Are Married Women More Deprived Than Their Husbands?*
SARA CANTILLON AND BRIAN NOLAN**
(Received 19.2.96; Accepted 12.6.96)
ABSTRACT
Conventional methods of analysis of poverty assume resources are shared
so that each individual in a household/family has the same standard
of living. This article measures differences between spouses in a large
sample in indicators of deprivation of the type used in recent studies
of poverty at household level. The quite limited overall imbalance in
measured deprivation in favour of husbands suggests that applying such
indicators to individuals will not reveal a substantial reservoir of hidden
poverty among wives in non-poor households, nor much greater depriva-
tion among women than men in poor households. This points to the need
to develop more sensitive indicators of deprivation designed to measure
individual living standards and poverty status, which can fit within the
framework of traditional poverty research using large samples. It also
highlights the need for clarification of the underlying poverty concept.
INTRODUCTION
Conventional methods of analysis of poverty and income inequality take
the household or the narrower family as the income recipient unit, and
assume resources are shared so that each individual in a given house-
hold/family has the same standard of living. Ignoring the within-house-
hold distribution in this way has been increasingly criticised on the basis
that it obscures gender differences in the causes, extent and experience of
poverty, but these criticisms have as yet had little impact on mainstream
poverty measurement practice. Jenkins (1991), in reviewing the case for
opening up the ‘black box’ that is the household and assessing strategies
for doing so, also noted increasing dissatisfaction with the suitability
of money income as the measure of household members’ experiences. He
identified reliance by those investigating the within-household distribution
Jnl Soc. Pol., 27, 2, 151–171 Printed in the United Kingdom 151
© 1998 Cambridge University Press
* We are grateful to participants at seminars at the Economic and Social Research Institute, the
Irish Economics Association Annual Conference 1996, and the HCM Network on Comparative
Social Policy and Taxation Meeting, Cambridge 1996.
** University College, Dublin and The Economic and Social Research Institute, Dublin.
on qualitative studies based on small samples as one reason why many
mainstream poverty researchers relying on secondary analysis of large
household surveys remained unconvinced by their evidence. Here we
respond to these challenges by employing data from a large sample on
non-monetary indicators of deprivation, of the type employed in a num-
ber of recent studies of poverty at household level, directly to measure dif-
ferences between spouses in the extent of deprivation.
Assuming equal living standards within the household in measuring
poverty means that either all members of a given household will be
counted as poor or all will be counted as non-poor, and each member of a
poor household will be assessed as equally poor. Critics argue that the
result is that women’s poverty within households with incomes above the
poverty line remains hidden, as does the extent to which women within
poor households disproportionately suffer the consequences in terms of
reduced consumption (Millar and Glendinning, 1987). The feminist
critique of reliance on the household as recipient unit is of course driven
by a much broader concern about inequality between husband and wife
in access to and control over resources: as Jenkins (1991) puts it, it is not
simply inequality in outcomes but inequality in process which is at issue.
Research on the way money and spending are managed within families
(notably Pahl 1983, 1989; Vogler and Pahl 1994) has focused attention
on differences in power and responsibilities between spouses, on the dif-
ferent allocative systems which operate, and on the distinction between
management and control of resources. Material deprivation is itself only
one aspect of being poor; indeed it need not be central to the way in
which one conceptualises poverty, as we bring out below. However, devel-
oping ways to measure intra-household differences in outcomes in terms
of living standards is an indispensable element in opening up the house-
hold ‘black box’, and the need to do so is demonstrated by recent studies
showing the substantial effects on poverty and income inequality of vary-
ing the assumption about the extent to which resources are shared
within the household (Borooah and McKee, 1994; Davies and Joshi,
1994).
Differences in living standards within the household, like household
resource allocation systems, have for the most part been investigated via
in-depth studies of small numbers of cases (for example Graham, 1987;
Charles and Kerr, 1987), which yield valuable insights but are difficult to
generalise and have had limited impact on mainstream poverty measure-
ment. Vogler and Pahl (1994) are an exception, looking at financial
allocative systems and relating these to reported deprivation for spouses
in a sample of 1,211 couples. However, their primary focus is on allocative
152 Sara Cantillon and Brian Nolan
systems rather than deprivation per se, and it is not possible to relate their
deprivation measure to the non-monetary indicators which have been
employed in mainstream research on poverty at household level.
The use of non-monetary indicators of deprivation in poverty measure-
ment at household level was pioneered by Townsend (1979) and Mack
and Lansley (1985). Other recent studies employing non-monetary
deprivation indicators in measuring poverty include Townsend and
Gordon (1989), Freyman et al. (1991) and Gordon et al. (1995) with
British data, Mayer and Jencks (1988) with US data, Muffels and Vrien
(1991) using Dutch data, and Hallerod (1995) with data for Sweden.
Callan et al. (1993) used Irish data to implement Ringen’s (1987) pro-
posal that both income and deprivation criteria be used to identify house-
holds excluded from society due to lack of resources, and Nolan and
Whelan (1996) use the same data to provide an in-depth analysis of the
relationship between deprivation indicators, household income and
wider resources. Here we are able to use this data, with the type of depri-
vation indicators employed in research on poverty at household level, to
look at intra-household differences: specifically, to measure differences
between spouses in the extent of deprivation being experienced. The
results serve to demonstrate the advantages of seeking direct measures of
individual living standards, rather than trying to infer them from income
or expenditure data. While the indicators of deprivation used at house-
hold level are seen to have limitations for this purpose, this is itself a nec-
essary first step to building bridges between measurement of deprivation
at household and at intra-household levels.
The article is structured as follows. Section 2 describes the data.
Section 3 compares the responses of husbands and wives on whether
they lack a range of possessions and activities. Section 4 uses respon-
dents’ replies as to whether they could not afford or did not want the
items that they lack to develop alternative summary measures of
‘enforced’ deprivation. Section 5 looks at the extent to which differences
in measured deprivation between spouses are related to characteristics
such as family composition, income and social class, and whether the
wife has access to an independent income. Section 6 summarises the con-
clusions and draws out their implications for the way poverty is concep-
tualised and measured.
THE DATA
The data employed were obtained from a specially designed large-scale
household survey carried out throughout Ireland in 1987 by the
Economic and Social Research Institute. The effective response rate was
Are Married Women More Deprived Than Their Husbands? 153
64 per cent, comparable with other large-scale surveys covering similar
sensitive areas, and extensive validation has shown the sample to be rep-
resentative of the population in terms of a range of characteristics such as
the age and sex distribution, labour force status, numbers in receipt of dif-
ferent social security schemes, and the distribution of taxable income.
The survey design, response, reweighting and validation are fully
described in Callan et al. (1989).
The survey obtained information on household composition, demo-
graphic characteristics, labour force status, income by source, and on a
set of indicators of style of living. These indicators of style of living were
designed primarily to complement income in assessing the living stan-
dards/poverty status of households, and the approach developed to using
them for that purpose has been set out in Callan et al. (1993) and
extended in Nolan and Whelan (1996). However, the individual
responses also provide a rare opportunity to look at differences in living
standards between members of a household, and our aim in this article is
to exploit that potential by comparing the responses of spouses. The sur-
vey obtained information on twenty items or activities which were to be
considered as possible indicators of deprivation, listed in Table 1. Some of
these items will be common to all members of a family or household – for
example a fridge or a bath/shower – and will not be of use in comparisons
between spouses, but some do clearly relate to the individual, while oth-
ers are more difficult to categorise as familial versus personal.
Following the approach developed by Mack and Lansley (1985),
respondents were shown a card listing these items/activities and asked:
(1) ‘Which of the things listed you do not have or cannot avail of?’
(2) ‘Of the things you don’t have, which ones would you like to have but
must do without because of lack of money?’; and
(3) ‘Which ones you believe are necessities, that is things that every
household (or person) should be able to have and that nobody should
have to do without.’
Here we confine attention to married persons where both spouses are liv-
ing in the household and both completed the detailed individual ques-
tionnaire without any missing responses on any of the items or the differ-
ent elements of the question, which gives a substantial sample of 1,763
couples.
It will be clear that the items themselves were not chosen with intra-
household differences in living standards and deprivation as the primary
focus, nor was the way the data was collected structured with that issue
154 Sara Cantillon and Brian Nolan
to the forefront. For example, interviewers were not asked to ensure that
each spouse was interviewed alone, or explicitly that respondents focused
on their own situation rather than that of their family for specific items
where this might be in doubt. Small-scale intensive studies have shown
the sensitivity and subtlety required to tease out differences between
spouses in activities and attitudes (Graham 1987; Pahl 1989). However,
the fact that the indicators are for a large nationally representative sam-
ple, embedded in a wealth of other information about the individuals and
households concerned, are offsetting strengths, and our aim is to see
what can be learned about differences between spouses from these types
of indicators employed in poverty research at household level.
DIFFERENCES BETWEEN SPOUSES IN STYLE OF LIVING INDICATORS
Of the twenty items/activities available to us, Table 2 shows that half by
their nature appear unlikely to have much potential as indicators of indi-
vidual rather than familial living standards, whereas the other half do
seem to have some such potential. Allocation of some items is not always
clear-cut a priori: a roast once a week and a meal with meat, chicken or
fish every second day have been counted as potentially personal, for
example, because small-scale studies have suggested that women some-
times limit their own consumption of food, particularly meat, so that the
rest of the family can have more (Delphy and Leonard, 1992), though
Are Married Women More Deprived Than Their Husbands? 155
TABLE 1. Life-style items/activities measured in 1987 ESRI Survey
Item
Refrigerator
Washing machine
Telephone
Car
Colour television
A week’s annual holiday away from home (not staying with relatives)
A dry damp-free dwelling
Heating for the living rooms when it is cold
Central heating in the house
An indoor toilet in the dwelling (not shared with other households)
Bath or shower (not shared with other households)
A meal with meat, chicken or fish every second day
A warm, waterproof overcoat
Two pairs of strong shoes
To be able to save some of one’s income regularly
A daily newspaper
A roast meat joint or its equivalent once a week
A hobby or leisure activity
New, not second-hand, clothes
Presents for friends or family once a year
