Population (english edition)
I.N.E.D
I.S.B.N.sans
200 pages
p. 451 à 486
doi: en cours
Veille sur la revue
Veille sur l'auteur
I.N.E.D
I.S.B.N.sans
200 pages
p. 451 à 486
doi: en cours
Veille sur la revue
Veille sur l'auteur
Vous consultez
Volume 58 2003/4-5
Re-Emerging Diversity: Rapid Fertility Changes in Central and Eastern Europe After the Collapse of the Communist Regimes
Tomás Sobotka [*] Tomás Sobotka, Population Research Centre, Faculty of Spatial Sciences, University of Groningen, P.O. Box 800, 9700 AV, Groningen, The Netherlands, Tel: + 31 50 363 7321, Fax: + 31 50 363 3901,
This article provides a detailed analysis of recent fertility changes in 15 countries of central and eastern Europe and in the former East Germany. It focuses on the period after 1989, which witnessed a profound transformation in childbearing patterns, including a rapid decline in fertility rates, the postponement of childbearing, and an upsurge in the proportion of extramarital births. These shifts went hand in hand with changes in union formation, abortion and contraceptive prevalence. While the intensive decline of the total fertility rates seems to indicate a uniform reaction of former Communist societies to the ongoing social and economic changes, the analysis reveals that there was increasing diversity in fertility patterns across the region. The article pays particular attention to the interplay between postponement of childbearing and period fertility levels. The progression of the postponement—indicated by an increase in the mean age of women at first birth—has varied widely between countries. We hypothesize that the more rapid postponement of parenthood was related to the success of the transition period, and to the extent it brought new opportunities and choices for young people and shifted the institutional structure of many societies considerably closer to the structure of western European countries.
Cet article présente une analyse détaillée de l’évolution récente de la fécondité dans 15 pays d’Europe centrale et orientale ainsi que dans l’ex-RDA. L’étude porte essentiellement sur la période postérieure à 1989, marquée par une transformation en profondeur des structures de la fécondité, notamment la baisse rapide des taux de fécondité, le report des naissances et l’augmentation de la proportion de naissances hors mariage. Ces changements sont allés de pair avec ceux qui ont concerné la formation des couples, le recours à l’avortement et l’emploi de la contraception. Si la forte chute des indices synthétiques de fécondité pourrait faire penser à une réaction uniforme des anciennes sociétés communistes face aux transformations économiques et sociales dont elles sont le théâtre, l’analyse révèle une diversification croissante des modèles de fécondité dans toute la région. L’article met particulièrement l’accent sur l’interaction entre le report des naissances et le niveau de la fécondité du moment. L’ampleur de ce report – qui se manifeste par l’élévation de l’âge moyen des femmes à la naissance de leur premier enfant – a considérablement varié d’un pays à l’autre. Nous avançons l’hypothèse que le report des naissances a été le plus rapide dans les pays qui ont le mieux surmonté les difficultés de la transition, où celle-ci a ouvert aux jeunes de nouvelles perspectives d’avenir et de nouveaux espaces de choix, et où elle a étroitement rapproché la structure institutionnelle de la société de celle des pays d’Europe occidentale.
Este artículo presenta un análisis detallado de la evolución reciente de la fecundidad en 15 países de Europa Central y del Este, y también en la ex-RDA. El análisis se basa esencialmente en el periodo posterior a 1989, marcado por una transformación profunda de las estructuras de la fecundidad, es decir, por una disminución rápida de las tasas de fecundidad, un retraso de los nacimientos y un aumento de la proporción de nacimientos fuera del matrimonio. Paralelamente a estos cambios se han producido otros relativos a la formación de parejas, al recurso al aborto y al uso de anticonceptivos. La fuerte disminución de los índices sintéticos de fecundidad recuerda las reacciones uniformes típicas de los antiguos regímenes comunistas ante transformaciones económicas y sociales de los que fueron protagonistas, pero el análisis revela una diversificación creciente de los modelos de fecundidad en toda la región. El artículo hace hincapié en la interacción entre la postergación de los nacimientos y el nivel de fecundidad del momento. El nivel de tal postergación – que se manifiesta a través del aumento de la edad de las mujeres al primer nacimiento - ha variado considerablemente de un país a otro. La hipótesis que avanzamos sugiere que la postergación de los nacimientos ha sido más rápida en aquellos países que han superado mejor las dificultades de la transición, en aquellos países donde ésta ha abierto nuevas perspectivas de futuro y nuevas opciones para los jóvenes, y donde las estructuras institucionales de la sociedad se asemejan más a las de los países de Europa Occidental.
Demographic developments in central and eastern Europe have long attracted specialist attention for the varied effects of the population policies pursued in these countries during the communist era. The political upheavals occurring since the collapse of the Soviet Union have re-established economic and social conditions as the key influences in the process of family building. Using a set of indicators for fertility, contraception and union formation, Tomás Sobotka analyses the different paths taken by these countries in the 1990s. Over and above the apparent uniformity of the decline in fertility since 1989, the author distinguishes between, on the one hand, the countries of central Europe, where the decline seems to result primarily from a shift towards later child-bearing, and on the other, the group formed by Romania, Bulgaria, the Russian Federation and the republics of the former USSR, where completed fertility is tending to fall sharply, mainly due to the growing proportion of one-child families. The trends observed in the first group can be interpreted as an alignment with western patterns of behaviour and lifestyles, and those in the second group as the result of the economic and social crises affecting these countries since the demise of the communist regimes.
The avalanche of events that led to the demise of authoritarian regimes swept through the countries of central and south-eastern Europe in the “revolutionary year” 1989 and reached its climax with the dissolution of the Soviet Union in 1991. The political division of Europe into east and west, clearly distinguishing the members of the two competing political blocs and two different social and economic systems for more than four decades, had come to an end, paving the way for a less clear-cut political, social and economic differentiation of Europe.
The radical social and economic transformations generated a strong impetus for the subsequent change in the demographic behaviour of the populations in this region. Fertility patterns, initially characterized by early and almost universal childbearing and by a strong attachment to the two-child family norm, changed rapidly over the 1990s. A substantial decline in period fertility has taken place in all previously communist countries of Europe, which now form the region of lowest fertility in the world, with total fertility rates (TFRs) ranging between 1.1 and 1.4 children per woman in 2000. An interesting discussion has evolved around several issues: Why did total fertility rates decline to such low levels ? What was the impact of the postponement of childbearing on period fertility ? Is eastern Europe experiencing the same transformation as the western European societies since the mid-1960s ? Are fertility patterns in eastern Europe becoming more heterogeneous ?
This article aims to provide detailed evidence on fertility changes in the countries of central and eastern Europe during the 1990s and to discuss the evolving regional differentiation in fertility patterns. In doing so, it touches on most of the questions posed above. First, a general overview looks at fertility changes in a longer time perspective. Developments in fertility are then linked with evidence on changes in family formation and with data on contraceptive use and induced abortion. The final discussion focuses on increasing differences in the timing of childbearing and the effects of fertility postponement on parity-specific fertility rates.
I. Data and methods
Data were collected from various sources. Basic data on fertility, abortions and various demographic indicators originate from the Council of Europe (2001, 2002) and from Eurostat (2001-2003). Detailed data on births by biological birth order and age of the mother, and data on the age distribution of women, were obtained from Eurostat and from official vital statistics data and publications. Indicators related to cohabitation, contraceptive use and mother’s status at first birth were compiled from published tables of the FFS (Fertility and Family Survey) Standard Country Reports and RHS (Reproductive Health Surveys) conducted in many countries of the region throughout the 1990s.
In total, 16 units — 15 countries and the former German Democratic Republic — have been selected for the comparative analysis. To allow a concise comparison of a large amount of data, some figures are presented as arithmetic means
[1] for four distinctive geographical regions: central Europe (Croatia, the Czech Republic, East Germany, Hungary, Poland, the Slovak Republic, and Slovenia), south-eastern Europe (Bulgaria and Romania), the Baltic states (Estonia, Latvia, and Lithuania), and the post-Soviet countries, regrouping the remaining European states of the former Soviet Union (Belarus, Moldova, the Russian Federation and Ukraine). Only a limited amount of data is available for Croatia, East Germany and the former Soviet countries. Apart from Croatia and Slovenia, other successor states of the former Yugoslavia as well as Albania were excluded for lack of data, because of specific demographic developments influenced by the civil war (Bosnia-Herzegovina, Serbia and Montenegro) or because of their distinctive history or cultural tradition
[2]. In order to provide a wider perspective on fertility changes in central and eastern Europe, a comparison with the following European regions is frequently provided: western Europe (Austria, Belgium, France, West Germany, the Netherlands, Switzerland, and the United Kingdom); northern Europe (Denmark, Finland, Norway, and Sweden), and southern Europe (Greece, Italy, Portugal, and Spain). Presentation of recent fertility transformations is complemented by an estimate of the effects of fertility postponement on the level of period TFR, using a method proposed by Bongaarts and Feeney (1998).
II. Complex changes in fertility and living arrangements
A demographic distinctiveness of eastern Europe
[3] as a relatively homogeneous region had gradually developed between the mid-1960s and the mid-1980s. In 1985, there was a clear demographic division between eastern and western Europe (Monnier and Rychtaríková, 1992, p. 157). A cleavage along political boundaries had replaced the geo-cultural line identified by Hajnal in 1965 (Ní Bhrolcháin, 1993, p. 463). Political blocs constituted specific levels of spatial organization, manifested also by contrasting fertility development (Decroly, 1993). In the mid-1980s eastern European patterns of fertility behaviour were characterized by a strong attachment to the two-child family norm with only a small proportion of women remaining childless, by early family formation, by an early start of childbearing (as compared with the western European standard), and by a subsequent short spacing of births. Reproductive life was marked by a low prevalence of modern contraception and a high incidence of induced abortion. Premarital conceptions were common
[4], while extramarital births were relatively rare.
The 1990s brought a complex transformation of this reproductive model. An overview of fertility change between 1989 and 2000 in 16 analysed units is provided in the Appendix. The following section deals in more detail with recent fertility changes. Particular attention is paid to the postponement of childbearing, the increasing differences in period fertility by birth order, the spread of extramarital childbearing and the diffusion of cohabitation, and the changes in abortion rates and contraceptive prevalence.
1. Long-term changes in period fertility
Since the 1960s total fertility rates in central and eastern Europe contrasted with those in other parts of Europe. In Hungary, Estonia and Latvia, the TFR had fallen below 2.0 already at the beginning of the 1960s, a time when many parts of Europe experienced a moderate baby boom. In the 1970s and 1980s, various population policy measures
[5] together with other factors (e.g. a limited supply of effective contraception and the easy availability of jobs due to non-existent unemployment) preserved eastern Europe from the fertility decline experienced in all other European regions. A sharp fall in the TFR during the 1990s shifted the position of eastern Europe on the European fertility map within one decade from the “highest fertility” to the “lowest fertility” region (Figure 1). The level of fertility in eastern European regions in 1980 and 1998 was close to that of southern Europe, where the post-war fertility decline started later and was more pronounced than in northern and western Europe. Nevertheless, apart from central Europe, the TFR decline in the former Communist countries was concentrated entirely in the short period of the 1990s. In several countries of central Europe, especially in Slovenia and in the former GDR, the decline in period fertility had started already in the 1980s. It was particularly strong in East Germany, following shortly after the fall of the Berlin Wall in November 1989 and German reunification in October 1990 (see Sardon, 1998). Consequently, the TFR in East Germany reached the record low level of 0.77 in 1993-1994.
Figure 1
Total fertility rate in European and eastern European regions, 1980-2000
Source: Council of Europe (2001, 2002).
A steep decline in period fertility over the 1990s suggests a seemingly uniform reaction to the uneasy and complex transformation of the former Communist societies. However, a more detailed look at the changes in the timing and composition of fertility depicts a growing diversity in fertility patterns across eastern Europe.
2. Changes in the timing of fertility characterized by the postponement of childbearing
An early age at childbearing was one of the most distinctive features of the eastern European reproductive pattern. Around 1990, eastern Europe was still hardly affected by the rise in the mean age at childbearing which was so prevalent in other parts of Europe. Women were bearing their first children on average between age 22 (Bulgaria, Moldova) and 24 (Croatia, GDR) as contrasted with a range of 25 to 28 in the rest of Europe. Over the 1990s, postponement of first births spread gradually in eastern Europe
[6]. Although the large contrast with other European regions in the timing of first births remained virtually unchanged (Figure 2), there were substantial differences in the pace of postponement across central and eastern Europe.
Figure 2
Mean age of mothers at birth of first child in European and eastern European regions, 1980-2000
Note: post-Soviet countries: data for Belarus and Russia only; Western + Northern Europe: data for Denmark (until 1996), Finland, France, Norway (since 1986), and Sweden; Central Europe: excluding East Germany.
Sources: Eurostat (2002, 2003), Council of Europe (2001, 2002), Toulemon and Mazuy (2001).
The “aging of fertility” had been manifested earlier in some central European countries (Hungary, Slovenia, the former GDR) where, together with the Czech Republic, it gained considerable momentum during the 1990s, with the mean age at birth of the first child increasing by 2 to 3 years between 1989 and 2000. Former Soviet countries and the Baltic states initially experienced a slight decline in the mean age of first-time mothers from the end of the 1980s, but this trend reversed around 1994. The available data indicate that in the second half of the 1990s first births were being delayed to some extent in all former Communist countries. In central Europe, the Baltic states and south-eastern Europe, the postponement was already progressing at a faster pace than in western and northern Europe.
The shift in the timing of childbearing proceeded hand in hand with the decline in first-order total fertility rates (TFR1)
[7]. The TFR1 and the mean age of women at the birth of their first child (MAB1) are likely to follow a similar transition characterized by four distinctive stages (Figure 3):
- Initial stage: High TFR1 and relatively low MAB1.
- Onset of the postponement manifested by a substantial decline in the TFR1 and a parallel increase in the MAB1.
- Continuing postponement: TFR1 stabilizes at a low level, while the mean age of first-time mothers continues to grow.
- Postponement stops: As the mean age of first-time mothers approaches 30, the postponement gradually slows down and as a result the TFR1 increases again. The delay of childbearing proceeded in most countries hand in hand with a gradual decline in fertility quantum (level) of first births, therefore the TFR1 usually does not fully return to the initial level.
Figure 3
First-order TFR (TFR1) and mean age of mothers at birth of first child. selected countries, 1970-2001
Sources: Eurostat (2002, 2003), Council of Europe (2001, 2002), CNPS (1990-1997), CSU (2000b), FSU (1981-1989), GUS (1991-1995).
In Figure 3, the ongoing transition in several former Communist countries is compared with that of the Netherlands, which has already reached the final stage, and of Spain, where the postponement of first births is still in progress (Stage III). In most countries of eastern Europe, the timing shift is progressing relatively fast and the depressing effect on first-parity fertility rates is more pronounced. The very high levels of the TFR1 in Russia around 1990 were associated not only with a high level of first-order fertility, but also with the advancement of childbearing — a temporary decline in the mean age at first birth, which inflated the TFR1 values to a level close to 1.0. The figure illustrates the persistence of postponement, discussed by Kohler et al. (2002, p. 660): once initiated, the postponement transition in the low-fertility countries is continuous and irreversible, leading to a shift towards relatively late childbearing.
A more detailed view of changing fertility patterns by age is provided in Figure 4, which displays age-specific fertility rates in central Europe, Romania, Lithuania, Moldova, Russia and the European Union in 1989 and 2000. The quite homogeneous fertility profiles of eastern European countries in 1989 contrast with the much older schedules of the EU countries. In eastern Europe, childbearing was highly concentrated into a narrow age span of young women, and almost half of total fertility was realized between age 19 and 24. Moldova was an extreme case of very high fertility at those ages. Since 1989, mothers in the EU countries have become even older and the fertility profiles of eastern European regions have been more differentiated along the age axis. In all of them, the fertility decline caused a “flattening” of the fertility schedules, thus ending the clear concentration of childbearing into a few age groups. Childbearing patterns have become more heterogeneous, mirroring an increasing diversity of fertility strategies among women. As a result, the interquartile interval of the distribution of first births increased substantially, from 4.0–5.0 years around 1990 to 5.0–6.5 years in 2000. The age profile of childbearing in central Europe moved closer to the EU profile. Mothers in two post-Soviet countries, Russia and Moldova, remained very young, while Lithuania and Romania occupied an intermediate position. Most central European countries also experienced a process of gradual “making up” for the delayed births among women above age 30, and a particularly sharp reduction of fertility among teenage women. In other regions this decline was only gradual, resulting in increasing differences between countries. In Slovenia, the cumulative fertility rate among women below age 20 declined to the level of 36 per 1,000 in 2000, whereas in Bulgaria it was more than six times higher (234 per 1,000).
Figure 4
Age-specific fertility rates (ASFR) of women in selected countries and regions in 1989 and 2000
Note: Central Europe: arithmetic average of age-specific fertility rates (ASFR) in the Czech Republic, East Germany, Hungary, Poland, the Slovak Republic and Slovenia; EU profile: arithmetic average of ASFR in Finland, the Netherlands and Spain.
Sources: Eurostat (2002, 2003), Avdeev and Blum (2002), CNPS (1990), CSU (2000b), DASS (1999), GUS (1991), SB (1994, 2001) and unpublished national data.
3. Increasing differences in parity distribution
For several decades until the 1980s, the first-order TFRs were high, as almost all women had at least one child. Only 5 to 10% of women in different eastern European countries born in 1955 remained childless. Since the 1980s central Europe has gradually emerged as a region with low levels of TFR1, dropping to an average of 0.58 in 2000. Other regions, especially the post-Soviet countries, were characterized by slightly higher fertility at birth order 1 throughout the 1990s.
The position of different countries with regard to period fertility at the second birth order is in contrast with the situation at parity one. Generally, many countries that retain higher TFR1 have seen a pronounced decline in the second-order TFR (TFR2), and countries with a low TFR1 rank higher with respect to TFR2. This difference appears to be partly related to the uneven pace of the postponement of childbearing across eastern Europe. In countries where a marked decline of the TFR was largely driven by timing effects, there was a very strong reduction in first-order fertility rates. This pattern is typical of central European countries, where many women are postponing childbearing. However, once they give birth to their first child, most of them decide to have a second one. In contrast, in countries where the postponement was slower, especially in southeastern Europe and the post-Soviet countries, the fertility decline was driven mostly by quantum effects, and was very intensive with regard to second and higher birth orders.
These emerging contrasts may be illustrated by the analysis of period fertility for the first and second birth orders in the Czech Republic and Romania. The Czech Republic witnessed a pronounced decline in TFR between 1993 and 1996, while in Romania the most intensive decline took place already in the 1990-1992 period. Since 1995, the TFR in Romania has been higher than in the Czech Republic (upper left quadrant of Figure 5). The latter country experienced a particularly strong reduction in first-order TFR, which was lower by about 0.15 than in Romania in the second half of the 1990s (upper right quadrant). In contrast, the second-order TFR has been higher in the Czech Republic ever since 1989. Does this mean that Czech women will reach very high levels of childlessness, while most Romanian women will have only one child ? The use of the Bongaarts and Feeney (1998) adjusted order-specific TFRs (lower-left quadrant) and the parity-specific indexesof total fertility
[8] (PATFRs in the lower right quadrant of Figure 5) suggest a different interpretation. Both indicators show that, contrary to the order-specific TFR values, the first-order fertility rate was higher in the Czech Republic until 1996 and the two have been almost equal since then, reaching values around 0.8. The second-order fertility has been, in conformity with the order-specific TFR values, higher in the Czech Republic over the whole period since 1989. The PATFRs display a gradual reduction in this difference, while the adjusted TFRs show a continuation of the contrast. In the case of parity 1, birth probabilities are considerably less affected by the postponement of childbearing and the PATFR1 index therefore displays smaller fluctuations and higher values than the TFR1 (see Sobotka, 2003). According to the adjusted TFRs, the continuation of fertility rates and the pace of the postponement of the late 1990s would imply that about 80% of Czech and Romanian women would ultimately have one child. However, some 65 to 70% of Czech women and only 45% of Romanian women would have a second one.
Figure 5
Total fertility rate (TFR), unadjusted and adjusted order-specific TFRs, and parity-specific indexes of total fertility (PATFR) for the Czech Republic and Romania in 1989-2000
Note: adjustment by Bongaarts-Feeney method.
Sources: data from CSU (2000a, 2000b), CNPS (1990-1997), and Eurostat (2003).
How can the contrasting trends and values of different period fertility indicators be interpreted ? Clearly, timing changes in fertility affect different indicators in different ways and they particularly distort the TFR. The comparison of the Czech Republic and Romania has revealed that their major difference in fertility trends is not first-order fertility, but the proportion of women who choose to have a second child. The Romanian pattern of parity-specific fertility change also characterizes Bulgaria and the former Soviet countries, where most women want to have at least one child despite the difficult living conditions. However, fewer women among those who had a first child are willing to have another one. This indicates a move towards the acceptance of the one-child family model and its spread. Avdeev and Monnier (1995, p. 34) make two remarks on the attitudes among Russian women that illustrate this point. A widespread preference characterized as “at least one child, at most two” is combined with a strong effort to have really at least one: “Come what may, Russian women have one child at least, unless they are sterile”. The higher progression rate to the second child in central Europe may be attributed to a more stable social and economic environment, but selection effects play a role there as well. We may assume that women who are having children now are especially those behaving more “traditionally”, while women currently postponing births will not embrace the two-child family norm so often in the future.
The hypothesis on increasing parity differences in fertility structure across eastern Europe gains further support from the statistics on the distribution of live births by birth order that are available for more countries than the detailed data needed for the computation of order-specific TFRs. In Bulgaria, Belarus, Russia and Ukraine, the proportion of first births in the total number of births has increased to 55-60%, implying a strong orientation towards the one-child family model (Figure 6). Besides these countries, there has been a notable increase in the proportion of first births in Romania, Moldova and Poland, countries that used to have a large share of families with three or more children. In all central European countries, but not in East Germany
[9], first births accounted for less than 50% of all births during the 1990s.
Figure 6
First births as a proportion of total births, 1988-2000 (in %)
Notes: (1) most recent value for 1998 only; (2) most recent value for 2001 only.
Source: Council of Europe (2002).
The changes in the third and higher-order total fertility (TFR3+) have also been differentiated across the region. Bulgaria, Russia and probably also Belarus and Ukraine have reached the lowest levels of fertility at the third and higher birth order. In these countries, third and subsequent births contribute only about 10 to 12% to the overall TFR, indicating that families with three and more children have become uncommon there. The decline in higher-order fertility has been very marked also in Romania and Moldova. On the other hand, in several countries of central Europe (Hungary, Poland, and the Slovak Republic), the TFR3+ still accounted for about 25% of the overall TFR in 2000. Higher fertility rates at birth orders 3+ may be associated with the strong adherence to Catholicism in Poland and in some regions of Slovakia, and with higher fertility among the distinctive Roma minority in Hungary and Slovakia.
4. Upsurge in extramarital childbearing
A rapid increase in the proportion of children born out of marriage was — besides the changes in the level and timing of period fertility — the most characteristic feature of recent fertility transformations in eastern Europe. The spread of childbearing outside marriage is associated with two related developments: a decline in marriage rates and the increasing popularity of informal unions.
The proportion of children born out of wedlock did not exceed 10% in most countries of eastern Europe until the 1980s, though in several countries, particularly Estonia, Slovenia, and the former GDR, an earlier diffusion of non-marital childbearing and cohabitation took place. During the 1990s all eastern European countries saw an upsurge in the proportion of children born out of marriage, reaching an average level of 28% in 2000, which is a typical value in western European countries (Figure 7). However, the differentiation in the intensity of extramarital childbearing has increased enormously.
Figure 7
Proportion of children born out of marriage in European and eastern European regions, 1980-2000 (in %)
Source: Council of Europe (2001, 2002).
The diffusion of extramarital childbearing was particularly strong in the group of Baltic states and south-eastern European countries. Although its prevalence is influenced by a number of factors, the traditional cultural-religious division plays an important role in this differentiation. The Catholic Church is especially known for its emphasis on traditional family values and for its frequent interventions in the domain of human reproduction for the purpose of influencing sexual morality, sex education, family planning, and abortion laws. Three very secularized countries with a strong Protestant tradition — Estonia, the former GDR and Latvia — experienced the most dynamic increase in extramarital childbearing during the 1990s. With the proportion of extramarital births ranging between 40% (Latvia) and 54% (Estonia) in 2000, the situation in these countries resembles that of (also traditionally Protestant) northern Europe. As more than half of first children were born out of marriage, a disconnection of marriage and reproduction was taking place during the 1990s(
[10]). Among the historically Catholic countries, there are large differences between the more secularized societies (the Czech Republic, Hungary, Slovenia) where the proportion of extramarital births reached values comparable to those in western Europe, and the more traditional and religious societies, in particular Poland and Croatia, that experienced only a gradual increase in the share of extramarital births. Their position resembles that of the Catholic countries of southern Europe.
The increase in extramarital childbearing was accompanied by a declining propensity to marry, and particularly to marry at an early age. The total first marriage rate of women (TFMR
[11]) in eastern Europe, until 1990 by far the highest among the European regions (Figure 8), has recently reached the low levels typical of western and northern Europe. The average value for eastern European countries plummeted to the level of 0.51 in 2000. As in the case of first births, widespread postponement of first marriages had a depressing effect on the first marriage rates, which would otherwise be considerably higher.
Figure 8
Total first marriage rates (TFMR) of women in European and eastern European regions, 1980-2000
Note: data for Belarus, Moldova and Ukraine are not available; data for Russia are available until 1996 only.
Source: Council of Europe (2001, 2002).
To a large extent, the intensity of the decline in marriage rates mirrored the speed of the increase in extramarital childbearing. In most central European countries and Estonia and Romania, marriage rates had been declining slightly already during the 1980s, while in the other regions they remained very high, typically around 0.9 until 1990. We may speculate that the cultural-religious division also played an important role in the fall of marriage rates. In Estonia, Latvia and East Germany the TFMR reached for a time the extremely low level of 0.4. In the Czech Republic, Hungary and Slovenia, the TFMR had fallen below 0.5 by the end of the 1990s. The Christian Orthodox countries, and the more traditional Catholic ones, experienced a less pronounced decline in first marriage rates. Generally, postponement of first marriages has been even more pervasive than the postponement of first births. Extramarital births have become common especially among young women below the age of 25, while older women are more likely to marry before having children. In Slovenia, Estonia and Bulgaria, as in the Scandinavian countries, the mean age of women at first birth has become lower than the age at first marriage.
5. Increase in cohabitation or expansion of single motherhood ?
Was the spread of extramarital childbearing associated with the diffusion of alternative forms of the family, and particularly with an increase in cohabitation, or did an expansion of single motherhood take place in central and eastern Europe ? As data on informal unions are not regularly collected, we have to rely on the evidence provided by two large survey projects, FFS (Fertility and Family Surveys) and RHS (Reproductive Health Surveys) that were conducted at least once in almost all the analysed countries during the 1990s(
[12]). The major difficulty of using these two sources for a cross-country comparison is that they were conducted over a period of 9 years (1990-1999). Dynamic changes in eastern Europe during the 1990s imply that the picture obtained by combining these data is distorted, as it provides a comparison of countries at different stages of demographic and social change. However, keeping this fact in mind, this comparison is still valuable since these are the only extensive data sets on cohabitation as well as on contraceptive prevalence and life transitions in eastern Europe.
Figure 9 compares the proportion of extramarital births with the proportion of women living in consensual union at ages 25-29 during the 1990s(
[13]). In most former communist societies, relatively few women were cohabiting at those ages (typically fewer than 5%), especially in the early 1990s, with highly traditional Poland (0.5% in 1991) being an extreme case. Nevertheless, there were substantial regional differences, with some countries showing a higher prevalence of both cohabitation and non-marital childbearing. Especially in Estonia and Slovenia, cohabitation had been largely accepted already during the Communist era. In addition, two units — Bulgaria and the former GDR — displayed a relatively low prevalence of cohabitation (less than 10%) combined with a high proportion (over 30%) of extramarital births. Childbearing among single women seems to be widespread there
[14]. Despite a large heterogeneity, the direction of change towards increased extramarital childbearing coupled with the spread of cohabitation is clearly illustrated for three countries with comparable data on developments over time: the Czech Republic (1993 and 1997), Romania (1993 and 1999) and Russia (1996 and 1999). Additional evidence, provided by the 1999 European Values Study, suggests that the increase in the popularity of cohabitation since the early 1990s has been spectacular. Despite small national sample sizes, the survey shows that in the countries of central Europe (excepting Poland and Slovakia) and in the Baltic states (excepting Lithuania), cohabitation has become a common living arrangement among young people aged 20-29 (Lesthaeghe and Surkyn, 2002).
Figure 9
Extramarital births (in %) and proportion of women cohabiting at ages 25-29 in eastern European countries (various years)
Note: see Appendix for acronyms of countries. Data for Estonia refer to native-born women only.
Sources: FFS (1991-1998) and RHS (1993-1999) surveys (% of women cohabiting), Council of Europe (2000, 2001) for extramarital births.
The FFS surveys also investigated the partnership status of women at first birth. A distinction was made between marriage, consensual union and no partnership. The results depict a large variety not only in the proportion of children born within marriage, but also in the division of extramarital births between women in consensual unions and single mothers, i.e. without a partner (see Figure 10 for women aged 20-24). The low status of cohabitation in two traditionally Catholic societies, Poland and Lithuania, seems to be a strong barrier to living and having children in unions other than marriage, and explains a small prevalence of extramarital births, which were almost entirely taking place among single mothers. Among countries with a large proportion of first children born outside marriage, only in Estonia is cohabitation a typical setting for extramarital childbearing(
[15]). In Slovenia, cohabitation was only prevailing slightly over single motherhood, while East Germany, where a large proportion of first births takes place outside of marriage and particularly among single women
[16], constitutes a special case (see Konietzka and Kreyenfeld, 2002).
Figure 10
Partnership status at birth of first child among women aged 20-24 with at least one child
Note: data for Estonia refer to native-born women only.
Sources: FFS (1991-1998) surveys.
6. Fewer abortions, better contraception: a stealthy contraceptive revolution
Till the late 1980s the reproductive behaviour of women in eastern Europe was characterized by a limited choice of modern contraception, a lack of sex education, and a high prevalence of induced abortion. Abortion was provided on request for free in most countries from the second half of the 1950s
[17] (see Frejka, 1983; Blayo, 1991; David, 1999; Stloukal, 1999). As most societies were tolerant of pre-marital sex, young people experienced sex relatively early and usually before marriage
[18]. First sex occurred mostly without the use of any contraceptive method and pregnancy often followed soon after the onset of reproductive life. Since women pregnant for the first time usually did not opt for an abortion, marriage — commonly entered by a pregnant bride — and birth soon followed. Abortion was widespread among women with two or more children. Generally low control of people over their reproductive life resulted in a high proportion of “mistimed” and “unwanted” births. The important question is whether the former Communist countries experienced the “contraceptive revolution” and the move toward the ideal of the “perfect contraceptive society” during the 1990s. In other words, has the specific “abortion culture” of eastern Europe disappeared ?
Except for the special case of Poland, where a strict anti-abortion law was passed in 1993
[19], available statistics indicate a significant decline in the total induced abortion rate (TIAR
[20], see Table 1). The simultaneous strong decline in fertility rates and abortion rates during the 1990s contrasts with the previous substitution effect between induced abortion and fertility, which was characteristic of eastern Europe when contraceptive prevalence was low. Decline in fertility rates was usually achieved through an increase in the prevalence of abortions and an increase in fertility was frequently associated with a decline in abortion rates. A portion of the recent decline in abortion rates may be caused by the incomplete registration of abortion, particularly in countries where abortions are also performed in private facilities and in countries where the official health care system has deteriorated. Nevertheless, the official data depict the overall trends well. There was a lasting division between countries with fairly low abortion rates and countries where the incidence of induced abortion still remained high at the end of the 1990s. All central European countries and Lithuania belong to the first group, with TIARs below 1.0. The most recent figures for other countries with available data put the TIAR between 1.0 and 2.1. Values of TIAR around 0.5 registered in 2000 in the Czech Republic, Slovenia and the Slovak Republic are close to the abortion rates of many western European countries including France and Sweden. A spectacular decline of abortion rates occurred in Romania between 1990 (TIAR of 6.1) and 1998 (TIAR of 1.6). Extremely high abortion rates in 1990 followed the legalization of abortion in December 1989 under conditions of still very limited access to contraceptive information and means.
Table 1
Estimated total induced abortion rates, 1985-2000
According to various RHS and FFS surveys, the proportion of women aged 25-29 living in union who used any contraceptive method varied between 52% in Poland and 76% in Hungary
[21]. Bulgaria is a notable exception, with a very low level of contraceptive use, reported by only 42% of women, and yet not extremely high abortion rates (Table 2).
Table 2
Current contraceptive use among women in union aged 25-29 in Central and Eastern Europe during the 1990s (in %)
The heterogeneity in the patterns of contraceptive use in eastern Europe is manifested in two ways. Some countries display a persistent popularity of traditional contraceptive methods — coitus interruptus and periodic abstinence — which were still used by 30 to 50% of women in Ukraine (1999), Romania (1993 and 1999) and Poland (1991). In these countries, as well as in Lithuania (1994-95) and Bulgaria (1997-98), the use of modern contraception (condoms, the pill and the IUD) was lower than 40%. Only in central Europe (with the notable exception of Poland) were traditional methods currently used by a small minority of women. Another differentiation was in respect to the usually used contraceptive means. The pill has become the main method of contraception among younger women in central Europe (again with the exception of Poland in 1991), while in the Baltic states and the post-Soviet countries, the proportion of women using the intrauterine device (IUD) outnumbered by far the small proportion of women using the pill. Time trends, provided by data for the Czech Republic and Romania, reflect an increasing popularity of the pill, which is replacing the traditional methods of contraception, and also the IUD in the Czech Republic. These trends, common across the whole region are, however, taking place at a very different pace.
III. Timing shifts and fertility decline
1. The influence of timing effects on fertility decline
As total fertility rates declined to very low levels in almost all former communist countries by the end of the 1990s, there was an increasing diversity in the timing of childbearing, both within countries — manifested by an extension of prime childbearing years into a larger age span — and between countries, with marked differences in the mean age at childbearing related to the different pace of postponement. One reason why the timing shifts are important for understanding fertility change is the effect of the timing change itself: postponement of child-bearing depresses the total fertility rate to a lower level than it would reach in the absence of timing changes. Thus the fall in the TFR during the 1990s was driven by the “real” reduction in fertility level (quantum) as well as by timing effects.
Since the intensity of the timing changes varied considerably across the region, the “depressing” effects of these changes on the TFR differed accordingly. Bongaarts and Feeney’s (1998) adjusted TFRs presented in Table 3 help to shed some light on these differences
[22]. Postponement appears to be the major cause of decline in period fertility rates for several countries of central Europe, in particular for the Czech Republic, Hungary and Slovenia. Five central European countries with available data had adjusted TFRs around 1.7 in 1998-2000. Similar levels were characteristic of the three Baltic states (1.61–1.67), while the evidence for south-eastern Europe and Russia (data for 1994-95) put the adjusted total fertility at a relatively low level of 1.4–1.5. Low values are likely to be typical of other post-Soviet countries, where the delay of childbearing has been only modest so far. Due to the different pace of the postponement, countries with roughly equal period TFRs in 1998-2000, such as Bulgaria and the Czech Republic, display different levels of the adjusted TFR (1.47 and 1.67 respectively).
Table 3
Total fertility rate (TFR), unadjusted and adjusted by the Bongaarts-Feeney method, all birth orders and at birth order 1 (TFR1), 1990-2000
Although the Bongaarts-Feeney procedure may be based on overly simplistic assumptions, it provides a less dramatic perspective on the period fertility level than the (non-adjusted) TFR values of 1.1 to 1.4. It supports the hypothesis that the major differentiation in fertility level between countries lies in the progression rate to second-order births. The non-adjusted TFRs show large differences in first-order fertility, reaching as low as 0.53 in the Czech Republic. The adjusted TFR at order 1, however, mostly comes close to 0.8, still indicating a potentially strong increase in childlessness, but also suggesting that most differences in first-order TFRs were caused by timing effects. Furthermore, countries with low values of the adjusted TFR — in our analysis Russia, Bulgaria, and Romania — differ from other countries by their low fertility rates at parity 2. Detailed analysis in the previous section, comparing the Czech Republic and Romania, revealed that this differentiation also holds when fertility indicators based on age and parity-specific birth probabilities are used. The TFR decline in central Europe was to a large extent driven by timing shifts; in contrast in Romania, Bulgaria, and particularly in the former Soviet countries, timing effects were considerably smaller. While fertility rates at parity 1 remain quite high, and while first births make up to 60% of all births there, the progression rates to second births have declined rapidly. Having children relatively early and bearing just one child may be a common strategy to cope with the difficult living conditions.
2. Explaining diversity in the postponement of childbearing
How can we explain the increasing diversity in the timing of child-bearing across countries ? One explanation may treat these differences as a re-emergence of traditional differences described in the case of marriage timing by Hajnal (1965). Indeed, all countries with an early pattern of childbearing and high teenage fertility in 2000(
[23]) lie east of, or on, the imaginary line between Trieste and St Petersburg. Furthermore, most countries located to the west of this line have recently experienced a marked increase in the mean age of childbearing.
But why would such a distinctive division between the early and the late marriage and childbearing pattern reappear under radically changed social conditions that by no means resemble the family systems prevailing in the early twentieth century ? A more realistic explanation would relate recent changes to the complex transformation of former Communist societies. An intensive debate among demographers has evolved around the question of whether the economic crisis, unemployment and other constraints were the main causes of fertility decline (e.g. UN, 2000b), or whether new opportunities and changing values were mostly responsible for changing fertility patterns (e.g. Zakharov, 2000). Most researchers recognize that both factors played an important role. Lesthaeghe and Surkyn (2002, p. 215) for instance, pointed out that it is the “entire restructuring of society” that accelerated ideational and demographic changes.
There exists a large body of evidence suggesting that despite the widespread emergence of negative economic phenomena such as unemployment and poverty, eastern European countries have become enormously differentiated in terms of the successes and failures of their reforms and the living standards of their populations (Ellman, 2000). While the post-Soviet countries, especially Moldova and Ukraine, got trapped in massive economic crises, bringing the living conditions of their populations close to those of many developing countries, central European societies were fairly successful in implementing a market economy and institutional reforms. Thus it is plausible that the impact of the new opportunities, including prolonged education and the pursuit of leisure activities and consumerist lifestyles, has been strongest in central Europe, while the effects of the economic downturn affected the populations in former Soviet countries most. How are these differences translated into fertility trends ? The more successful social and economic transformation brings the whole institutional structure of the former communist societies closer to the structure of western European countries. This generates a profound change in values, living arrangements and fertility patterns, a sort of “westernisation” of demographic patterns.
Lesthaeghe and Moors (2000, p. 124) see the postponement of marriage and parenthood as the main feature of the Second Demographic Transition. Can we observe any relationship between the pace of the postponement of childbearing and societal transformation in central and eastern Europe ? The increase in the mean age of mothers at the birth of their first child between 1989 and 1999 is to a surprisingly large extent associated with more successful development. Figure 11 illustrates this relationship using an indicator of real wage change over this period. Countries experiencing the most severe decline in real wages (over 50% in Moldova, Russia, Lithuania and Ukraine) experienced only a minor postponement of childbearing while, with the exception of Poland, postponement progressed relatively rapidly in all the countries with a decline in real wages of less than 20% over this period. Poor economic prospects, uncertainty, and low level of social security may contribute to the persistence of an early childbearing pattern rather than stimulate its postponement. According to the “theory of the value of children”, proposed by Friedman et al. (1994, p. 383), marriage and childbearing are widely available strategies to reduce uncertainty regarding the future, especially among women whose “alternative pathways for reducing uncertainty are limited or blocked”. This hypothesis gained support in Kohler and Kohler’s (2002, p. 255) analysis of fertility decline in Russia, which found that women or couples most affected by labour market crises often had a higher probability of having another child.
Figure 11
Relationship between change in mean age of women at birth of first child and percentage change in real wages. Central and Eastern European countries, 1989 to 1999
Note: see Appendix for acronyms of countries.
Sources: UNICEF (2001) for data on wages, Council of Europe (2002) for fertility data. Change in the mean age for Ukraine was estimated on the basis of the most recent data (1993) and trends in related indicators (mean age at childbearing and overall fertility rates by age) in the following period.
IV. A new fertility landscape in central and eastern Europe
It would be simplistic to relate the nature of fertility changes in central and eastern Europe only to the relative success of their social and economic transformation. Obviously, there were many complex factors influencing childbearing decisions among women. Some countries such as Poland and Romania have seen a strengthening of religiosity, with Churches frequently promoting traditional family values, while other societies such as the Czech Republic and eastern Germany are among the most secularized in the world. In some countries, women have been increasingly facing discrimination at the workplace and gender inequality has widened. Central Europe and the Baltic states experienced a shift in the structure of employment toward a service economy, while south-eastern Europe and the former Soviet countries faced an increase in the proportion of people working in agriculture. These and many other factors have affected recent developments in fertility. Nevertheless, more opportunities, an increased freedom of choice and a generally higher control of people over their lives — including control over their reproduction thanks to the greater availability and affordability of modern contraceptives — are connected with the “westernisation” of fertility patterns.
What appears at first sight to be a uniform decline in fertility rates across central and eastern Europe was in fact a complex transformation of fertility patterns following the collapse of the previous political system. Since 1989, within a period of ten years, the former communist countries in Europe have become widely differentiated with regard to their social and economic conditions as well as to the reproductive behaviour of their populations. Changes in fertility are part of a broader transition in family life marked by the spread of alternative family forms, non-marital births, postponement of childbearing and the decline of fertility and marriage rates, which had been taking place in western European societies since the 1960s. In contrast to western Europe, changes in central and eastern Europe have frequently progressed at breathtaking speed, as they were partly driven by the emerging economic constraints. The interconnectedness of the changes in values, fertility behaviour and family formation, however, clearly points out the influence of a broader social transformation in all former Communist societies.
The unprecedentedly low fertility rates in eastern Europe are to a large extent associated with the ongoing delay of childbearing among women. At the same time, decline in fertility quantum has been substantial as well and will ultimately result in a gradual decline in cohort fertility. As first-time mothers are on average still fairly young compared with those in other parts of Europe, postponement of first births will proceed further and fertility rates are likely to remain very low for another 10-15 years. A gradual economic recovery may paradoxically stimulate a further decline in fertility rates in countries where the fertility decline has been driven by factors associated with the economic crisis. Potential speeding-up of the postponement in the former Soviet countries, and to a smaller extent also in south-eastern Europe, may bring about a further reduction of the already very low fertility rates.
Appendix
Basic indicators of period fertility in Central and Eastern Europe, 1989-2000
Acknowledgements
This study was realized in the framework of the project “Towards a dynamic scenario model of economic determinants of European population development”, funded by the Netherlands Organization for Scientific Research, project number A 510-03-901. This article stems from a working paper entitled “Ten years of rapid fertility changes in the European post-communist countries: evidence and interpretation”, published by the Population Research Centre, University of Groningen (WP 02-1). An earlier version was presented at the European Population Conference held in Helsinki on 7-9 June 2001. I am thankful to many colleagues who supplied me with data for various countries of central and eastern Europe, in particular to Vlada Stankuniene, Boris Burcin, Hans-Peter Kohler, Sergei Scherbov, Elena Koytcheva, Taeke Gjaltema, Sergei Zakharov, Aura-Mihaela Zamfirescu and Krystof Zeman. I appreciate useful comments and suggestions from Frans Willekens, Leo van Wissen, Michaela Kreyenfeld, and Sabu S. Padmadas. Two anonymous reviewers provided many valuable suggestions on an earlier version of this article.
BIBLIOGRAPHIE
·
Avdeev A., A. Blum, I. Troitskaya, 1995, “The history of abortion statistics in Russia and the USSR from 1900 to 1991”, Population: An English Selection, 7, pp. 39-66.
·
Avdeev A., A. Blum, 2002, Démographie de la Russie et de son empire sur la toile, <http:// dmo. econ. msu. ru/ demogrus/ >.
·
Avdeev A., A. Monnier, 1995, “A survey of modern Russian fertility”, Population: An English selection, 7, pp. 1-38.
·
Blayo C., 1991, “Les modes de prévention des naissances en Europe de l’Est”, Population, 46(3), pp. 527 -546.
·
Bongaarts J., G. Feeney, 1998, “On the quantum and tempo of fertility”, Population and Development Review, 24(2), pp. 271-291.
·
CDEC 2000, 2001, Naselenie Rossii 1999 (2000) [Population of Russia 1999 (2000)], Institut Narodnokhoziaistvennogo Prognozirovaniia RAN, Centr Demografii i Ekologii Cheloveka, Moscow, <http:// demoscope. ru/ center/ sitemapnn. html>.
·
CNPS, 1990-1997, Nascuti-vii în anul 1989 (1990…1996), [Live births in 1989 (1990…1996)], Bucharest, Comisia Nationala pentru Statistica (National Commission for Statistics).
·
CNPS, 1998, Unpublished data on abortions and age structure provided by the National Commission for Statistics, Bucharest.
·
Council of Europe, 1996, 2001, 2002, Recent Demographic Developments in Europe 1996 (2001-2002), Strasbourg, Council of Europe Publishing.
·
CR POPIN, 2001, Czech Republic Population Information, <http:// popin. natur. cuni. cz>.
·
CSU, 2000a, Demografie, [CD-ROM containing Czech Demographic Yearbooks for 1992-1998], Prague, Cesky statisticky úrad (Czech Statistical Office).
·
CSU, 2000b, Unpublished data on births by birth order and age of mother, 1989-1999, provided by the Czech Statistical Office, Prague.
·
DASS, 1999, Unpublished data on births by birth order and age of mother and on age structure of women 1990, 1996 provided by the Department of Statistical and Sociological Analysis of the Republic of Moldova, Chisinau.
·
David H.P. (ed.), 1999, From abortion to contraception. A resource to public policies and reproductive behavior in Central and Eastern Europe from 1917 to the present, Westport, Connecticut, Greenwood Press.
·
Decroly J.-M., 1993, “Boundaries, political systems and fertility in Europe”, Population: An English selection, 5, pp. 101-119.
·
Ellman M., 2000, “The social costs and consequences of the transformation process”, in Economic Survey of Europe 2000, No. 2-3, New York and Geneva, Economic Commission for Europe, United Nations, pp. 125-140.
·
Eurostat, 2001-2003, NEW CRONOS Database, accessed in March 2001, May 2002 and May 2003.
·
Federa, 2000, The Anti-abortion Law in Poland: The Functioning, Social Effects, Attitudes and Behavior. Report – September 2000, Federation for Women and Family Planning <http:// www. federa. org. pl/ english/ reports/ report00>.
·
FFS, 1991-1998, Fertility and Family Survey. Tables from the Standard Country Reports (printed publications) and Standard Country Tables (accessed through the Internet) for the Czech Republic, Estonia, Germany, Hungary, Latvia, Lithuania, Poland and Slovenia. Geneva, Population Activities Unit, UN/ECE, <http:// www. unece. org/ ead/ pau/ ffs/ ffs_standtabframe. htm>.
·
Frejka T., 1983, “Induced abortion and fertility: A quarter century of experience in Eastern Europe”, Population and Development Review, 9(3), pp. 494-520.
·
Friedman D., M. Hechter, S. Kanazawa, 1994, “A theory of the value of children”, Demography, 31(3), pp. 375-401.
·
FSU, 1981-1989, Pohyb obyvatelstva v Ceské socialistické republice 1980-1988, [Population movement in the Czech Socialist Republic], Prague, Federální statisticky úrad (Federal Statistical Office).
·
Grünheid E., J. Roloff, 2000, “Die demographische Lage in Deutschland 1999 mit dem Teil B ‘Die demographische Entwicklung in den Bundesländern – ein Vergleich’”, Zeitschrift für Bevölkerungswissenschaft, 25(1), pp. 3-150.
·
GUS, 1991-1999, Roczniki statystyczne: Demografia 1991-1992. Rocznik statystyczny demografii 1993 -1996. Rocznik demograficzny 1997-1999 (Demographic yearbook), Warsaw, Glówny urzad statystyczny (Central Statistical Office).
·
Hajnal J., 1965, “European marriage patterns in perspective”, in D.V. Glass, D. Eversley (eds.), Population in History, London, pp. 101-143.
·
Infostat, 2000, Population Development in the Slovak Republic 1999, Bratislava, Institute of Informatics and Statistics, Demographic Research Centre.
·
Katus K., L. Sakkeus, A. Puur, A. Põldma, 2000, “Demographic development of Estonian Population: Recent changes in the context of long-term trends”, in T. Kucera et al. (eds.), New Demographic Faces of Europe, Berlin, Springer – Verlag, pp. 125-140.
·
Klinger A., 1991, “Les politiques familiales en Europe de l’Est”, Population, 46(3), pp. 511-526.
·
Kohler H.-P., F.C. Billari, J.A. Ortega, 2002, “The emergence of lowest-low fertility in Europe”, Population and Development Review, 28(4), pp. 641-680.
·
Kohler H.-P., I. Kohler, 2002, “Fertility decline in Russia in the early and mid-1990s: The role of economic uncertainty and labour market crises”, European Journal of Population, 18(3), pp. 233-262.
·
Konietzka D., M. Kreyenfeld, 2002, “Women’s employment and non-marital childbearing: A comparison between East and West Germany in the 1990s”, Population, English edition, 57(2), pp. 331-358.
·
Kreyenfeld M., 2001, Employment and fertility - East Germany in the 1990s, Doctoral Thesis, University of Rostock.
·
Kulczycki A., 1995, “Abortion policy in postcommunist Europe: The conflict in Poland”, Population and Development Review, 21(3), pp. 471-505.
·
Lesthaeghe R., G. Moors, 2000, “Recent trends in fertility and household formation in the industrialized world”, Review of Population and Social Policy, No. 9, pp. 121-170.
·
Lesthaeghe R., J. Surkyn, 2002, “New forms of household formation in Central and Eastern Europe: Are they related to newly emerging value orientations ?”, in Economic Survey of Europe 2002/1, New York and Geneva, Economic Commission for Europe, United Nations, pp. 197-216.
·
Monnier A., 1990, “The effects of family policies in the German Democratic Republic: A re-evaluation”, Population: An English Selection, 2, pp. 127-140.
·
Monnier A., J. Rychtaríková, 1992, “The division of Europe into East and West”, Population: An English Selection, 4, pp. 129-160.
·
NíBhrolcháin M., 1993, “East-West marriage contrasts, old and new”, in A. Blum, J.-L. Rallu (eds.) European population II. Demographic Dynamics, Montrouge, John Libbey Eurotext, pp. 461-479.
·
Philipov D., H.-P. Kohler, 2001, “Tempo effects in the fertility decline in Eastern Europe: Evidence from Bulgaria, the Czech Republic, Hungary, Poland and Russia”, European Journal of Population, 17(1), pp. 37-60.
·
Rallu J.-L., L. Toulemon, 1994, “Period fertility measures. The construction of different indices and their application to France, 1946-89”, Population: An English Selection, 6, pp. 59-94.
·
RHS, 1993-1999, Reproductive Health Survey. Selected data published in the following RHS Reports: 1993 Czech Republic RHS (Final Report, 1995), 1993 Romania RHS (Final Report 1995), 1996 Romania Young adult RHS (Final Report 1998), 1999 Romania RHS (Preliminary Report 2000), 1996 Russia’s women’s RHS (Final Report 1998), 1999 Russia’s women RHS (Preliminary Report 2000), 1997 Moldova RHS (Final Report 1998) and 1999 Ukraine RHS (Preliminary Report 2000). Surveys were organized by the Centers for Disease Control and Prevention, Atlanta, in cooperation with other institutions.
·
Sardon J.-P., 1998, “Fécondité, bouleversements politiques et transition vers l’économie de marché en Europe de l’Est”, Espace, Populations, Sociétés, 1998-3, pp. 339-360.
·
Sardon J.-P., 2001, “Demographic change in the Balkans since the end of the 1980s”, Population: An English Selection, 13(2), pp. 49-70.
·
SB, 1991, 1994, Statistisches Jahrbuch für die Bundesrepublik Deutschland 1991, 1994, Wiesbaden, Statistisches Bundesamt.
·
Smallwood S., 2002, “The effect of changes in timing of childbearing on measuring fertility in England and Wales”, Population Trends, No. 109, pp. 36-45.
·
Sobotka T., 2003, “Tempo-quantum and period-cohort interplay in fertility changes in Europe. Evidence from the Czech Republic, Italy, the Netherlands and Sweden”, Demographic Research, 8, article 6, <www. demographic-research. org>.
·
SORS, 2000, Population of Slovenia, CD-ROM, Publ. No. 740, Ljubljana, Statistical Office of the Republic of Slovenia.
·
Stloukal L., 1999, “Understanding the ‘abortion culture’ in Central and Eastern Europe”, in H.P. David (ed.) From abortion to contraception. A resource to public policies and reproductive behavior in Central and Eastern Europe from 1917 to the present, Westport, Connecticut, Greenwood Press, pp. 23-37.
·
Toulemon L., M. Mazuy, 2001, “Les naissances sont retardées mais la fécondité est stable”, Population 56(4), pp. 611-644.
·
UN, 1997, 1998, 2000a, 2001, Demographic Yearbook 1996, 1997, 1998, 1999, New York and Geneva, United Nations.
·
UN, 2000b, “Fertility decline in the transition economies, 1989-1998: Economic and social factors revisited”, in Economic Survey of Europe 2000, No. 1, New York and Geneva, Economic Commission for Europe, United Nations, pp. 189-205.
·
Unicef, 2001, A decade of transition, The MONEE project, Regional monitoring report No. 8-2001, Florence, Unicef, Innocenti Research Centre.
·
Van Imhoff E., N. Keilman, 2000, “On the quantum and tempo of fertility: Comment”, Population and Development Review, 26(3), pp. 549-553.
·
Zakharov S.V., 2000, “Fertility trends in Russia and the European newly independent states: Crisis or turning point ?”, Population Bulletin of the United Nations, Special Issue Nos. 40/41, 1999, pp. 292-317.
·
Zeng Yi, K. C. Land, 2001, “A sensitivity analysis of the Bongaarts-Feeney method for adjusting bias in observed period total fertility rates”, Demography, 38(1), pp. 17-28.
NOTES
[*]
Population Research Centre, University of Groningen, The Netherlands.
Population Research Centre, University of Groningen, The Netherlands.
[1]Arithmetic means were chosen rather than means weighted by the population of given countries because of the large differences in population size between various central and eastern European countries. Weighted means would mostly mirror the situation in a few large countries of this region, e.g. Russia among the former Soviet Republics (146 million out of 209 million in habitants in 2000).
Arithmetic means were chosen rather than means weighted by the population of given countries because of the large differences in population size between various central and eastern European countries. Weighted means would mostly mirror the situation in a few large countries of this region, e.g. Russia among the former Soviet Republics (146 million out of 209 million in habitants in 2000).
[2]For an overview of demographic changes in the Balkan countries during the 1990s, see Sardon (2001).
For an overview of demographic changes in the Balkan countries during the 1990s, see Sardon (2001).
[3]For simplicity, the term eastern Europe is often used for the whole group of European former Communist countries included in the analysis. This follows a concept of eastern Europe that evolved from the peculiar development of the region after the Second World War.
For simplicity, the term eastern Europe is often used for the whole group of European former Communist countries included in the analysis. This follows a concept of eastern Europe that evolved from the peculiar development of the region after the Second World War.
[4]In the Czech Republic, about half of the first marital births were conceived before marriage during the 1980s (FSU, 1981-89). In Poland, 49% of first children were born within 9 months of the marriage in 1990 (GUS, 1991).
In the Czech Republic, about half of the first marital births were conceived before marriage during the 1980s (FSU, 1981-89). In Poland, 49% of first children were born within 9 months of the marriage in 1990 (GUS, 1991).
[5]A large variety of pronatalist and social policy measures included child benefits (often increasing progressively with the number of children in the family), paid maternity leave and the provision of daycare (see Klinger, 1991). Housing, which was always in short supply, was primarily distributed to married couples with children. Moreover, restrictions on the access to abortion were imposed in several countries at the end of the 1960s and in the 1970s (Blayo, 1991). In Romania and the Soviet Union, employed unmarried men and childless women had to pay a special “childlessness” tax.
A large variety of pronatalist and social policy measures included child benefits (often increasing progressively with the number of children in the family), paid maternity leave and the provision of daycare (see Klinger, 1991). Housing, which was always in short supply, was primarily distributed to married couples with children. Moreover, restrictions on the access to abortion were imposed in several countries at the end of the 1960s and in the 1970s (Blayo, 1991). In Romania and the Soviet Union, employed unmarried men and childless women had to pay a special “childlessness” tax.
[6]This article conceptualizes the postponement of first births in a period perspective, as an increase in the mean age of women at the birth of their first child that continues at least for three consecutive years. The mean age is derived from the distribution of first-order age-specific fertility rates for ages 15 to 49.
This article conceptualizes the postponement of first births in a period perspective, as an increase in the mean age of women at the birth of their first child that continues at least for three consecutive years. The mean age is derived from the distribution of first-order age-specific fertility rates for ages 15 to 49.
[7]Total fertility rates for birth order i are computed as the sum of age-specific fertility rates for that birth order (i.e. the ratio of births of order i to the total number of women aged x) by single years of age from 15 to 49 in a given year.
Total fertility rates for birth order i are computed as the sum of age-specific fertility rates for that birth order (i.e. the ratio of births of order i to the total number of women aged x) by single years of age from 15 to 49 in a given year.
[8]Birth probabilities by age and birth order are calculated as the ratios of the number of births at birth order i in year t among women reaching age x during that year to the population of women at parity i-1 at the beginning of the year t who will reach age x during that year. The parity and age-specific index of total fertility (PATFR) is a multiplicative indicator computed on the basis of these birth probabilities (see Rallu and Toulemon, 1994).
Birth probabilities by age and birth order are calculated as the ratios of the number of births at birth order i in year t among women reaching age x during that year to the population of women at parity i-1 at the beginning of the year t who will reach age x during that year. The parity and age-specific index of total fertility (PATFR) is a multiplicative indicator computed on the basis of these birth probabilities (see Rallu and Toulemon, 1994).
[9]Statistics on the distribution of births by biological birth order are not available in former East Germany since 1990. However, survey data indicate that East German women have retained relatively high first-order fertility rates, but their progression rate to the second birth has fallen sharply (Kreyenfeld, 2001).
Statistics on the distribution of births by biological birth order are not available in former East Germany since 1990. However, survey data indicate that East German women have retained relatively high first-order fertility rates, but their progression rate to the second birth has fallen sharply (Kreyenfeld, 2001).
[10]In 1998, the number of births per 1,000 women aged 15-49 in East Germany was for the first time higher among unmarried women (27.8 per 1,000) than among married women (27.6 per 1,000) (Grünheid and Roloff, 2000, p. 35, Table 12).
In 1998, the number of births per 1,000 women aged 15-49 in East Germany was for the first time higher among unmarried women (27.8 per 1,000) than among married women (27.6 per 1,000) (Grünheid and Roloff, 2000, p. 35, Table 12).
[11]TFMR is computed as the sum of the ratios of first marriages by individual years between 15 and 49 to the total number of women at those ages. It is an indicator of the proportion of women who would eventually marry before age 50, should the age-specific first marriage rates of a given year apply up to that age.
TFMR is computed as the sum of the ratios of first marriages by individual years between 15 and 49 to the total number of women at those ages. It is an indicator of the proportion of women who would eventually marry before age 50, should the age-specific first marriage rates of a given year apply up to that age.
[12]Among the 16 countries considered in the article, only Croatia, Belarus and Slovakia had no FFS or RHS conducted during the 1990s.
Among the 16 countries considered in the article, only Croatia, Belarus and Slovakia had no FFS or RHS conducted during the 1990s.
[13]The author was inspired by a similar figure for European countries, compiled by Lesthaeghe and Moors (2000, p. 157, Figure 23).
The author was inspired by a similar figure for European countries, compiled by Lesthaeghe and Moors (2000, p. 157, Figure 23).
[14]Most countries applied policies that discriminated against consensual unions, e.g. by reserving advantageous loans or provision of housing for newly-wed couples. The East German policy, which granted a special status to single mothers since 1976, was an exception, and this created an incentive for extramarital childbearing (Monnier, 1990; Konietzka and Kreyenfeld, 2002).
Most countries applied policies that discriminated against consensual unions, e.g. by reserving advantageous loans or provision of housing for newly-wed couples. The East German policy, which granted a special status to single mothers since 1976, was an exception, and this created an incentive for extramarital childbearing (Monnier, 1990; Konietzka and Kreyenfeld, 2002).
[15]According to Katus et al. (2000, p. 133), cohabitation has spread steadily in Estonia since the 1960s. Among native-born women born in the early 1970s, it accounts for more than 90% of first partnerships (see also FFS, 1994). While about four-fifths of first births are conceived outside registered unions, births to single mothers formed only 7 to 10% of all births over a long period of time. In this sense, Estonia exhibits similar trends to Scandinavian countries.
According to Katus et al. (2000, p. 133), cohabitation has spread steadily in Estonia since the 1960s. Among native-born women born in the early 1970s, it accounts for more than 90% of first partnerships (see also FFS, 1994). While about four-fifths of first births are conceived outside registered unions, births to single mothers formed only 7 to 10% of all births over a long period of time. In this sense, Estonia exhibits similar trends to Scandinavian countries.
[16]Among women aged 20-24 having at least one child in 1992, only 39.6% had a first child within marriage, while 37.6% were single and 22.8% were cohabiting at the time of birth. Among 25-29 year-old women, one-third had a first child while they were living without a partner.
Among women aged 20-24 having at least one child in 1992, only 39.6% had a first child within marriage, while 37.6% were single and 22.8% were cohabiting at the time of birth. Among 25-29 year-old women, one-third had a first child while they were living without a partner.
[17]In some countries, access to abortion was later limited, or abortion policy changed over time as a part of pronatalist measures. Romania enforced a ban on abortion for most women after 1967 and allowed no access to contraception, but it represented an exception among eastern European countries.
In some countries, access to abortion was later limited, or abortion policy changed over time as a part of pronatalist measures. Romania enforced a ban on abortion for most women after 1967 and allowed no access to contraception, but it represented an exception among eastern European countries.
[18]The FFS surveys reveal that the median age at first sex among women born in the second half of the 1960s varied between 17.6 years in the Czech Republic (birth cohorts 1968-72) and 20.1 years in Lithuania (birth cohorts 1965-75). In several countries (Moldova, Romania), first sex often took place within marriage, but in most countries it was typically a pre-marital experience. For instance in the Czech Republic only 0.4% of sexually experienced women reported they had first sex after marriage (RHS, 1993).
The FFS surveys reveal that the median age at first sex among women born in the second half of the 1960s varied between 17.6 years in the Czech Republic (birth cohorts 1968-72) and 20.1 years in Lithuania (birth cohorts 1965-75). In several countries (Moldova, Romania), first sex often took place within marriage, but in most countries it was typically a pre-marital experience. For instance in the Czech Republic only 0.4% of sexually experienced women reported they had first sex after marriage (RHS, 1993).
[19]For a valuable analysis of the abortion debate, see Kulczycki (1995); for more recent developments, see FEDERA (2000).
For a valuable analysis of the abortion debate, see Kulczycki (1995); for more recent developments, see FEDERA (2000).
[20]The TIAR is a period indicator of the average number of induced abortions per woman that would prevail in a hypothetical population of women if they experienced the age-specific abortion rates of a given year throughout their reproductive life (ages 15-49).
The TIAR is a period indicator of the average number of induced abortions per woman that would prevail in a hypothetical population of women if they experienced the age-specific abortion rates of a given year throughout their reproductive life (ages 15-49).
[21]These figures are not adjusted for non-response, which was quite high in some countries (e.g. in Poland).
These figures are not adjusted for non-response, which was quite high in some countries (e.g. in Poland).
[22]The adjustment method proposed by Bongaarts and Feeney (1998) estimates the total fertility rate free of timing distortions caused by the postponement or advancement of childbear-ing. It is calculated for each birth order separately as 
, where ri,t is the change in the mean age at childbearing at birth order i between the beginning and the end of the year t. The overall adjusted TFR is then computed as a sum of the order-specific adjusted TFRs. The use of this method has been challenged by some researchers who suggest that the assumptions behind the adjustment formula may be unrealistic (e.g. Van Imhoff and Keilman, 2000) and that the interpretation of the adjusted TFR is problematic (Smallwood, 2002). In particular, the Bongaarts-Feeney formula assumes that the period schedule of fertility rates is moving towards younger or later ages by the same pace among all women, and the shape of the fertility schedule by age remains intact. Although this is not a realistic assumption, Zeng Yi and Land (2001) concluded that the Bongaarts-Feeney adjustment is usually not sensitive to its underlying assumption concerning the invariant shape of the fertility schedule, and Sobotka (2003) found that despite large fluctuations, it provides values considerably closer to the completed cohort fertility than do the non-adjusted TFRs. Given the availability of data, it provides a relatively simple estimate of the intensity of timing distortions of period fertility in most countries of central and eastern Europe. To reduce random fluctuations, we present average values of the adjusted TFR for periods of 3 to 4 years.
The adjustment method proposed by Bongaarts and Feeney (1998) estimates the total fertility rate free of timing distortions caused by the postponement or advancement of childbear-ing. It is calculated for each birth order separately as , where ri,t is the change in the mean age at childbearing at birth order i between the beginning and the end of the year t. The overall adjusted TFR is then computed as a sum of the order-specific adjusted TFRs. The use of this method has been challenged by some researchers who suggest that the assumptions behind the adjustment formula may be unrealistic (e.g. Van Imhoff and Keilman, 2000) and that the interpretation of the adjusted TFR is problematic (Smallwood, 2002). In particular, the Bongaarts-Feeney formula assumes that the period schedule of fertility rates is moving towards younger or later ages by the same pace among all women, and the shape of the fertility schedule by age remains intact. Although this is not a realistic assumption, Zeng Yi and Land (2001) concluded that the Bongaarts-Feeney adjustment is usually not sensitive to its underlying assumption concerning the invariant shape of the fertility schedule, and Sobotka (2003) found that despite large fluctuations, it provides values considerably closer to the completed cohort fertility than do the non-adjusted TFRs. Given the availability of data, it provides a relatively simple estimate of the intensity of timing distortions of period fertility in most countries of central and eastern Europe. To reduce random fluctuations, we present average values of the adjusted TFR for periods of 3 to 4 years.
[23]A mean age of women at the birth of their first child lower than 24 years and a cumulative fertility rate of teenage women above 100 per thousand.
A mean age of women at the birth of their first child lower than 24 years and a cumulative fertility rate of teenage women above 100 per thousand.
© Cairn 2007
Vie privée |
Conditions d’utilisation |
Conditions générales de vente
À propos | Éditeurs | Bibliothèques | Aide à la navigation | Plan du site | Raccourcis
À propos | Éditeurs | Bibliothèques | Aide à la navigation | Plan du site | Raccourcis
Retour en haut de page
Ici s'achève le contenu de la page, la suite ne concerne que les attributs d'affichage