UN/POP/PRA/2000/19
15 August 2000
ENGLISH ONLY
EXPERT GROUP MEETING ON POLICY RESPONSES TO
POPULATION AGEING AND POPULATION DECLINE
Population Division
Department of Economic and Social Affairs
United Nations Secretariat
New York, 16-18 October 2000
LOOKING FOR EUROPEAN DEMOGRAPHY, DESPERATELY? *
Patrick Festy **
*This document was reproduced without formal editing.
**Institut National d’Études Démographiques (INED), France. The views expressed in the paper are those of the author and
do not imply the expression of any opinion on the part of the United Nations Secretariat.
000809UN.doc
L
OOKING FOR
E
UROPEAN DEMOGRAPHY
. D
ESPERATELY
?
Patrick F
ESTY
(Institut National d’Études Démographiques, France)
According to the UN population projections (1998 revision, medium variant), the European
population would have reached a maximum by the end of the 20
th
century and would
continuously decline in the five following decades, at an ever-increasing rate. From 730
million people in 2000, it would have lost 100 millions by 2050, despite a rise in fertility (the
total fertility rate would have gained .35 child per woman, from 1.42 in 1995-2000 to 1.77 in
2045-2050), a prolonged life expectancy at birth (from 73.3 to 80.1 years) and some
immigration. Ageing would be the other major feature, with a decline in the proportion of
children below 15 (from 17.6% in 2000 to 14.4% in 2050) and a rise in the proportion of
elderly people aged 65 or more (from 14.7 to 27.6%).
Of course, these results are correct, i.e. population trends in numbers and age structure are
consistent with the values taken by the components of the demographic movements: fertility,
mortality and migration. The latter are just assumptions, which we will not contest. But in
fact, there were no projections, through assumptions and calculations, made for Europe, only
projections for each national population, then an aggregation that resulted in a European total.
Aggregation of births, deaths or migrants can be seen as the result of European levels of
fertility, mortality and migration applied to European population, but they were not reached
this way. Consequently, we will not question these implicit global assumptions and wander
whether they are reasonably likely or not; instead, we will question the process of aggregation
and wander whether demographic levels in European countries are homogeneous enough for
the sum of them to be considered meaningful. Our analysis of the diversity of demographic
situations in Europe will reveal different configurations of family or health elements, with
various policy implications.
Europe is a large puzzle of 47 countries that would be too difficult to understand if we had no
simplification scheme, ab initio. We will assume that there are two Europe, a western one, the
population number of which the European Union dominates, and an eastern one, which
includes Russia as a major member. A widely shared view is that the demographic regime
which prevails presently in western Europe (the “second demographic transition” model,
maybe) should extend to the eastern part in the future, thus resulting in converging trends
between the two regions. By 1995, there were 728 million people in Europe, of which slightly
more than a half lived in the 15 countries of the European Union (372 million persons).
But the European Union itself is diverse. An opposition is generally made between the North
and the South, with the idea of a possible diffusion from North to South of family patterns and
health systems, likely to have a strong incidence on demography. Four countries dominate EU
by their population numbers: Germany with more than 80 millions in 1995, France, Italy and
the United Kingdom with nearly 60 millions each; we will concentrate most of our analyses
on them, before we contrast EU to the rest of Europe, where Russia (nearly 150 million of
persons) will be given a major role.
In the recent years (1990-1995), total fertility rate was 1.57 child per woman in Europe; life
expectancy at birth was 72.6 years for both sexes. The European Union had a lower fertility
than the rest of Europe (1.50 against 1.64), but a higher life expectancy (76.5 against 68.7).
Inside the European Union, there were marked differences in fertility, with low rates in
Germany and Italy (more generally, in central and southern Europe) and relatively high rates
in France and UK (and in northern Europe). Disparities in EU were high enough for Russia
(and the average of the rest of Europe) to be inserted in the 1.28-1.78 bracket. On the
contrary, life expectancies were very similar in the four western countries and the distance
was large between them and Russia. (Table 1)
T
ABLE
1. F
ERTILITY AND MORTALITY INDICATORS IN
E
UROPEAN REGIONS AND SOME COUNTRIES
European Union Rest of Europe
Total France Germany Italy UK Total Russia
Total fertility rate
(number of births per woman)
1.50 1.72 1.30 1.28 1.78 1.64 1.55
Life expectancy at birth
(in years, for both sexes)
76.5 77.1 76.0 77.2 76.2 68.7 66.5
Source: Population Division, Replacement Migration: Is it A Solution to Declining and Ageing Populations?
United Nations 2000.
There is a global diversity in fertility levels, throughout Europe, with a limited gap between
East and West. There is a clear divide in mortality, with much more favourable levels in the
European Union than in the rest of Europe. From a slightly different perspective, the
European Union is diverse on fertility, but fairly homogeneous on mortality.
A. D
IVERSITIES IN
E
UROPE ON FERTILITY
1. Diversity in the European Union on fertility
Low period fertility indices can partly be explained by tempo effects: fertility is low now,
because people prefer to postpone the births they will have to later (better?) times. An
increase in the age at birth is a signal for this movement, which also results in period rates
being lower than cohort rates. The total number of children to women in birth cohorts can
remain unaffected while postponement results in a lowering of contemporary fertility. A
possible explanation for divergences in period fertility within the European Union could be
linked to differentials in these tempo effects.
It is not really the case. Mean ages at birth are rather similar throughout western Europe. Still
more importantly, the four more populated EU countries share the movement towards later
childbearing; only Italy emerges with an amplified increase in mean age at birth. Cohort
fertility rates, which result in period rates when combined with tempo variations, are nearly as
diverse as period rates have already been showed to be. (Table 2)
Diversity in EU fertility is not just a postponement effect that would explain short-term
movements and levels of period indices. It is a lasting phenomenon, written in cohort histories
and likely to be maintained in the future. Reasonably enough, the UN population projections
have taken some account of that reality.
T
ABLE
2. T
EMPO AND QUANTUM COMPONENTS OF PERIOD FERTILITY RATES
IN COUNTRIES OF THE
E
UROPEAN
U
NION
France Germany Italy UK
a) TFR (birth cohort 1963) (births per woman) 2.02 1.57 1.57 1.89
b) Mean age at birth (birth cohort 1963) (years) 28.0 27.5 28.4 28.0
c) Variation in mean age at birth (bc. 1958-1963) (years) +0.6 +0.6 +1.0 +0.5
d) Expected period TFR [a*(1-c/5)] (births per woman) 1.78 1.38 1.26 1.70
e) Actual TFR (period 1990-1995) (births per woman) 1.72 1.30 1.28 1.78
Sources: UN, cited in table 1;
Conseil de l’Europe, Évolution démographique récente en Europe, 1999.
Low fertility can also be linked to the movement away from marriage, which many western
European countries have experienced for the recent decades. Of course, marriage is no longer
a pre-condition for childbearing in most of these populations, but it remains true that married
couples have a higher fertility than non-married people, even those who live in a “marriage-
like” cohabitation.
This time, the decomposition is very efficient; not because of marriages, the proportion of
which is low but not very different throughout the European Union, but because of the
proportion of children borne from married parents, which is low where total fertility is
relatively high and high where TFR is specially low. When children borne in wedlock are
referred to ever-married women, the fertility index so calculated is much more uniform
throughout EU countries than TFR itself. That index is lower than TFR in France and UK; it
is higher than TFR in Germany and Italy. The range for TFR is .50 child (from 1.28 to 1.78);
that for the marital fertility index is but .11 (from 1.41 to 1.52). (Table 3)
T
ABLE
3. M
ARRIAGE AND MARITAL FERTILITY COMPONENTS OF FERTILITY RATES
IN COUNTRIES OF THE
E
UROPEAN
U
NION
France Germany Italy UK
a) Total fertility rate (period 1990-1995)
(births per woman)
1.72 1.30 1.28 1.78
b) Proportion of ultimately married women
(birth cohort 1963)
.75 .80 .83 .81
c) Proportion of children borne from married parents
(period 1990-1995)
.66 .87 .93 .69
d) Estimated number of “legitimate” births to
ultimately married women [a*c/b]
1.52 1.41 1.43 1.52
Sources: UN, cited in table 1;
Conseil de l’Europe, cited in table 2.
The countries with the highest TFRs, France and UK, are those in which low nuptiality has
been compensated for by the rise of fertility out of marriage. In the countries with very low
period fertility, the decline in marriages has directly been consequential for fertility, because
marriage remains the place for childbearing, without any substitutes for it. We must go on
developing along these lines, to understand trends and diversities in the European Union.
a. Declining marriages
Marriage has declined throughout Western Europe. The Nordic countries, France or UK have
experienced dramatic changes. In Sweden, the proportion of men and women who will
definitely not marry will reach some 40% in recent birth cohorts and women who marry will
not do it before the average age of 27.5 years; in Finland, France or Norway, the proportion of
never married will climb up to 30% and mean age at first marriage for women to 26.5 years.
By contrast, the Mediterranean countries are facing mild transformations: the proportion of
never married women has not yet reached 20% in Italy in the 1960s birth cohorts; still further
from Sweden, Portugal has exhibited only very recently a decline in period marriage rates
which has hardly affected birth cohorts.
1
Between northern and southern Europe, the central part of the region (Austria, Germany or
Switzerland) is in a median position. The global picture is that of a regular gradation, which
evokes the idea of a geographical diffusion of a matrimonial model from North to South.
b. Rising divorces
Divorces have been more and more numerous throughout western Europe for the last three
decades. The 1970s were also a period of changes in family laws which made easier the
access to divorce almost everywhere, through more rapid and simplified procedures. But
comparing the magnitude of juridical and statistical changes, or the tempo of reforms and
numbers, suggests that the evolution of attitudes and behaviours came first and resulted in
legislative action, rather than the other way round.
Like for marriage trends, levels of divorce rates gradually decline from northern to southern
Europe. According to the most recent period rates, 40 to 50% of marriages would be
concluded by a divorce in the Scandinavian countries and UK, 30 to 40% in the central and
western parts of the region and less than 20% in the Mediterranean countries.
c. The spreading of cohabitation
Declining marriages and rising divorces do not mean that couples refuse to live together any
longer. Everywhere in western Europe, informal living arrangements have developed, with
men and women living “like husbands and wives” without being married.
The spreading was exceptionally rapid in Sweden, where the marriage-burst and the divorce-
boom were so sudden. As many as 80% of women borne in the early 1950s had chosen
cohabitation as their first form of living together before the age of 25. The proportion raised
to 90% in the following cohorts, but there already remained little room for any further
evolution. A few years have been enough for cohabitation to become not only predominant
but almost exclusive and for marriage to be left a residual importance.
Trends were more gradual and slightly time-lagged in the rest of the Nordic countries. Then
came the time for western countries like France, Germany or UK. Levels are coming closer
and closer to the Swedish record: 90% of the women in Finland and 70 to 80% of those in the
other countries have chosen cohabitation as their entry gate into conjugal life.
By contrast, the movement has hardly started in the Mediterranean countries. Some increase
has occurred across cohorts, but as few as 10 to 20% of young women have chosen
cohabitation for their first living together experience, in Italy, Portugal or Spain. Marriage
remains heavily predominant in these countries. Maybe convergence towards higher levels is
on its way, but there remains a long way to go.
2
d. Marriage, couple, children
The proportion of couples who conceived a child out of wedlock and who married before the
end of pregnancy, so as to legitimise their child, was high everywhere in western Europe by
the 1960s: about 60%; but trends since then have been divergent. In northern Europe, France
or UK, the proportion has sharply declined: to 10% or less; on the contrary, in southern
Europe, high levels have been maintained, alike the ones in the 1960s: for instance, in Italy
unmarried couples turn to marriage six times out of ten when a pregnancy occurs. In the
between, but very near of the Mediterranean countries, central Europe has experienced a
limited decline in the nuptiality of pregnant unmarried women: in Germany (western Länder)
and Austria, more than half of conceptions out of wedlock result in a marriage before the end
of the pregnancy. The case of Germany is the more so remarkable as an increased proportion
of children borne from unmarried parents are legitimised later: more than 40% against 30%
by the end of the 1970s. The movement in France is in the reversed direction. Three quarters
of children conceived out of marriage are integrated into marriage, before birth or after, in
Germany. It is twice more than in France, while the two countries were at the same high level,
twenty-five years ago.
On this point, there is a clear-cut distinction between the northwestern part of Europe and the
central and Mediterranean countries. It is quite different from what has occurred in nuptiality
trends, but very similar to the fertility divide. Central Europe and southern Europe had already
experienced sharp fertility declines when nuptiality still maintained relatively high levels. In
these countries, marriage remains the highly favoured frame for family building and a very
limited space has been left for fertility out of its traditional legitimate form. No compensatory
mechanism has been able to work between declining births from married parents and rising
births from unmarried ones.
e. Implications
Differential fertility within the European Union does not only result in unequal rates of
population growth and population ageing. It also has social implications linked to the very
significance we have just analysed.
By 2050, population numbers in France and the United Kingdom will still be over their 2000
mark; Germany will have lost 9 million persons and Italy 16 millions. The proportion of
elderly people aged 65 or more will reach 25% in France and UK, 28% in Germany and 35%
in Italy, instead of 16% fifty years earlier (18% in Italy). The rapidly declining and ageing
Italian population makes a large difference with the stabilised numbers and the slowly
increasing proportions in France and UK.
But it could be still more important to note that children in Italy will probably be borne and
raised in marriage, while those in France and UK will have experienced more informal and
more unstable family links, through cohabitation, separation, reconstituted families and so on.
One step further would lead us from “more informal” to “weaker”; but some sociologists
suggest that things evolve the other way round, from “more informal” to “more personal”
then “stronger”. Cohabiting fathers who care to recognise their children will be closer to them
than married fathers who receive their paternal authority automatically from their marital
status. We are unable to add arguments in one direction or another, but the issue is crucial for
the future of our societies, which are based on an intergenerational contract. Parents
contribute to the education of their children with the implicit hope that they will be rewarded
later, when the generation of their children supports them in their older ages. It has been
admitted up to now in Europe that the most effective way to get this exchange working is to
rely mostly on private transfers from adults to their young children and to have public systems
to transfer resources from the active to the inactive population or from the healthy to the
unhealthy people.
Let us take two realistic case studies, which oversimplify the situation various EU countries
will face in the future. Will children take in charge similarly their old-aged dependant fathers,
whether these ones raised them, together with their spouses, or paid a reasonably generous
child support after their divorce? Will they be efficient caregivers for their elderly parents,
whatever their present situation: still married to their only spouse, separated from him (her) or
living with a second (third…) partner?
Answers to these questions are not only important for these individuals but for society as a
whole. If marital status does matter, for better or worse, two countries like France and Italy
will probably be much more different in 2050 than what population numbers tell us. Present
disparities in their fertility trends and levels point to disparities in family systems, which
should exclude aggregation and globalisation.
2. East-West European fertility differentials
Fertility levels in eastern Europe, exemplified by the Russian case, are not very different from
those in the European Union. More precisely, period rates for 1990-1995 or cohort rates for
women borne in 1963 are midway from the relatively high levels in France and UK and the
extremely low levels in Germany and Italy (Table 4).
T
ABLE
4. T
EMPO AND QUANTUM COMPONENTS OF PERIOD FERTILITY RATES
IN
R
USSIA AND COUNTRIES OF THE
E
UROPEAN
U
NION
France Germany Italy UK
Russia
a) TFR (birth cohort 1963) (births per woman) 2.02 1.57 1.57 1.89
1.71
b) Mean age at birth (birth cohort 1963) (years) 28.0 27.5 28.4 28.0
24.7
c) Variation in mean age at birth (bc. 1958-1963) (years) +0.6 +0.6 +1.0 +0.5
-0.6
d) Expected period TFR [a*(1-c/5)] (births per woman) 1.78 1.38 1.26 1.70
1.92
e) Actual TFR (period 1990-1995) (births per woman) 1.72 1.30 1.28 1.78
1.55
Sources: UN, cited in table 1;
Conseil de l’Europe, cited in table 2.
Discrepancies are elsewhere. First, mean age at childbirth is much lower in Russia: an
enormous three-year gap with the western countries, which are pretty homogeneous on this
point. Second, mean ages were still declining in Russia when they had reversed that trend for
a long time in EU countries. Third, cohort rates, variations in mean age at childbirth and
period rates apparently failed to be consistent in Russia, but the expected TFR should have
been compared to earlier actual rates, because of earlier childbirth.
3
Other discrepancies are linked to the marital and non-marital components of total fertility.
The European Union is diverse on that point because of the unequal contribution of unmarried
parents, substantial in France or UK and very limited in Germany or Italy; Russia stands in
the middle but diverges substantially by the high proportion of ever-married persons
(Table 5). That specificity is reinforced by the prevalence of young ages at marriage, which is
consistent with early childbirth.
T
ABLE
5. M
ARRIAGE AND MARITAL FERTILITY COMPONENTS OF FERTILITY RATES
IN COUNTRIES OF THE
E
UROPEAN
U
NION
France Germany Italy UK
Russia
a) Total fertility rate (period 1990-1995)
(births per woman)
1.72 1.30 1.28 1.78
1.55
b) Proportion of ultimately married women
(birth cohort 1963)
.75 .80 .83 .81
.91
c) Proportion of children borne from
married parents (period 1990-1995)
.66 .87 .93 .69
.83
d) Estimated number of “legitimate” births
to ultimately married women [a*c/b]
1.52 1.41 1.43 1.52
1.41
Sources: UN, cited in table 1;
Conseil de l’Europe, cited in table 2.
A. Avdeev, A. Monnier, La nuptialité russe: une complexité méconnue, Population, 4-5, 1999.
More generally, family building in Russia differs from the western models by the coexistence
of three traits:
–marriage is frequent and takes place at young ages. This ancient specificity is shared by all
the populations East of the St Petersburg-Trieste line, already drawn by J. Hajnal. The
political and economic crisis of the 1990s has resulted in a sharp decline in marriage rates, but
it could be a short-term reaction to acute uncertainties. Some people thought that early
marriage was a way for young persons to get some facilities like an independent housing
during the soviet period, but it is an over simplification which severely underestimates the
importance of a deeply rooted practice of young marriage in Russia;
–divorce is frequent too. Half of marriages result in a divorce. By the mid-1960s, after a
limited reform of the judicial procedure, numbers have rocketed to 30 divorces per 100
marriages, at a time when there were 10 to 15 in most western countries. A continuous rise
since that period has put divorce frequency in Russia at one of the highest levels of the world,
with Sweden and the United States;
–the combination of early and frequent marriage with high divorce rate is very unusual. In
western Europe, the rise in divorce has paralleled the decline and delay of marriages. It is
made possible in Russia by the existence of a dense family network, which interferes
continuously with the daily life of couples and families. The cohabitation of different adult
generations in the same household (young couples frequently live with parents or in laws) and
the intergenerational support (child sitting, shared food, financial aid, etc.) are not only the
result of economic difficulties but also the product of old traditions, brought from rural
settings to urban modern places. Cohabitation in the same household helps solving the
housing shortage but it also permits early marriage as a first step towards material
independence; it is also one of the safety nets offered to former spouses after their divorce.
Most importantly, it gives the young generation a stable frame for their socialisation after the
separation of their parents.
Family network is probably the central part of the Russian family system, with important
implications for the future, when the ageing of the society will increase the weight of the
elderly and will put more pressure than ever on families. Private solidarity will find traditional
channels more easily than in most western European countries. The heavy strain presently put
on family networks by the transition to market economy must be watched with great concern.
B. D
IVERSITIES IN
E
UROPE ON MORTALITY
1. Diversity in the European Union on mortality
Contrary to what has been just shown concerning fertility, there is a remarkable homogeneity
in mortality levels throughout the European Union. We have already exhibited life
expectancies at birth in table 1, but even such a: volatile index as infant mortality rate has a
limited range of values: in the four more populated EU countries in 1997, it was as low as 4.7
p. 1000 in France and as high as 5.9 in UK, at the two extremes.
A classical decomposition of life expectancy by age and sex does not introduce major changes
in that preliminary observation. Differentials between men and women are somewhat more
contrasted than average life durations are: from 5.2 years in UK to 7.9 in France. The gap
between the two extremes has increased since the eighties, because male over-mortality
declined more rapidly in UK, but these discrepancies are long lasting and relatively moderate.
EU countries are all characterised by an important male excess mortality. Similarly,
differentials associated with age are limited: for instance the expectation of life at the age of
65 is higher in France than in the other three countries, but the difference never exceeds two
years. On all these aspects, the homogeneity of EU countries cannot be questioned. All of
them have a long survival; in technical terms, their survivorship curve declines so late that it
looks like a rectangle. (Table 6)
T
ABLE
6. A
GE AND SEX COMPONENTS OF LIFE EXPECTANCY AT BIRTH
IN COUNTRIES OF THE
E
UROPEAN
U
NION
(1995)
France Germany Italy UK
Men Women Men Women Men Women Men Women
Life expectancy at birth (years) 73.2 79.7 73.9 81.8 74.6 81.0 74.3 79.5
Men-women differential 6.5 7.9 6.4 5.2
Life expectancy at 65 (years) 14.7 18.4 16.1 20.6 15.5 19.4 14.8 18.3
Men-women differential 3.7 4.5 3.9 3.5
Source: Conseil de l’Europe, cited in table 2.
A long survival has direct consequences on numbers and age structure in any population, but
its impact on the other features of the society are at best indirect. For instance, the distinction
between actives and non-actives is linked with age, and demographers often contrast people in
working age with those out of that bracket, but the divide between active and inactive people
is more pertinent for social policy. The most direct and socially pertinent associate with age
and mortality is health. Mortality is just an indication of a more important issue: how healthy
are the survivors? That is the reason for the increasing popularity of the decomposition of life
expectancy in “good” and “bad” years, the latter ones being costly for individuals and the
society.
The decomposition according to health status gives some contradictory results. A large
majority of life is free of any disability and the average number of these good years are
distributed among the EU countries as life expectancies are: France and Italy are in a slightly
better position than Germany and UK. But disparities between the first and the last countries
are larger (4.0 years between France and Germany instead of 1.5 on e
0
); not only is life
longer, but it is also better in France, with only 15.4 years with some disability against 17.9 in
Germany. Unfortunately for the French people, their bad years are much worse than in
Germany, with 7.9 years of severe disability instead of 5.2. Taken at face value, this last
indicator leads to the following conclusions: the most costly years of life are more numerous
in France and Italy than in Germany and UK, as if longer life expectancies had been gained in
bad physical conditions. Years of severe disability weigh so heavily on health budget that we
could no longer consider all these countries as homogenous. What a pity these data look so
unreliable; they would be so valuable for policy making! (Table 7)
T
ABLEAU
7. H
EALTH COMPONENTS OF LIFE EXPECTANCY AT BIRTH
IN COUNTRIES OF THE
E
UROPEAN
U
NION
(1994)
France Germany Italy UK
Life expectancy at birth (years) 77.7 76.2 77.5 76.6
Disability free life expectancy (years) 62.3 58.3 60.5 59.9
Mean number of years in mild disability 7.5 12.7 11.2 12.7
Mean number of years in severe disability 7.9 5.2 5.8 4.0
Source: J. M. Robine et al., Les espérances de santé dans l’Union Européenne. Analyse des données du Panel
des Ménages de la Communauté Européenne, REVES paper n°320, Octobre 1998
Let us conclude with a result obtained for the Netherlands. By the age of 65, people still have
6.5 disability-free years to live and 10 other years in disability. If the cancer were eliminated,
they would gain disproportionately more bad years than good ones (respectively 1.6 and .6); if
heart diseases were eliminated, they would maintain the balance between bad and good years
(respectively 1.7 and 1.2); if arthritis and back complaints were eliminated, their total life
expectancy would remain unchanged, but they would transform bad years in good ones (an
estimate of .9).
4
Longer life can be a proxy for healthy life for demographers, but it should not
be for politicians.
2. East-West diversity on mortality
The East-West divide on life expectancy is so clear-cut that it hardly deserves any comment.
It is a relatively recent phenomenon, which started in the mid-1960s and which has
continuously amplified since. Some improvement in the Russian situation after 1995 has been
stopped in 1999 (an epidemics of flu?).
In thirty years, the male situation has progressively deteriorated, while the female one
stagnated. It has resulted in an ever-growing excess mortality for men: the differential
between men and women, which was already important in the sixties, is now over 13 years,
twice as much as in France. (Table 8)
T
ABLE
8. A
GE AND SEX COMPONENTS OF LIFE EXPECTANCY AT BIRTH
IN
R
USSIA AND COUNTRIES OF THE
E
UROPEAN
U
NION
(1995)
France UK
Russia
Men Women Men Women
Men Women
Life expectancy at birth (years) 73.2 79.7 74.3 79.5
58.3 71.7
Men-women differential 6.5 5.2
13.4
Life expectancy at 65 (years) 14.7 18.4 14.8 18.3
10.8 14.9
Men-women differential 3.7 3.5
4.1
Source: Conseil de l’Europe, cited in table 2.
Most of the deterioration has concentrated on the adult ages, with a possible improvement at
young ages and a slower than average worsening for the elderly. But the data quality can
probably questioned at these ages. Nevertheless, excess mortality after 65 is not much
different from that in France and UK. The quality of survival at working ages has probably
declined sharply for the last decades.
Causes of death point to two factors responsible for the gap between eastern and western
countries: heart diseases and violence (Table 9). The former played a major role in decennial
trends and differentials. A slow rise in frequency in Russia contrasted with a sustained decline
in France and UK. The advantage of the French population has been gradually increased; UK,
who hardly did better than Russia in 1970, has followed the French example. Costly
techniques and health care for the elderly have been responsible for progress in western
Europe, but could not be afforded in the East.
T
ABLE
9. C
AUSE OF DEATH COMPONENTS OF STANDARDISED MORTALITY RATES
(
P
. 100 000)
IN
R
USSIA AND COUNTRIES OF THE
E
UROPEAN
U
NION
France (1992) UK (1992)
Russia (1993)
Men Women Men Women
Men Women
Infectious diseases 28 11 5 4
33 7
Cancer 360 158 324 210
361 160
Diseases of circulatory system 332 209 554 350
1 292 839
Diseases of respiratory system 86 40 144 82
168 86
Diseases of digestive system 56 31 35 29
63 29
Other diseases 100 74 106 88
83 55
Injury, poisoning, violence 114 51 46 20
400 95
TOTAL 1 075 575 1 214 782
2 401 1 235
Source: V. Shkolnikov, F. Meslé, J. Vallin, La crise sanitaire en Russie, Population, juillet-octobre 1995.
Violent deaths are directly linked to alcoholism. Ups and downs in their numbers coincide
with the anti-alcoholic campaigns. Together with alcohol, the worsening of economic
conditions and the deterioration of the political system are responsible for these trends.
Divergences with EU countries on this point exhibit consequences of more general
differences in ways of living.
There are so deep structural differences between eastern and western Europe in mortality that
a real convergence cannot be expected in the short or even the medium-term. Once more, the
aggregation of populations with so different health conditions and, probably, health
determinants and systems, looks like a statistical artefact. Ageing, which will be equally
important in the eastern and the western parts of Europe, can hardly be considered as a single
issue, when so different answers are likely to be given by societies.
C
ONCLUSION
The European population will probably decline in the next five decades by 100 millions out of
730. The group of people aged 65 or more will grow in numbers and, still more, in
proportions, up to 27.6% in 2050. A slowly rising fertility (from 1.4 to 1.8 children per
woman) and a continuously prolonged life expectancy, up to 80 years, will fuel these
movements. But does that population exist or is it just a statistical aggregate?
There is much diversity in the European family. Intergenerational links, which will have a
crucial role in ageing societies, have become more and more personal in the Nordic countries,
France and UK, while they stayed deeply anchored in marriage in the Mediterranean countries
or embedded in dense networks in eastern Europe, specially in Russia. Not only does that
variety result in different fertility levels but it also has differentiated implications for socio-
economic issues. Private solidarity is stronger, when one moves from personal-type to
marriage-type, then network-type families and the need for social solidarity will be the more
intense when the private ones are the more problematic. If this diversity were to be
maintained along the next fifty years, the tackling of the ageing problem would probably
imply differentiated answers throughout Europe.
On health matters, diversity leaves room to an East-West divide. While most European
countries have experienced a substantial increase in life expectancy since the 1970s, the
eastern countries have stagnated, due to the collapse of their health system and economic
regime. There is some uncertainty in the quality of life gained during that phase of costly
struggle against cancer and heart diseases. The statistical indices should take more
consideration of these aspects, so that politicians be in a better position if a choice is to be
made between more years of survival and better years of life. It is doubtful whether countries
in so different a stage of their health transition would give similar consideration to this topic.
All European countries in the future will share ageing, but the diversity in the components of
demographic change throughout Europe precludes any hastily generalisation. Policy
implications of that diversity will mean national or regional adaptations to long-lasting socio-
economic systems.
E
NDNOTES
1
ODE, Évolution récente de la nuptialité féminine en Europe occidentale, L’Observatoire Démographique
Européen vous informe, n° 7, janvier 1998.
2
M. Macura, Y. Mochizuki Sternberg, J. Lara Garcia, Europe’s fertility and partnership : selected developments
during the last ten years, paper to the FFS Flagship Conference, Brussels, May 2000.
3
Actual TFR in the period four years earlier (1986-1991) is 2.02.
4
W. J. Nusselder et alii, The elimination of selected chronic diseases in a population : The compression and
expansion of mortality, The American Journal of Public Health, 1996, 2.
