Familialism and financial resources in old age. Setting
the scene for the provision of care in Portugal
Alexandra Lopes
London School of Economics
Department of Social Policy. LSE Health and Social Care
Houghton Street. London WC2A 2AE. United Kingdom
Abstract. The paper examines some trends in the living arrangements and in
the living conditions of the Portuguese elderly . The analysis draws on panel
data from the European Community Household Panel and provides a
sociological characterization of the living realities of the Portuguese elderly in
order to ground in empirical evidence the discussion on the interest and
viability of tele-care in the Portuguese context. The main argument of the paper
is that in a society still strongly marked by traits of familialism, all mechanisms
that are based on the idea of the elderly remaining in their homes are of
potential interest in the sense that they can complement and improve the quality
of the care provided by families. The second argument of the paper addresses
the issue of the financial conditions of the Portuguese elderly and raises
awareness for the limitations of any market-based solutions as potentially
triggering further inequalities within the care system.
1 Introduction
Despite the erosion of the traditional modes of functioning of the Portuguese society,
Portugal can be argued to remain a strongly familialist system in the ways it organises
for the provision of welfare. This trait becomes particularly clear if analysed from a
cross-national comparative perspective. By familialism one means a system of
welfare provision where the families and the households are taken as the primary
locus for social aid, but more than that, where families are assumed not to fail when
performing that role.
The labelling of the Portuguese system as a familialist system shapes quite
considerably the debate about the provision of care for the elderly in this country.
While in other European countries the debate on care for the elderly often involves
taking as a starting point the absence of family resources and in fact a majority of
elderly living alone, in Portugal, similarly to other South European countries, that
same debate has to take as a starting point the prevalence of traits of familialism in the
ways elderly people organise their lives, namely the comparatively higher incidence
of cohabitation between generations.
This paper aims at giving a contribution to the debate on the solutions for the
provision of care for the elderly, by analysing the living arrangements and the living
conditions of the Portuguese elderly. The focus is on the presence of strong familialist
Lopes A. (2004).
Familialism and financial resources in old age. Setting the scene for the provision of care in Portugal.
In Proceedings of the 1st International Workshop on Tele-Care and Collaborative Virtual Communities in Elderly Care, pages 119-135
DOI: 10.5220/0002665001190135
Copyright
c
SciTePress
traits in the ways the Portuguese elderly organise their lives. The argument that flows
along the paper is that designing solutions for the provision of care in Portugal has to
take into account those traits. In that sense, it is reasonable to expect that developing
solutions that facilitate the lives of families as welfare providers may tackle one of the
main problems related to care for the elderly in Portugal. Families are still there and
are still a relevant welfare provider. The fact that they are experiencing increasing
difficulties in performing that role should signal the need to invest in policies that
alleviate that burden by promoting simultaneously a better quality of life both for the
elderly and for their carers. The use of new technologies and in particular of long-
distance assistance mechanisms can have a particularly fertile ground for
implementation in the Portuguese context.
This question however has another side: the side of the financial implications of
the solutions to be implemented. It is legitimate to ask who will pay for the
application of home-based technologies to the provision of care for the elderly. What
this paper tries to discuss is the impossibility of putting the financial burden on the
elderly or on the families. By analysing the financial situation of the Portuguese
elderly population and of their households the paper will demonstrate how generalised
is the lack of financial resources to cope with any market-based solution. This has
immediate and obvious implications for the state as welfare provider that can not be
disregarded when debating this topic.
The paper is organised in four parts. In the first part it puts forward the key
arguments underlying the analysis and identifies the core research questions being
addressed. In the second part we find a brief description of the methods and data used
in the analysis. This paper draws on data from the European Community Household
Panel. The third part is the core of the paper and lays down the main findings of the
analysis. It starts with some considerations on the living arrangements of the elderly
followed by some analysis of the engagement of the Portuguese society in caring after
the elderly. Both components of the discussion will support the argument about the
importance of a familialist approach to the provision of care for the elderly in the
Portuguese context. After that it moves on to some analysis of the financial situation
of the Portuguese elderly and their households trying to capture, even if broadly, some
indicators about the availability of financial resources in old age and about the
feasibility of market-based solution within the context of the Portuguese society. The
fourth and last part of the paper brings together some concluding remarks about the
findings highlighting the policy implications of those when debating the introduction
of new mechanisms of provision of care for the elderly in Portugal.
2 Arguments
In a context of increasing demands for caring services and solutions but, at the same
time, of increasing constraints on the financial ability of the states to meet those
needs, it has been argued that the search for alternative mechanisms of care provision
that maximize the permanence of the elder person at home will be one of the most
feasible ways to address the overall issue of care for the elderly in the near future [5].
That will respond not only to the limitations of resources available by developing
solutions that can be potentially cheaper than institutional care, but also to the
120
demands for better quality of care and better quality of life in old age. Among these
alternative mechanisms one can discuss the interest of tele-care solutions.
The interest of a paper on living arrangements and living conditions of the elderly
in a workshop about tele-care is grounded in the fact that if we are to stand for
mechanisms of care that are based on the elderly staying in their homes, we need to
know where they are living, what are their living arrangements and their living
conditions.
Living arrangements are one of the most important dimensions of quality of life
and well being in old age. They are strongly correlated with the availability of family
care, as well as social and economic support, therefore determining the ways the
elderly will tackle old age related needs. It has been widely recognized that in order to
plan care services it is very important to have information on patterns of living
arrangements among the elderly, in order to better assess not only the nature of the
needs of these but also the existing resources to tackle those needs [3] [7].
This paper argues that mechanisms of care provision such as tele-care are
particularly interesting in the Portuguese context because of the persisting traits of
familialism that characterize the living arrangements of the elderly in our society.
Families remain the main source of support in old age and therefore we are still in
time to structure the provision of care to elderly taking the family as the starting point.
If the traditional family support is under constraint and if families are experiencing
more and more difficulties in acting as welfare providers it is largely because they are
not being targeted as welfare providers. The dissemination of mechanisms that can
help families deliver care and that can make it less burdensome to juggle care and
paid work are argued to fit the sociological characteristics of the elderly population in
Portugal.
However, designing solutions for the delivery of care is not dependent solely on
the sociological profile of the living arrangements of the elderly people. It must take
into account other factors, among which the availability of financial resources in old
age. The paper tries to provide some evidence on the disadvantaged financial situation
of the majority of old age people in Portugal and on the limitations that represents in
terms of policy design. Any type of alternative solution to institutionalised care will
necessarily involve state funding based programs. Any other approach would mean
severe equity problems that would in any case respond to market imperatives but fail
to address the main issue of increasing needs for care among the elderly. Provision of
tele-care and other long-distance assisted care mechanisms has to consider the
financial realities of the elderly and has to be thought of from the perspective of cost-
effectiveness when the state is the likely fund provider.
3 Methods and data
The paper uses data from the European Community Household Panel (ECHP). The
ECHP is a longitudinal panel covering the EU population, which commenced in 1994
and follows up its sample members annually. The ECHP has been interrupted after its
8
th
wave. At the time the analysis was carried out there had been released the 5 first
waves of the panel (1994 to 1998). In the first 2 waves the panel did not cover the 15
EU member states (Austria joined in the second wave, in 1995, and Sweden and
121
Finland in 1996). Also, there were some problems of reliability with the data recorded
for Luxembourg. In our research we only consider the national samples of 11
countries: Germany, Denmark, Netherlands, Belgium, France, United Kingdom,
Ireland, Italy, Greece, Spain and Portugal.
The ECHP is the first cross-national panel survey that has been administered in a
comparable way across the EU, to households rather than to individuals. It contains
data on personal characteristics of all the members of each household, at the same
time that retains data for the household as a unit. And these are the main advantages
of the ECHP compared to any other survey available: because it is a household survey
it collects information on all members of respondent households, which is particularly
useful in the analysis of living arrangements. Because the same questions were asked
in each country, the results are directly comparable overcoming the most common
problems of harmonization of data in cross-country analysis. In addition, because it is
a panel, it allows for some reliable analysis of trends over time, namely for the
analysis of changes in living arrangements and of the factors that trigger those
changes. Finally it is a relatively large panel compared to some other datasets. Wave 1
contains information on more than 9000 males and over 12000 females aged 65 or
over.
The paper draws on some descriptive analysis of data from the last wave available
(1998). At the time the paper was written the analysis of the longitudinal dimension of
the panel was in progress and therefore still not available for dissemination. It was
considered that the results to include in the paper should refer to the closest point in
time, therefore focusing on 1998. Some glimpses of dynamic analysis are included in
the paper and when that is the case data from previous waves are recovered.
The findings of the paper are organized in two main parts. The first part focuses on
the characteristics of the living arrangements of the elderly people in Portugal
highlighting the permanence of traits of familialism and setting the scene for the
implementation of policies of care provision from a sociological perspective. The
second part focuses on the analysis of the financial conditions of the elderly and
particularly on some poverty analysis.
The numerical findings include some multivariate analysis with the estimation of
some models for the likelihood of some specific types of living arrangements. This
however responds more to an exercise of synthesis than to any real attempt to model
realities. Along the paper, and for clarification purposes, information about the
variables used for the analysis and the specific approaches developed will be put
forward.
4 Findings
The argument put forward in this paper is two folded. On one hand it is argued that
the permanence of strong traits of familialism in the ways the Portuguese elderly
organise their lives makes it necessary to consider the family as a unit of welfare
provision in any discussion about mechanisms of care. On the other hand it is argued
that both the elderly and their families do not have the financial resources to purchase
those mechanisms if they are implemented on a market-basis provision.
122
The section on findings is therefore organised according to these two axis. Firstly we
will find some evidence about the importance of familialism in the living strategies of
the Portuguese elderly. This is done by analysing their living arrangements but also
by analysing the nature of the engagement in caring after the elderly by the
Portuguese families. Secondly we will find some evidence about the financial
conditions of the elderly and of their households.
The living arrangements of the Portuguese elderly
Where and with whom are the Portuguese elderly living?
The living arrangements of the elderly were classified according to a six categories
typology, which is the result of the combination of theoretical criteria and data
availability. The categories are as follows:
living alone
living with spouse
living with spouse and adult children
living just with adult children
living in complex household with dependent children
living in other complex household without dependent children
Different studies use different typologies, most of the times depending on the
research questions being addressed. As for the first 4 categories they are self-
explanatory. In this typology we would highlight the meaning of the last 2 categories.
First, the term “complex household” was chosen to designate a household structure
that is more complex than the nuclear family, either because involving the
cohabitation of more than 2 generations - which is often the case of the complex
households with dependent children, meaning that the elderly is living with at least
one adult child, his or her spouse and grandchildren – or because involving bonds
among the household members different from parenthood – some examples are
daughters in law or other relatives such as brother/sister.
As for the observed crude differences among countries our attention is drawn
immediately to the South European cluster (to a certain extent followed by Ireland,
but less clearly). It is in these countries that one seems to observe a clear trait of
familialism reproduced in the elderly people’s lives in the form of a relatively high
share of individuals living in some kind of complex extended household. It is also in
these countries that one finds the lowest shares of elderly living alone as well as signs
of another well documented phenomenon in familialist systems which is the late
departure from the parental home by the younger cohorts reflected in a
proportionately higher incidence of adult children/parents cohabitation [2].
Focusing on the Portuguese elderly sub sample, we have tried to identify the
profiles of the elderly according to their living arrangements. The interest of this
analysis is more than merely descriptive. Understanding the characteristics of the
different living arrangements from the perspective of the socio-demographic
characteristics of the elderly should allow for some preliminary considerations on the
potential factors associated to each living arrangement and on the implications of that
in terms of policy design.
123
Table 2 below summarizes data on a set of variables taken as descriptors of the
different living arrangements considered. In the next few paragraphs we will put
forward some considerations on the meanings of those data.
Several studies have shown the strong association between gender and living
arrangements in old age, as well as with the patterns of change in living arrangements
as age progresses once reached old age. One should note that, at least partially, these
features should not be imputed to any specific institutional or cultural milieu in the
Table 1. Cross–sectional frequencies for the living arrangements of the elderly people in 11
ECHP countries, in 1994 and in 1998 (row percentages)
Living arrangements
Countries
Year
Valid
n
Alone
With
spouse
With spouse
and adult
children
With adult
children
In complex
household
1
1994 1626 38.1 52.0 5.7 a) 2.2 Germany
1998 1644 42.2 48.6 3.6 a) 3.9
1994 557 39.4 58.4 a) a) - Denmark
1998 558 48.2 50.0 a) a) a)
1994 1039 37.5 56.4 4.4 a) - Netherlands
1998 984 43.5 52.6 a) a) -
1994 731 41.4 45.6 7.2 a) a) Belgium
1998 740 45.9 43.0 4.2 a) 4.3
1994 1590 33.0 54.3 6.3 3.9 2.4 France
1998 1614 37.7 49.1 4.3 4.0 4.8
1994 1423 40.4 50.5 4.3 3.3 a) UK
1998 1438 45.9 44.1 2.8 3.4 3.7
1994 657 37.1 34.0 15.4 8.4 4.8 Ireland
1998 684 40.6 31.0 9.3 8.4 10.7
1994 2027 27.8 45.5 11.8 4.3 10.6 Italy
1998 2019 31.9 39.3 9.0 5.2 14.7
1994 1336 21.5 46.4 10.4 4.0 17.6 Greece
1998 1372 27.4 41.1 7.0 4.5 20.0
1994 1630 16.7 41.4 19.4 7.1 15.4 Spain
1998 1709 19.4 35.9 15.4 7.7 21.5
1994 1287 21.0 44.7 14.1 5.9 14.1 Portugal
1998 1236 26.9 38.1 9.5 5.1 20.3
Source: ECHP, waves 1 and 5 (1994 and 1998)
Obs. Cases are weighted
Notes:
1)
given the very low frequencies observed in the 2 categories defined for complex
households (with or without dependent children), it was considered that the aggregation of
both categories would bear more significance for the cross-national comparison. If not, the
small numbers observed for those categories would make it impossible to display any data
at all for a large number of countries.
a)
the frequencies observed are below 40 (non weighted cases) (Eurostat regulations on
data presentation determine that we do not present the respective proportions for those
categories)
124
sense that they are often the “natural” consequence of the demographic behaviour of
older cohorts.
It is not surprising to find in that respect that among the elderly living alone a bit
more than 75% are females. This is undoubtedly related to the fact that women have a
higher life expectancy than men and tend to be because of that proportionately more
exposed to widowhood, adding to the fact that women tend to marry older men,
reinforcing the effect of their higher life expectancy.
Although this is a far too well recognised trait of the living arrangements of the
elderly, it is important to highlight it once more from the perspective of the
implications of gendered patterns of living arrangements for policy design. If one
looks at the distribution within each gender category, it should be almost self-evident
the importance of recognising that more than 60% of the male elderly are living with
a spouse, while among females that group hardly reaches 40%. When it comes to
discussing the provision of care for the elderly, it becomes evident that for males the
most likely provider will be a spouse, frequently an elderly herself. As for females,
that likelihood is significantly reduced, while it increases the likelihood of living
alone and therefore the exposure to the risk of absence of an informal carer if in need
and the need to resource to descendents or other kin. We will come back to this issue,
but what the gender divide seems to suggest is that the issue of living arrangements is
Table 2. Individual-based variables describing the living arrangements of the Portuguese
elderly, in 1998 (column percentages)
Living arrangement
Individual-based variables
Alone Couple Couple
with
adult
children
With
adult
children
Complex
household
with dep.
children
Complex
household
without
dep.
children
Male 23.8 57.5 66.7 18.9 30.6 30.7 Gender
Female 76.2 42.5 33.3 81.1 69.4 69.3
65-69 4.9 9.5 15.6 12.2 9.6 9.3
70-74 31.6 45.4 47.5 28.9 36.9 32.6
75-79 31.6 28.5 26.2 20.0 28.0 23.7
80-84 18.9 12.4 7.1 23.3 14.6 16.7
Age
group
85+ 13.1 4.2 3.5 15.6 10.8 17.7
Married - 96.1 99.3 - 29.3 30.7
Divorced/separated 2.7 - - - 1.9 3.3
Widowed 87.1 1.8 - 97.8 66.2 43.3
Marital
status
Never married 9.2 1.9 - - 2.5 22.8
Yes, severely 24.0 25.1 30.5 26.7 33.8 26.4
Yes, to some
extent
25.2 25.1 17.0 23.3 17.8 24.1
Hampered
in daily
activities
No 50.9 49.8 52.5 50.0 48.4 49.5
Yes, after children 1.2 1.5 - 2.2 12.7 -
Yes, after adult - 4.2 9.9 10.0 2.2 7.5
Engaged
in caring
after
someone
No 98.5 94.3 90.1 87.8 85.4 92.0
Source: ECHP, wave 5 (1998)
125
largely an issue for females. Women are far more often confronted with changes in
living arrangements here reflected in a larger variety of living forms than men.
The same way as gender, age seems to be a significant discriminator element of
living arrangements in later life. We would highlight in particular the age distribution
among those living alone. It is very relevant that 95% of those living alone are 70 or
more years old (undoubtedly an effect of their exposure to widowhood), and that a bit
more than 30% are actually over 80. If one accepts that the likelihood of developing
age-related needs increases with age, this group of old people living alone must be a
special target in terms of policy design. However, and from a comparative
perspective, the age distributions of the other living arrangements also show how
relevant is the presence of the very old elderly in the households where extended
family resources are available (around 30% among the 2 types of complex
households).
The gendered divide mentioned above is largely reinforced by age, which shows as
referred before the effect of the demographic behaviour of older cohorts, namely in
terms of male and female life expectancy. The graphs that follow provide a very
intuitive description of this phenomenon.
If for both men and women the likelihood of living alone increases with age, the
growth is more marked among women. In fact, all along the age line males show as
the most frequent living arrangement living with a spouse (with or without adult
children). A further cross-tabulation with marital status would show that even among
those men in other living arrangements, namely in complex households, the share of
individuals that cohabit with a spouse if far greater that among women.
Fig. 1. Living arrangements of Portuguese male elderly by age, in 1998
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
65-69 70-74 75-79 80-84 85+
living in complex household
without dep. children
living in complex household
with dep. children
living with adult children
living in couple with adult
children
living in couple
living alone
Source: ECHP, wave 5 (1998)
126
Fig. 2. Living arrangements of the Portuguese female elderly by age, in 1998
0%
20%
40%
60%
80%
100%
65-69 70-74 75-79 80-84 85+
living in complex household
w ithout dep. children
living in complex household
w ith dep. children
living with adult children
living in couple w ith adult
children
living in couple
living alone
Source: ECHP, wave 5 (1998)
One should also note from the graphs above that the likelihood of living in
complex/extended households increases with age, this for both gender groups. This
increase is however more pronounced among females, again undoubtedly an effect of
their greater exposure to widowhood.
As for marital status variations the single most evident trait to highlight is
obviously the strong correlation between widowhood and living alone. This is the
single most impacting event in old age as has been largely demonstrated [3].
Widowhood often means that the individual changes to living alone status. That is
what our data seem to suggest with more than 87% of those elderly living alone
declaring being widowed. Yet, it seems to be equally significant to highlight that
within the distribution of widowed elderly around 40% are not living alone,
cohabiting with other kin. Around 30% are actually living in some type of complex
extended household.
One of the crucial variables to analyse in this paper is the effect of the existence of
some type of health problem that may impose some limitations in the normal life of
the elder person, in that sense acting as a proxy indicator for the existence of specific
needs for care. One would expect to find a higher proportion of elderly living in some
form of extended household among those who declare some hampering condition
limiting their daily activities. Interestingly enough we do not observe such a clear
pattern in our data. In fact, and in a similar way across all types of living
arrangements, around 50% of the elderly declare having some hampering condition,
either severe or moderate.
However, this does not mean that the distributions found in our data are not
substantially significant. It is our belief that the data may be biased (to what extent we
could not say) by the way the variable is measured. The assessment of the existence
127
of an hampering condition is made on the basis of the self-perception the elder person
has on his/her condition. It is documented how one’s self-perception of health status
is a culturally determined construct [1]. It would be a legitimate hypothesis to rise that
elderly living alone tend to perceive themselves as more frail and in a worse health
condition than those that benefit from some type of family network. In that sense we
could expect to inflate our results in both directions according to the elderly sub-
group we are focusing.
Our interpretation of the trend of no association between the existence of an
hampering condition and the type of living arrangement of the elder person displayed
in our data is that it could also be reflecting the increasing difficulties of families to
cope with the demands falling on them and in that sense leaving severely exposed to
the risk of social exclusion a significant portion of our elderly population. In any case
it is worth highlighting that the majority of people who actually declare some
hampering condition are living with a spouse. This substantiates what other
researchers have been claiming about the couple as the main unit of care provision in
old age.
This idea results reinforced by the analysis of the engagement in caring after other
people among our sample of elderly. From table 2 we would highlight two main
aspects. On one hand it seems quite clear that the extended household tends to work
as the locus for the exchange of care. This means that if it is true that by living with
his kin, the elder person is benefiting, at least in theory, form a support network that
will help dealing with old age related problems, it is also true that in these households
the elderly tends to have himself a role as carer, namely of children. The second
aspect to highlight is precisely the reinforcement of the couple as a care unit.
Table 3 below summarises the relative impact of this set of factors when trying to
understand what are the structuring elements of living arrangements among the
Portuguese elderly. We have estimated 2 relatively simple logistic regression models
for the likelihood of finding a Portuguese elder person living in a specific type of
household. Model 1 estimates the likelihood of living alone, model 2 the likelihood of
living in some type of extended household (with or without children). We have
elected these two types of households as they can be considered the antipodes of a
familialist society.
The first summary measure to highlight is the non-significance of the gender divide
in the likelihood of finding the elder person either living alone or in
complex/extended households. This does not contradict our initial comments on the
gender differences. Those differences exist in absolute numbers. What our model
reinforces is the idea that those absolute differences among men and women do not
have a gender origin and are simply the result of the differentiated incidence of other
socio-demographic phenomena in the two groups (particularly widowhood and life
expectancy).
Contrary to gender, age and marital status do show a very significant impact in
both our models, although in different ways. As for age, younger cohorts of elderly
are more likely to be living alone (corresponding to the age interval where it is more
likely for the bereavement of a spouse to take place). Older cohorts of elderly show
more likely to be living in some type of complex household (one could infer that with
age the likelihood of needing specific care increases therefore the likelihood of
getting that care from the extended family network).
128
Marital status has a strong effect, as expected. We would call the attention to the
effect of widowhood as increasing both the likelihood of living alone and of living in
complex household.
Suffering from an hampering condition does not show any significant effect on the
choice of living arrangement, which should be read along the same lines as what was
argued before about the same topic.
Finally, we observe a significant impact from the caring status variable. It is
important to remember that we are focusing here on the elder person as a carer and
not as a recipient of care. The relevant bit of information to highlight once more
seems to be the exchange of care implied in the cohabitation of the elder person with
the extended family (the granny phenomenon).
Table 3. Summary models to estimate the likelihood of finding an elder person living alone
and living in a complex household in Portugal, in 1998
Living alone Living in complex household Factors associated to
living arrangements in
old age
Coefficient estimates
(t-statistic)
Odds
ratio
Coefficient estimates
(t-statistic)
Odds
ratio
Age group in 1994
65 – 69 (base)
70 – 74 0.761** (2.44) 2.141 -0.116 (0.51) 0.891
75 – 79 0.899*** (2.84) 2.457 -0.117 (0.50) 0.889
80 – 84 0.394 (1.20) 1.483 -0.089 (0.34) 0.915
85 + 0.157 (0.46) 1.170 0.478* (1.76) 1.613
Gender
Male (base)
Female -0.243 (1.36) 0.785 0.162 (1.14) 1.175
Marital status
Married (base)
Widowed 5.857*** (11.33) 349.595 0.973*** (6.53) 2.646
Never married 5.066*** (9.26) 158.488 1.970*** (8.87) 7.167
Separated/divorced 5.407*** (8.28) 222.930 1.610*** (3.76) 5.003
Hampering condition
No (base)
Yes 0.070 (0.48) 0.932 0.068 (0.55) 1.073
Caring status
Not caring (base)
Caring after children -1.645*** (3.23) 0.193 1.256*** (3.88) 3.512
Caring after adult -3.544*** (3.46) 0.029 0.298 (1.04) 1.347
Constant -5.994 - -2.136 -
Source: ECHP, wave 5 (1998)
Notes: * significant at 0.1; ** significant at 0.05; *** significant at 0.01
129
How much are Portuguese families looking after the elderly?
So far we have looked at where do elderly people choose to live (be that a free choice
or a constrained choice). Now we suggest looking at familialism from the perspective
of those that engage in caring after an elder person. In familialist systems the
household appears as the main locus for the provision of social welfare. The question
to ask follows logically and addresses to what extent do Portuguese families and
households reflect that trait of familialism when we focus on taking care of the
elderly.
Table 4 summarises data for the amount and the nature of engagement in care after
an elder person in 11 ECHP countries in 1998.
One could be initially surprised by the absence of any significant overall incidence
of engagement in caring after an elder person among the South European countries
(traditionally labelled as the familialist countries). That should not be the case in the
sense that it is well documented that the rates obtained for this type of broad question
are heavily influenced by concepts of care that vary enormously across countries[4].
The reality of the Northern European countries in particular is significant to illustrate
this phenomenon. Caring and the figure of the carer are well established in the public
arena of these countries, making it more straightforward for their citizens to identify
themselves as carers, even if the amount of care delivered is very reduced and most of
the time only in the form of visits to give some emotional support to an elder person
living alone. In familialist systems, on the contrary, care is usually identified with
Table 4. Descriptive variables on the amount and nature of the engagement in care after elder
people in 11 ECHP countries, in 1998 (percentages within each country)
Intensity of care giving
(% of those engaged in
care)
Location of care (%
of those engaged in
care)
Effects on paid
work (% of those
engaged in care)
Country %
caring
after
an
elderly
Less
than
14
hours
per
week
14
up to
28
hours
per
week
More
than
28
hours
per
week
Cared
after
person
lives in
household
Cared
after
person
lives
elsewhere
Caring
prevents
from
taking
paid
work
Caring
does
not
prevent
from
taking
paid
work
%
females
among
those
engaged
in care
after an
elderly
Germany 3.0 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.
Denmark 6.9 66.9 19.7 13.4 30.0 69.3 15.5 84.5 63.2
Netherlands 7.5 45.9 44.9 9.2 29.9 70.1 15.3 84.7 58.4
Belgium 6.8 68.7 15.6 15.7 30.2 68.7 13.5 86.5 57.1
France 4.1 64.3 22.2 13.5 35.6 63.3 6.8 93.2 63.8
UK 15.4 n.a. n.a. n.a. 31.4 64.4 n.a. n.a. 58.0
Ireland 5.1 31.2 17.5 51.3 56.9 42.6 33.3 66.7 67.6
Italy 6.6 41.6 30.7 27.7 45.7 49.8 15.9 84.1 64.3
Greece 4.1 34.0 46.1 19.8 65.2 32.2 14.8 85.2 78.4
Spain 5.8 17.3 25.9 56.8 66.6 32.3 20.3 79.7 72.8
Portugal 5.1 23.8 23.0 53.2 83.0 15.7 32.7 67.3 85.1
ECHP 6.2 53.5 22.5 24.0 49.2 48.6 18.1 81.9 64.8
Source: ECHP, wave 5 (1998)
Obs. Cases are weighted
Note: n.a. means data were not available for the particular country on the variable
130
more demanding tasks of support in daily activities.
To support our argument we should analyse in detail the data on the nature of the
provision of care in the subsequent columns of table 4. There we will find significant
variations across countries and patterns of care provision that clearly reinforces the
familialist character of the Portuguese society.
Our general remark on this would be that in Portugal caring after an elder person is
definitely a household matter, meaning that engaging in care after an elder person
usually involves cohabitation between the carer and the recipient of care. This is
associated to a proportionately higher intensity of care giving, which in turn seems to
have a stronger impact on the carer’s ability to participate in the labour market.
Although put forward in a very simple way, this is a pattern of care giving that has
lots of implications both for the elder person receiving care and for the household
providing care. The immediate consequence to raise is of a financial nature and has to
do with the income losses caused by the inability of at least one household member to
participate in the labour market as a consequence of his or her engagement in care
giving. Other consequences will cover aspects related to the burden on the carer and
on the entire household and on the emotional consequences the care giving context
often has. There is no research of a kind in Portugal on this topic, but research carried
out in other countries has shown how important it is to include in policy design not
only the needs of the elderly but also the needs of their carers [6]. In that sense any
mechanisms that can somehow alleviate the burdens on the informal family carers
making it more flexible to juggle care with other social roles, namely paid work and
leisure activities are expected to have positive effects in the overall quality of life of
the household engaged in care giving and, if not directly at least indirectly, in the
quality of life of the elder person receiving care.
The last column in table 4 adds up on this by showing how comparatively strong is
the gender bias in the Portuguese pattern of engagement in caring after an elder
person. This, if not surprising within a familialist context, does raise some worries
about the availability of informal carers in the near future. The behaviour of younger
cohorts of females as far as their participation in the labour market is concerned has
been changing significantly, with increasing levels of participation but also, as
surveys on values have been demonstrating, with a increasingly stronger orientation
towards values of self-fulfilment and individual economic independence. The joint
effect of the impact caring has on being able to keep paid work and the decreasing
availability of females to sacrifice their professional lives, may pose serious problems
of availability of carers in the near future.
Financial conditions in old age
The analysis of the financial conditions of the Portuguese elderly sub sample has
made use of rather aggregated information. We work with household income as a
more reliable alternative to personal income. By considering household income we
account for the transfers within the household and therefore get a more accurate
picture of the effective resources available for the elder person. Since the overall goal
of our analysis is to assess, even if in a crude way, the living conditions of the elderly,
the equivalent adult household income appears as the most reliable measurement. In
terms of calculations this means that the summary variable for the household income
131
is transformed according to the number of equivalent adults in the household. To
equalise the income variable we have used the modified OECD scale
1
.
The income distribution problems that characterise the Portuguese population in a
comparative perspective, namely within the EU, are well know and are well
documented. We will not focus on that. The same can be said for the problems of
income distribution within the Portuguese society. Evidence has been put forward on
the inequality problem associated to the income distribution across the Portuguese
society. A good reference for that is the work of Rodrigues, where among other
conclusions, it is demonstrated the comparative weak situation of the elderly [8].
1
The OECD equivalent modified scale for computing the household income per equivalent
adult gives a weight of 1.0 to the first adult in the household, 0.5 to the other adults and 0.3
to each child. By dividing the total household income by the OECD equivalent modified
scale we take into account in the analysis the differences in dimension and composition of
households.
Table 5. Measures of inequality within the income distribution of the elderly Portuguese
sub sample, in 1998
Measures of Inequality
1
Deciles Ratio (P90 / P10) 4.50
Share Ratio (S80 / S20) 5.74
Gini Index 0.37
Source: ECHP, wave 5 (1998); own calculations
1
The measures used are:
Deciles ratio: the ratio between the 90
th
percentile and the 10
th
percentile of the
distribution.
Share ratio: the ratio between the total amount of income of the 20% of the population with
higher incomes and the total amount of income of the 20% of the population with lower
incomes.
Gini index: it is a widely used measure of inequality particularly sensitive to transfers in the
middle of the distribution.
Table 6. Proportion of total equivalent adult income in each decil of the distribution
Decil Proportion of total income Lorenz Curve Coordinates
1
st
0.0328 0.0323
2
nd
0.0464 0.0792
3
rd
0.0537 0.1329
4
th
0.0612 0.1941
5
th
0.0679 0.2620
6
th
0.0770 0.3390
7
th
0.0913 0.4303
8
th
0.1132 0.5435
9
th
0.1534 0.6969
10
th
0.3032 1.0000
Source: ECHP, wave 5 (1998); own calculations
132
In this paper we focus on the distributional aspects within the elderly population.
The reason for this approach is the belief that one of the main challenges when
designing policies for any group is to offer solutions that maximise the equal access
of all to whatever goods are being offered. In the field of care provision this argument
seems more relevant than any other.
Tables 5 and 6 provide for some summary data about the income distribution
inequalities within the elderly sub sample.
The information contained in both tables is self-evident and gives us a clear picture
of the deep inequalities within the elderly population. The Gini index (0.37) and the
Lorenz Curve coordinates are particularly clarifying in that respect and show how
unequal is the income distribution among the elderly, with more than 45% of the total
income concentrated in the 20% better off elderly. This becomes more significant if
we remember that the income variable we are analysing refers to total income after
social transfers. We are talking about the final disposable income of the elderly.
What this picture reinforces is the idea that when targeting elderly people we need
to take into account profoundly distinct segments of elderly. More than that, we need
to keep in mind that any solution that involves a substantial provision of care
solutions by the market will raise serious equity problems in such an unbalanced
income distribution.
Our interest though, and because we are tackling the problem of care provision
from the perspective of a familialist framework, involves some more detailed analysis
that focus on the variations of income across different living arrangements. It is
argued that the choices of the elderly in terms of living arrangements are often
responding to financial constraints, in that sense working many times as strategies to
alleviate poverty. Table 7 below displays summary income data for different types of
living arrangements.
The comparison of both the mean and the median incomes of different types of
living arrangements give us a first idea of how important the family network can be in
the life of the elder person as a mechanism to alleviate poor financial conditions.
Table 7. Mean and median equivalent adult income by type of living arrangement among
the Portuguese elderly, in 1998 (in thousand escudos)
Type of living arrangement Mean income Median income Valid n
Living alone 693.4 504.0 408
Living in couple 858.8 584.2 674
Living in couple with adult children 1009.4 819.0 141
Living with adult children 1034.6 783.6 88
Living in complex households with dep.
children
1026.6 854.8 157
Living in complex households without dep.
children
1012.4 811.2 215
Total elderly sample 875.8 629.2 1716
Source: ECHP, wave 5 (1998); own calculations
133
We would highlight in particular the better off situation of the elderly living in
complex households with dependent children.
If we add to the table above information about the relative position of the elderly in
terms of the national poverty line, the effect of different living arrangements as
poverty alleviators results even more clear.
Having defined the national poverty line as 60% of the national median income,
table 8 above shows how strong the effect of the family resources is for the elderly in
alleviating poverty. One should note in particular the extremely high proportion of
elderly below the poverty live among those living alone and, although less, among
those living in couple. In any case, for both living arrangements we have elderly
people on their own. These are the groups that contribute more to the total amount of
poor. By contrast, those living in extended households seem to benefit from transfers
of family resources that significantly reduce their exposure to poverty.
These are all extremely important elements to take into account when discussing
the potential of any mechanism of care provision. Not only we have to take into
account the average disposable income of families and of the elderly in particular
(signalling clearly the unfeasibility of pure market solutions) but also the differences
among groups of elderly and the risks of introducing further inequalities in the system
if the solutions developed target the small segment of those that are better off.
5 Closing remarks
We have seen that the living arrangements of the Portuguese elderly, and despite
some signs that the traditional models are under increasing pressure, are still of a
markedly familialist nature. This in itself is of major interest when discussing care
solutions based on the elderly staying at home. The majority of the elderly in Portugal
are not living alone, although that is a living arrangement in clear expansion
(therefore to be taken into account in terms of policy design).
The consideration of the people with whom the elderly are living is in that sense a
very important issue when debating the characteristics of the technologies to
Table 8. Proportion of individuals below the national poverty line by type of living
arrangement, in 1998
Type of living arrangement Proportion of poor by
living arrangement
Proportion of the total
population of poor
Living alone 52.7 12.8
Living in couple 36.5 14.6
Living in couple with adult children 22.7 1.9
Living with adult children 26.1 1.4
Living in complex households with dep.
children
13.4 1.2
Living in complex households without dep.
children
20.0 2.6
Total elderly sample 34.7 -
Source: ECHP, wave 5 (1998); own calculations
134
implement and the extent to which they require interaction with the human element.
This human element is very likely to be there in the Portuguese case so it should be
triggered as a resource available. This seems to be equally important from the
perspective of the existing carers, making it worth to discuss the potentially
complementary character of these new technologies to the care provided by the
families.
However, any serious discussion on the advantages of implementing a particular
solution to care after the elderly has to go hand in hand with considerations about the
financial viability of that solution. In this paper we have limited our contribution to
putting forward the existing financial conditions of the Portuguese elderly,
highlighting their limited resources to deal with any market-based solution. The
overall income asymmetries of the national population are clearly reproduced (if not
reinforced) among the elderly sub sample. Debates on policy design must always be
based on equity principles. It is our belief, given the data available, that there is a
clear danger of reinforcing existing inequalities if we are to base the provision of
these new mechanisms of care on market forces.
References
1. Fry, C.L.: Culture, Age and the Infrastructure of Eldercare in Comparative Perspective.
Journal of Family Issues 21, 6 (2000) 751-776
2. Guerrero, T.J., Naldini, M.: Is the South so Different? Italian and Spanish families in
comparative perspective. In: Rhodes, M. (ed.): Southern European Welfare States. Between
Crisis and Reform, Frank Cass, London (1997) 42-66
3. Iacovou, M.: Health, Wealth and Progenity. Explaining the Living Arrangements of Older
European Women. Institute for Social and Economic Research, Colchester (2000)
4. Kendig, H., Hashimoto, A., et al. (eds.): Family Support for the Elderly. The International
Experience. Oxford University Press, Oxford, New York, Tokyo (1992)
5. OECD: Caring for the Frail Elderly People. Policies in Evolution. OECD, Paris (1996)
6. Pickard, L.: Carer Break or Craer-blind? Policies for Informal Care in the UK. Ageing and
Society 20, 6 (2001) 441-458
7. Pickard, L., Wittenberg, R., et al.: Relying on Informal Care in the New Century? Informal
Care for the Elderly People in England to 2031. Ageing and Society 20, 6 (2000) 745-772
8. Rodrigues, C.F.: Income Distribution and Poverty in Portugal (1994/95). INE (available
online at www.ine.pt
)
135