The vast majority of paid and unpaid domestic workers are women. Explanations for women's employment frequently resort to crude biological reductionism, which purports to show that women have a range of natural attributes and aptitudes such as 'nimble fingers', 'quick eyes' and greater manual dexterity; rendering them, therefore, more suitable for assembly-line production, exemplified by the women-intensive textile and garment sector. This biological reductionism finds a parallel expression in notions of women as natural carers, and in the concentration of women workers in the lower grades of the caring professions, such as nursing, social work and teaching. One continuity appears in the undervaluing of work carried out by women. The work of a maid is a concentrated example of gendered work characterised by low pay and invisibility. At the same time, foreign domestic workers are recruited in processes that involve both the sending and receiving countries, much of which is processed and monitored, and subject to international and bilateral agreements on immigration, labour rights and social protection. A significant social dimension appears in the presentation of domestic workers as 'young girls'. Many, in fact, are not young girls, and there is significant variation in the age structure between different migrant groups and receiving countries. However, the representation of migrant women as 'girl-children incapable of making decisions about their contracts, pay and terms of employment' (Chin, 1997: 379) is an important social construct, and a conscious strategy adopted by agents, governments and households in the management of women carers.
The growing demand for elderly care has led to an increase in the number of domestic workers employed for this purpose. This is one globally recognisable market response to demographic trends towards an ageing population, which are most pronounced in the high-income economies of Europe and in the East Asian 'newly industrialising countries' (NICS). In the UK, it is estimated that between 1995 and 2031 the number of elderly people aged 65 and over will rise by nearly 57 per cent. The number of the very elderly (aged 85 and over) is projected to rise more rapidly, by around 79 per cent. Almost half of the growth in overall numbers is expected to occur between 2020 and 2031 (PSSRU, 1998: 46). In southern European countries, in parallel with the growing number of elderly people, there is a significant reduction in the average size of the family, with people tending to marry later and to have fewer children. While many of the current elderly are those who had large families in the pre-war period, future generations will progressively have to draw on much smaller numbers of children as potential carers (Mestheneos & Triantafillou, 1993; Dell'Orto & Taccani, 1993; Twigg, 1996).
Trends towards an ageing population have converged with attempts by governments to 'reform' pensions and care for the elderly, typically by increasing the role of the private sector, by the devaluation of the state pension, and by raising the retirement age. These policies have provoked major conflict between trade unions and governments in a number of European countries (Italy, Greece and France), and the fear of retirement has become a critical personal and political issue sweeping across Europe. In the UK, the privatisation of care for the elderly accelerated during the 1980s and 1990s as part of the shift from residential to 'community care' (see Hughes, 1995). Private domestic help tends to be highest amongst single elderly people, and 11 per cent of single elderly people with some level of dependency in the UK indicated this as a recourse. It is further projected that by the year 2031, private domestic help will make up the second-largest category of non-residential care in the UK (PSSRU, 1998: 81).
The nature of care for the elderly is shaped by the level of dependency, frequently measured in terms of the ability to perform activities including those instrumental to daily living. This is used by local authorities in the UK and elsewhere in order to assess need for residential care, and gives us an indication of the kind of work carried out by caters. High indices of dependency include the inability to bath and shower oneself, to dress and undress, to get in and out of bed alone, to go to the toilet alone, and in some cases to feed oneself. The proportion with no dependency tends to fall markedly with age. Department of Health figures show that over 75 per cent of elderly residents are in private care homes (PSSRU, 1998: 101). Another study of trends in residential and nursing home care indicates that the mean length of stay of permanent residents in private residential homes was substantially longer than the previous decade, and that residents in nursing homes were most likely to have been admitted from hospital (Darton & Miles, 1997: 9-10).
Twigg (1996) makes the distinction between 'informal helping' and 'heavy-duty caring' in order to explain different tasks in home-based care, which require different levels of intensity and effort. These range from help with practical domestic tasks such as preparing meals, shopping and housework, to help with personal and physical tasks such as dressing, bathing, going to the toilet and getting in and out of bed. Help with physical tasks is very labour-intensive, and is associated with long hours of caring. Many of the tasks that denote high levels of dependency have been excluded from government funding in the UK. One study of elderly people who had some level of dependency pointed to their spouse (38 per cent), another household member (17 per cent), or a relative outside the household (42 per cent), as the most significant carets. In 17 per cent of the sample, health and social services were mentioned, and the use of paid help featured in 15 per cent (PSSRU, 1998: 74).
The increased role for the private sector in care for the elderly raises important questions about the relationship between market, community and need. It is clear from the above that household members, rather than the community in general, are shouldering the major responsibilities for elderly care. It is also the case that the demand for care, while driven by need, is--as with other goods and services--a function of the person's income and the price of the good. While need is an important criteria, in privatised care this is not necessarily the case. Demand might not be effected because of inadequate income. Significantly, since the specific care-demand is shaped by the person's age, gender, occupation and health status, this has particular implications for elderly people, given that more of them tend to be in the lowest-income groups, and can least afford private care.
The commoditisation of reproductive labour
Passing the costs of increased elderly care needs to households has important implications for working class and professional women, who have to negotiate complex productive and reproductive demands on their time. Reproductive labour is generally the labour used by households to produce the services necessary for the care and maintenance of the current labour force, and in the reproduction of the next generation of labour. This includes childbearing and child rearing, as well as daily household maintenance.
One significant and growing dimension of this labour could be the care of close elderly relatives. It is women who come under the greatest pressure to care for the elderly. In culturally diverse societies, powerful prevailing ideas and social practices ensure that wives, daughters, daughters-in-law, granddaughters and sisters provide the necessary reproductive labour for the care of the elderly. At the same time, women are a growing section of the world's labour force. In Malaysia, for example, women accounted for a third of manufacturing employment in 1970, and nearly half by 1990 (Chin, 1997: 369). In Singapore, women were actively encouraged to re-enter employment after they had had children, and between 1980 and 1994 the number of married women employed in the labour force increased from 29 per cent to become 45 per cent of the total labour force. The government of Singapore encouraged middle-class households to employ foreign domestic workers as the cornerstone of its return-to-work policy. For many professional women, foreign maids became essential for housework, childcare and for the care of aged parents. Indeed, according to Yeoh et al. (1999: 120), they 'could not work but for the employment of maids'. Similar trends can also be observed in Hong Kong, where many of the employers are also women from the professional and managerial class (Ozeki, 1997: 685).
Much of the literature points to the transfer of reproductive labour as a recognisable response to the 'double burden' faced by professional women worldwide. Parrenas (2000: 561) writes that 'reproductive labor has long been a commodity purchased by class-privileged women'.
The conclusions drawn from the above analyses are that the 'transfer of social reproduction is a viable alternative only for middle and upper class families' (Heyzer & Wee, 1994: 39). Foreign domestic workers can also be understood as an example of globalisation in the personal service sector. Saskia Sassen (1991, 1996) suggests that one characteristic of contemporary globalisation is the incorporation of diverse immigrant groups into the informal labour markets of Europe and North America, typically as low-waged service workers or as marginal entrepreneurs. In this literature, the transfer of reproductive labour, or the 'international transfer of caretaking' (Parrenas, 2000: 561) lies at the heart of an analysis of modern capitalism (also see Chin, 1998; Hondagneu-Sotelo, 2001; Chang & Abramovitz, 2000; Parrenas (ed.), 2001; Anderson, 2001).
The extent to which the commoditisation of reproductive labour becomes prevalent...