Fertility & reproductive health abstracts

 Strand organisers: Stuart Basten, Melanie Channon, University of Oxford

Fertility and Gender - Monday 7 September 1.30pm

Partners’ relative earnings and fertility intentions                                                       Agnese Vitali1, Maria Rita Testa2, ESRC Centre for Population Change, University of Southampton1, Vienna University2

The role of gender roles in reproductive decision-making has been so far under-investigated, especially for what concerns its association with fertility intentions, i.e. the strongest predictors of reproductive behaviour. For example, the Theory of Planned Behaviour and the Traits-Desires-Intentions-Behaviour do not explicitly refer to the role of gender-related factors in shaping individual reproductive intentions. This paper studies the association between changing gender roles and fertility intentions in Europe by looking at the role of relative earnings among partners, a measure worth of investigation given that between 20-25% of couples in developed countries today are such that women out-earn their partners. We formulate hypotheses on the expected sign of the correlation between fertility intentions and relative earnings on the basis of the economic theory and the gender theory of fertility. We use data from the European Social Survey (2010/11) and model the probability of intending to have a child in the next 3 years using logistic regressions run separately for women and men. Explanatory variables include: earning arrangement (man as main earner, equal earners, woman as main earner), woman’s age and educational attainment, age and education differential between partners, n. children (0, 1, 2, 3+), experience of economic difficulty in the last 3 years. Results suggest that fertility intentions are only weakly associated with the earning arrangement. Women and men at parity 0 and 1 are significantly more likely to intend to have another child with respect to women and men at higher parities, irrespectively of their relative earning. 

Email: a.vitali@soton.ac.uk 

Fertility and socioeconomic gender equality between a couple – A Bayesian analysis
Beata Osiewalska, Cracow University of Economics, Poland 

Connections between couples’ socioeconomic status and fertility have recently attracted noticeable attention, especially in the context of on-going changes of the gender roles and a growing variety of family models. Although gender differences have been examined quite deeply, couples’ procreative behaviour treated as a mutual result of male and female socioeconomic characteristics remains under-researched. Previous studies have suggested that the proper inference about procreative behaviour of a couple should be performed not only by analysing the characteristics of both partners, but also by considering at the same time the childless population, as well as the population of parents. Therefore, the aim of this study is to investigate couples’ procreative behaviour with regard to the gender socioeconomic (in)equality between partners, taking into account that the behavioural drivers could differ among parents and childless couples. The results showed that including characteristics of both partners significantly improved the ability to describe fertility behaviour. In particular, the overall level of socioeconomic status of a couple occurred to have negative influence on fertility timing, while the U-shape relation was found for their completed fertility. Beyond that, the overall level of socioeconomic status differentiates the procreative behaviour of a couple more than the distribution of this status between partners, yet it is necessary to include both dimensions to fully describe possible relations. The Bayesian Zero-Inflated Poisson framework, which allows considering two states (childlessness and parenthood) within one statistical model, is applied. The empirical illustration is based on the Generations and Gender Survey dataset.  

Email: beata.osiewalska@uek.krakow.pl

Socio-economic differentials in the uptake and effects of (in)formal childcare on second birth hazards in Belgium
 Karel Neels, Jonas Wood, Tine Kil, University of Antwerp

The positive association between fertility and female employment in OECD countries suggests that family policies have played an important role in reducing the ‘parent-worker’ conflict. The empirical literature, however, finds only small positive effects of family policies on fertility, but has typically failed to consider eligibility and uptake of family policies at the individual level, as well as population heterogeneity in the uptake and effect of these policies. Using longitudinal individual-level data from the 2001 Census and the National Register, we document socio-economic and educational differentials in the uptake of formal childcare (kindergarten, daycare mothers) and informal childcare arrangements (family or household members) in Belgium in 2001 and analyze the effect on second birth hazards in the period 2002-2005. In line with theoretical expectations, results show that 70 per cent of the higher educated women make use of formal childcare arrangements compared to 35 per cent of the middle and 20 percent of the lower educated women. Among middle educated women informal care is the dominant mode of childcare whereas lower educated women have the highest probability of not using any type of formal or informal childcare arrangement. With respect to the effect of childcare on second birth hazards, the analyses show that focusing on each mode of childcare in turn yields inconsistent results, whereas a typology of childcare arrangements simultaneously considering uptake of different modes of childcare as well as combined uptake shows strong positive effects of informal and particularly formal childcare arrangements on second birth hazards which are particularly articulated among higher educated women. 

Email: karel.neels@uantwerpen.be 

Fertility intentions in Scotland: gender differences, social attitudes and family formation
Annemarie Ernsten, Centre for Population Change, University of St Andrews

This paper argues that studies of fertility intentions need to place intentions within a longitudinal framework, which recognises the importance of gender differences and attitudes regarding family formation. It examines the role of various determinants of fertility intentions, including social attitudes (relating to children, employment and housing) and housing characteristics. Previous studies have noted how factors impacting fertility decisions (for example, education and labour force status) have different effects for men and women. However, there has been little research into the impact of social attitudes, and their gendered dimensions, on fertility intentions. The present study addresses this research gap. The analyses use data from the British Household Panel Survey to address the following research question: Do socio-economic differences and shared social attitudes regarding family formation impact the fertility intentions of men and women in similar ways? The analytical sample comprises childless respondents of reproductive ages. First, latent class analysis is used to generate groups with shared social attitudes. Next, these attitudinal groups, along with other potential determinants, are entered in a multinomial logistic regression model predicting fertility intentions. The results show a significant impact of shared social attitudes on fertility intentions for both men and women. Gender differences in the impact of socio-economic variables on fertility intentions are confirmed, with women’s intentions being more sensitive to education and men’s to employment status. The paper concludes by discussing the wider implications of the findings for the understanding of gender differences in the relationships between social attitudes and family formation.  

Email: ae22@st-andrews.ac.uk

Fertility: Methods and Theories – Monday 7 September 4.45pm

Using ‘pseudo panels’ to establish causal link between HIV and fertility in Kenya     M.A. Magadi, School of Social Sciences, University of Hull

The relationship between HIV/AIDS and fertility is a complex one, partly because causality can run in either direction. While HIV/AIDS can affect fertility desires and outcomes, it is also possible for fertility to affect the risk of HIV/AIDS and disease progression. This paper focuses primarily on fertility as the outcome of interest and HIV/AIDS as a contributing factor. Repeated cross-sectional Kenya Demographic and Health survey (KDHS) data are used to construct “pseudo panels” (Deaton, 1985) based on birth cohorts by ethnicity. The pseudo panels aim to allow analysis of possible causal link between HIV/AIDS factors in an earlier survey (2003 KDHS) and fertility behaviour of similar cohorts in a subsequent survey (2008 KDHS). Preliminary analysis provides no evidence of a significant link between HIV prevalence in 2003 and births in the next five years preceding 2008 KDHS. However, there was evidence that cohorts with higher HIV prevalence in 2003 were more likely to desire to stop childbearing in 2008. More salient associations were observed between perceived HIV risk in 2003 and fertility intentions/behaviour in 2008. Cohorts with perceived low risk of HIV had significantly higher fertility during the 5-year period preceding 2008 KDHS and also reported higher ideal family size and desire for more children. Further multivariate analysis based on fixed effects models will determine the extent to which observed patterns may be attributable to key demographic/ socio-economic differences or infer possible causal links.

Deaton, A (1985). Panel data from time series of cross-sections. Journal of Econometrics vol. 30(1-2): 109-126.   

Email: m.magadi@hull.ac.uk

Why is there a two-child norm in Pacific Asia? 
Stuart Basten, University of Oxford

Despite having some of the lowest period total fertility rates in the World, Pacific Asia is generally characterised by having a persistent two-child norm in the literature. Recently, for example, Sobotka and Beaujouan (2014) included these territories within their discussion of a European two-child norm in order to suggest that such a norm existed elsewhere. This paper explores in greater depth the evidence which, for example, Sobotka and Beaujouan cite, and argues that a key determinant of the so-called 'two-child norm' is the fact that in most Pacific Asia Surveys, only married women are included. Given that the 'retreat from childbearing' in the region has alternatively been called 'a retreat from marriage' - especially in the context of ever increasing FMAM - this is critical. The paper concludes that survey instruments which focus on married women in their early twenties are, in fact, increasingly examining the fertility behaviour of outliers rather than truly representative samples. 

Email: stuart.basten@spi.ox.ac.uk 

The role of transnational groups of reference in understanding individuals’ childbearing intentions
Joanna Marczak, London School of Economics 

Background: Social scientists have long highlighted the importance of socioeconomic conditions in theorising and investigating key issues related to fertility trends. Researchers have so far focused entirely on processes occurring ‘within’ nation states to examine factors relevant for childbearing decision making with consequent assumptions that improving socioeconomic conditions at national level could reverse downward fertility trends in low-fertility countries, especially if family size ideal remains fixed at around two children. Aims: This paper investigates how framing of socioeconomic factors differs between individuals’ narratives of childbearing intentions and that conceptualised in demographic theories and research. We empirically investigate whether transnational framing can add explanatory element to the enquiry of fertility trends in increasingly interconnected societies. Data and methods: This paper is based on completed PhD research. Qualitative in-depth interviews (n=42) with Polish mothers and fathers living in London and Krakow were conducted in 2010/2011 to explore childbearing intentions’ rationales and thinking processes underlying individuals’ rationalizations. Conclusions: The data indicate that transnational groups of reference are imperative in understanding individuals’ childbearing intentions and factors vital for their realisations, which contrasts with the current exclusive focus on national framing in demographic research. Notwithstanding scholars’ assumptions that improving socioeconomic conditions at national level can contribute to rebound in fertility rates, our data indicate that such improvements alongside the existence of transnational groups of reference could simply raise individuals’ expectations regarding socioeconomic conditions necessary for childbearing. However, the relationship between transnational groups of reference and childbearing intentions varies depending on subpopulations. 

Email: j.marczak@lse.ac.uk

The Consequences of Fertility Timing – Tuesday 8 September 9.00am

Session organised by Dr. Alice Goisis, London School of Economics 

Does teenage pregnancy affect social mobility? Decomposition using the KHB-method
Heini Väisänen, London School of Economics and Political Science

Background: Parents’ socioeconomic position (SES) is associated with that of their children. Teens from less advantaged backgrounds more often become pregnant, and among pregnant teens higher socioeconomic background is associated with a higher likelihood of abortion. Teen parents often do not achieve high SES. This study examines the degree to which intergenerational social mobility is mediated through teen fertility behaviour. Methods: I compare whether educational attainment at age 30 is associated differently with parental SES among those who have no teen pregnancies, those who have an abortion, and those who have a birth using a unique set of Finnish register data of three birth cohorts (born in 1950s-70s). The KHB-method was used to decompose the direct and indirect relationship between parental SES and education via teen pregnancies; and the degree to which the effect of teen pregnancies on education was mediated through later pregnancies. Results: Children of upper-level employees were the most likely to pursue higher education. 7-14% of the total effect travelled through teen births and abortions. Teenage births mediated a larger proportion of the effect than abortions. Teen pregnancies were strongly associated with lower odds of having higher education, but 13-31%, of the total effect was mediated through later pregnancies. Conclusions: The accumulation of disadvantage among teen parents is of concern in Finland. Fertility behaviour in early 20s played an important role too. Policy makers should make sure that all women who wish to pursue high education can do so regardless of the timing of their pregnancies. 

Email: h.e.vaisanen@lse.ac.uk 

Determinants and later life outcomes associated with early parenthood: cross-European comparisons
Katherine Keenan, Emily Grundy, London School of Economics

In Europe early parenthood is associated both with early and later life socio-economic and health disadvantages, reflecting selection effects and the ‘causal’ effect of early parenthood. Given the diverse family-building contexts in Europe over time and place, this paper aimed to investigate regional, gender and cohort variations in these relationships among women and men in Europe. We used data from waves 1-3 of the Survey of Health, Ageing and Retirement in Europe (SHARE), which includes retrospective life course information for 10 countries from four European regions (East, North, South, and West). We investigated gender, regional and cohort variation in associations between childhood circumstances and early parenthood, and between early parenthood and later outcomes including employment, wealth, experience of divorce, large family size, and quality of life among parents aged 50-79 years. Results of multivariable regression models showed that low childhood socio-economic position (SEP) was consistently associated with earlier parenthood, but that associations were stronger for women and in non-Eastern regions. Early parenthood was strongly associated with subsequent divorce/separation and eventual high parity in both men and women. Early motherhood was associated with subsequent lower wealth and lower quality of life at ages 50-79, although not in the Eastern region. Associations between early parenthood and labour market participation differed for men and women. Findings indicate the importance of considering gender and contextual influences on early parenthood and subsequent life trajectories. 

Email: k.keenan1@lse.ac.uk 

Reproductive trajectories of teenage mothers: a Franco-British comparison
John Tomkinson, French Institute for Demographic Studies (Ined), University of Strasbourg 

In the European context of a postponement of first birth timing, becoming a teenage mother represents an act seemingly contradictory to increasingly normalised reproductive behaviours. Does an early entry into motherhood play a determining role in subsequent reproductive outcomes? This paper examines to what extent women having become mothers during their adolescence differ from those having had their first child at a later age in terms of total lifetime fertility and timing of subsequent births, and this for women in France and the UK. Using two retrospective surveys, Understanding Society and the French Enquête Famille et Logements (Family and Households survey), we reconstruct reproductive trajectories of over 100,000 women in the two countries in order to see what happens in terms of fertility after a teenage birth. The research questions of this paper are:  1. How has the fertility of teenage mothers evolved over time and how does it differ between the two countries?  2. Does a teenage birth lead to a higher level of lifetime fertility?  3. With respect to the timing of subsequent births, is a teenage birth a “shock” to the tempo of childbearing or a “trigger” event which sets in motion the childbearing of a woman? 

Email: john.tomkinson@ined.fr 

Son Preference – Tuesday 8 September 11.00am

Skewed sex ratios and sex-selective abortions in Nepal: An analysis using census data
Melanie Channon1, Gyanendra Bajracharya2, Mahesh Puri3, Stuart Basten1, University of Oxford1, Central Bureau of Statistics Nepal2, CREHPA3

Sex-selective abortion and the skewed sex-ratios which result are an emerging issue in Nepal. In the wake of abortion being legalised and increasing availability of prenatal sex-determination technologies, it is feared that sex-selection may become widespread. The Nepali context makes pervasive use of sex-selection likely, as sons are more highly valued than daughters on both religious and socio-economic grounds. Furthermore, the ongoing reduction in fertility means that couples feel pressure to bear a son at a low parity. This paper uses data from the 2011 Nepal population census to estimate the absolute number of sex-selective abortions that have occurred. Additional analysis of census data also allows a detailed breakdown of the geographic distribution and socioeconomic correlates of sex-selective abortions. Multivariate analysis will be carried out to look at the determinants of the district level child sex ratio. It is estimated that roughly 10,000 sex-selective abortions occurred in Nepal in the year before the census, with a further 17,000 in the four years before that. The majority of sex-selection is concentrated in the Kathmandu Valley and other urban areas as well as on the Indian border. It is anticipated that use of sex-selective abortions may spread as Nepal urbanizes and develops.  

Email: melanie.channon@ageing.ox.ac.uk 

Son preference and intentions to move to parity two: Evidence from Taiwan Georgia Verropoulou1 & Stuart Basten, 1University of Piraeus, 2University of Oxford

A number of studies have identified the role of son preference in family formation, especially through so-called the ‘stopping mechanism’ whereby having the desired number of sons curtails further reproduction. Taiwan is a country formerly characterised by highly skewed sex ratios at birth which have now, largely, been normalised. Taiwan currently has one of the lowest fertility rates in the world, leading to projections of rapid population ageing and decline. Recently, policies designed to support childbearing have been introduced and some optimism for their future success has been drawn from the fact that the ‘ideal’ number of children stated in Taiwanese surveys is over two. In the present paper we use data from the 2006 and the 2010 rounds of Women’s Marriage, Fertility and Employment Survey to investigate fertility intentions of childless and parity one women and whether in the latter case these are influenced by the gender of the first child. While we found some evidence for a ‘two-child norm’ among childless women, this could be an unrealistic ideal as the majority of women with one child do not intend to have another. Descriptive analysis as well as use of logistic regression models indicate that the decision to cease childbearing is shaped by a strong son preference; unemployment of the male partner seems also important. 

Email : gverrop@unipi.gr 

Religion or national context? Examining Hindu-Muslim differentials in the demographic manifestations of son preference across South Asia
Ridhi Kashyap, Melanie Channon, University of Oxford 

This paper investigates how demographic manifestations of son preference vary between Hindus and Muslims across South Asia to explore if religious or national context is a more salient determinant of son preference in the region. The literature on South Asia has overwhelmingly focused on India, and largely argued that son preference exists as a family-building behaviour among Hindus (Bhat and Zavier 2003; Nasir and Kalla 2006). Indeed, the presence of strong son preference among Hindus compared with Muslims has often been proposed as an explanation for the paradoxically lower levels of child mortality among Muslims compared with Hindus, despite their otherwise lower socio-economic status. Among Hindus, it is argued, strong son preference particularly disadvantages Hindu female children leading to higher overall levels of child mortality (Bhat and Zavier 2005; Bhalotra et al. 2010). Empirical evidence for this claim however has been mixed for India. While data on fertility preference suggest that Hindus show higher son preference (Bhalotra et al. 2010), as indicated by a reported preference for families with a greater number of sons than daughters, sex-differentials in child mortality exist similarly in both communities (Guillot and Allendorf 2010). This study develops a more nuanced understanding of how son preference affects childbearing behaviour for Hindus and Muslims by using data from the Demographic and Health Surveys for India, Bangladesh, Pakistan and Nepal. We explore outcomes such as differential stopping behaviour, sex ratios at birth, and sex differentials in child mortality to characterize the demographic manifestations of son preference in each of the countries. Our analysis suggests that national contexts matter more than religion in determining the demographic manifestations of son preference in South Asia. We find evidence for similar patterns of differential stopping behaviour and sex differentials in child mortality among Indian Hindus and Muslims. In Nepal Hindus act similarly to North Indian Hindus in terms of high levels of differential stopping behaviour, but with less skewed sex ratios at birth and childhood. Pakistani Muslims show higher levels of fertility and less clear evidence for differential stopping behaviour given their high fertility levels, whilst Bangladeshi Muslims show lower fertility as well as limited evidence of differential stopping behaviour. 

Email: ridhi.kashyap@nuffield.ox.ac.uk 

European Fertility – Tuesday 8 September 5.00pm

Spacing between children and trends in mean age of successive birth orders: quite different stories!                                                                                                        Marion Burkimsher, University of Lausanne, Switzerland

The increase in age at childbearing over recent decades is well-researched; where birth registration data by biological birth is recorded, then trends in mean age at 1st, 2nd, 3rd and subsequent births can also be monitored. This data has been modelled from marital birth order in Switzerland since 1969 (Burkimsher 2011). In addition to the rise in age of each successive birth order, there has been a decline in the gap between mean age of successive birth orders. The 1st-2nd gap is now less than 2 years, and the 2nd-3rd and 3rd-4th gaps are only 1.2 years. The variability in age at first birth, which was the least variable in 1969, rose sharply in the 1990s and is now the most variable. Data on spacing between children is less commonly available, but in Switzerland the number of children borne and their years of birth was recorded for the whole population in the census of 2000. This shows that the gap between children has remained remarkably constant over the past decades. The mean interval between first and second children is around 3 years and is a little higher between second and third and third and fourth children. These apparently contradictory results are explained by different sub-populations developing different fertility behaviour, with educational level being the probable driver. Increasingly, women who end up with larger families have their first child in their 20s, while women who start after 30 end up with only one or two children.  

Email: drmarionb@gmail.com 

The educational gradient of fertility timing intentions: Regional differences in Europe
Maria Rita Testa, Fabian Stephany, Wittgenstein Centre for Demography and Global Human Capital (IIASA, VID/OEAW, WU), Vienna Institute of Demography of the Austrian Academy of Sciences, Vienna University of Economics and Business

Previous investigations have found that fertility intentions, unlike actual fertility, are positively correlated with education. Social norms, postponement of childbearing or conflicting career and family plans of highly educated women are possible explanations for this phenomenon. The aim of this study is to investigate the conditions under which a positive relationship between women’s educational level and childbearing timing intentions is observed. Using fertility timing intentions of individuals from 34 European countries we conduct a large comparative study with more than 115,000 observations. Results indicate a positive educational gradient if a significant gap in fertility intentions between low and high educated women is observed. It can be noticed that the gap widens with higher parities for some countries. For central, central-east and south-east regions of Europe we observe a strengthening of the educational gradient, mainly driven by higher fertility intentions of well-educated women at parity one or two and above. 

Email: fabian.stephany@wu.ac.at 

A changing relationship between fertility and economic development at the sub-national regional level? Theoretical considerations and evidence from Europe Jonathan Fox1, Sebastian Klüsener2, Mikko Myrskylä2, 3, Freie Universität Berlin1, Max Planck Institute for Demographic Research2, London School of Economics3

Recently presented evidence for nation states suggests that the long-standing negative relationship between socioeconomic development and fertility turns positive at high levels of development. This paper introduces the sub-national regional dimension to this debate. It focuses on Europe, which comprises a number of highly developed countries with comparatively high fertility rates. In the theoretical section, we revisit considerations about how modernization processes in the 19th and 20th centuries led to a negative association between economic development and fertility outcomes across sub-national regions of developed countries. This is followed by the discussion of important recent trends, which offer possible reasons underlying a potential reversal of the fertility-economic development relationship at the sub-national regional level. We then investigate empirically whether a convex relationship between the regional total fertility rate and economic development is detectable within European countries. For this we analyse data for 21 European countries covering the period 1991-2011. Using a country-by-year fixed effects model, we find evidence for a convex relationship between fertility and per capita income in a large number of the observed countries. Consistency checks, in which we attempted to control for tempo effects, support the view that our findings are not just driven by postponement effects. However, the factors that predominantly determine this turn-around seem to vary across countries. The same is true for the level of economic development and the fertility levels at which the turn-around is observed. 

Email: kluesener@demogr.mpg.de 

Socio-economic development and emerging sub-national geographies of fertility: the case of Spain
Albert Sabater1, Elspeth Graham1, John Macinnes2, University of St. Andrews1, University of Edinburgh2 

Recent studies suggest that advances in socio-economic development are now associated with reversals of fertility decline, both at a global scale and within Europe. This paper uses spatial analysis techniques and time-series cross-sectional data for years 1980 to 2010 to investigate fertility within Spain, one of the lowest-low fertility countries in Europe. At the national scale, Spain experienced some fertility recovery in the early 2000s (TFR rose from 1.15 in 1998 to 1.46 in 2008), but fertility returned to lowest-low levels following the economic downturn and emigration. We ask whether there are sub-national variations in the associations between fertility, development and other correlates and how these have changed over time. Our aim is to determine which regional characteristics are important for fertility change. Our analyses indicate that the local associations between fertility and its correlates are not homogenous across regions. While development (HDI) is positively associated with TFR in urban provinces, a negative relationship is found in intermediate and rural provinces. We seek to explain these spatial differences by examining the strength of local associations and find that - more important than development - it is the contribution of migrant fertility which shows the greatest impact on local fertility via both direct and indirect effects. Our analyses contribute to existing debates on emerging sub-national geographies of fertility in Spain and elsewhere, as well as providing further insights into both the impact of the recent economic recession on fertility and the circumstances in which a reversal of fertility decline might occur 

Email: asc6@st-andrews.ac.uk 

UK Fertility - Wednesday 9 September 11.30am 

All support is not equal: Allomaternal support, economic status, and the decision to have a second child in the United Kingdom
Susan B. Schaffnit, Rebecca Sear, Department of Population Health, London School of Hygiene and Tropical Medicine 

Women receive support for raising children from partners, family and non-relatives. The availability of such support may plausibly influence women’s fertility decisions. We test associations between support and women’s second birth decisions in the UK, using Millennium Cohort Study data, paying special attention to socioeconomic position (SEP). The support measures used capture support from different sources (families, partners, non-relatives); and different types of support (practical, emotional). In our study, we stratify our sample by SEP and: (1) describe patterns of allomaternal support; (2) test how the absence of key allomaternal support (partners) affects the availability of other support; (3) test how allomaternal support after a first birth relates to the probability of second birth. Our results demonstrate: (1) high SEP women receive more partner and formal support than low SEP women, low SEP women tend to receive more family support; (2) family and friend support increases in the absence of partners, particularly for low-SEP women; (3) broadly, receiving practical support, regardless of source, associates with a lower probability, while receiving emotional support relates to higher probability, of second birth. The latter associations were largely consistent across SEP, with one exception: childcare from families positively predicts births for low SEP women, but negatively for high SEP women. Results highlight that all support is not equal in the decision to have children, and that, despite differences in the sources of support available to women of different SEP, the influence of support on the likelihood of births is broadly similar across SEP. 

Email: susie.schaffnit@lshtm.ac.uk 

We can’t just ‘Call the Midwife’! Are variations in fertility rates more about ethnicity, country of birth or local geography of residence?
Paul Norman, School of Geography, University of Leeds 

We know that in the UK there are age-specific fertility differences between ethnic groups, that there may be fertility differences by country of birth (especially for recent immigrants), and at a variety of scales there are geographical differences which will be related to concentrations of different population subgroups (not only ethnic groups and recent immigrants, but by social class, students, etc). Age-specific fertility rates for each ethnic group and local government area are needed as inputs for a cohort component projection model of future populations by ethnic group in the UK but our previous estimates of ethnic-geographic specific fertility probably underestimated differences between groups and areas. There is a need then not only to revise these estimates but also to incorporate further dimensions which may be influential including mother’s country of birth. To date, subnational variations in ethnic fertility have been based on scaling local rates for all women from vital registration by national level ethnic indicators from survey data. Newly commissioned vital events data on births by mother’s country of birth will allow geographic variations in fertility to be explored for 2001 to 2011. We can thereby apportion changes in fertility over time which may be attributed to different population subgroups. 

Email: p.d.norman@leeds.ac.uk 

Why does fertility remain high among certain UK-born ethnic minority women?
Hill Kulu, Tina Hannemann, University of Liverpool 

This study investigates fertility among the descendants of immigrants in the UK and examines the causes of high fertility among certain ethnic minority groups. Previous research has shown high total fertility among the UK-born Pakistani and Bangladeshi women, but the reasons for their high fertility have remained far from clear. Some researchers attribute elevated fertility levels among the UK-born ethnic minorities to cultural factors, whereas others argue that high fertility is the consequence of their poor education and labour market prospective. Using data from the first wave of the Understanding Society study and applying multivariate event history analysis this study shows that relatively high second-, third- and possibly also fourth-birth rates are responsible for the high total fertility among women of Pakistani and Bangladeshi origin. However, there is little variation in the first-birth rates among the UK-born women. The fertility differences between ethnic minorities and ‘native’ women slightly decrease once the socio-economic and cultural characteristics, particularly the number of siblings and religiosity, are controlled. Nonetheless, significant fertility differences between those groups persist. Factors related to culture and family background account for some elevated fertility among ethnic minorities in the UK, whereas the role of education and employment seem to be negligible. 

Email: tina.hannemann@liverpool.ac.uk 

Squeezing childbearing into the life course? Childbearing following postponement to later ages in Britain
Ann M. Berrington1, Juliet A. Stone1, Eva Beaujouan2, University of Southampton1, Wittgenstein Centre (IIASA, VID/ÖAW, WU)2 

Mean age at first motherhood in Britain rose from 23.7 to 27.9 years between 1970 and 2011. Reasons why it can be difficult for women starting their childbearing in their mid to late thirties to fully recuperate births following postponement include declining fecundity with age, competing commitments, lack of a suitable partner and cultural deadlines as to the age at which men and women should become parents. Previous research has found that more educated women are more successful in recuperating their fertility at later ages. Explanations for this finding include selection effects, partner characteristics, and the desire to concentrate their childbearing within a shorter span of time. This paper examines the relationship between age at entry into motherhood and subsequent childbearing for British cohorts born between 1940 and 1969. The paper goes beyond existing research in examining not only educational differentials in the overall likelihood of progressing to second and higher order births, but also whether the pace of parity progression has increased, especially among more educated members of recent birth cohorts. Using fertility histories collected within the General Household Surveys and Understanding Society surveys we find that the strong, negative association between age at first birth and subsequent childbearing remained virtually unchanged, suggesting that women in recent cohorts are no more likely to recuperate their fertility. Whilst the pace of childbearing is faster for those who delay entry into motherhood we find only limited evidence of cohort change in these relationships. 

Email: a.berrington@soton.ac.uk