Radical
Statistics
Engendering poverty research. How to go beyond the feminization of poverty
Elisabetta Ruspini
Introduction
This article discusses the gender dimension of poverty risks and some of the methodological issues connected to the implementation of a gender-sensitive approach in poverty research. More specifically, it focuses on the relationship between poverty and inequalities in the process of distribution of resources within households.
The debate on the definition and 'measurement' of poverty rarely includes a specific discussion of gender, treating poverty principally as a gender-neutral phenomenon. It often ignores gender differences in the causes, extent and experience of economic deprivation.
But poverty has always been a gendered phenomenon: historical data provide a remarkably consistent picture of women's experience of deprivation over the last hundred years (Glendinning and Millar, 1991). Other findings (e.g. Scheiwe, 1994; Ruspini, 2000) indicate that women's poverty is a quite different experience from that of men's: the structural causes of women's poverty are to be found in the interaction between dependency, social exclusion and social change within the three resource systems: family, labour market and welfare systems. These causes are rooted:
1. In the private domestic sphere. Power within marriage and the family is directly related to the control of financial resources and to the participation in the labour market. Generally speaking, husbands have more power than wives in the control of family resources (Pahl, 1989). It has also been observed that the poverty of a poor family is not shared equally by the members. In poor households women more often have responsibility for managing the money, whilst in households where the main source of income is earnings, men will more often control the finances (Pahl, 1995). Other studies on families below the poverty line find that the women primarily bear the burden of a family's poverty, inasmuch as they inhibit their own consumption. Thus, the decline in the quality of life coming from poverty affects the woman more than the man, in that the woman manages the limited resources to satisfy the whole family's demands (Lee 1999).
2. In the sexual division of labour which assigns to women a primary role (largely hidden and unpaid) and a secondary position in the labour market. Women's lives are still shaped by the family responsibilities they have traditionally been expected to take on: these tasks shape women's work patterns, the type of occupation they work in, their earnings and their social security benefits. 'Moral obligations' create a time conflict between paid and unpaid work, care tasks and employment (Leccardi 1996)1: such inequalities in the distribution of unpaid work remain even when the man is unemployed. As a consequence, women often take a poorly paid job, in order to meet family needs (Graham, 1987) or find employment which fits in with their domestic chores: part-time work; work which is near their homes or their children's schools; and work in anti-social hours, that is, when their partners are at home to care for the children. At the same time, their partners may oppose their employment unless the home environment does not suffer (Payne, 1991). The major investment by women in unpaid work reduces their potential for earning income and increases their risk of poverty (Daly, 1992): thus, the earning gap between the sexes is a direct consequence of the home responsibilities assumed by women. Women's income stability over the life course may also be greatly affected by economic dependency (Sorensen and McLanahan, 1987): the greater the dependency, the greater her potential loss of income should she lose her spouse by death or (increasingly common) divorce.
3. In the welfare sphere. All this is reflected in the private, occupational and public systems of welfare provision. The inequalities which women experience in paid work are mirrored in their different access to, and levels of, income replacement benefits (Glendinning and Millar, 1991). Institutional arrangements can negatively affect women's and mothers' access to income and increase their poverty risks or may limit the number of available alternatives and therefore make it difficult to find ways out of poverty (Scheiwe, 1994).
4financial institutions, e.g. insurance companies and mutual societies as well as employers and the state, provide mechanisms for distributing resources between women and men and between generations. Tax systems subsidise some of these transfers through the use of tax relief on contributions paid by employers and/or employees as well as on the funds generated by the insurance or mutual companies themselves. Financial markets, as well as labour markets, have an impact on the distribution of resources between men and women but the assumptions on which they operate - especially with regard to access to credit and to mortgages - have not yet been subject to the detailed scrutiny to which labour markets have been subjected (Land 2000). As an interesting example, Pahl (2000) shows that gender differences in the use of credit cards are associated with differences in employment status. When the man and the woman are both in full time employment they are equally likely to use a credit card. However, women in part time employment are less likely, and women without employment very significantly less likely to use a credit card, by comparison with their employed husbands.
Social inequalities are also shaped by social change. Life course changes are far more pronounced for women than for men both in the family and in the labour market. The most dramatic changes in the life course are portrayed in the family sphere. Family life course changes include population ageing, the weakening of kinship networks (as a consequence of both increasing individualism and falling birth-rates), the increase in separation/divorce and in non-marital unions, the increase in births out of wedlock, and a decline in nuptiality. Changes in the labour market have created a complex scenario marked by a reduction in the number of opportunities for obtaining permanent jobs and by a parallel increase in flexibility: unemployment and marginal forms of employment, a decline in life-time occupations, and a shrinkage in permanent contract career ladders (Mayer, 1993; Morris, 1994). Obviously the transformations currently underway do not have the same impact either on the diverse groups which make up society or on the individuals within each family: some subjects - such as women - are more vulnerable than others. Families where the head of the household is a woman, often separated or divorced and with dependent children, are indeed far more vulnerable than families where the head of household is a male or where an adult male is present.
Thus, specific risk factors affect women (and men) in particular. The expectation is that critical family events - such as lone parenthood, divorce or separation, particularly in the presence of children - will be stronger predictors of poverty transitions for women than for men because of the female economic dependence on a male breadwinner combined with caring responsibilities at home.
The different interaction of beginning events also determines a different duration of poverty spells for the sexes: female poverty trajectories are longer than men's, since they are strongly related with critical family events and women's burden of care and domestic work (Ruspini, 2000). Women are also more likely to enter poverty situations such as lone parenthood or low income, to remain in those situations and to exit from them with greater difficulties. Finally, poverty among women persists and is structurally reproduced over time and generations: one of the vehicles for the transmission of poverty arises from the absence of sharing of family responsibilities.
Thus, women cannot simply be 'added in' to existing analyses: instead, a different analytic framework is required. What is needed is the elucidation of the different processes by which both women and men fall into, experience and escape poverty, as opposed to the paradigm we have now, which is an analysis of the way in which households experience poverty. Our understanding of women's position in society requires us to look not only at individual women's position in the labour market but also at their familial position, the resource they derive from it and their level of dependency within households (Sorensen and McLanahan, 1987). As Millar argues (1999), in order to develop gender-sensitive ways of 'measuring' poverty what is needed is a way of placing individuals within households and measuring both their contribution to the resources of that household and the extent of their dependence on the resources of others within the household. Thus, the most crucial question for poverty research is how to open the 'black box' in order to understand whether and how women's poverty is concealed (Millar and Glendinning, 1989; Arber 1990). We shall ask to what extent do all household members share the same level of economic welfare and what methodological implications emerge from the study of intra household resource distribution.
The challenge is how to proceed methodologically. In order to offer an answer to this crucial question, I will discuss three of the central methodological issues required to analyse poverty and especially women's deprivation: identification of the poor and how to engender the poverty line approach; appropriate units of measurement; and the non-monetary dimension of poverty.
Engendering the poverty line approach
There are many alternative ways of defining poverty and to identify the poor: any 'measure' of poverty involves a large number of choices. Moreover, poverty rates are very sensitive to the definition of poverty itself. The adoption of either one or other method to define low income may heavily influence both the absolute number and the structure of the population which is 'poor'.
The poverty-line approach poses in particular a number of problems. In fact, there is both the crucial question of where to draw the line and that of how to conceptualise and measure income and money resources (Daly, 1992). Moreover, only relying on income ignores the fact that there are other resources that can profoundly affect people's standard of living: gift exchanges, income from relatives and friends, services/benefits in kind such as savings produced through women's domestic labour in food preparation, cleaning and other household tasks. It is indeed extremely difficult to estimate either the magnitude and the distribution of income received from 'hidden' transfers or from home production: this is particularly problematic in the case of women, since in practice, the largest source of home-produced goods today is the work that housewives and mothers do in the home (Ruggles, 1990). Thus, the 'poverty line' approach is inherently gender-blind. The question is, how to engender it
Let us start from the choice of the equivalence scale: an adjustment in needs is important, since economies of scale may arise as a household increases in size. One can assume that, due to economies of scale, the needs of a family for resources grow with each additional member, but not in a proportional way. With the help of equivalence scales, each family type in the population is assigned a value in proportion to its needs (Foerster, 1993).
The choice of an 'official' equivalence scale is controversial, since it can substantially affect the composition of the poverty population. This choice is even more controversial in cross-national research, since it must account not only for differences across households but also of country-specific differences (Burkhauser et al., 1994). Thus, there is a considerable range of methods which can be used to derive equivalence scales and a large number of scales are used in OECD countries.
Another priority is to use and compare different assumptions about sharing. There are various possible solutions to the problem of measuring within-household resources. The traditional approach assumes complete pooling of income, so that the living standards of all members of the same household are assumed to be the same. One alternative is to make the opposite assumption ( of no pooling at all ( and measure poverty rates on that basis (Millar, 2000). For example, Daly (1995) distributed households income among individuals by applying two conditions of income sharing: the first is that household members share the aggregate income equally; under the second assumption family income is allocated unequally among members, using the adjustments inherent in the OECD equivalence scale. Not surprisingly, Daly's empirical evidence shows that the assumption of equal sharing yields the lower poverty rates. It is female poverty rates which are most affected: women are particularly hit by hidden poverty.
It is also necessary to define a threshold or poverty line to distinguish households and individuals who are poor from those who are not. Due to the fact that poverty is a contested concept - a phenomenon difficult to understand, define and measure - it is not possible to draw one single, valid poverty line, below which all individuals or households are undeniably poor. As already said, poverty lines can be set using a great variety of alternative methods, and figures depend crucially on the poverty line chosen. In my analysis of female poverty dynamics in Germany and Great Britain, I defined the poverty lines as 40 percent, 50 percent and 60 percent of the median monthly household equivalent income and of the median monthly individual labour income.3 Those below the 40 percent line may be classified as 'very poor', those below the 50 percent line as 'poor' and those below the 60 percent line 'near poor'. The choice of using different poverty lines is an appropriate correction for the arbitrary choice of using only one poverty interval: poverty rates are indeed very sensitive to the poverty line definition itself (Buhman et al., 1988).
The unit of measurement: aggregate and individual poverty measures
The most crucial question for poverty research is how to open the 'black box' in order to understand whether and how women's poverty is concealed. I shall ask to what extent do all household members share the same level of economic welfare?
My answer is that there are two possible ways of exploring the concealment of women's poverty within households. First, through careful choice of the unit of measurement. Second, through increased availability of variables which specifically aim to depict the three processes involved in the acquisition and expenditure of resources within the family: the entry of resources into the household (sources of income); how resources are allocated and controlled; and how resources are expended (Daly, 1992).
Early 'measurement' of poverty (e.g. by Rowntree) was at the household level, and much still is. The reason for focusing on aggregate units rather than on individuals is based on two assumptions. Firstly, that resources are shared equally within the household: under that assumption, the household is treated as an inseparable economic unit in which every member shares in the common resources, and a member's personal resources are used to meet the collective needs of the other members. Secondly, it is implicitly assumed that all members of poor households experience deprivation caused by poverty to the same extent.
It has indeed long been a tradition to assume that family resources are distributed equally among its members, with the family serving as the fundamental unit. Also, the inclination has been to take the head of a family as the major subject (Lee, 1999).
Ample evidence on financial management within marriage has thrown considerable doubt on the prevalence of egalitarian income pooling by couples. Various patterns of money management have been reported, some of which involve the male breadwinner having discretionary funds to spend on himself unmatched by similar resources for wives (Pahl, 1989; 1995). Other evidence has shown that family members have unequal access to resources (such as food and clothing) as well as to space, warmth and light, and men tend to be the 'privileged' consumers. Wives tend to be responsible for basic necessities and children's needs, and to control a larger proportion of the budget the smaller it is. Access to resources within households is determined by the relative status of family members: it is structured by norms of behaviour that are a consequence of power relationships which are themselves structured by gender, age, class and race (Brannen and Wilson, 1987).
If the unequal sharing of resources is occurring, then conventional methods of poverty 'measurement' will lead to an underestimate of female poverty and an over-estimate of male poverty (Findlay and Wright, 1994). Thus, in order to make gender differences explicit in research on poverty, an aggregate measure of income is not particularly useful. The study of poverty on an aggregate rather than individual basis obscures the particular circumstances of women. By contrast individual measurement assumes that no sharing or pooling of resources (through the common use of rooms and amenities) takes place within the family.
The use of family income to 'measure' poverty masks the economic weakness of women. If, on the one hand, the family plays a crucial role in pooling different sources of income and in mediating hardship, on the other hand it can also mask an unequal distribution of resources. In fact, women add a great volume of non-market work that helps families to cope with the lack of resources resulting from either unemployment or job instability. If a couple's income and resources are not pooled, then married or cohabiting women, whose own income is low, may suffer from poverty that is hidden within official statistics (Ward, et al. , 1996).
Acquisition and expenditure of resources within the family
The availability of variables which permit the exploration of how income and other resources are converted into standards of living within the family is a crucial element for understanding women's poverty. If it is true that family members pool their resources to some degree, the pooling clearly presents problems in the measurement of income levels, as we do not know to what extent people pool their resources and how "to relate this pooling to patterns of control and responsibility for different items of expenditure by the various household members" (Millar and Glendinning, 1989:376).
Unfortunately, such variables are very rare. What follows are the main points that emerged from my comparative research experience, which used national household panel data:
- The British Household Panel Survey makes it possible to understand how household finances are arranged. It contains two crucial variables: the first describes which person in the household is responsible for big financial decisions and the second describes how financial decisions are organised.
- The European Community Household Panel data offer the possibility of identifying the household's main source of income and, also, make it possible to gain insights into both the characteristics and the relevance of personal financial support from others outside the household.
- The German Socio-Economic Panel survey, while not permitting the question of how household resources are organised to be investigated, contains several variables concerning financial assistance and private transfers from outside the household, however, most of these are only available from 1994. The data set also offers some information concerning child-care and other kinds of help offered by persons outside the household.
- The Bank of Italy Survey on Household Income and Wealth contains information concerning the exchange of financial resources within family networks and intergenerational transfers of resources.
- The Panel Study on Belgian Household contains some variables concerning the exchange of financial resources within family networks.
- The Swedish Household Market and Non-Market Activities survey offers some information on child-care by persons outside the household and some very general indicators about financial help which comes from outside the household.
A final suggestion comes from Millar (1999; 2000) who argues that the issue of how to place individuals within the household and capture not only their contribution to the resources of that household but also their dependence upon those resources also lies in the examination of sources (not just levels) of income. A good example of such an approach is a study of low pay and poverty in Britain (Millar et al., 1997), which estimated the contribution which different individual sources of income made to reducing the risk of household poverty. Figures show that, of all low-paid workers (defined as hourly earnings of less than half the median for all workers) very few (only 8 per cent) were able to lift their households out of poverty by means of their own income alone. If it is the man who is low paid, he had a much greater chance of being able to lift his household out of poverty than did the low-paid women (2.4 per cent). Finally, low-paid lone mothers are the least likely to be able to keep their households out of poverty (29 per cent stay poor) and where they do manage to stay out of poverty it is state support, through benefits, which does this.
Alternative measures of poverty
Any understanding of poverty requires a focus not just on numbers: the individual experience of poverty should also be described (Alcock, 1993). Poverty research that uses income-based measures and indices of material well-being, fails to address the multi-dimensional nature of deprivation and focuses attention on narrow policy objectives (mainly on raising income levels) (McKendrick, 1998). The non-monetary dimension of deprivation is very important if we wish to capture the gender nature of poverty, since it makes it possible for us to understand the consequences of economic hardship and the connection between low incomes and lack of resources.
There are less quantifiable aspects of poverty, such as not being able to see friends and relatives, which are not only different for women and men but also differ between diverse groups of women. A comparison of activities pursued by women and men on benefit shows that while for both sexes activities outside the home were severely curtailed by living on benefits, women, on the whole, were even less likely to participate in such activities than men (Bradshaw and Holmes, 1989).
To take account of the multi-dimensional aspect of poverty, non-monetary or hardship indicators can be used to supplement the income or expenditure values. This approach tries to make a direct assessment of deprivation by collecting data on a certain number of specific fields, for example food, clothing, housing conditions, possession of certain consumer goods, health, education, social contacts and leisure activities. This method was pioneered by Townsend (1979), who defined poverty not simply as a lack of money, but also as exclusion from the customs of society. It was developed and improved by Mack and Lansley (1985): they defined being in poverty as a situation in which people had to live without the things which society as a whole regarded as necessities.
Previous empirical evidence from ECHP data (Ruspini, 1998) showed that lone parents were a particularly vulnerable group when compared to married mothers as they were less likely to have access to consumer assets. It was also quite difficult for them to save, take an annual holiday, replace furniture or invite friends or family round. As Millar (1989) said, the lack of such items suggests that lone mothers may be more socially isolated than two-parent families. The initial economies which are immediately made in a situation of reduced domestic income involve a reduction of social and leisure activities: holidays, hobbies, entertainment. Moreover, the lack of a private means of transportation (and/or of a telephone) drastically limits the possibility of going out and seeing friends or relatives, both for the mother and for her children. The fact that they received benefits from the State did not guarantee an adequate standard of living.
Thus, economic poverty has important negative implications for the lives of poor women, lone mothers and consequently, their children. Living in poverty inevitably restricts the activities in which children can participate: Cohen et al. (1992) documented that poor families could not afford to send their children on school trips or outings with friends. Others said there were few play facilities for children and they had no money to travel further afield. Moreover, coping with little money creates difficulties for relationships within couples and between parents and children (Oppenheim and Harker 1996).
Discussion: towards a gender-sensitive definition of poverty
I would like to conclude with a summary of the methodological reflections presented here, which may be useful when considering the issue of gender issue in poverty research:
- The researcher is required to make very careful methodological choices and should be fully aware of the meaning and impact of these choices on her/his research.
- Concerning the identification of the poor, it is important to integrate alternative research traditions. Given the limitations of the official definitions of poverty, a combination of alternative poverty measures is needed.
- A crucial methodological challenge for women's poverty research is how to open the family 'black box'. There are various possibilities, such as combining different poverty measures, that is, tackling the problem by collecting income data for both households and individuals. As already discussed, the gap between women and men increases if one takes into consideration individual income data. Another possibility concerns the greater availability of variables specifically aimed at depicting the three processes involved in the acquisition and expenditure of resources within the family. Particularly, we should consider ways of accounting for assets and other non-cash resources when measuring poverty.
- Thus, there is a need to design suitable, adequate and comparable data sets for the study of female poverty and its dynamics. New data collection (both quantitatively and qualitatively oriented) may be required in order to address some questions.
ACKNOWLEDGEMENTS
This article is a shorter and revised version of an article submitted to the 'International Journal of Social Research Methodology: Theory and Practice'. Many debts were incurred in the course of writing this article. In particular, I would like to thank Jane Millar, Julia Brannen, Rosalind Edwards and Christina Pantazis for reading various drafts and offering useful suggestions.
NOTES
- There is little evidence suggesting any significant change in men's attitude; women's entry into paid work has not been matched by an increase in the sharing of unpaid work (Lewis, 1993; Sabbadini and Palomba, 1993).
- I am indebted to Hilary Land for this insight.
- It would be possible to use the mean instead of the median. The median has been chosen because it is less affected by the extreme values of the income distribution.
REFERENCES
Alcock, P. (1993), Understanding Poverty, Hampshire and London: MacMillan Press.
Arber, S. (1990), 'Opening the black box: understanding inequalities in women's health', in P. Abbott and G. Payne (eds) New Directions in the Sociology of Health, Brighton: Falmer Press.
Brannen, J. and Wilson, C. (1987), 'Introduction', in J. Brannen and C. Wilson (eds) Give and Take in Families. Studies in Resource Distribution, London: Allen & Unwin, 1-17.
Bradshaw, J. and Holmes, H. (1989), Living on the Edge: a Study of the Living Standards of Families on Benefit in Tyne and Wear, Tyneside: Child Poverty Action Group.
Buhman, B., Rainwater, L., Schmaus, G. and Smeeding, T. (1988), 'Equivalence scales, well-being, inequality and poverty: sensitivity estimates across ten countries using the Luxembourg Income Study (LIS) database', The Review of Income and Wealth, 34 (2), 113-42.
Burkhauser, R.V., Smeeding, T.M. and Merz, J. (1994), Relative inequality and poverty in Germany and the United States using alternative equivalence scales, Luxembourg Income Study Working Paper Series, Working Paper n. 117, Luxembourg: CEPS/INSTEAD.
Cohen, R., Coxall, J., Craig, G. and Sadiq-Sangster, A. (1992), Hardship Britain: Being Poor in the 1990s, London: Child Poverty Action Group.
Daly, M. (1992), 'Europe's poor women? Gender in research on poverty', European Sociological Review, 8 (1), 1-12.
Daly, M. (1995), Sex, gender and poverty in the British and (West) German welfare states, Paper presented at the Conference 'The Cost of Being a Mother, the Cost of Being a Father', European Forum, Florence, European University Institute, March 24-25.
Findlay, J. and Wright, R.E. (1994), Gender, Poverty and Intra-household Distribution of Resources, Discussion Paper n. 913, London: Centre for Economic Policy Research.
Foerster, M.F (1993), Comparing poverty in 13 OECD countries: traditional and synthetic approaches, Luxembourg Income Study Working Paper Series, Working Paper n.. 100, Luxembourg: CEPS/INSTEAD.
Glendinning, C. and Millar, J. (1991), 'Poverty: the forgotten Englishwoman', in M. MacLean and D. Groves (eds) Women's Issues in Social Policy, London: Routledge.
Graham, H. (1987), 'Women's poverty and caring', in C. Glendinning and J. Millar (eds) Women and Poverty in Britain, Hemel Hempstead: Harvester Wheatsheaf.
Land, H. (2000), 'La ricostruzione della dipendenza delle donne', in F. Bimbi, E. Ruspini (eds) Povertà delle donne e trasformazione dei rapporti di genere. Inchiesta, 128, aprile-giugno 2000, 85-90.
Leccardi, C. (1996), 'Rethinking Social Time: Feminist Perspectives', Time and Society, 5 (2): 169-86.
Lee, A. (1999), Income distribution within household and women's poverty. Web page http://www.auckland.ac.nz/apec/papers/LeeA.html
Lewis J. (1993), 'Introduction: Women, Work, Family and Social Policies in Europe', in J. Lewis (ed.) Women and Social Policies in Europe. Work, Family and the State, Aldershot: Edward Elgar: 1-24
Mack, J. and Lansley, S. (1985), Poor Britain, London: George Allen & Unwin Ltd.
Mayer K.U. (1993), Changes in European life courses and their social, political and economic determinants, Paper presented at the Conference Life Course and Social Inequalities in Europe, European Science Foundation and European Consortium for Sociological Research, November 22-26, Schloss Ringberg, Tegernsee.
McKendrick, J. (1998), 'The 'Big' Picture: Quality in the Lives of Lone Parents', in Millar, J. and Ford, R. (eds) Private Lives and Public Responses. Lone Parenthood and Future Policy in the UK, London: Policy Studies Institute), 78-103.
Millar, J. (1989), Poverty and the Lone-Parent Family: The Challenge to Social Policy, Aldershot: Avebury.
Millar, J. (1999), Gender, Poverty and Social Exclusion, Paper presented at the Conference Beyond The Feminization of Poverty, November 29-30, University of Padova.
Millar, J. (2000), 'Genere, povertà e esclusione sociale', in F. Bimbi, E. Ruspini (eds) Povertà delle donne e trasformazione dei rapporti di genere. Inchiesta, 128, aprile-giugno 2000, 9-13.
Millar, J. and Glendinning, C. (1989), 'Gender and poverty', Journal of Social Policy, 18 (3), 363-81.
Millar, J., Webb, S. and Kemp, M. (1997), Combining Work and Welfare, York: Joseph Rowntree Foundation.
Morris L. (1994), Dangerous Classes. The Underclass and Social Citizenship, Routledge: London and New York.
Oppenheim, C. and Harker, L. (1996), Poverty: The Facts. 3rd Edition, London: Child Poverty Action Group.
Pahl, J. (1989), Money and Marriage, London: Macmillan.
Pahl, J. (1995), 'Denaro, potere e accesso alle risorse nell'ambito del matrimonio', Polis, IX (2), 179-96.
Pahl, J. (2000), 'Donne e denaro nell'economia elettronica', in F. Bimbi, E. Ruspini (eds) Povertà delle donne e trasformazione dei rapporti di genere. Inchiesta, 128, aprile-giugno 2000, 14-21.
Payne, S. (1991), Women, Health and Poverty. An Introduction, Harvester Wheatsheaf, Hemel Hempstead, Hertfordshire.
Ruspini, E. (1998), Living on the poverty line: lone mothers in Belgium, Germany, Great Britian, Italy and Sweden, MZES Working Paper, MZES: University of Mannheim.
Ruspini, E. (2000), 'Longitudinal research in the Social Sciences', Social Research Update, Issue 28, Guildford: University of Surrey.
Ruggles, P. (1990), Drawing the Line. Alternative Poverty Measures and their Implications for Public Policy, Washington D.C.: The Urban Institute Press.
Sabbadini, L.L. and Palomba, R. (1993), Differenze di genere e uso del tempo nella vita quotidiana, in: Paci M. (ed.) Le dimensioni della disuguaglianza - Rapporto della Fondazione Cespe sulla disuguaglianza sociale in Italia, Bologna: Il Mulino.
Scheiwe K. (1994), 'Labour Market, Welfare State and Family Institutions: The Links to Mothers' Poverty Risks. A comparison between Belgium, Germany and the United Kingdom', Journal of European Social Policy, 4 (3), 201-24
Sorensen, A and McLanahan, S.S. (1987), 'Married Women's Economic Dependency, 1940-1980', American Journal of Sociology, vol. 93, n. 3, 659-87.
Townsend, P. (1979), Poverty in the United Kingdom: A Survey of Household Resources and Living Standards, Allen Lane and Penguin Books: London.
Ward, C., Dale, A. and Joshi, H. (1996), 'Income dependency within couples', in L. Morris and E. Stina Lyon (eds), Gender Relations in Public and Private, New Research Perspectives, London: MacMillan Press, 95-120.
Elisabetta Ruspini
Dipartimento di Sociologia e Ricerca Sociale
Facoltà di Sociologia, Università di Milano-Bicocca
Via Bicocca degli Arcimboldi, 8
Edificio U7
20126 Milano, Italy
Fax n. 0039-02-64487561
E-Mail: elisabetta.ruspini@unimib.it
