Max Planck Institute for Demographic Research

Populations worldwide are aging, intensifying the challenge of meeting long-term care (LTC) needs. While formal LTC programs aim to address this demand, less is known about how they interact with informal caregiving and affect caregiver health and labor outcomes. This paper exploits the staggered expansion of Medicaid Home and Community-Based Services (HCBS) waivers across U.S. states to examine these effects. Using data from 1998 to 2018 with approximately 20,600 observations, our sample consists of individuals aged 40-70 who are initially healthy and report no activities of daily living (ADL) limitations, allowing us to assess their responses to expanded formal care availability for their parents. We implement a two-way fixed effects difference-in-differences (TWFE DiD) framework to estimate causal effects. We find that HCBS expansions increase the likelihood of providing any informal care by 5 percentage points (14% relative to the mean), personal care by 2 percentage points (20%), and errands by 5 percentage points (14%). These increases are temporary, vanishing within two to three years. Short-run health declines and small labor force reductions follow the same pattern, with no persistent adverse effects. Our findings challenge the view that formal and informal care are solely substitutes. HCBS can complement certain non-intensive caregiving activities without long-term harm, highlighting the need for LTC policies that account for the dynamic relationship between formal services and family care.

Abstract Age-specific fertility rates (ASFRs) provide the most extensive record of reproductive change, but their aggregate nature obscures the individual-level behavioural mechanisms that drive fertility trends. To bridge this micro–macro divide, we introduce a simulation-based Bayesian framework that couples a demographically interpretable, individual-level simulation model of the reproductive process with sequential neural posterior estimation (SNPE). We show that this framework successfully recovers core behavioural parameters governing contemporary fertility, including preferences for family size, reproductive timing, and contraceptive failure, using only ASFRs. The framework’s effectiveness is validated on cohorts from four countries with diverse fertility regimes. Most compellingly, the model, estimated solely on aggregate data, successfully predicts out-of-sample distributions of individual-level outcomes, including age at first sex, desired family size, and birth intervals. Because our framework yields complete synthetic life histories, it significantly reduces the data requirements for building microsimulation models and enables behaviourally explicit demographic forecasts.

This study explores the relationships between migration, gender, and memory trajectories over the life-course within an intersectional framework. Memory is an important dimension of cognitive decline, a critical concern in aging populations, and varies significantly across different demographic groups. In this paper we address the understudied influence that the intersection of gender and migration status has on memory trajectories and evaluate the role of age at migration. Using random-effects growth curve models, our findings reveal persistent memory disparities by migration status, particularly in early older age, though differentials by migration status generally reduce after age 70. Immigrants, overall, exhibit lower baseline memory functioning levels and steeper declines. Contrary to expectations, immigrant women are not uniquely vulnerable; instead, gender differences persist across groups, with men generally faring worse. Later age at migration predicts poorer memory performance, though a late-life advantage emerges for men who migrated at ages 6-17 warranting further investigation. These results underscore the importance of considering baseline levels, longitudinal patterns, and intersectional factors in cognitive aging research by migration status.

It is well-established that deaths peak in winter and show throughs in summer. However, it remains unclear how mortality patterns will unfold as the climate warms, bringing fewer cold days and more hot days. One concern is “harvesting,” where a short-term surge in deaths among the most vulnerable people is then followed by a period with fewer deaths than usual because those individuals would have died soon anyway. Under global warming, it is possible that higher mortality rates in summer will result not only from an increase in extreme heat events but also from a seasonal shift in excess deaths that would have previously occurred in winter. Combining mortality data from the Italian Statistical Office with temperature data from the Copernicus Data Store for Italy at the provincial level from 2011 to 2019, we employ Poisson regression models to estimate the effects of temperature extremes on mortality among individuals aged 60 and above. The results reveal that temperatures outside the comfort zone, both lower and higher, are associated with increased monthly mortality rates, with the strongest effects seen in the most extreme temperature ranges. We find evidence of a harvesting effect, particularly for moderately warm days (≥ 85th to < 95th percentile). However, even after high winter mortality, extremely hot days still lead to significant increases in deaths—especially among individuals aged 80 and above. This suggests that while some short-term mortality displacement occurs, it is not enough to offset the full impact of extreme heat, highlighting the continued vulnerability of older populations.

The risks of poverty entry vary considerably across European countries, and within them by educational group. To what extent are these differences related to the association between health condition onset and poverty entry? This study analyses the associations and population contribution of the onset of limiting long-standing health conditions on the short-term risk of entry into objective and subjective poverty in 24 European countries. A series of logistic regression models fitted to the weighted EU-SILC data show that, net of key covariates measured the year before, the link between the onset of ill health and income poverty entry is weak, but the association between ill health and the risk of subjective poverty entry is substantial, particularly in Eastern European countries. Nevertheless, at the population level, only a small proportion of all poverty entries could be potentially attributed to this association. Similarly, poverty entry differences across educational groups could not be attributed to the differences in health conditions and their associations with poverty entries. Countries with strong associations between health and poverty entries, such as Greece and Latvia, should develop their social insurance systems to both buffer individuals from the resources they lose due to health conditions and compensate them for their increasing needs due to these conditions. • Health explained only a small share of the variation in poverty entries • The link between health conditions and objective poverty entry was small. • The association between health and subjective poverty was substantial.

Spousal caregiving is linked to increased symptoms of poor mental health among older people. Less is known about the effects of physical health conditions with different progression and level of impairment on spousal mental health among younger people. Using Finnish total population register data, we identified 106 673 partnered individuals aged 30-70, newly diagnosed in 2000-2015 with progressive conditions with increasing impairment (Parkinson's disease, multiple sclerosis, dementia), acute conditions with sudden impairment (stroke) and acute conditions without cognitive impairment (myocardial infarction, prostate cancer, female breast cancer), and employed event-study difference-in-difference models to assess short-term and long-term effects on spouses' 6-month probability of psychotropic medication purchases (Anatomical Therapeutic Chemical (ATC) codes N05B, N05C and N06A) by type of health condition. Incident stroke had a sudden effect on spousal psychotropic medication purchase probability (2.1 percentage points increase among female and 1.4 percentage points increase among male spouses). Effect size attenuated thereafter, although more slowly among female spouses, yet remained over 1 percentage point in the long term. Smaller (about 0.7 percentage points among female and 0.5 percentage points among male spouses), but also persistent, effect was observed following the diagnosis of acute conditions without cognitive impairments. For progressive conditions, little changes occurred around the time of diagnosis. Acute onset of physical illness may have sudden and persistent adverse impact on spousal mental health in midlife even when prognosis is good and expectations for spousal care are likely to be limited.

Socioeconomic inequalities in mortality are large and persistent. While the differential timing and impact of the smoking, alcohol, and obesity epidemics among socioeconomic groups likely influenced past trends in socioeconomic mortality inequalities, the evidence is scarce. We estimated the combined impact of smoking, alcohol, and obesity on past trends in educational inequalities in remaining life expectancy at age 30 (e30) in England and Wales, Finland, and Italy (Turin). To do so, we used long-term timeseries of annual individually-linked mortality data by educational level (low, middle, high), sex, and age (30+). We multiplicatively aggregated estimates of smoking-, alcohol-, and obesity-attributable mortality by educational level to obtain "lifestyle-attributable mortality" (LAM) by educational level. We compared trends in educational inequalities in e30 with and without LAM using segmented regression. We found that smoking-, alcohol-, and obesity-attributable mortality individually contributed 23%, 14%, and 10%, respectively, to the average educational inequality in e30 of 4.4 years in 1992-2017, and 44% combined (males: 51%; females: 34%). LAM contributed 57%, 63%, and 43%, respectively, to the increase in educational inequalities in e30 among Finnish males (1987-2008), Finnish females (1987-2017), and Italian males (1990-2018); tempered the decline in inequalities among British females (1992-2017); and was responsible for the reversal in 2008 from increasing to declining inequalities among Finnish males. Targeting socioeconomic inequalities in smoking, alcohol, and obesity could, thus, substantially reduce socioeconomic inequalities in e30, and the increasing time trends in these inequalities. The observed country differences in the importance of these lifestyle factors demonstrate the need for context-specific strategies.

Seasonal mortality fluctuations significantly affect national life expectancy, yet their role in regional inequalities remains underexplored. Understanding this is crucial for targeted health policies aimed at reducing spatial mortality differences. We quantify the impact of seasonal excess mortality on regional life expectancy levels (e0) and inequalities in Italy. Using monthly mortality data from the Italian National Statistical Institute by region (N = 20), sex and age between 2005 and 2019, we assessed e0 losses due to seasonality by using the general approach of comparing observed e0 with minimum achievable levels. Seasonal effects on inequalities in e0 were quantified by comparing standard deviations with and without excess seasonal mortality. We then analysed regional contributions using decomposition analysis. Seasonal excess mortality led to an average loss in e0 of 1.4 years (1.36–1.70) across Italian regions. This effect was most pronounced in southern and insular regions (Campania and Sicilia), especially for winter-related excess mortality. Our analysis showed that spatial inequality in e0 would have been reduced on average by 11.2% by eliminating excess seasonal mortality, mostly due to excess winter mortality (−7.5%). However, during the years with high mortality burdens (2005 and 2015), seasonality contributed to spatial mortality inequality by 10% and 5.2%, respectively. The regional contributions to e0 inequality revealed that Campania and Sicilia also had the strongest role in increasing spatial mortality variation throughout the period. The pronounced regional inequalities in e0 losses due to seasonal excess mortality contributed significantly to shaping regional e0 variation in Italy, mostly due to different impacts of winter mortality within the countrySupplementary InformationThe online version contains supplementary material available at 10.1007/s10680-025-09753-7.

Abstract This study investigates the cost of job loss to household incomes, and the extent to which initial earnings losses are compensated through the labor market, within the household, and by the social security programs. Using survey and administrative data from Denmark, Finland, Germany, and the UK (1990–2018), we estimate short- and long-term effects of job loss with a dynamic difference-in-differences model. Job loss reduces household income by 17 per cent in the UK while only 5 to 6 percent in other countries during the first year. These losses gradually diminish and disappear over the long run. Across all countries, market (i.e. re-employment) is the main source of compensation, while the role of household and state compensations varies in line with the national compensation strategies. State compensation is crucial in mitigating immediate income losses, while market compensation becomes even more important over time. Household compensation mainly substitutes for weaker market and state protections.

Under-five mortality in sub-Saharan Africa has declined markedly, yet ethnic gaps persist and may even widen despite overall gains. This study uses two consecutive rounds of Demographic and Health Surveys (2007–2022) from twelve countries to trace how child characteristics, maternal education, household wealth, and urban-rural residence drive ethnic differentials in child survival. Country-specific total births ranged from 23,109 to 127,545 per survey round. Employing survey-weighted quasi-Poisson models with person-year offsets and a sequential Shapley-value decomposition, the study estimates crude and adjusted rate ratios for pairwise ethnic comparisons. It predicts mortality rates under counterfactual socioeconomic conditions. Results show enduring disparities, for example, among the Luo versus the Kalenjin in Kenya (adjusted RR 2.15, 95 % CI 1.83–2.51) and the Hausa versus the Yoruba in Nigeria (adjusted RR 1.39, 95 % CI 1.22–1.56), with maternal education accounting for roughly one-quarter to one-third of inequality. At the same time, the role of household wealth varies by context. Sensitivity analyses examining administrative boundary effects revealed that geographic mediation accounts for 15.4 % of ethnic disparities in Ghana's centralized system and 35.1 % in Nigeria's decentralized federal structure, suggesting that apparent ethnic disparities substantially reflect differential healthcare access across administrative units rather than intrinsic cultural factors. In Ghana, targeted policies combining fee removal and culturally tailored maternal care have noticeably narrowed ethnic gaps. These findings underscore that eliminating preventable child deaths by 2030 requires interventions that address both the socioeconomic and cultural determinants of ethnic disadvantage.