OXPOP
Oxford Centre for Population
Studies
Working Paper Series no. 17
Facing the 21st
Century: new
developments, continuing problems.
(European Population Forum, Geneva 2004. A shortened
and revised version of this paper will be published by the
United Nations early in 2005)
David Coleman, Professor of Demography, University of Oxford
Department of Social Policy and Social Work
Barnett House, 32 Wellington Square
Oxford OX1 2ER
+44 (0)1865 270345 tel
+44 (0)1865 270324 fax
david./coleman@socres.ox.ac.uk
http://www.apsoc.ox.ac.uk/oxpop
1
European Population Forum 12 ­ 14 January 2004
Keynote address on Population and Development in Europe during the last decade: an
overview
Facing the 21st
Century. New developments, continuing problems
Introduction
At the end of the last decade of the 20th century, demographic patterns in the UN
ECE area, and the problems and opportunities that they present, have become if
anything even more incoherent than they were around 1990. Some issues remain as
they were then: continuing demographic marginalisation on the world scene,
persistent low fertility, generally high and improving levels of survival, continuing
immigration pressure, persistent demographic divergence (Coleman 2002). There
have also been radical changes, some failures of hopes and expectations, and some
substantial re-alignment of demographic perceptions, definitions and policies
affecting population.
The demographic responses to the collapse of communist regimes at the beginning of
the decade have been profound. International migration now dominates European
population dynamics. The significance of population ageing, long understood by
demographers, has finally burst onto political and public consciousness, provoking
much alarm, despondency and media hype. The conceptual world of demography has
also been transformed. The idea of the `second demographic transition', relating the
rise of individualistic behaviour to the radical retreat of marriage in the face of new
`living arrangements' has promoted research in new lines, facilitated by advances in
the technical analysis of the life-course and in many other areas. In this paper only a
selection of these interesting developments can be considered. There is, for example,
little on mortality, and nothing on the new concepts developed recently to understand
the cross-currents of urbanisation, and counter-urbanisation that are transforming
Europe's social and geographical landscape (e.g. Champion et al. 2003).
Trends and concepts of the decade
The Second Demographic Transition
The concept of the `Second demographic transition' is undoubtedly the theory of the
decade, bidding fair to dominate demographic thinking at the beginning of the new
century as the `first' demographic transition dominated that of the last. This theory
that launched a thousand research projects is described as `the' mainstream concept
among population scholars dealing with demographic change in European societies'
(EAPS 2002 p. 3). Developed by van de Kaa (1987) and Lesthaeghe (1987) as
recently as 1986, it is an ambitious model. It describes and explains the substantial
and unprecedented progress of cohabitation, lone parenthood, childbearing outside
marriage and low fertility observed in many countries since the 1960s and the parallel
retreat from marriage and from traditional norms of sexual restraint. All these
demographic trends have been consolidated during the 1990s (see Kiernan 2002,
Heuveline et al 2003) and as the theory predicts, are increasing almost everywhere in
the developed world, although still at different levels of prevalence (Fig 1-4, T 4).
2
Figure 1 Total First Marriage Rate, groups of European countries 1960 - 2001
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
Scandinavia mean
Southern Europe mean
NW Europe mean
CEE mean
FSU mean
former Yugoslavia mean (ex Bosnia)
Figure 2
Total Divorce Rate trends, groups of European countries 1960 - 2000
0.00
0.10
0.20
0.30
0.40
0.50
0.60
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
Southern Europe
Scandinavia
CEE
Balkans (FR Yugoslavia only)
Western Europe
Source: Council of Europe
3
Percent of persons cohabiting, selected countries, 1996.
Source: Eurobarometer data from Kiernan 2002 table 5.1
0
5
10
15
20
25
30
35
40
45
50
Denmark
Sweden
Finland
France
Netherlands
WestGermany
Belgium
GreatBritain
EastGermany
Luxemburg
Austria
Spain
Portugal
Ireland
Greece
women aged 20-24
men aged 25-29
Figure 3
.
Figure 4
Births outside marriage per 1000 live births, selected countries 1945 -2001.
Source: Council of Europe, National Statistical Yearbooks
0
100
200
300
400
500
600
1945
1948
1951
1954
1957
1960
1963
1966
1969
1972
1975
1978
1981
1984
1987
1990
1993
1996
1999
Austria
Bulgaria
Czech Republic
Denmark
Estonia
FRG
Greece
Italy
Poland
Sweden
During the 1990s the concept developed actively (Lesthaeghe 1995); post-materialist
demography has now evolved further into `post-modern' demography (van de Kaa
2001). But in essence the theory proposes that the new freedom of sexual behaviour,
the diversity of forms of sexual partnership, and the relaxation of traditional norms
and constraints observed in many Western societies since the 1960s, are intimately
related and share common causes. They are held to be irreversible and likely to
become universal in developed societies. The underlying causes are a development of
the socio-economic progress in literacy and income that made possible the first
demographic transition, of which these events are a logical continuation. The new
transition is made possible by parallel trends in further economic growth, intellectual
emancipation through education and the concomitant ease of diffusion of ideas. Its
underlying theory, derived from the work of Maslow (1954) and Inglehart (1990),
posits an emancipation from traditional deferential modes of behaviour once material
needs and anxieties are mostly satisfied though the achievement of prosperity and, in
Europe at least, the personal security offered by the welfare states which that
prosperity sustains.
On this view an educated and liberal-minded population, no longer constrained by
material anxieties, is able to emancipate itself from traditional rules and strictures,
unconstrained by deference to religious authority or parental sanction. More
concerned with `self-realisation' than with duty or obedience to authority whether
parental, civic or divine, individuals choose modes of life suited to their convenience.
Under these circumstances, the single state, cohabitation, lone motherhood and the
avoidance of parenthood are more practical and feasible and become more socially
acceptable. Conduct formerly frowned upon becomes tolerated or `normal'. The
spread of these `post-material ` values in society, measured by questionnaire batteries,
correlated with the spread of secular opinions, unconventional attitudes and toleration
4
of behaviour formerly regarded as deviant, immoral or criminal. These general
empirical social trends have been analysed under various different ideological
umbrellas, not all of them sympathetic (Murray 1990, Fukuyama 1999). The specific
`post-materialist' ideology underlying SDT theory has not been without its critics,
some of whom find little difference between `materialism / post-materialism' and the
conservative / liberal poles of personality, and those who find little retest consistency
or behavioural predictive value in the results (Degraaf and Evans 1996, Marshall
1997, Dennis and Erdos 1992). Others have disputed the complete novelty of the
trends it describes, seeing more continuity with developments in the past (Cliquet
1991).
Numerous empirical studies in western countries support the theory. At national level
there is indeed a syndrome of Second Demographic Transition behaviour. National
populations with a high prevalence of (for example) divorce also tend to have lower
levels of marriage, higher prevalence of cohabitation and of births outside marriage,
and abortion ratios, although the statistical association is not always very strong
(Figure 6). The prevalence or even the possibility of such behaviour is of course
modulated by national government policies on family welfare and on legal provisions
for divorce and abortion, which are far from uniform (Tomka 2003). Most important,
those individuals who score higher on `post-materialist' responses are more likely to
be engaged in unconventional living arrangements such as cohabitation (Lesthaeghe
and Moors 1996) and to `do their own thing' in many other ways. Also, the
incorporation of these questions into the European Values Survey and other enquiries
have shown that the prevalence of these responses increased during the 1990s in many
countries, in step, to put it no more strongly, with the spread of the relevant
demographic behaviour. Not surprisingly, therefore, the trend towards 'post material'
values and attitudes is presented by the votaries of the 'second demographic transition'
as an historically inevitable universal development of irresistible force. Identification
of 'leader countries' which others follow has proved difficult; there does not seem to
be one single trajectory. Liberating forces need not lead to convergence, unless all
agree to be liberated in the same direction.
The second demographic transition and low fertility
However, although this model may be successful in accounting for differences in
living arrangements and other preferences within and between European populations,
is less satisfactory when confronted with low fertility. Low fertility (meaning fertility
well below replacement level) is also claimed to be part of the SDT `syndrome'.
Indeed few things could be more bound up with concepts of traditional duty, or
attended with so much cost and inconvenience, as bearing and caring for children. It
would be reasonable, indeed logically necessary given the underlying theory, for
populations that score highest on post-material ideational responses and which
manifest strongly the other SDT attributes, to have the lowest fertility as well. That is
quite strikingly not the case. Comparing national populations, the relationship
between the patterns and trends of period fertility levels and other `SDT' behaviour
are the reverse of what might have been expected (Figure 8). Populations most
enthusiastic for non-traditional living arrangements within the developed world (NW
Europe and the Neo-Europes) tend to have the highest fertility, where the lowest
might be expected. Populations with very low fertility are typically those where
traditional attitudes towards sexual relationships and living arrangements persist. Thus
all the countries of Southern Europe with the partial exception of Portugal, together
5
with Germany, Japan, Korea and other developed countries outside Europe, have low
or very low levels of divorce, cohabitation and illegitimacy while at the same time
they have the lowest fertility rates in the world: (Bettio and Villa 1996). That
inconsistency will be taken further in a later section which considers fertility.
Figure 5 Correlation of different SDT variables.
Illegitimacy Ratio and Total Divorce Rate, W. European
countries 2000. Source: Council of Europe 2002
y = 0.0048x + 0.2232
R
2
= 0.3801
0.0
0.1
0.2
0.3
0.4
0.5
0.6
5 15 25 35 45 55
Births outside marriage per 100 live births
TotalDivorceRate
Total First Marriage Rate by Total Divorce Rate, W.
European countries, 2000. Source: Council of Europe 2002
y = -0.1848x + 0.6718
R
2
= 0.1047
0.50
0.55
0.60
0.65
0.70
0.75
0.1 0.2 0.3 0.4 0.5 0.6Total Divorce Rate
TotalFirstMarriageRate
There are, however, other impediments to the unconditional acceptance of second
demographic transition theory in its complete form. The `first' demographic transition
dealt with the central demographic concerns of birth and death rates and their
consequences for population size, growth and ageing. Its reverberations are still with
us. Its second incarnation is an altogether lesser affair. The SDT, as argued above,
does not deal with the central demographic issue of birth rates. It has nothing to say
about death rates or with migration or with any large-scale demographic processes.
Instead it is more concerned with micro-sociological events: sexual behaviour and
living arrangements. It may also be asked to what extent a `transition' is expected to
6
be complete and irreversible. The first appears to be both. Even in the `progressive'
populations most enthusiastic for the SDT remain highly heterogeneous. Only in very
few countries are more than half of all births outside marriage and no population has
yet abandoned marriage. In most populations, most marriages are still ended by the
death of one of the partners rather than divorce. By contrast, the first demographic
transition is complete: family size is tightly clustered around an average of two, 90%
of births survive to age 60, and barring migration the end of population growth has
arrived.
Figure 6. Total Fertility Rate and Total Divorce Rate, W European Countries 2000. Source:
Council of Europe 2002.
y = 0.8457x + 1.2794
R2
= 0.277
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Total Divorce Rate
TotalFertilityRate
Total Fertility Rate and Illegitimacy Ratio, W European countries, 2000. Source:
Council of Europe 2002
y = 0.0099x + 1.2991
R2
= 0.4639
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
5 10 15 20 25 30 35 40 45 50 55 60 65 70
Births outside Marriage per 100 live births
TotalFertilityRate
It is certainly true that SDT behaviour is generally increasing even in those
populations that have proved most resistant to change, in Southern Europe and Asia.
But so far it remains a regionally limited phenomenon, still concentrated in its more
developed form North of the Alps though widespread in the English-speaking world
overseas. For whatever reasons, the secular trend in second demographic transition
behaviour was faltering, or even declining in a few populations by the end of the
1990s. Lack of increase in divorce may be simply an unsurprising consequence of the
smaller and more selective part of the population that has married in recent decades.
But the upward trend in total first marriage rates seen in some countries recently
requires more explanation (Figure 7). Little work has been done, however, on
analyzing the tempo and quantum effects of marriage rates and therefore not too much
should be made of this observation. Application of the Bongaarts-Feeney methods to
total first marriage rates (with the usual cautions) has shown that delay in marriage
7
accounts for the greater part of the apparent reduction in propensity to marry in the
middle 1990s. For example in Bulgaria, Hungary and the Czech and Slovak
Republics, adjustment returns the TFMR from about 0.6 to about 0.8 although the
final value for Bulgaria returns to a low level (Philipov 2003 pp 108 ­ 109).
Figure 7.
Total First Marriage Rate, selected countries 1960-2001
1945
1947
1949
1951
1953
1955
1957
1959
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
Denmark f Finland f Germany f France f Iceland f Spain f UK f
1.20
1.00
0.80
0.60
0.40
0.20
0.00
Those third-world populations which have reduced vital rates in a spectacular manner,
have so far shunned cohabitation and births outside marriage. Advocates of the SDT
idea present this marked international diversity, for example in cohabitation and births
outside marriage (Figures 5, 6) in somewhat Hegelian fashion, as evidence of
populations inexorably proceeding to a common destination, only at different speeds.
They may be right, but this `transition' may stall half-complete, much as did the
earlier Reformation which made little progress beyond the Alps. Some populations in
the Western world have remained notably resistant to it, particularly Asian
immigrants and most of all Muslims. Arranged marriage with spouses from the
countries of origin is prevalent in many of those immigrant populations. This is, in
part, because even co-religionists who have been brought up in the `enlightened' and
emancipated West are not considered to be suitable marriage partners lest they have
imbibed some of the values under discussion here. In fact the younger generation of
Muslims in Europe is showing signs of reverting to a more traditional Islam,
demonstrated publicly through the wearing of headscarves and other outward signs of
ethnic solidarity. That is not the way of the Second Demographic Transition.
Transitions must, presumably, be sustainable. The underlying theory of the SDT
posits radical ideational change made possible by economic progress. Are the
ideational insights, once attained, irreversible irrespective of the standards of material
security which made their realisation possible on a large scale? Wealth emancipates
populations from anxieties about material needs and, in Europe, supports the welfare
states and social housing policies on which choices of living arrangements at least
partly depend. Or a variety of those welfare programmes have already been checked
or reversed in many western societies from Sweden to New Zealand. High levels of
divorce and lone parenthood transfer some of the costs of the consumption of women
8
and the production of children to the general taxpayer, They may not be affordable in
the long run. In the UK for example, divorce adds about 15% to the UK benefit bill.
In the UK divorce creates three households for every two that existed before, and it is
the quickest route out of owner-occupation into state subsidized `social' housing
(Holmans et al. 1987).
Can modern economies afford the long-term costs of the second demographic
transition as well as the unavoidable and permanent drag on economic growth
presented by population ageing? Furthermore, lone parenthood tends to inflict
psychosocial as well as material handicaps upon children brought up in fragmented or
unconventional households, compared with those in intact families ­ specifically in
respect of school performance, discipline and subsequent parenting (Kiernan 1992,
Ermisch and Francesconi 2001a and b, Osborne et al. 2003). This, however, may be a
special feature peculiar to the circumstances of the UK and the US, where a higher
proportion of children in non-marital households are brought up with only one parent
than is the case in Europe. The age of entitlement may only temporarily have
insulated people from the consequences of their reproductive actions and thereby only
transiently permitted a wider spectrum of behaviour.
The empirically observable demographic trends charted above are, of course,
indisputable. It is less clear, however, that they can all be swept up as evidence for the
diffusion of ideas and attitudes specified by the SDT theory. A plurality of
explanations may be more suitable, depending on circumstances. For example, is the
continued high level of teenage childbearing, mostly to mothers without partners,
observed since the 1970s in the UK and the US a component of the enlightened and
self-realising behaviour envisaged by the SDT? Or is it instead, as both those
government believe, a pathological manifestation, harmful to the interests of mothers
and children, against which policy measures are appropriate? More generally, is the
ideational change theory sufficient for a full explanation of the upward trends in
divorce, cohabitation and the rest? Less ambitious, more conventional economic
theories based on rational choice, not necessary contradictory to SDT theory, have
also pointed out the drawbacks of conventional marital unions and the advantages of
ambiguity when women are financially independent of men through their own work
and as well, if not better, educated (Ermisch 1991, Ermisch and Francesconi 2000).
The Second Demographic Transition in Central and Eastern Europe
The recent trends in Central and Eastern Europe, and in the European republics of he
former Soviet Union, provide another example where a plurality of explanations may
be needed to account for diverse situations. A persuasive case is made out for the
diffusion of new modes of behaviour ­ rising cohabitation, divorce and births outside
marriage ­ in many of these populations during the 1980s, before the collapse of
communism, as part of a process of modernisation. This argument is particularly
convincing in relation to the more prosperous parts of the region (Sobotka et al. 2003,
Lesthaeghe and Surkyn 2002). However this view may encounter more difficulties
when confronted with some of the material realities of the less prosperous sectors of
the former Communist countries. First, in general it is difficult to see how post-
materialist sensibilities, normally regarded as requiring for their nurture a secure
material situation, could flourish in the serious economic downturn and heightened
employment and political insecurity of the post-communist period. Furthermore, in
some senses the restricted choices available in communist times went hand in hand
9
with a high level of certainty and assurance about crucial life events ­ guaranteed
education and employment, for example (Philipov 2003 p. 156). Accordingly it has
been suggested that communist society, with its certainties, might have been a more
fruitful environment for post-materialist views than was its aftermath (Kyveldis
2001).
Elsewhere, however, especially a more local level, apparently `classical' SDT
behaviour (e.g. high levels of births outside marriage) might be due not to individual
empowerment but to quite different, and more socially pathological, developments,
related to the prevalence of `anomie' and disorganisation among the poorer elements
of the population distressed and unsettled by recent changes.. The highly
discontinuous rapid increase of births outside marriage in Bulgaria and Romania, after
over two decades of negligible change, are particularly noteworthy (Figure 4). These
populations were among the poorest of the CEE countries under communism, are still
substantially rural and have so far failed to make effective economic or political
transitions, remaining in a weak economic position (slund 2001). Yes despite this
unpromising theoretical environment for SDT, their proportions of births outside
marriage exceed those in more prosperous, more Westernised countries in Central
Europe. Furthermore demographic changes of this type are apparent even in the most
geographically isolated and socially marginalised populations in those countries -
Turks and Gypsies ­ who are perhaps the least obvious candidates for progressive
ideational change.
Finally, it is necessary to account for the considerable heterogeneity of the changes in
SDT-like demographic processes during the communist period. For example, in
Estonia even in the 1960s most of these innovations were becoming prominent:
cohabitation ranked third in Europe; births outside marriage, early sexual experience
were common in a society where female education was surpassing that of males,
where female work participation rates were high (in the socialist fashion) and attitudes
were becoming more individualistic. Nonetheless the fertility level did not decline but
increased almost to replacement rate (Katus 2003).
Methodological innovations in the 1990s
Analysis of this increasingly diverse behaviour and its correlates has been greatly
enhanced during the 1990s by methodological innovations. The most prominent is
event history analysis, a set of formal techniques for integrating information about life
events described as `the most important new paradigm in demography (Willekens
1999), shifting attention from macro to micro analysis and from structure to process.
As Willekens points out, life events: leaving the parental home, sex, cohabitation,
childbirth, employment, migration are milestones that dominate the concerns and
actions of individuals. Using a statistical basis of proportional hazard techniques,
event history analysis uses non-linear models to integrate and compare these events
and histories, and relate them to individual circumstances. It permits more realistic
analyses beyond the reach of even the most sophisticated classical macro-approaches
using multi-state life tables (Courgeau 2002). As noted below, however, such
concentration on individual-level processes risks loosing sight of the traditional large
scale concerns of demography which are just as important as ever. However, multi-
state modelling (Goldstein 1995), scarcely known in the 1980s but which has become
much more widely used in demography in the 1990s, has enabled individual level data
and aggregate `ecological' variables (which may have no equivalent at the individual
10
level) to be incorporated into the same statistical analyses (see e.g. Lievens 1998).
These techniques have enabled the true pattern of demographic events to be
determined, independently of other influences, sometimes revealing that the true
picture is quite the reverse of that which might be supposed by inferring individual
results from aggregated data.
At the end of the day, event-history analysis cannot of itself be the final demographic
paradigm. Its function is to clarify sequences and relationships between life events. It
only provides a means for analysing their causes in terms of sequences of events; it
provides no mechanism for the analysis of ultimate causes or consequences of the
events in terms of independent variables. That must surely be regarded as the higher
aim of demography. In short, in the broad sense it is not a demographic theory. The
marked enthusiasm for event-history analysis in some circles seems to have been at
the expense of analysis of larger-scale analysis of, for example, the effects of
population change on the environment or on individual behaviour. That risks
confirming the accusations that demography as an `object science' (Dykstra and van
Wissen 1999) is a theory free subject, being more concerned with description and
technical analysis, albeit now at a the highest possible level, than with explanation.
Event history analysis presents the raw material for theory to work on; it provides no
explanation of its own.
Brief mention should be made of some valuable new sources of data intended to
analyse new demographic developments in the light of second demographic transition
ideas, providing retrospective life-histories suitable for event-history techniques.
These are the set of standardised international enquiries into partnership and
reproductive behaviour promoted by the Family and Fertility Surveys of the UN ECE
mounted in 24 countries in the early and mid 1990s (Cliquet 2002), now to be
followed by a new round of `Gender and Generation' surveys adopting a developed
theoretical perspective.
The future of fertility and the end of growth
Sub-replacement period fertility became universal in Europe in the 1990s, with the
possible exceptions of Iceland, Albania and Macedonia. It was first achieved in some
western countries in the 1930s, a point often overlooked. But in the post-war period,
after the `baby boom', below replacement fertility was resumed in Western countries
in the early 1970s and has now persisted for a generation. Greece, Portugal and Spain
joined this group with extraordinary synchrony and speed a decade later, falling to
low levels not surpassed until the post-communist crisis of Central and Eastern
Europe and the former Soviet Union took period fertility to the lowest levels yet
recorded. The conventional - but far from perfect- measure used here is the period
Total Fertility Rate (TFR): the sum of the age-specific fertility rates in a given
calendar year. That measure indicates the average family size per woman, were those
age-specific rates to persist over the real fertile lifespan from age 15-49, or, as the late
Gérard Calot would have preferred to emphasise, the weighted propensity of recent
cohorts of women to replace themselves in the population. The European countries
fall naturally into a small number of groups through a consideration of their
geography, culture, language and demographic behaviour (Grasland 1990, Pinnelli et
al. 2002). In 2001 the mean TFR of the four Southern countries had fallen to 1.32, that
of the 6 CEE countries 1.24, and that of the European Republics of the former Soviet
Union (FSU) to 1.22 (Figure 8).
11
Figure 8 TFR trends, groups of developed countries
1.0
1.5
2.0
2.5
3.0
3.5
4.0 1950
1952
1954
1956
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
Southern unweighted mean
FSU unweighted mean (ex Mol)
CEE unweighted mean
Northern Europe
Western Europe
USA
These fertility patterns not only differ in their overall average level but also in their
components of childlessness and parity distribution ­ there are several routes to low
and very low birth rates. In NW Europe below-average family size has been driven
down partly by high levels of childlessness ­ 20% is normal and in Germany up to a
quarter of women reach age fifty without children. France is an exception. At the
same time quite a few women ­ up to 34% - in France and Northern Europe still go on
to have third or even fourth children (Table 1),
Table 1 Distribution of women born around 1960 by children
ever-born (percent)
Children ever-born
0 1 2 3 or more total
Russian Federation 8 30 44 18 100
Sweden 14 16 40 30 100
Italy 15 25 42 18 100
Germany 26 25 30 19 100
France 14 20 32 34 100
United States 15 19 32 34 100
Japan 3 44 40 13 100
Source UNPD 2003 table 28 p 83 (mostly from Eurostat)
even though childbearing starts particularly late. In Southern Europe by contrast,
childlessness until recently remained relatively uncommon but family size seldom
exceeds two children. In Italy and Spain, however childlessness is increasing to nearly
20%, without the compensation of higher parity births so that fertility at all parities is
12
low. In Eastern Europe most of the populations are in retreat from the former pattern
of near-universal and early childbearing, with families completed by age 30 (UNPD
2003). In Russia especially, recent rates suggest that 90% of women will have a child
but only 75% will have any others; according to Avdeev (2001), the Russian one-
child family is here to stay.
These `lowest- low' levels, defined variously as period levels below 1.3 (Billari et al.
2003) or 1.5 (Caldwell et al. 2003) seem now to have reached their nadir. Average
fertility in two of these `lowest-low' fertility regions ceased to decline around 1996,
although that in the 6 CEE countries is still drifting down a little, thanks to continued
small falls in two of the Catholic countries of the region in which decline was late
(Poland and Slovakia). In the national championships for the lowest fertility ever,
Bulgaria in 1999 and Latvia in 1998 share first prize (both 1.09) unless the example of
the former German Democratic Republic be admitted (0.77 in 1994).. Lowest-low
fertility is even more widespread in smaller (but still very large) sub-national
populations, below the national level. Large areas of Northern Italy and Spain, for
example ­ exceeding in population size many of the countries of Western Europe -
have had period TFRs less than 1.0 for most of the 1990s.
It is no longer true that Western countries have a declining birth rate
The Western part of the continent has reached an apparent fertility equilibrium. Birth
rates generally ceased to decline over 20 years ago, at a somewhat higher level than
elsewhere in Europe. Up to 2001 the TFR remained about 1.75 in Northern Europe,
about 1.6 in Western Europe. At the least a superficial stability, and during the 1990s
a slight tendency to increase, have been the typical trends (Figure 9).
Figure 9.
TFR trends, industrial higher-fertility countries 1945-2001
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
1945
1947
1949
1951
1953
1955
1957
1959
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
Denmark France Netherlands NZ USA
Our conventional period measure of fertility, however, is vulnerable to shifts in the
timing of births. As is well known, when births are delayed, the period measure will
be deflated if the same number of births become distributed over a larger number of
years. Completed family size can remain as it was, as long as births postponed earlier
are produced later in life. Whether they will be is the big unanswered question. Delay
13
in the production of the first birth has been strong and universal in Western Europe
since the 1960s, and has become marked in the CEE and FSU countries since the mid-
1990s. Recuperation is less certain.
The completed family sizes of women approaching the end of their childbearing
career, at age 40, have also moved steadily downwards from their post-war peak. In
the most extreme case, the last cohort of Germans to replace themselves in the
population was born around 1933. However the picture today, dominated by women
in their peak childbearing years ten to fifteen years ago, is somewhat different. In 13
out of 38 European countries for which data are available, women born around 1960
had produced at least 2 children on average by 2001, that is by the time they passed
age 40. Five were in Western Europe; Iceland, Ireland, France, Norway and Sweden.
Twelve years previously, in 1988, the picture was one of continual decline in all these
countries except Sweden. Now a more diverse pattern of cohort fertility has emerged.
In Denmark, Norway and France completed family size has remained at about 2, even
increasing slightly over the decade. The low German average has stabilised at just
under 1.7, about to be joined, and possibly undershot, by the averages of Spain, Italy
and Austria where decline up to 2001 has continued apace to well below two children
(Frejka and Calot 2001), (Figure 10).
Figure 10 Completed Family Size, selected countries, women born 1960.
Spain
Sw
Average Completed Family Size in 2001, women born 1930 - 1960. Source: Council of Europe.
1.5
1.7
1.9
2.1
2.3
2.5
2.7
1930
1932
1934
1936
1938
1940
1942
1944
1946
1948
1950
1952
1954
1956
1958
1960
Austria
Denmark
France
Germany
Italy
Norway
eden
These data on completed family size at 2001 reflect the social and economic
environment of the 1980s and early 1990s, when these women were producing most
of their children. The crucial question relates to the intentions, and the behaviour, of
women now in the late 20s and early 30s whose families have yet to be completed.
Continued delay in childbearing makes recovery to replacement rate increasingly
unlikely in the Southern populations, at least for cohorts now in their 30s which have
so much to recuperate and so little time to do it (Frejka and Calot 2001, Lesthaeghe
2001). Recuperation to sub-replacement levels much closer to two is more plausible
in Northern populations and especially in the Netherlands. The Dutch (with the
14
Japanese) are the oldest mothers of first babies in the world, but (unlike the Japanese)
they also have one of the highest levels of fertility over age 30.
1960 ­ a bad base for comparison
Before continuing it is desirable to put these data into longer-term perspective. In
most analysis of period fertility and mean age at first birth, data series start in 1960 ­
partly because this is the first year of the very convenient time-series produced
according to the Calot method by Eurostat (2002) and by the Council of Europe
(2002) in their annual publications. But if one were starting to analyse Western
demographic trends, one should not be starting from there. The 1960s were the peak
of the post-war baby-boom, with the highest post-war TFRs since the 1920s and the
lowest mean age at marriage and of first birth for centuries. In Western populations,
mean age at marriage and childbirth was traditionally high, well into the 1930s. The
transient baby-boom peak of period fertility was also much higher than the TFR in the
1950s and the 1930s in many countries. The 1930s mark the true end of the first
demographic transition in that it saw fertility fall to replacement levels ­ and in many
cases below replacement levels ­ for the first time since the beginning of the
transition. This decline should not be attributed just to the depression years: it was a
linear continuation of previous trends. Fertility fell to low levels even in those parts of
Europe which were experiencing low unemployment, high levels of economic growth
and new investment, such as the South-East region of England. Mortality rates were
of course somewhat higher then; so that replacement fertility was a little higher than
2.1. Table 2 below shows TFR around 1930, 1950 and 2001 together with the net
reproduction rate, a period indicator of the number of surviving daughters produced
by each women and by extension of the implied rate of population replacement.
Table 2 TFR and Net Reproduction Rates, selected countries 1930 to 2000
1930s 1950 2000
TFR NRR TFR NRR TFR NRR
1935 England and Wales 1.78 0.76 2.18 1.02 1.65 0.79
1935 Scotland 2.17 0.91 1.48 0.71
1932/4 Australia 2.15 0.96 3.06 1.48 1.70 0.82
1933 New Zealand 2.16 0.98 3.38 1.63 2.01 1.96
1931 Canada 3.19 1.32 3.46 1.74 1.48 0.71
1933 United States 2.14 0.94 3.03 1.60 2.14 1.05
1934 Sweden 1.67 0.75 2.32 1.04 1.54 0.75
1933 Germany 1.64 0.91 2.09 0.85 1.38 0.69
1935 France 2.06 0.87 2.93 1.26 1.89 0.91
1930/2 Italy 3.29 1.24 2.47 1.09 1.24 0.58
1926/7 Ukraine 5.09 1.68 2.81 1.29 1.10 0.88
1929/31 Japan 4.76 1.54 3.65 1.19 1.36 0.65
Note: NRR for `1950' is actually for period 1950-55 (from UN). Sources:
Glass and Blacker 1938 Table V, Council of Europe 2002 Table 3.6 (NRR)
UN World Population Prospects 2002 Vol. 1, National Statistical Yearbooks.
Needless to say there are many important differences between the low fertility of the
1930 s and that of the present day. Unlike today there was no long history of low
period or cohort fertility, so the age-structure was still youthful and could generate
many years of natural increase even without replacement fertility. Today, that positive
15
momentum is nearly exhausted in many European countries, especially those in the
East which did not experience a `baby boom'.
Understanding low fertility
Trying to understand low fertility has preoccupied Western demographers for some
time. It used to be assumed, for no very good reason, that fertility would stabilise at
replacement level at the end of the first demographic transition. That has not
happened, and recent conclaves of demographic Magi could find no reason why it
should (Demeny 1997, UNPD 1997, UNPD 2002). So what is going to happen next?
Have European populations come to rest at a new equilibrium and regime of fertility,
no longer declining to zero, but stabilised as levels below, or very substantially below,
the replacement level? Might it resume its decline, given that economic theories of
fertility predict no lower limit in modern circumstances (Namboodiri and Wei 1997).
Or can fertility recover, even to the replacement level? The social and economic
implications of continued decline are obvious. The implications even of stabilisation
at current levels are very substantial.
The first problem is to decide exactly what is going on. As we all know, current
fertility rates are not quite what they seem. These single-number period indices
condense information from the multiple processes underlying fertility: the timing of
its onset, the intervals between births, the chances of going from one birth to the next
or remaining childless, the final completed family size. Low fertility may have come
of age in the 1990s but its study is in a state of rapid flux, with unprecedented
developments in technical analysis in the last decade leading to somewhat divergent
interpretations of the simple data presented above. How to interpret today's current
low levels of births, and what they imply for the future, is by no means agreed.
The single figure measures naturally tell us nothing about the pace of childbearing or
the distribution of births, or of childlessness, among the women in the populations
being measured. It is well known that low period fertility has been accompanied and
to a considerable but disputed degree is a consequence of substantial delays in the
onset of childbearing in all European populations, as in almost all other parts of the
world in recent decades. At the same time more women have remained childless and
larger families, of three children or more, have become less numerous. The average
family size of women who have completed their childbearing has fallen everywhere to
2 or in some cases substantially less. The rapid onset of delay in childbearing was
most marked in Southern and especially in Eastern Europe and the FSU. There,
childbearing traditionally started much earlier than in Western Europe. Hence the
scope for delay, and for consequent deflation of period measures of fertility and
marriage, has been much greater.
It is not yet clear where the process will end, and how far it is `postponement' of
births which may eventually be delivered, or a reduction in the quantum of children
ever to be born. Delay in childbearing continues in most countries (Figure 7),
although in some the pace is slowing down (France, Netherlands). So substantial has
been the delay in Italy and Spain that they have overtaken most NW European
countries to become, together with Japan, the oldest mothers in the world. The scope
for this effect in the CEE and FSU is also much greater than elsewhere. There, until
the late 1980s, marriage and (near-universal) childbearing traditionally began much
16
earlier than in Western Europe, so the scope for tempo effects as well as quantum
changes is much greater ( Kohler et al. 2002, and this conference).
Index wars
Much technical progress has been made in addressing this problem of measurement,
greatly aided by the advances in computing power in the 1990s. The essential problem
is to see how much of the decline in period measures such as the TFR is due to
postponement of births and how much to a reduction in quantum. If much of the fall
can be assigned to a delay in timing then the eventual consequences for population
decline and again may be considerably less than the current figure suggests. There is
little agreement as to how this can be done, and even indeed if it is meaningful to do
so at all rather than using real cohort measures. It is impossible here to go into the
minutiae of these index wars. Very briefly, one approach is to calculate current parity-
progression ratios (the probability of progressing from a given number of births to the
next one), on a period basis. Summing these ratios gives a parity-based estimate of
TFR free of the tempo distortions that arise when age-specific rates are employed.
But all that requires additional assumptions of various degrees of plausibility, and
usually additional data on duration of time since the previous birth. Such data are
seldom available (Ni Bhrolchain 1987, Feeney and Yu 1987, Murphy and Berrington
1993, Kohler, Billari and Ortega 2002).
An alternative approach, which makes fewer demands on data, is to adjust the period
age-based TFR by measuring the delay in timing of births by birth-order as precisely
as possible and to re-assign the delayed births to the calendar year, and age of women,
in which they would have occurred had the delay not arisen (Bongaarts and Feeney
1998). This method of adjustment suggests that period measures in Western Europe
understate the `real' recent level of fertility by between 0.2 and 0.4 children. That still
leaves most of them below the replacement rate, although to a less alarming degree
than that indicated by the conventional, un-adjusted TFR. Similar calculations applied
to Eastern European populations suggest that at least until the mid 1990s the great
falls in period TFR in many of those countries (not Russia) represented very little
change in final quantum, as tempo adjustment restored them to an average of nearly
two children (Philipov and Kohler 2001). On the other hand, similar calculations
applied to the republics of the former Soviet Union as well showed hardly any tempo
effect up to 1995. There, it seems, births were being reduced across the board,
implying a substantial reduction in completed family size or `quantum' (Macura et al.
2002, Avdeev 2001).
Recuperation
Delay in childbearing implies a degree of recuperation of fertility at older ages, if any
of the delayed births are actually to be produced. While cohorts of younger women
have (often) not completed their fertility, nonetheless the pace at which they are
producing their families permits certain conclusions to be made. If the recuperation of
fertility is to be sufficient to maintain a constant completed family size, then
biologically and socially plausible patterns of family building must be implied in the
uncompleted part of the fertile life span (Lesthaeghe 2001). Comparisons of the
reduction of fertility before age 30, and its recuperation after age 30 in recent cohorts
of mothers compared with those born in 1950 allow the extent of recuperation to
specified levels of completed family size, achieved by given ages, to be estimated
(Lesthaeghe 2001). The Netherlands and France emerge well from this analysis, with
17
increased birth rates among women in their early and late thirties almost sufficient to
compensate for the sharply reduced birth rates of younger women, compared with the
1950s baseline. Others, however, notably Italy and Spain, show much weaker
recuperation of fertility at older ages and their trajectories indicate a completed family
size continuing will below replacement, a continuation at best, and more likely a
decline, from the cohort rates at the moment. There, recovery to the fertility levels of
1950 in (for example) would require an accelerated pace of childbearing in the late
30s and 40s beyond anything known to have happened before (Frejka and Calot
2001).
Pessimistic projections of the future birth rate
At least there is some agreement that the `real' level of fertility is rather higher than it
seems in most countries and that period rates can be expected to rise somewhat. That
outcome already incorporated in most population projections. There is less consensus
on the more tricky problem of the explanation of the independent causes of low
fertility, their likely duration and thus the likely future of fertility in Europe. It is fair
to say that, in this debate, the pessimists (or perhaps realists) are in the majority for all
the reasons noted above. There is a near-consensus that European birth rates will not
return in the near future to replacement rate, or even anything like it in Southern
Europe and in parts of the CEE and FSU (Lesthaeghe and Willems 1999, Avdeev
2001). National population projections assume continuation of low fertility with only
moderate recovery through the process of recuperation; sometimes as in the case of
Germany, too pessimistically even for the pessimists themselves. Some assume a
continuation of current levels. Expert opinion is widely used for the targets adopted
for national population projections. Other projections have used birth intentions from
surveys or have taken national plans into consideration, with fertility theories coming
a poor third (de Jong 2002, Figure 15). Most national agencies in Europe assume an
increase in period fertility by 2020, but none to a TFR as high as 2.0 and only eleven
as high or higher than 1.75 (de Jong 2002 Figure 18).
Can birth rates return to a higher level?
Perhaps it is more appropriate to ask why fertility should not fall further. As noted
above, the tacit assumption that the transition would end at a convenient stabilisation
at replacement level fertility for ever after, is now looking rather inappropriate (UN
2002). Instead attention has been directed to how low fertility can go; to zero,
according to some interpretations of economic theory (Namboodiri. and Wei 1997,
given the great direct and indirect net material costs of children in modern society.
Some irreversible socio-economic trends, themselves desirable, seemed inevitably
likely to continue to depress fertility. The classical Malthusian relationship suggested
that increases in men's real wages and the economic growth which fuelled them
would tend to increase marriage and birth rates in stable societies, and vice-versa
when times were harder. The Becker model (1981) is a sophisticated re-working of
Malthus. It posits, without explanation, a desire for child `services', mostly of a
psychological nature. In a stable situation these would be satisfied more easily (once
basic needs were satisfied) by the better off. Other things being equal, economic
upturns would have the same positive effects on fertility throughout society. This
theory can say nothing, however, about the number of children required to satisfy the
desire for them.
18
New Home Economics models took into account the increasing importance of
women's paid work in the post-war period and of their contributions to the household
budget. That new development nullified or reversed the old Malthusian relationship
based on men's income only, because women did not find that child care and paid
work outside the home were easily compatible: Choosing one or the other involved a
substantial opportunity cost: foregone income, or foregone children (Joshi 1990). As
married women entered the workforce, fertility must necessarily decline ceteris
paribus. Furthermore women's economic independence made marriage itself less
attractive; it became rational to postpone such commitment or avoid it. Cohabitation
became a more suitable compromise in societies that tolerate it (Ermisch 1996); in
those that do not, birth rates fall even further. Relatively high women's workforce
participation became established in northern parts of Europe and in North America by
the 1970s, its effect fortified by equal pay legislation. Counter-cyclical relationships
were then expected to prevail whereby the upturns of the business cycle, or longer
term trends bringing women into work, would neutralise, or even reverse (Butz and
Ward 1979), the formerly expected positive effect of economic growth cycles on birth
rates.
Opinion is divided on how well the transition in the labour force and of the inputs to
household budgets fits the European and North American fertility patterns from the
1970s: certainly not well enough for some critics (Murphy 1992, Robinson 1997). The
upturn of the baby boom from the 1950s was fuelled by unprecedented economic
growth in the Western world (Crafts and Mills 1995). In the absence, at that time, of
Figure 11 Women's workforce participation rate and TFR, 1970 and 1994. Sources of
data: Council of Europe and Labour Force Surveys.
4.
3.
3.
2.
2.
1.
YEAR: 1970
percentage of females aged 15-64 in employment
706050403020
totalfertilityrate
0
5
0
5
0
5 Rsq = 0.3169
FRG
USAUK
SWI
SWE
ESP
POR
NOR
NL
LUX
ITA
IRE
ICE
GRE
FRA
FIN
DEN
CAN
BEL
AUT
YEAR: 1994
perc entage of f em ales aged 15-64 in employ ment
908070605040
totalfertilityrate
2.2
2.0
1.8
1.6
1.4
1.2 Rsq = 0.5165
FRG
USA
UK
SWE
ESP
POR
NOR
NL
LUX
ITA
IRE
ICE
GRE
FRA
FIN
DENCAN
BEL
AUT
high levels of female workforce participation, it may be regarded as the last fling of
the old direct Malthusian relationship between male wages and births. In those
classical terms the subsequent end of the baby boom, after the mid-1960s, is quite
inexplicable. Real wage growth continued but high birth rates did not. New Home
Economics models that take into account the rapid increases in women's workforce
participation, and of equality of pay, of the 1970s and onwards fitted the new situation
19
rather better (Ermisch 1979). Up to the 1980s the expected time-series and cross-
sectional negative relationship between period fertility levels in the western world,
and women's workforce participation rates, was well established (Figure 11). Modern
emancipated societies where women joined the workforce and thus did what they
wanted, then had the lowest birth rates.
However by the 1990s the position had changed radically; the cross sectional
relationship was significantly and obviously reversed (Figure16). Taking all the
Western countries together, the only national populations with relatively high birth
rates (TFR over 1.5) are those with relatively high levels of female workforce
participation (Coleman 1999, Ahn and Mira 2002). Women's paid work had become
at least partly compatible with childbearing and child-care in some of the Western
countries, notably the United States and North-western Europe, although much less so
in Germany, Austria and the South. Modern emancipated societies where women go
out to work now have the highest birth rates. This contrast between countries is
mirrored at the individual level: for example since the 1980s, in Norway and some
other countries, women at work are just as likely as those keeping house to have a
third child (Kravdal 1992). In Sweden's roller-coaster fertility trends, pro-cyclical
fertility is observed, with levels of female earnings now positively related to levels of
childbearing (Andersson 2000). It must be kept in mind, however, that a high
proportion of women's work, especially in the peak childbearing years, is part-time,
and opportunities for part-time work vary greatly between countries, being
particularly restricted, for example, in Germany and Italy. The correlation with
fertility is less strong when only full-time work, or hours of work, is considered.
Economic models of fertility have yet to provide a specification of the nature of
demand for children, or of `child services' or `child quality', to explain why modern
parents should want any at all. Perhaps no wonder that Robinson (1997) summarised a
review of the economic theories of fertility by commenting: ` the economic model
seems bogged down in a simplistic demand-oriented framework, with the unnecessary
and confusing `quality of children' notion clouding everyone's thinking' (p 70-71).
The limited success of economic models of fertility has given indirect support to
theories of a more cultural or ideational nature about the chief influences on fertility
in prosperous countries. It has also focussed attention on the effects of family policies
and `institutional' effects more broadly; all the other intended and unintended
consequences of government action (Gauthier 1998, DiPrete et al., in press). In theory
at any rate, economic theories of fertility are not without room for ideational factors
and ideational theories leave space for economic effects, not the least in the creating
the right conditions for wider freedom of choice. At least it is clear that the social,
cultural, political and policy environments in NW Europe and in the US seem in their
different ways to be much more `family-friendly' than that of Southern Europe,
Eastern Europe and the Asian industrial countries. Teasing out which is the most
important of these factors has proved to be very difficult, a point considered further
below.
Both economic and ideational concepts are deployed in individual-level analyses of
fertility behaviour through life-course analyses and more traditional multivariate
approaches. Inevitably these reveal a higher level of indeterminacy than aggregate-
level approaches. There is a strong element of randomness and unpredictability about
individual family building, thanks to the uncertainties to which individuals are prone,
20
the negotiations between couples (Voas 2003), and the chances of accidental
pregnancy or of failure to conceive a wanted child. Individuals are not much better
than demographers at predicting family size. Surprisingly high proportions of births in
modern societies ­ perhaps a quarter- are still reported to be unwanted in respect of
timing, and some are regretted altogether. On average the errors roughly balance out,
usually on the side of deficit, the more so when births are delayed. A higher-level
element of indeterminacy may also be important: simply following fashion in
childbearing might lead to random switches from relatively high to relatively low
birth rates unrelated to and unexplained by the conventional rational choices and `risk
factors' normally associated with family formation (Kohler 2001).
Gender equity
The success of cultural and ideational models in accounting for some recent
demographic behaviour was noted above, together with reservations about the degree
to which these `Second Demographic Transition ` ideas might be applicable to the
explanation of low fertility. Even though some analysts have been at pains to
emphasise that economic models and ideational models of fertility need not be
contradictory, in practice the notions and data of the one are not easy to incorporate in
the other and the literature remains somewhat antithetical.
Just as `New Home Economics' models have helped to put women back into the
economics of fertility, the notion of `gender equity', which has become prominent
during the 1990s (McDonald 2000) may do the same from a more cultural viewpoint.
Equality between the sexes is part of the progressive package of modern ideational
change. The central position of women in both economic and ideational models may
help reconcile their apparently conflicting interpretations, in societies where men are
becoming more like women, and women more like men.
The symmetry of sexual equality in the public and in the private realms is thought to
be crucial. Asymmetry between them, in modern societies, leads to very low fertility.
In most Western countries, opportunities for women in employment, education and
public life (so -called 'economic gender equity') have advanced considerably,
promoted, to varying degrees, by legislative change and equal opportunities policies.
However, if these trends are not matched by equivalent changes in that society's ideas
about equality in the private realm, on domestic tasks, the care of young and old
family members and in sharing other responsibilities, then fertility may be severely
constrained. Women trying to work in both realms become overloaded with too many
tasks. While cultural and legislative changes have permitted, at least to some degree,
greater freedom of movement to women in the public realm, division of labour in the
home is still much more traditional in the societies of Southern and Eastern Europe,
and in the developed countries of Asia. That is, women do all the work.
The attractions of the labour market, the welfare arrangements for young and elderly
dependants, if any, the familial pressure faced by women to produce a child and care
for elderly parents may all pull in different directions. Low levels of fertility
combined with low levels of childlessness, notably in Southern European countries,
thus arise from a tension between unequal levels of gender equity in different social
institutions of society. New opportunities similar to men's in education and work are
in practice curtailed by obligations to children, to husbands and to the older
generation inherited from an older regime of 'familistic' values. In those
21
circumstances women will satisfy the obligation to produce a family to the minimal
extent, for the most part avoiding childlessness but restricting subsequent children
severely (McDonald, 1997). At the same time, workforce participation rates will
remain lower than they might be while husbands' participation in housework and
childcare remain minimal (Kiernan 1996). The same `familistic` culture, which at the
micro-scale expects families (i.e. women) to care for their own family members
young and old, has larger scale connotations. In such societies political pressure for
the high levels of public family support and old-age care, taken for granted in North-
West Europe, has been weak until recently (Gauthier 1996 and 2003). `Familist'
culture surviving into the modern world means in practice more work for women, less
help from public resources.
In Italy and Spain and Greece, three-generation families, although comprising a
minority of households, are still more prevalent than in Northern Europe. Furthermore
adult children there, as in developed Asian countries, appear much more content to
remain living with their parents than in Northern Europe. Despite being at home and
working, adult children can enjoy conditions of considerable personal and economic
freedom, be well looked after by devoted mothers, and avoid the inconvenience of
moving out to live in separate households (Castiglioni & Dalla Zuanna, 1994,
Palomba 2001). Their situation in Southern Europe parallels that of the `parasite
singles' of Japan. Many factors contribute to this: a preference by mothers, however
exploited, to keep their children close, interminable degree courses, a difficult housing
market and youth unemployment in a highly protected labour market. All this delays
marriage, within which almost all childbearing is still located. Major policy changes
to support working women if chronic low fertility is to be avoided in those areas.
Insofar as those would appear to require major shifts in traditional culture, they may
be a long time coming.
Women do it their way?
The rise of female autonomy, with women overtaking men at school and college,
potentially earning the same or more, erodes the value of marriage for women
(Ermisch 1986). On purely rational choice grounds, the compromise of cohabitation
or living apart together may seem preferable. Better-off women, if they want a child,
may be able to dispense with a partner altogether. Are some women only prepared to
have children if they can have at least some of their babies in a looser, less
constrained relationship? As noted above, relatively high birth rates are only found in
those societies which tolerate a relatively high proportion of births outside marriage
(Figure 12). Most but not all occur within cohabiting unions. In such countries many
women, it seems, want to have children more than they want a firm commitment to a
partner (Kravdal 1997). These are, not surprisingly, mostly those countries with well-
developed welfare systems and where non-traditional attitudes to marriage, divorce,
and cohabitation prevail. Does it follow from this that higher fertility in Southern
Europe and the Asian industrial countries must await cultural changes facilitating
cohabitation and the emancipation of `illegitimacy'?
American exceptionalism
The North American examples sit awkwardly in an account devoted primarily to
Europe. This is partly because the US example seems to contradict the lessons of
Europe, partly because the fertility patterns in Canada and the United States, once
very similar, have diverged in the 1990s. Canadian TFR has descended to mid-
22
European levels, while birth rates in the US have increased to become the highest of
any major country in the developed world (TFR 2.03 in 2001). The US example is
Figure 12. Source of data: Council of Europe
TFR and Births outside marriage 2000
Total Fertility Rate
2.22.01.81.61.41.2
IllegitimacyRatio
700
600
500
400
300
200
100
0
UK
Sw itz
Sw eden
Spain
Portugal
Norw ay
Netherla
Luxembur
Italy
Ireland
Iceland
Greece
Germany
France
Finland
Denmark
Bulgaria
Austria
important because it appears to show an alternative route to relatively high fertility
without NW Europe's welfare culture. US fertility patterns are complicated by the
salience of higher fertility from some ethnic minority groups. In 2001 Hispanic TFR
was 2.75, and black fertility, though somewhat convergent, was 2.05 (Hamilton et al.
2003 t. 1). But even the non-Hispanic white TFR was 1.84. That is a high level
compared with Europe, where of course overall fertility is somewhat elevated by the
fertility of its own ethnic minorities, which are not usually distinguished in the
statistics. American society has developed strong currents of female autonomy and
independence; it provided the feminist cheerleaders of the Western World from the
1960s onwards. Women's workforce participation is very high and their prospects for
advancement good, reinforced by law as well as informal pressure.
Although American culture is diverse, companionate marriages are favoured and New
Men approved. Cohabitation is common, though less so than in Scandinavia. In a low-
tax environment, private arrangements for childcare, although time consuming and
often inconvenient, can be more widely afforded, and illegal immigration provides
sources of domestic help for the better paid. Less conventionally, it may be mistaken
to ignore the extent to which welfare does help child-care (Folbre 2001). However the
US shares with the UK a pathological side of the `Second Demographic Transition' in
having undesirably high levels of teenage childbearing: 45 and 27 per 1000 teenagers
23
respectively in 2001. Were that to fall to average European levels, overall US TFR
would be about 1.83. Furthermore, it has been claimed that the US fertility advantage
compared with Europe, and the differentials between racial and ethnic groups, are
substantially accounted for by a high level of unplanned and unwanted pregnancies,
equivalent to about 0.7 of a child (Table 3). By contrast, in Italy it is estimated from
FFS data that unplanned births declined from 0.57 per woman among the 1946-50
birth cohorts of children to 0.28 among children born in 1991 ­ 1995, despite Italian
preferences for obsolete methods of family planning (Castiglioni et al. 2001). The
total unintended fertility rate in the US in 1994 (average number of unintended births
per women) were as follows:
Table 3 Total unintended
fertility rate of which :
(a) total mistimed
fertility rate
(b) total unwanted
fertility rate
White non-
Hispanic
0.50 0.38 0.13
Black non-
Hispanic
1.19 0.69 0.51
Hispanic 0.96 0.56 0.40
Source: Henshaw, in Frejka and Kingcade 2003 table 28.
It should be noted that this is much higher than some previous estimates; for example
that only 7% of US births in 1978-82 were unwanted, plus another 22% mistimed
(Menken 1985) which seems very low by any standards. These high rates are claimed
to follow from the deficient family planning system and limited sex education in the
US ,compounded by a low level of functional literacy among the young and least
educated (Frejka and Kingcade 2003). As none of the underlying factors is expected
to change fast, higher US fertility is expected to persist. An official campaign during
the 1990s to reduce the very high level of teenage births appears to have successfully
reduced the rate by about 20% without making any discernible impact on overall
fertility.
Can governments influence the birth rate?
Neither ideational nor conventional economic models of fertility based on earned
income can explain adequately the patterns, trends and differentials of birth rates. On
the other hand, people in developed countries can make few decisions without
encountering fiscal or other incentives or disincentives for their actions which attract
subsidies or taxes, direct and indirect, through housing, education and health
provisions, quite apart from policies focused on the family. In Europe, governments
spend between 40% and 55% of annual national income. Effects on the rational
choices of individual citizens are inevitable, including those related to family
formation and reproduction.
Welfare support links ideational and conventional economic models, being a product
of the former though public opinion and politics, and having its effects, if any, by
ameliorating the time and money budgets of the household. These `institutional'
factors; various fiscal and workforce measures, and in some countries, family welfare
policies, may have permitted the better realisation of desired family size through a
combination of direct and indirect payments, more flexible working arrangements and
24
child care facilities aimed, at least ostensibly, at family and women's welfare rather
than demographic targets. Others may hamper family formation. Because of these
multiple intended and unintended (potential) effects, it has been difficult to come to
clear conclusions on the question whether government action does, or does not,
eventually influence the birth rate.
Family welfare measures usually described as `family policy' are more or less
universal although to a very varied degree. Whatever their ostensible purpose, in
terms of actual provisions there is little but the rhetoric to distinguish some of them
from `pronatalist ` policies insofar as they are designed to make it easier for women
(and families as a whole) realise their desires for a family without falling into poverty
and to ease the apparently conflicting demands of child care and workforce
participation. More specifically, most are not intended to increase demand for
children, but to make it easier to satisfy it.
The bottom line of the demographic assumptions relating to family policy ­ as
opposed to its pure welfare aims ­ is that most women want to have at least two
children and will attempt to have such a family size. This is supported by most of the
available evidence from pan-European Surveys such as Eurobarometer (Eurostat
1991) and the European and World Values Surveys (Table 4), the Family and Fertility
Survey and national surveys such as the British General Household surveys. Despite
Table 4 Ideal number of children (mean)
World Values Survey Eurobarometer
1981 1990 1995/6/7 / 8 1989
France 2.90 2.62 2.13
Britain 2.71 2.50 3.68 2.14
W Germany 3.14 2.25 3.08 1.97
Italy 2.59 2.44 2.20
Netherlands 3.19 2.63 2.23
Denmark 3.39 2.56 2.13
Belgium 3.33 2.48 2.01
Spain 3.37 2.53 2.84 1.94
Irish Republic 4.39 3.52 2.79
USA 2.66 2.74 2.61
Canada 2.77 2.79
Japan 2.68 2.75 2.63
Sources: World Values Survey
http://www.worldvaluessurvey.org/services/index.html
Eurobarometer 32 the Family and the Desire for Children, Graph 1.
http://europa.eu.int/comm/public_opinion/archives/ebs/ebs_048_de.pdf
note: sample size about 1000 in each case.
the well-known difficulty of interpreting the answers to such questions, replies have
been relatively consistent over the three decades or so over which such surveys have
been held. That being the case, it has been assumed that if family policies can look
after the interests of women, incidentally helping them to have the number of children
that they want or to have them earlier, then population will look after itself. However,
there are some recent suggestions that desired family size in some European countries
has started to slip below the average of two. If that continues some re-thinking will
25
need to be done (Lutz et al. 2003). Underlying desire for children is taken to be
axiomatic in the economic rational-choice theories of Becker and others. At
equilibrium, families with higher income should have more children, as they have
more of everything else. This seemed to be contradicted by the negative association of
status and education with family size which emerged in the demographic transition.
But even by the mid 20th
century the general relationship between family status and
fertility, measured by men's rather than women's income and education, was no
longer negative and linear (Ni Bhrolchain 1993b) but had become more U ­ shaped.
Since the 1950s higher-income families no longer had the lowest fertility. Now,
women of higher occupational status(in their own right) and women with higher
education, while more likely to end up childless, may also end up with larger than
average families if they have any children at all. The evidence is ambiguous, however,
both in Scandinavia and in the US (Kravdal 2002, Retherford and Luther 1996,
Martin 2000). In Britain, better-educated women have the smallest income loss
though childbearing, even for fourth births, although they are also most likely to avoid
or postpone childbearing (Joshi 2002). There seem to be two opposing effects, higher
qualification tending to promote further childbearing while also delaying it,
moderating its positive effect (Rendall and Smallwood 2003). This pattern does not
point to the extinction of the family.
Do family policies encourage families?
Family polices and their recent development have been admirably reviewed by
Gauthier (2003) for this meeting. There is no need to repeat her findings but we might
ask again: what effect do they have on fertility? A glance at the comprehensive data
reported by Gauthier and by McDonald (2002) does not show an overwhelmingly
obvious correlation with high contemporary birth rates. It is true that Finland and
France appear in the upper distribution of all three kinds of benefits and their TFR is
among the highest of the European countries, that of France having recently increased
to 1.9. At the other extreme, Spain, Greece and Italy with their very low TFR feature
prominently in the lowest categories of family support, although Spain and Portugal
do well on maternity leave. In such countries, lacking cash transfers or tax reliefs, the
arrival of successive children reduces living standards of families substantially. The
UK as usual has a somewhat anomalous position: cash benefits for children as a
percent of GDP are little below those of France or Sweden, while levels of public
childcare for young children and of maternity leave are among the lowest in Europe.
Its relatively high fertility has consequently been a source of surprise and even of
irritation to overseas observers; its recent decline may be taking it down to a more
appropriate international position given the level of family support.
Systematic quantitative comparisons are made difficult by the variety of effects, direct
and indirect, intended and unintended, of many government policies even within one
country. It is difficult to take account of them all. Gauthier concludes `there is no way
of measuring the degree of child- and family- friendliness in countries' (page 14). Pre-
war pronatalist policies, widespread in Europe, developed before high levels of
female workforce participation, were in general too meagre to be effective (Glass
1967). Policies in the post-war communist regimes of Eastern Europe copied some of
the features of the pre-war Nazi programme. Measures included family allowances,
free childcare in crches, preferential access to state housing for married couples with
children. In the absence of access to modern contraception until recently, reliance on
abortion permitted some element of state control over fertility, however unwanted that
26
fertility might be. Fertility could increase in response to policy change at least in the
short run, for example in East Germany in the 1970s, mostly a result of temporary
shifts in the tempo of fertility rather than enduring increases in the quantum of
completed family size (e.g. Stloukal 1998). However, fertility rates in Eastern Europe
showed little tendency to decline after around 1970 and TFR remained around 2 until
the end of Communism, that is about 0.5 above the average level in Western Europe.
The weakening of these policies after the end of communism was followed by sharp
reductions in the birth-rate, which remains universally low today (Macura 1999,
2000). That fall is hardly be surprising in view of the parallel sharp decline in national
income and the novel rise of unemployment and general economic and political
uncertainty. Economic recovery in Central Europe in the late 1990s, however, has not
yet brought any recovery of the TFR. The family-policy history of these countries is
not a good guide. Any `success' was gained in an unfree environment, without access
to modern contraception and without free markets ­ or much choice - in housing or
labour outside that provided by the state, or even, for most, the option of emigration.
It is inconceivable that democratic political systems would wish to copy them.
In Western Europe, some comparative analyses have showed rather weak net results
of family policy on fertility (Gauthier 1996, Blanchet and Ekert-Jaffe 1994); for
example that an increase in cash transfers would increase the TFR by only 0.02
children (Gauthier and Hatzius 1997). Other results are more positive, finding a clear
positive statistical association between the child benefit package and fertility levels
(Finch and Bradshaw 2003). However these careful studies have been unable to
include all relevant factors such as housing. A common message is that workforce
policies and those affecting gender equity are probably more important than others
and any analysis ­ or policy - which omits them will not yield results. Those Italian
municipalities that throw cash at women hoping to increase the output of babies will
do so in vain.
At the national level, an early analysis of French policy concluded that it had elevated
the TFR by up to 0.3; that is taking it from an expected 1.5 to the actual 1.8 of the
early 1980s (Calot and Chesnais 1983). A similar but more modest result (an increase
of 0.2) was obtained by Ekert (1986). Analysis of the extensive Swedish family
support system have suggested a close connection between its cash benefits, tax
breaks and especially its paid maternity leave provisions, and the level of period
fertility. Comparison with the situation in Hungary underlines the importance of
gender roles. In Sweden, effects of policies are quite similar for men and women,
unlike in Hungary where more traditional gender roles are preserved. Hence `public
policies that aim to reduce the conflict between employment and parenthood do
facilitate the combination of these competing roles for individuals, especially women'
(Olah 1998 p. 24) underlined by the higher average fertility and progression to second
birth in Sweden. Policy effects in Sweden are modulated by economic trends, which
are positively related to fertility rates (Santow and Bracher 2000). Malthus lives, it
seems.
Swedish comprehensive system of family benefits, child care and work practices
(especially the easy availability of part-time work) make continuous working life a
natural choice for young women, and fathers are more involved in childrearing. The
less flexible work arrangements in France pressed French women to choose between
27
working life and family, putting a double work load on Frenchwomen who worked.
Partly thanks to day-care arrangements, career-minded Swedish women are more
likely to progress to a third child than in France (Corman 2000) although for all
women the proportions are similar. Third-birth rates are crucial to the achievement of
reasonable total fertility rates, especially in populations where childlessness is quite
common (14% among the 1960 cohort of women in both countries). Since the mid-
1990s, however, the Swedish economy has fallen on evil times, with (for Sweden)
unprecedented levels of unemployment that have hit younger people hard and reined
back the generous but expensive welfare system. Because welfare is so strongly
linked to work, unemployment depresses family formation severely. The Swedish
fertility rate is now more modest although recovering, while that of France has
increased (1.9). However, similar analyses of the effects of policy on fertility in
Norway, where total fertility rates float on a sea of oil money, come to similar
positive conclusions (Rnsen 2001) in a country where the TFR remained high (about
1.85) until very recently.
Policies in Sweden and Norway have been described as `state feminism'. More than
elsewhere, men and women share childcare. Policies to promote gender equity no
doubt help, but those polices could not be pursued in a democracy without substantial
pre-existing support.. This contrasts with countries such as Austria ` which has
hesitated to put its money on gender equity and ...has retained much of the old sprit of
family-role specialisation' (Hoem et al 1999 p 22) and where third births have
declined markedly among working women. In this respect, Austria seems to resemble
other German-speaking countries ­ Germany and Switzerland ­ also not famous for
gender equity and where female workforce-participation and birth-rates are both low.
The emergence of a positive relationship between birth-rates and female workforce
participation rates may reflect the success of family-friendly employment policy in
Northern Europe in moderating the opportunity costs of childbearing, and the dire
fertility consequences of persistent gender inequity in the South and in Japan
(McDonald 2002).
Despite the uncertainty, it remains true that some developed counties obviously
provide a more `family-family' environment than do others. Three roads, it was
claimed in a recent analysis, can lead to comparatively high fertility rates:
(1) high wages, ample childcare, and well-developed maternity leave
(2) high gender equity and well developed parental leave
(3) high male wage, high percentage of female part-time employment but with less
parental leave. Socio-economic systems and childcare support need `institutional
complementarity' (Fukuda 2003, p. 23); a good fit promotes fertility. If bad, public
policies will have little effect, as they have so far in Japan.
Pronatalist measures
The governments of most European countries and of all the Neo-Europes avoid
explicit pronatalist motivation. However, as low fertility goes on and on and
population ageing starts to loom large on government agenda, the merits of a
demographically-aware family welfare policy become more apparent. Accordingly
some national and local authorities, notably in birth-starved regions of Italy, have
revived specifically pro-natalist measures usually involving cash payments, most
recently to mothers of second children in the Naples area. However, modern views on
women's empowerment do not take kindly to any suggestion of instruction
28
concerning their reproductive `duty', and no concept more clearly contrary to the
spirit of the second demographic transition could be imagined. However pronatalist
policies, at least of a mild kind, were and are by no means restricted to autocracies;
they were the norm in pre-war Western Europe. Pronatalist elements of policy (parity-
specific benefits), of a non-oppressive kind, are maintained within family welfare
measures are maintained at a subdued level in France and one or two other European
democracies, and are explicit in Japan and Singapore.
In 2001 the governments of 24 out of 48 developed countries reported that their
fertility was `too low' (compared with 7 out of 34 in 1976). Most expressed worries
about population ageing, not about declines in the working age or overall population
size. But only 14 had policies intended to elevate the birth rate (United Nations 2002).
Eleven of these were former communist countries (where TFR has fallen to between
1.2 and 1.5), the twelfth was Singapore, where ingenious but persistently fruitless
efforts in this direction have become famous. In Western Europe only the
governments of Austria and of Luxemburg had policies to raise fertility. None of the
other low-fertility countries of Western Europe admitted to such polices even though
almost all declared their national fertility rate to be too low.
These responses cannot be taken entirely at face value. For example the declared
intention of Canada's immigration policy is to increase the population by 1% per year,
but the return claims a population policy of `no intervention'. Even the Italian
response even claims only `mild' concern about ageing and none about the decline in
the workforce, despite widely publicised worries on these matters (e.g. Golini et al.
2000), the prevalence of provincial policy initiatives, and the assertion by some Italian
authorities that mass immigration is needed to rectify population ageing.
Genetics, evolution and biology
In all this discussion, some basic desire for children is assumed. Where does it come
from? Why is fertility not tending to zero, given that the logic of costs, negative
utility, increasing individualism and the like all point downwards. In the 1990s, this
fundamental problem of why modern educated individuals have any children at all,
and whether there is any reason why the average should be two, has moved closer to
centre stage. It is being addressed for the first time from genetic and evolutionary
viewpoints as well as more conventional ones (Friedman et al., 1994, Hobcraft and
Kiernan 1995; Joshi and David 1996, Hobcraft 2003). Economic theories must
assume an axiomatic desire for children and child services. Evolutionary theory
purports to provide an enduring reason for that desire, not related to any `material
utility' of children, much questioned since the beginning of the decade and clearly
negative in modern settings. On the new view, fertility in the human species, as in all
others, is maintained ultimately for the benefit of genes, not of people (Coleman
1999). This is achieved through basic needs of biological origin not just for sexual
gratification but, more important, for the protection and nurturing of children. This is
not the place to discuss these ideas but to note their recent salience (Udry 1996, Foster
2000, Morgan and King 2001) and the reaction that they have provoked in some
circles. If demography has had an ideology it has tended to be sociological,
traditionally hostile to attempts from any non-social discipline to interpret `social
facts'. Accordingly this new thinking has provoked strong responses (e.g. the
exchanges in European Journal of Population 17, 1 (2001); and Vetta and Courgeau
(2003). We are still a long way from establishing strictly defined connections between
29
basic human behaviour, gene differences and neuronal pathways, if indeed it can ever
be done.
Population Ageing and Population Decline
Persistent low fertility, together with increases in survival, leads naturally to older
population structures. Sooner or later it will go hand in hand with population decline.
Natural decrease has already arrived in some western countries; in Germany, Italy and
Greece only net immigration prevents actual population decline. Within the next thirty
years all European countries are expected to follow Italy into `natural decline', where
deaths exceed births, some very gradually, others more rapidly depending primarily
on past and future levels of fertility (Lutz et al 2003). That will go hand in hand with a
more aged population. Population decline, as opposed to natural decrease, will be
deferred until later, because net immigration is expected to continue. Alone in the UN
ECE area, the population of the United States is not expected to decline at all but to
continue to increase, indeed at an impressive rate. The populations of Europe as a
whole, and the United States, are thus on divergent paths. This disparity in growth
between the US and Europe has excited financial journalists who see much
economic, and geopolitical, significance in the future demographic disparities
between even an expanded EU, and a United States continuing to grow to 500 million
and beyond. Both, however, face similar pressures from neighbours.
Demographic marginalisation
The populations of Europe's Muslim neighbours in North Africa and Turkey, minor
players economically and demographically in the early 20th
century, have now
accelerated past the European total. The Mediterrnaean now comprises the greatest
demographic, economic and cultural gulf in the world ( Chesnais 1995). The southern
Mediterranean countries also continue to send large numbers of their citizens to
Europe. Behind them the vast demographic potential of sub-Saharan Africa, its
growth only moderately checked by AIDS, continues the population increase
propelled by huge demographic inertia and very late transition. Of the 14 countries
not yet even started the demographic transition, 12 are in Africa and 1 (Yemen) in the
Middle East. Their potential is impressive: the population of Yemen is projected to
exceed that of the Russian Federation shortly after 2050 (Demeny 2003).
At the other end of the UN ECE area, the US looks across a parallel if less striking
demographic and economic gulf to Mexico and Latin America. There the disparity is
less marked: Latin America has made more economic progress and fertility is falling
faster. Furthermore, US population is growing much faster than that of Europe,
though its more youthful age-structure, its higher fertility and its high levels of
immigration - due substantially to the transfer of large parts of the Latin American
population, together with their higher birth rates, to the USA.
Finally in the Far East of the region an old demographic gulf, much discussed in the
Cold War period, continues un-noticed. Russia's Muslim Asiatic hinterland, separated
not by river or sea but by steppe, a nagging security problem in Soviet times,
continues to gain ground demographically, despite chaos and civil war, as the Russian
population falls. Projections to 2050, however, do not suggest that these republics will
30
Figure 13 Major UN ECE regions and their neighbours ­ demographic contrasts
across the Mediterranean, the steppes of Central Asia and the Rio Grande.
31
Southern Europe and North Africa: Projections 1990 - 2050 (millions)
0
0
0
0
0
0
1990
1995
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
Southern Europe
North Africa
Russia and the Central Asian Republics: projection 1990-2050 (millions)
100
120
140
160
1990
1995
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
Russian Federation
Central Asian Republics
USA, Central and South America. Projections 1990 - 2050 (millions)
100
150
200
250
300
350
400
450
500
550
1990
1995
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
USA
Central America
South America
35
30
25
20
15
10
80
60
40
approach the Russian total by that year. The legacy of communism has obviously
changed radically the status even of revived religion, Russian hegemony continues
and despite much return migration substantial Russian nationality population remains
in Central Asia (Figure 13).
Population ageing
The population of the UN ECE, like that of the whole world is getting older and all its
countries, sooner or later, will have to face and manage the consequences. That is an
inevitable consequence of low vital rates. In the last century, fertility decline has been
the main engine behind population ageing. Fertility, now constant or even rising, will
eventually make no further contribution to population ageing. Instead, continued
mortality improvements will be the dominant or only factor and has already become
so in some Western countries (Calot and Sardon 1999). At an average family size of
1.8, about the level of the higher fertility European countries such as France and
Norway, the percentage aged 65 and over would stabilize at about 23% at current
mortality rates. At 1.6, about the European average, that percentage would rise to
about 28%. With persistent really low fertility, like that seen in the Southern and
Eastern European countries, the percentage would rise to well over 30%. In the
absence of migration, stable populations with these rates would have rates of
population decline varying from 0.25% per year to 0.75% per year and potential
support ratios from 2.0 to 1.1 (Table 5). Differences in fertility that may appear small
will generate, over decades, highly divergent age-structures. Of course we are not
dealing, yet, with stable populations, which will take some decades to emerge. For the
Table 5 Stable populations with given levels of period fertility
TFR 1.55 TFR 1.78 TFR 2.07
Rate of Pop change (per 1000) - 10.0 - 5.0 0.0
Mean age 46.8 43.9 40.9
% pop under 15 years 13.0 15.7 18.7
% pop aged 15-64 59.4 60.7 61.3
% pop aged 65 and over 27.6 23.7 20.1
Overall dependency ratio 68.3 64.9 63.2
Aged dependency ratio 46.5 39.0 32.7
Potential Support Ratio 2.2 2.6 3.1
_____________________________________________________________________
Note: Expectation of life at birth in all cases 80.0 years for females, 76.0 for males.
Source: calculated from Coale and Demeny 1983, pp 79, 129.
time being, the age-structure in many Western populations is still favourable to
positive natural increase, if only marginally, thanks to the inheritance from the earlier
20th
century of large age-groups of reproductive age, particularly from the baby boom
period. But that demographic bonus is fast running out. Most western countries are
on the cusp of population ageing and sooner or later will acquire a negative
momentum generating natural decline (Lutz et al. 2003).
32
If fertility remains stable or increases, mortality decline, for as long as it lasts, will
eventually become the only continuing agent of population ageing in the future. The
reverse is true in the East. There, in the absence of a 1960s baby boom, momentum
from the age-structure is in any case more modest. Very low birth rates combined, in
many countries of the region, with persistently high death rates conspire together to
push most populations into natural decline already.
The current position is made to seem particularly unfavourable, because we
are coming to the end of a short and transient period of unusually benign age-
structure. During the latter half of the 20th
century, the more developed
countries could enjoy the benefits of low dependency from fewer children
together with a relatively low proportion of pensioners inherited from earlier
times. That demographic bonus is now disappearing and will not return; it is
now being enjoyed - for a while - by third world countries. The older
population can be projected with reasonable confidence; all have been born.
The standard picture of projected population ageing is very well known and
needs little elaboration here. Thanks to persistent low fertility over many
years, the contrasting levels and paces of ageing, and their international rank
order have acquired a distinct pattern since 1990 (Table 6).
Southern European countries now have the most rapidly ageing populations,
having overtaken populations such as the UK and Sweden in the 1990s. The
CEE countries are only kept temporarily out of the same league by their more
recent fertility decline and by their high death rates. Today the potential
support ratio (PSR; the number of people of potential working age for every
person of retired dependent age, stands at about four in most European
countries if calculated on the conventional age-limits of fifteen to sixty four
and 65 and over, and over 5 in the Russian Federation and the US. If
calculated on the more realistic basis of ages 20 to 59, and 60 and over, the
figure is closer to 3 (3.1 in EU15 in 2001). That is not the real ratio of actual
workers or taxpayers to retired dependents, which is much less (about 2.5 to
1). Only about 64% of Europeans aged 15-64 are actually at work. All
Western countries begin with a PSR of around about four calculated on the
former basis but will decline by 2050 to rather different levels: over two in the
case of France and the UK, well under two in the case of Germany, and in the
case of Italy and Spain, less than one and a half (Figure 20). These results
imply quite big differences in future problems for the economy, productivity,
dependency, and other indicators.
This future diversity means that it makes little sense to propose common
`European' solutions. Despite the claims of the European Commission (2001)
there is no `European' problem, rather a wide variety of national ones that
require their own individual policy responses. The low fertility countries of
Southern Europe, with their low fertility, `familist' traditions, rejection of
`second demographic transition' behaviour, low female workforce
participation rate, high unemployment and rigid labour regulations, resemble
somewhat the low-fertility societies of Asia. In demographic terms, the
countries of NW Europe, with their higher birth rates, high levels of
cohabitation and births outside marriage, weak kinship structures and high
female workforce participation resemble much more the populations of the
33
English-speaking world overseas, although they differ radically from the US
(less from Canada) in terms of welfare provision and taxation. These contrasts
are well illustrated by projections of the future populations of potential
workforce entry age (20-24). In NW Europe the expectation is of relative
constancy (including an element of migration, of course), while in others steep
decline is expected (Figure 14).
Table 6 Evolution of the age-structure 2000 - 2050, selected countries
France Germany Italy Netherlands Spain
65+ % PSR 65+ % PSR 65+ % PSR 65+ % PSR 65+ % PSR
2000 16.0 4.08 16.4 4.15 18.1 3.73 13.6 5.01 17.0 4.01
2010 16.6 3.95 20.2 3.31 20.6 3.22 15.3 4.49 18.5 3.69
2020 20.5 3.05 22.5 2.91 23.9 2.71 19.8 3.33 21.3 3.14
2030 23.8 2.52 27.7 2.16 28.6 2.11 24.2 2.52 26.5 2.37
2040 26.2 2.21 30.9 1.84 34.5 1.57 27.1 2.14 33.3 1.68
2050 26.7 2.15 31.0 1.83 35.9 1.47 26.5 2.22 37.6 1.36
indicator values in 2050
TFR 1.9 1.61 1.61 1.81 1.64
e0m 80.6 80.7 79.5 79.6 79.4
e0f 87.3 86.2 85.6 84.9 85.9
net migration 40 180 60 30 30
UK Norway Hungary Russia USA
65+ % PSR 65+ % PSR 65+ % PSR 65+ % PSR 65+ % PSR
2000 15.8 4.13 15.4 4.21 14.6 4.69 12.5 5.56 12.3 5.35
2010 17.0 3.94 16.0 4.18 16.0 4.38 13.3 5.50 12.8 5.21
2020 20.2 3.21 20.0 3.23 19.7 3.41 16.7 4.17 15.9 4.03
2030 24.3 2.48 23.5 2.56 21.5 3.02 21.8 3.00 19.2 3.20
2040 27.2 2.13 26.5 2.16 24.8 2.47 23.5 2.69 19.8 3.12
2050 27.3 2.12 26.2 2.21 28.8 1.99 27.0 2.17 20.0 3.11
indicator values in 2050
TFR 1.91 2.07 1.85 1.85 1.85
e0m 80.6 80.8 76.0 70.9 79.2
e0f 85.6 86.7 82.4 77.4 84.1
net migration 95 10 1 50 1100
Source: UN World Population Prospects 2000, 2002 Medium Variant
Population age-structure is not the only factor that determines the impact of
population ageing on society and economy. Those effects are modulated by
levels of workforce participation, by labour productivity, by the structure of
pension systems and their sustainability or cumulated indebtedness. These
differ considerably between countries and differences are likely to persist into
the future. Impact of demographic and non-demographic factors has been
brought together in an index of the impact of ageing which considerably alters
our picture of the most vulnerable populations in the future Many of the
assumptions about the future values of relevant variables may be questioned,
and the UN projections used are not ideal. Nonetheless such exercises may put
the spotlight more accurately on the danger areas.
34
Figure 14
95
85
90
80
75
65
70
60
55
45
50
40
35
Population aged 20-24, selected European countries 2000 -
2050 (2000 = 100). Source: United Nations.
100
105
110
115
120
2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050Norway France Germany
Italy Spain UK (UN)
UK (GAD 2001)
Figure 15 Index of weighted population ageing
Austr
UK
US
Canada
S
Ja
Germany
Ne
B
France
It
Spa
Source: J
Ageing Vulnerability Index 2003
Public Fiscal Benefit Elder
Overall Index Burden Room Dependence Affluence
Rank Score Rank Rank Rank Rank
alia 1 -1 2 2 4
2 7 1 1 6
3 18 3 4 3
4 42 6 6 5
weden 5 48 4 3 8
pan 6 50 9 9 1
7 52 7 5 11
therlands 8 62 8 7 9
elgium 9 63 5 8 10
10 81 10 10 12 8
aly 11 84 11 11 2 12
in 12 93 12 12 7 7
weight 1/3 1/3 1/6 1/6
ackson and Howe 2003, Figure 18
6
11
1
2
10
3
5
4
9
35
What can we do about it?
There is no 'solution' to population ageing short of a return to much higher rates of
population growth or mass age-specific euthanasia. Today's age-structure is non-
sustainable. Immigration cannot solve problems of population ageing except at rates
of immigration so high that they would generate economically and environmentally
unsustainable population growth rates and permanently and radically change the
cultural and ethnic composition of the host population (Coleman 2000). Figure 22
shows the population size consequent on the migration needed to preserve the current
potential support ratio in the UK up to 2100. Today's population of 60 million would
have doubled to 120 million by 2050 because the UK would be importing 1.2 million
persons per year. By 2100, up to five million new immigrants would be needed every
year, and the population would have risen to 312 million, not much less than the
present population of the whole EU.
The reductio ad absurdum of all this is what one might call the `Korea syndrome': the
level of immigration required in order to preserve the current potential support ratio in
the Republic of Korea and its consequences for population growth. In order to
preserve the present potential support ratio of Korea ­ 10: 1 ­ the population would
need to increase to 6.2 billion people by the year 2050. Just by coincidence, this
happens to be the entire population of the planet at the present time, so we would all
have to go there.
Figure 16. Source: UK Government Actuary's Department
Net Annual Immigration required to
maintain UK Potential Support Ratio, 2000 -
2100 (millions)
0
1
2
3
4
5
6
1998
2000
2010
2020
2025
2030
2040
2050
2060
2070
2080
2100
millions
Of course immigration does have a favourable impact upon the age structure and can
in theory maintain population size or workforce size, although to keep those constant
would require an implausibly precise level of management: immigration has always
been easier to start than to stop. The problem is that the effect is not very great and
immigration is an inefficient way of achieving this end. Immigrants themselves age
and then require more immigrants, as it were, to replace their number. Also, there is a
36
tendency for immigrant birth rates to converge to those of the host population
although by no means a complete one.
Population ageing is not a consequence of a failure of migration: birth rates
are a more effective way of changing age-structure, without population
growth. But population ageing cannot be `solved' by fertility either, although
it is easier to moderate it by that route. Even if the birth rate rose up to
replacement level, the potential support ratio would only increase to about 3.
But that would imply no further population growth and eventually, an end to
further population ageing given constant mortality. To keep the PSR at the
current level of four, an average family size of about three and a half would be
required, implying population growth of about 1.8% per year.
Managing population ageing
What matters, however, is not so much demographic abstractions such as the potential
support ratio but rather whether the future costs of dependency are sustainable in the
economic and social environment of the future. Workforce, retirement and pension
reforms within the demographic system offer many flexible and promising ways of
adapting to population ageing and preserving the viability of economic systems some
of the measures are desirable in their own right. Together with improvements in
productivity, such measures offer the prospect of a reasonably effective and
affordable management of this burden as long as birth rates are not too low, even
though they definitely cannot offer a `solution' (Daykin 1999).
First we need to consider first the `real' support ratios, that is the actual number of
taxpayers in relation to aged dependant people. In making such calculations we need
to take into account the future reduction of dependency arising from the decline of the
youthful dependant population. Actual retirement ages today is are already
substantially below `official' retirement age, about 58. Early retirement, late entry
into the workforce and modest workforce participation rates already give actual
support ratios of about 2.5 taxpayers per pensioner in the UK., and even fewer
elsewhere, not the nominal 4.1 of the potential support ratio (Government Actuary's
Dept 1999). This lower ratio is being managed today without notable problems.
No one management factor can ameliorate the situation all by itself except
with considerable discomfort. We therefore need to address simultaneously as
many of these contributing factors as possible.
First, there is considerable scope for increasing the actual workforce within the
existing boundaries of working age. European countries have very high levels
of demographic reserves. In the EU, only 62% of the nominal `working'
population aged 15-64 is economically active. This is the lowest of any major
industrial area in the world. In Spain and Italy, only just half the population
aged 15-64 actually has a job (Table 6). Mobilising this spare capacity, in
which existing foreign population is disproportionately over-represented
already, would address many social problems as well as helping to meet
workforce needs (see Punch 2000). Because of low current employment rates,
the potential workforce in Germany is sufficient to meet its likely needs until
2020 (Fuchs 1995). An increase of EU workforce participation rates to the
levels already achieved in Denmark, for example, would add 32 million
37
people to the EU workforce (Figure 17). Similarly, a return to the levels
actually achieved among men in the 1960s, would go a long way to meet
adverse future ageing changes.
Combining Eurostat projections of populations and of modest increases in
workforce participation, it can be shown that he actual workforce in all EU
countries will continue to increase for the time being and in all cases except
that of Italy will not decline below current level before 2020 (Feld 2000).
However, improvements in workforce participation rates cannot have further
enhancing effect once they have reached their maximum level, beyond say
2020. Labour market reform is also needed to permit a gradual, not an abrupt,
transition from work to retirement and the moderation of the tax penalties for
working after pensionable age, and less pressure at work for older workers.
The most potentially effective measures relate to later retirement age, which
improves both sides of the dependency equation. While formal retirement age
is 65 in many EU states, actual retirement age is about 58 or 59. Preservation
of today's actual support ratio would require actual retirement age to rise by
between 5 and 6 years, to 66. On that basis, managing the additional costs of
Table 7 Employment and Unemployment, selected countries 2001, ages 15-64
Employment rate Unemployment rate
both sexes males females both sexes
Denmark 75.9 80.2 71.4 4.2
France 62.7 76.0 55.7 8.6
Germany 65.7 72.6 58.7 7.8
Netherlands 74.1 82.7 65.3 2.1
Spain 57.5 70.9 42.7 10.4
Switzerland 79.1 87.6 70.6 2.5
UK 71.6 78.2 64.9 4.7
Source: Eurostat Labour Force Survey 2001 t 13, t 53
Figure 17.
Potential increase in EU 15 workforce , 1999,
given Danish participation rates (millions)
0
5
10
15
20
25
30
35
40
Economically active Actually employed
all males females
Source: Calculations based on Eurostat data.
38
elderly dependency simply requires people to stop work when they are
'expected' to. If 65 is taken to be the actual current retirement age, the current
support ratio would be fully preserved with an increase of retirement age to
between 72 and 76, depending on the population. Full preservation of the
support ratio is, however, neither necessary nor, probably, possible. Neither
would it do to minimize the difficulties of substantial increases in average
retirement age, especially for manual workers, when popular pressure is to
reduce it or to take early retirement. But s expectation of life after 65
continues to increase, working life must keep pace with it. Fortunately, recent
findings suggest that most additional years of life are active years. In the
longer run, it will not be possible for populations to live longer without
working longer.
Enhanced productivity growth , highly desirable in their own right, would
rectify much of the threatened economic shortfall and would in any case e a
natural economic response to a relative shortage of workers. The European
Commission's Annual Review of the Demographic Situation in Europe in
1995 (European Commission 1996) noted that additional productivity growth
required to meet the additional demands on the economy created from
pensions would between 0.1% and 0.3% annually up to 2005, increasing to
0.5% per year by 2025. That apparently small increase is, in terms of
productivity growth, substantial, and not easy to achieve. Economic diversion
on this scale to pensions costs would, for example, reduce a real annual GDP
growth rate from (say) 3% to 2.5%. Similar conclusions have been reached by
other economists in the US (Lee et al 1988, Lee 2000).
All these measures together could restore the future position to about the
current level in most European countries at least up to 2020, (UN ECE 1999).
However it would not do to minimize the political problems obstructing
necessary reforms. In many European countries, workforce participation is
constrained by poor support for (would-be) working women with families, by
interminable degree courses (favoured in the 1980s to disguise youth
unemployment), by pressure to keep and even expand early retirement. As
most of the population has a vested interest in them, there is great reluctance
to accept the revision of unsustainable generous pay-as-you-go unfunded
pension schemes giving up to 80% of wages over a long retirement, which are
commonplace in Europe. In the last two or three years general strikes in Italy,
transport disruption in France, political turmoil in Germany provoked by the
most modest proposed future reforms, all show how difficult it is going to be
to wean Europe's workers off the state they have enjoyed during the fat years
of the late 20th
century.
Immigration and its discontents.
Migration ­ the new engine of demographic and social change
Migration is now in the front line of demographic analysis. Twenty years ago
migration flows into Europe were ebbing. All that has changed. In 1992 international
migration to Europe reached an all-time peak, estimated at an inflow of 2.7 million
from all sources (Widgren 1994) a precarious estimate including illegal immigration ,
39
an estimate which has not, however, yet been dethroned. Net inflow to the European
Union as estimated by Eurostat only briefly fell below 0.5 million in the middle 1990s
and since then has risen to nearer one million. This net figure is similar to the gross
inflows to the United States, although into a larger population. However, migration
can go down as well as up (Figure 18) and some national trends underline this - in
1997 and 1998 net international migration to Germany became negative, as Polish
worker recruitment fell and many Yugoslavs granted temporary protection in 1992
returned home. The flows that make up this total, in and out, are in some cases quite
separate and independent although most have tended to rise. Some flows have been
inherited from the past; others are quite new.
The flows are too complex to describe in detail here. Excellent overviews are
available in respect of the Council of Europe countries (Salt 2003) and for OECD
member states (OECD 2003), together with two new global compendia: from the UN
Population Division (2003) and from the International Organisation for Migration
(2003). Comprehension of migration processes is muddied by the poor and
incompatible statistics available (Salt et al 1993, Poulain 1996), by the uncertainties
Figure 18.
EU 15 net foreign immigration 1960 - 2001 (thousands)
-600
-400
-200
0
200
400
600
800
1000
1200
1400
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
thousands
of migration theory and by the great difficulty of making projections on any rational
basis (see Hilderink et al. 2001). Migration is powerfully affected by the economic
and demographic situation in the sending countries, by labour demand in the host
countries, by the political situation in each and the historical connections between
them. The size of immigrant groups already in the host country and their propensity to
perpetuate chain migration by arranged marriage and attracting and absorbing illegal
or irregular immigrants is becoming more salient. Political effects on migration flows
have tended to be ignored by economists preoccupied with their models (Hollifield
2000).
40
Labour migration
Labour migration is the focus of debate by economists and in the media. Most
immigrants to Europe and the US are not, in fact, workers, but nonetheless labour
flows promote and perpetuate subsequent flows of dependents and spouses. Since the
later 1990s gross labour migration inflow to Europe, long-term and temporary, has
increased despite the economic slowdown in the Euro zone (Salt and Clarke 2003).
Totals of work-permit immigration from those countries which provide data rose from
about 200,000 in 1989 to a peak of 650,000 following German re-unification. Then
the inflow fell to 400,000 in 1995, rising again to over 500,000 by the end of the
decade (Figure 19). These are all gross inflows; few data on net flows of labour are
available. This is all movement from outside the EU; movement between EU
countries does not require a permit. The reasons are mixed. The Information
Technology sector needs labour because its rapid growth has outstripped the ability of
domestic resources to train suitable personnel. On the other hand, skilled medical
workers, especially nurses, need to be recruited overseas more as a reflection of poor
working conditions arising, from under-funding over many years, creating a
dependency on foreign workers. There the problem is not so much a shortage, as
many trained nurses no longer work in the profession, but `a shortage of nurses
willing to work under the conditions being offered them' (OECD 2003 p 23).
Believing that it might improve their competitive position, many European countries
have changed their migration laws in the 1990s to favour the recruitment of skilled
professionals overseas (OECD 2001). However the interest may be transient and the
results disappointing: IT jobs were withdrawn for the official UK list of shortage
occupations within 18 months despite official claims that 200,000 were needed. The
German `green card' for overseas IT specialists remained under-subscribed, until
2001 anyway. The number of foreign nurses in the British NHS remained constant
despite substantially increased recruitment. At the other end of the scale, recruitment
of unskilled workers, often on a temporary basis, expands to work in agriculture,
building and domestic services, despite persistent high unemployment in Europe
particularly among foreigners themselves. Unregulated, casual trades also attract
many illegal immigrants, who have also become prominent in prostitution and the
drugs trade.
Figure 19.
Gross labour Inflows, selected Western European countries 1989 - 2000
source: OECD 2003 table A.2.1
200
250
300
350
400
450
500
550
600
650
700
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
grosslabourinflow(1000s)
0
10
20
30
40
50
60
70
80
percentageofgrosslabourinflowtoGermany
Gross labour inflows, 1000s)
Percent of inflow to Germany
41
Some European governments, notably that of the UK, have been persuaded that a
greater inflow of potential workers benefits the economy even if they are not recruited
to specific jobs; accordingly all kinds of conditions for immigration have been relaxed
with a substantial increase on inflows to the UK. Others have maintained a more
selective approach and have not opened their doors to the new EU accession
countries. The economic evidence on which this is based seems rather tenuous and
some of it is negative. For example, immigration can be shown to increase both local
(Dustmann 2003) and remote (Hatton and Tani 2003) unemployment. Immigrants
naturally increase overall GDP insofar as they increase population but may not
increase GDP per head, which is what matters from the viewpoint of individual
welfare. Some studies, e.g. in Sweden, show that all immigration taken together
imposes a net economic cost on the country; others show benefits to the middle class
but a reduction of income among the less well paid, e.g. Denmark and the USA
(Storeslettten, Wadensjo 1999). In general, immigrants from poor countries tend to
impose a net cost on their host societies; the net benefits tend to come from
immigrants from developed countries. Furthermore workers leave as well as enter.
The net effect of migration on the labour force is generally unknown; many countries
lack emigration data. But in the UK, which does collect departure data on a sample
basis, outflow can be substantial. During the 1990s, almost as many persons left the
UK as entered it for purposes of work, and the net inflow of labour migration to the
UK was only 10,000 in 2000 and 36,000 in 2001 (when 90,000 UK workers
departed). One wonders why.
Figure 20. Source: OECD 2003 chart 1.2, reproduced by permission. For notes see original.
0 20 40 60 80
Australia 2
Switzerland
United Kingdom 3
Portugal
Canada
Norway 4
Denmark
France 5
United States 6
Sweden 7
Workers Family reunification Refugees
100
42
Family migration
Chains of migration pulled in by existing immigrant communities (OECD 1978),
often founded by workers, have given a momentum to flows which then owe little to
economic considerations. Since the 1970s most legal migration into Europe and North
America - up to 80% - has been not of persons arriving ostensibly to work, but of
their dependants, relatives, spouses and new marriage partners, as well as persons
arriving independently as students and more recently as asylum claimants, many of
whom of course contrive to stay one way or another. The magnitude - and the variety
­ of this pattern is seen in Figure 20. For example, in France in the early 1990s, over
80% of migration from North Africa as family or humanitarian, into Austria about
90% (Biffl 2003). Even in North America, where the economic effects of migration
are most loudly praised, the priority given to `Family Preference visas' ensure that
labour visas, even though recently increased, still account for only 20% of legal
entries (163,000 H1B visas were issued in 2001, out of a quota of 195,000. And in
Canada the same priority reduced the economic migrants to a variable and residual
category within the quota until recently.
In the late 1980s and thought the 1990s, the growth of immigrant minority societies in
Europe, thought immigration and strong natural increase has created powerful poles
of attraction for further migration through marriage and other aspects of chain
migration. Minority populations from North Africa, the Middle East and Asia,
Muslims and Hindus, tend to prefer arranged marriages to persons of the same
religion, race, nationality, caste and in the case of (e.g.) Pakistani Muslims, close
relatives such as first cousins or nieces. Among many groups, life in the West has not
weakened these preferences. Immigrants may cling to traditional ways more tightly
than those who remain in the country of origin. Some, notably Muslims, reject or
distrust Western secularism and prefer to import brides of guaranteed provenance and
purity, free of liberal and disturbing Western ideas and attitudes, who have had no
opportunities to disgrace family honour. Such migration also increases the size, and
the local power, of immigrant populations. This process is notable among Turks but
less among Moroccans in Belgium (Surkyn 1998), among Turks but less among
Algerians in France (Tribalat 1995), among Pakistanis and Bangladeshis in Britain. In
the Netherlands, marriage migration is now the most important route of entry (except
asylum) for Turks and Moroccans (Figure 21).
This represents a failure of integration and its practice erodes it further . Chain
migration slows down demographic change and integration, high levels of in-marriage
are associated with other indicators of enclosure or integration (Lesthaeghe et
al.1998). The intensity of marriage migration also responds to the desire, among large
numbers of people in the third world to immigrate to the West. There is widespread
advertising for this purpose in the press and various rackets in the host countries. The
rise of spouse migration has created a powerful and accelerating dynamic in migration
to Europe, which in the late 1908s and 1990s has eclipsed the migration of established
spouses in immigration streams to Europe, for example in the Netherlands and in the
UK, especially after the abolition of the `primary purpose rule' of checks intended to
prevent the abuse of family migration. Nowhere have the expectations of integration
been more frustrated.
43
Asylum and illegal migration
The discovery by intended migrants from poor and troubled countries of the world of
an open-ended commitment made fifty years ago to clear up post-war chaos and
distress has opened the way to a cumulative total of over 6.6 million asylum claims
made in Europe since 1980 (because of duplicate claims, the number of individuals
will be smaller). That total is about the same cumulated number as of labour migrants
since 1980. While most claims are rejected, it is believed that most claimants
nonetheless manage to remain in Europe. This parallel migration stream has greatly
augmented flows into Europe and diversified the foreign-origin populations in all
Figure 21.
Source: data from Home Office
Spouse migration by sex and NC origin, UK 1973-2001
0
5000
10000
15000
20000
25000
30000
35000
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
All husbands All wives NC wives NC husbands
European countries. Patterns and problems of asylum have been treated in depth by
Penninx (2003) for this meeting and need no repetition here (see UNHCR 2003).
The most salient conflict for Europe, in former Yugoslavia, paradoxically produced
few actual asylum claims at the time, although many since. The usual asylum
screening was clearly inappropriate and would have been overwhelmed. Most of the 2
million or so given shelter were given `temporary protection'; many have indeed
returned to their former country, if not their former homes. While claims from
particular countries ebb and flow in relation to the political situation, once started
from a poor country they persist, and new flows constantly arise. Yugoslavia apart, it
has not been easy to make a clear connection between the intensity of political unrest
in the world and changes in overall asylum claims.
Europe is in a state of confusion over asylum. No political systems want claimants;
they are seldom welcome in large numbers to the mass of the electorate. But moral
principles of governments, their adhesion to international agreements, pressure group
44
agitation, all over-ride popular opinion. In persistently affirming the commitment to
asylum, governments send a welcoming message to the world which in practice they
Figure 22.
Migration trends to Western European countries 1980 - 2002, various
definitions (1000s)
0
500
1000
1500
2000
2500
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
thousands
gross inflow including some asylum (SOPEMI) net immigration EU15
asylum (W Europe) gross inflow EU 15, non-nationals
do their best to frustrate through visa systems, action against traffickers and continual
attempts to simplify, and usually make stricter, the asylum system. Yet in the 1990s
no European country has seriously considered the question of radical reform of the
asylum principle. Most claims are rejected, although relatively few rejected claimants
are believed to be removed. Asylum continues to be a process of mass population
movement, as never intended, as well as a means of protection.
Asylum claiming is closely connected with illegal entry and over-staying; many
claimants enter illegally and claim at a later date when apprehended. Those apart, a
variety of uncertain approaches (IGC 1995, Widgren 1994, Salt 2000) contribute to a
consensus of between 350,000 and 500, 000 illegal entrants to Western Europe each
year, with an illegally resident population of perhaps 3 million. Rather more secure
estimates for the US put the annual inflow at about 350,000 and the stock at 5
million.. Repeated amnesties secure the position of earlier illegal immigrant cohorts
and do nothing to reduce further inflows (Orrenius and Zavodny 2003).
Changes in attitudes and policy, and their effectiveness.
Policy is more important in determining migration patterns than any other component
of demographic change, and no phenomenon related to demographic change has had
more impact on politics. But the political pressures are contradictory, and policy may
change radically following elections which remove one government and replace it
with another. Conflicting beliefs confuse the picture. Popular opinion in general
opposes mass migration, especially from outside Europe, but political elites take a
more permissive view. Immigrant and race relations pressure groups can exert more
power than these diffuse concerns and can focus it effectively on Ministers. Hence
liberal democracies do not find it easy to control immigration (Freeman 1994). As
noted above some governments have been persuaded of the economic merits of mass
migration, the UK government has even adopted American rhetoric in praise of the
cultural diversity which mass migration brings with it. It is often claimed that
45
immigration is in any case impossible to stop or reverse. A conviction of impotence is
thus allied with economic self interest in the new idea of `managed' migration
(Council of Europe 2000, IOM 2003).
These elevated sentiments seldom take into account the social and ethnic
transformation of urban areas in Europe, not those on the whole where elites live,
which disquiets popular opinion. It is customary to dismiss such concerns as `racism'
and `xenophobia'. In the last few years, however, electorates have made their views
felt by supporting right-wing or far right wing parties, shocking the political
establishment in North Italy, in Austria, in the French Presidential election, in the
Dutch general election of 2002, nearly overturning the German government in 2003
and stalling its new, more open migration law, returning a centre-right government
with a restrictive immigration agenda in Denmark and disturbing consensus even in
Switzerland. Even where the right-wing parties have failed or later fallen apart, much
of the drift of their policy is absorbed by mainline parties. Even in the UK, a
government still devoted to talking up and stoking up legal migration has adopted
asylum legislation (the fourth since 1993) of a severity which could not have been
contemplated by its Conservative, supposedly more right-wing, predecessor.
The growth of all forms of migration, new and growing migration pressures from
abroad, the changing political situation at home and abroad, the ascendancy of the EU
over national interests from 1998, has provoked a continual series of new laws. Some
are designed to regulate and restrict, especially asylum, although others are
permissive. Former communist countries have had to start from scratch, as have
former emigration countries. For example new and generous Irish legislation of 1996,
formalising entitlement to asylum in Ireland for the first time, has put the Irish
Republic on the asylum map to such an extent that it now has one of the highest
inflows per head of population in Europe, giving Irish society with a wholly new
multiracial element. The UK government's incorporation of the ECHR into English
law has further strengthened the hand of judges against ministerial attempts to stem
asylum flows, assisted by numerous pressure groups funded by the government funds
and public lotteries.
Finally, migration has gained further prominence in political debate a through a volte-
face in attitudes to it by many of Europe's governments and media. As European
population growth has started to run out of steam and population ageing has risen up
the political agenda, opinions have become more complicated. Originally `unwanted
foreigners', immigrants are now also regarded by some as demographic salvation;
needed to supply a missing workforce, provide care for the growing army of elderly
and in general save European populations from the trouble of reproducing themselves.
Unique events ­ war and state collapse
The biggest migration surprises were provoked by the unique political crises in
Central and Eastern Europe in 1989, in the Soviet Union in 1991 and in Yugoslavia in
1992. One of the surprises was that the expected mass-migration of newly liberated
Eastern European and former Soviet citizens did not materialise. Migration rates have
certainly increased, however, especially through short-term work schemes between
Poland and Germany, increased work permit migration, students and illegal
immigrants everywhere, including the most surprising places (numerous Russian
prostitutes in Istanbul, for example). Over the decade, CEE and FSU populations in
46
Western Europe, formerly sparse, have increased substantially, although they still
only comprise 0.1% to 5% of total population (net of naturalisation) (Table 8). The
higher percentages in Table 6 arise from guest-worker populations of long standing
from former Yugoslavia. The US and Canada have also gained new CEE and FSU
population, including most of the emigrants from Armenia and Georgia.
Table 8.
Population stocks - citizens from Central and Eastern Europe in selected OECD countries 2000
and all foreign citizens in selected CEE countries (thousands and percent)
citizenship birthplace
Austria Germany Italy N'lands Sweden Switzerl'd Canada US
(workers) 1999
All foreign 242.2 7296.8 1388.2 651.5 477.3 1384.4 4971.1 19767.3
CEE as % foreign 58.5 19.9 11.7 3.7 15.3 24.5 10.4 5.7
All foreign as % total 10.5 8.9 2.4 4.1 5.4 19.3 17.4 7.9
CEE as % total 6.1 1.8 0.3 0.1 0.8 4.7 1.8 0.4
Bulgaria Czech R Hungary Poland Romania Slovak R
total foreign
population 101.0 204.7 128.4 39.4 69.5 24.8
percent of total pop 1.2 2 1.3 0.1 0.3 0.5
source: OECD 2003 table 1.24,
1.25
note : these totals will be considerably under-stated by naturalisation in the case of Western countries,
and by non-registration of illegal immigrants in both but especially in the CEE countries
Ethnic migration
Perhaps surprisingly, ethnic migration continued to dominate migration in and out of
the Eastern part of the area following the opening of frontiers and the break-up of old
collectivities. Some of these flows were continuations of flows fitfully established
during the Cold War: aussiedler to Germany, Jews (especially from Russia) to Israel
and America. New ethnic flows, previously prevented by the Iron Curtain, have also
arisen, notably of gypsies and others claiming asylum in Germany. The negotiated
return of 30,000 of them to Romania proved to be an expensive business. A new CEE
gipsy presence is seen throughout European cities, and is likely to grow as the large
gipsy minority is one of the few components of CEE population with vigorous rates of
natural increase, and with many reasons for wanting to migrate.
A real surprise was the rise of Russia, for a while, to become Europe's biggest
immigrant destination, around 1994. The end of the Soviet Union left the 24 million
people of Russian `nationality' in newly independent post-Soviet republics, the
natives of which were by no means all favourably disposed towards their old masters.
Migration to Russia, perhaps a third of it forced, topped a million a year though had
now fallen. This flow moderates Russian population loss, partly compensating for the
 million deficit of births over deaths.
The effect on the sending areas too should not be forgotten. During the 1990s, ancient
Jewish communities were being rapidly eliminated by emigration. The German origin
settlements in Mitteleuropa and further East, which dated from the 18th
century, are
47
coming to an end with results by no means uniformly happy. The population of
Armenia has declined by a third.
If that old diversity of the CEE is being lost, it was more than compensated in the
1990s by the arrival of entirely new populations. These are still small in number,
officially between 0.1 and 2% of total population, although the official figures are
likely to be substantial under-estimates (Table 8). More open borders to the West,
and the dissolution of previously formidable border security, have attracted
considerable migration flows from the South and the East, creating completely new
Asian and African populations in Eastern Europe. Initially, most were seeking easier
entry to the more attractive Western economies, where they can work illegally or
claim asylum. That in turn increased immigration flows to Western Europe,
particularly to Germany, which has taken about half Europe's legal immigrants in the
1990s and up to two-thirds of its asylum claimants. Many of these transit migrants,
most with illegal or irregular status, baulked at entry to the West, may become
permanent residents. Further, the CEE countries have felt obliged to accede to the
Geneva Convention at various times in the 1990s, preparing to re-enter the
mainstream Europe of the Council of Europe and the EU, and for the first time these
countries have started generating statistics on asylum claimants and refugees (OECD
2002). That will reinforce these new and unexpected flows of displaced persons and
refugees from similar origins. Once settled, further growth can be expected from the
arrival of dependants, spouses and new partners.
Estimates of the likely long-term added inflow from new accession countries to the
EU arising from the free movement provisions are very mixed. For example Fertig
and Schmidt (2001) made astonishingly exact predictions of between 66,740 to 73,
583 immigrants to Germany per year from the CEEC-10 countries from 1995 ­ 2015.
Other estimates range between 20,000 and 200,000 per year. The official view from
the European Commission (2001, and Boeri and Buecker et al. 2000) is that
enlargement will lead to an eventual migration of about 3 million from the new
accession counties plus Bulgaria and Romania when they eventually join. Another
study suggested that without transitional restrictions between 4 and 5 million would
eventually migrate to Germany alone over 15 years from all the CEE countries. Salt et
al.1999 concluded that it was impossible to quantify the scale of future immigration,
but that the most likely figure lay between the broad margins of 55,000 and 278,000
per year from all the accession states to the rest of the EU ­ the latter being a
substantial increase in present flow. The higher of these figures is in line with the
cumulative totals cited above. The poor economic and integration situation in the
Euro zone in 2003 prompted most member states to impose a staged block on free
CEE migration for up to 7 years. A bigger flow than expected, therefore, will move to
the UK and to the Republic of Ireland, the only remaining countries that continue to
offer immediate access.
Parallels between this migration and that from Southern Europe in the 1980s are not
appropriate. Once free movement from Spain, Portugal and Greece was eventually
permitted, inflows from those countries were modest. But they had already had a
previous major guest-worker migration in the 1950s and 1960s. Their GDPs per
capita, about 50% of the EU average in the 1960s, had risen to  by the time the
migration restrictions were lifted. By then the guest-worker migration was in reverse
and migration pressures had disappeared. None of the is true in the CEE countries,
48
where pc Gross National Income is still just under half the EU level (47%) on a
purchasing power parity basis (World Bank 2002; European Commission 2003),
where there are large and unproductive agricultural populations and, with the
exception of Poland, where guest-worker migration is still at an earlier stage.
Unemployment in the new accession countries is even higher than in the Euro zone, in
Q4 2002 being 18% in Poland, 17% in Slovakia, 9% in the Czech Republic and 8% in
Hungary (UN ECE 2003 Table 3.4.2)
The demographic effect of migration ­ immigrant, foreign and ethnic minority
populations in 2000
About 175 million people throughout the world now live in a country that is not their
birthplace (UN 2002). Of these, about 106 million live in the `developed world': 56
million in Europe (8.5% of population), 46 million in the countries of the English-
speaking world overseas (13.8%, mostly ­ 35 million ­ in the US) and 4 million in the
Asian industrial countries. The `European' figure of 56 million, however, is
misleadingly high. It includes over 24 million residents of the republics of the former
Soviet Union. Most of these were born in other republics of the former Soviet Union,
very few from outside its borders. Until 1992 few would have been classified as
having been born in a foreign country. The events of 1992 transformed their status
overnight into that of foreigner.
In about 2001, Western Europe had the largest number of immigrants in any region of
Europe; that is persons born abroad (18.8 million) and the highest proportion relative
to population (10.3%). France and Germany have 6 and 7 million respectively, about
Table 9. Foreign Population 1971 ­ 2001 (thousands and percent)
1971 1981 1991 2001
Country thousands % thousands % thousands % thousands %
Austria 195 2.6 288 3.8 439 5.7 761 9.4
Belgium 663 6.9 861 8.7 905 9.1 862 8.4
Denmark 100 2.0 102 2.0 161 3.1 259 4.8
France ... ... 3714 6.9 3608 6.3 3342 5.8
Germany 3054 3.9 4453 5.7 5343 6.7 7297 8.9
Italy 122 0.2 211 0.4 566 1.0 1271 2.2
Netherlands 247 1.9 521 3.7 692 4.6 668 4.2
Norway 76 2.0 83 2.0 143 3.4 184 4.1
Spain 148 0.4 183 0.5 279 0.7 896 2.2
Sweden 411 5.1 422 5.1 484 5.6 477 5.4
Switzerland 1080 17.4 915 14.4 1130 16.7 1424 19.8
UK ... ... 1638 2.9 1892 3.3 2580 4.4
Sum 6097 13390 15641 20020
NB France, Greece, Portugal UK, 2001 data Labour Force Survey. Italy 2001 data 2000
Source Council of Europe 2002 t 1.8 and Eurostat 2002
49
10% of the population as in the Netherlands (1.6 million). That is almost as high as
the proportion of immigrants in the US (12.4 %). In Switzerland and in the small
population of Luxemburg, over 20% were born abroad, about the same as in the
countries of the English-speaking world overseas (e.g. Canada 19%).(Table 9)
In Northern Europe, demographically dominated by the United Kingdom, the foreign-
born population comprises a slightly smaller proportion of the total; 7.8%. Latvia and
Estonia have much higher proportions as a result of planned `Russification' in Soviet
times. The rise of Ireland's foreign-born population to 8% is new; a proportion are
returned expatriates of Irish origin.
These data on birthplace ignore citizenship. Most immigrants are foreign citizens..
Returning citizens, or those entitled by ancestry to citizenship form only a small part.
East European aussiedler of (often remote) German origin, mentioned above, have
moved to West Germany in large numbers since the Basic law established their
entitlement under the ius sanguinis: 2.3 million from 1987 to 1996. Their return
peaked as recently as 1992 and continues (105,000 in 1999). They are not included in
statistics of the foreign population but they are included in the statistics on
naturalisation. The Southern European countries share similarly low proportions of
immigrant population according to official figures. However the 2-5% reported for
Italy, Portugal and Spain, conceal a large number of illegal residents.
Citizenship and naturalisation.
Citizenship is the usual indicator of immigrant status in continental Europe, and a
misleading one, in both directions. Naturalisation makes foreigners disappear, often
in large numbers. It accounts for the apparent absurdity of the decline in the foreign
population in France from 1982 (3.71 million) to 1999 (3.26 million), in the
Netherlands from 1990 (692,400) to 1999 (651,500) and in Belgium over the same
period. It might be argued that this is only appropriate, as foreigners acquire new
citizenship by having shown a commitment to and knowledge of the new homeland.
This transubstantiation, however, may be very partial, especially in those countries
were naturalisation is seen not as a reward for assimilation to the local culture but
rather as a mechanism to encourage integration. Naturalisation has been powerful in
reducing the apparent size of the foreign origin population (Table 10). In Europe in
the mid 1990s, the proportion of foreign-born population who retained foreign
citizenship had fallen to 48% in the Netherlands while remaining as high as 92% in
Denmark.
Table 10 Foreign and immigrant population compared, mid 1990s
Population Born abroad Born abroad Foreign nationality Foreign as %
Country (millions) (thousands) (percent) (thousands) (percent) of immigrant
Denmark 1996 5.236 259.2 5.0 237.7 4.5 91.7
Netherlands 1995 15.531 1407.1 9.1 679.9 4.4 48.3
Norway 1996 4.381 246.9 5.6 157.5 3.6 63.8
Sweden 1996 8.841 943.8 10.7 526.6 6.0 55.8
Great Britain 1991 54.889 3746.1 6.8 1791.0 3.3 47.8
Sources: OECD 1998 table B.1.5, B.1.6; Council of Europe 1998, UN 2002
50
The broader net effect of international migration, the growth of the immigrant-
descended population, would be reflected by citizenship statistics if the children of
foreign immigrants likewise retained the foreign citizenship of their parents. That
would give an estimate analogous to that of the `ethnic' population categories used in
the English ­ speaking world including the UK, but avoided elsewhere in Europe with
the partial exception of the Netherlands.
Growth of immigrant and minority populations
Many immigrant ethnic minority populations come from countries with incomplete
fertility transitions. Immigration and high fertility, allied to a youthful age-structure,
have generated rapid growth. In 1951 the UK ethnic minority population, judging
from birthplace data, is likely to have been below 50,000. By 1971 it had grown to 1.4
million and by 2002 to 4.7 million, an annual growth rate during the 1990s of 2.9%.
In 2001, over half was born in the UK and the rest, including most adults and heads of
household, were born abroad. In the US the most dramatic growth has been that of the
Hispanic population. Most are of more recent immigrant origin. When the `Hispanic'
category was first introduced into the US Census questionnaire in 1970, 9.1 million
were enumerated (4.5%). By 2000 this had grown to 35.3 million (12.5% of the total
population), with the biggest ever increment (57.9%) occurring in the last intercensal
period since 1990 (US Bureau of the Census 2001b), an annual rate of 4.5% per year.
This growth arose mostly through immigration even though the Hispanic TFR is
almost 50% higher (2.95) than the US average (2.03). The Asian population, entirely
of immigrant origin, grew from 6.9 million, (2.8%) in 1990 to 10.2 million (3.6%) in
2000 (US Census Bureau 2002). In the US 2000 census 63.1 million people identified
themselves as belonging to a non-white racial group (22.4% of the total population).
An additional 6.8 million people (2.4%) claimed descent from two or more races;
about 80% `white and other' (US Census Bureau 2002a, Goldstein and Morning
Figure 24.
Projected growth of population of immigrant or
foreign origin 2000 - 2050 as percent of total
population
0
5
10
15
20
25
30
35
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
percent
Germany medium variant USA medium variant
Netherlands base scenario Denmark base scenario
UK high migration
Sources: Alders 2001a,b, US Bureau of the Census 1996, Think Tank on Integration 2002, Ulrich
2001.
51
2002). In the UK, 18.7% of the ethnic minority population aged 0-15 in 2000 was of
mixed origin, mostly white / Caribbean. (Coleman and Smith, in preparation).
Current levels of immigration into many Western counties, together with the
relatively higher rates of increase of some populations of immigrant origin, are
transforming the structures and compositions of those societies. Population
projections of foreign-origin or ethnic minority populations are available only for a
few countries (Figure 24). They show a striking similarity in the rate of increase of the
ethnic or foreign origin population, to reach between 20 and 30 percent of the total
population by mid-century and for the most part still increasing rapidly. The foreign
origin populations here include foreigners of Western origin as well as non-Western:
about half the starting total in the case of the Netherlands, for example. These are
projected to increase only slightly before stabilising or declining. The native
population of those countries would remain static or decline slightly over the same
period. In most of the projections, higher immigrant birth rates are assumed to
converge on those of the general population. It is continued immigration that makes
the difference. Variant projections with net zero migration assumptions show only
modest further increases in foreign population (Figure 25). If immigration to the
developed world continues at recent levels, the ethnic and social composition of the
population of many European countries, as well as that of the US, will be radically
and permanently transformed.
Concluding remarks
This paper has covered a wide area and does not address a single thesis. These
concluding remarks will address a small selection of topics, and attempt to bring
together some of the themes discussed above.
Figure 25.Projected growth of population of immigrant or foreign origin
2000 - 2050 as percent of total population, with zero net
migration
0
5
10
15
20
25
30
35
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
percent
Germany medium variant USA medium variant Denmark base scenario
Germany zero mig US zero mig Denmark zero mig
Sources: Alders 2001a,b, US Census Bureau Census 1996, Think Tank on Integration 2002, Ulrich
2001.
52
The demographic turbulence created by the political and economic upheaval in
Eastern Europe at the beginning of the decade remains its biggest event of the decade.
In 1990 the huge impact of the collapse of communism in 1989 was only just
becoming apparent and the demise of the Soviet Union was still to come,
Czechoslovakia and Yugoslavia had not yet separated and Germany was not yet
unified. These areas have seen profound demographic upheavals and the jury is still
out on what the final outcome will be. Certainly the relative demographic, social and
political uniformity which characterised the communist bloc in 1990 has been
shattered and its component populations are moving apart in divergent directions;
some to a prosperity and stability which has taken them into the EU, albeit with some
holding of breath, others still languishing in a post-communist economic, political and
social tristesse of persistent high mortality, lowest-low birth rates and actual, not
prospective, population decline.
Further West, international migration dominated European population dynamics in the
decade, often in unlooked-for ways. While the great Soviet outflow did not
materialise in the expected way, the Yugoslav catastrophe forced two million to
move, helping migration reach a peak in 1992 from which it receded only to rise again
following the more resistible rise of globalised asylum-seeking. Perceptions have
changed and changed again. Migration, whose rapid rise in the early art of the decade
provoked unprecedented across-party cooperation in Germany to stem it, lives under
two parallel and incompatible pressures. Elite opinion favours more migration to
promote the economy and gives high priority to preserving the principle of asylum.
Mass opinion in Europe increasingly opposes it as communities change visibly under
its impact.
In terms of public and political perception, the 90s was the decade of population
ageing, as media, governments and pension systems finally woke up to the long-
heralded prospect of the unsustainability and insolvency of their retirement, labour
market and pensions arrangements. Never has the discipline of demography enjoyed
such favour from the media, even if its reward is to be vulgarised as `demographics'.
Fads have come and gone. The concept of `replacement migration' to prevent
population ageing, was shown by UNPD to involve immigration and population
growth levels implausible even to the most enthusiastic. Yet these hypothetical
calculations were taken as necessary prescriptions by serious politicians. Saner
counsels have now abandoned these wilder shores of demography. However some
governments, mostly English speaking, cling to the notion that immigration offers
economic, if not demographic salvation and have promoted it to record levels in the
last few years. Meanwhile most of Europe tries to moderate flows except those of the
highly-skilled.
The conceptual landscape in demography has changed radically. The second
demographic transition maintains its progress. The trends that it describes have
become more prevalent in the developed world and more popular as a scientific
paradigm. Only formulated in 1987 to account for the rise of cohabitation, divorce
and illegitimacy amidst the ebbing of marriage, this concept has seen its predictions of
the globalization of such behaviour in the modern world at least partly vindicated. In
gaining wide acceptance, it has promoted a whole new industry of demographic
research based on ideational change as the driving force. New techniques have been
developed for the analysis of the life course, for the `correction' of period fertility
53
measures, for probabilistic population projection, for the integration of macro and
micro scale effects, which have transformed the analytical landscape. Accordingly,
demography has become much more complicated in the 1990s, and only limited areas
can be explored here. We know much more about the workings and relationships of
its component parts. But the future remains an unknown country.
In the meantime, Europe has been radically re-arranged. The fall of the iron curtain
and the collapse of the Soviet empire and various of the inventions of Versailles and
Trianon ­ Czechoslovakia and Yugoslavia - have greatly increased the number of
European countries even while Europe's overall population is on the edge of decline.
Only one has disappeared, East Germany being swallowed by the Federal Republic.
Fortunately the communist sub-national republics provided good statistical series. The
new re-arrangement of European countries by the UN may make geographic sense but
has created some very heterogeneous regions, as Northern and Southern Europe have
both acquired new and relatively poor former communist members. The Russian
republics of the Soviet Union, formerly a block unto themselves, now join Europe,
taking its population to (750) million and its boundaries from Rejkyavik to
Vladivostock.
Convergence and Divergence in the UN ECE area
We might ask if any tendency is apparent for the diverse populations of the UN ECE
area to converge on a common demographic pattern. So far the answer appears to be
rather little (Coleman 2002). Neither is it regarded as being inevitable (de Beer and
Van Wissen 1999). Modern countries maintain long-standing differences of many
kinds. These include political systems and electoral preferences, national
characteristics of various kinds (Inkeles et al 1996), values and attitudes (Inglehart
2000). Whether welfare institutions, which may have important demographic effects,
are converging remains controversial (Greve 1994, Gauthier 2002, Tomka 2003). One
striking examples of non- convergence is the persistence of large differentials in
standardized mortality rates from accidents and violence in 31 industrial countries,
especially traffic accidents, despite near identity in the cars driven, the standards of
road and signs, the use of seat belts, laws against drink-driving and so on (Adams
1985, Chesnais 2003). If symmetrical gender equity is an important determinant of
fertility levels (McDonald 2001) then convergence may indeed be slow, as such a
change in the fundamental relations between men and women would require a major
cultural transformation in Southern and probably Eastern Europe.
In the East, a unique political system has put these populations under un-natural
constraints for decades (see Lutz, Scherbov and Volkov 1994 , part III; Meslé,
Schkolnikov, Hertrich and Vallin, 1996). These constraints, summarised by Elwood
Carlson (in press) as the 'State Socialist Mortality Syndrome' have driven them off the
trajectory of improvement in survival which would have been expected earlier in the
century. On top of that is the recent increase in the death rate in many (not all) of the
countries in transition. The convergence imposed by communism is given way to
centripetal forces taking the former communist block in different directions, often to a
position closer to their former neighbours.
The achievement of low mortality in the West is far from uniform. Despite the
triumphs of oldest old survival, mortality is not in retreat everywhere. An early group
of low mortality achievers; The Netherlands, Denmark and Norway for example, have
54
failed to make much recent progress, and none at all among Danish women. The
statistical mediocrity of survival in very successful economies such as that of
Germany, with high expenditure on medical services, suggests that further
convergence cannot be expected simply on economic grounds (Vallin 1995). Models
of economic inequality (Wilkinson 1996) compete with dietary hypotheses. The
superior survival of the populations of the 'olive oil belt' of poorer Southern Europe
compared with the richer sausage beer and chips belt of Northern Europe confounds
economic explanation of mortality differentials, but does connect persuasively with
parallel advantages shown surprisingly in such places as Albania (Gjonça 2000) and
(FYR) Macedonia.
It may be that the 20th century saw so many upheavals that a forecast of the outcome
is beyond the reach of theory; the turbulence created in the 1990s has unhelpfully
muddied the demographic waters. The expectation for the future must be one of
constrained variety, more in some areas than others. Past trends may be some guide,
in that demographic characteristics do seem to be geographically located and some
salient characteristics today connect with the particularities of the past, for example
the familism and household patterns of Southern Europe (Reher 1998) the
Scandinavian history of births outside marriage, now shared by Estonia. The ancient
division of Europe by Hajnal's line, which outlasted the Ottoman and Russian
Empires and was if anything re-inforced by the Soviet one, looks as though it will not
survive the end of the iron curtain.
The demographic consequences of the EU and its expansion.
In the same decade that the Communist block collapsed, the European Union made
exceptional gains in acquiring additional countries as member states and additional
power over policies indirectly, if not directly, affecting fertility migration and health.
Likely consequences include the convergence of family-formation and family-
building behaviour under the influence of increasingly common welfare and
entitlement polices. The EU Commission believes that EU demographic and other
social characteristics are already converging (European Commission 1995) and will
need similar policy responses (Hantrais 1998) as a result. Demographic convergence
is already regarded as one indicator of a desirable harmonisation of European social
conditions. Although so far no European Directive relating to standard European
birth rates, death rates or household structure has been formulated, the possibility of
harmonized EU demographic optima has already been raised (Gesano, 1999). Since
the Amsterdam Treaty and the Tampere meeting, Migration and Asylum Policy has
moved from the `Third Pillar ` of responsibility of member states, to community
competence. A European migration policy has been proposed (European Commission
2000) despite the very different demographic and workforce conditions of different
European states (Coleman 2002). The Cohesion Fund and the Structural Fund are of
course intended to accelerate the removal of economic inequalities in the EU area,
between and within countries; other things being equal that must be expected to bring
family formation more into line.
However if the problems with the Euro continue, the overall economic effects may
not be favourable upon families. Furthermore the effects of EU fiscal and competition
rules on the new accession countries may, at least in the short run, be highly
disruptive to their economies and some areas of their economy, especially agriculture
and rust belt industries. The example of the former East Germany after nearly fifteen
55
years of unification is not encouraging. Nonetheless some wish to see the expansion
of the EU to embrace the more fragile economies and unsettled political systems of
the Balkans and even of Turkey. In that case its boundaries would include an even
greater share of the population of the Eastern hemisphere of the UN ECE region, but
its goals of harmonizing the economic and social institutions of the member states,
and thereby indirectly at least the behaviour of its populations, would correspondingly
diminish. Demographic diversity in Europe will be with us for some time.
56
References
Alders, M. 2001a: Bevolkingsprognose 2000-2050: recente ontwikkelingen in de
migratie en veronderstellingen voor de toekomst. Mandstatistiek van de bevolking
2001/3, 31-40.
Alders, M. 2001b: Allochtenprognose 2000-2050 de toenamevan het aantalniet-
westerse allochtonen nader bekeken. Maandstatistiek van de bevolking 2001/4, 29-
33.
Ahn, N. and P. Mira (2002). "A note on the changing relationship between fertility
and female employment rates in developed countries." Journal of Population
Economics 15(4): 667 - 682.
Andersson, G. (2000). "The Impact of Labour-Force Participation on Childbearing
Behaviour: Pro-Cyclical Fertility in Sweden during the 1980s and the 1990s."
European Journal of Population 16(4): 293 - 333.
slund, A. (2001). Building Capitalism: the transformation of the former Soviet Bloc.
Cambridge, Cambridge University Press.
Avdeev, A. (2001). The extent of the fertility decline in Russia: is the one-child
family here to stay? IUSSP Seminar on International Perspectives on Low Fertility:
trends, theories and policies. Tokyo March 2001.
Becker, G. (1981, 1991). A Treatise on the Family (First and second edition).
Cambridge, Massachusetts., Harvard University Press.
Bettio, F. and P. Villa (1996). A Mediterranean Perspective on the Breakdown of the
Relationship between Participation and Fertility. Discussion Paper No 5 1996. Trento,
Universit degli Studi di Trento Dipartimento de Economia: pp 54.
Boeri, T. and M. Keese (1992). "Labour Markets and the Transition in Central and
Eastern Europe." OECD Economic Studies 18: 133 - 159.
Boeri, T. and H. Bücker (2000). the Impact of Eastern Enlargement on employment
and Labour Markets in the EU Member States. Berlin and Milan, DIW, CEPR, FIEF,
IGIER, HIS.
Bongaarts, J. and G. Feeney (1998). "On the Quantum and Tempo of Fertility."
Population and Development Review 24(2): 271 - 291.
Butz, W. P. and M. P. Ward (1979). "Will US fertility remain low? A new economic
interpretation." Population and Development Review 5(4): 663 - 689.
Calot, G. and J.-C. Chesnais (1983). L'Efficacité des politiques incitatrices en matire
de natalité. IUSSP colloquium, Lige, IUSSP.
Calot, G. and J.-P. Sardon (1999). "Les facteurs du vieillissement démographique."
Population 54(3): 509 - 552.
57
Castiglioni, M., G. D. Dalla Zuanna, et al. (2001). "Planned and Unplanned Births and
Conceptions in Italy 1970-1995." European Journal of Population 17(3): 207 - 233.
Champion, A. (1999). "Migration and British Cities in the 1990s." National Institute
Economic Review 4/99(170): 60 - 77.
Champion, A. and G. Hugo, Eds. (2003). New Forms of Urbanisation - beyond the
Urban-Rural Dichotomy. London, Ashgate.
Chesnais, J..-C. (1996). La crepuscule de l'occident. Paris: Robert Laffont.
Chesnais, J.-C. (2003). "Les morts violents dans le monde." Population et Sociétés
(395).
Caldwell, J. C. and T. Schindlmayr (2003). "Explanations of the fertility crisis in
modern societies: a search for commonalities." Population Studies 57(3): 241 - 264.
Coale, A. J. and P. Demeny (1983). Model Life Tables and Stable Populations.
Princeton, Princeton University Press.
Cliquet, R. L. (1991). The Second Demographic Transition: fact or fiction? Strasburg,
Council of Europe.
Cliquet, R. (2002). Origin and History of the FFS Project: achievements and
limitations. Dynamics of Partnership and Fertility in Europe: insights and lessons
from comparative research. M. Macura and G. Beets. New York and Geneva, United
Nations. 1: 17 - 26.
Coleman, D. A. (1999). Reproduction and Survival in an Unknown World: what
drives today's industrial populations, and to what future? Hofstee Lecture No. 5. The
Hague, Netherlands Interdisciplinary Demographic Institute (NIDI).
Coleman, D. A. (2000). Who's afraid of low support ratios? An unofficial response
from the UK to the UN Population Division Report on Replacement Migration. in UN
Population Division (eds) Expert Group Meeting on Policy Responses to Population
Ageing and Population Decline. pp 15-3: 15-31. New York, United Nations.
Preliminary version, no ISBN.
http://www.un.org/esa/population/publications/popdecline/Coleman.pdf
Coleman, D. A. (2001). 'Why Europe does not need a European immigration policy'.
Memorandum and Oral Evidence submitted to the House of Lords Select Committee
on the European Union. Session 2000-01, 13th Report. A Community Immigration
Policy Report with evidence. HL Paper 64. London, The Stationery Office: 19-36.
Coleman, D.A. (2002) 'Populations of the Industrial World - a convergent
demographic community?' International Journal of Population Geography 8, 319 -
344.
58
Coleman, D. A. (2002). "Replacement migration, or why everyone is going to have to
live in Korea: a fable for our times from the United Nations." Philosophical
Transactions of the Royal Society B 357: 583-598.
Coleman, D.A. and M.D. Smith (in preparation) Projections of the Ethnic Minority
Populations of the UK: First Steps.
Cornia, G. A. and R. Paniccia (1998). The Transition's Population Crisis: Nuptiality,
Fertility and Mortality Changes in Severely Distressed Economies. Population and
Poverty in Developing Countries. M. Livi-Bacci and G. De Santis. Oxford, Clarendon
Press: 217 - 249.
Council of Europe (2002). Recent Demographic Developments in Europe. Strasburg,
Council of Europe.
Courgeau, D. (2002). New Approaches and Methodological Innovations in the Study
of Partnership and Fertility Behaviour. Dynamics of fertility and partnership in
Europe: insights and lessons from comparative research. M. Macura and G. Beets.
New York and Geneva, United Nations. 1: 99 - 114.
Crafts, N. F. R. and T. C. Mills (1995). Europe's Golden Age: An Econometric
Investigation of Changing Trend Rates of Growth. Discussion Paper No 1087.
London, Centre for Economic Policy Research.
DaVanzo, JULIE., and GWENDOLYN. Farnsworth, eds. Russia's Demographic
"Crisis."Santa Monica, CA: Rand.
Davis, K. (1963) The theory of change and response in modern demographic history.
Population Index 21, 345 ­ 366.
Daykin, C. D. and D. Lewis (1999). "A Crisis of Longer Life: Reforming Pension
Systems." British Actuarial Journal 5, Part 1(21): 55 - 113.
de Beer, J. and L. van Wissen, Eds. (1999). Europe: One Continent, Different Worlds.
Population Scenarios for the 21st Century. Dordrecht, Kluwer.
Degraaf, N. D. and G. Evans (1996). "Why are the Young more Post-materialist? A
cross-national analysis of individual and contextual influences on post-material
values." Comparative Political Studies 28: 608 - 635.
Demeny, P. 1997. "Replacement-level Fertility: The Implausible Endpoint of the
Demographic Transition." In The Continuing Demographic Transition, ed. Jones, G.
et al. Oxford: Clarendon Press.
Demeny, P. (2003). "Population policy dilemmas in Europe at the dawn of the
twenty-first century." Population and Development Review 29(1): 1 - 28.
de Jong, A. (2002). Latest National Fertility Forecasts in Europe. Annex 3, Eurostat
Working Party on Demographic Projections. Luxemburg, European Commission.
59
Dennis, N. and G. Erdos (1992). Families without Fatherhood. London, Institute of
Economic Affairs Health and Welfare Unit.
Dustmann, C., F. Fabbri, et al. (2003). "The local labour market effects of
immigration in the UK." Home Office Online Report 06/03.
Dykstra, P. and L. van Wissen (1999). Introduction: The Life Course Approach as an
Interdisciplinary Framework for Population Studies. Population Issues: An
Interdisciplinary Focus. L. van Wissen and P. Dykstra. New York, Kluwer Academic:
1 - 22.
EAPS (2002). Newsletter No 32 December 2002. The Hague, European Association
for Population Study.
Ekert, O. (1986). "Effets et limites des aides financires aux familles: une experience
et un modle." Population 41(2): 327 - 348.
Ekert-Jaffé, O., H. Joshi, et al. (2002). "Fertility, Timing of Births and Socio-
Occupational Status in France and Britain: Social Policies and Occupational
Polarisation." Population 57(3): 475 - 508.
Ermisch, J. (1986). The Economics of the Family: applications to divorce and
remarriage. CEPR Discussion Paper Series No 140. London, Centre for Economic
Policy Research.
Ermisch, J. (1979). "The relevance of the "Easterlin Hypothesis" and the "New Home
Economics" to fertility movement in Great Britain." Population Studies 33(1): 39-58.
Ermisch, J. (1996). The Economic Environment for Family Formation. In Europe's
Population in the 1990s. D. A. Coleman , ed.. Oxford, Oxford University Press: 144 -
162.
Ermisch, J. F. (1991). Lone Parenthood: An Economic Analysis. Cambridge,
Cambridge University Press.
Ermisch, J. and M. Francesconi (2000). "Cohabitation in Great Britain: not for long,
but here to stay." Journal of the Royal Statistical Society A 163(Part 2): 153-171.
Ermisch, J.F. and M. Francesconi (2001a) 'Family Structure and Children's
Achievements'. Journal of Population Economics, 14(2), pp 249-270.
Ermisch, J. F., and M. Francesconi ( 2001b) 'Family Matters: impacts of family
background on educational attainments'. Economica, 68(270), pp 137-156.
European Commission (1995). The Demographic Situation in the European Union.
1994 Report. Luxemburg: Office for Official Publications of the European
Communities. (section 1.3)
60
European Commission (2000). On a Community Immigration Policy. Communication
from the Commission to the Council and the European Parliament. COM(2000) 757
Final. Brussels, Commission of the European Communities.
European Commission (2001). The Free Movement of Workers in the Context of
Enlargement. Information Note 6, March 2001. Brussels, European Commission.
European Commission (2003). Purchasing Power Parities and related economic
indicators for EU, Acceding and Candidate Countries and EFTA. Eurostat, Statistics
in Focus, Economy and Finance, 64/2003. Luxemburg, Eurostat Press Office.
Eurostat (1991). Eurobarometer 32. The Family and the Desire for Children.
Luxemburg, Office for the Official Publications of the European Communities.
Eurostat (2002). European Social Statistics. Labour Force Survey Results 2001, Data
2001. Luxemburg, Office for Official Publications of the European Communities.
Eurostat (2002). European Social Statistics: Demography. Luxemburg, Office for the
Official Publications of the European Communities.
Evans, G. with N D De Graaf (1996) 'Why are the Young more Postmaterialist? A
Cross-National Analysis of Individual and Contextual Influences on Postmaterial
Values', Comparative Political Studies, 28, 608-35, 1996.
Feeney, G. and J. Yu (1987). "Parity Progression Measures of fertility in China."
Population Studies 41: 77 - 102.
Feld, S. (2000). "Active Population Growth and Immigration Hypotheses in Western
Europe." European Journal of Population 16(3 - 40).
Fertig, M. and C. M. Schmidt (2001). Aggregate-level migration studies as a tool for
forecasting future migration streams. International Migration: Trends, policies and
economic impact. S. Djajic. London, Routledge.
Finch, N. and J. Bradshaw (2003). Fertility and Supporting the costs of children.
Recent fertility trends in Northern Europe, Voksenasen Hotel, Oslo, Social Policy
Research Unit, University of York.
Folbre, N. (2001). How the cost of children is distributed. International Perspectives
on Low Fertility: Trends, Theories and Policies. Tokyo, IUSSP.
http://demography.anu.edu.au/Publications/ConferencePapers/IUSSP2001/Program.ht
ml
Foster, C. (2000). "The Limits to Low Fertility: a Biosocial Approach." Population
and Development Review 26(2): 209 - 234.
Friedman, D., M. Hechter, et al. (1994). "A Theory of the Value of Children."
Demography 31(3): 375 - 401.
61
Frejka, T. and Calot, G. (2001) Cohort Reproductive Patterns in Low-fertility
Countries. Population and Development Review 27, (1) 103 - 132
Fukuyama, F. (1999). The Great Disruption, Profile Books.
Frejka, T. and G. Calot (2001). "Cohort Reproductive Patterns in low-fertility
countries." Population and Development Review 27(1): 103 - 132.
Frejka, T. and W. Kingcade (2003). US Fertility in International Comparison: An
Exploration to Aid Projections. The Direction of Fertility in the United States. US
Bureau of the Census. Alexandria, Virginia, Council of Professional Associations on
Federal Statistics: 55 - 150.
Fukuda, N. (2003). "Comparing Family-Friendly Policies in Japan and Europe:
Are We in the Same or in a Different League?' " Journal of Population and Social
Security. Population Study 1: 31-45.
Gauthier, A. H. (1996). The state and the family : a comparative analysis of family
policies in industrialized countries. Oxford: Oxford University Press.
Gauthier, A. H. and J. Hatzius (1997). "Family Benefits and Fertility: an econometric
analysis." Population Studies 51(3): 295 - 306.
Gauthier, A. H. (2002). "Family Policies in Industrialised Countries: Is there
Convergence?" Population E 57(3): 447 - 474.
Gesano, G. (1999). Maastricht Requirements for European Populations? Demotrends -
quadrimestrale sulla realt demografica italiana, 1999 (1), 1.
Gjonca, A., C. Wilson, et al. (1997). "Paradoxes of Health Transition in Euopre's
Poorest Country: Albania 1950 - 90." Population and Development Review 23(3):
585 - 610.
Glass, D. V. and C. P. Blacker (1938). Population and Fertility. London, Population
Investigation Committee.
Glass, D.V. (1967). Population Policies and Movements in Europe (1940). London:
Oxford University Press.
Goldstein, H. (1995). Multilevel Statistical Models. London, Edward Arnold.
Goldstein, J. R. and A. J. Morning (2002). Back in the Box: The Dilemma of Using
Multiple-Race Data for Single-Race Laws. The New Race Question: How the Census
Counts Multiracial Individuals. M. Waters and J. Perlmann. New York, Russell Sage:
119 - 136.
Goldstein, J., W. Lutz, et al. (2003). The Emergence of sub-Replacement Family Size
Ideals in Europe. European Demographic Research papers 2003 no 2. Vienna, Vienna
Institute of Demography, Austrian Academy of Sciences.
62
Golini, A. (1998). "How low can fertility be? An empirical exploration." Population
and Development Review 24(1): 59 - 74.
Golini, A., A. Mussino, et al. (2000). Il malessere demografico in Italia. Bologna,
Societa Editrice Il Mulino.
Government Actuary (1999). National Insurance Fund Long Term Financial
Estimates Cm 4406. London, The Stationery Office.
Grasland, C. (1990). "Systmes démographiques et systmes supranationaux: La
fécondité européenne de 1952  1982." European Journal of Population 6(2): 163 -
192.
Greve, B. (1994). Development in the EC and its Impact on the Welfare State in
Europe - trends towards convergence in the last ten years. History of European Ideas,
19(1- 3), 147 - 152.
Hamilton, B. E., P. D. Sutton, et al. (2003). "Revised Birth and Fertility Rates for the
1990s and New Rates for Hispanic Populations, 2000 and 2001." National Vital
Statistics Reports 51(12): 96.
Hantrais, L. (1997). Exploring Relationships between Social Policy and Changing
Family Forms within the European Union. European Journal of Population, 13, 4, 339
- 379.
Hatton, T. J. and M. Tani (2003). Immigration and Inter-regional Mobility in the UK,
1982 - 2000. Canberra, Australian National University: 21.
Heuveline P., J.M. Timberlake, F.F. Furstenberg Jr., 2003, Shifting Childrearing to
Single Mothers: Results from 17 Western Countries, Population and Development
Review, 29, 1: 47-71.
Hilderink, N., L. van der Gaag, et al. (2001). Analysis and Forecasting of
International Migration by Major Groups. To be published as a Eurostat working
paper. The Hague, NIDI.
Hobcraft J., 2002, Moving beyond elaborate description: Towards understanding
choices about parenthood, in Macura and Beets (2002): 131-143.
Hobcraft, J. (2003). Reflections on Demographic, Evolutionary, and Genetic
Approaches to the Study of Human Reproductive Behavior. Offspring: Human
Fertility Behavior in Biodemographic Perspective. K. W. Wachter and R. A. Bulatao.
Washington, National Academies Press.
Hobcraft, J. and K. Kiernan (1995). Becoming a Parent in Europe. Evolution or
Revolution in European Population. European Population Conference Milan 1995.
European Association for Population Studies. Milan, FrancoAngeli. 1 . Plenary
Sessions: 27 - 66.
63
Hoem, B. and J.M. Hoem (1997) Sweden's family policies and roller-coaster fertility.
Demography Unit, Stockholm University Stockholm Research Reports in
Demography 115.
Hollifield, J. F. (2000). The Politics of International Migration: How can we 'Bring
the State back in'? Talking Across Disciplines: Migration Theory in Social Science
and Law. C. Brettell and J. F. Hollifield. London, Routledge.
Holmans, A. E., Nandy, S. and A.C. Brown (1987). "Household formation and
dissolution and housing tenure: a longitudinal perspective." Social Trends 17: 20-28.
Inglehart, R. (1990). Culture Shift in Advanced Industrial Society. Princeton, New
Jersey, Princeton University Press.
Inglehart, R. and W. E. Baker (2000). "Modernization, Cultural Change, and the
Persistence of Traditional Values." American Sociological Review, Vol. 65, No. 1,
65( 1. Looking Forward, Looking Back: Continuity and Change at the Turn of the
Millennium): 19-51.
InterGovernmental Consultations on Asylum (1995). Illegal Aliens: A Preliminary
Study. Geneva, Intergovernmetnal Consultations on Asylum.
International Organization for Migration (1995). Trafficking and Prostitution: the
growing exploitation of migrant women from Central and Eastern Europe. Budapest,
Migration Information Programme, International Organization for Migration.
International Organization for Migration (2003). World Migration 2003: Managing
migration - challenges and responses for people on the move. Geneva, International
Organization for Migration.
Jackson, R. and N. Howe (2003). The 2003 Aging Vulnerability Index. Washington
DC, Center for Strategic and International studies and Watson Wyatt Worldwide.
Joshi, H. (1990). "The Cash opportunity costs of childbearing: an approach to
estimation using British data." Population Studies 44: 41-60.
Joshi, H. (2002). "Production, Reproduction and Education: women, children and
work in a British perspective." Population and Development Review 28(3): 445 -
474.
Katus, K. (2003). Post-transitional fertility developments: new perspectives
introduced by Central and East European nations. Population of Central and Eastern
Europe. Challenges and Opportunities. I. E. Kotowska and J. Jozwiak. Warsaw,
Statistical Publishing Establishment: 117 - 138.
Kiernan, K. (1992). "The Impact of Family Disruption in Childhood on Transitions
made in Young Adult Life." Population Studies 46: 213 - 234.
Kiernan, K. (2002). The State of European Unions: an analysis of partnership
formation and dissolution. Dynamics of Fertility and Partnership in Europe, Volume
64
1. Proceedings of the Family and Fertility Surveys Flagship Conference, Brussels
2000. M. Macura and G. C. N. Beets. New York and Geneva, United Nations. 1: 57-
76.
Kohler, H.-P. (2001). Fertility and Social Interaction - an economic perspective.
Oxford, Oxford University Press.
Kohler, H. P., F. C. Billari, et al. (2002). "The emergence of lowest-low fertility in
Europe during the 1990s." Population and Development Review 28(4): 641 - 680.
Kohler, H.-P. and J. A. Ortega (2002). Tempo-adjusted period parity progression
measures: assessing the Implications of Delayed Childbearing for Cohort Fertility
in Sweden, the Netherlands and Spain. Demographic Research 6, 7.
http://www.demographic-research.org/.
Kravdal, . (1992). "The weak impact of female labour force participation on
Norwegian third-birth rates." European Journal of Population 8(3): 247 - 263.
Kravdal, . (1992b). "The emergence of a positive relation between education and
third birth rates in Norway with supportive evidence from the United States."
Population Studies 46(3): 459 ­ 476.
Kravdal, . (1997). "Wanting a Child without a Firm Commitment to the Partner:
Interpretations and Implications of a Common Behaviour Pattern among Norwegian
Cohabitants." European Journal of Population 13(3): 269 - 298.
Kyvelidis, I. (2001). "Measuring Post-Materialism in Post-Socialist Societies."
European Integration online Papers 5 (2): 14.
Lesthaeghe, R. and G. Moors (1996). Living Arrangements, Socio-Economic Position
and Values among Young Adults: a Pattern Description of France, West Germany,
Belgium, and the Netherlands 1990. Europe's Population in the 1990s. D. A.
Coleman. Oxford, Oxford University Press: 163 - 221.
Lesthaeghe, R. (1995). The Second Demographic Transition in Western Countries:
An Interpretation. Gender and Family Change in Industrialized Countries. K. O.
Mason and A.-M. Jensen. Oxford, Clarendon Press: 17 - 62.
Lesthaeghe, R. (2001). Postponement and recuperation: Recent fertility trends and
forecasts in six Western European countries. IUSSP Working paper series.
International Perspectives on Low Fertility - trends, theories and policies. Paris,
IUSSP.http://demography.anu.edu.au/Publications/ConferencePapers/IUSSP2001/Pro
gram.html
Lesthaeghe, R. and P. Willems (1999). "Is low fertility a temporary phenomenon in
the European Union?" Population and Development Review 25(2): 211 - 228.
Lesthaeghe, R. and J. Surkyn (2002). New Forms of Household Formation in Central
and Eastern Europe; Are they related to newly emerging Value Orientations?
65
Interuniversity papers in demography IPD-WP 2002-2. Brussels and Ghent, Interface
Demography.
Lievens, J. (1998). "Interethnic marriage: Bringing in the context through multilevel
modelling." European Journal of Population 14(2): 117-115.
Lutz, W., S. Scherbov, et al., Eds. (1994). Demographic Trends and Patterns in the
Soviet Union before 1991. London, Routledge.
Lutz, W., Sanderson, W. and S. Scherbov. 2001. "The End of World Population
Growth." Nature 412: 543-545.
Lutz, W., B. C. O'Neill, et al. (2003). "Europe's population at a turning point."
Science 299(1991 - 1992).
McDonald, P. (1997). Gender Equity, Social Institutions and the Future of Fertility.
Working papers in Demography No 69. Canberra, Australian National University
Research School of Social Sciences: 25.
McDonald, P. (2000). "Gender Equity in theories of fertility transition." Population
and Development Review 26(3): 427 - 440.
McDonald, P. (2002). "Sustaining Fertility Through Public policy: The Range of
Options." Population 57(3): 417 - 446.
Macura, M. (1999). Fertility Decline in the Transition Economies 1982 - 1997:
Political, Economic and Social Factors. Economic Survey of Europe. UN ECE. New
York and Geneva, United Nations. 1: 181 - 194.
Macura, M. (2000). Fertility Decline in the Transition Economies 1982 - 1997.
Economic and Social Factors revisited. Economic survey of Europe 2000/1. UN ECE.
Geneva, United Nations. 1: 189 - 207.
Macura, M., Y. Mochizuki-Sternberg, et al. (2002). Eastern and Western Europe's
Fertility and Partnership Patterns: selected developments from 1987 to 1999.
Dynamics of fertiilty and partnership in Europe: insights and lessons from
comparative research. M. Macura and G. Beets. New York and Geneva, United
Nations. 1: 27 - 55.
Marshall, G. (1997) Repositioning Class: Social Inequality in Industrial Societies.
London: Sage.
Martin, S. P. (2000). "Diverging Fertility among U.S. Women who delay childbearing
past age 30." Demography 37(4): 523 - 533.
Maslow, A. (1954). Motivations and Personality. New York, Harper and Row.
Morgan, S. P. and R. B. King (2001). "Why Have Children in the 21st Century?
Biological Predisposition, Social Coercion, Rational Choice." European Journal of
Population 17(1): 3 - 20.
66
Murphy, M. (1992). "Economic Models of fertility in Post-War Britain - A
Conceptual and Statistical Re-interpretation." Population Studies 46: 235 - 258.
Murphy, M. and A. Berrington (1992). The Construction, validation and interpretation
of period parity progression ratios from large scale developed society household
structure data: an exploratory analysis. Fertility in Britain: new perspectives. M. ni
Bhrolchain. London, HMSO.
Murray, C. (1990). The Emerging British Underclass. London, Institute of Economic
Affairs Health and Welfare Unit.
Namboodiri, K. and L. Wei (1997). Fertility Theories and their Implications
Regarding How Low can Low Fertility be? 23rd International Population Congress,
Peking.
ni Bhrolchain, M. (1987). "Period parity progression ratios and birth intervals in
England and Wales 1941 - 1971: a synthetic life table analysis." Population Studies
41: 103 - 125.
Ni Bhrolchain, M. (1993b ). Recent Fertility differentials in Britain. New Perspectives
on Fertility in Britain. Studies on Medical and Population Subjects No 55. ed. M. Ni
Bhrolchain. London, HMSO: 93 - 108.
OECD (1978). The Migratory Chain. Paris, OECD.
OECD (1998). Trends in International Migration. Annual Report 1998 edition. Paris,
OECD.
OECD (2001). Trends in International Migration. SOPEMI 2001. Paris, OECD.
OECD (2003). Trends in International Migration. SOPEMI 2002. Paris, OECD.
Okolski, M. (2003). Demographic Processes Before and During the On-going
Transition. Warsaw: unpublished manuscript 17 pp plus appendices.
Olah, L. S. (1998). Do public policies influence fertility? Evidence from Sweden and
Hungary from a gender perspective. Stockholm Research Reports in Demography
130. Stockholm, Stockholm University Demography Unit.
Orrenius, P. M. and M. Zavodny (2003). "Do amnesty programs reduce
undocumented immigration? Evidence from IRCA." Demography 40(3): 437-450.
Osborne, C., S. McLanahan, et al. (2003). Is There an Advantage to Being Born to
Married versus Cohabiting Parents? Working Paper # 03-09-FF. Princeton, Center for
Research on Child Wellbeing: 41. http://crcw.princeton.edu/workingpapers/WP03-09-
FF-Osborne.pdf
Palomba, R. (2001). Postponement of family formation in Italy, within the Southern
European context. IUSSP Seminar on International Perspectives on Low Fertility:
67
Trends, Theories and Policies, Tokyo March 2001.
http://demography.anu.edu.au/Publications/ConferencePapers/IUSSP2001/Program.ht
ml
Philipov, D. and H.-P. Kohler (2001). "Tempo Effects in the Fertility Decline in
Eastern Europe." European Journal of Population 17(1): 37 - 60.
Philipov, D. (2003). Fertility in times of discontinuous societal change. Population of
Central and Eastern Europe. challenges and Opportunities. I. E. Kotowska and J.
Jozwiak. Warsaw, Statistical Publishing Department: 665 - 690.
Pinnelli, A., A. De Rose, et al. (2002). Interrelationships between Partnership and
Fertility Behaviour. Dynamics of fertility and partnership in Europe: insights and
lessons from comparative research. M. Macura and G. Beets. New York and Geneva,
United Nations. 1: 77 - 98.
Punch, A. and D. L. Pearce, Eds. (2000). Europe's population and labour market
beyond 2000. Strasburg, Council of Europe.
Reher, D. S. (1998). "Family Ties in Western Europe: Persistent Contrasts."
Population and Development Review 24(2): 203 - 234.
Rendall, M. S. and S. Smallwood (2003). "Higher Qualifications and timing and
subsequent pace of childbearing: a study of women born in England and Wales
between 1954 and 1958." Population Trends(111): 18-26.
Retherford, R. D. and N. Y. Luther (1996). "Are fertility differentials by education
converging in the United States?" Genus 52(3-4): 13-38.
Robinson, W. C. (1997). "The Economic Theory of Fertility over Three Decades."
Population Studies 51(1): 63 - 74.
Rnsen, M. (2001). Fertility and family policy in Norway: is there a connection.?
IUSSP Working paper series. International Perspectives on Low Fertility - trends,
theories and policies. Paris, IUSSP.
Salt, J. (2000). "Trafficking and Human smuggling: a European perspective."
International Migration 38(3 (special issue 1 / 2000)): 31 - 56.
Salt, J., A. Singleton, et al. (1993). Europe's International Migrants. Data Sources,
patterns and trends. London, HMSO.
Salt, J. and J. Stein (1997). "Migration as a Business: the case of Trafficking."
International Migration 35(4).
Salt, J., J. Clarke, et al. (1999). Assessment of possible Migration Pressure and its
Labour Market Impact following EU enlargement to Central and Eastern Europe.
London, Migration Research Unit, Department of Geography, University College
London.
68
Salt, J. (2002). Current Trends in International Migration in Europe. Report presented
to the Ministers responsible for migration matters. Strasburg, Council of Europe.
Salt, J. and J. Clarke (2003). Labour Migrant Flows in Europe: Patterns and Trends.
European Population Committee December 2003. Strasburg, Council of Europe.
Santow, G. and M. Bracher (2000). Deferment of first birth and fluctuating fertility in
Sweden. Stockholm University Research Reports in Demography 141. Stockholm,
Demography Unit, Stockholm University.
Sobotka, T. (2003). Understanding lower and later fertility in Central and Eastern
Europe. Population of Central and Eastern Europe. Challenges and Opportunities. I.
E. Kotowska and J. Jozwiak. Warsaw, Statistical Publishing Department. 691-724.
Sobotka, T., K. Zeman, et al. (2003). "Demographic Shifts in the Czech Republic
after 1989: A Second Demographic Transition View." European Journal of Population
19(3): 249 - 277.
Surkyn, J. (1998). Social Integration of ethnic minorities: Indicators at the family
level. IDP-Working paper 1998-2. Brussels, Interface Demography, Vrije Universeit
Brussel.
Stloukal, L. (1998). "An APC analysis of demographic responses to population policy
measures: the case of the Czech and Slovak Republics." Genus LIV(1-2): 87-121.
Storesletten, K. (2003). Fiscal Implications of Immigration - A Net Present Value
Calculation. Scandinavian Journal of Economics 105(3): 487-506.
Think Tank on Integration in Denmark (2002). Population Development 2001 - 2021;
possible developments. Copenhagen, Ministry of Refugee, Immigration and
Integration Affairs. http://www.imm.dk.
Tomka, B. (2003). "Western European welfare states in the 20th century:
convergences and divergences in long-run perspective." International Journal of
Social Welfare 12(4): 249 - 260.
Tribalat, M. (1995). Faire France - Une enqute sur les immigrés et leurs enfants.
Paris, La Découverte.
Udry, J. R. (1996). Biosocial models of low-fertility societies. New York, The
Population Council.
Ulrich, R. E., 2001: Die zukünftige Bevölkerungsstruktur Deutschlands nach
Staatsangehörigkeit, Geburtsort und ethnischer Herkunft: Modellrechnung bis 2050.
Paper prepared for the Independent Commission on Migration. Berlin/Windhoek: 38
UNECE, Ed. (2003). Economic Survey of Europe 2003 No. 1. New York and
Geneva, United Nations.
69
United Nations (2002). National Population Policies 2001. New York, United
Nations.
United Nations Population Division (1997). Expert Group Meeting on Below-
Replacement Fertility, New York 4-6 November 1997. New York, United Nations.
United Nations Population Division (2002). The Future of Fertility in Intermediate-
level countries. New York, United Nations.
United Nations Population Division (2003) Partnership and Reproductive Behaviour
in Low-Fertility Countries. New York, United Nations.
United Nations (2002). World Population Prospects: the 2000 Revision. Volume 1:
Comprehensive Tables. New York, United Nations.
United Nations (2003). World Population Prospects: the 2002 Revision. Volume 1:
Comprehensive Tables. New York, United Nations.
US Bureau of the Census (1996). Population Projections of the United States by Age,
Sex, Race and Hispanic Origin 1995 - 2050. Current Population Reports P25-1130.
Washington DC, US GPO.
US Bureau of the Census 2001a: Census 2000 Brief: Overview of Race and Hispanic
Origin. Washington DC. www.census.gov/population/www/cen2000/briefs.html
US Bureau of the Census, 2001b: Census 2000 Brief: The Hispanic Population.
Washington DC. www.census.gov/population/www/cen2000/briefs.html
US Bureau of the Census 2002: Census 2000 Brief: The Asian Population: 2000.
Washington DC. www.census.gov/population/www/cen2000/briefs.html
van de Kaa, D. J. (1987). "Europe's Second Demographic Transition." Population
Bulletin 42(1).
van de Kaa, D. J. (2001). "Postmodern fertility preferences: from changing value
orientation to new behavior." Population and Development Review (supplement) 27:
290 - 331.
van Imhoff, E. (2001). "On the impossibility of inferring cohort fertility measures
from period fertility measures." Demographic Research 5(2): 23-64.
Ventura, S. J., T. J. Mathews, et al. (2001). Births to Teenagers in the United States,
1940 - 2000. National Vital Statistics Reports; vol 49 no 10. Hyattsville, Maryland,
National Center for Health Statistics.
Voas, D (2003) Conflicting Preferences: a reason Fertility Tends to be Too High or
Too Low? Population and Development Review 29, 4.
Widgren, J. (1994). The Key to Europe: a comparative analysis of entry and asylum
policies in Western countries. Stockholm, Fritzes.
70
Varese, F. (2001). The Russian Mafia : private protection in a new market economy /
Federico Varese. Oxford, Oxford University Press.
Vetta, A. and D. Courgeau (2003). "Demographic Behaviour and Behaviour
Genetics." Population E 58(4-5): 401 - 428.
Wadensjö, E. (1999). Economic Effects of Immigration. Immigration to Denmark:
International and national perspectives. D. A. Coleman, Wadensjö, E., Jensen, B. and
S. Pedersen. Aarhus, Aarhus University Press: 290-328.
Watson, P. (1996). Marriage and Mortality in Eastern Europe. East-West Life
Expectancy gap in Europe. Environmental and Non-Environmental Determinants. C.
Hertzman, C. Kelly and M. Bobak. Dordrecht, Kluwer: 143 - 160.
Willekens, F. (1999). The Life Course: Models and Analysis. Population Issues: an
interdisciplinary focus. L. J. G. van Wissen and P. A. Dykstra. Dordrecht, Kluwer: 23
- 51.
Witte, J. C. and G. G. Wagner (1995). "Declining Fertility in East Germany after
unification: A demographic response to socio-economic change." Population and
Development Review 21(2): 387 - 397.
World Bank (2002). World Development Indicators 2002 World View Table 1.1 size
of the economy. Washington DC, World Bank.
http://www.worldbank.org/data/wdi2002/tables/table1-1.pdf
71