BACKGROUND: Much attention has been paid recently to the large decline in period fertility in many countries since 2010. In the Nordic countries and England and Wales much of it can be attributed to the decline in first-birth rates, suggesting higher childlessness. OBJECTIVE: The aim of this paper is to quantify the contribution of childlessness to changes in cohort completed fertility between the 1960 and 1979 cohorts, for sixteen countries from Europe, East Asia, and North America. METHODS: Using Kitagawa exact decomposition, I decompose the changes in cohort completed fertility between two components: the proportion of women who remained childless up to the age of 45 (‘nulliparous’ women) and the average number of children born to women who became mothers by age 45 (‘parous’ women). The data primarily come from the Human Fertility Database. RESULTS: The contribution of childlessness to the change in fertility of the 1960–1979 cohorts is varied. In many countries a rise in childlessness made a large contribution to the decline in their cohort fertility (e.g., Japan, South Korea, Spain, and Poland), but in others its contribution is small. CONCLUSIONS: Increases in childlessness played a larger relative role in cohort fertility decline in countries which experienced large declines. CONTRIBUTION: This paper decomposes the changes in cohort fertility into two components, using a simple exact decomposition and more-recent birth cohorts than in Zeman et al. (2018), and ascertains conditions where the childlessness component was large.
BACKGROUND: Period lifespan variability is sometimes interpreted as an indicator of lifespan uncertainty for individual cohort members. Is this an admissible interpretation for living cohort members, given that future death rates are not precisely known? OBJECTIVE: To answer this question, we express individual lifespan uncertainty as a function of variation in ages at death in a cohort lifetable and uncertainty about future death rates. We then quantify the importance of each component to total individual lifespan variability of a birth cohort. A small contribution of the forecast variance would support the interpretation of lifespan variability as individual lifespan uncertainty. METHODS: Using Human Mortality Database data for seven European countries, we decompose the total variance in the remaining lifespan at birth, age 40, and age 80 of an individual cohort member into cohort lifetable variance and forecast variance for cohorts born 1871–2019. We do so under Lee–Carter cohort forecast variance and under an empirical forecast variance measure which allows for mortality shocks. RESULTS: Even when accounting for the possibility of future mortality shocks, the forecast variance component tends to contribute less than 10% to the total lifespan variance of an individual. Exceptions are France and the small and thus volatile population of Iceland. Conditioning on the absence of future mortality shocks, the forecast variance contributes around 1% to total lifespan variance. CONTRIBUTION: We give credibility to the interpretation of period lifetable lifespan variance as individual lifespan uncertainty by showing that the lifespan variance of a cohort is mainly determined by expected lifetable variance with smaller contributions from forecast variance.
BACKGROUND: China’s continuous decline in births has attracted scholarly and policy attention. However, total fertility rates are affected by tempo effects and may not accurately reflect the underlying fertility level. OBJECTIVE: We examine cohort trends in childlessness and the number of children ever born (CEB) among Chinese women, focusing on fertility postponement, recuperation, and the effects of structural changes. METHODS: Using nationally representative data, we calculate age-specific childlessness proportions and the number of CEB by cohort. We analyze fertility postponement and recuperation, and employ decomposition techniques to assess how age-parity-specific progression ratios, urban–rural composition, and educational structure contribute to cohort fertility differences. RESULTS: Childlessness has increased across cohorts, while CEB has declined. The rise in childlessness is stronger among urban and more educated women, whereas the decline in CEB is greatest among rural and less educated women. Fertility postponement has intensified, while recuperation has weakened. Decomposition shows that increased childlessness is primarily driven by reduced first-birth progression at younger reproductive ages, while the decline in CEB is largely attributed to shifts in women’s urban–rural and educational composition. CONCLUSIONS: Childlessness has risen and fertility levels have declined, with differences across urban–rural and educational groups. These patterns highlight the need to consider social and structural factors in policies aimed at supporting fertility. CONTRIBUTION: This study provides a systematic cohort-based analysis of childlessness and fertility trends in China, providing valuable empirical evidence on cohort fertility.
BACKGROUND: As more adults assume caregiving responsibilities for their aging parents, a key question emerges: Does coresidence increase financial strain for adult child caregivers? OBJECTIVE: To examine whether coresiding with an oldest-old parent (aged 80+) is associated with higher financial strain among adult child caregivers. METHODS: This study uses data from the Hispanic Established Population for the Epidemiologic Study of the Elderly (2010–2011), a population-based survey of Mexican Americans in five southwestern US states (N = 659). Logistic regression and propensity score–matched logistic models are used to examine the association between coresidence and caregiver financial strain, accounting for selection into coresidence. RESULTS: Coresiding caregivers had more than twice the odds of experiencing financial strain compared to non-coresiding caregivers (OR = 2.17 [95% CI: 1.22–3.82], p = 0.007). This relationship remained robust in the matched sample (OR = 2.02 [95% CI: 1.06–4.02], p = 0.038). Poor parental health independently increased the odds of caregiver financial strain in both the full (OR = 3.63 [95% CI: 1.38–12.5], p = 0.019) and the matched sample (OR = 2.20 [95% CI: 1.01–5.36], p = 0.061). CONCLUSIONS: These findings suggest that coresidence often functions less as a financial buffer and more as a conduit for high-intensity care, thereby elevating caregiver financial strain. Policies aimed at promoting caregiver financial autonomy – including tax credits, paid family leave, and expanded home- and community-based services – should take into account the distinct vulnerabilities of coresiding caregivers. CONTRIBUTION: This study provides evidence that coresiding with an oldest-old parent more than doubles the odds of financial strain among Mexican American adult child caregivers, challenging the assumption that intergenerational coresidence functions primarily as an economic buffer for low-income families. The findings inform policies designed to support the growing population of families navigating population aging, long-term care infrastructure, and housing affordability constraints.
BACKGROUND: The number of centenarians worldwide has increased considerably since 1950. However, most demographic research on population longevity has focused on high-income countries. As a result, little is known about centenarians in low- and middle-income countries, such as Mexico. OBJECTIVE: This study assesses the quality of demographic data sources on centenarians in Mexico over time, estimates their population size in 2010, and analyzes their geographic distribution. METHODS: We used individual-level data from the census and vital statistics records from 1990 to 2023. Using the extinct cohort method, we estimated the population of centenarians alive in the 2010 census. To examine spatial distribution, we computed centenarian rates by state. RESULTS: Our findings suggest an increase in the number of deaths occurring at ages 100 and over after 2000. We estimate that in 2010 there were 6,300 female and 3,600 male centenarians in Mexico, which is half the number reported in the census. The female-to-male ratio is lower than that in low-mortality countries. Southern states show higher centenarian rates than northern states, which could be associated with differences in age reporting. CONCLUSIONS: The number of centenarians has increased in recent decades, with marked differences in longevity and data quality across the country. CONTRIBUTION: We provide evidence of the increase in the number of centenarians and improvements in data quality for their study in a middle-income country such as Mexico. In contrast to high-income countries, we found a lower female-to-male ratio, highlighting the need to expand research on population longevity and age reporting worldwide to better understand their heterogeneity.
BACKGROUND: Family migration shapes early childhood development (ECD), but the link between parental voluntary versus forced migration experiences and ECD remains underexplored. Germany, with its diverse migration streams, offers an opportunity to investigate this interrelation. OBJECTIVE: This study examines the link between parental migration status and ECD of children aged 2–4 and investigates whether the association between parental mental health and ECD differs by migration background. METHODS: Using the German Socio-Economic Panel (2004–2021; N = 1,826), we fit multivariate linear regression models, comparing ECD between children of voluntary and forced migrants. ECD is measured using an adapted version of the Vineland Adaptive Behaviour Scales. Further models account for maternal and paternal mental health scores, using the Mental Component Summary (MCS). RESULTS: Children of forced migrants showed lower early childhood development across all domains compared to children of voluntary migrants, with the largest gaps observed in gross motor skills (β = −0.354, 95% CI: −0.472, −0.237) and social relations (β = −0.239, 95% CI: −0.356, −0.122). Adjustment for parental mental health attenuated these differences but did not eliminate them. Interaction analyses provided no evidence that the association between parental mental health and ECD differed by migration background. CONCLUSIONS: Developmental disparities associated with a forced migration background emerge early and persist beyond parental mental health, indicating that individual family resources do not fully explain these inequalities. CONTRIBUTION: This study demonstrates that early developmental inequalities among children of forced migrants are not primarily driven by parental mental health, highlighting the importance of broader structural conditions.
BACKGROUND: The fifth United Nations sustainable development goal calls for gender equality and the empowerment of all women and girls. However, progress toward this goal remains difficult to assess, as empowerment is multidimensional, context-specific, and challenging to quantify. OBJECTIVE: This study develops a framework to examine how women’s empowerment manifests across intersecting domains and sociocultural contexts in sub-Saharan Africa (SSA). METHODS: Using the Demographic and Health Survey, with data from 303,717 women in 34 SSA countries, we employ latent class analysis to identify distinct empowerment typologies by leveraging 23 survey indicators of agency across multiple dimensions and contexts. We then evaluate how well a smaller set of commonly applied survey items and country-level indicators of empowerment accurately predict the identified typologies. RESULTS: Women who report taking part in household decisions also report a lower acceptance of intimate partner violence, but rejecting such violence does not always align with decision-making power. Education and employment are not consistently associated with greater domestic agency, and many women face healthcare barriers regardless of socioeconomic status. Commonly applied survey items of empowerment are able to predict only women in the less empowered typologies. CONCLUSIONS: These findings highlight the complex expression of women’s empowerment and illustrate how it unfolds unevenly across public and private domains, as well as across diverse sociocultural contexts. CONTRIBUTION: This study provides a framework to improve the monitoring of gender equity and emphasizes the need for a holistic approach in development research while highlighting the challenge of defining what constitutes an empowered woman alongside rapid socioeconomic change.
BACKGROUND: China has witnessed a continuous rise of couple migration and family migration. These migration developments possibly signal individuals’ changing considerations of marital and parenthood responsibilities in their migration decisions over time. OBJECTIVE: We focus on one specific type of migration, first labor migration, and examine the gender and birth cohort patterns of how marriage and family status are associated with first labor migration among rural-origin Chinese individuals born between 1951 and 2000. METHODS: Data are drawn from the 2018 round of the China Household Income Project. We calculate the median values for ages at first labor migration, first marriage, and first childbearing and use discrete-time logistics models to examine the gender and birth cohort differences in how first labor migration is related to marriage and family status. RESULTS: First labor migration is negatively related to marriage and family status for both genders. The pattern of parents’ lower migration risk is more marked among mothers than among fathers. Over birth cohorts, a lower risk of first labor migration is observed among both fathers and mothers. A cohort trend toward lower risk of first labor migration is observed for married individuals but among women only. CONTRIBUTION: Using a precise definition of migration and a sample not conditional on migration outcomes, we have demonstrated Chinese individuals’ changing patterns of life course events across birth cohorts and showed that the negative relationships between marital as well as childbearing responsibilities and first labor migration have grown stronger over birth cohorts, especially among women.
BACKGROUND High-quality population data are expensive to collect and limited in substantive scope, constraining timely and innovative research. OBJECTIVE We evaluate nonprobability data from the KINMATRIX survey by comparing it with external data on kinship structure and relationships to kin. METHODS KINMATRIX is a web-based quota survey of over 12,000 younger adults (aged 25-35) in 9 European countries and the United States. Respondents reported egocentric kinship networks, yielding more than 250,000 ego-kin dyads. We compare indicators in three domains-family network size (number of living kin), family relationships (emotional closeness, geographical distance, coresidence, and contact frequency), and family complexity (parental separation and number of half-siblings)-with estimates from probability-based surveys, demographic projections based on official vital statistics, and national registers. RESULTS KINMATRIX estimates of family network size tend to be lower than demographic projections based on official data, with smaller gaps for aunts and uncles and larger gaps for siblings and cousins. Cross-national differences in the number of parents, siblings, aunts, uncles, and cousins are broadly consistent across data sources, whereas patterns differ for grandparents. Regarding family relationships, the correspondence between KINMATRIX and probability surveys varies across indicators, but established north-south contrasts are generally preserved. Regarding family complexity, KINMATRIX yields lower rates of parental separation than probability-based data, but it broadly captures cross-national differences. CONTRIBUTION Our comparisons point to nonprobability data having both promise and pitfalls in family demography, highlighting the need for study-specific benchmarking and targeted validation.
OBJECTIVE: To provide an in-depth examination of the birth squeeze, the alteration of male and female birth rates by the age-sex composition of the population. METHODS: Mathematical analyses using reasonable simplifying assumptions are made to find male and female total fertility rates (TFRs), and from them an index of the magnitude of the birth squeeze under exponential, linear, and sinusoidal birth trajectories. RESULTS: Explicit equations are shown for male and female TFRs and for birth squeeze measures under the three birth trajectories. In all cases, the magnitude of the birth squeeze was approximately proportional to (1) a meaningful measure of population change, (2) the difference between the male and female mean ages of fertility, and (3) the sex ratio at birth. Despite the strong influence of cycle length T in sinusoidal models, the cycle time of the maximum birth squeeze is consistently found near time 30, and the minimum near time 30 + T/2. CONTRIBUTION: Birth squeezes are a potentially important, though rarely considered, factor in fertility analyses. Here, easy-to-apply methods are presented that show how male and female fertility rates are affected by population composition, and that measure the magnitude of birth squeezes. New, interpretable regularities are derived that relate birth squeezes to meaningful demographic measures and provide new insights into fertility dynamics.
BACKGROUND Life expectancy in China has been increasing in recent decades, but regional inequalities persist. Unbalanced regional development has resulted in large-scale population movement. Over time, the changing regional distribution of the population changes the proportion of the population exposed to different regional mortality levels. OBJECTIVE The increase in national life expectancy can be attributed to changes in both regional mortality and the regional composition of the population. We aim to quantify the respective contributions of changes in mortality and in the composition of the population. METHODS We apply a decomposition method to data from the Chinese Disease Surveillance Points system from 2010 to 2020 and disaggregate the improvement in Chinese life expectancy into mortality and composition components. RESULTS Improvements in mortality were observed across all regions, positively contributing to the increase in national life expectancy. However, only the most and least economically developed regions-urban-east, urban-central, and rural-west-made positive contributions through the compositional component. All other regions contributed negatively, partially offsetting the progress made by the improvement in regional mortality. CONTRIBUTION Our work is the first to decompose the increase in Chinese national life expectancy into its regional contributing components of changing mortality and population exposure. This disaggregation of the different components at the regional level provides insights as to how to better address regional health inequality issues in China.
BACKGROUND The postponement of first childbirth is a well-established trend across many high-income countries. While delayed childbearing often reduces the likelihood of having additional children, some women compensate by shortening the interval before a second birth. However, the capacity to recuperate fertility after a late start varies, particularly according to educational attainment, and the underlying mechanisms remain unclear. OBJECTIVES This study examines how partnership trajectories shape the timing and likelihood of first and second births, with a focus on educational differences. It explores whether stable partnerships are associated with earlier childbearing and whether they facilitate fertility recuperation after delayed motherhood, with particular attention to educational differences. METHODS Using data from the 2018 Spanish Fertility Survey, event-history analysis is applied to a subsample of 11,813 women to estimate the timing and likelihood of first and second births. RESULTS Women with more than one coresidential partner over their life course are less likely to have a first child and tend to postpone motherhood. However, they also experience shorter intervals between the first and second child, suggesting fertility recuperation. This pattern is more evident among university-educated women with multiple partnerships. CONTRIBUTION This study contributes to understanding fertility recuperation by linking partnership trajectories to fertility outcomes and highlighting the moderating role of education. The findings also provide new evidence from Spain, where very late fertility patterns may constrain fertility recuperation due to biological limits.
BACKGROUND The COVID-19 pandemic and related mitigation policies potentially were related to romantic and sexual partnering among straight and queer college students in 2020-2024. OBJECTIVE We examine percentages of those hooking up, dating, and forming long-term romantic relationships in college during 2019-2024; the number of intimate events and relationships; partner meeting contexts; and how patterns varied by gender and sexual orientation. METHODS We present results from surveys of 11,993 college students collected at 34 colleges between fall 2019 and spring 2024. We use random effects regression models to examine how partnering and meeting contexts changed among students surveyed each academic year. We examine differences separately by gender and sexual orientation. RESULTS Percentages of those hooking up remained relatively flat, while dating and long-term relationship formation rose temporarily over the period examined. Differences could be attributed to the changing age composition of students, with older students more common after 2019-2020. After standardization by age, percentages of those hooking up first declined and then recovered, but not fully. Dating percentages were flat before declining in the last year examined, and relationship formation was also flat with a brief rise in 2022-2023. Students met partners online more commonly during the height of the pandemic. CONCLUSIONS The COVID-19 pandemic was related to partnering in college in part because average age increased, partners increasingly met online, and hooking up declined. CONTRIBUTION We analyze a new large dataset to examine trends in sexual and romantic partnering on campus, providing new insights into college intimacy both before and during the pandemic.
BACKGROUND Previous research shows that ethnic minorities in Britain are disproportionately affected by housing insecurity. However, studies rarely distinguish between migrant generations, making it difficult to assess intergenerational life course disadvantage. OBJECTIVE The objective of this study is to examine the long-term residential mobility and housing trends among migrant generations across different country-of-origin backgrounds. By so doing, we aim to gain further understanding of housing outcomes and evaluate the residential assimilation and stratification theories. METHODS We use the Office for National Statistics Longitudinal Study for England and Wales, a large census-based dataset. We use pooled binary and multinomial logistic regression methods to analyse residential and housing changes by origin group and migrant generation from 1971 to 2011. RESULTS First, we find that for most groups, the 1.5 and second generations have a higher likelihood of moving than the first generation. Second, the likelihood of moving to homeownership increases across migrant generations; the descendants of Indian and Pakistani immigrants are even more likely to move to homeownership than the native population. Third, we find that residential mobility decreases over time for nearly all groups. CONCLUSIONS Overall, our findings support both the assimilation and stratification theories, with intergenerational socioeconomic support and housing market instability being potential factors. Groups which, on average, have higher socioeconomic status and accumulated wealth are more likely to move up the housing ladder. However, for groups which are, on average, more socioeconomically disadvantaged, wealth alone does not explain housing tenure differences, and these differences are not as pronounced as the literature suggests.
BACKGROUND Measuring sex and gender is central to demographic research and studies of social inequality. Best-practice guidelines for surveying English-speaking adults recommend a two-step approach, which uses two items to distinguish sex at birth from current identity. Additional research is needed to understand whether specific formatting decisions affect response distributions or measures of data quality. OBJECTIVE We explore three understudied formatting choices: (1) Should sex or gender terms be used for current identity responses, (2) should both items appear on the same page or different pages in online surveys, and (3) in which order should the response options be presented? By answering these questions, we provide practical guidance for researchers applying the two-step measure, and evidence on how robust the measure is to various formatting differences. METHODS We draw on three survey experiments conducted on two US adult samples: a sexual and gender minority (SGM) sample (n = 2,402) and a general population sample (n = 1,491). CONCLUSIONS Our results suggest researchers using the two-step approach can improve data quality by (1) using gender terms for current identity, and (2) using a page break in online surveys. Varying the response option order does not have measurable effects on data quality in our samples, which frees researchers in most survey contexts to consider nontraditional response orders that have other advantages.
BACKGROUND Economically insecure conditions may be driving young people's decisions to delay, forgo, or reduce childbearing. Yet establishing a causal link between economic resources and fertility is methodologically fraught. One approach has been to use survey experiments to manipulate hypothetical economic conditions and gauge respondents' fertility preferences. OBJECTIVE Unlike survey experiments that manipulate economic circumstances and then ask about fertility preferences, our experiment varies both hypothetical economic circumstances and hypothetical parity to test the hypothesis that US women will prefer family scenarios with more (compared to fewer) children under conditions of high economic resources and with fewer (compared to more) children under conditions of low economic resources. METHODS In an online conjoint survey experiment with a nationally representative sample of US women aged 18-35 (n = 1,794), respondents chose between different family scenarios comprised of randomly varied attributes, including parity and economic resources. We examine how these two attributes interact to predict preferred family scenarios. RESULTS There was no evidence that respondents preferred family scenarios with more (compared to fewer) children under conditions of high economic resources or preferred family scenarios with fewer (compared to more) children under conditions of low economic resources. CONTRIBUTIONS We provide new experimental evidence about whether economic resources and parity interact to predict preferred family scenarios. Our findings are consistent with scholarship suggesting that although economic factors shape decisions about family life, the rational-actor framework is insufficient for understanding contemporary US fertility declines.
BACKGROUND Much debate exists about how women's employment interacts with divorce and separation, with conflicting findings and opposing theoretical interpretations. OBJECTIVE This study revisits the employment-separation link from a dynamic perspective, examining how employment influences separation risk (independence), whether women increase employment before separation (anticipation), and how they adjust work hours afterward (adjustment). METHODS Data come from the Comparative Panel File (harmonised panel data) from six countries (Australia, Germany, Russia, South Korea, Switzerland, and the United Kingdom), covering the 1990-2021 period, 73,213 married or cohabiting women, and 9,847 separations. Discrete-time event-history models estimate separation risks, and longitudinal growth models track employment changes before and after separation. RESULTS Across all countries, employed women-especially women employed full-time-face a higher separation risk than non-employed women. These associations persist after controlling for potential confounders such as household income, health, life satisfaction, and religiosity. Part-time work shows weaker but still positive associations with separation. Employment trajectories indicate limited changes around separation, though modest pre-separation increases in work engagement appear in some contexts. CONCLUSIONS The study provides evidence for the independence hypothesis regarding the influence of women's employment on separation across diverse societal contexts, with more mixed support for anticipation and adjustment effects.
BACKGROUND: Parenting leave policies shape how caregiving and paid work can be reconciled around the time of childbirth. They have important implications for fertility, employment, and gender equality. Still, there are limited quantitative cross-country data capturing long-term policy changes that impact how long parents can temporarily be away from work to care for their children, and how leave can be shared between them. OBJECTIVE: The European Parenting Leave Policies (EPLP) Dataset provides harmonised regulations on maternity, co-parent, paid parental, and job-protected leave across 21 European countries from 1970 to 2024. It focuses on policies that shape how long birth mothers and their co-parents can take leave. METHODS: Statutory leave entitlements were compiled from national legal sources, official government publications, and secondary literature. We followed a consistent set of data collection rules to enable comparison across countries and over time. Because the dataset focuses on time away from the job, it considers only rights for employed parents. It includes 33 variables and also documents country-specific reform timelines. CONTRIBUTION: The EPLP Dataset fills a gap in existing data sources by providing quantitative data across 55 years on policies that shape how long parents stay at home around birth and how leave is shared between them. In addition to leave duration and benefits, it covers recent policy instruments such as incentives for parents to share leave, and timing and flexibility of leave use. The dataset enables cross-national comparisons and the analysis of changes over time, and can be used to study the effects of policy reforms.
BACKGROUND A stationary population may offer advantages for societal, economic, and ecological sustainability. It requires that the number of births aligns with expected deaths under given mortality patterns. However, reproductive rates such as the TFR are not designed to assess whether the number of births observed is sufficient to replace expected deaths, necessitating direct assessment of birth-death balance for population sustainability. OBJECTIVE We aim to assess whether current fertility levels are sufficient to replace expected deaths using a new demographic indicator, the ratio of births observed to births needed (Bo/Bn). METHODS We analyze global and regional demographic sustainability from 1950 to 2100 using data from the World Population Prospects 2024. The Bo/Bn ratio is derived from stationary population theory, comparing actual births to those required under given mortality conditions. RESULTS Globally, births observed exceeded those needed by 18% in 2025, but this surplus is declining. According to the United Nations medium variant projection, by 2060, global births will be insufficient to replace expected deaths for the first time in modern history. By 2100, births are projected to fall 12% short of the level needed. Regional variation is substantial: Middle and Eastern Africa maintain Bo/Bn ratios above 2.0, while Eastern Asia exhibits a critically low ratio of 0.49. Currently, 43% of the global population live in countries with insufficient births, and this proportion will reach 75% by 2100. CONTRIBUTION The Bo/Bn ratio provides a direct measure of demographic sustainability that complements traditional fertility indicators, offering crucial insights for policy planning in an era of global demographic transition.
BACKGROUND Fertility in Spain has declined and shifted to later ages alongside diversification of family formation, raising questions about how partnership life courses shape completed fertility by late reproductive age. OBJECTIVE We aim to assess how key features of partnership trajectories-timing, duration, and sequencing-relate to motherhood and parity by age 41 among women in contemporary Spain, and whether associations differ by education. METHODS We analyze 7,656 women, born 1962-1977, in the Spanish Fertility Survey (2018). A feature-selection approach identifies salient aspects of union histories (e.g., postponement or absence of union, stability, cohabitation, higher-order unions) associated with (1) entry into motherhood and (2) progression to higher-order births by age 41, with results stratified by educational attainment. RESULTS Delayed or foregone union formation during the reproductive years is strongly associated with lower fertility, particularly a lower likelihood of entering motherhood. Stable marriage is consistently linked to higher probabilities of both becoming a mother and progressing to higher-order births. Cohabitation and post-separation trajectories show more nuanced and generally weaker associations with fertility in these cohorts. Associations between partnership status at later reproductive ages and fertility outcomes are more pronounced among highly educated women. CONCLUSIONS Partnership life courses are central to understanding low and late fertility in Spain, and our results shed light on how traditional and emerging partnership dynamics jointly shape fertility outcomes by late reproductive ages. CONTRIBUTION We move beyond a narrow event-based focus to provide an integrated, trajectory-based account of how partnership timing, duration, and sequencing relate to motherhood and parity in a sustained low-fertility setting, documenting educational differences in these associations.
