Du grain à moudre|
Commodification and masculinisation in post-harvest systems for rice in South Asia
Note de l’auteur
This is a reworked and updated version of the paper “Commercialisation, commodification and gender relations in post-harvest system for rice in South Asia” which was first published in Economic and Political Weekly (June 18th 2005, 2530-42). I am very grateful to Maitreyi Krishna Raj and Mina Swaminathan for their comments then. And to Christine Verschuur and Marie Monimart at the stimulating conference Du grain à moudre. I am grateful to Ursula Huws for her engaged response to the original draft, and to P. K. Ghosh and Dr Rupinder Kaur for the field research they carried out under the joint NCAER/Queen Elizabeth House, Oxford, research project on Trade liberalisation and India’s informal sector which helped me update the analysis of rice processing.
- 2 A rice marketing firm may do one or a combination of the following activities: buy, sell and broker (...)
1For forty years the production of rice worldwide grew at rates greater than that of population. Rice production more than doubled – from about 265 million tonnes in 1961 to about 560 in 2001 (Rai 2003). When the output of a product that has been the basis of direct use for subsistence and social reproduction – as rice has been in Asia – expands, the marketed surplus rises disproportionately to the growth rate of production. Post-harvest activities that were part and parcel of the reproductive activity of household labour (in the hands and under the feet of women – even if under the control of men) then also become commercialised. They move from being a set of crafts to being mass-produced. Firms expand in number and labour markets sprout up as firms become differentiated in size, scale and activity2. Food security comes to depend on the market, as well as on the social and political structures in which all markets are embedded.
2Gender relations pervade these regulative social structures. While the gender division of tasks in rice production in South Asia is quite diverse (Mencher and Saradamoni 1982; Kapadia 1993; Jackson and Palmer-Jones 1998; Jackson 1999), that of the post-harvest system appears to be much less so. One of the strongest patterns in the post-harvest system takes the form of a trend in post-harvest female livelihoods: while the absolute number of these may be decreasing, static or increasing, their proportion relative to those of men is declining over time. Men dominate control over commercial assets and technological change displaces female labour disproportionately to that of men (Harriss-White 2004a).
3This essay is a reflection on the process of masculinisation of the markets and the post-harvest system for rice. It is a social process that was first described and modelled in general terms and descriptively as early as 1972 by the Danish economist Ester Boserup. She even gave it a name: productive deprivation. The essay has two parts. In the first, Boserup’s argument is developed from an analytical perspective that combines class, caste and gender, and then tested against four regional cases. In the second part, a counter-theory of commodification pioneered by Ursula Huws is developed for the class-differentiated rice markets of South Asia. We find first that Boserup’s “productive deprivation” is a class-specific process and that masculinisation co-exists with a high level of female economic participation. Second, we find that the process of proliferation of markets and commodities is instituted in ways which have perpetuated petty scales of production and unwaged women’s work in conditions of economic and social insecurity.
4Economic and social development unavoidably entails the disintegration of the division of labour among the two sexes traditionally established in the village. With modernisation of agriculture and with migration to the towns, a new sex pattern of productive work must emerge, for better or worse. The obvious danger is, however, that in the course of this transition women will be deprived of their productive functions, and the whole process of growth will thereby be retarded (Boserup 1989, 5).
- 3 By this, even then Boserup must have meant upper caste Hindus.
- 4 This process of marginalisation continues to this day (Hensman 2001), despite the rise in female em (...)
- 5 Demand disincentives include rules on maternity benefits, child care and equal pay. Supply disincen (...)
5The process of deprivation proceeds from the agriculturalisation of the peasantry and the stripping of crafts from the work of agricultural households. As craft production becomes specialised, it increases in scale, production being organised either in households (according to household divisions of task and authority) or through male wage labour. As the division of labour deepens and exchange becomes fundamental to social reproduction so tasks are progressively defined by categories of worker, in which skilled categories are dominated by men (Boserup 1989, 69-76). Boserup then turns her attention to towns. South Asian towns are male domains, either through selective male migration, because of the seclusion of women, or both. The prospects for women’s work in towns is related to the rural gender division of labour in the non-farm economy; while in northern India “men even do the shopping” (86), in the south, in what she recognises as the “semimale town”, retail trade may be in the hands of women. However, “to most Hindus the idea of female participation in trade is an abomination” (87)3 and “modern sector” bureaucracy, industry and markets are dominated by men. Even in South India, she observed a “deepening cultural resistance to women’s participation in trade” (98). Female work is then confined to un-secluded women from the lowest castes who provide artisanal, home-based, petty production as well as a variety of services. Boserup shows convincingly that women are progressively marginalised from wage work in factories and that female activity rates decline with development (192)4. Both demand and supply factors play their role in this. Employment regulations for women increase their cost while the inflexibility of modern industrial discipline is incompatible with the rearing of children (110-17)5. She reaches a powerful conclusion: “If women are hired at all… it is usually for the unskilled, low-wage jobs, men holding skilled jobs. Thus the roles assigned to men and women even in the modern sector indicate a widening difference between the productivity and earnings of each” (139-40).
6Boserup did bring broad brush global evidence to bear on her thesis. But neither her theory nor her evidence enabled the process of productive deprivation to be related to class formation. This is our first task here. Given the spectacular growth of female employment in the Indian service sector, to test for productive deprivation within a single commodity-branch of the Indian economy might be considered inappropriately restrictive. Yet rice is the most important agricultural commodity and a fundamental part of India’s agrarian structure, food culture and food security. The steps involved in the marketing, processing, storage and transport of rice are set out in Appendix I (which is heavily derivative of Kaur 2004). There is rather little evidence about the gendering of the rice markets of India. What exists is not continuous in time, it is patchy in space and the product of individualistic research methodologies. It can however be used to explore Boserup’s argument about masculinisation and productive deprivation. First, the class positioning of women will be examined using evidence for Tamil Nadu and West Bengal in the 1980s and 1990s (Harriss 1993; Harriss-White 1996; 1999). Second, the impact of technical change on such institutional arrangements will be explored with evidence from Bangladesh in the late 1970s, Tamil Nadu in the 1980s, and West Bengal and Punjab in the current era (Harriss 1979; Harriss-White forthcoming; Ghosh and Sudarshan 2004; Kaur 2004).
7The emergence of classes of employers and employees, of capital and labour, does not happen in a vacuum. Far from it. Not only are grain markets strongly regulated through gender relations but they are also structured through other aspects of identity, such as caste and religion. These aspects of culture are reworked as regulators of economic behaviour, a process Kate Meagher has labelled “identity economics” (Meagher 2010). Thus, women participate in paddy-rice markets in four ways, according to their caste and class position.
Women from pauperised, female, female-headed and/or low caste petty producing and trading households are confined to: seasonal operation, subsistence orientation, particular stages and activities within the post-harvest system (especially processing and retailing), localised territorial linkages, weekly marketplace sites and unlicensed and/or illegal transactions in cash rather than on credit. Their participation is conditioned by the life cycle and when children are no longer dependent, this activity is abandoned. They experience marketplaces as physically dirty and ritually polluting places; so they regard non-participation as “development”.
- 6 The female casual labour force can number up to 700 in cotton ginning and wholesaling firms (see Ha (...)
- 7 “Coolie” in Tamil means wages for casual labour.
- 8 In any case this would be almost impossible to measure accurately since the division of tasks in mi (...)
Low caste, female casual wage workers from the assetless class form the largest sub-stratum of labour in grain milling and premilling processing. In South India, marketing systems rest on the backs of these women, who form from 40 to 60% of the labour force (Harriss-White 1996, 266). The average mill employs 15 such women, but up to 70 have been encountered6. Out-caste women are allowed to turn paddy on the large sun-drying yards surrounding the rice mills, as the kernel is still protected from ritual pollution by its husk. Women are debarred from being mill mechanics and it is unusual to find them handling heavy consignments of scalding paddy during the parboiling process. Female coolie7 is prevalently but incorrectly regarded by mill owners as a household supplement for their employees. Wage differentials are sharp, from two thirds to a half that of male wages in rice mills, and in no way reflect gendered productivity8. Female mill work is almost everywhere deliberately casualised. The sexual harassment of the mill workforce by management is not unknown.
- 9 With the commercialisation of labour, however, the practice of payment in tea and meals is dying, o (...)
In smaller family firms (of “backward” or higher castes), unwaged female family members have provided that part of labour which takes the form of prepared food9. The female family labour of a rice milling and trading firm will almost certainly “subsidise” some of the costs of reproduction of their male labour-force. This means that the social reproduction of male labour across the generations is not entirely born by female labour within working class households but extends to the women of some small family businesses as well.
- 10 Male family labour (up to 13 members have been encountered) whose work is loosely specified, may wo (...)
- 11 Female education leads to the lowering of birth rates, though not to reduced gender discrimination (...)
- 12 See for example Laidlaw’s 1995 study of Jain families.
A typical milling and trading firm is owned by the males of a joint family, and practises a diversity of trading and transformation activities. As a consequence, there is a complex of patriarchal relations between male family labour, together with the hiring of a permanent male labour force and casual labour of both sexes10. Women in large milling and trading firm-families are used for the caste-based reproduction and expansion of firms, first by means of their dowries on marriage and second (but rarely) through the practice of fictitious (benami) registration of a trading company in a woman’s name, generally for the purposes of tax avoidance. In the first case, the higher education of such women is a good example of the economic inefficiency of gender institutions. For such women, education is a status good and leads to neither economic participation nor control over assets and over major economic decisions11. Women are positioned according to a paradigm of service and subordination in which their piety is factored into a business family’s reputation12. The prevalent lack of ownership or control over property, or over any collateral which determine creditworthiness, makes the economic role of women belonging to the accumulating oligopolies an indirect one, in contrast to the role of poor women in petty trade and the casual labour force.
In sum: While a model of women as labour and men as owning assets is altogether too simple, women shape the economic returns of rice milling through their compliance with norms of seclusion: their work is symbolic but has real effects. It is clear that women in business families do indeed experience the productive deprivation of which Boserup wrote. But the relevance of Boserup’s theory of productive deprivation is confined to this specific class of woman. In almost the entire post-Independence period – and with few exceptions throughout the subcontinent – India’s rice markets have depended on low caste female toil while higher (business) caste men control economic assets, income and profit.
8The gender division of tasks is a social institution and masculinisation is thus an example of institutional change. Institutional change is argued by institutional economists such as Douglass North (1992) to be driven by changes in relative prices, in technological change (or transfers), or both. Partha Dasgupta (1993) has developed a gendered account of technological and institutional change. Like Ester Boserup, he sees productivity increases focused on “male activity”: the gender differential in productivity will thus widen over time. He predicts:
Where household demand for goods and services reflects male concerns, we would expect technological inventions in farm equipment and techniques of production to be forthcoming in response. Where cultivation is a male activity… we would not observe much in the way of process innovation in threshing, winnowing, grinding grain and preparing food. (1993, 335)
9Is this a prediction that is robust throughout South Asia? Four case studies will be used here to interrogate the relationships between technical change, prices and gendered institutional change: Bangladesh in the late 1970s, Tamil Nadu in the 1980s and West Bengal and Punjab in the early 21st century.
- 13 Soaking up to 48 hours, parboiling in small batches (30 minutes to 3 hours), drying, raking and bir (...)
- 14 Parboiling insures against the vagaries of rainfall; it reduces the proportion of broken rice grain (...)
10As late as the late 1970s, two thirds of Bangladesh’s paddy was milled by women, in their homesteads (baris), by means of the foot-operated wooden hammer mill (the dheki). One to two women could de-husk a maund (37,3 kg) a day. Women were also completely responsible for pre-milling processing13. Parboiling originated as a form of short-term under-water storage when paddy was harvested during the monsoon. It also gelatinizes the rice kernel – with many nutritional benefits14. Since one woman could process 4-5 maunds a day, dheki husking was the constraint on the scale of operations, a fact of no great significance when the preparation of paddy was simply for subsistence of other processes. Women also winnowed and organised the use and disposal of husk and bran. Not all this labour was unwaged: estimates vary according to the region, but between 5% and 60% of households hired women out, and about 15-20% of households hired women in for the production of rice for subsistence. Paddy-processing wage-labour was paid in combinations of cash, rice and meals. These labouring women were among the most disadvantaged: pauperised widows, divorcees, and wives of sick or landless husbands (Arens and Van Beurden 1977; Wood 1976; Adnan et al. 1976). In the late 1970s, roughly 2,7 million woman-years of employment were estimated to be generated by the domestic processing of paddy for the population of 20 million working women (Harriss 1979).
- 15 The distress consequent to displacement is most acute in the north of the subcontinent (see Greeley (...)
11By the late 1970s, about 20% of paddy production was passing through small huller mills (the “Lewis Grant huller”, adapted in the early 20th century from a coffee grinder). Following state-regulated rural electrification, this cost-effective huller was diffused widely throughout the entire subcontinent. Huller mills were growing at 7% every year, sustaining local “innovation rents” with rates of return averaging 70%. These mills were operated by men who supervised a team of up to 10 women working on the drying yard, sieving the residues and separating husk and bran. The money costs of hulling were one twelfth that of dheki processing. Productivity per woman in the hulling mill was up to 33 times that of the dheki and in pre-milling processing was 1,5 times that of the bari. The result was a large scale displacement of human energy and its replacement by energy from fossil fuel. The first wave of mechanised technology was the most female-labour-displacing of all, as it replaced foot-operated machines (or elsewhere in South Asia the hand operated pestle and mortars) which had been worked entirely by women15. This uncelebrated process destroyed tens of millions of seasonal livelihoods for women in South Asia as a whole.
12The modernisation of rice milling received less attention than the Green Revolution in production, but the two processes happened side by side. Since the late 1960s, technical change in rice processing has involved foreign transfers and imports of technology developed for the radically different “factor endowments and ratios” of the U.S. and Japan. This was both preceded and accompanied by foreign technical assistance, the quality of whose evidence-base and policy advice left a great deal to be desired (Harriss 1976; Pacey and Payne 1984), and domestic Indian legislation outlawing the “indigenous” technology of the huller in 1970 (Harriss and Kelly 1982). Blinkered technical appraisals, indicating that the new technology increased the productivity of both capital and labour, wrenched mills from their institutional and logistical contexts and led to the adoption/imposition of imported technological packages. The package included bulk parboiling (sometimes pressure-parboiling), conveyor-belt transport, silo-storage, paddy cleaning, rubber-roll shelling and cone-polishing. State institutions (cooperatives and parastatals) regulated these modern rice mills (MRMs), but un-coordinated with other state and private institutions of finance, pricing, procurement and logistics upon which the package had to depend. The low capacity utilisation and higher cost per unit of marketing and processing were the inevitable early result, and led to the MRMs’ long term dependence upon state subsidies.
- 16 Gita Sen uses the Bulletin of Food Statistics to indicate 91,000 mills in 1975 (Sen 1983, 22).
13In India as a whole, meanwhile, wherever a marketed surplus was being generated, the “illegal” huller had expanded in numbers – from about 34,000 in the early 1960s to over 100,000 in the early 1980s16. The huller replaced female labour by non-renewable energy priced at levels which reduced processing costs, and increased labour productivity. It thus restricted demand for female wage workers to dry paddy and attend to the mill (Table 1). The introduction of the modern rice mill – under state patronage and unable to compete against the huller without state subsidy – generated the second wave of labour displacement. Like the first, this was also disproportionately biased against women – as seen in Table 1. Small and doubtful gains in technical efficiency were made at the expense of the most vulnerable segments of society, who could least forego employment (Harriss 1976).
14“The market” initially resisted the original technological package, a rational response, but over the years has adopted the techniques of “modern rice milling” in stages and by component. This has resulted in economies of scale and accentuated the structural concentration of ownership of assets in marketing systems. These assets are male-controlled.
15By the start of the 21st century the India-wide distribution of hullers and modern mills was extremely uneven (and no doubt underestimated) – see Table 2. In the largest rice producing state, West Bengal, only 9% of mills were modern ones; in the southern rice-producing belt, the proportions of MRMs varied from 16% to 60%; in the rice exporting belt of the north west, MRMs accounted for 30% of mill numbers in Punjab and 55% in Haryana; while in Bihar they amounted to 1%. The regional distribution of technologies matters because of its consequences for female livelihoods. While to my knowledge there is no rigorous research explaining this distribution, the factors that might need taking into account include: the agrarian structure, which generates the marketed surpluses supplying MRMs, shapes demand for custom milling for subsistence and even shapes residual demand for pounded rice; the location of final demand (in the northwest, rice tends to be an “inedible” cash crop which is bulked for relatively large scale processing and export to other regions); the extent of state subsidies; and the regulation of the relation between MRMs and the public distribution system.
- 17 Labour regards this as a tax because so few are able to redeem it at 60 if they live that long.
16In West Bengal, rice mills currently process a rough estimate of 20-25% of production, about half of which is sold to the public distribution system. Changes to the labour process in rice mills are differentiating this economically powerful sub sector of MRMs. The ratio of family labour to other kinds is increasing. Even as the tendency to professionalize the tasks formerly carried out by family members added to the wage labour, the displacement of wage workers is reducing the total wage labour force to an even greater extent. The absolute numbers of regular, salaried or “permanent” labour declined by 50% in the period 1990–2002. By the early 21st century, some 70% of labour worked on casual contracts, only half of which were thought to be registered on muster rolls, and therefore eligible for provident funds or employees State insurance17. Unions were unable to claim much by way of labour rights under the framework of the Labour Laws. Instead, they negotiated directly with bosses about contingent rights at work – and organised resistance to incremental lay-offs. Their limited set of insecure rights varied from region to region, as labour supply is regulated by ethnicity, and labour compliance through the use of physically oppressive technology which ensures submissive behaviour.
17The latest technological components, husk-fired mechanical driers (HFMD), substituting for the free public good of sunshine, reduce the pre-milling processing from 3-5 days to 24 hours and increase the milling season from 250 days to the entire working year. While enabling the harvest to be de-linked from the cycles of the monsoons and helping supply to keep pace with the appetite of the mill, the weather-proofing technology also satisfies the need to work at high capacity utilisation to cover the increased fixed costs component of more highly capital-intensive machinery. MRMs are still eligible for subsidised loans for technical upgrades. HFMD have had a massive impact on the labour process (see Table 3), the burden of displacement being borne entirely by female casual labour. This on-going displacement also pre-empts the unionisation of female labour. Even though wages paid to casual labour are gender-neutral, the system is increasingly male.
18Women are still found working without wages for the homestead-based pre-milling processing of custom-milled rice for subsistence. Women work for wages on the same physical process in the slightly scaled up micro-conglomerates of capital that are forming in and around the huller mills. Hullers have spread like wildfire after having been deregulated in 1996. They are now thought to process 80% of the rice harvest (Harriss-White 2008).
- 18 Rice mills in Punjab mill basmati rice exclusively for export while some 1,900 are agents for FCI, (...)
19By contrast, only 2 to 3% of production goes through custom mills in Punjab, the rest passing through “modern rice mills”. Most of these operate on contract to a range of state trading agencies (see Table 4)18. Since the 1960s, Punjab has been a significant net contributor to the public distribution system. The system of rice mills is stratified by caste and by regulative framework: a minority is owned and managed by cooperatives, most basmati rice mills is owned by Hindu business castes, while a minority of contract mills has been set up using Sikh agrarian capital. The labour process has developed as almost entirely male. It shows extremes of differentiation (Table 5). In the basmati export sector, annual returns in 2002 vary from Rs 4 to 11,2 crores per business family, down to Rs 17,000 for male rice mill labour – a factor of roughly 1,000: 1. By far the biggest component of the labour force (86%) consists of casual, right-less male wage labour, who are not entitled to bonuses, provident fund, employees state insurance or Diwali perks. Wages are agreed through mutual collusion in trade associations and without any consultation with labour. The Minimum Wage (Rs 85) is paid – but on piece rate for a day of about 12 hours rather than the 8 stipulated in law, so that in practice, the daily wage is two thirds of the minimum. Almost all such labour is from the landless agricultural labouring class, from backward castes, scheduled castes and tribes who migrate for work from Bihar and Eastern Uttar Pradesh. Organised through kinship and locality, and recruited in groups via contractors, the labour force is disciplined through a combination of on-site residence and debt. The continuum of work status required by rice mills (Tables 4 and 5) thinly disguises the fact that the vast majority of the labour force is working on informal and insecure terms and conditions – with no rights to work, at work or to social security. Women seem to be invisible in Punjab: they are absent from the accounts of milling costs, and stay in the villages from which migrant mill workers originate. They are, however, key not only to the inter-generational reproduction of this labour but also to its day to day maintenance “in time of need”, when workers are incapacitated from work by sickness or accidents, income plunges and they are forced home.
- 19 A revolutionary component should increase output per unit of input and thus lower total costs per u (...)
20To sum up: Dasgupta’s prediction is not borne out by the Indian trajectories of technical change. These four comparative analyses of the dynamics of rice processing show that massive technological change and process innovation has taken place not in the male activities but in the female activities of “grinding grain” and “preparing food”. Furthermore, against North’s hypothesis, masculinisation is not a rational response to relative prices. The newest technical components are only cost-effective at market prices under conditions of high capacity utilisation, conditions which are difficult to achieve without logistical and price support (subsidies) from the state. Yet, inexorably, they are being adopted in the very reverse of a green-revolutionary development19. North, acknowledging workers’ agency, has also hypothesised that technology is generally adopted so as to maximise the use of less skilled workers who do not have the bargaining power to disrupt production (1992, 65). The case studies, however, show that technology and modes of organisation are transferred which do away with precisely those people least capable of bargaining or of withholding labour – female labour. These gendered changes are hard to explain using the toolkit of institutional economics. The expulsion of armies of low caste women from the rice markets may be more a status-reducing expression of the contaminating caste relations of a merchant than it is a status-enhancing expression of patriarchy.
21While the post-harvest activities of north-western regions, of strategic importance to the state-regulated food security system, have always been highly masculine, the question remains: what work do the women whose livelihoods are displaced find?
- 20 The precise gendered impact will depend on the tasks mechanised, the prior gender division of labou (...)
- 21 There are strong barriers to female entry into the “non farm rural economy” – whether it is mining (...)
22In the first case considered here, dheki operating women and their households faced a major food security crisis – one no less serious for being dispersed in time and space. The dheki labour displaced in Bangladesh in the late 1970s and 1980s found exiguous work making quilts, grinding tobacco or tending poultry (Greeley 1987). In the complete absence of social security, so many took to begging that the state had to expand food-for-work infrastructure schemes – at times with foreign food aid. The rice mill labour being displaced at present in West Bengal seeks refuge in employment in brick kilns, in construction and in the residual shock absorber – agriculture. There is now overwhelming evidence that South Asian agriculture is being feminised (Kapadia and Lerche 1999). Against the early predictions about the femalelabour-displacing impact of mechanisation in paddy cultivation (Mencher 1985), this earlier trend is being reversed and it is male labour that tends to be displaced in agriculture20. This widely noted feminisation has been attributed to the male labour displacing impact of mechanisation in lift irrigation, ploughing and harvesting; to male withdrawal from joint tasks performed by both genders; to increasing local off-farm income-earning opportunities for men21; and to the tendency for men to migrate temporarily in search of work (women often being prevented from doing this by child care and other genderinelastic, domestic work).
- 22 One of the hypotheses considered for the period of the 1930s to the 1960s by Gita Sen (1983).
23From the argument of this essay, it appears that is not simply that male off-farm livelihoods are pulling men from agriculture, but that the displacement of women from female off-farm livelihoods is pushing them back to agriculture and helping to depress female wages there22. Most definitely marginalised from rice processing, they are not necessarily and always marginalised from productive activity; but their incorporation is on adverse terms.
- 23 The jury is still out over the debates on whether lower pay for women than for men reflects product (...)
24This research confirms that men are consolidating their hold over the economic power points in rice markets, and that women in elite business families are egregious examples of productive deprivation. Boserup predicted a widening gender gap in productivity, which is confirmed in this study of sectoral and regional unevenness. So is another gap she did not examine: a greater range in the returns available to male livelihoods in rice markets – in Punjab up to 1,000:1. However, while Boserup’s theory aspired to cover an entire society, the case material shows that women still provide much more than “artisanal… production and a variety of services” (1989, 192). In regions of southern India, they still provide the bulk of the wage work force. Though the regions of the North West have always had a masculine labour force, elsewhere, each wave of incremental technical change has led to the net displacement of labour, but has been biased hugely against casual female labour. Whether or not this displacement destroys livelihoods or “merely” reduces drudgery does not only depend on the degree of commercialisation of the product. It also depends on the extent to which the labour relations of post-harvest processing have themselves become commercialised. Where it destroys livelihoods in rice processing, women are pushed out into work on casual contracts and usually at under-nourishing rates of pay23.
25The state aids and abets this process. It uses the carrot of incentives, in the form of subsidies for technological upgrading, which concentrate assets and differentiate the labour process; it uses the stick of disciplinary regulation, outlawing an appropriate milling technology until recently; and it creates a nexus of rents which sustains a structure of accumulation in family businesses in which the rights of labour are avoided and evaded as assiduously as are tax obligations.
26However, the social realist economist Tony Lawson has made the case that we cannot ignore, that there is one changing reality with many theories of it, each of which are not only part of this reality but partial in their scope (Lawson 2003). Ester Boserup’s argument is steeped in modernisation theory. Its conceptual currency is the deepening of the division of labour, the advent of factory production and the definition of a rural sector different from that of towns. It fails to examine the specifically capitalist logic of this process. This has now been innovatively theorised in a general way by Ursula Huws (2004). In the second part of this essay, her theory of gendered commodification under capitalist production relations is developed in an application to Indian rice markets. It is another way of thinking about gender.
27“It is a law, based on the very nature of manufacture, that the transformation of the social means of production and subsistence must keep extending.” (Marx 1999, 222)
28In the second part of this essay, we examine the gendering of the process of commodification of rice in India, building on the above exploration of the changes in gender relations, which occur while rice is commercialised and while the technologies of processing are “modernised”. To help us do this, we develop the theory of capitalist commodification offered by the feminist sociologist of labour, Ursula Huws. Capital does not only “keep extending” according to a logic of expanded reproduction – which requires a continual increase in consumption, investment and competition with labour-displacing technology, leading to both the concentration and centralisation of capital. Huws suggests that capital also “keeps extending” by non-stop commodification, including that of services.
29The process of commodification involves a sequence of relations, the first few of which are familiar from Boserup’s account. First, un-valorised productive and reproductive tasks are carried out for their essential usefulness – their use value. They are then replaced by craft work for sale and/or by paid services. In turn, these are replaced by mass produced commodities, in conditions with economies of scale. In the process, new mechanisms of control “Taylorise” the production of these goods and workers are gradually de-skilled. The new mass commodities in turn require new services; these are also industrialised, involving further de-skilling alongside the managerialisation of professional services. Each wave of commodification is accompanied by new technology. This new technology is not immiserising or labour-displacing to society as a whole. Indeed, the creation of new commodities – including commodified forms of services – requires labour, for example advertising and marketing before purchase, insurance, repair and servicing after purchase. This demand continually stimulates the labour market. As a consequence of this process, the home is transformed from a site of production of use value to a site of the consumption of commodities (Huws 2004, 18). But not quite – and not only.
- 24 Huws gives examples of collecting goods from supermarket shelves; bagging your own vegetables; self (...)
- 25 Leisure, which used to be spent creatively – such as the enjoyment of active music-making – has bee (...)
30First, since capital strips itself of all unprofitable tasks, new forms of work (unvalorised tasks/work for use value) are loaded onto the consumer24. Huws adopts the term consumption work for this activity and argues convincingly, after Gershuny, that this process of industrialisation of domestic work does not increase leisure time25. Since the wage is no longer a family wage, women have to work in order to pay for the ever wider range of reproductive technologies, all in commodity form, which replace production for use (including pleasure). Public goods and services, as well as profoundly influential ideologies of cleanliness, pleasure, and of women as providers of domestic peace and care, all make it difficult for people not to possess these “reproductive commodities” (grinding machines, fridges, washing machines, etc). Because the home is a special site where these commodities are owned, either individually or by the corporate household, there will be a structural excess capacity in them nationwide. As the division of labour and the process of reproductive commodification both deepen, consumers’ knowledge about the commodities they use decreases, and the commodities themselves have the potential to become increasingly dangerous (Huws 2004, 44-45). Of course, from this transformed consumption work come requirements for yet more specialised services, which themselves become commodified – for example, the semi-automated systems for the repair and maintenance of domestic appliances organised for European consumers via call centres in India – and so the cycle of commodification and transformed consumption work continues.
31Second, since capital always searches for the cheapest, most docile labour, the home becomes a new site for wage work and for the production of commodities (2004, 33). Information technology enables the control of out-sourcing and creates new forms of alienation, under which the worker may own the means of production – the machine – while the terms of work are still set by the employer.
- 26 There is a third, though Huws does not explicitly develop this. The speed and comprehensivity of th (...)
32There are at least two consequences that are relevant to the argument about gender and rice processing in South Asia that has been developed in the first part of this essay. First the new technology displaces jobs disproportionately allocated to women (in Huws’ European case studies, these are clerical jobs). It leads to the degradation of work, due to the reductive nature of commodity production: physically repetitive, mentally restricting and constraining social relations. It leads to conditions in which it is very difficult to organise labour – in that sense it is “de-socialised”. It also leads to job losses, as outsourcing is spatially and socially relocated. Second, this isolated wage work has implications for the physical and mental health of (women) workers, who are stuck in precisely the same process that requires them to be ever more central to the provision of domestic harmony. Huws calls this the Taylorisation of domestic life (2004, chapter 11). The increasingly frequent breakdown of the family unit in so-called “advanced” (but actually unsustainable) capitalist societies exacerbates the condition of women; and these conditions persuade a small but growing number of women and men not to form families at all26.
33While the theoretical focus of Huws’ analysis is the working household under advanced capitalism, Huws herself makes global connections between two elements. On the one hand are the twin processes of commodification and accumulation and the search for cheap labour; and on the other, the outsourcing of production and the transfer of (cyber-) work to women of different classes (as in highly educated call centre labour in India), at different stages of life (as with unmarried women in Bangladesh and China), and/or in conditions where rights at work and rights to social security are absent (as in sweatshops everywhere).
34In certain respects, Huws’ theory resembles Boserup’s theory, but their predictions are different. Both are concerned with female agency in the context of threatening processes, and both describe the process of industrialisation of craft production. However, while Boserup argues that the productive labour involved is progressively disproportionately male, Huws reasons that the waves of commodification use female wage workers not only in factory production but also sited at home. For Boserup, if women in agrarian societies are educated, their education does not result in mass entry into the labour force, quite the reverse. Women are forced progressively into the condition of “productive deprivation”, and this creates “domestic tension”. Meanwhile, Huws not only acknowledges such tension in principle, but deconstructs the social tensions resulting from the transformation of the purposes of the home – first from the site of production-for-use to the site of consumption, and then in addition to the site of women’s wage work.
35Do their predictions differ because developing countries (Boserup) are different from advanced ones (Huws)? Because Boserup wrote in 1972 and Huws in the 21st century? Or because Boserup analyses neither capitalism nor commodification? Rice is one of the key nutrient bases for the global system of commodification theorised by Huws. What is happening to the rice sector in India and what light does it shed on these questions?
- 27 A family’s credit in business “is its stock in the broadest sense, which includes social position, (...)
36Let us try to extend Huws’ insights to conditions of mass poverty in the labour force, in developing countries such as India. As was discussed earlier, women occupy a range of class positions. The structure of the difference between gender and class and its implications for labour market supply have been theorised – for North Indian conditions – by Kalpana Bardhan (1993). To summarise, in the capitalist class, women are first defined through kinship norms and practices of village exogamy and caste-hypergamy. These deprive women of regular support from their natal families. Work in the family business is sex-segmented and women are secluded, such that their income and price elasticity of supply is zero. However, their compliant and pious behaviour affects the business family’s commercial reputation27. This in turn shapes returns from the real economy, for example through access to credit for working capital and access to investment capital through resource flows in marriage alliances. While women are not exploited, health and demographic analysis shows only too well that they may be oppressed.
37Bardhan then turns to consider the “working peasantry”, where-women may organise the hiring-in and out of labour and may supervise production. Their status within a household will affect their supply to the labour market. In labouring households, uxorilocal marriage means a woman has access to support from her natal family but she, like her husband, is exploited by capitalist men through wage work. Of all the class sites discussed here, her elasticity of supply is highest. The working woman, by reproducing wage labour across the generations, also supplies a “subsidy” (even if it is one which cannot be valued) to the capitalist. Exploited in her own wage work, perhaps as a domestic servant or agricultural labourer, she is also commonly vulnerable to sexual oppression from capitalist men. Hence a woman’s experience of oppression is primarily from the capitalist class and only secondarily from relations within the working class household.
- 28 Along with the peasantry and religion, the “opium of the masses” (Marx 1999, 595).
38Clearly then, in Indian conditions, the structures of exploitation and oppression which are mapped onto labour supply prevent poor working people, especially poor working women, from buying income elastic reproductive commodities such as electric grinders and fridges. And while poor working women do not have individual purchasing power for the so-deceptive “labour saving” devices central to Huws’ theory, some bourgeois women have had those labour-saving devices purchased for domestic use – and in some cases, for display and status rather than for use. So some women from the capitalist and professional ancillary classes in developing countries (possibly those reading this essay) are available for capital and the state, go to work and are captured in – and by – exactly the process that Huws theorises for working women in advanced countries. But the majority of bourgeois women who are unemployed “home-makers” (with or without labour saving devices) continue to transform consumption work into petty services, paid for by the surplus value which is extracted by their men through the family business. The home is then a site not only of production for use/production for exchange by men, but also of demand for, and consumption of, reproductive services (superficially expressed by women in the division of domestic bourgeois labour, but using money transferred by their male kin). This demand is often, but not exclusively, for female domestic wage labour. In labour-surplus economies, petty production and petty services are then built solidly into the structure of commodification, even when they face direct competition from commodity and factory production – exactly the conditions which have been widely predicted to destroy petty production28.
39This might be just as well: the multipliers from waves of Huws-style commodification involve technological transfers from economies with different “factor proportions” to those of societies with labour surpluses. The extent to which the process of capitalist accumulation (triggered by commodification) generates the kind of jobless growth, much discussed in the Indian literature on trade liberalisation, sets further distinctive limits on the Huws-style process of commodification.
- 29 This is one of the explanations for the period of industrial stagnation between 1965 and 1980 – see (...)
40In sum: in developing countries, the irony is that poor labour-market returns constrain, though they do not prevent, both a Boserup-style process of productive deprivation and a Huws-style process of waves of commodification, broadly centred on the home29. Instead, a remarkably stable structure of demand is created for services and basic “wage goods” either produced or supplied under conditions of petty production. Petty production accounts for a higher percentage of GDP than does the corporate sector; it is a form of production every bit as modern as the corporation; it is the most common form of production and it has proliferated despite the predictions to the contrary of all the major social theorists of modernity (Harriss-White 2010a).
- 30 See Rogaly et al. (1999) for the debate over growth rates.
41Let us put the most basic wage good under the lens of commodification. In the absence of other research, evidence from my own research in West Bengal in 1982, 1990 and 2000-2002 will be used as illustration. Figure 1 shows the result of over two decades of institutional involution in the systems of local petty trade centred on huller (husking) mills and long distance (state mediated) trade through the rice mills. During this period, growth rates for rice production doubled for about a decade and then declined30. West Bengal went from being a rice deficit state to self sufficiency in 1993, and surplus thereafter.
42Figure 1 shows that large numbers of new livelihoods have appeared during the past decade or so: new village-level sub-agents for paddy brokers, new petty parboiling firms, new petty milling operations, new petty post-milling rice-puffing firms and new itinerant traders of rice. Over the last 15 years, family firms have started to employ labour at a number of sites, in freight and haulage, huller/husking mills and newly decentralised wholesale-cum-retail firms. Further, new rental markets have appeared where the owners of huller mills have invested in mini drying yards and parboiling machinery. Third, the process of commodification of products also grinds ineluctably onwards. Rice is differentiated, not only through varieties (which are still diverse, though showing signs of a reduction in the range of cultivars), but also through types of pre-and post-milling processing for raw rice, several kinds of parboiled rice, puffed dry rice (muri) and rice flour ground from the 2% of rice which emerges from milling in broken forms. Fourth, even waste is commodified for recycling (Beall 1997); even the small stones that are used as adulterants have been commercialised for re-adulteration.
- 31 Husk in rural huller milling is not yet commodified, as it is retained for domestic parboiling and (...)
- 32 Most bran from huller mills however, is lost to the agro-industrial system and is retained in villa (...)
43By-products are being commodified. One, rice mill husk, is no longer traded as an organic manure but is now recycled as a fuel for steam exchangers in rice mills. New specialist markets have developed for any surplus husk31. The by-product of the commodified by-product (burnt rice mill husk) is also being commodified as a raw material for both the high tech silica industry and for low-tech brick kilns. Specialised firms are developing to supply each kind of industry. Another by-product, rice mill bran, is the basis of a relatively highly capitalised and large-scale oil extracting industry. This commodity, oil, has only recently been declared fit for human consumption: it was previously an intermediate good for the production of detergents and paints. The by-product of this by-product, de-oiled bran cake, even finds its way to Europe as commercial cattle feed32.
44So rice now enters the human food system to be consumed as rice, cooking oil and meat.
45Obsolete technology is not rejected but instead feeds second-hand markets, well evinced in the thriving market for majestic cast–off steam engines (some from the Indian Railways), which are converted for husk-firing.
- 33 Instead, due to poor infrastructure, both physical and telecommunications, it remains internalised (...)
46The process of commodification is not yet as multiplicative of derived specialised professional services such as advertising, insurance, legal and accounting advice, as it is in advanced capitalist economies. The repair and maintenance of mill machinery is also not yet being commodified as a specialist service sector, but is carried out by permanent in-house labour33.
47As Karl Polanyi argued (1944), markets respond to demand rather than need and are inappropriate – even socially dangerous – mechanisms to match human needs. Societies are therefore under social and political pressure not to commodify, or under counter-moves from society to de-commodify, the provisioning of essential needs. In India, this de-commodification took the form of the Essential Commodities Act of 1955, which includes, as one of its many instruments, the Public distribution system of foodgrains. Through a state-administered system, running parallel to its grain markets, the state protects and subsidises the costs of reproduction of that part of the labour force living below a stingy poverty line.
- 34 State interventions such as the Public distribution system or Noon meal schemes have been found to (...)
48However, in contemporary “peripheral capitalist societies”, just as elsewhere, the sphere of public goods and services is being privatised and commodified. This is only possible because public goods and services are not confined to non-rivalrous and non-excludable goods and services, but include those about which there has been a justified public consensus that they be provided by the public sector for non-commercial reasons. The process has been theorised by Leys (2001). Public services are reconfigured as standard commodities; service providers are reduced to wage labour, from which surplus can be extracted; the expression of public service values has to be converted into demand for commodities and the state acts to underwrite the risk to capital of this transformation. Under this process, dominated by big business, and, thanks to WTO rules, increasingly by multinational corporations, labour in the state sector is being displaced throughout the capitalist world system. But this process is always constrained by the limited proportion of society that can afford “open” market prices for commodified public goods, or “user fees” for privatised services. The privatisation of public goods and services always has regressive outcomes34.
- 35 BPL means below the poverty line.
49In the Indian rice system, the de-commodified public distribution system is under attack from a process of commodification that takes many forms (well addressed in Swaminathan 2000). The most obvious is the planned privatisation of publicly owned storage and the reduction of food reserves. Another form of attack is deliberately planned price convergence to the open market, as when the price of administered food-grains prices for all but BPL households35 becomes asymptotic to market prices. Yet another attack consists of barriers to off-take by poor households, exemplified by the large size of the limiting consignments which poor households may purchase on subsidy – too large for casual day-labourers to afford.
- 36 See Mooij (1999), for detailed evidence for Karnataka and Kerala.
50When the state abnegates its responsibilities and does not enforce its own laws, markets substitute for the defective or absent supply of public goods and services. Sometimes they take petty forms, for example the night watch when property is insecure, or the private disposal of industrial waste when public hygiene is poorly provided. But they may also be organised on a large scale, for example the production and sale of diesel engines to substitute for fickle state electricity. The process of private substitution for state failure carries with it no necessary implications for scale. Where the state does not implement its own regulatory interventions, institutions of collective regulation fill the normative vacuum and create authority for market order. To the extent that this authority requires resources and skilled and specialised labour, the parallel system of regulation then follows the inexorable logic of commodification. Commodification also proceeds apace in the informalised and illegal markets for public goods and services. Conventionally theorised in terms of rent-seeking, leakages from fraud and plunder far exceed those due to corruption (Roy 1996). Tax-liable resources, fraudulently retained, will further lubricate the process of commodification. For example, even in the “liberalising” post-harvest rice system, the most common and enduring of such markets are those for grain “leaked” from the public distribution system, in transport or from storage; grain that was never exported as planned; or grain that seeped into parallel markets from the differences between official milling outturn ratios and higher actual ones36. However, by far the more significant phenomena are the criminal markets, developed for capital flight and black finance (Srinivasan 2004). Though Arun Kumar argues that such phenomena are concentrated at the top of the Indian economy, there is other persuasive evidence that it is widespread and decentralised (Kumar 1999; Roy 1996; Khan and Jomo 2001; Harriss-White 2003). So the re-commodification of the decommodified state-administered distribution of rice is proceeding apace by informal means as well as by formal ones.
- 37 Mills are polluted and drying yards involve back breaking work in hot sunshine.
51In the post-harvest rice system, it is only the homes of the propertied commercial elite and elite consumers of rice which are turned into a site of consumption of Huws-style “reproductive commodities”. The systemic commodification we have described here is not sited in the home, but rather in workshops, where the work is scaled-up and more specialised than bari production, or in mills and factories where skills are differentiated. While work conditions are harsh and while labour is certainly next to impossible to organise37, in neither case to date is work “de-socialised” – as it is depicted in Huws-style commodification – because in terms of political organisation, it was never socialised in the first place.
- 38 Old age is defined by rural people in terms of the needs generated by incapacity and disability; an (...)
52The impact of commodification is well established: a heavy double burden of productive and reproductive work, with deleterious implications not so much for women’s mental health (due to isolation, Huws had theorised) as for their physical health (Elson 1995; Jackson and Palmer Jones 1998, Bhushan, 2003). Even here, there is an additional burden, for women not only endure their own ill health, they are also the prime providers of care at times of disability or ill health of other family members – including migrant workers forced to return with burdens of occupation-related accidents and illness (Erb and Harriss-White 2002; Rogaly et al. 2002). They are the ones who perform the “help in time of need” that the state provides to public sector workers, and that public-private insurance provides to the propertied elites (Harriss-White 2010b)38.
53From our limited evidence for West Bengal and North India, it seems to be mostly young men of low caste who benefit, through wage work and self-employment, from the proliferation of livelihoods (producing and trading of new products, by-products and informalised products) in the involuting rice system. The system is financed through a cascade of money advances from commercial capital. Where family labour is used, women are, for the most part, reported as being under the control of men. It is only when men migrate for work that women emerge as managers of the household’s residual productive assets (Coppard 2004; Mosse 2004; Picherit 2009; Breman, Guerin and Prakash 2009). Women retain prominent roles in petty retailing. They lack collateral, but also access to banks, and thus remain dependent clients of male rice wholesalers, and their capacity to accumulate independently is seriously constrained (Choudhury 2001). As a result of the polarised accumulation trajectories created by commodification, de-oiled bran cake, fed to European cattle, and basmati rice wholesaled in Europe can be confected into a ready made meal for consumers in relatively food-secure Europe, while the Orissan or Bihari woman, who supports the family of the migrant wage-worker in the bran factory or rice mill, is one of the most enduringly vulnerable to food insecurity on the planet.
54Meanwhile new technology may be gendered in ways which have perverse effects. The next stage of technical change in agro-commerce, automation, urgently requires computerisation. Until very recently, in small-town South India, local women who have obtained relevant qualifications have been barred, because of their gender, from computer operation on commercial premises. The agro-commercial elite has rejected the female gendering of high-tech work, located socially and physically above their male (family) labour force. The diffusion of automation has therefore been hindered, and computerisation has been confined to the kind of simple accounting operations, which can be learned by trial and error by under-educated and uncommodified male family labour.
- 39 Three forces can be seen to drive this process: the differentiation of demand (carefully constructe (...)
55In sum: Just as we summarised in Lawson’s argument at the end of the second section of this paper, Indian rice markets reveal both Boserup’s and Huws’ theories as being partial. In both cases, what they exclude limits their social relevance and their predictions. Boserup conceptualised neither commodification nor capitalism, while Huws has not yet focused on gendered commodification under conditions of low returns to labour. Huws’ focus on the repeated waves of commodification centred around the home enables her to examine its far reaching impact on the household, the changes it creates in women’s work, and the different conditions which “make a cybertariat” in centres of advanced capitalism and in peripheral capitalist societies like India. Despite these limits, her theory is rich enough to be developed in several new directions. In particular, from the material discussed here, a link can be established between commodification and the combination of petty commodity production and commercial capital, which is a persistent and highly gendered feature of the capitalist accumulation through which the process unfolds in India39.
56The object of this essay was not to examine commodification across all reproductive commodities in India but in one of the most basic of all: rice. The evidence we have marshalled shows that the combination of masculinisation and productive deprivation is confined to the one class where the reproductive commodities, which both generate Huws-type commodification and result from it, are being amassed. Economically powerful and sizable in absolute terms, this class is relatively small in comparison with the total population. But it is this class which also generates demand for petty services and petty commodity production, much of which is women’s work. As mill-owners and wholesalers, this class is deeply engaged in a dynamic, state-subsidised process of upgrading using technology originally transferred from a different factor environment. This displaces exactly the kind of labour that ought, in theory, to be maintained: female labour, the cheapest and most compliant of all.
57Rampant commodification can be observed throughout the post-harvest system for rice, even under conditions of mediocre growth in production, involving: investment goods and technologies (stores), intermediate goods (machinery), labour, products, by-products, waste, adulterants, informalised public goods and regulative resources. The forms taken by this commodification are highly varied. New forms of advanced capitalist organisation coexist stably with older forms, along-side an obstinately persisting production for direct use. A range of technologies, of laws (of selectively limited reach) and informalised procedures, of social norms and institutions coexist to regulate this process. These are the characteristics of a market economy which is socially as well as state-regulated.
58In this process, commodities show all signs of becoming standardised. However, they are far from always being the mass products of factories (although steam and diesel engines, rice mill machinery and some rice certainly are mass-produced under factory conditions). But in South Asia as elsewhere, the polar classes of capitalism coexist with a more or less independent stratum of petty commodity production. The outcome, in many cases, is that a commodified good, produced at home using family labour, competes with an identical one produced in a factory using wage labour. What is more, a given task can be performed by wage labour with varying rights at work and rights to social security, under a range of contracts and wages. In a socially regulated economy, the segmentation of markets can be such that competition between these forms of labour process may not take place at all. If it does, it requires market preconditions where scale economies in factory production compete directly with unwaged labour in petty production.
- 40 In the regulation of wage work, non-local (migrant) men of low social status occupy similar positio (...)
59Gender is a social structure crucial to the viability of the space for petty production, since the services that are the outcome of women’s agency, under exploitive and oppressive conditions, both underwrite and constrain bourgeois commodification. It is women’s unwaged work that is pitted against the economies of scale reaped by capitalist enterprise employing wage workers40.
- 41 The phrase is by Leys (2001).
- 42 See Chibber (2003) for big family business and corporate capital and Harriss-White (2003), for inte (...)
60The recognition of the gendering of petty production carries with it some conundrums for policy. The state shackles petty production by subsidizing the mass production of reproductive commodities such as rice. It gives petty production incentives by not enforcing minimum wages laws and by allowing what many labour economists call “disguised wage labour” to bring in lower returns to work than those to wage work itself. On the one hand, if working women were to demand an end to state-subsidies for the production of “reproductive commodities”, the competitive advantage of petty production would be boosted. On the other hand, the moment women and children in family enterprises were to start to demand minimum wages, petty production (and male household heads) would be thrown into crisis. One can safely predict that every attempt will be made to prevent either claim from happening. The state may have an interest in preventing the former, since the actually existing state has long been embroiled in a “market driven politics”41 in which capital seeks to seize incentives, such as subsidies, and to resist all disciplinary regulation42. It may have an interest in preventing the latter since, in an era of zero or negative elasticities of labour absorption, in registered industry as well as in unregistered agriculture, the petty informal sector is increasingly crucial to the state’s political legitimacy. Bereft of a coherent development project for petty production, the state (re)labels it as “the tiny sector”, “self employment”, “micro-enterprise”, “micro-credit”, “livelihoods”, or “self-help groups” (Harriss-White 2010a). Although interventions proliferate, by taking this ad hoc approach, the state’s regulative and welfare responsibilities towards labour can be reduced to highly selective patronage, and its infrastructural responsibilities toward business and capital can be minimised.
61My final comment concerns the theoretical situation of gender relations. While Huws has focused on what commodification does to women (and to gender relations) in the sphere of consumption and at work “at home” and around the home, gender is one of a set of non-class, non-State means whereby capitalism is instituted or structured. Other prominent structures are ethnicity, caste, religion, nation, locality, language, age and the life cycle. Feminist scholarship has been very successful in mapping capitalism, and now commodification, onto gender and vice versa.
- 43 See also Harriss-White 2003 for a general attempt for the Indian economy.
62Gender actually intersects with these other social institutions to regulate markets and to structure accumulation. In this essay, an attempt has been made to explore the intersection between rice processing, gender, class and the state. Laurence Pujo (1997) has started to theorise the intersection of gender with other institutions, using her experience with the gendered rice markets of Guinee43. Locality and ethnicity vary in their significance and regulate the economy in different ways at different scales of aggregation. With respect to locality, information and access to capital is heavily urban and male-biased. Ethnicity is a strong entry barrier in business, particularly to women – even in a marketing system which Boserup deemed to be “female”. Little is known about the ways in which divine authority and the practices attributed to religious groups (including “othering”) regulate the process of commodification, gender relations and the control of women – and the sites of production, distribution, consumption/demand and reproduction. Pujo observes that allegiance to religion structures network transactions and may also lead to contractual behaviour which differs among coreligionists and others. While kinship alliances affect entry into marketing systems, less is known about the role of the life cycle in market behaviour. Pujo found that women gain independence with age and can accumulate at a faster pace; however the great majority of women stop trading altogether when their dependents leave the households (Pujo 1997, 291-301).
63With the understanding of feminists, gender can be seen as a special case of the general phenomenon of the non-class, non-state power structures through which commodification and class relations are constructed. If that is the case, then how do we theorise the sites and the spaces in which the other institutions work, as Huws and other feminist theorists have done with gender and the spheres of consumption and production? How to relate the emancipatory and oppressive practices expressed through all these major social institutions? How to establish their roles in market exchange? Until this work is done, it is impossible to develop a general understanding of the social regulation of the economy. Nor, with few exceptions, have these institutions or structures been related systematically to the practices of the state and the arenas and processes of policy (Fernandez 2008). Formulated in terms neutral to these structures, unless it purposively intends to transform them, in practice policy works through them. It is due both to the ethnocentricity and to the reductive nature of economics that these theoretical, practical and political questions are neglected. Gender has been a most notable exception.
Table 1: The gendered production of rice, 1980s, S. India
Livelihoods per annual output of 81.60 tonnes of rice
Huller mill 0.2-0.8 tonnes per hr
Modern rice mill 2-3 tonnes per hour
Source: Harriss and Kelly 1982
Table 2: Percentage of distribution of rice mills as on 01.01.2001
@ Number insignificant
Source: Report of the Ministry of Food Processing Industry 2002; Kaur 2004
Table 3: Impact of the mechanical drying, W. Bengal
Livelihoods per Rs 1 crore of gross output (2002)
Permanent labour (M)
Casual labour (M)
Casual labour (F)
Processing Time (hrs)
Source: data in Ghosh and Sudarshan 2004
Table 4: Employment structure in rice processing (Punjab)
Per cent of the total
Regular family (management)
Regular (but not permanent) employees:
a. With most service benefits
b. With only PF facility
Seasonal employees (without any benefits)
Unskilled daily wage workers
Sub-total (Formal Sector Workers)
Seasonal family workers in informal sector units
Source: Survey of Rice Processing Industry, Kaur 2004
Table 5: Male livelihoods in rice mills in Punjab 2002 per unit
Source: data in Kaur 2004.
64The milling of rice involves removing the outer layer of husk and the intermediate layer of bran. The milling outturn varies from 60-69% depending on the pre-milling processing, the milling technology, the moisture of the paddy and the degree of polish required.
65The tasks associated with paddy milling vary widely. At one extreme is the custom mill, processing 0.2 tonnes per hour with one pass through the mill for the payment of a fee; in which the byproducts are retained by the owner of the consignment. At the other extreme is the modern rice mill (MRM). Paddy is passed through mechanical driers or sun-dried down to 12-14 % moisture. Dust, stones and other adulterants are removed, paddy is milled using a rubber roll sheller, rice is polished, graded and packaged, while husk and bran are separated.
66In Eastern India and regions supplying Eastern India, paddy may be par-boiled according to a range of techniques, prior to drying and milling. In general it is soaked prior either to steaming or boiling. Parboiling gelatinizes the starch, making the grain translucent, hard and resistant to breakages during milling. As a result milling recovery rates for head rice and total rice yields are improved. The rice kernel also absorbs oil, vitamins and minerals from the surrounding layer of bran, making it more nutritious.
67Paddy husk, a by-product in the process, is used as fuel for boiling and mechanical drying. Husk from raw rice mills is sold. Other by-products are rice bran, which is either returned as animal feed or commercialised as a raw material for solvent oil extraction, broken rice (which is sold or crushed or powdered prior to sale) and husk ash which is used as an organic fertiliser and a base for cosmetics. (see Kaur 2004)
Adnan, S., R. Islam and Village Study Group. 1976. Social change and rural women: possibilities of participation. BIDS Working Paper no 7. Dhaka.
Agnihotri, S. 2001. Sex ratio patterns in the Indian population. New Delhi: Sage.
Arens, J. and J. van Beurden. 1977. Jhagrapur: poor pesants and women in a village in Bangladesh. Birmingham: Third World Publications.
Athreya, V. 2001. Census 2001: some progress, some concern. Frontline. 26th April.
Bardhan, K. 1993. Gender and class: the structure of the difference. In Gender and political economy: explorations of South Asian systems. (ed.) A. Clark. New Delhi: Oxford University Press.
Beall, J. 1997. Households, livelihoods and the urban environment: sociological perspectives on solid waste management in Pakistan. PhD thesis. London School of Economics.
Benewick, R., M. Blecher and S. Cook (Eds.). 2003. Asian politics in development: essays in honour of Gordon White. Frank Cass: London.
Bhushan, B. 2003. Are mental health and social anxiety related to the status of working women? Farzaneh. 6(11): 66-71.
Boserup, E. 1989. Women’s role in economic development. Brookfield Vt.: Gower.
Breman, J. 2003. The labouring poor in India: patterns of exploitation, subordination and exclusion. New Delhi: Oxford University Press.
Breman, J., I. Guérin and A. Prakash. 2009. India’s unfree workforce. New Delhi: Oxford University Press.
Chibber, V. 2003. Locked in place. State building and late industrialisation in India. Princeton: Princeton University Press.
Choudhury, S. 2001. The public distribution system in India and informal traders: rice markets in Calcutta. M. Phil. thesis. Oxford University, Queen Elisabeth House.
Clark, A. (Ed.). 1993. Gender and political economy: explorations of South Asian systems. New Delhi: Oxford University Press.
Coppard, D. 2004. Labour and drought: a social history of agrarian relations in a Chottanagpur Village of eastern India. Paper to the conference on Livelihoods at the Margins, SOAS, London.
da Corta, L. and Venkateshwarlu, D. 1999. Unfree relations and the feminisation of agricultural labour in Andhra Pradesh, 1970-95. Journal of Peasant Studies. 26(2/3): 71-139.
Dasgupta, M. 1987. Selective discrimination against children in Punjab. Population and Development Review. 13.
Dasgupta, P. 1993. An inquiry into wellbeing and destitution. Oxford: Clarendon.
Elson, D. (Ed.). 1995. Male bias in the development process. Manchester: Manchester University Press.
Erb, S. and B. Harriss-White. 2002. Outcast from social welfare: adult disability and incapacity in rural South India. Bangalore: Books for Change.
Fernandez, B. 2008. Engendering policy in India. D Phil Thesis. Oxford University.
Ghosh, P. and R. Sudarshan. 2004. Rice milling in West Bengal. Paper to the NCAER/QEH conference on Trade Liberalisation and India’s Informal Sector. New Delhi: NCAER (IIC).
Greeley, M. 1987. Post harvest losses, technology and employment: the case of rice in Bangladesh. Boulder: Westview Press.
Harriss, B. 1976. Paddy processing in India and Sri Lanka: a review of the case for technological innovation. Tropical Science. XVIII(3): 161-186.
–––. 1979. Post harvest rice processing systems in rural Bangladesh: technology, economics and employment. Bangladesh Journal of Agricultural Economics. II(1): 23-50.
–––. 1993. Markets, society and the state: problems of marketing under conditions of smallholder agriculture in West Bengal. Development Policy and Practice. Working paper 26. Open University.
Harriss, B. and C. Kelly. 1982. Food processing: policy for rice and oil technology. IDS Bulletin. 13(3): 32-44.
Harriss-White, B. 1995. Efficiency and complexity: distributive margins and the profits of market enterprises. In Prices, products and people: analysing agricultural markets in developing countries. (Ed.) G. Scott. 301-324. Boulder: Lynne Rienner.
–––. 1996. A political economy of agricultural markets in South India: masters of the countryside. New Delhi: Sage.
–––. 1999. Agricultural growth and the structure and relations of agricultural markets in West Bengal. In Sonar Bangla agrarian change and agricultural growth in West Bengal and Bangladesh. (Eds.) B. Rogaly, B. Harriss-White and S. Bose. 381-412. New Delhi: Sage.
–––. 2003. India working: essays on economy and society. Cambridge: Cambridge University Press.
–––. 2004a. Labour, gender relations and the rural economy. In Rural India facing the 21st century. Chapter 1-5. B. Harriss-White and J. Janakarajan. London: Anthem.
–––. 2004b. Nutrition and its politics in Tamil Nadu. South Asia Research. 24(1): 51-71.
–––. 2008. Rural commercial capitalism: agricultural markets in West Bengal. New Delhi: Oxford University Press.
–––. 2010a. Capitalism and the common man: four decades of development in Africa and South Asia. Work in Progress Paper 2. CSASP, Oxford University.
–––. 2010b. Work and wellbeing in the informal economy: the regulative roles of institutions of identity and the state. World Development. 38(2): 170-183.
Harriss-White, B. and S. Janakarajan. 2004. Rural India facing the 21st century. London: Anthem.
Harriss-White, B. and P. Nillesen. 2004. Life chances: development and female disadvantage. In Rural India facing the 21st century. Chapter 3-3. B. Harriss-White and J. Janakarajan. London: Anthem.
Harriss-White, B. and G. White (Eds.). 1996. Liberalisation and the new corruption. Special Issue. IDS Bulletin. 27(2).
Hensman, R. 2001. Organising against the odds: women in India’s informal sector. In Working classes, global realities. Socialist Register 2001. (Eds.) L. Panitch and C. Leys. 248-259. London: Merlin Press.
Hodgson, G. 2001. Why economics forgot history. London: Routledge.
Hodgson, G. and T. Knudsen. 2004. The replication of habits. Paper to the international workshop on Research in institutional economics. University of Hertfordshire. http://www.herts.ac.uk/business/geoffrey-hodgson.ws
Huq, M.A. 1976. Exploitation and the rural poor. A working paper on the rural power structure in Bangladesh. Comilla: Bangladesh Academy for Rural Development.
Huws, U. 2004. The making of a cybertariat: virtual work in a real world. London: Merlin Press.
Jackson, C. 1999. Men’s work, masculinities and the gender division of labour. Journal of Development Studies. 36(1): 89-108.
Jackson, C. and R. Palmer-Jones. 1998. Work intensity, gender and wellbeing. UNRISD Discussion Paper 96. Geneva: United Nation Research Institute for Social Development.
Jackson, C. and R. Pearson. 1998. Feminist visions of development: gender analysis and policy. London: Routledge.
Jeffery, P. and A. Bose. 1998. Appropriating gender. Routledge: New York.
Jeffery, P. and R. Jeffery. 1998. Gender, community and the local state in Bijnor India. In Appropriating gender. (Eds.) P. Jeffery and A. Bose. Routledge: New York.
Kapadia, K. 1993. Mutuality and competition: female landless labour and wage rates in Tamil Nadu. Journal of Peasant Studies. 20(2): 296-316.
Kapadia, K. and J. Lerche. 1999. Introduction. Journal of Peasant Studies. 26(2/3): 1-9.
Kaplinsky, R. 2003. Is globalisation all it’s cracked up to be? In Asian politics in development: essays in honour of Gordon White. (Eds.) Benewick, R., M. Blecher and S. Cook. 255-272. Frank Cass: London.
Kaur, R. 2004. Rice processing industry: a case of Punjab. Paper to the NCAER/QEH conference on Trade Liberalisation and India’s Informal Sector. NCAER (IIC) New Delhi.
Khan, M. and K. Jomo. 2001. Rents, rent seeking and economic development in Asia. Cambridge: Cambridge University Press.
Kumar, A. 1999. The black economy in India. New Delhi: Penguin.
Laidlaw, J. 1995. Riches and renunciation. Religion, economy, and society among the Jains. Oxford: Clarendon Press.
Lawson, T. 2003. Theorising ontology. Feminist Economics. 9(1): 161-9. Leys, C. 2001. Market driven politics. London: Verso.
Marx, K. 1999. Capital: an abridged edition. Oxford: Oxford University Press.
McCartney, M. 2009. The political economy of liberalisation in India: 1991-2008. Routledge: London.
Meagher, K. 2010. Identity economics: social networks and the informal economy in Africa. James Currey: Oxford.
Mencher, J. P. 1985. Landless women agricultural labourers in India: some observations from Tamil Nadu, Kerala and West Bengal. In Women in rice farming. Proceedings of a conference on Women in rice farming systems. IRRI, Philippines, September 26-30.
Mencher, J. and K. Saradamoni. 1982. Muddy feet, dirty hands: rice production and female agricultural labour. Economic and Political Weekly Review of Agriculture. 17(52): 149-67.
Mooij, J. 1999. Food politics and policy in India. The public distribution system in South India. Delhi: Oxford University Press.
Mosse, D. 2004. Adivasi seasonal labour migrants in western India. Paper to the conference on Livelihoods at the Margins, SOAS, London.
North, D. 1992. Institutions, institutional change and economic performance. Cambridge: Cambridge University Press.
Olsen, W., B. Harriss-White, P. Vera-Sanso and V. Suresh. 2010. The experience of slum dwellers in Chennai under the economic and environmental insults of 2008-9. Work in Progress Paper 4. CSASP, Oxford University.
Pacey, A. and P. Payne. 1984. Agricultural development and nutrition. London: Hutchinson.
Panitch, L. and C. Leys. (Eds.). 2000. Working classes, global realities. Socialist register 2001. London: Merlin Press.
Pearson, R. 1994. Uneven development and the textiles and clothing industries. In Capitalism and development. (Ed.) L. Sklair. 189-210. London: Routledge.
Picherit, D. 2009. Between village and work site: the circulation of labour, clientelism and the politics of low castes in Andhra Pradesh. Doctoral Thesis. University Paris 10-Nanterre.
Polanyi, K. 1944. The great transformation: the social and economic origins of our time. Boston: Beacon Press.
Pujo, L. 1997. Towards a methodology for the analysis of the embeddedness of markets in social institutions: an application to gender and the market for local rice in eastern Guinee. D. Phil. thesis. Oxford University.
Rai, M. 2003. Genetic diversity in rice production: past contribution and the potential of utilization for sustainable rice production. http://www.fao.org/DOCREP/006/Y4751E/y4751e08.htm
Ramachandran, V.K. and M. Swaminathan (Eds.). 2003. Agrarian studies: essays on agrarian relations in less developed countries. New Delhi: Tulika.
Rogaly, B. 2003. Who goes? Who stays back? Seasonal migration and staying put among rural manual workers in Eastern India. Journal of International Development. 15(5): 623-632.
Rogaly, B., J. Biswas, D. Coppard, A. Rafique, K. Rana, and A. Sengupta. 2000. Seasonal migration, welfare regimes and adverse incorporation: a case study from East India. Paper for the Global Social Policy Regional Workshop. DFID, Koitta, Bangladesh.
Rogaly, B., D. Coppard, A. Rafique, K. Rana, A. Sengupta and J. Biswas. 2002. Seasonal migration and welfare/illfare in Eastern India: a social analysis. Journal of Development Studies. 38(5): 89-114.
Rogaly B., B. Harriss-White and S. Bose. 1999. Sonar Bangla agrarian change and agricultural growth in West Bengal and Bangladesh. New Delhi: Sage.
Roy, R. 1996. State failure: political-fiscal implications of the black economy. IDS Bulletin. 27(2).
Scott, G. (Ed.). 1995. Prices, products and people: analysing agricultural markets in developing countries. Boulder: Lynne Rienner.
Sen, A. 2002. Agriculture, employment and poverty: recent trends in rural India. In Agrarian studies: essays on agrarian relations in less developed countries. (Eds.) V. K. Ramachandran and M. Swaminathan. New Delhi: Tulika.
Sen, G. 1983. Paddy production, processing and women workers in agriculture – the South versus the Northeast. Working Paper no 186. Trivandrum, Centre for Development Studies.
Sklair, L. (Ed.). 1994. Capitalism and development. London: Routledge.
Srinivasan, K. 2004. Capital account convertibility and related issues. New Delhi, Transform India Group.
Swaminathan, M. 2000. Weakening welfare: the public distribution of food in India. New Delhi: Leftword Books.
Vera-Sanso, P. 2010. Ageing, poverty and neoliberalism in urban South India. NDA Findings no 5. http://www.newdynamics.group.shef.ac.uk/innovation-in-envisioning-dynamic-biomechanical-data1.html
Vijayabaskar, M. 2001. Industrial formation under conditions of flexible accumulation: the case of a global knitwear node in southern India. Ph.D. Dissertation, Centre for Development Studies, Thiruvananthapuram, Jawarhalal Nehru University, New Delhi.
Williamson, O. E. 1985. The economic institutions of capitalism. New York: Free Press.
Wood, G. 1976. Economic activity. In Exploitation and the rural poor. A working paper on the rural power structure In Bangladesh. M. A Huq. Comilla: Bangladesh Academy for Rural Development.
2 A rice marketing firm may do one or a combination of the following activities: buy, sell and broker, store, transport and process, and finance production, trade and post-harvest processing. Instead of being patterned, the activity profile of firms studied so far in India show tendencies towards complexity, diversity and uniqueness (Harriss-White 1996).
3 By this, even then Boserup must have meant upper caste Hindus.
4 This process of marginalisation continues to this day (Hensman 2001), despite the rise in female employment in South Asian export processing zones.
5 Demand disincentives include rules on maternity benefits, child care and equal pay. Supply disincentives include fixed working hours and the location of sites. Pearson (1994, 339-58) comments critically that employment regulations do not act as disincentives where they have no reach in the vast informal sectors of developing economics. Nor are sites and times a constraint under flexible production.
6 The female casual labour force can number up to 700 in cotton ginning and wholesaling firms (see Harriss-White 1996, chapter 7).
7 “Coolie” in Tamil means wages for casual labour.
8 In any case this would be almost impossible to measure accurately since the division of tasks in milling is sex-sequential.
9 With the commercialisation of labour, however, the practice of payment in tea and meals is dying, or itself being commercialised.
10 Male family labour (up to 13 members have been encountered) whose work is loosely specified, may work part-time or seasonally, or as part of a multiple occupational profile. Permanent workers (averaging 3 but up to 7, whose task specification may be quite refined but whose terms and conditions are varied) may work at the simultaneous performance of more than one activity within the firm and also on their own account. Rates of pay are unsystematised, and accentuated by both patronage and debt bondage. Male casual labour (averaging 9, but sometimes up to 40) may be permanently attached to a trading firm but employed on a daily, weekly, seasonal, group contract or piece rate basis for manual work. While tasks are highly specified, contracts, terms and conditions, and rates of pay vary greatly. Lastly, male child labour is used at key points in the grain marketing system (messages, carrying food and drink for negotiations, cleaning), and may be paid secure though very low wages. For some children, such work is an apprenticeship – though there is no reason why such apprenticeships should replace formal school, because the children of rich traders participate in both activities (Harriss-White 1996).
11 Female education leads to the lowering of birth rates, though not to reduced gender discrimination in regions of South Asia where this is practised (Dasgupta 1987; Jeffery and Jeffery 1998). But it is primary rather than tertiary education which achieves this result.
12 See for example Laidlaw’s 1995 study of Jain families.
13 Soaking up to 48 hours, parboiling in small batches (30 minutes to 3 hours), drying, raking and bird-scaring for 1 to 3 days (Harriss 1979).
14 Parboiling insures against the vagaries of rainfall; it reduces the proportion of broken rice grains and raises the rice recovery to 72%; the kernel absorbs proteins, vitamins and minerals and the bran layer absorbs oil; the kernel becomes more resistant to pests in storage and loses fewer solids to gruel in cooking.
15 The distress consequent to displacement is most acute in the north of the subcontinent (see Greeley 1987).
16 Gita Sen uses the Bulletin of Food Statistics to indicate 91,000 mills in 1975 (Sen 1983, 22).
17 Labour regards this as a tax because so few are able to redeem it at 60 if they live that long.
18 Rice mills in Punjab mill basmati rice exclusively for export while some 1,900 are agents for FCI, PUNSUP, Markfed, Punjab Food and Civil Supplies Dept and ancillary industries owned by the State and Central Warehousing Corporations (Kaur 2004).
19 A revolutionary component should increase output per unit of input and thus lower total costs per unit of output.
20 The precise gendered impact will depend on the tasks mechanised, the prior gender division of labour and that associated with any change in cropping pattern (Da Corta and Venateshwarlu 1999).
21 There are strong barriers to female entry into the “non farm rural economy” – whether it is mining of weaving and especially when migration is necessary for wage work. And export processing zones, where female labour is incorporated into the most oppressive work in global value chains (Jackson and Pearson 1998) still create a relatively small proportion of a developing country’s livelihoods.
22 One of the hypotheses considered for the period of the 1930s to the 1960s by Gita Sen (1983).
23 The jury is still out over the debates on whether lower pay for women than for men reflects productivity (impossible to verify in sex sequential production systems) – or whether it reflects supply conditions where work is a residual which needs to be compatible with the priors of household reproductive tasks – or whether markets, rather than being liberating, are an independent field of patriarchal oppression or a combination of these factors.
24 Huws gives examples of collecting goods from supermarket shelves; bagging your own vegetables; self-service auto fuel stations; automatic cash dispensers (2004, 27).
25 Leisure, which used to be spent creatively – such as the enjoyment of active music-making – has been completely replaced by pleasure-consumption packaged through commodities such as televisions, CDs, iPods, etc.
26 There is a third, though Huws does not explicitly develop this. The speed and comprehensivity of this process means that elderly people can no longer be a repository of useful experience. While Huws sees elderly people as part of women’s burden of care, it can be argued that the “retired” body is also yet another fertile field or object of commodification. The retired body is also a class specific body: in many societies working people never retire (see Vera Sanso 2010).
27 A family’s credit in business “is its stock in the broadest sense, which includes social position, its reputation and the moral and religious as well as the business conduct of all its members” (Laidlaw 1995, 355).
28 Along with the peasantry and religion, the “opium of the masses” (Marx 1999, 595).
29 This is one of the explanations for the period of industrial stagnation between 1965 and 1980 – see McCartney 2009. But it also points to the class specificity of demand for the goods generating high growth in the 21st century.
30 See Rogaly et al. (1999) for the debate over growth rates.
31 Husk in rural huller milling is not yet commodified, as it is retained for domestic parboiling and cooking purposes.
32 Most bran from huller mills however, is lost to the agro-industrial system and is retained in villages for animal and poultry feed.
33 Instead, due to poor infrastructure, both physical and telecommunications, it remains internalised into the labour process of individual firms.
34 State interventions such as the Public distribution system or Noon meal schemes have been found to be socially progressive transfers (Harriss-White 2004b). Educated and skilled labour then plays its role as the “global reserve army” (Kaplinsky 2003), or is let loose unprepared and unprotected in the bear garden of the informal economy (Breman 2003).
35 BPL means below the poverty line.
36 See Mooij (1999), for detailed evidence for Karnataka and Kerala.
37 Mills are polluted and drying yards involve back breaking work in hot sunshine.
38 Old age is defined by rural people in terms of the needs generated by incapacity and disability; and the harsh conditions of work in the post harvest system brings on old age at ages far below that of the Indian state’s stingy old age pension which is 60 years – see Olsen et al. (2010).
39 Three forces can be seen to drive this process: the differentiation of demand (carefully constructed, in India as elsewhere, through the media, advertising, etc.); the state-backed import, transfer and adaptation of technology; and relations of corruption and fraud (see Harriss-White 2003).
40 In the regulation of wage work, non-local (migrant) men of low social status occupy similar positions to women (Vijayabaskar 2001; Rogaly et al. 2000; 2002; Rogaly 2003).
41 The phrase is by Leys (2001).
42 See Chibber (2003) for big family business and corporate capital and Harriss-White (2003), for intermediate capital.
43 See also Harriss-White 2003 for a general attempt for the Indian economy.
Table des illustrations
|Titre||Table 2: Percentage of distribution of rice mills as on 01.01.2001|
|Légende||@ Number insignificantSource: Report of the Ministry of Food Processing Industry 2002; Kaur 2004|
|Titre||Table 5: Male livelihoods in rice mills in Punjab 2002 per unit|
|Légende||Source: data in Kaur 2004.|
© Graduate Institute Publications, 2011