditions and a loss of state benefits (Benton, 1991; Thomas and Thomas, 1994). Such a deterioration in working conditions is not inevitable. In other areas, the outcome is thriving enclaves of small informal enterprises that have adapted to changing market demands for specialized products while sustaining relatively high wages and good working conditions (Benton, 1990; Capecchi, 1989). This is because the effect of increased sub-contracting depends on the local economy and the local labour market conditions as well as the local political circumstances.

Tax and Social Contributions

Although some argue that this condition alone is sufficient to explain the growth of informal employment (e.g. Matthews, 1983), it is but one, albeit important, part of the ‘cocktail’ of explanatory factors which cannot be ruled out. As taxes rise in a nation or region, so the differential cost of engaging in and using formal rather than informal employment increases for both consumers and firms (Frey and Weck, 1983; Gutmann, 1978; Renooy, 1990). However, not all individuals will avoid such burdens by employing or offering themselves as informal workers. Instead, some consumers will engage in self-provisioning. Neither do all companies automatically switch some or all production to informal modes. It depends on what other options are open to them. Indeed, it is not just the level but also the structure of tax and social contributions which influences the configuration of informal employment. In places where taxes and social contributions are raised more from companies than individuals, organized informal employment might well be more prevalent, whilst if contributions are raised more from individual workers, then individual informal employment might be more likely (Barthelemy, 1991). The way in which the level and structure of taxation influences the configuration of informal employment, nevertheless, is dependent upon a range of other factors such as attitudes towards tax evasion, labour and welfare laws and the interpretation and enforcement of rules and regulations.

Social Regulators

Local and Regional Traditions, Social Norms and Moralities

Economic processes are often mediated by cultural norms, traditions and moralities. In some areas, informal employment is more acceptable than in others, such as when feelings of resentment and of being let down by the state lead to less acquiescence to the laws of the state (Howe, 1988; Legrain, 1982; Leonard, 1994; Van Geuns et al., 1987; Weber, 1989) or when informal employment is undertaken more for social than economic reasons (Barthelemy, 1991; Cornuel and Duriez, 1985; Komter, 1996). Moreover, tax morality (that is, the willingness to cheat on taxes) varies not only cross-nationally (Frey and Weck, 1983) but also intra-nationally, where tax morality is lowest in big cities (Chavdarova, 1995). The precise ways in which these cultural processes mediate economic activity, however, are shaped by the economic and institutional conditions of the locality as well as by other social regulators.

The Nature of Social Networks

Areas characterized by populations with dense social networks generally undertake a greater proportion of the total work using informal employment than areas with sparse or dissipated social networks, who use relatively more formal purchases (Legrain, 1982; Mingione, 1991; Morris, 1994; Van Geuns et al., 1987). Similarly, individuals with denser social networks tend to have more opportunity for informal employment than those without (e.g., Leonard, 1994; Warde, 1990). Dense social networks alone, however, are insufficient to lead to informal employment if their members are unable to pay for such goods and services (Lysestol, 1995). As Morris (1993) discovers in the UK, the long-term unemployed often only mix with other long-term unemployed and thus are less likely to be offered opportunities for engaging in informal employment. However, it must be remembered that she was studying an area characterized by an industrial structure undergoing contraction and dominated by large enterprises. Dense social networks in conjunction with other factors (for example, in areas with a wide socio-economic mix or rural areas) can lead to informal employment alleviating poverty, if not providing a complete survival strategy (e.g., Jessen et al., 1987; Weber, 1989).

Socio-Economic Mix

An area that combines a population with high incomes but little free time with a population who have low incomes but much free time will usually witness both high absolute and relative levels of informal employment (Barthelemy, 1991; Portes, 1994; Renooy, 1990). This, however, need not always be the case. Strict enforcement of social security regulations coupled with a low tax regime would not necessarily lead to such a configuration, since suppliers would feel restricted from participating and customers would receive lower marginal benefits from using informal rather than formal employment.

Educational Levels

The more educated the population, the more likely they are to supply informal labour and to perform autonomous well-paid informal employment, whilst those with fewer qualifications engage in more exploitative poorly paid informal employment (e.g. Fortin et al., 1996; Lemieux et al., 1994; Renooy, 1990). Again, however, there will be exceptions depending on the way in which this human capital factor is mediated by other economic, institutional, social and environmental conditions in any area (for example, the structure of tax and welfare benefits and the degree to which they are enforced).

Institutional Regulators

Besides economic and social regulators of the level and character of informal employment, there are also a range of institutional conditions that regulate such work.

Labour Law

By definition, informal employment is created by government rules and restrictions (e.g. Gutmann, 1978). As labour law differs between nations, so too does the level and nature of informal employment. In some nations, for example, industrial homeworking is illegal whilst in others it is not (Phizacklea and Wolkowitz, 1995), which influences the type of work undertaken informally and who participates. Such legal differences thus have a heavy influence on the level and nature of informal employment.

Welfare Benefit Regulations

The unemployed in countries with poor access to permanent state benefits are more likely to engage in informal employment as a survival strategy because they have little to lose if caught (Del Boca and Forte, 1982), thus helping to explain the higher levels of informal employment in southern EU nations and the USA. In contrast, those eligible to claim more generous benefits will have the major disincentive of losing them if caught engaging in informal employment, as Wenig (1990) highlights in Germany. Again, however, this general tendency will be mediated by other economic, institutional, social and environmental factors such as local cultural traditions, social networks and the buoyancy of the local labour market as well as by the extent to which such regulations are enforced.

State Interpretation and Enforcement of Regulations

In addition to actual differences in rules and regulations, how they are interpreted and enforced will influence the level and nature of informal employment. Some authorities, both national and local, not only actively support informal employment so as to enable firms to compete in international markets and/or to help individuals and families raise adequate incomes, which would be difficult if ‘regular’ regulations were strictly enforced (Cappechi, 1989; Lobo, 1990a; Portes and Sassen-Koob, 1987; Van Geuns et al., 1987; Warren, 1994), but also passively tolerated its existence through lax enforcement (e.g. Lobo, 1990b; Van Geuns et al., 1987). This again is insufficient alone to facilitate the growth of informal employment.

Corporatist Agreements

Labour is not only regulated by the state but by industry, company and individual agreements which can limit the rights of employees to take on outside jobs, do overtime and so on. Informal employment can be an attempt to circumvent these rules, which thus become a factor in local and national variations in the level and character of informal employment (e.g., Lobo, 1990b).

Environmental Regulators

Settlement Size

Although little is known about how urban settlement size influences the magnitude and character of informal employment, some studies show that rural areas undertake more informal employment than urban areas (e.g. Jessen et al., 1987; Legrain, 1982), whilst other show the opposite (e.g., Fortin et al., 1996, Mogensen, 1985). The implication, therefore, is that urbanity alone is an insufficient determinant of the level of informal employment. It depends upon the presence or absence of the range of additional factors discussed above. Furthermore, it appears to be the case that rural areas are characterized more by autonomous forms of informal employment (e.g. Renooy, 1990), whilst organized informal employment seems to be more concentrated in urban areas (e.g., Portes and Stepick, 1993; Sassen, 1991). Again, however, this depends upon a range of additional economic, social and institutional characteristics of the area and how these combine to produce particular configurations of informal employment.

Type and Availability of Housing

Housing tenure seems to be an important determinant of the quantity and quality of informal employment in a locality. Areas with high levels of privately owned households undertake a wider range of informal work than areas dominated by rented accommodation (Morris, 1994; Pahl, 1984; Renooy, 1990). Moreover, owner-occupiers have been shown to be more likely to use informal employment to get a job done than tenants (Mogensen et al., 1985; Renooy, 1990). Again, whether this always holds depends upon the presence or absence of the other factors discussed above.

In sum, the level and character of informal employment in a place is the result of the way in which economic, social, institutional and environmental conditions combine in multifarious ‘cocktails’ in different places to produce specific local outcomes. As Mingione (1990: 42) asserts in relation to the spatial variations in Greece, ‘Everywhere, the mixes appear to be very complex and fundamentally based on a range of combinations of local opportunities.’ The temptation, therefore, is to compensate for the previously overgeneralized and simplistic typologies concerning the urban geography of informal employment by concluding with such an idio-graphic approach. We could conclude, similar to Smithies (1984: 89), that ‘Black economies are as different from one another as “legitimate” economies’. However, to leave it at this would be to go too far to the other extreme, ignoring some important parallels and commonalities in the extent and character of informal employment between and within urban areas.

Take, for example, deprived urban areas in the advanced economies. These localities are usually characterized by high unemployment, economic vulnerability, little support or sympathy for people (especially the unemployed) working informally, a poor socio-economic mix, below-average educational qualifications, authorities stringently implementing rules and regulations concerning benefit fraud, tax evasion and labour laws, and low levels of private ownership of housing. The result of this cocktail of factors is low levels of informal employment and that informal employment which does exist tends to be mostly of the exploitative variety (Barthe, 1988; Foudi et al., 1982; Howe, 1988; Jordan et al., 1992; Morris, 1995). This is because the population possesses the free time but lacks the additional resources and opportunities necessary to engage in a wide array of informal employment. They lack the skills, tools and materials to participate in such work, the social networks to hear about informal employment opportunities, fear being reported to the authorities and have less scope for undertaking such work in the areas in which they live. The few opportunities for informal employment which do exist, moreover, tend to be in the form of exploitative organized informal employment (for example, contract cleaning, fruit picking, unlicensed taxi-driving, labouring, kitchen work). This would not perhaps be problematic so far as the urban geography of social inequality is concerned if these areas occupied only a minor space in the urban landscape of the advanced economies. The serious issue for socio-spatial disparities, however, is that this is the dominant way in which informal employment is configured in the majority of deprived urban areas in the advanced economies. This has major issues for the socio-spatial divisions of not only informal employment but standards of living in general.

Given the dominance of this particular cocktail in the deprived urban areas of our advanced economies, it can be argued that the principal outcome of integrating informal employment into the analysis of the growth/decline of urban economies is that it further exacerbates, rather than reduces, the overarching socio-spatial inequalities. On the whole, those urban areas witnessing contraction of their formal economies are also witnessing a decline of their informal work and vice versa. This applies both on an intra- and inter-urban level. Although, in specific situations, there are exceptions to this rule, the dominant way in which the relationship between formal and informal employment manifests itself in the contemporary urban landscape is one which reinforces rather than mitigates the socio-spatial inequalities produced by formal employment. What, therefore, are the available policy options towards informal employment and what are the implications of pursuing each option?

Policy Options and Implications

To examine the policy options and their implications, we here examine the three most popular policy approaches in the literature: the regulatory approach, which seeks to eradicate informal employment through tougher rules and regulations; the neo-liberal approach, which seeks to integrate formal and informal employment by deregulating formal employment; and the new economics approach, which seeks a radical restructuring of work and income and, in so doing, to abolish the necessity for people to undertake exploitative informal employment whilst liberating them to more fully participate in other forms of work. Each is here evaluated in turn in terms of, first, their vision for work and welfare and, secondly, the practicability and desirability of their approach towards informal employment. Before commencing, however, it is necessary to note that few, if any analysts advocate the development of informal employment. That is a common misconception. Instead, and in all approaches, the intention is to formalize informal employment as a means of harnessing its productive energies.

Given the way in which the informal labour market generally tends to reinforce rather than reduce the socio-spatial disparities produced by the formal labour market, it might seem superficially that the overwhelming popularity of the regulatory approach with its desire to eradicate informal employment through tougher regulatory conditions is the most appropriate response. Seeking the regulation of informal employment due to its fraudulent (e.g., Feige, 1990; ILO, 1996) and/or exploitative nature (e.g., Portes, 1994), the goal of this approach is to replace such work with full employment and, failing this, a formal welfare ‘safety net’. The principal problem, however, is that little consideration is given to the notion that full-employment and/or a comprehensive welfare state might be unachievable and thus, that people need to be given alternative ways of engaging in meaningful and productive activity both as a social entitlement and as part of a policy of active citizenship. There are also major problems with the feasibility of seeking the eradication of informal employment. Not only are there ‘resistance cultures’ to this policy in governments, such as in southern Europe (Benton, 1990; Cappechi, 1989; Mingione, 1991), but there is also the problem that because informal employment is deeply entrenched in everyday urban life and frequently undertaken as much for social as for economic reasons (e.g., Fortin et al., 1996; Komter, 1996), it will be near impossible to eliminate. Moreover, given that it is simpler to eradicate organized than autonomous informal employment, and that the former is more likely to be undertaken by marginalized populations, the unintended consequence of such an approach is likely to be the further exacerbation of socio-spatial inequalities.

The implication, therefore, is that any attempt to eradicate informal employment through more stringent regulations will not only fail but will extenuate current socio-spatial inequalities in our urban economies. Take, for example, current attempts to introduce tougher regulations through ‘welfare-to-work’ programmes in the USA and UK. Although little, if any, research has been conducted in either nation in terms of its impacts on informal employment, it seems likely that the principal consequence will be merely to exacerbate socio-spatial disparities by constraining the ability of the unemployed to supplement their benefit through informal employment whilst leaving the already employed free to continue with their informal working. Research, nevertheless, is badly required on the implications of ‘welfare-to-work’ programmes before any firm conclusions can be reached. What, for example, happens to people who fall out of welfare-to-work programmes? Do they depend on kin for support, find formal jobs, get by in informal employment, turn to crime or end up in prison? Currently, nobody knows the full consequences of the introduction of these tougher welfare approaches.

The second approach, similar to the regulatory approach, also wants to see a return to full employment but has a different means of achieving it. Rather than bolstering the flagging edifices of welfare capitalism by regulating informal employment, this neo-liberal approach advocates de-regulating formal work and welfare so as to close the gap between formal and informal employment (e.g., Matthews, 1983; Sauvy, 1984; de Soto, 1989). Although stripping away formal regulations would greatly reduce informal employment since such a form of work is by definition a product of such regulations, there is little reason to believe that this would do much to relieve socio-spatial inequalities. Indeed, the only possible reason that greater economic competitiveness and socio-economic equity would be achieved through de-regulation is because there will be a levelling down rather than up of material and social circumstances as urban economies engage in a race to the bottom.

The final policy stance towards informal employment is the ‘new economics’ approach. Similar to the de-regulationists, the existence of informal employment is seen as an indictment of the present system, but not of the over-regulation of the economy and welfare provision by the state. Instead, it is an indictment of the present system of distributing income and work in society, of which exploitative informal employment is a symptom. This approach, therefore, advocates a radical restructuring of work and income and, in so doing, seeks to abolish the necessity for people to undertake exploitative informal employment whilst liberating them to more fully participate in other forms of work. Assuming that full-employment and comprehensive welfare provision are impractical and undesirable goals, this perspective seeks to replace the employment-focused mentality of the above approaches with a ‘whole economy’ approach based on ‘full-engagement’ (e.g., Lipietz, 1992; Mayo, 1996) through the introduction of a citizen’s income and greater assistance to help people help themselves. However, there are major difficulties over the practicality and desirability of implementing this approach (see Williams and Windebank, 1998). So far as the basic income scheme is concerned, there are many imponderables concerning not only both the cost of implementing it and whether there is the political will to do so, but also whether it will significantly reduce either the level of informal employment in general or exploitative informal employment more particularly. Moreover, there is the issue that the new economists are racing ahead with implementing self-help initiatives before they have a clear conception of the barriers to participation of different groups.

In sum, and having reviewed three different approaches towards urban informal employment, we can see that the option chosen is very much dependent upon the wider issue of the future for work and welfare desired. Should we try to return to the commodification of both work and welfare in the form of full employment and the welfare state, despite all of the evidence suggesting that this is not practical or desirable? Or, alternatively, should we pursue a future of work and welfare in which the formal labour market is de-regulated and the welfare safety net taken away so as to try to recapture full employment? Or should we accept that full employment is gone and seek to restructure work and welfare using the whole economy and do so in a manner which helps people to help themselves and pursue greater self-reliance? These are ultimately the questions that need to be answered when deciding what can and should be done about informal employment in the advanced economies. What is certain, however, is that the laissez-faire option is not available. To do this will only consolidate through the informal labour market the socio-spatial disparities and injustices perpetrated by formal employment.