Why is ‘women and work’ an issue? In the past, as in the present, most women, like most men, have worked for a living. But women’s work, its particular historic development, its current content and location, and its meaning to its subjects and objects, is different from the work of men. Work is a gendered experience.

Women’s work is distinguished primarily, though not exclusively, by their responsibility for certain tasks associated with daily and intergenerational reproduction. The cooking, cleaning, childcare, nursing and nurturing involved is a distinct labour in many ways: not least because it is unpaid. It has remained privatized even in advanced industrial economies whether organized by a plan or by the market, and in societies where planners’ preferences or market prices index value, work which is neither planned nor marketed is undervalued, indeed often to the point of invisibility. It is not deemed to be work at all.

The differences between men’s and women’s work does not stop here. Paralleling the division of labour between men and women in unpaid work in the home is a division of labour by sex in paid work. Men and women are not randomly distributed across the employment structure. Rather there are men’s jobs in which primarily men are engaged and women’s jobs where the labour force is overwhelmingly female. There are few mixed occupations where men and women can be found in the same proportions as in the labour force as a whole.

A third significant dimension of women’s jobs concerns their terms and conditions. They are less likely to be complemented by expensive capital equipment, thus less productive, more likely to be temporary and insecure, less likely to be organized, and less associated with prospects for promotion. Above all they are less well paid.

The implications of these characteristics for the economic position of women are clearly pernicious. In the home women are unpaid and unappreciated; in the workplace their relatively low wages make them vulnerable to poverty and deny them independence from state or family subsidies which are desperately needed if there are others dependent on their wages. Their higher labour market flows and sometimes disproportionate representation in unstable jobs make them especially subject to recession-enforced periods of nonparticipation or unemployment.

Massive empirical evidence documents these characteristics of women’s work as universals within the advanced industrial world (OECD 1985; ILO 1985). There are outliers. Modern Sweden, the USSR, and most countries during wartime have achieved extraordinary female involvement in paid production. Ireland is out at the opposite extreme. Eastern European countries have less sexual segmentation and different jobs are feminized, but a sexual division of labour exists nonetheless. The deployment of women in some countries has to be seen as mediated by religion, but this usually compounds rather than mitigates the privatization, segmentation and subordination suggested here as universals.

The history of women’s work can also be characterized in terms of stylized facts. Thus the painful and problematic transition of women from production for use to waged work, closely documented in the British case, can and has been paralleled in other national experiences (Pinchbeck 1930; Tilly and Scott 1978). Even countries at very different levels of development can be interpreted as exhibiting aspects of the same sexual divisions of labour if the comparison is with the historical experience of the now economically advanced countries.

So overwhelming is the evidence of a generic women’s experience of work in both the past and the present that it has prompted explanations in terms of some universal, cross–cultural, historical model of female subordination often described as patriarchy. This and other sometimes integrated universalist explanations appealing to biological or psychological differences between the sexes constitutes the first explanatory framework reviewed below.

Models of Explanation

Patriarchy

Patriarchal models of explanation see male dominance and female subordination as an enduring characteristic of all societies, hence readily explaining the commonalities of experience. An important move here is the detachment of patriarchy from other class systems of dominance and subordination and the denial, in opposition to classical marxism, of the former’s dependence logically and historically on private property (Delphy 1977; Hartmann 1976).

Much of the empirical debate has been anthropological, involving the operational definition of patriarchy and its identification in preclass societies. One question of interest here is whether a sexual division of labour in work itself constitutes patriarchy, or if separate but equal roles for men and women are feasible.

The enormous changes that have taken place in women’s political, legal and economic status suggest that an immutable patriarchy is indefensible, and most recent presentations within this framework argue that while patriarchal power relations remain a constant their particular expressions, and perhaps their intensity, change with economic and social development. Much of this literature is concerned to specify the form of interaction between patriarchal power relations and class society while retaining the essential autonomy of male authority and control. Adherents of this model of explanation must argue that the subordination of women in advanced capitalism is explained not by the dominant mode of production but by patriarchal power relations operating through the family and the political system and within the social relations of production. Capitalism may exploit the divisions among the working class arising from patriarchy, but above all capitalism must adapt itself to a given system of sexual hierarchy. Thus the conditions of women’s employment are primarily determined by the dominance of men and capital must adapt to a sex-differentiated wage hierarchy so that men’s power in the domestic and political spheres is not contradicted in the workplace. Adaptation takes place despite the interests of capital in eroding male power and establishing a homogeneous competitive labour force. Alleged mechanisms of control have received attention: ideology and socialization as well as concrete institutions like schools and trade unions.

One of the major criticisms of patriarchy as an explanatory framework is that it contains no material explanation of women’s position and ultimately the argument must devolve on biologism, psychoanalytic theory or cultural catalepsy. Another possibility is to see patriarchy itself as an economic class system with men as the appropriators of female labour-time: a prospectively rich but empirically underdeveloped approach that is unfortunately thwarted by the suspicion that much of women’s potentially alienable labour time benefits children rather than fathers.

Whatever the origins of patriarchy, they are clearly independent of the capitalist mode of production. There is therefore truth in the argument that women’s economic subordination has its roots in precapitalist forms and cannot be explained solely by functionalist reference to capitalism. The third conceptual framework reviewed below owes much to this challenge to classical marxism.

Neoclassical Economics

Neoclassical economic theory tries to explain women’s economic subordination in terms of rational utility-maximizing behaviour. The emphasis has been on the narrowly economic issues of women’s labour supply, occupational distribution and relative pay. Neoclassical analysis of labour supply postulates an individual allocating his/her time between work and leisure. Leisure here is a catch-all term for all uses of time other than paid work but it is a peculiarly male label in view of the reality of domestic labour and the latter’s impact on women’s ‘leisure’. Indeed, much of the neoclassical interest in female labour supply has occurred precisely because the question ‘Why do paid work?’ can meaningfully be asked of women, for they have the option of work in the home. The added dimension to women’s choice set challenged neoclassical economists, who soon realized that female labour supply could not be modelled analogously to that of men and significant innovations with widespread repercussions for mainstream economics followed. Two relevant developments are sketched below.

First, recognition that simple reduced form models of female labour supply, drawing only on samples of working women, involved serious biases, prompted the development of structural models in which labour supply decisions were modelled as involving discrete quantitative choices about whether or not to work and how many hours to work. These second generation labour-supply models involved new techniques for handling discrete choices (logit, probit and tobit) and the introduction of the shadow or imputed wage to capture the influence of unobservable but relevant returns, for example, to non-market activities (Heckman 1974; Heckman et al. 1981; Killingsworth 1983).

As longitudinal data have become increasingly available, particularly in the United States, investigation has focused on lifetime experience and appropriate techniques for handling analyses of attachment to the labour force over the whole life cycle have been developed. Much empirical work has now been undertaken for the USA and UK providing important information regarding female labour supply.

However, the widespread use of personal characteristics and family circumstances as proxies for the tastes of individuals or the shadow wage of domestic work, indicates the limitations of the approach (Greenhalgh and Mayhew 1981). Belief that these variables genuinely represent exogenous tastes or comparative advantages surely involves some dubious propositions about sex and race-linked biological endowments! However, the attachment of certain behavioural proclivities to variables like race, sex and the presence of small children may simply reflect rational expectations about discrimination or family opposition to waged work, or even historical differences in the relative earnings potential of men and women. Then labour supply cannot be taken as independent of demand-side variables or the organization of the family and a historic and interactive analysis between production and social reproduction must be undertaken.

A second development stemmed from recognition that husbands’ and wives’ decisions about work, leisure and homeproduction are interdependent, which contested the basis of neoclassical economic theory in the individual decision-maker. Simultaneously neoclassical economists, looking for new areas of behaviour to subject to their choice theoretic framework, lighted on fruitful terrain: the household. Becker’s (1965) analysis of household decisionmaking in terms of his theory of the allocation of time produced the New Household Economics (NHE).

The NHE views the household as a production unit and consumption decisions as dictated by the drive to maximize utility. Various activities can contribute to utility and these activities require inputs of time and other goods. The ultimate product is the utility derived. All kinds of household decisions from the mundane to the consideration of whether to have children of certain qualities and quantities, have been cast in this framework. Economists’ techniques can then be mobilized to describe the optimal allocation of time under certain assumptions about tastes and relative costs.

The value of the NHE lies in its explication of the link between labour supply decisions and consumption decisions and therefore between women’s primary responsibility for work in the home and their partial and discontinuous involvement in paid work. Unfortunately the insight is not maintained. Other relevant problems are sidestepped or trivialized: for example, the deep difficulties involved in synthesizing a collective preference ordering from individuals’ preferences are not addressed here despite the move from the individual to the household. One way out is to postulate a set of rules for aggregating and weighting individuals’ preferences; another is to understand the collectivity’s preference ordering as ‘given’ by an individual representative: ‘the benevolent dictator’ of welfare economics. Both methods abstract from the conflicts and complementarities among household members and, essentially, collapse household and individual decision-making. Moreover, feminists see the benevolent dictator as mirroring the dominant patriarchal form of family organization in society and, not surprisingly, are leery of closing the model by accepting the very hierarchy of authority that they want to question. Perhaps the recent interest in bargaining models of family relations (Pollak 1985) will eventually help with these difficulties.

More importantly, the NHE cannot explain why the most efficient allocation of time by family members should involve a sexual division of labour between paid and unpaid work. Specialization is explained by comparative advantage, but the latter’s suggested origins often make the argument circular: women hiring men as breadwinners because they earn more, but women earning less because they leave the labour market to have children (Becker 1981). Alternatively the argument is sometimes shored up here by biologism: higher productivity in domestic labour simply being read off the female sex, or women’s prior investment in children, since they carry them in the womb, being held to make them more inclined to further childcare investment (ibid). So comparative advantage is deduced from the existing sexual division of labour and then used to explain that division.

Neoclassical economics’ treatment of the occupational distribution of women workers and their relative terms and conditions, suffers from similar defects. Briefly both occupational choice and relative wages are viewed as the outcomes of rational utility–maximizing behaviour. Indeed the former is often held as an ‘explanation’ of the latter as women’s disproportionate representation in poorer paying jobs accounts for a substantial part of the male–female wage differential. Sometimes occupational choice and relative wages are analysed in the framework which subsumes them both into sex specific choices about levels of investment in human capital. Employers’ discrimination has usually entered the argument only to explain the residual after the impact of other sex specific worker characteristics have been deducted.

Why do women choose certain occupations in preference to alternatives despite the fact that the resulting occupational distribution is (demonstrably) a major factor in their relatively low pay and poor terms of employment? An obvious defensive manoeuvre, and one which neoclassical economists have not ignored, despite its disturbing implications for the premise of rational-maximizing behaviour, is to cite pre–market discrimination as engineering women’s choices, or cultural predispositions to study certain subjects as leading girls into less well-paid jobs. But these arguments are non sequiturs as far as economic explanations of women’s subordination are concerned.

Human capital theory has also been used to explain such seemingly paradoxical choices as rational-maximizing behaviour. Polachek (1975, 1976) has argued that women choose occupations for which earnings losses during spells of non-participation are minimized. Since women plan intermittent paid work because of intended childrearing, they prefer occupations where skills depreciate only slowly when not employed. Women’s primary responsibility for children is assumed in this model, as are differences in the rates of decay of human capital across occupations. Polachek’s work has been subject to both theoretical and empirical criticism internal to his own paradigm (England 1982; Beller 1982). Radical reservations have also been expressed about the notion of productivity embedded in human capital theory. It is essentially a sexist concept since it only counts as productive those skills which the market rewards and many skills which women have go unrecognized (Dex 1985).

To summarize: neoclassical economic theory confuses descriptions with explanations of the subordination of women. Moreover, cultural and even biological factors have sometimes been used as prime movers in the argument, despite the inconsistency of this procedure with a belief in the prime explanatory power of rational maximizing behaviour. This is not to deprecate the value of much neoclassical work on female labour supply, or the insights of the NHE, but as a model of explanation of either women’s specialization in the home or the sexual division of labour in paid work, the approach is ultimately nugatory.

Neo-Marxist Economics

The neo-marxist approach developed out of the challenge to classical marxism’s treatment of women’s subordination posed by the persistence of patriarchal social relations. Interest was initially focussed on the articulation of unpaid work in the home to production relations in a capitalist economy and the meaning of the former for value categories: the domestic labour debate (Dalla Costa and James 1972). How domestic labour should be integrated into value accounting represented a challenge to marxian value theory and to classical analyses of the natural price of labour (Himmelweit and Mohun 1977). The domestic labour debate’s attention to work done to reproduce workers and their labour power promoted new interest in the process of social reproduction more broadly defined (Kuhn and Wolpe 1978). Consequently developments in the theory of the family have both drawn on and contributed to developments in the theory of the state Both have contributed to significant progress in neo-marxist understanding of advanced capitalism. It is interesting to note that neo-marxist interest too is now turning to bargaining models of the family.

Attention also spread to analyses of the relationship between women’s primary responsibility for the reproductive work of the home and their position in paid labour. Concepts for classical marxism, such as the reserve army of labour, as well as from contemporary labour economics, such as labour market segmentation, have been especially useful in analyses of the impact of secular restructuring and the business cycle on women workers. Recent studies have tried to test whether women do constitute a buffer labour reserve over the business cycle or if their segregation in the less volatile sectors and occupations affords them relative protection in hard times (Bruegel 1979). A third hypothesis which has received attention is whether women’s cheapness in conjunction with their predominance in growing sectors is causing women to be substituted cyclically and secularly for men (Rubery 1987).

These studies have developed a dual systems theoretic approach which sees women’s economic subordination as the outcome of interaction between social reproduction and production. Although debate continues as to the precise specification of this interaction, key characteristics of the neo-marxist approach to understanding women’s work include: (1) an insistence that social reproduction be taken seriously as work and as an integral part of the economy; (2) a conviction that social reproduction is relatively independent of the organization of production and does not respond smoothly, accommodatingly or predictably to the needs of the economy; (3) the deduction therefore that the relationship between the spheres of production and social reproduction can only be understood historically and is not predetermined; and (4) also that the relationship must be analysed within a nonfunctionalist perspective (Humphries and Rubery 1984).

The approach is less methodologically hidebound and much more open-ended than that of neoclassical theory. Nevertheless developments within this literature have both drawn from and fed back into neoclassical economics. Moreover both approaches are forced to respond to the contemporary dramatic increases in female participation rates in advanced industrial economies and attempt to predict the implications for the unequal burden of domestic work, for occupational segregation, and for the terms and conditions of women’s work.

Conclusion

As suggested above, interest in women’s work is not only analytical. Social commentators of the past, as well as of the present, have attributed major significance to shifts in the allocation of women’s labour time between the home and the workplace, though not always in agreement as to their implications. Contrast the oft-quoted view of Engels that ‘The emancipation of women will only be possible when women can take part in production on a large social scale, and domestic work no longer claims anything but an insignificant amount of her time’ (1891, p. 221) with the more traditional position of Marshall: ‘If we compare one country of the civilized world with another, or one part of England with another, or one trade … with another, we find that the degradation of the working classes varies almost uniformly with the amount of rough work done by women’ (1961, p. 565).

Perhaps the reconciliation of these two distinguished views lies in consideration of the terms and conditions of women’s paid work and how it is coordinated with childcare and domestic labour. Significantly it is college-educated women with their more interesting, betterpaid jobs, who are able to purchase domestic help and high-quality childcare, who find work most enriching. For women as for men, one objective must be more interesting work. More contentious, in terms of the distribution of necessary labour time between the sexes, is the target of a more equal distribution of domestic responsibilities.

See Also