Fertility strand abstracts

Accounting for low fertility in Scotland relative to England.

Elspeth Graham¹, Paul Boyle¹, John Ermisch², Vernon Gayle³ and George Bouliotis^1
1University of St Andrews, ²University of Essex, ³University of Stirling

In the three decades after the Second World War, fertility in Scotland was consistently higher than fertility in England and Wales. Around 1980, the situation was reversed and Scotland's fertility became the lowest in the UK. By 2004, Scotland's total fertility rate was 1.60 children per woman compared to 1.79 in England. This paper investigates these relative fertility differences using retrospective fertility history data from the British Household Panel Survey (BHPS). Using a demographic accounting approach we examine the fertility behaviour of those born since 1965 who entered their childbearing years at or after the fertility reversal of 1980. We compare the rates of childlessness and time to first and subsequent births and demonstrate some significant differences in fertility behaviour between Scotland and England.

Email: efg@st-andrews.ac.uk

Fertility and women's education: A cohort analysis.

Mike Brewer¹, Anita Ratcliffe² and Sarah Smith^3
1Institute for Fiscal Studies, London, ²Centre for Market and Public Organisation, University of Bristol, ³CMPO and IFS

We use data from the UK Family Expenditure Survey (1968 - 2003) to analyse trends in fertility across successive cohorts of women, using age of mother and children in the household to infer mother's age at birth and birth order; we explore possible sources of measurement error. The advantage of the FES is we can look at trends over a 35-year period, and explore the link between fertility and the education of the mother.

The main findings are:

• Beginning with women born in the 1940s, average family sizes have fallen by approximately 0.5 child.
• The first phase (starting with women born in 1935) saw a fall in third and subsequent births
• The second phase (starting with women born in 1945) saw postponement of childbearing and an increase in childlessness.
• Women who go on to further/higher education (18+) have very different fertility to those who do not; they typically start having children later and, on average, have smaller families. But, the overall decline in average family size cannot be accounted for simply by the rise in female participation in further/higher education. A striking trend over the last 25 years has been the increase divergence in fertility by education.

These findings have a number of implications, including:

• Predicting the effect of recent increases in the number of women going on to higher education is not straightforward since fertility depends not (just) on education, but on subsequent career choices.
• If family circumstances at birth matter for child development, the increasing divergence of fertility by education amplifies the advantage of children born to better-educated mothers.

Email: m.brewer@ifs.org.uk

Changes over time in later childbearing and childlessness in Britain: The 1958 and 1970 cohorts compared.

Roona Simpson
Centre for Research on Families and Relationships, University of Edinburgh.

Changing fertility patterns in Britain over the last three decades have been characterised by a rising average age at first birth and higher levels of childlessness (ONS 2005). The postponement in timing of first births has been identified as a main factor in overall fertility decline, related to both smaller family sizes and increased rates of childlessness (McDonald 2000, Sobotka 2004). In Britain, the mean age of women having their first birth in 2004 was 27, a rise of over 3 years from 1971, while around one in five of women currently reaching the end of their fertile life are childless, compared to one in ten women born in the mid-1940s (ONS, 2005). However, this delay in childbearing is socially selective, and motherhood in Britain in particular is characterized by high levels of polarization by factors such as educational qualifications and occupational status: the better educated a woman is, the more likely she is to defer childbearing (Joshi 2002), while levels of childlessness are also highest amongst highly educated women (Ferri et. al. 2003).

Providing detailed information on who is delaying childbirth and remaining childless will contribute to our understanding of the social processes underlying these trends. This paper reports the findings of analysis conducted on the 1958 and 1970 cohort studies to investigate the socio-economic characteristics of those delaying childbirth. It looks firstly at those remaining childless in their early thirties, and compares cohorts to assess changes over time. It also compares the characteristics of childless men and women at this age: despite increasing attention to fertility decline, there is relatively little attention and research looking at the situation of men. The analysis also follows those women born in 1958 who were childless in their early thirties to age 46, the end of their reproductive life. Here, the characteristics of those who subsequently became mothers are compared with those experiencing lifetime childlessness.

Email: roona.simpson@ed.ac.uk

Fertility variations in Scotland.

Paul Boyle¹, Dan Allman², Ian Dey², Zhiqiang Feng¹, Elspeth Graham¹, Lynn Jamieson², Fran Wasoff^2
1University of St Andrews, ²University of Edinburgh

This paper explores the role of socio-cultural factors in shaping fertility ideals, expectations and behaviour in Scotland. The paper is based on interim findings of research sponsored by the ESRC and the Scottish Executive into fertility variations in Scotland. The paper draws mainly on evidence obtained through a module incorporated into the 2005 Scottish Attitudes Survey. The module included questions on the reproductive biography and family circumstances of respondents, current fertility ideals and expectations, attitudes to factors commonly associated with fertility (the costs of child-rearing, work-life balance issues, income and housing constraints etc), the richness of family and friendship networks, and perceptions of local environments as hostile or friendly to families. In exploring geographical variation in fertility the study draws on a wide range of variables from linked datasets.

Email: p.boyle@st-andrews.ac.uk 

Government induced fertility decline.

Patrick Carroll
PAPRI

Government policies in several areas have had an impact on fertility which is comparatively little studied by demographers. But these same policy areas will need to be reconsidered if and when government policy is to be amended to enable a recovery in the birth rate to take place. Government policies in four main areas are considered:

• Health. Contraception and abortion in the UK are largely funded by the National Health Service. The historic provision of these services is examined in relation to the decline in the birth rate since the 1960s.
• Taxation and Social Benefits affecting Families with Children. Income tax in relation to couples is to be examined over the epoch since the 1960s with regard to the impact on marriage and fertility alongside trends in universal benefits, insured contributory benefits and means-tested discretionary benefits in the UK for families with children.
• Housing and Housing Finance. Developments affecting housing tenure such as the privatisation of council housing and modern policy on social housing are to be examined. Housing finance and trends in interest rates are also considered as to their impact on fertility.
• Educational and Career Opportunities for Women. Equal access for women to higher education, professional opportunities and higher paid posts leads to change in relative incomes for men and women with implications for the birth rate. Fertility differentials between households with higher earning females and male partners are to be examined.

PAPRI Pension and Population Research Institute, 35 Canonbury Road, London N1 2DG

De-Secularisation?: Religiosity, fertility and politics

Eric Kaufmann
Birkbeck College, University of London

This paper aims to test the thesis that the world's population will become increasingly religious and politically conservative in the twenty-first century, reversing decades - even centuries - of secularisation. The dynamic for this change will be secular-religious fertility differences. Recent cross-national research confirms the relationship between religiosity and political conservatism (Norris & Inglehart 2004; Guth et al 2005). The study will use data from the General Social Survey (1972-present), European Social Survey (2004), European Values Survey (1981-2000) and possibly the ONS Longitudinal Study (1971-2001) to generate some speculative projections. The paper will argue that religious-secular fertility differences remain marked in Europe, though this is currently offset by large-scale religious apostasy between generations. While the short and medium term outlook in Europe is toward a steady-state of falling or low religiosity among (white) dominant ethnic groups, this is not true in the USA and elsewhere, and also less accurate for the growing ethnic minority population within Europe. In addition, secularisation-as-church attendance is far more apparent in Europe than the secularisation of religious belief. The wider ideological and political implications of these shifts will also be touched upon.

Email: eric.kaufmann@birkbeck.ac.uk 

Interpreting sex ratios at birth: the case of Singapore

Elspeth Graham
University of St Andrews

Throughout much of Asia, high male-to-female sex ratios at birth have persisted despite increasing levels of economic prosperity, urbanisation and female education and employment. Male bias at birth has been interpreted as the outcome of an enduring cultural preference for sons and population researchers have argued that fertility decline accompanying economic modernisation has resulted in the intensification of male bias in countries such as India, China and the Republic of Korea. In the Asian context, greater prosperity and smaller family size, it seems, not only fail to weaken traditional gender norms but also conspire to increase discrimination against daughters.

This paper examines the trend in sex ratios at birth in a Newly Industrialised Asian Country, Singapore. As one of the wealthiest countries in Asia, Singapore provides a good test case for investigating the intensification hypothesis. The findings confirm a close association between the fertility rate and the sex ratio at birth. However, the recent dramatic increase in male bias apparent in other Asian populations is absent in Singapore. Rather, sex ratios at birth fluctuate around the upper limit of what is conventionally taken to be the normal range. This trend, it is argued, presents particular problems of interpretation, which challenge demographers to consider the influence of biology as well as culture on the proportion of male-to-female births in a population.

Email: efg@st-andrews.ac.uk 

Stopping sooner or starting later? Fertility decline in Uzbekistan.

David Clifford
University of Southampton

The Ex-Soviet states of Central Asia have been relatively neglected in the demographic literature. This gap is unfortunate given the importance of context to an understanding of fertility change, and the particular socialist history of the region. This study provides an analysis of fertility change in Uzbekistan, the most populous of the Central Asian republics. Substantively, interest lies in assessing whether fertility decline has been affected through a 'starting later' pattern, characterised by the postponement of childbearing, or a 'stopping sooner' pattern, characterised by a reduction in childbearing at later ages. The demographic measures calculated are designed to distinguish between these two scenarios. Cohort and period trends in the ages at first marriage and birth are complemented by cohort perspectives on the first birth interval, subsequent birth intervals, and cumulated fertility at different ages. Data are extracted from the 2002 Uzbekistan Health Examination Survey (UHES). There is evidence that the country has followed both models of fertility decline: earlier birth cohorts (1953-57 through to 1963-6) followed a 'starting later' pattern; later cohorts (1963-67 through to 1973-77) followed a 'stopping sooner' pattern. This latter trend is rooted in the complex set of changes associated with the end of Soviet-style socialism: in particular, greater economic hardship during the post-independence years served to discourage childbearing beyond the culturally-expected first birth. The research therefore serves to reinforce the importance of local context to an understanding of fertility change.

Email: dmc104@soton.ac.uk 

Socio-economic development, the social structure, and Kenya's fertility transition.

Ekisa Anyara and Andrew Hinde
University of Southampton

Recent studies have shown that Kenya's high fertility of 8.1 children in 1978 had declined by 39% by 2003. The findings from an analysis of the role of the proximate determinants in the decline proved inconclusive and explanations for the decline remain unclear. Large declines are in evidence in some regions, especially Nyeri, Meru and Nairobi, which have higher Human Development Index scores as compared to regions with low Human Development Index scores. Some other regions, particularly Mombasa and Nyanza, have experienced a fertility decline despite persistently low contraceptive prevalence. Further, both rural and urban regions with the lowest fertility report the highest proportion of births outside marriage. These features suggest a complex pattern of regional variations in Kenya's fertility decline. However, little evidence exists on the role of socio-economic development and the social-structure in explaining regional fertility variations in Kenya. This paper determines the impact of socio-economic development and social structural factors on regional fertility levels in Kenya. The Kenya DHS data collected in 1989, 1993, 1998 and 2003 are used and consistent regional boundaries across all surveys are maintained. An extra-Poisson model is used to model births in the last four years prior to the survey date as a function of a set of socio-economic development and socio-structural variables.

Email: ela@soton.ac.uk 

The highest fertility in Europe - for how long? The analysis of fertility change in Albania based on individual data.

Arnstein Aassve¹, Arjan Gjonca² and Letizia Mencarini^3
1Institute for Social and Economic Research, University of Essex, ²London School of Economics,
³University of Florence.

Our knowledge about demographic behaviour in Albania is limited. Though Falkingham and Gjonca (2001), using census data, provide very useful insights into the fertility transition in Albania from 1950 to 1990, nothing is known about fertility behaviour during the nineties. From being one of the most isolated countries of the world, Albania has embarked on a remarkable transition which involves dramatic political and economic change. The new and emerging situation of Albania is bound to have profound impact on society and the behaviour of individuals within it. The introduction of the Albanian Living Standard and Measurement Survey (ALSMS) in 2002 provides unique information about demographic change. Using this information we analyse fertility behaviour in terms of the quantum and tempo during the period following the collapse of the communist era. We use simple non-parametric survival analysis together with more sophisticated event history models. A preliminary version of the paper demonstrates strong cohort and period effects, but which are quite different depending on birth parity. The Albanian LSMS is a rich data set and we plan to extend the analysis by introducing more comprehensive specifications of fertility behaviour, including controls for migration. We will also extend the analysis by estimating a random effect version of the hazard regression and possibly perform simulation analysis to assess the sensitivity of parameter estimates.

Email: aaassve@essex.ac.uk 

From first to second child in Italy: Do child care arrangements matter?

Francesca Fiori
University of Rome "La Sapienza"

This study provides a contribution to the study of lowest low fertility in Italy, a country characterized by TFRs below 1.3 since the early nineties. Using event history analysis applied to data from the first two waves of the ILFI (Indagine Longitudinale sulle Famiglie Italiane: ILFI, University of Milano Bicocca, University of Trento, University of Bologna. Scientific coordinator: Prof. Antonio Schizzerotto), the study focuses on the transition from first to second birth - the event which has a particularly crucial impact on lowest-low fertility.

Since one of the causes of very low fertility in Italy is seen in the lack of institutional support for mothers, which is only partially mitigated by the help provided by informal networks of relatives and friends, this work aims to understand how formal and informal support in childcare in the three years following the birth of the first child may influence second birth risks, controlling for the usual socio-economic and demographic characteristics of mothers.

Email: francesca.fiori@uniroma1.it