Working Papers Series #13
ISSN 2045-5763 (online)

Intimations of the ‘Informal Economy’
In the Work of Henry Mayhew, P T Bauer and Richard Salisbury

John D Conroy1
Australian National University

© 2012 John D Conroy
Creative Commons License
Open Anthropology Cooperative Press

Forum discussion in the OAC network.
Download as PDF, EPUB, MOBI.



This paper considers the idea of informality in market exchange, as introduced into the economic development literature by Keith Hart in the 1970s. In addition to Hart (1971, 1973) it will discuss three writers who may be considered his intellectual forerunners. Each, to a greater or less degree, anticipated the idea of informal economic activity and described it in a particular historical period and place. They are the mid-Victorian journalist Henry Mayhew (London, c.1850), the libertarian economist P. T. Bauer (British West Africa, c.1948) and the economic anthropologist R. F. Salisbury (colonial New Guinea, c.1952-1963). The principal texts relied upon are Mayhew’s monumental London Labour and the London Poor (4 vols, 1851-61), Bauer’s Economics of Under-Developed Countries (1957) and Salisbury’s From Stone to Steel (1962) and Vunamami: Economic Transformation in a Traditional Society (1970).

Keywords: informal economy, informal sector, Keith Hart, Henry Mayhew, P. T. Bauer, Richard Salisbury


Table of contents

  1. The Idea of the Informal Economy
  2. Keith Hart and the Informal Economy in Newly-Independent Ghana
  3. The Informal Economy in History and Literature: Henry Mayhew
  4. P T Bauer and the Transition from Subsistence to Exchange
  5. Richard Salisbury and Services as ‘Leading Sector’
  6. Conclusion
  7. Bibliography


Outline of the project

This paper forms the introductory part of a monograph in which Papua New Guinea (PNG) is the main case study. It will also include a section on ‘original’ or ‘subsistence affluence’, along with another on traces of an informal economy in the pre-contact and colonial anthropological literature of New Guinea, and in the limited economic history material available. The question there is why the evidence of urban informal economic activity was so limited at the time of the Development Strategy report (the Faber Mission) in 1972-73, on the eve of Independence.

The final section will deal with the Faber Mission and the reception of its recommendations on the informal sector. It will review such evidence as I can find that it was (in that respect) of any continuing influence. Primarily, however, it will deal with the emergence of informal economic activity in the intervening 40 years, during which population almost trebled and a lot of other changes took place. These included the impoverishment of large sections of population despite periodic injections of mineral wealth. Subsistence ‘affluence’ has diminished progressively, urban populations have become more entrenched (‘trapped’) and less connected with rural origins, the State has developed the full apparatus of grand and petty corruption, social differentiation (‘class formation’) has proceeded a long way in both urban and rural areas, while ‘national capitalism’, such as it is, consists largely of rent-seekers rather than genuine entrepreneurs.

And yet the PNG informal economy is still ill-developed and marked more by violence than creativity. My study will look for explanations of this phenomenon by harking back to anthropology and history to understand the limited precedents for any fine-grained division of labour in a PNG informal sector, and consider the extent to which the continuation of colonial attitudes and policies by the national elite offers some explanation of what has occurred.


  1. The idea of the informal economy

This paper is concerned with economic activity, seen in terms of ‘formality’ and ‘informality’, in the context of market exchange2. The growth of market exchange is associated with the emergence of bureaucracy, which creates rules to give form to its own activities and to those of the regulated. In this way market activity becomes enumerated, and the status of being enumerated is a crucial marker of ‘formality’. By contrast, ‘informal’ activities occur outside the strictures of bureaucracy and indeed may transgress its requirements. Informal activities may avoid certain obligations but must also forgo at least some of the benefits of conformity with the requirements of bureaucracy.

This study commences with a discussion of the pioneering account of an informal economy, in early-independent Ghana (c. 1966), by the British anthropologist Keith Hart (1971, 1973). It will then focus on informality in market exchange, as described historically in a number of other ‘emerging’ economies. These are mid-Victorian London (c.1850) and the late-colonial dependencies of British West Africa (c.1948) and Papua New Guinea (c.1960). The paper will consider the forms associated with market exchange in these disparate settings, and the deviations from form, asimplied by the term ‘informal economic activity’, occurring in each. Some incongruity is apparent among the three ‘cases’ offered for comparison with Hart’s work, which are widely separated in time and space. Nonetheless, their authors, respectively, Mayhew (1851, 1861), Bauer and Yamey (1951, 1957) and Salisbury (1971), have strong claims to be regarded as forerunners of Keith Hart’s informal economy. They are almost certainly not the only such. Hart has himself acknowledged Mayhew, together with Clifford Geertz (1963) and Oscar Lewis (1966), among his precursors, citing, in particular, the latter’s concept of the ‘culture of poverty’ (Hart 2010, 143).

Whatever about the contributions of earlier scholars, the idea of the ‘informal economy’, as it has developed now over a period of some 40 years, originated with Keith Hart. In a recent account of its intellectual history he noted that

‘The informal economy (or informal sector) became current in the early 1970s as a label for economic activities that escape state regulation. It arose in response to the proliferation of self-employment and casual labour in Third World cities; but later the expression came to be used with reference to industrial societies, where it competed with similar epithets – the ‘hidden’, ‘underground’, ‘black’ economy, and so on’. The social phenomenon is real enough and of some antiquity, but its definition remains elusive (Hart 2010, 142).

This paper focuses on Hart’s original conception of the informal economy, and deals only incidentally with the subsequent elaboration of it by him and others in the context of ‘de-industrialization’ and globalization3. This more narrow focus is adopted because in its original form the idea has proved influential among development economists and policymakers. For example, in terms of theory it led to the re-examination of propositions in the classic paper by Arthur Lewis, Economic Development with Unlimited Supplies of Labour (Lewis 1954) as well as in Michael Todaro’s influential model of rural-urban migration (Todaro 1969). In terms of policy it was taken up by international agencies (including ILO and the World Bank) as a remedial measure for urban unemployment (see, for example, ILO [1972]). It continues to condition perceptions of the processes of urbanization, employment and income distribution in developing countries and may still be applied fruitfully to them. Nor has Hart modified his original conception in any way that would discourage its application to current conditions of development and underdevelopment.

As an introduction to Hart’s ideas it may be useful to consider his views on bureaucracy, since bureaucratic activity substantially determines the character, and designation, of economic activity as either formal or informal. Hart noted Hegel’s contention that ‘Society … should be managed by an educated bureaucratic elite in the national interest’, and that Max Weber had ‘recognized such a synthesis in Germany’s historical experience of the alliance between Rhineland capitalists and Prussian bureaucracy’ (Hann and Hart 2011, 30). For Hart, bureaucracy is an essentially positive construct, ‘invented as part of a democratic political project to give citizens access to what was theirs by right’, although in practice it is often seen to operate as ‘the negation of democracy’ (Hart 2004).

When Hart arrived in Ghana in the mid-60s, he was influenced by Arthur Lewis’s dualistic model of developing economies. As he recounted, ‘the conceptual pair ‘formal/informal’ grew out of [my] attempt to figure out what happened to agricultural labour when it migrated to cities whose markets were only weakly organized by industrial capitalism. The formal and informal aspects of an economy are linked of course, since the idea of ‘informality’ is entailed by the institutional effort to organize society along formal lines … [Thus] “form” is the rule, an idea of what ought to be universal in social life; and for most of the twentieth century the dominant forms have been those of bureaucracy, particularly of national bureaucracy, since society has become identified to a large extent with nation-states’ (Hann and Hart 2011,114-15, emphasis in original).

Another source of Hart’s formal/informal duality was found in the work of Clifford Geertz, in which national bureaucracy again played an important role. In Pedlars and Princes,

‘Geertz identified two economic ideal-types in a Javanese town. The majority were occupied in a street economy that he labeled “bazaar-type”. Opposed to this was the “firm-type” economy consisting largely of western corporations who benefited from the protection of state law. These had form in Weber’s sense of “rational enterprise” based on calculation and the avoidance of risk. National bureaucracy lent these firms a measure of protection from competition, thereby allowing the systematic accumulation of capital. The “bazaar” on the other hand was individualistic and competitive, so that accumulation was well-nigh impossible’. (Hart 2004)

The commercial impotence of petty ‘bazaar’ entrepreneurs, notwithstanding they ‘were rational and calculating enough to satisfy Max Weber on ideological grounds’, was due to their being ‘denied the institutional protection of state bureaucracy’. In consequence of this exclusion, ‘their version of capitalism remained stunted at birth’ (Hann and Hart 2011, 113). The duality of the formal and informal was produced by bureaucratic action impacting actors differentially. Hart developed these ideas in the setting of early-independent Ghana, where the state’s reach ‘only extended so far into the depths’, creating a situation in which economic activity was indeed ‘individualistic and competitive’, and accumulation difficult.


  1. Keith Hart and the Informal Economy in Newly-Independent Ghana

During fieldwork in Ghana over the period 1965-68 Hart worked in Nima, a slum district of Accra where 40 per cent of active males and fully 95 per cent of females were ‘not touched by formal employment’ (Hart 1973, 63). Intending to prepare an orthodox ethnography of the Frafra ethnic group, he found himself caught up in their precarious economic existence and opted instead to document their livelihoods. As rural-urban migrants, generally with little Western education and limited formal skills, most Frafra ‘got by’ on wages from unskilled jobs, earnings from self-employment and ‘transfers’.

Hart’s observations stressed high levels of ‘unemployment’ among the Frafra (in the then conventional sense of non-participation in regular ‘waged’ jobs), the inadequacy of earnings in such jobs, the consequent incidence of multiple job-holding (‘moonlighting’) among waged workers, and their frequent resort to additional, non-wage, sources of income.Those Frafrawithout access to ‘formal’ employment (as we now think of it) subsisted on earnings from casual employment and by self-employment in the production of a wide variety of goods and services. Most people in the local economy of Nima, and in similar neighbourhoods of Accra, actively sought out multiple sources of income. Many benefited from ‘transfers’ received by virtue of ethnicity and/or membership of social networks (as well as incurring their own responsibility to make such transfers). Resort to credit was pervasive and frequent, serving to smooth consumption and to meet contingencies, but impelling many into a downward economic spiral, leading to desperation and the disruption of lives. Hart (1973, 65) described their situation memorably:

‘[F]aced with the impossibility of making ends meet, the urban worker of 1966 often ran up considerable debit accounts, used some of his pay to settle a few bills, went on a short-run binge until penniless, and spent the majority of the month in penury and increasing debt, relying on extended credit facilities and a wide range of putative kin and friends to provide occasional meals, and even lodging, if necessary. The pattern of everyday economic life for these workers was thus a hand-to-mouth existence characterised by unevenness of expenditures over a pay period, flexibility of consumption units, and the proliferation of credit in all commodities. Haushalten (budgeting), one of Max Weber’s two types of rational economic activity, is not widespread in places like Nima’.

Hart stressed the unreliability of non-wage incomes in Nima rather than their inadequacy. Returns to non-wage employment varied widely from period to period, around mean levels often higher (and occasionally much higher) than unskilled wages. Because of this unreliability, men were often reluctant to surrender low-paying waged-employment, opting instead to juggle the demands of such work with the opportunistic, though sporadic, activities of the informal economy. The latter he grouped in a classification distinguishing between ‘legitimate’ and ‘illegitimate’ activities, with the notion of ‘legitimacy’ being ‘derived essentially from Ghana’s laws … coincid[ing] with the morality of “respectable” Ghanaians’. Such notions were, however, not necessarily congruent with the mores of the residents of Nima, a district ‘notorious for its lack of respectability, for the dominance of a criminal element, and for the provision of those goods and services usually associated with any major city’s “red-light district”’ (Hart 1973, 74).

Among ‘legitimate’ activities Hart distinguished numerous categories of primary (including urban agriculture), secondary (petty manufactures, artisanship) and tertiary activities, conducted by an ‘urban proletariat’ (Hart 1971) or ‘sub-proletariat’ (Hart 1973). The latter included activities at the apex of the informal economy, requiring relatively large capital inputs (‘transport operators, landlords, cornmill owners, commodity speculators’). There was a profusion of small scale distributive roles, distinguished by whether operators were itinerants from ‘upcountry’ or resident in the city, by the locus of their trading (‘market stalls, roadside booths, hawking’), by the commodities traded, and by position in the distribution chain. Traders differed in terms of their time-input and scale of operations but were typically flexible in response to opportunity. ‘Petty traders, brokers, wholesale merchants, commission agents, and occasional dealers – all these roles are played in varying degrees by large numbers of the urban sub-proletariat’ (Hart 1973, 72).

Concerning ‘illegitimate’ activities, Hart (1973, 68) noted that ‘a consideration of income opportunities outside formal employment must include certain kinds of crime. The incidence of illegitimate activity in Nima was, to my knowledge, all-pervasive’. His classification (1973, 69) of criminal ‘income opportunities’ may be summarized as follows:

  1. Services: hustlers, spivs, receivers, usurers and pawners at illegal interest rates, drug-dealers, prostitutes, procurers, smugglers, bribe-takers, influence pedlars and protection men
  2. Transfers: thieves, larcenists, peculators and embezzlers, confidence tricksters, gamblers

In considering the formal/informal distinction,‘[t]he key variable is the degree of rationalisation of work – that is to say, whether or not labour is recruited on a permanent and regular basis for fixed rewards. Most enterprises run with some measure of bureaucracy are amenable to enumeration by surveys and – as such – constitute the ‘modern sector’ of the urban economy’. The unenumerated, those who were self-employed or casual workers, constituted the informal economy (Hart 1973, 68). Yet there were multiple relationships between the formal and informal economies. For example, in terms of ‘legitimate’ goods and services, the informal economy provided a very large proportion of the daily needs of the residents of Nima and other such suburbs, but it was also a substantial ‘exporter’ of goods and services to middle class neighbourhoods. Similarly, in regard to ‘illegitimate’ goods and services, one might distinguish between the thriving local market within Nima and its dealings with the ‘bourgeois’ world of greater Accra, which was the willing recipient of many services as well as the unwilling source of certain ‘transfers’ (owing to theft and burglary, for example). Urban crime was thus a powerful agent of redistribution4.

Hart’s account of Nima demonstrated that, taking the legitimate with the illegitimate, ‘income and expenditure patterns are more complex than is normally allowed for in the economic analysis of poor countries’. In this early work Hart felt unable to resolve the issue with which he had commenced his research: whether ‘informal economic activities possess some autonomous capacity for generating growth in the incomes of the urban (and rural) poor’. Only the application of more refined tools of analysis would enable economists to understand the structure and dimensions of the urban economy and how much the informal economy, in both its legitimate and illegitimate manifestations, was contributing to it (Hart 1973, 61, 89)5.

Finally, it will be useful to relate Hart’s observations of Nima to his understanding of Weberian bureaucracy. The poor in Accra were not ‘unemployed’, in the sense this term was understood at the beginning of the 1970s. Rather,

‘[t]hey worked, often casually, for erratic and generally low returns; but they were definitely working. What distinguished these self-employed earnings from wage employment was the degree of rationalization of working conditions. Following Weber (1981), I argued that the ability to stabilize economic activity within a bureaucratic form made returns more calculable and regular for the workers as well as their bosses. That stability was in turn guaranteed by the state’s laws, which only extended so far into the depths of Ghana’s economy. The ‘formal sector’ consisted of regulated economic activities and the ‘informal sector’ of all those, both legal and illegal, lying beyond the scope of regulation. I did not identify the informal economy with a place or a class or even whole persons. Everyone in Accra, but especially the inhabitants of slums like Nima where I lived, tried to combine the two sources of income’ (Hart 2004).

Hart contended that the stabilization of economic activity within a bureaucratic framework, assuring a degree of certainty to both labour and capital, occurred only in the upper reaches of the economy of Accra. This observation may be taken as a motif for the workings of the economy of mid-Victorian London, as observed by Henry Mayhew, in which uncertainty and insecurity marked the lives of a multitude6.


3. The informal economy in history and literature: Henry Mayhew

Henry Mayhew was a Victorian journalist whose three volume study London Labour & the London Poor7, was published over the decade from 1851.To read Mayhew’s reports is to be reminded that the early period of industrialization in Britain involved much the same phenomena – urbanization and population movement, technological change and social dislocation – as are at work in the contemporary developing world. Victorian London had a significant and complex informal economy, supporting (as the novelist’s biographer tells us) ‘that toiling mass which we may briefly glimpse in the pages of Dickens but which he could not present in all its enormity’ (Ackroyd 1990, 678). ‘It was all there before [Dickens], on every corner in every street; and perhaps what is more remarkable still is the extent to which the poor represented not so much an underclass as a state within a state’. For Mayhew’s London was not part of that England of the ‘two nations’ depicted by Disraeli (1845). The ‘state within the state’ contained an underclass denied even such protection as the Factory Acts afforded the industrial worker or the Master and Servant Act the domestic. This was a ‘sub-proletariat’ sharing many characteristics with Hart’s Frafrain Nima8. In hindsight, both factory workers and servants should, by virtue of employment contracts and legislation, be regarded as ‘formal economy’ workers, while those denied protection by legislation were ‘informal’. Formal sector workers were only incidentally the subjects of London Labour and the London Poor.

Others had also remarked this aspect of London society. A contributor to the Quarterly Review commented in 1855 that

‘the most remarkable feature of London life is a class decidedly lower in the social scale than the labourer, and numerically very large, though the population returns do not number them among the inhabitants of the kingdom, who derive their living from the streets … for the most part their utmost efforts do little more than maintain them in a state of chronic starvation’ (cited by Jones 1971, 12).

Why this class existed in London (and essentially only in London among English towns and cities) is clarified by a reading of Outcast London, a study of the period by Gareth Stedman Jones (Jones 1971). Industrialisation and the factory system had emerged in northern cities and towns, whereas London was a commercial and financial hub and ‘a “national emporium” at the heart of the transport and distribution network’ (Jones 1971, 20). High rents in the city and its distance from coalfields rendered large-scale factory production increasingly uncompetitive, with London actually experiencing a degree of de-industrialization during the first half of the nineteenth century, although technical change enabled certain industries to survive, and even to thrive. ‘The invention of the sewing-machine in 1846 and the band-saw in 1858, and the adoption of mass sewing and cutting from 1850 provided the technical pre-conditions for a large scale ready-made clothing industry’, while ‘the use of steam power in sawmills from the end of the 1840s enormously accelerated furniture production’ (Jones 1971, 107).

Such innovations set the stage for ‘sweated’ production, maintaining competitiveness by economizing on wages and rents (the latter through rapid expansion of home-working). Innovation in management, and the control of trades by large wholesalers, facilitated ‘vertical disintegration’, breaking down production into multiple tasks suitable for semi- and unskilled labour. This gave greater flexibility to entrepreneurs and facilitated sub-contracting by ‘small masters’, although at the cost of the casualization and impoverishment of much of the remaining industrial workforce, for:

‘Once the technical problems had been solved … and the conditions of mass demand for cheap ready-made goods had been established by rising working class prosperity, manufacturers were able to take advantage of a cheap, overfilled, unskilled labour pool of women and immigrants who were prepared to work at sub-subsistence wages’ (Jones 1971, 22).

Manchester at mid-century, by contrast, was described as having been ‘the great symbol of the hopes and fears of the age’. Social thinkers had expectations for ‘the new society being forged by the Industrial Revolution’, in which much was expected of a ‘new northern factory proletariat’, among whom were emerging ‘working-class institutions embodying self-help, sobriety, and religious dissent’ (Jones 1971, 11). For comparison with London Jones quoted Beatrice Webb. She spoke of the industrial north and of ‘the other side of the process through which bad workmen and bad characters are attracted to the large town’. Webb saw this as due to an ‘outcasting force’, for ‘there are no odd jobs in a small community which depends on productive industries. Unless a man can work regularly he cannot work at all. Then a bad character is socially an outcast, the whole social life depending on the chapel and the co-op’. Jones concluded that ‘[t]he extensive survival of small-scale production in Victorian London determined that its economic structure, its social and political character, and its patterns of poverty remained largely distinct from those of other nineteenth-century industrial regions’ (Jones 1971, 32). For whereas industrialization, regulated by Victorian bureaucracy, had fostered a modern, formal labour force in the north, in London ‘the effect of the Industrial Revolution … was to accentuate its “pre-industrial” characteristics’ (Jones 1971, 26).

The world’s largest city, with over 2 million people in 1851, London was about twice the size of Paris. Its concentration of unemployed and casual labour was unprecedented. Seasonality of economic activity underlay the casualization of labour, contributing to ‘the glut of unskilled labour for which the capital was renowned’. In the ‘sweated’ trades a large part of the workforce ‘existed in ‘a casualized limbo, filling in its time between short periods of employment by invading an already overfilled general unskilled labour market’ (emphasis added). Some higher skills were transferable; for example the unemployed piano maker might turn to cabinet making, ‘but the marginal cabinet-maker could only turn to street-selling, firewood chopping or the docks’ while for the unemployed docker ‘the only solutions were either some form of casual employment, reliance upon his wife’s wages, or in the last resort, charity and the poor law’ (Jones 1971, 42-3).

In peopling his novels with the poor of these classes, Dickens ‘had no desire gratuitously to alienate his largely middle-class public, and took care to retain the social and sexual proprieties of the period’. There were ‘some things that could not be said’. That was left to Mayhew, whom Dickens knew personally and on whose work he may occasionally have drawn for colour and authenticity. As Ackroyd remarks, ‘there were terrible stories in Mayhew’s account, stories of the poor and desperate … more terrible than anything described by Dickens himself’ (Ackroyd 1990, 677). Certainly, there were degrees of desperation and poverty among the London poor and some identifiable groups were relatively better off. Upward mobility is one theme of the narrative, just as, in Nima, Hart was able to describe individuals who had accumulated capital and become relatively large operators in Accra’s informal economy.

Mayhew was more than a chronicler; he was a humanely engaged and analytical observer. He described six ‘distinct genera’ among the working poor of the streets. These included ‘sellers’, ‘buyers’ and ‘finders’. Then there was a diverse group of ‘performers, artists and showmen’ (with ‘many of the characteristics of the lower class of actors’) as well as ‘artisans or working pedlars’. These were described as ‘struggling men, who, unable to find any regular employment at their own trade, have made up a few things, and taken them to hawk in the streets, as the last shift of independence’. Finally, there was an army of ‘labourers’, engaged for the performance of a single service, or by the hour or the day, but all beyond the jurisdiction of Master and Servant legislation (Quennel 1984, 29-31).

Each of these six categories is sub-divided in extraordinary detail, showing the human diversity and ingenuity, as well as the desperation, to be found in the street economy of mid-century London. Mayhew practiced a kind of political arithmetick, attempting, in an age before social surveys and random sampling, to estimate the numbers of the poor sustained by one wretched occupation or another, calculating their earnings and contribution to gross output, cataloguing literally hundreds of trades and occupations in the process, often with wry humour. Thus a buyer of hare and rabbit skins, reflecting on the seasonality of his trade, tells Mayhew that ‘hareskins is in … from September to the end of March, when hares, they says, goes mad’ (Quennell 1984, 284). But rather than smiling at the quaintness of Mayhew’s descriptions, the modern reader might instead reflect on the spirit of inquiry and originality of thought that underlay his observations and calculations.

Mayhew has come, on the left at least, to be regarded as a kind of proto-ethnographer (Swift 2011). No doubt this evaluation owes as much to his sympathy for the underdog as to his analytical qualities. But Mayhew’s curiosity is certainly that of the social scientist. Consider, for example, his speculations concerning the origins of human occupations; in effect, a research agenda:

‘It would be … a curious inquiry to trace the origins of the manifold occupations in which men are found to be engaged in the present day, and to note how promptly every circumstance and occurrence was laid hold of, as it happened to arise, which appeared to have any tendency to open up a new occupation, and to mark the gradual progress, till it became a regularly-established employment, followed by a separate class of people, fenced round by rules and customs of their own, and who at length grew to be both in their habits and peculiarities plainly distinct from the other classes among whom they chanced to be located (Quennell 1984,317)’.

This passage, drawn from Mayhew’s discussion of the Thames ‘dredgermen’, shows an appreciation of what Hart calls ‘form’. The working lives of Mayhew’s dredgers did not conform with the emerging requirements of the industrial economy, bound by a legal and increasingly bureaucratic framework, and quite different from the ‘rules and customs’ distinguishing dredgermen as a group. These rules were of their own making, the result of historical evolution and the physical environment of the river, where particular skills and knowledge, together with some minimum capital requirements, acted to establish their conventions and limit their numbers. Unlike many others among the poor, the dredgers were not new arrivals but rather descendants of generations of watermen. They had adapted their traditional way of life to the exigencies of a new age and were as yet imperfectly incorporated into its ‘forms’.

Dredgers originated among fishermen and some still fished seasonally. As shipping and commerce on the river increased they attached themselves to the margins of this trade, using dragnets to dredge for coal and other saleable items lost or thrown overboard (‘barges are often sunk in the winter and on such occasions they make a good harvest’). Mayhew estimated that, in response to advertised rewards, as many as a hundred boat-owners between Putney and Gravesend vied to search for lost valuables. Upon intelligence of accidents or suicides, their boats would congregate in search of corpses. For recovering the dead dredgers were paid ‘inquest money’, sometimes also a reward, as well as seeking ‘other chances’, for ‘no body ever recovered by a dredgerman ever happens to have any money about it, when brought to shore’ (Quennel 1984, 323).

Of its nature opportunistic and irregular, the dredging trade brought highly variable returns, though its practitioners enjoyed greater autonomy and higher incomes than many members of the London labouring poor. On the river banks and mudflats, ‘mudlarks’ pursued more meagre livelihoods. Like the dredgers they also escaped official regulation, perhaps because society could not provide for them or because they feared society’s remedy. Many mudlarks were children or the very old, the former supplementing family earnings and the latter desperate to avoid the workhouse, together searching for scraps of coal and other salvageable items.

There were numerous other occupations described by Mayhew in which economic change and the increasing reach of the State impinged more directly upon livelihoods. These influences took many forms; some groups found their status to be defined by reference to parallel occupations, conducted within the fold of formality and yielding higher returns. For an example, consider‘the cabinet-makers … [who] consist, like all other operatives, of two distinct classes, that is to say, of society and non-society men … those whose wages are regulated by custom and those whose earnings are determined by competition’ (Quennell 1984, 546). ‘Society-men’ were employed in formal establishments under the Master and Servant Act, by masters who maintained standards (and earnings) by restricting numbers. Mayhew recorded that ‘as a general rule … I find the society-men of every trade comprise about one-tenth of the whole’. This ‘artisan aristocracy’ (Jones 1971, 338) resisted the competitive pressures for ‘sweated’ production.

Those outside the fold of guilds such as the cabinet society were ‘slop-workers’, ‘the cheap men or non-society hands – who constitute the great mass of paupers in this country’ (Quennell 1984, 548). They were employed at subsistence wages by ‘garret-masters’, described as ‘a class of small “trade-working masters”, supplying both capital and labour’. Cashflow problems, typically requiring the immediate sale of products, left them at the mercy of the ‘slaughter-house men’. This latter group was a new breed of distributors and diversified shop-keepers, emerging during the transition from an age of craftsmanship to an era of mass production for a growing middle and lower-middle class market, whose members were perhaps more price-sensitive and less discriminating. ‘Small-master’ tradesmen were obliged to ‘scamp’ their work, for ‘the slaughterers cared only to have them viewy and cheap’ (Mayhew III, Ch 7). They also felt compelled to employ growing numbers of young apprentices, displacing higher-paid tradesmen and sending many into the street occupations for survival.

Other occupations were characterized by vulnerability to the arbitrary actions of authorities, no doubt struggling to cope with the demands of change and burgeoning urban populations. For example, street people might suffer from opaque enforcement of regulations. The ambiguous legal status of street sellers (the first of Mayhew’s six ‘distinct genera’), was among the causes of the famous antagonism held for the police by Costermongers, a major and readily identifiable subgroup of the street-folk9. In London, as Mayhew observed, legal tensions surrounding the operations of street traders dated back to Elizabethan times. Traders were, in principle, free to hawk their wares through the streets, but the operations of street markets were cloaked in ambiguity; in this uncertainty lay the source of traders’ informal status. Municipal law held that no market for provisions could be held within seven miles of The City, but certain markets were nonetheless tolerated ‘by custom and usage’. Mayhew commented that ‘the right to sell provisions from stands in the streets of the metropolis … is merely permissive’. In practice, police tolerated trading from fixed stands in the absence of complaints from shopkeepers. Where ‘injury and nuisance’ to shopkeepers could be demonstrated, police closed markets and moved traders on, causing them considerable loss. Mayhew asked, rhetorically, ‘whether any particular body of householders should, for their own interest, convenience, or pleasure, have it in their power to deprive so many poor people of their only means of livelihood’. However, as he pointed out, such actions sometimes proved a double-edged sword. In the case of Leather Lane market, when shopkeepers secured its removal, the decline in their own trade persuaded them to petition its return, following which their takings were restored. As seen in many cities even today, synergy may exist between formal and informal, such that people drawn to a street by its informal trade will also patronize its shops.

To return to the notion of Mayhew as ‘proto-ethnographer’, this title may be justified by his detailed accounts of the character and habits of three notable groups in the London street economy. These were Costermongers, Jews and the Irish (with some members of the latter two groups accepted as Costermongers). Among the Costers, ‘one-half of the entire class are costermongers proper, that is to say, the calling with them is hereditary, and perhaps has been so for many generations; while the other half is composed of three-eighths Irish, and one-eighth mechanics, tradesmen, and Jews’. Others of the Irish, newly arrived from their famine-stricken homeland, were among the poorest (though better able to fend than orphans or the aged and infirm among the native English). Their desperation set an upper limit on returns in some street trades, causing them, for example, to displace Jewish traders from the low end of the used clothing and footwear markets. Many Jews in informal occupations were relatively young or newly arrived in England. These did not usually identify as Costers, nor were they accepted as such. Members of a community with a high degree of trust, the poorer and less-educated among them might make their start in relatively open and accessible branches of the street trades before taking up more promising opportunities. In Sketches by Boz (1836), Charles Dickens described such a scene:

‘The coach-office is all alive, and the coaches which are just going out, are surrounded by the usual crowd of Jews and nondescripts, who seem to consider, Heaven knows why, that it is quite impossible any man can mount a coach without requiring at least sixpenny-worth of oranges, a penknife, a pocket-book, a last year’s annual, a pencil-case, a piece of sponge, and a small series of caricatures’.

But by mid-century, after the street-Jews had been undersold by Irish traders in the vending of oranges, Mayhew’s inquiries yielded the intelligence that

‘even when the orange and hawking trade was at the best, the Jews rarely carried it on after they were twenty-two or twenty-three, but that they then resorted to some more wholesale calling, such as the purchase of nuts or foreign grapes, at public sales. At present, I am informed, they are more thickly than ever engaged in these trades, as well as in two new avocations, that have been established within these few years – the sale of the Bahama pineapples and of the Spanish and Portuguese onions’ (Mayhew I, Ch.5).

Costermongers, particularly those of the ‘hereditary’ class, were genuinely a ‘tribe’, with their own dress, argot and customs. In passages reminiscent of Hart’s description of the erratic Weberian haushalten of the improvident denizens of Nima, Mayhew provides an account of the Costermonger equivalent. Feckless by middle-class standards, intemperate, promiscuous, pugnacious and antagonistic to authority, they were not ‘capitalists’. Indeed, ‘not more than one-fourth of the entire body trade upon their own property. Some borrow their stock money, others borrow the stock itself, others again borrow the donkey-carts, barrows or baskets, in which their stock is carried round, whilst others borrow even the weights and measures by which it is meted out’ (Quennell 1984, 76).

But while Costers were frequently in debt, some were capable of improvement and became community leaders. There are strong parallels between the descriptions of upward mobility in Mayhew and Hart. Mayhew reported that ‘many of the better-class costers have risen into coal-shed men and greengrocers, and become settled in life’, a process sometimes facilitated by marriage outside the group. As an informant put it, ‘some marry the better kind of servants – such servant-maids as wouldn’t marry a rag and bottle shop, but doesn’t object to a coal shed’. For the most part, however, Mayhew observed that Costers ‘have generally no foresight’ and that ‘only an exceptional few are provident’. For such a Coster, the combination of thrift and industry

‘enables him in some few cases to become “a settled man,” … He perhaps gets to be the proprietor of a coal-shed, with a greengrocery and potato business attached to it … He may too … have a sum of money in the savings’-bank, or he may invest it in the purchase of a lease of the premises he occupies, or expend it in furnishing the rooms of his house to let them out to single-men lodgers; or he may become an usurer, and lend out his money to his less provident brethren at 1040 per cent per annum; or he may purchase largely at the markets, and engage youths to sell his surplus stock at half profits’ (Mayhew I, ‘Of the providence and improvidence of Costermongers’).

Compare this with Hart’s description (1973, 71) of upward mobility among the inhabitants of Nima. While debt and chronic insecurity were the lot of most,

‘Entrepreneurial activities in the services sector – transport operators, landlords, cornmill owners, commodity speculators, and so on – represent the apex of informal economic opportunities to the sub-proletariat. Their essential characteristic is that they are frequently part-time roles, entered by individuals who have accumulated savings by some other means and, in the absence of an advanced capital market, re-invest income under their own management in taxis and lorries, accommodation, bulk purchases of maize, and the like. Successful performance … [requires] specialised ‘know-how’ and the ability, through diversification of investments and delegation of tasks, to accommodate the considerable risks attendant on these one-man enterprises. Though income from such activities may be very high, they are often combined with a formal job’.

Hart does not mention money-lending here, though we know it to have been pervasive in Nima. It is difficult to imagine his entrepreneurs succeeding without deploying capital shrewdly through the provision of credit, as was done by Mayhew’s successful Costers. The patterns are similar, as are the activities pursued. In both, the successful draw on social networks for custom and support. One difference is the reliance on formal employment to underpin the successful Frafra entrepreneur. But perhaps, in Accra, growth of government employment in the decade after Ghana’s independence offered opportunities not available to the entrepreneurial poor in mid-Victorian London.

Mayhew’s investigations led him to conclude that the number of street people in London was some 40,000, including about 30,000 working Costermongers, of whom some 12,000 were men, while women and children numbered 6,000 and 12,000, respectively. Thus many children were economically active; Mayhew noted that ‘every coster trading through the streets with a barrow is accompanied by a boy, who helps in selling and moving the stock. In this way they learn the trade and usually become independent by the age of 16’. In addition to the Costermongers there were ‘2000 Street-Sellers of “Green-Stuff,” as Watercresses, Chickweed, and Groundsell, Turf, &c.; 4000 Street-Sellers of Eatables and Drinkables; 1000 selling Stationery, Books, Papers, and Engravings in the streets; and 4000 other street-sellers vending manufactured articles, either of metal, crockery, textile, chemical, or miscellaneous substances’ (Mayhew II, ‘Introduction’).

Mayhew believed his estimates were more accurate than the official data, claiming that street people had been grossly under-enumerated in the census of 1841. This recorded ‘only 2,045 “hawkers, hucksters, and pedlars,” in the metropolis, and no costermongers or street-sellers, or street-performers at all’. This inaccuracy was due to ‘not one in twenty of the costermongers, or of the people with whom they lodged, troubl[ing] themselves to fill up the census returns – the majority of them being unable to read and write, and others distrustful of the purpose for which the returns were wanted’ (Mayhew I, Of the number of Costermongers and other street folk). As in Nima in the 1960s, the non-enumeration of a significant class of the urban population was a marker of the informality of their economic activities. Mayhew drew attention to the ambiguity inherent in their status: They were ‘a body numbering thousands … allowed to live in the prosecution of the most ancient of all trades, sale or barter in the open air, by sufferance alone … [and] classed as unauthorized or illegal and intrusive traders, though they “turn over” millions in a year’ (Mayhew II, ‘Introduction’).

Here Mayhew was making a case for the valuable contribution of ‘hawkers, hucksters and pedlars’ to the needs of the formal economy. But, as in the case of Hart’s account of livelihoods in Nima, with its division between ‘legitimate and ‘illegitimate’ activities, he accepted also a responsibility to describe the underbelly of the economy of the London poor. There is a difference of approach, however. Hart was not censorious in his account of activities offensive to the morality of ‘respectable’ Ghanaians. He was making a serious point about the diverse source of livelihoods in Nima, the understanding of which should not be clouded by normative considerations. By contrast, Mayhew was, enlightened observer or not, nonetheless a Victorian and an ‘improver’. Hart was playing a longer game; for him the sources of inequality and injustice were to be found in ineluctable historical forces. But whatever about Mayhew’s moral stance, he was able to convey a sense of the importance of ‘illegitimate’ practices and sources of livelihood to at least some of the poor. These sources were by definition informal.

Sharp practice and adulteration abounded in the street trades, from the collection of cigar-ends by the destitute for sale and re-incorporation into the stock of ‘best Havannahs’, to the procurement of used tea leaves for resale, after drying and even (in the case of green tea) dyeing, to the use of bogus weights and measures to defraud the unwary. A persistent theme in Mayhew’s many accounts of street business is that traders, held to the meanest of margins in a brutal commerce, were apt to pass on their disadvantage. The ‘improvident’ coster, along with hiring his barrow, might also rent a ‘slang’ quart pot, containing one and a half pints rather than two. He would pay a higher daily rate than for the genuine article, as a premium ‘for the risk’ of its being detected and traced to the owner (Quennell 1984, 77). Such practices did not begin in mid-Victorian London, of course. But perhaps the ‘formality’ involved in the policing of weights and measures became more difficult to maintain in a rapidly-changing and growing metropolis.

While not as comprehensive or systematic as Hart’s catalogue of ne’er-do-wells, Mayhew’s gallery of miscreants and ‘lurkers’ is full of interest. Mayhew subjects them to the same scrutiny, and occasional humour, as marks his records of the law-abiding. For example, he gives an account of the stealers, ‘finders’ and restorers of dogs. Perhaps because of the eminence of some of the owners thus put upon by the ‘dog appropriators’ (Mayhew records ‘royalty, foreign ambassadors, peers, courtiers and ladies of rank’ among the aggrieved) the trade occasioned a Commons committee in 1844, and subsequent legislation (Quennell 1984, 229-237). Along with the earnings of the thieves, pickpockets, burglars and receivers described by Mayhew, much of this income amounts to what Hart (1973, 86) has described as the ‘transfer’ or ‘redistribution’ (within Accra) of ‘static wealth lying idle in the homes of the bourgeoisie’.

As well as being the unwilling source of such transfers, elements of ‘respectable’ society were avid, if discreet, clients of the substantial demi-monde of Victorian London, whose members were largely drawn from the ranks of the poor. Though Mayhew discusses prostitution and crime only incidentally in his initial volumes, these still contain abundant evidence of the importance of activities classified by Hart as ‘illegitimate’, as well as numerous categories of people living on their wits (for example, Mayhew’s ‘patterers’, ‘screevers’ and ‘cheap johns’, analogous to Hart’s ‘conmen’ and ‘spivs’). This paper has already referred to Mayhew’s Coster ‘usurers’, and the resort to ruinously expensive credit was a pervasive theme in his narrative.

The political fixers and ‘influence peddlars’ who feature in Hart’s pages are not prominent in Mayhew’s account, however. Whether this was a step too far for Mayhew (for his method required citing the testimony of informants and might have risked legal action or other censure) or whether London was less subject than Nima to such abuses, is open to conjecture. Peter Bauer, whose work will be discussed at length below, remarked on ‘the pervasive significance of the politicization of economic life in LDCs [less developed countries]’, an aspect of ‘the general politicization of life’ in such countries’. After Independence, many new governments moved to increase the role of the state in economic affairs, with measures such as ‘state monopoly of major branches of industry and trade’ and ‘the establishment of many state-owned and state-operated enterprises’, shoring up political support through the creation of public sector jobs (Bauer 1984, 142).

In sum, though, while making allowances for differences in context, there are indisputably strong parallels between these two ‘case studies’. ‘Illegitimate’ or illegal activities were of great importance in the informal economies of both Nima and mid-century London. As for ‘illegitimate’ informal activity, so for the legitimate; the parallels between 1960s Nima and 1850s London are persuasive and their respective chroniclers were concerned with essentially the same phenomena. Hart took the discussion further by providing a framework of analysis which continues to influence research and policy formation. For example, international institutions, which in the 1970s saw merit in encouraging the informal economy as a palliative for unemployment, now devote their energies to ‘formalizing the informal’. ‘Formalization policies’ include measures to facilitate small firms in meeting bureaucratic requirements for registration and incorporation10. Mayhew also offered policy prescriptions intended to improve the lot of people in the London informal economy and to recognize their contribution to the broader economy and society; some of these appear startlingly relevant to contemporary concerns.

Finally, Mayhew saw the importance of seasonality in the London labour market, and the connection between seasonal and casual labour, noting that both were exacerbated by the small-scale system of production and the sweating of labour. Stedman Jones concluded that ‘[i]n embryo at least, Mayhew provided a theory of the specificity of the London economy which in turn made intelligible the economic behavior of the London poor’ (Jones 1971, 263). Mayhew’s prescience is evident from his identification of themes which continue to inform our understanding of the ‘development’ process. These include the impact of technical innovation on traditional occupations, of the modernization of retailing on patterns of trading and opportunity for the poor, of the importance of informal sector production in meeting the consumption needs of the poor (as well as members of more ‘respectable’ society) and of the impact of changing legislation and regulation on patterns of informal activity.


  1. P T Bauer and the Transition from Subsistence to Exchange

We now turn to the emergence of the idea of the informal economy in the specialist economic literature, and consider a relatively isolated figure among early post-war development economists. The term ‘isolated’ is appropriate, since the minutiae of ‘grassroots’ economic activity among colonial peoples were then an unusual focus of research for orthodox economists11. A widely-prevailing assumption, that neoclassical economic theory could be applied to analysing the broad range of human societies, appeared to render unnecessary any close study of the specifics of particular cultures or indigenous economic systems. However Bauer’s interest in the transition from subsistence to wider exchange led him to document processes unnoticed or ignored by others. Walter Elkan (1982, 247) would later claim that Bauer’s early work ‘foreshadowed the discovery of the “informal sector”’ while Bauer’s collaborator Basil Yamey (1987, 22) also believed that ‘[Bauer] was the first economist to recognize the extent and economic significance of what has come to be known as the informal sector’.In what follows his work will be examined with a view to assessing the significance of his contribution.

Bauer was isolated also in being a notable contrarian, placing himself in the classical or Smithian tradition during the heady post-war days of interventionist development planning. Deepak Lal (2002, 75) described him as having been ‘virtually a pariah in his chosen sub-discipline – development economics’, because of his opposition to the ‘fashionable planning and dirigisme of the 1950s and 1960s’, which employed a kind of ‘priceless economics’. Lord Bauer (he was later ennobled by Mrs Thatcher) has now, a decade or so after his death, achieved the status of libertarian ikon in such circles as the Cato Institute, the Institute for Economic Affairs and the Centre for Independent Studies12.

Some of Bauer’s work of relevance to the informal economy was published jointly with Basil Yamey. In retrospect, Bauer acknowledged the importance of this partnership, although, by his customary use of the first person singular he may have encouraged some to think otherwise. This perception was reinforced by Yamey’s habitual self-effacement. The original fieldwork in late-colonial territories and the observations stemming from this (essential to my account of the emergence of the ‘informal economy’) appear to have been Bauer’s sole responsibility. Yet the principal documents relied on here also bear Yamey’s name (Bauer and Yamey 1951, 1957). Yamey’s contribution may be sufficiently evident from the text and footnotes in this study and from their separate accounts of the relationship (Bauer 1984), (Yamey 1987, 2002, 2005). Peter Bauer was eventually recognized by inclusion in the first World Bank volume on the history of development economics, Pioneers in Development (World Bank 1984), to which he made a characteristically pungent contribution. Indeed one might say of Bauer that he parlayed a relatively small corpus of empirical work into a lifelong career, and that with each retelling his description became more polished, his conclusions more emphatic and his barbs more sharp. It is also necessary to say that his anecdotes grew more familiar.

Bauer is included in this study for two primary reasons. The first is the importance of his pioneering studies of internal trade and entrepreneurial activity among the people of colonial West Africa and Malaya in the early postwar years (Bauer 1948, 1954). The second is his contestation of propositions on the occupational distribution of labour during the course of economic growth (Bauer and Yamey 1951, 1957), derived from William Petty (1690) by way of A G B Fisher (1933) and Colin Clark (1940). Insofar as this concerned the role of the service sector in development it is highly relevant to the informal economy. Bauer’s setting was late-colonial, with particular reference to British West Africa (especially contemporary Ghana and Nigeria) and, to a lesser degree, British Malaya. Colonial governments played a major role, in administration and as providers of formal employment, while European trading houses, often with plantation interests, dominated large-scale economic activity. As in other such situations, a subsidiary class of ethnic-minority traders (for example, Chinese, Indians, ‘Levantines’) acted as intermediaries with the broader population. Bauer pointed to the spontaneous and substantial expansion of smallholder plantings of export crops, especially rubber and cocoa, in West Africa and Malaya, occurring due to the enterprise and energy of native peoples. Such agricultural investment was accompanied by intensive trading and other service activities and this indigenous economic complex was his primary focus. Bauer’s study of the indigenous economy and its interaction with the national and international economies has provided an account of an intricate and fascinating set of phenomena. In hindsight, we may now choose to characterize this as an ‘informal economy’, though whether he had in mind so clear-cut a distinction is doubtful.

Bauer’s concern was to document the entrepreneurial capacities and energy of some colonized peoples and to expose the ‘hollowness’ of popular ‘stereotypes’ concerning their incapacity for planning, capital accumulation and entrepreneurship. This project was subject to caveats however; for example, economists needed to take account of such matters as ‘the effects of different systems of land tenure, or of family obligations, on agricultural productivity and the supply of labour in non-agricultural occupations’. These were issues beyond the purview of neoclassical theory, which operated at a level of abstraction such that ‘the principal long-term determinants of income and growth, such as the factors underlying the growth of capital, the size of population, the attitude towards work, saving and risk-bearing, the quality of entrepreneurship and the extent of markets, were considered as institutional forces or as facts given, as data, to the economist’ (Bauer and Yamey 1957, 10).

The indigenous economy of British West Africa was indisputably ‘other’ than that of the government, the trading houses and the ethnic intermediary class which comprised the ‘official’ economy. But it was not the degree of ‘otherness’ implied by dual-sector models, conceived in terms of widely-divergent ‘capitalist’ and ‘subsistence’ sectors. In his review of W Arthur Lewis’ (1955), Theory of Economic Growth, Bauer (1956, 633) criticized the Lewis model for its ‘too sharp’ distinction between ‘stagnant’ agricultural and dynamic capitalist sectors. This characterization, he believed, neglected the capacity for capital accumulation in the subsistence sector (evidenced, for example, by smallholder investment in commercial cropping). Similarly, it is difficult to imagine Bauer’s being sympathetic to any ‘too sharp’ distinction between formal and informal economic activity (just as Hart had emphasized the propensity for individuals to operate in both, playing a variety of roles). Bauer might perhaps have accepted the formal/informal labels as names for the end-points on a continuum, but not as ‘sectors’ on either side of a dichotomy. His colonial economy was an organic entity, amenable to analysis using a set of ‘versatile general-purpose tools’. These were ‘the basic elements of supply and demand analysis … the theory of inflation, the concepts of substitution at the margin, and of the complementary or competitive relationship between factors of production’, which would prove ‘equally enlightening whether used to elucidate events and circumstances in advanced or in under-developed countries’ (Bauer and Yamey 1957, 8).

Intimations of the informal economy are unmistakable, nonetheless, in a controversy concerning the sectoral distribution of labour, initiated by Bauer and Yamey in 1951. Commencing with a paper in the Economic Journal and elaborated in their book The Economics of Under-developed Countries (Bauer and Yamey 1951, 1957)13, they challenged a proposition associated with Colin Clark’s Conditions of Economic Progress (1940) and AGB Fisher’s Economic Progress and Social Security (1945). What came later to be known as the ‘Clark-Fisher hypothesis’ had earlier origins, in William Petty’s Political Arithmetick (Petty, 1690). Colin Clark (1984), writing in Pioneers in Development, acknowledged

‘That agriculture should show a decline in its relative importance in employment and in national product, with manufacturing showing first a rise and then a decline in favour of services, was a generalization (emphasis added) first made as long ago as the seventeenth century by Sir William Petty’.

Bauer and Yamey (1951, 741) disputed the proposition that (as they phrased it) ‘economic progress is generally associated with certain distinct, necessary and predictable changes in occupational distribution, in particular with a relative increase in the numbers engaged in tertiary activities’. They contended that a combination of data deficiencies and inadequate observation underlay this hypothesis, charging that it could not bear empirical scrutiny and indeed reflected a degree of economic determinism. Later they described the Clark-Fisher hypothesis as challenged by two characteristics of developing country labour markets. These were the ‘imperfect specialization’ of individual workers – a concept acknowledged as originating from Adam Smith – and their ‘occupational fluidity’ (Bauer and Yamey 1957). These elements contributed to a greater allocation of labour to tertiary or ‘service’ occupations during the early stages of development than was anticipated by the Clark-Fisher hypothesis. This proposition appears to have had considerable influence on the research of Richard Salisbury, who appears below as another ‘forerunner’ of Hart’s informal economy.

Imperfect specialization of labour was masked by the unreliability of conventional census data concerning occupations. In practice, however, the phenomenon was apparent from simple observation of the livelihoods of people in agriculture. ‘Many of the so-called farmers (emphasis added) spend a large part of their time in small-scale transport, porterage and trade both during the farming season and much more so outside the season. They may trade not only in the goods they themselves have produced, but also in goods purchased by them for resale’. Occupational specialization should be termed ‘imperfect’ because such people operated in both the agricultural and service sectors. Further, imperfect occupational specialization was not confined to agriculture, anymore than it was confinedtorural areas or low-status individuals. Doctors, lawyers and traditional chiefs often had trading interests alongside their primary occupations, as did government employees and the domestic servants of Europeans. Like Keith Hart, Bauer observed ‘moonlighting’ in early post-war West Africa, with wage-earning individuals also active in the informal economy.

‘Fluidity of occupations’ was cited as a second stumbling block for the Clark-Fisher hypothesis. As with ‘imperfect specialization’, this characteristic is not easily captured in official data. It was the result of easy entry to many service tasks requiring only low levels of skill and capital for their performance. Mirroring the situation of some street trades in Mayhew’s London, ‘people can generally move with little sacrifice or difficulty within a wide range of occupations in accordance with changes in prospective net advantages. These activities include various forms of small-scale trading, the supply of the less-mechanized forms of transport service, and the provision of personal services generally. Many farmers are at no disadvantage in small-scale trading provided they have the small amount of capital which is required in this relatively unskilled activity’.

Because of imperfect specialization and fluidity in occupational choice, ‘the economic activity of many people in under-developed countries is better described as the performance of a number of different tasks than as the pursuit of a definite occupation’. Yet this rich and diverse set of ‘subsidiary’ activities was typically not captured by census data even though they constituted a large share of aggregate labour time, just as ‘the trading activities of children and of many wives are not likely to find their way into the statistical tabulations’. Recording such people as employed in agriculture led to substantial under-enumeration of the service sector.

Bauer and Yamey (1957, 42) took flourishing tertiary sector employment in West Africa and some other contemporary developing economies as an instance of ‘a well known theme in economic history’, that is, ‘the crucial role of trade and transport in quickening the process of economic growth and in extending the money economy’. As for an explanation of the importance of service occupations in the aggregate allocation of labour time during that ‘quickening’, they observed that what resources are used in trade, transport and other service industries, at a given level of techniques, depends on relative costs and scarcities. ‘In an economy such as West Africa, where capital is scarce and expensive and unskilled labour abundant and cheap, the large volume of resources in distribution and transport consists very largely of labour. This tendency … permeates West African trading arrangements’ (1951, 745).

Thus, ‘once the level of economic activity has risen from that of a subsistence economy to that of an emerging exchange economy … the task of distribution may require a substantial volume of resources’. In such an economic environment ‘the indispensable tasks of assembly, bulking, transport, breaking of bulk, and dispersal’ may require a significant proportion of available labour power. Conditions in West Africa (‘dispersed farmers and holdings, poor natural communications and long distances, and the difficulties of prolonging storage in the open’) suggested also the need for substantial resources to be applied to transport (1951, 743). They invoked Adam Smith to explain ‘imperfect specialization’ in terms of the narrowness of markets. Markets were limited due to the poverty of consumers, but also to the relative poverty of traders, much of whose behavior was explained by scarcity of capital. Traders faced high transport costs and lacked capital to finance the accumulation and storage of stocks. Cheap labour was substituted for expensive capital in transporting commodities as well as by economizing on the costs of storage (‘which in effect is transport through time’). More recently, such calculation has underpinned the adoption of ‘just in time’ processes in modern manufacturing.

But if traders lacked capital, this constraint bound many consumers as well. The possession of even a tiny amount of capital offers a trading opportunity for some, in offering services to the poor which higher-income consumers would provide for themselves. Petty traders, in effect, hold inventory for the poor by dealing from a small stock in minute quantities, just as coal-dealers in poor neighbourhoods of Mayhew’s London broke down fuel into small parcels for use over a day or two. Examples, which remain valid in some twenty-first century contexts, include ‘the West African petty retailer selling perfume by the drop or cigarettes by the piece, the woman spending a full day selling a dozen mangoes in a Caribbean market, and the woman selling paraffin in small quantities from door to door in a Thai village’. Bauer and Yamey explained these as ‘examples of an intermediary performing the functions of proximate stockholder, which in a wealthier country would be performed by each consumer for himself’. Proximate stockholding might also extend to modest items of capital equipment which were, nevertheless, beyond the resources of individuals to buy for themselves. Reminiscent of Mayhew’s ‘improvident’ Costers, hiring the tools of their trade day by day, bicycles might be offered for hire to transport people or goods, or even the services of bicycle pumps, enabling petty entrepreneurs to conserve working capital.

Bauer and Yamey (1951, 745) gave other examples demonstrating the importance of services provided by ‘imperfectly specialized’ labour. These included trading in empty containers (manufactured items such as petrol drums, tins, cans, bags and bottles, as distinct from traditional containers) which had been reconditioned for re-use. These might become household utensils or other commodities, though most often they were re-used in the storage and movement of goods. In a reminder of Mayhew’s Rag and Bottle shops, ‘[t]hose who seek out, purchase, carry and distribute second-hand containers maintain the stock of capital. They prevent the destruction of the containers, usually improve their condition, distribute them to where they can best be used, and so extend their usefulness, the intensity of their use, and their effective life’. Such recycling activities were ‘highly economic in substituting superabundant for scarce resources; within the limits of available technical skill nothing is wasted in West Africa’. Neither was there much waste obvious in Mayhew’s London, where a ragged army of ‘finders’ and ‘buyers’ sought out a wide range of materials for reprocessing and re-use.

The same logic applied in agriculture: for peasant agriculturalists the scarcity of capital ‘is reflected in the absence of suitable storage facilities and of cash reserves’. This had consequences for trade and distribution. Thus, ‘each producer has to dispose of small quantities of produce at frequent intervals as they become available during and immediately after the harvesting season. This postulates a large number of intermediaries who … employ methods of transportation using relatively little capital and much labour. Donkey and bicycle transport are examples, while in some cases there is still head loading and human porterage, especially in the short-distance movement of local crops. The available transport equipment is used continuously with the assistance of large quantities of labour (subject to frequent breakdowns owing to poor roads and low technical skill)’.

Notwithstanding the importance of ‘imperfect specialization of labour’, some specialization did occur even at relatively early stages in the emergence of exchange, and in forms perhaps surprising to observers from high-income countries. Thus, ‘in many under-developed countries narrowness of markets, which discourages occupational specialization, is found together with a more elaborate or minute division of labour in certain economic activities than is usual in richer countries. Some services which in richer countries are usually performed by consumers for themselves are … provided commercially, often in large volume in the aggregate’. For example, because personal services, such as those of tailors, barbers and beauticians were relatively cheaper in such countries, ‘the better-off people of the poor country are likely to buy more of these services than their counterparts in richer countries’. For this same reason domestic servants were seen as ‘conventional necessaries for all but the poorest’ in poor countries. Again, where particular skills or training were scarce, the holders of such skills could convert them into an income-earning opportunity (as with illiteracy, where letter-writers and readers were in demand, often offering their services in marketplaces).

As development proceeds, however, ‘occupational specialization generally becomes more marked, and at the same time the fluidity of labour between occupations is reduced. Markets are extended with improvements in transport and increased purchasing power, so that specialization becomes more profitable. Moreover the increase in capital requirements and the improvements in techniques of production limit the movement of people between economic activities’. Such changes represent incremental steps away from occupational fluidity.

In terms of the occupational distribution of labour, the particular interest of Bauer and Yamey’s observations lay in pointing to the importance of ‘tertiary’ employment at very early stages of economic development, rather than at later stages as suggested by the Clark-Fisher hypothesis. It must be pointed out that their ‘tertiary’ activities were of a different order from those which Clark saw as becoming predominant with the progress of development. In summary, however, Bauer and Yamey asserted that tertiary sector employment, far from being the consequence of economic development, was a necessary element in the transition from subsistence to exchange, because:

‘in West Africa, as in other emerging economies, the indispensible task of commodity distribution is expensive relatively to available resources … the multiplicity of traders is the result of the mass use of unskilled labour instead of capital in the performance of the task of distribution. There is an extensive demand for the services of intermediaries, and there is a large section of the population available to perform these services at a low supply price in terms of daily earnings’.

Thus, pace Clark and Fisher, Bauer and Yamey concluded (1951, 747) that ‘there are severe limitations and qualifications to the view that a high proportion of labour in tertiary production is both a consequence of and a pointer to a high standard of living’. The qualification they offered to this view was based on an examination of occupational distribution during relatively early stages of the transition to market exchange – an observation of importance to our understanding of the informal economy, but scarcely a refutation of the Clark/Fisher hypothesis as a long-run proposition. In hindsight, their qualification of Clark/Fisher has not proved influential and contemporary research continues to explore the causal mechanisms of ‘the shift to services employment’ rather than the direction of that shift14. For this reason neither the brief debate in the 1950s on the Bauer and Yamey thesis, nor the considerable literature since accumulated, is explored here.

For the purposes of this paper, the importance of Bauer and Yamey’s intervention does not lie in its attempt to challenge a well-established generalization concerning growth and structural change, but rather in its pioneering account of indigenous trading and service activities. While these are observed with a sharp eye for social and cultural context, Bauer does not deal with what Hart termed ‘illegitimate’ activity. This major feature of life in Nima is entirely absent from Bauer’s account. References to petty manufacturing and the varieties of ‘transfer’ documented by Hart are also few. Bauer’s account of the informal economy appears to some degree unbalanced by his preoccupation with the role of trade and transport in the transition from subsistence.

Bauer’s argument in his discussion of occupational specialization turned on the difficulties of enumeration; this is a central criterion in defining informal economic activity. In Bauer’s analysis, increasing specialization tended to lead to what we would now see as more formalized occupational roles. Bauer’s ‘imperfect specialization’ and ‘occupational fluidity’ of labour also emerged strongly in Mayhew’s account of the ‘casualized’ labour market of Victorian London. Bauer also emphasized the potentially-multiple activities of people in situations of ‘occupational fluidity’. This emphasis was consistent with Hart’s account of people performing multiple occupational roles. Bauer, like Hart, detected a high incidence of ‘moonlighting’ among those with formal waged employment, though Hart’s emphasis on the importance of the reliability of income streams was lacking in Bauer’s account.

Whereas Hart’s narrative dealt almost exclusively with an urban informal economy, Bauer’s story gives more attention to informal activity in the rural sector, as well as dealing with the role of trade in creating rural-urban economic linkages. Bauer’s account gives rather more attention than Hart’s to the economic activities of women (leaving aside the ‘red light’ elements in Hart’s narrative). On the other hand, Mayhew provided considerably more information on the economic activities of women than either of them. Mayhew’s work, like Hart’s, dealt largely with the urban economy, though he had numerous examples of rural-urban interactions, dealing (for example) with seasonal labour and itinerant trading.

Hart’s emphasis on transfers within social networks as an urban reflection of traditional relationships, while not entirely absent from Bauer’s narrative, suggested the former’s anthropological perspective. Bauer’s application of ‘versatile general purpose [economic] tools’ provided us, for those aspects of informal activity which interested him, with a fine-grained explanation of non-agricultural livelihoods. Bauer was more positive than Hart about the economic value of informal employment (though here again his selective attention to trade and services should be remembered). Hart’s more limited economic analysis stressed interactions between formal and informal ‘sectors’, couched in terms of ‘export’ and ‘import’ transactions. Bauer’s account dealt with the colonial economy as a more organic entity – a continuum of economic activity ranging from (as we might now style it) the ‘formal’ to the ‘informal’ – but all amenable to a common framework of analysis. Much later in his career Bauer (1984, 38) was happy enough to accept Walter Elkan’s encomium, that he had ‘foreshadowed’ the ‘discovery’ of the informal sector. But while Bauer was certainly a forerunner of the idea, Elkan was correct to imply that he was not indeed its ‘discoverer’.


  1. Richard Salisbury: Services asLeading Sector

In 1972, the (then) Territory of Papua and New Guinea was among the earliest states in which Keith Hart’s idea of an urban informal economy was introduced into policy discussion. Hart brought the idea to Papua New Guinea (PNG) as a member of a consulting team, the ‘Faber Mission’. This had been commissioned by UNDP, to advise on economic policy for a soon-to-be self-governing PNG (Overseas Development Group 1973). On arrival, Hart was taken aback to find that informal economic activity appeared conspicuous by its absence in PNG’s few small urban centres. His recommendation that its growth should be encouraged found a receptive audience among economic nationalists, and the idea of the informal ‘sector’ achieved a degree of influence, at least at the level of rhetoric, on early planning priorities and in the constitution of the new State. This was drafted with several clauses implying the duty of government to support and encourage informal economic activities (Papua New Guinea 1975).

Papua New Guinea will, coincidentally, figure in this narrative as the country in which another forerunner of the informal economy developed his ideas, before the advent of Hart and the Faber Mission. On the basis of anthropological fieldwork in two contrasting and widely-separated districts of PNG, Richard Salisbury came to the view that service, or tertiary, activities could play the role of ‘leading sector’ in the transition from subsistence agriculture to market exchange. This paper will examine his work to consider whether, and in what fashion, he anticipated the idea of the informal economy, particularly in his account of the economic development of the Tolai people of the Gazelle Peninsula of New Britain.

At the time of Salisbury’s work with them, some ten years before the visit of Keith Hart, the Tolai were possibly the most prosperous and progressive population group in PNG. For reasons to be explained, they had achieved a relatively high degree of intra-regional trade and exchange and a number of non-traditional sources of income. Conditions for a flourishing informal economy were probably more favourable on the Gazelle Peninsula than elsewhere at the time. It will be useful to consider the extent to which informal economic activity had indeed developed there in the late-colonial period and, if so, the degree to which Salisbury appreciated its significance. First, however, it is necessary to introduce the two studies on which Salisbury’s ideas were founded.

The Siane people of the Eastern Highlands of New Guinea and the Tolaiof the Gazelle Peninsula, on the island of New Britain, both experienced rapid social and economic change after the Pacific war. Salisbury lived among the Siane in the period 1952-53, quite soon after the introduction of steel tools had revolutionized aspects of their agriculture and social relations. The Siane were at an early stage of transition from a ‘pure’, non-monetized subsistence economy (Salisbury 1962), whereas the Tolai, with whom he worked between 1961 and 1963, were further along that transition path, with an agricultural system combining subsistence and export crops. Tolai experience of international trade and the monetized economy, colonialism and Christianity had begun much earlier, during the 1870s (Salisbury 1970).

The introduction of the steel axe to the Siane released labour-time, which ‘was used for politicking, ceremonials, legal disputes, and fighting’. These proved to be elements in a process of ‘political consolidation’ which occurred, for reasons specific to each case, among both the Siane and the Tolai. In each of these societies, larger social groupings coalesced around emergent leaders exercising more effective political control. This in turn led to ‘organizational innovation’ and ‘the development of new consumer demands’ (Salisbury 1970, 10-11). In his view, increasing levels of political activity, combining with capacity to make decisions at the local level, produced in each of these societies a socially-beneficial consolidation of political authority, from which further economic progress could flow.

In a little-known paper published in 1971, Salisbury drew together findings from his 1962 Siane study (From Stone to Steel: Economic Consequences of a Technological change in New Guinea) and his 1970 Tolai publication (Vunamami: Economic Transformation in a Traditional Society). The 1971 paper shows some congruence with the ideas of other ‘forerunners’ of the informal economy and with those of Hart himself. Titled Development through the Service Industries, it contended that ‘under certain conditions [the service industries] may be the dynamic sectors of an economy’. Salisbury explicitly rejected the view that ‘following Colin Clark’s schema of development’, services ‘are looked for … as properly emerging only after primary and secondary industry have achieved certain levels of production’. In fact, in both of his PNG cases, ‘service industries have proved the leading sector, in the Tolai case for almost one hundred years’ (Salisbury 1971, 57). These views echoed Peter Bauer who, as we have already seen, rejected the ‘Clark-Fisher hypothesis’, contending instead that services were crucial from the very beginning, ‘in quickening the process of economic growth and in extending the money economy’. In the very last sentence of From Stone to Steel, Salisbury (1962, 213) offered his Siane study as a building block towards understanding ‘the complex mechanisms which Bauer and Yamey have called the Economics of Under-developed Countries’. This may fairly be taken as acknowledging an intellectual debt.

Nor was this the only such debt. Salisbury commenced his anthropological career as an unabashed formalist, crediting the influence of the neoclassical economist and methodologist Lionel Robbins (1935). In the spirit of Robbins, Salisbury (1962, 3-4) set out to describe activities ‘in which [the Siane] appear to organize their behavior in terms of a rational calculation of the quantities of goods or services produced, exchanged or consumed, in such a way as to allocate scarce means to meet competing ends’. Keith Hart, himself no formalist, would later comment that Stone to Steel had demonstrated ‘that a formalist premise was not incompatible with rich and nuanced ethnographic arguments’ (Hann and Hart 2011, 68).

In his 1971 paper Salisbury attempted a theoretical model of the role of the service sector in development, to generalize the empirical findings of his Siane and Tolai studies. The model is described as emerging from ‘a social evolutionary perspective’, a branch of social theory given to identifying ‘stages’ in the course of human progress. Many writers, from Adam Smith to Karl Marx, and from Lewis Henry Morgan to Walt Rostow, have discerned such stages. From the work of Marshall Sahlins (1958), Salisbury drew the observation that ‘in generally undifferentiated subsistence societies the first full-time specialist occupations to occur are political and religious ones’. Further, in such societies, ‘the actual proportion of the labour force employed in service production would be much higher than the figures of full-time specialists would indicate because of part-time involvement in politics and religion’ (Salisbury 1971, 57). This sounds very much like Peter Bauer’s explanation of errors in colonial labour force data resulting from ‘imperfect specialization’.

Salisbury concluded that, if trading or gift-exchange were also regarded as tertiary activity, the proportion of subsistence ‘manpower’ employed in services among the Siane may have been as much as 20 percent. Ceremonial gift-exchange is a means by which aspirants to political leadership in Melanesia strive for status and prestige. Such exchanges were a thoroughly traditional element in the process of Salisbury’s ‘political consolidation’. For the Siane and the Tolai, increased levels and frequency of gift-exchange were a culturally-appropriate means of enjoying the additional leisure time made available by labour-saving innovation. While a 20 percent allocation of labour-time to this and other ‘services’ might seem high in a subsistence agricultural economy, Bauer and Yamey would not have been surprised by such a finding. It was substantially greater than the share of ‘manpower’ engaged in manufacturing, including craft-goods (Salisbury 1971, 57).

Salisbury compared his case-study data for services with statistics for the historical growth of service sectors in a number of industrial economies. He interpreted these to support an argument that ‘in a context of social welfare … the volume of services generally available within the society is a better measure of “real development” than is an index based on money or on quantities of goods’ (Salisbury 1971, 58). From this assertion he developed a somewhat convoluted and eclectic argument, derived from the ‘development of under-development’ school of political economy, and incorporating the ‘agricultural involution’ of Clifford Geertz. In Salisbury’s view, ‘under-development’ occurs when subsistence economies are diverted from what is, in some sense, the ‘normal’ (that is, organic or correct) path of development. An extended quotation here may be the best way to convey the flavour:

‘[U]nderdevelopment is itself a process that takes something away from non-developed countries like New Guinea or nineteenth century Africa (e.g., Frank 1967). Such countries absorb, or are absorbed by, complex technologies and political structures but undergo a reduction in the local provision of services. The process of under-development is one that makes simple but labour-efficient agriculture more labour-intensive (c.f., Geertz 1963b), absorbing a larger proportion of the labour force, but taking the provision of services away from the rural or poorer areas, and centralizing them in either capital cities or metropolitan countries. This centralization … retards the “normal” pattern of development and differentiation within rural or colonial areas’ (Salisbury 1971, 58).

So the Siane and Tolai would, in the absence of the centralizing forces of ‘under-development’, proceed along a ‘normal’ (that is, decentralized) path in which village ‘services’ (whose definition we shall shortly examine)form the leading sector. Political consolidation and the retention of decision-making authority at the local level are crucial elements in this scenario. But, if made subject to centralizing forces, Tolai and Siane would likely fall prey to what Clifford Geertz, writing of Java, had described as ‘agricultural involution’. So bald a paraphrase probably does rough justice to Salisbury’s argument, but further comment will be withheld until his application of it to Papua New Guinea has been discussed.

One of the more interesting aspects of Salisbury’s work was his detailed measurement of Siane and Tolai labour use. By preparing ‘time budgets’, he was able to measure how people who might otherwise be regarded simply as subsistence cultivators actually apportioned their time between various activities. Such an approach, had it been employed by Peter Bauer in West Africa, could have illuminated his discussion of the ‘imperfect specialization’ of labour and the dangers of using official data as a guide to its sectoral distribution. Salisbury was able to show (1971, 59) that among the Tolai most subsistence cultivators (and especially men) ‘usually work part-time, often voluntarily and without pay, at what may be called service activities, providing entertainment, education, organization and assistance to their neighbours where specialized agencies to provide them do not exist’. This is an important clue to Salisbury’s understanding of the ‘service’ sector. It was clear from Salisbury’s time budgets that official data tended to ‘overestimate the proportion of labour self-employed in agriculture’ and ‘underestimate the proportion of service work performed’ (1971, 59).

For the Siane, whose subsistence production was non-monetized, Salisbury collected relatively straight-forward time budget data (measured as ‘labour cost’ in terms of time). For the Tolai his task was more complicated, though also more illuminating, due to their use of traditional ‘shell-money’ currency. This circulated alongside Australian currency, but (1971, 59) ‘with different implied rates of exchange in different spheres of trade’. This carried the advantage that ‘subsistence production consumed locally can be compared either quantitatively or by cash value with store-bought foods. Imputed wage rates (in shell-money) can be calculated for traditional service activities’ (1971, 59). He was able not only to measure Tolai time inputs as between various activities, but also to establish approximate income differentials between subsistence, service and other forms of employment.

Salisbury’s vision of the role of services in a subsistence economy undergoing transformation, and the nature of the services he had in mind, is further clarified in his conclusion to the 1971 paper:

‘In countries where there is an approximate internal self-sufficiency in foodstuffs, encouragement of local (and particularly rural) self-help projects in education, marketing, artistic endeavour, housebuilding, athletics, or leisure generally, service industries may well turn services into a leading sector, may absorb unemployment, provide incentives to labour-saving innovations in productive activities, encourage entrepreneurship, facilitate investment and make changes in the average standard of living in such societies, without the need for large additional outside investments. Development through service industries can be a pattern for grass-roots development’ (1971, 65).

Salisbury envisaged many of these services being provided by local government- and church-related entities, as a dividend of the ‘political consolidation’ process, and flowing from the energies of political and community organizers. At the local level, ‘the process of economic development can be seen as involving both technological improvement and the desire by the people to provide increased services for themselves’ (1970, 351). This process occurs in an emerging public sphere, based on newly-created local institutions. Nonetheless, as we shall see, Vunamami is replete with instances of private, for-profit informal enterprise, whether based on groups or individuals. Moreover, his Tolai local government enterprises often had more the character of informal public-private partnerships than a corporate form recognizable to Western eyes. With the exception of the Tolai Cocoa Project, to be discussed below, such activities generally lacked the element of formality bestowed by bureaucratic intervention.

The role of services in the transition from subsistence is a major theme of Vunamami. The Tolai experienced four successive instances of labour-saving agricultural change over the period 1875-1960 (Salisbury 1971, 60). Each of these permitted either the production of a given quantum of output with less labour input, or the expansion of output for the same level of labour input. The four changes were presented as ‘sequences’, a term used by Salisbury in both his Siane and Tolai studies, and equivalent to ‘stages’, as commonly found in the social evolutionary literature. Indeed Salisbury (1962b, 338) explicitly compared his sequences with Rostow’s ‘stages of growth’. In the first sequence, commencing about 1875, Tolai took the opportunity to exchange surplus coconuts for trade goods with Europeans. In the second, commencing during the 1890s, they began to expand their plantings of coconut palms for this trade. In the third, from the late 1930s, Tolai adopted the hot-air copra dryer to add value to their crop. Fourth, and most recently before Salisbury’s arrival, from the late 1950s they adopted cocoa as a new export crop, together with associated mechanized processing. The effect of each of these changes was, broadly speaking, to release a larger proportion of labour inputs to be applied to service activities.

Rather than the discrete movement of individuals from one occupation to another, it is more accurate to think in terms of changes in the proportions of aggregate labour input,as individuals adjusted their own labour inputs between multiple activities. Labour-saving change, occurring in a situation where (apparently) there was little demand for increased material output, would have produced ‘technological unemployment’ if workers had not been able to diversify, spending more time productively in service activities. While there was some increase in the ownership of material goods, the most notable effect was non-material. ‘“Prosperity” was evidenced more by greater leisure and a varied quality to social life than the ownership of material goods, although this did increase also’ (Salisbury 1971, 60). In this context, ‘leisure’ appears to have included engagement in group activities, such as ceremony and exchange, in local government, and in church and community affairs (described above as the dividends of political consolidation). While these may not be regarded as ‘leisure activities’ in a Western individualistic sense, their potential for building social capital suggests their value in the social systems of the Siane and the Tolai.

As Salisbury explained in Vunamami, this shift to leisure ‘would occur only if existing demand for food and subsistence requisites were already met – in other words, in a relatively affluent society’ (1970, 11). Labour was not unlimited in supply; in fact it was scarce, relative to the natural resource endowment and productive potential of the Tolai lands. Salisbury’s data for land-use and labour inputs in Vunamami supported the contention that the Tolai were ‘affluent’ in subsistence terms. Thus, for 1961, he reported that ‘Vunamami, thanks to its women, still is … virtually self-sufficient in food crops, at a cost of only fifteen and a half acres of land and 11 percent of the labour of its farmers’, while ‘the remaining agricultural labour of Vunamami, and the remaining acreage, were about equally divided between copra and cocoa, between planting [ie, expansion of commercial crops] and cropping’ (Salisbury 1970, 146).

That activities, deemed unprofitable at the margin by Tolai, were abandoned when more rewarding opportunities arose was attributed by Salisbury to their ‘employment situation’. ‘With an elastic demand for labour (even self-employed labour) in service activities, which absorbed available labour, and no local shortage of food supplies, people were happy to forgo drudgery where they could’ (1971, 60). This situation encouraged investment in, and the movement of labour to, activities for which demand was elastic. In terms of proportions of aggregate labour input, measured by time-budgets, there was a movement of Siane labour from subsistence agriculture to services (Salisbury 1962, 152). Demand for ‘self-employed’ labour is derived from demand for the output of self-employed service workers. In referring to an ‘elastic’ demand for self-employed labour, Salisbury appeared to mean there was no shortage of relatively remunerative outlets for the energies of any workers ‘displaced’ from subsistence activities by technical change. He contrasted this fortunate situation with a more constrained case, that of land shortage where unemployment and food shortages exist. In such situations, ‘workers can always be found to move into relatively unproductive or low-paying jobs vacated by upwardly-mobile workers. Under these conditions technological innovation does not restructure the whole economy but merely adds a new sector to an old-style one’. In this case, new and more productive economic activity is simply superimposed on a stagnating traditional sector beset by Geertzian ‘agricultural involution’.

Salisbury concluded (1971, 60) that ‘full employment and self-sufficiency in food supplies’ are ‘pre-conditions’ for the applicability of his model. Under the ‘affluent’ circumstances of the Tolai, labour tended to flow from primary production to service industries, where returns were higher. Salisbury’s data for income relativities revealed the incentives underlying these shifts. Taking the wage on expatriate-owned plantations as a base, a Tolai farmer drying and selling his copra earned about two and a half times the cash wage of the labourer (suggesting why few Tolai worked on plantations). Even the least skilled ceremonial dancers or clerical workers earned about double the plantation cash wage, while members of Tolai fishing crews earned about as much as the plantation wage, though in more congenial circumstances. At the other end of the scale, ‘the richest men in both shellmoney or cash are top-flight artists, or political and financial organizers’ (Salisbury 1971, 61).

Salisbury (1971, 61-2) contrasted the situation of the Tolai, in an affluent mixed economy of subsistence and cash-cropping,undergoing transition to market exchange, with that occurring in some cash-cropping peasant societies. These were typified by monocrop export production systems and reliance on imported food staples. This was a situation of ‘over-specialization’, characterized by ‘steadily declining wage-rates, lack of local services, extreme vulnerability of the local economy to changes in world-market prices for its product, and an inability to innovate or to diversify the economy….’ Such circumstances led to agricultural involution, which Salisbury believed might be avoided if local service activity could be maintained, and with it the flexibility necessary to diversify economic activity.

Salisbury’s discussion of involution and its application to PNG is not entirely satisfactory. The agricultural involution described by Clifford Geertz derived from the particular conditions of ‘wet-rice’ cultivation in colonial Java and Bali, where peasant farmers grew irrigated rice on land they were also forced to lease, in a system of periodic rotations, to large commercial sugar plantations. This rice cultivation system was distinguished by ‘its marked tendency (and ability) to respond to a rising population through intensification … absorbing increased numbers of cultivators on a unit of cultivated land’. Geertz specifically ruled out the possibility of such involution in shifting agriculture, such as Tolai subsistence gardening: ‘such a course is largely precluded to swidden farmers … because of the precarious [ecological] equilibrium of the shifting regime’ (Geertz 1963b, 32).

Nonetheless Salisbury saw a kind of agricultural involution as an ever-present threat to the Tolai, should their path to development become blocked by ‘centralization’ and an inability to continue transferring labour into productive service activities. The Gazelle had been in danger of succumbing during a stagnant period from 1921 to 1935, when ‘increasingly direct administration from Australia and a lack of technological innovation brought Tolai society into an incipient phase of involution. Both technological and political change were needed to reverse the process’ (Salisbury 1971, 65n). Adopting cocoa as a new export crop and investing in fermentaries for post-harvest processing provided the necessary escape. Complementary political development also occurred, as Tolai became involved increasingly in a broad range of community activities, creating local-level institutions and building social capital. Absent such beneficial change, however, and with continuing population growth, the ‘precarious’ ecological equilibrium of Tolai subsistence agriculture would have been disturbed. Conditions on the Gazelle would guarantee a quite short road to over-population and falling incomes for the Tolai, unless significant innovation in food crop production became possible.

Salisbury saw a further dire possibility (1971, 64). He warned of the simultaneous dangers of ‘agricultural involution in rural areas and service development in the capital’. Urban centralization of services was inimical to progressive rural communities and was, in his analysis, causal in the emergence of what he understood as agricultural involution. But declining rural living standards would also cause service provision to become ‘low-productive and technologically involuted’. For the idea of an ‘involution of services’ Salisbury acknowledged discussions with a colleague, the economist Benjamin Higgins (himself a colleague of Geertz in Indonesia in the 1950s). Paraphrasing Higgins, Salisbury noted that ‘service industries of low productivity in many under-developed countries absorb large proportions of the manpower, but disguise under-employment without removing it. They then create what might be called “an involution of services” – taking in each other’s washing – leading to increasingly labour-intensive technologies, and none of the economically stimulating effects described in this paper’ (Salisbury 1971, 65n).

Courtesy of Ben Higgins, Salisbury had now identified two forms of involution. He had also distinguished two informal service sectors, one rural and productive, the other urban and marked by a degree of poverty and disguised unemployment. The former could, under the right conditions, flourish in a society unconstrained by shortage of land for subsistence. The latter would emerge from an increasingly land-short subsistence economy from which people were forced to migrate to urban areas. But this was a warning rather than a prediction of outcomes. Aside from the community-level services which provided the dynamic of his model, he gave numerous examples of private informal economic activity among the Tolai, none of which appears to have developed involutionary characteristics by 1961. Rather, these activities enabled many people to profit from rising income levels in their own communities and to benefit from proximity to the thriving urban centre of Rabaul. This could be done while retaining security of access to one’s own land and resources, although some peri-urban populations (in particular in the village of Matupit) were beginning to experience land shortage and were adopting new sources of livelihood in consequence (A L Epstein 1963).

Although the scale was much smaller, there were some parallels between the late-colonial economies of Salisbury’s Gazelle and British West Africa, as described by Bauer and Yamey (1957). In both the native peoples had been quick to take up opportunities for export cash-cropping (though large, foreign-owned plantations were relatively more important on the Gazelle). Large expatriate trading houses dominated the formal economies of both, while a subsidiary class of traders (in the case of the Gazelle, the Chinese) acted as intermediaries with the local population. A relatively large expatriate population in Rabaul (at the time, the second-largest urban centre in PNG) provided market opportunities for the Tolai. Early informal economic opportunities had arisen from the rich agricultural base, and in particular from product specialization due to the ecological diversity of the Gazelle region. Even in pre-contact times this had fostered a system of trade and marketplaces, multiplied by external influences from the 1870s on. Subsistence households on the Gazelle engaged in a substantial trade in traditional foodstuffs, initially based on seasonality and the availability of different varieties of staple foods from a range of sources. By 1961 Salisbury’s observations led him to the conclusion that this ‘earlier interdependence of different ecological areas’ was being replaced by a new set of relationships at the Rabaul and other markets. For example, people in peri-urban villages of Rabaul were increasingly reliant on wage-labour and the purchase of food crops, while rural villages had ‘individuals who specialized in producing bulk crops for the market, and who themselves purchased from other producers the vegetables they needed for subsistence’. Thus, if peri-urban villagers could be classed as ‘proletarians’, some rural villagers had become ‘commercial farmers’ (Salisbury 1970, 214).

In Salisbury’s account of Vunamami, women were dominant as sellers in the marketplaces of the Gazelle though sales by individuals were typically small and (underlining the ‘affluence’ of the Tolai subsistence economy) market attendance was valued as much for the opportunity to socialize as for profit earned. Transactions between Tolai were largely conducted in shell money, and since this could be exchanged for a wide range of traditional commodities and services, it seems legitimate to classify such trade as informal economic activity. Typically, much of vendor earnings was spent immediately, to buy other varieties of food for household consumption. Transactions made with Europeans, Chinese and other non-Tolai yielded cash, and since the production of vegetables for profit was ‘normally a male affair’ this cash would usually be surrendered to husbands.

Men were in general more profit-oriented than women, often initiating plantings of subsistence and introduced crops for market. Some entrepreneurs contracted for the bulk supply of foodstuffs to commercial plantations and institutions (schools, missions, hospitals), bulking up consignments from individuals and trucking produce on commission. They performed intermediary functions which appeared still to be lacking from the marketplace trading of women. Outside the markets, there was some street trading in Rabaul while at public occasions such as sporting events, vendors selling cigarettes, buns, soft drinks and sweets were active.

Other informal economic activities were traditional, and usually conducted for shell money. These included hiring traditional specialists (composers of songs and dances, canoe builders, makers of musical instruments, also ritual specialists who performed rituals of benefit to communities and were paid by them. Mortuary rites had become larger in scale, as had ceremonial associated with male cults. Specialists in magic and ritual, including the workers of spells, may also have continued to operate in 1961, though Salisbury’s informants were reticent on this subject.

Tolai entrepreneurs had established processing facilities for the export crops, copra and cocoa. While the bulk of Tolai cocoa was processed in the fermentaries of the Tolai Cocoa Project (itself a dividend of the ‘political consolidation’ on which Salisbury laid such store), some plants were owned individually and by small groups. Established with government assistance, the Cocoa Project had formal governance and access to commercial bank funding. Due to the bureaucratic intervention which had made these facilities available, the Project had the characteristics of ‘formal’ enterprise. On the other hand, private fermentaries were informal in character, as were the more numerous private copra dryers, for which capital and technical requirements were less onerous. In addition to processing their own produce, owners allowed other growers to use these for a fee. Trucking was tried by many individuals and groups; ownership of vehicles conferred prestige and the hope of profit, though typically inadequate provision was made for capital replacement. Nonetheless, as in Hart’s Nima, some men succeeded in maintaining their vehicles and operating profitably. Small carpentry contracting was a widespread private business, relatively lucrative though sporadic and combined with agriculture and other activities. Contractors normally preferred to quote construction costs using the client’s own materials, limiting their investment to tools and the hire of labour. Pig raising had become a commercial activity, with most customers from among the Chinese community, while many households raised a few chickens for sale. Small ‘trade’ stores were common, though often short-lived or spasmodic in operation, with maintenance of working capital and credit control the principal problems. Beer was obtained and re-sold illegally and a local banana liquor was distilled. Commercial pati (parties) were frequent events, with music and drinking, and could be quite profitable for a shrewd entrepreneur.

Despite this modest profusion of informal ‘micro and small enterprise’ (to use present-day jargon) there was little evidence in 1961 of progress to larger-scale operations. Salisbury (1970, 275) commented that the ‘lack of avenues for productive investment’ was an important deterrent to business growth in Vunamami. ‘The progression of an entrepreneur from petty marketing, to copra-drying, to truck ownership, was one that was readily achievable – so much so that an oversupply of trucks had already occurred and an oversupply of drivers seemed imminent’. But market forces did not appear to force people to branch out; ‘savings remained tied up in inactive bank accounts … or were employed in duplicating facilities used only by the local community’. The stimulus for a new sequence of Tolai economic growth had not yet become apparent. Salisbury thought it might lie in the development of new and larger investment vehicles (of which the Tolai Cocoa Project was an instance), operating over extended markets and mobilizing capital from wider groupings of Tolai. This would depend on the bureaucratic capacity to devise and regulate new forms of business entity which could ‘give shareholders security of investment equal to that which they currently obtain by their informal powers of sanctioning [local] managers’ (Salisbury 1970, 276).

The capacity of the Tolai to accumulate capital had been demonstrated in their first three growth sequences, during which innovations were self-financed from subsistence resources and earnings from employment in the monetary sector. This achievement reminds us of a persistent theme in the writings of Peter Bauer, the capacity of subsistence agriculturalists to finance their own investment (Bauer and Yamey 1957, especially pp131ff). Salisbury (1971, 62) provided convincing evidence for this claim, calculating that ‘for Tolai net investment is about 20 per cent of total output in all its various forms – labour for planting of cash-crop trees, labour and cash for construction, cash for machinery, cash and stocks for working capital, shell money, cash and taxes for social infrastructure, etc’. Even so, the planting and processing of cocoa beans in the fourth sequence required the Tolai to find external finance, for which the patronage of colonial bureaucracy was required. As Salisbury reported, ‘only the last stage required large-scale cash investments, for which bank credits became available when local savings were exhausted’ (1971, 60).

There can be little doubt that Salisbury recognized early manifestations of the informal economy phenomenon, occurring in and around Rabaul in the early 1960s, and was able to explain their emergence. Insofar as Salisbury discerned an informal economy in Hart’s sense of the term, it was primarily a rural informal economy. Moreover, unlike Hart’s urban informal sector, it was not driven by the spur of rural poverty. For reasons of geography (proximity to Rabaul) and infrastructure (the well-developed road system of the Gazelle) this rural informal economy was quite well articulated with the urban centre. There is a strong sense in Salisbury’s account of a thriving set of rural-urban relationships in which the informal economy is an important element. Further, Salisbury should be credited with having recognized the possibility of a future ‘involution of services’ (as suggested by Ben Higgins). In Salisbury’s view, consistent with the centralizing processes which would bring this about, such an informal economy would occur primarily in urban areas, (and would be driven increasingly by the threat of poverty).

Salisbury saw private, for-profit service activities among the Tolai as an element in their economic progress, moving from subsistence agriculture with limited intra-regional trade based on ecological diversity, to mixed subsistence and export-cropping with an emergent service or tertiary sector. True, Salisbury’s attention was directed primarily to communal (that is, politically-inspired) service provision as the dynamic element in his model, while his discussion of the private informal economy had something of the character of obiter dicta. Considered simply as a listing and description of informal economic activity among the Tolai at the time, the study of the peri-urban village of Matupit by A L Epstein (1963) is a more comprehensive account. But Salisbury’s conceptual framework, the focus on the service sector, per se, and its role in development gives his work its particular value for our purposes.

Salisbury set the discussion in a context of structural change in the Tolai economy, drawing on Colin Clark’s notion of the growing importance of services, while echoing Bauer’s objection to the classification of subsistence cultivators as simply ‘farmers’ when in fact they devoted (on his reckoning) perhaps 20 percent of their time to service activities. Thus, in the matter of the informal economy Salisbury’s intellectual debts, principally to Bauer and secondarily to Clark, are manifest. Regarding services, the originality of his argument lies, first, in applying Bauer’s logic to a novel set of circumstances, second, in his careful measurement of the uses of labour-time, third, in his recognition of the possibility of an ‘urban involution’ of services, and, finally, in his attempt to incorporate a political dynamic which had no part in Bauer’s thesis. To say that the argument is sometimes laboured, and that his formalist/social evolutionary machinery creaks rather audibly in places, is not to deny the power of the Siane and Tolai studies as records of ‘economic transformation in a traditional society’ (to recall the subtitle of Vunamami).


  1. Conclusion

This concluding section draws together the ideas of the ‘quartet’ – Hart, Mayhew, Bauer and Salisbury – to examine points of congruence and divergence and give credit to each, as seems appropriate. Hart’s conception of the informal economy arose from his capacity to see, with great clarity, certain realities associated with the getting of livelihoods in Nima. His vision was not obscured by categories devised for the regulation of an economy governed by particular ‘forms’. Neither was his vision coloured by judgments, drawn from the norms of ‘respectable’ society, of the propriety of the livelihoods he observed. As viewed from the streets of Nima, the ‘forms’ of the economy, as ordained by authority, were best interpreted as a vision of a world only partially realized – the world as it might become, rather than as it was in 1966. Whether or not Hart himself shared any such vision, he rendered a service by pointing to the discrepancies between vision and reality, and by questioning the categories imposed by theory and bureaucracy on an unruly actuality. Understanding and accounting for such discrepancies is surely a prerequisite for any purposive program. We do not find this same appreciation of the significance of bureaucratic intervention in the work of those other writers selected here as Hart’s ‘precursors’.

Central to the narrative of each was how people gained their livelihoods informally, outside the ‘official’ frameworks of the economies in which they lived and worked. Although differences in historical setting, environment and other circumstances operated to distinguish each case study, our task is to consider the degree to which genuine commonalities emerge from these accounts. All dealt with societies in transition, but while Mayhew was concerned with a period of gathering industrialization, political change and social dislocation, the others were placed at various points along a path from subsistence agriculture to market exchange, and at various stages of political development. The provision of services played a central role in each of these disparate accounts of informal economies, though with different degrees of emphasis on the role of the service ‘sector’ per se.

Salisbury and Bauer both accorded special status to services, with the former dubbing them the ‘leading sector’ and setting out a model of development based on the progressive accretion of capacity for service delivery on the foundation of an ‘affluent’ subsistence economy. For our purposes his obiter dicta, descriptions of the Tolai service economy, are the more interesting elements of his contribution. Bauer also privileged services, singling out transport and trade as crucial in the transition of economies from subsistence agriculture to market exchange. Both Bauer and Salisbury engaged with Clark and Fisher’s celebrated generalization concerning the increasing role of services in the course of economic growth. Bauer pointed to the under-enumeration of service activities in agricultural societies during the transition from subsistence, due to what he described as ‘imperfect specialization’. Salisbury, following Bauer, added empirical weight to this analysis by measuring the time-budgets of individual Tolai ‘farmers’, confirming Bauer’s results. Salisbury nonetheless accepted the broad sweep of the Clark/Fisher ’hypothesis’, which underpinned his ‘leading sector’ proposition, whereas Bauer regarded it as having been undermined. However Bauer’s view has not prevailed in the continuing economic debate, which accepts the reality of a shift to services and focuses instead on the causal mechanisms underlying the shift. Bauer did not falsify Clark/Fisher, rather he offered an intriguing qualification to the hypothesis touching on the transition from subsistence, and in the process provided a pioneering account of the economics of the informal sector. Neither Bauer nor Salisbury was concerned explicitly with ‘form’, as such, though readers coming to their work from Hart will find themselves in familiar territory. But this is also true of Henry Mayhew, from whose rich body of observation Hart might almost (like some latter-day Marcel Mauss) have derived the elements of his ‘informality’ construct without troubling to visit Africa.

Important environmental differences affected the societies described by the quartet, influencing the nature, and in particular the urgency, of economic activity. At one extreme, Salisbury’s Tolai pursued a relatively small range of livelihoods, buffered by the ‘affluence’ of the subsistence sector which provided the base for their economy. While he considered the possibility of an involution of urban services, Salisbury believed this could be avoided. At another extreme, Hart’s rural-urban migrants, the Frafra, suffered ‘extreme pressure on the land at home’ (Hart 1970, 105). For many of them, returning to the subsistence sector was not an available option. Making allowance for the radically different context of mid-Victorian London, we may still imagine why refugees from Ireland and Eastern Europe, artisans displaced by technological change, and the indigent elderly facing the severities of the workhouse might feel obliged to ‘hustle’ with a degree of desperation similar to that ascribed by Hart to many of the Frafra. The term ‘sub-proletariat’, which Hart applied to the Frafra in Nima, seems equally apposite to Mayhew’s London poor.

The circumstances of Bauer’s West Africans are rather more ambiguous. Unlike Hart, Salisbury and Mayhew, he drew upon the experience of an enormous geographic region and a population both substantial and heterogeneous. His generalized ‘West Africa’ resembled Salisbury’s Gazelle insofar as it still had a land frontier for the expansion of export cropping. However it also contained population sub-groups suffering land shortage, Hart’s Frafra among them. Much later, Michael Lipton would criticize Bauer’s analysis, charging that Bauer’s ‘classical’ prescription, that ‘enterprise, trade, enlargement of markets’ would act as ‘engines of development’, was persuasive only because he drew his generalizations from ‘lucky places’. Bauer’s analysis, based on experience in colonial West Africa and Malaya, had produced an optimistic scenario biased by their favourable environments (Lipton 1984). This may be a just observation, but for present purposes it is worth noting that Bauer’s West Africa had an economy where opportunity (in the guise of an expanding frontier and rapid growth of commercial agriculture) co-existed with deprivation among sections of the population, such as Hart’s Frafra. That low-cost labour was readily available to fill the niches of opportunity created by a flourishing export agriculture helps to explain the flowering of Bauer’s intricate and dynamic informal economy. By contrast, Tolai labour costs were relatively higher in 1961, providing fewer opportunities for such development on the Gazelle Peninsula.

Certainly, areading of Bauer does not suggest the desperate or involutionary character evident in at least some aspects of informality, as documented by Mayhew and Hart. This may simply reflect an optimism founded on Bauer’s trust in the ultimate success of the ‘engines of development’ posited by Smithian economics. In any event, both Bauer and Salisbury were enthusiastic supporters of the informal economy as they recorded it, whereas Hart remained unconvinced whether its net value to the economy was positive or negative. Mayhew, like Hart, reported positive contributions of the informal economy, both to higher income groups and to the poor themselves, but was nonetheless scathing of a social order that could permit such injustice.

Comparisons between the four bodies of work discussed here bring out contrasts, not only in geographic location but also in the specific locus of research. Hart introduced an urban informal economy, though there is enough in his account to convey a sense of the rural hinterland from which the Frafra came and with which they maintained economic and social relationships. Both Bauer and Salisbury dealt for the most part with rural informal activities, with the former describing the role of trade and transport in opening up chains of distribution deep into the African interior (though there are also numerous discussions of urban activities). Salisbury conveyed a sense of close economic relationships between the rural villages of the Gazelle and their quite accessible district town. There is in both cases a stronger sense of rural-urban relationships than appears in Hart’s work and the reader is entitled to conclude from the work of Bauer and Salisbury that the concept of informality may be extended fruitfully into the countryside. Mayhew’s observations confirm this; he gave numerous accounts of London pedlars making rural rounds, of traders seeking materials in the countryside, and of rural folk in town for trading. Much of this circular movement was seasonal in nature and was complemented by seasonal movements of labour, extending as far afield as Ireland.

A defining characteristic of the informal ‘sector’, the non-enumeration of informal workers and activities, was stressed by all four, and all were critical of the deficiencies of official data. The existence of ‘imperfect specialization’ and ‘occupational fluidity’ as underlying this problem was documented by frequent references in all studies to the multiple sources of income typically enjoyed by operators. Such fluidity leads to complexity and in the four cases there are substantial differences in the degrees of complexity observed. Not surprisingly, the Tolai informal economy revealed a comparatively small number of roles or livelihood activities, reflecting perhaps its recent emergence from a relatively undifferentiated subsistence condition. Intermediary functions and other occupational diversity were correspondingly underdeveloped. Other factors, including relatively limited demand for non-traditional services and good infrastructure (the latter limiting the scope for informal transport and storage options) may also have contributed to this comparative lack of complexity. Mayhew’s London revealed perhaps the most fine-grained and complex system of livelihoods, while both Hart’s Frafra and Bauer’s West Africa demonstrated considerable complexity.

A somewhat disappointing aspect of Hart’s account of Nima is the limited attention he gave to women in the informal economy (aside from ‘red light’ activity). This is surprising since the ‘Mammy’ traders of West Africa were well-known for their importance in marketing and distribution15. Bauer had a somewhat fuller account while Mayhew’s thoroughness is evident in his recording of many livelihoods peculiar to women, as well as those in which they competed with men. Salisbury gave particular attention, as we have seen, to the market trading of women on the Gazelle. He documented, in meticulous detail, their contribution to the subsistence agriculture which undergirded the changing rural economies of the Siane and the Tolai, while his discussion (in From Stone to Steel) of the impact of technological change on women is notable.

As might be expected of anthropologists, Salisbury and Hart tell us more about the importance of traditional (or quasi-traditional) mechanisms of mutual support and exchange than is available in the accounts of either Bauer or Mayhew. In Hart, this was often associated with patron-client relations. Given his emphasis on the unreliability of income streams in Nima, personal transfers and access to credit were important for consumption-smoothing, indeed for survival. For Salisbury, traditional exchange relationships, and the status rewards accruing to their successful manipulation, provided the motor force for much social change (and especially ‘political consolidation’) among the Siane and the Tolai. Mayhew has much more to say about credit than any of the others and despite the tribal solidarity of his Costermongers, we are left with the impression that in mid-century London ‘consumption-smoothing’ required access to usurers as often as to relatives. Indeed, many Costers accessed credit on a daily basis, simply to go about their trades.

In regard to illegal or ‘illegitimate’ activities, we learn almost nothing from Bauer or Salisbury (leaving aside the latter’s few references to banana ‘hootch’, ‘sly-grog’ and pati). Mayhew is reliably encyclopedic on the subject, from a socially-concerned and normative standpoint. But only Hart really confronts the issue of ‘pervasive’ illegitimate activity on his patch. Indeed the legitimate/illegitimate distinction is his, occurring in the context of a non-judgmental listing of informal occupations in Nima. This is done to show ‘how things work’, as an illustration of the phenomenon of multiple income sources, and as a negative example (in an accounting sense) of the economic flows between Nima and middle-class Accra. This suggestion, in effect for a flow-of-funds chart for Nima, is another sense in which Hart applies the term ‘transfers’. A related, but analytically distinct, phenomenon is the activity of influence-pedlars and political ‘fixers’, to whom Hart gave detailed attention. Aside from deploring the ‘politicization’ of economic life in developing countries Peter Bauer was uninformative on this subject, while Mayhew preserved a discreet silence. Salisbury, however, put ‘political consolidation’ at the centre of his theoretical apparatus. But while Hart was non-judgmental in regard to political fixing, Salisbury (from the perspective of 1961) appears to have viewed it as a constructive force.

All four observers discussed here produced accounts of continuing interest, providing, to a greater or less degree, insights into our contemporary experience and propositions of continuing explanatory power. Mayhew alone had no pretension to theorizing or generalization beyond his time and place, yet his enduring impact flows – these words are used without ironic or pejorative intent – from a virtuous polemicism. Salisbury certainly did pretend to theorization, but in his case the ‘pay-dirt’ lies hidden in the details rather than in the thesis. His rather cumbersome ‘model’ of service-led growth gained no traction, but nonetheless minutiae and footnotes in his argument (his ‘obiter dicta’) throw shafts of light onto the contemporary situation of PNG. Peter Bauer’s analysis was informative and stimulating. He certainly purported to generalize, and Lipton was no doubt correct to criticize his reliance on those colonial territories most promising for the application of the nostrums of classical economics. But whatever about the ‘Baverian’ worldview, as now celebrated in conservative think tanks, his application of a relatively simple set of conventional economic tools to the mechanics of the informal economy, reinforced by a sharp eye for social and cultural context, provided many satisfying explanations and insights.

Bauer might have been (as Basil Yamey believed) the first economist to see the scope and economic significance of the activities of the informal ‘sector’. But it was left to Keith Hart to recognize them as a phenomenon in their own right, and to provide an appropriately interdisciplinary framework for their analysis. Nowadays Hart evidences quite some ambivalence about this concept he introduced forty years back, and which continues to dog his footsteps. Observing its capacity for renewal, and adaptation to a wide range of circumstances and to changing times, he has come to see the informal economy as ‘a remedial concept’, serving to reconcile orthodox economic analysis with reality. It is as applicable to the remaining ‘socialist’ regimes (Cuba, North Korea) as to the market economies of the industrialized world and the ‘state capitalist’, ‘transitional’ or ‘emerging market’ systems of the developing world, for

‘As long as there is formal economic analysis and the partial institutionalization of economies around the globe along capitalist or socialist lines, there will be a need for some such remedialconcept as the informal economy. Its application to concrete conditions is likely to be stimulated by palpable discrepancies between prevalent intellectual models and observed realities’ (Hart 2007, italics in original).



Ackroyd, Peter (1991), Dickens, London, Minerva Mandarin

Bauer, P T (1948), Report on a visit to the Rubber Growing Smallholdings of Malaya, Colonial research Publications (Great Britain, Colonial Office) #1

Bauer, P T (1954), West African trade: a study of competition, oligopoly, and monopoly in a changing economy. Cambridge, CUP

Bauer, P T (1956), ‘Lewis’ Theory of Economic Growth: A Review Article’ American Economic Review, 46(4), 632-41

Bauer, P T (1984), Remembrance of Studies Past: retracing first steps, in Meir, Gerald M and Dudley Seers, Pioneers in Development, New York, OUP for the World Bank

Bauer, P T and Basil Yamey, (1951) ‘Economic Progress and Occupational Distribution’ Economic Journal, 61 (244) December, 741-755

Bauer, PT and BS Yamey (1957), The Economics of Under-Developed Countries. Cambridge, CUP

Clark, Colin (1940, and successive editions), The Conditions of Economic Progress, London, Macmillan

Clark, Colin (1984), Clark, ‘Development Economics: the Early Years’, in Meir and Seers (eds), Pioneers in Development, 59-77

Disraeli, Benjamin (1845 [1998]), Sybil, or the Two Nations, Oxford, OUP

Elkan, Walter, T C I Ryan and J T Mukui (1982), ‘The Economics of Shoe-shining in Nairobi’, African Affairs, 81 (323), April, 247-56

Epstein, A L (1963), ‘The Economy of Modern Matupit’, Oceania, 33(182), 182-215

Fisher, A G B (1933), ‘Capital and the Growth of Knowledge’, Economic Journal, 43 (171), September, 379-389

Fisher, A G B (1945), Economic Progress and Social Security, London, Macmillan

Frank, Andre Gunder (1967), ‘The Sociology of Development and the Under-development of Sociology, Catalyst, Summer, 20-73

Geertz, Clifford (1963a), Pedlars and Princes, Chicago, University of Chicago Press

Geertz, Clifford (1963b), Agricultural Involution, Berkely and Los Angeles, University of California Press

Hann, Chris and Keith Hart (2011), Economic Anthropology, Wiley

Hart, Keith (1970), ‘Small Scale Entrepreneurs in Ghana and Development Planning’, Journal of Development Studies, 6(4), 104-120

Hart, Keith (1973[1971]),‘Informal income opportunities and Urban Employment in Ghana’, paper delivered to Conference on Urban Unemployment in Africa, Institute of Development Studies, University of Sussex, 12-16 September, 1971 (Excerpted in Jolly, Richard, Emanuel de Kadt, Hans Singer and Fiona Wilson (1973), Third World Employment: Problems and Strategy, Harmondsworth, Penguin, pp66-70

Hart, Keith (1973), ‘Informal Income Opportunities and Urban Unemployment in Ghana’, Journal of Modern African Studies (March)

Hart, Keith (2002), ‘World Society as an Old Regime’, in C. Shore and S. Nugent eds Elite Cultures: anthropological perspectives, Routledge, London. Accessed online at The Memory Bank

Hart, Keith (2004), ‘Formal Bureaucracy and the Emergent forms of the Informal Economy’, online at ,subsequently published as Hart, Keith (2006), ‘Bureaucratic form and the informal economy’ in B. Guha-Khasnobis, R. Kanbur and E. Ostrom (eds) Unlocking Human Potential: Formality and Informality in Developing Countries, Oxford, OUP

Hart, Keith (2007), ‘The Informal Economy’ In Eatwell, John, M. Milgate and P. Newman (eds) The New Palgrave: A Dictionary of Economic Theory and Doctrine Vol. 2 , Macmillan, London pp. 845-46

Hart, Keith (2010), ‘Informal Economy’, in Hart, Keith, Jean-Louis Laville and Antonio Cattani (eds), The Human Economy, Cambridge, Polity Press, pp 142-153

Hill, Polly (1963), Migrant Cocoa Farmers of Southern Ghana, New York, CUP

IEA (The Institute of Economic Affairs) (2002), A tribute to Peter Bauer. London, IEA

ILO (International Labour Organization) (1972), Employment, Incomes and Equality: A strategy for increasing productive employment in Kenya. Geneva, International Labour Organisation

ILO (International Labour Organization) (2007), The Informal Economy: enabling transition to formalization (ISIE/2007/1), Geneva

Jones, Gareth Stedman (1971), Outcast London: a Study in the Relationship between Classes in Victorian London, Oxford, Clarendon

Lal, Deepak (2002), ‘A Dissident Vindicated’, in Institute of Economic Affairs, A tribute to Peter Bauer, London, IEA, pp 74-77

Lewis, Oscar (1966), La Vida; A Puerto Rican Family in the Culture of Poverty, San Juan and New York, New York, Random House

Lewis, W. Arthur (1954), ‘Economic development with unlimited supplies of labour’, The Manchester School, v.22, 139-191

Mayhew, Henry (1851 – 1862), London Labour and the London Poor, a cyclopaedia of the condition and earnings of those that will work, those that cannot work, and those that will not work (4 volumes)

Mayhew, Henry (1861), London Labour and the London Poor, Vol I, London, Griffin, Bohn and Company, University of Virginia E-text at

Mayhew, Henry (1861), London Labour and the London Poor, Vol II, London, Griffin, Bohn and Company, Tufts University Digital Collections at

Mayhew, Henry (1862), London Labour and the London Poor, Vol III, London, Griffin, Bohn and Company, Tufts University Digital Collections at

Meir, Gerald M and Dudley Seers (1984), Pioneers in Development, New York, OUP for the World Bank

Overseas Development Group (1973), A Report on Development Strategies for Papua New Guinea (The ‘Faber Report’). Port Moresby,The International Bank for Reconstruction and Development as Executive Agency for UNDP

Papua New Guinea, Independent State of (1975), Constitution of the Independent State of Papua New Guinea

Petty, William (1690 [1989)), Political Arithmetick, London, in Hull, Charles H (ed), Economic Writings of Sir William Petty, Cambridge, CUP

Quennell, Peter (1984), Mayhew’s London. London, Bracken. [An abridged edition, edited by Peter Quennell, of Mayhew, Henry (1861 [1984]), of London Labour & the London Poor (3 Vols)]

Robbins, Lionel (1935), An Essay on the Nature and Significance of Economic Science, (2nd rev. edition), London, Macmillan

Rostow, WW (1960), The Stages of Economic Growth: A non-communist manifesto. Cambridge, CUP

Sahlins, Marshall (1958), Social Stratification in Polynesia, Seattle, University of Washington Press

Salisbury, RF (1962), From Stone to Steel: Economic Consequences of a Technological Change in New Guinea. Melbourne, Melbourne University Press

Salisbury, RF (1970), Vunamami: Economic Transformation in a Traditional Society. Berkeley, Univ of California Press

Salisbury, RF (1971), ‘Development through the Service Industries’, Manpower and Unemployment Research in Africa, Vol. 4, No. 1, pp 57-66

Schettkat, Ronald and Lara Yocarini (2006), ‘The Shift to Services Employment: a review of the literature’, Structural Change and Economic Dynamics, 17, 127-47

Swift, Philip, ‘The “Perils of the Nation”: The Activist Anthropology of Henry Mayhew’, Open Anthropology Cooperative at

Todaro, MP (1969), ‘A model of labor migration and urban unemployment in less developed countries’, American Economic Review, 59, 138-148

Weber, M (1981), General Economic History, New Brunswick NJ, Transaction Books

WTO and ILO (World Trade Organization and International Labour Organization) (2009), Globalization and Informal Jobs in Developing Countries, Geneva

Yamey, Basil (1987), ‘Peter Bauer: Economist and Scholar’, Cato Journal, 7 (1) Spring/Summer,21-27

Yamey, Basil (2002), ‘Collaborating with Peter Bauer’, in IEA (2002), pp 96-99

Yamey, Basil (2005), ‘Peter Bauer, an Unusual Applied Economist’, Cato Journal, 25, 3, pp 449-453



1 John Conroy is an economist, not entirely unreconstructed although he still calls himself a ‘development economist’. He spent the 1970s in Papua New Guinea, finishing as Director of the PNG Institute of Applied Social and Economic Research. After six years in Indonesia, during the 1980s, he ran the Foundation for Development Cooperation through the 1990s, specializing in microfinance and ‘financial inclusion’. He abandoned that field in disgust after its invasion by private equity and is now a visiting fellow at the Crawford School of Economics and Government at the Australian National University.

2 I am grateful for discussions with Ron Duncan, Scott MacWilliam, Peter Drake and Andrew Elek; also to Michael Bourke and other participants in the Crawford seminar at ANU in August 2011, at which a preliminary version of this paper was presented. Most particularly I am grateful to Keith Hart for patiently fielding all the balls I have tossed up over the past several years.

3 For an early application of the concept to industrial societies by a sociologist, see Pahl (1984). For a spirited application to the contemporary world economy see Hart (2004).

4 Hart’s use of the term ‘transfers’ in this connection is somewhat confusing; ‘illegitimate’ transfers are to be distinguished from interpersonal transfers among members of kin- or social groups.

5 A more positive evaluation occurs in a scenario in which Hart suggested the processes of the Lewis model (Economic Development with Unlimited Supplies of Labour) might assure the eventual absorption of surplus urban African labour into formal employment, while being supported in the interim by ‘the inherent economic dynamism of concentrated urban markets, which [as in Nima] generate an almost infinite range of activities based on commodity exchange’ (Hart 1982, 162).

6 I am grateful to Keith Hart for having referred me to Mayhew in 1972.

7 In this study, reference is made to a Mayhew compilation edited by Quennell (1951 [1984]) and to online sources for the original three volume version (cited as Mayhew I, II and III). I have not drawn on volume IV, published in 1862, since Mayhew was not its sole author. Vol. IV was devoted to the London petty underworld, described adequately for our purposes in the first three volumes.

8 For which reason I adopt Hart’s usage here; he changed his description of the Nima underclass (as noted above in Section 2) from a ‘proletariat’ (Hart 1971) to a ‘sub-proletariat’ (Hart 1973). The former might be appropriate in a Two Nations framework, but the latter fits the reality of Mayhew’s London quite as well as it does Hart’s Nima.

9 Material in this section is drawn from Mayhew I, ‘Of the markets and trade rights of the Costermongers and of the laws affecting them’, ‘Of the removals of Costermongers from the streets’ and ‘Of the variety of street-folk in general and Costermongers in particular’ .

10 See for example ILO (2007) and WTO and ILO (2009).

11 Polly Hill (1963) was an honourable, if later, exception to this generalization.

12 Bauer has a posthumous Facebook page and was the subject of tribute volumes by Cato (in the US) and the IEA (in the UK) before his death. The term ‘ikon’ is appropriate, since some of the tributes paid him border on hagiography.

13 Most of the citations in this section, concerned with the sectoral distribution issue, are drawn from Chapter 3 of the 1957 book, except where the 1951 paper is cited explicitly.

14 See Schettkat and Yocarini (2006) for a review of the literature on ‘the shift to services employment’. On the basis of the empirical evidence they concluded that ‘Employment in the advanced economies shifts with a remarkable regularity towards services as income per capita rises … ’ (Schettkat and Yocarini 2006, 142).

15 Hart (pers. comm. 16 August 2011) has responded to this comment by noting that his study was of a particular ethnic group. ‘There are few women in my account because they were almost entirely absent from the Frafra migrant community in Nima which consisted of a sea of transient young men loosely attached to a few older men with established families’.