The notion of “lowest-low fertility” emerged in the early 2000s to describe an unprecedented demographic development: total fertility rates (TFR) falling to 1.3 children per woman or below in a small group of European countries (Kohler et al. 2002). At the time, the concept provided a useful diagnostic tool, signaling a historically unique demographic trajectory, with potential implications for population ageing, labour force dynamics, and longer-term demographic sustainability.

Two decades later, the empirical landscape has changed substantially. Fertility levels below 1.3, no longer confined to a handful of countries, have spread across diverse regions, including East Asia, Southern and Eastern Europe, and parts of Latin America. As a result, a growing share of the world’s population now lives under what would be classified as “lowest-low fertility.”

At the same time, the analytical usefulness of the concept is further undermined by the widening range of fertility levels it now includes. Originally associated with a relatively narrow band below 1.3 children per woman, the category today spans TFRs from around 0.7 to 1.3. From a demographic perspective, this is a considerable range: differences within it translate into markedly distinct long-term population outcomes, particularly when compounded over time. 

As the category has expanded geographically and substantively, its capacity to distinguish between demographic regimes has weakened. What was once a relatively coherent grouping now encompasses countries characterized by diverse demographic dynamics and varying levels of socioeconomic development and future prospects.

Diverging demographic trajectories

A limitation of the concept lies in its implicit assumption that similar fertility levels imply broadly comparable demographic futures. Increasingly, this assumption does not hold. Countries with TFRs below 1.3 display substantial variation in population trajectories, reflecting differences in migration, mortality, and age structure. For instance, while some countries with very low fertility face sustained population decline and rapid ageing, others continue to grow due to positive net migration. These differences are not marginal: they fundamentally shape the scale and pace of demographic change, as in the cases of Canada and South Korea, both “lowest-low” fertility populations.

Moreover, variation within fertility timing itself also matters (Sobotka 2017). In several Latin American contexts, recent declines are driven in part by the postponement of childbearing, particularly among younger cohorts, rather than by a uniform reduction in completed fertility. Such dynamics may imply different long-term outcomes than those observed in contexts where low fertility is more deeply entrenched across cohorts and the postponement transition is complete, providing less margin for recuperation.

At the same time, the persistence of the 1.3 threshold in public and policy discourse has contributed to a simplified narrative in which crossing a specific numerical boundary is equated with entering a zone of demographic crisis. This framing may inadvertently reinforce alarmist interpretations and direct attention toward policies aimed at increasing fertility, despite mixed evidence regarding their effectiveness.

Towards a multidimensional perspective

Reassessing the relevance of “lowest-low fertility” does not imply that fertility has become unimportant. Rather, it highlights the need to situate fertility within a broader analytical framework.

Population change reflects the interaction of multiple components—fertility, mortality, migration, and age structure—alongside the social and economic conditions that shape them. In many contexts, factors such as improvements in health, education, and productivity, or sustained migration inflows, may exert a stronger influence on demographic trajectories than modest variations in fertility. In fact, recent research further shows that, in some countries, changes in mortality can have a greater impact on population dynamics than shifts in fertility (Gietel-Basten et al. 2026).

A more comprehensive perspective would therefore move beyond single-threshold classifications and instead emphasize the diversity of demographic pathways observed across countries (Parr 2025). Such an approach is better suited to understanding contemporary population dynamics and to informing policy responses aligned with specific national contexts (Gietel-Basten & Pardo 2026).

References 

Gietel-Basten, S., & Pardo, I. (2026). Beyond Lowest-Low Fertility: Why Post-Transitional Populations Follow Divergent Paths. Population and Development Review. https://doi.org/10.1111/padr.70052

Gietel-Basten, S., Pothisiri, W., & Scherbov, S. (2026). Investments in health and mortality reduction to address population decline. Bulletin of the World Health Organization, 104(2), 94–102. https://doi.org/10.2471/BLT.25.293627

Kohler, H.-P., Billari, F. C., & Ortega, J. A. (2002). The emergence of lowest-low fertility in Europe during the 1990s. Population and Development Review, 28(4), 641–680. https://doi.org/10.1111/j.1728-4457.2002.00641.x

Parr, N. (2025). Total fertility rates with immediate and very long run zero population growth implications for European countries. Genus Vol 81, article 27. https://doi.org/10.1186/s41118-025-00268-x

Sobotka, T. (2017). Post-transitional fertility: The role of childbearing postponement in fuelling very low fertility levels. Journal of Biosocial Science, 49(S1), S20–S45. https://doi.org/10.1017/S0021932017000323

The post The end of lowest-low as we knew it: one threshold, many futures appeared first on N-IUSSP.

Traditionally, family structures have been authoritarian and age- and gender-stratified, with norms ensuring that power, authority and control rest with men and keep women in a subordinate position, whether in terms of property or of inheritance rights (Harper et al. 2020). Even where laws exist, fragile family support may inhibit women from claiming their legal rights, and compromise their reproductive rights. However, as societies modernize and become more egalitarian, women play an increasingly powerful role in choosing their partners and the type of union they enter, refusing sex or initiating sex, and deciding which contraceptive method to use, if any. 

Context matters: families and societies

Across low- and middle-income countries (LMICs), certain regions broadly conform to traditional hierarchical kinship structures (in particular South Asia and MENA – Middle East and North Africa), others are mixed (sub-Saharan Africa), and yet others are more modern and egalitarian (East and South-East Asia, Latin America). These structures play a strong role in determining the mores around women’s behavior, including in reproduction. 

We thus argue that contextual influences, whether represented by sub-national, national or regional affiliations, are associated with particular kinship structures, marriage systems, inheritance and other laws and norms. These, in turn, shape social norms surrounding female agency and empowerment, and directly affect reproductive choices available to women and girls. While countervailing influences may exist, societies that grant more equal status and agency to women and men are more likely than others to also offer an environment that gives greater reproductive choices to women.

Women’s agency – community and individual influences 

In a recent paper (Jejeebhoy and Sathar 2024), we reviewed 61 studies comprising 80 distinct sub-studies that explored relationships between women’s agency – decision-making, freedom of movement, control over resources, partner power balance and gender role attitudes – and contraceptive behavior (Figure 1). Most of them (67/80) support a positive association between indicators representing at least one domain of women’s agency and contraceptive outcomes after controlling for confounding factors. 

While fewer studies have explored inter-country-level or community-level associations between female agency and contraceptive practice, and just a few of these have explored both of these effects together with individual-level associations, our findings are illuminating. Plotting the association between one key measure of agency, namely decision-making around three major household items, and contraceptive use for 69 LMICs drawn from DHS surveys in countries in every region for which data were available reveals a strong positive correlation (Figure 2). 

Multi-country, country, and sub-national studies that go beyond the individual level and incorporate community-level information have corroborated the importance of community-level over and above individual-level factors in explaining the links between gender norms and roles and contraceptive outcomes (Metheny and Stephenson 2017; Mejía-Guevara et al. 2020). For example, they have highlighted that the decision to adopt a modern contraceptive was strongly influenced by perceptions about community reactions or community decision-making patterns, reflecting prevailing cultural norms surrounding the expected roles of women (Stephenson et al. 2007; DeRose and Ezeh 2010); or that, irrespective of personal beliefs, living in neighbourhoods with equitable attitudes or perceptions about community approval of family planning were positively associated with contraception (Okigbo et al. 2018; Dynes et al. 2012). Evidence, although not abundant, is sufficient to support the proposition that both individual and community level factors can influence women’s agency and contraceptive behaviour. 

Transformative factors 

Individual countries and societies are in flux, transitioning from more traditional to more modern patterns, and this affects women’s agency and gender roles. Transformative factors include changes in female educational attainment and opportunities for paid work outside the home, along with greater mass and social media exposure, and membership in women’s movements. While the links between education and contraceptive use are positive in most studies, contraceptive programs have an even greater effect on the uptake of contraception (Bongaarts and Hardee, 2019). Pathways of influence can also be indirect – clearly, women with agency will take the lead in adopting contraception, with or without a contraceptive/ family planning program. In settings with weak programs and restricted access to services, women’s agency can play an especially strong role in enhancing contraception (e.g. Brazil, Cavenaghi and Alves 2019). Conversely, a strong program endorsed at the state level can act as an equalizer and empower even the most disadvantaged women by expanding the reach of information and services to them (e.g. Iran, Abassi et al., 2009). 

Conclusions

Even after adjusting for other factors, women’s agency has a strong influence on contraceptive outcomes. Contraceptive use is likely influenced by both individual- and community-level factors. Transformational factors, especially increased exposure to education and family planning programs have independent effects on contraceptive outcomes, at times even weakening or cancelling out the effects of women’s agency. Three key conclusions emerge. First, context matters; kinship structures and social norms influence both female agency and sexual relationships, a major basis of the reproductive choices available to women and girls. Second, as associations are complex and can be multidirectional, there is a need to unpack the pathways in the relationships between and across women’s agency and patterns of contraceptive use. And finally, contraceptive endpoints must shift from mere levels of contraceptive use to measures that reflect more completely women’s own intent about whether to practice contraception, and the timing and choice of specific contraceptive methods . 

References

Abbasi-Shavazi M.J., McDonald P., Hosseini-Chavoshi M. 2009. The Fertility Transition in Iran: Revolution and Reproduction. Dordrecht: Springer Netherlands. https: //doi.org/10.1007/978-90-481-3198-3.

Bongaarts J., Hardee K. 2019. “Trends in Contraceptive Prevalence in Sub-Saharan Africa: The Roles of Family Planning Programs and Education.” African Journal of Reproductive Health 23 (3): 96–105. https://doi.org/10.29063/ajrh2019/v23i3.9.

Cavenaghi S., Alves J.E.D. 2019. “The Everlasting Outmoded Contraceptive Method Mix in Brazil and Its Legacy.” Revista Brasileira de Estudos de População 36 (December): 1–29. https://doi.org/10.20947/s0103-3098a0103.

DeRose L.F., Ezeh A.C. 2010. “Decision-Making Patterns and Contraceptive Use: Evidence from Uganda.” Population Research and Policy Review 29 (3): 423–39. https://doi.org/10.1007/s11113-009-9151-8.

Dynes M., Stephenson R., Rubardt M., Bartel D. 2012. “The Influence of Perceptions of Community Norms on Current Contraceptive Use among Men and Women in Ethiopia and Kenya.” Health & Place 18 (4): 766–73. https://doi.org/10.1016/j.healthplace.2012.04.006.

Harper C., Watson C., Kyomuhendo Bantebya G., Ghimire A., George R. 2020. “Historical Lessons on Gender-Norm Change, with Case Studies from Uganda and Nepal.” ALIGN.

Jejeebhoy S.J., Sathar Z. 2024. Revisiting Women’s Empowerment and Contraception. Population and Development Review supplement “Contraceptive Transitions: Explanations and Evidence” https://doi.org/10.1111/padr.12688

Mejía-Guevara I., Cislaghi B., Weber A., Hallgren E., Meausoone V., Cullen M.R., Darmstadt G.L. 2020. “Association of Collective Attitudes and Contraceptive Practice in Nine Sub-Saharan African Countries.” Journal of Global Health 10 (1): 10705. https://doi.org/10.7189/jogh.10.010705.

Metheny N., Stephenson R. 2017. “How the Community Shapes Unmet Need for Modern Contraception: An Analysis of 44 Demographic and Health Surveys.” Studies in Family Planning 48 (3): 235–51. https://doi.org/10.1111/sifp.12028.

Okigbo C.C., Speizer I.S., Domino M.E., Curtis S.L., Halpern C.T., Fotso J.C. 2018. “Gender Norms and Modern Contraceptive Use in Urban Nigeria: A Multilevel Longitudinal Study.” BMC Women’s Health 18 (1): 178. https://doi.org/10.1186/s12905-018-0664-3.

Stephenson R., Baschieri A., Clements S., Hennink M., Madise N.. 2007. “Contextual Influences on Modern Contraceptive Use in sub-Saharan Africa.” American Journal of Public Health 97 (7): 1233–40. https://doi.org/10.2105/AJPH.2005.071522.

The post Women’s empowerment and contraception in low- and middle-income countries appeared first on N-IUSSP.

Recently, David Lam wrote a N-IUSSP essay “How will two billion more people affect food, water, energy, and CO₂?” (Lam 2025). It is based on recent research and summarizes two of his papers. I am concerned that, although based on our current understanding of the past and contemporary projections for the future, Lam’s predictions are too optimistic.

In this paper Lam looks at four resources: food, water, energy and CO₂. The paper projects human population growth and fertility over the next six decades, using the UN (2024) database. These data predict that the richer parts of the world will have constant or decreasing populations, while population growth will mainly occur in the poorer countries. There is no surprise here.

A startling prediction

Lam’s surprising outcome of these projections is that global energy usage will only increase about 9 percent, and that global CO₂ production will decrease by more than 50 percent. (See Figure 3 in his article, reproduced below). Neither of these figures is affected significantly by population growth. 

In addition to CO₂ and energy, Lam looks at food production and water. As would be expected, demand for both of these resources increases with increasing human population, for food more so than water.

Reasons for concern

I have several concerns about the claims of this paper. The first relates to sustainability. The Ecological Footprint currently estimates that we are using the planet’s resources at a rate that is 1.75 times what would be sustainable. Any further increase in population is likely to increase earth’s overshoot. Secondly, the estimates of future usage of resources are based on current availability, which is likely to change. For instance, in many locations the water table is dropping – in some places, at a very rapid rate. Topsoil is also decreasing, which will also limit food production. Furthermore, what topsoil is left is being poisoned in many places by agricultural chemicals that are reducing fertility – both of crops and, perhaps, of humans.
Many biologists feel that we are entering the sixth mass extinction. They estimate that the human enterprise is causing loss of species at a thousand times the normal rate (Ceballos et al 2015). It is impossible to predict how this will upset the balance of nature and thus affect our species, but many fear that the outcomes of these extinctions will be detrimental and perhaps even disastrous.
The effects of climate chaos can only be guessed. However, it is very likely that the net effect will be disadvantageous to humans. Southwestern North America is already experiencing a megadrought that seems to be at least partially caused by anthropogenic climate heating (Park et al. 2020). Drought reduces food production, and similarly is likely to decrease hydroelectric power generation.
Politicians, who have control over so much of the world’s people and resources, are unpredictable. One person, in a position of high authority, can ruin the plans and predictions of more forward-thinking people. An example is the current US administration that is denying climate chaos and removing support for steps that would improve humans’ approach to sustainability. 
For his modeling, Lam chose IPCC scenario SSP2-4.5, the “middle of the road” projection of fossil fuel emissions. This predicts decreased carbon emissions, especially in higher-income countries. Unfortunately, recent elections in several high-income countries make this scenario less likely.
Finally, UN (2024) population forecasts, upon which Lam has based his predictions, tend to underestimate the rate of growth (O’Sullivan 2023). Even without this confounding error, projections of human population tend to be especially unreliable for periods longer than 5 decades, and Lam’s predictions are for a 65-year time span.

A welcome note of optimism, despite all

Despite all these concerns, I hope that David Lam is correct! Unfortunately (or perhaps mercifully), neither of us will live to find out how valid the predictions are. In any case, the article is a welcome change from the pessimism that I often feel from reading doomful news in the media.

References

Ceballos G, Ehrlich PR, Barnosky AD, García A, Pringle RM, Palmer TM. (2015) Accelerated modern human-induced species losses: Entering the sixth mass extinction. Sci Adv. Jun 19;1(5):e1400253. doi: 10.1126/sciadv.1400253. PMID: 26601195; PMCID: PMC4640606.

Lam, D (2025). How will two billion more people affect food, water, energy and CO₂. NIUSSP January 20, 2025.

Lam, David (2024). The Next 2 Billion: Can the World Support 10 Billion People? Population and Development Review. https://doi.org/10.1111/padr.12685.

O’Sullivan, JN. 2023. Demographic Delusions: World Population Growth is Exceeding Most Projections and Jeopardising Scenarios for Sustainable Futures. World, 4, 545-568. https://doi.org/10.3390/world4030034

Park Williams, A  et al. (2020) Large contribution from anthropogenic warming to an emerging North American megadrought. Science 368, 314-318. DOI:10.1126/science.aaz9600

United Nations (2024). World Population Prospects 2024. United Nations Department of Economic and Social Affairs, Population Division.  

The post The effect of two billion more people appeared first on N-IUSSP.

The United Nations (2024) projects that world population will peak at 10.3 billion in 2084, up 2.1 billion from 2024. Can the world provide food, water, and other resources for 10.3 billion people? How will two billion more people affect climate change? 

The demography of the next 60 years

UN projections imply that 2084 will mark the end of centuries of world population growth. Some countries, including China, are already experiencing population decline, and more will join them in coming decades. Population growth will be concentrated in countries currently classified as low-income (LICs) and lower-middle-income countries (LMICs) by the World Bank, while countries classified as upper-middle-income (UMIC) and high-income countries (HIC) will collectively experience significant population decline. Most population growth will be in Africa, a shift from population growth being concentrated in Asia. While Africa is projected to continue growing throughout the century, Asia’s population is projected to peak in 2054. China is projected to lose 600 million people between 2024 and 2084.

Past population growth and resource challenges 

Looking at the last 60 years, it is encouraging that food production has increased faster than population in all regions, and we have not experienced significant shortages in non-renewable resources (Lam 2023, 2024). The history of water and CO₂ emissions is less encouraging, with declining water levels in many aquifers and climate change threatening to undermine progress in all areas. A key factor is that food and non-renewable resources are highly responsive to market signals, while water and greenhouse gas emissions are sectors in which market signals are less effective. 

Simulating the impact of future population growth

Projections of future consumption depend on the relationship between consumption and income. Figure 1 shows per capita food, water, energy, and CO₂ by income group. For each consumption category, the per capita value for LICs is set to 1.0, with values for other income groups shown relative to that. The income gradient for food is relatively flat, with HICs having per capita food consumption 1.5 times higher than LICs. The income gradient for energy and CO₂ is much steeper. Per capita CO₂ emissions are 35 times higher in HICs than in LICs. 

Figure 2 shows the large differences in population growth across income groups. The combined population of LICs and LMICs is projected to increase 2.9 billion from 2019 to 2084, while that of UMICs and HICs is projected to fall 0.5 billion. (I use 2019 as the baseline for simulations to avoid the disruptions associated with COVID-19). Taking Figures 1 and 2 together, coming decades will see a large shift in population from countries with high consumption and emissions to countries with low consumption and emissions. 

I use two simulations to analyze future consumption and emissions. Simulation 1 asks what would happen to world food, water, energy, and CO₂ in 2084 if each country’s population remains at 2019 levels, but there is income growth and associated changes in consumption and emissions. Simulation 2 asks what would happen if each country followed UN median population projections, with income and consumption following the same path as Simulation 1. Lam (2024) details assumptions for factors such as income growth, declines in energy intensity of income, and declines in carbon intensity of energy, drawing on existing literature.  

Figure 3 summarizes the results, showing the percentage change in world population, food, water, energy, and CO₂ for 2019-2084 for the two simulations. Income growth causes a modest 15% increase in world food consumption and almost no increase in world water consumption in the absence of population growth (Simulation 1). When population growth is added (Simulation 2), food increases by 45% and water by 21%. In contrast, adding population growth has almost no effect on energy and CO₂ emissions relative to the simulation without population growth. Adding population growth slightly increases the decline in global emissions from 2019 to 2084, from -50% to -52%. 

Different assumptions about future declines in emissions per dollar of GDP would affect the magnitude of the decline (or increase) in global emissions. But it would still be the case that projections with and without population change would be almost identical. This is because declining populations in UMICs and HICs, with high per capita emissions, offset the impact of population growth in low-emission LICs and LMICs. 

It should be noted that population growth does increase CO₂ emissions in poorer countries in Simulation 2. Emissions would decline in both LICs and LMICs from 2019 to 2084 in the absence of population growth. Population growth reverses that trend from a 1.1% annual decline in CO₂ to a 0.4% annual increase in LICs. Slower population growth in LICs and LMICs would lead to faster declines in emissions, although the magnitude of the effect is mitigated by the low emission levels in those countries. If there were zero population growth in LICs and LMICs, leaving UMICs and HICs to follow UN median projections, global CO₂ emissions would be 15% lower in 2084 than is projected in Simulation 2.

The surprising bottom line is that the net effect of projected population changes over the next 60 years is an almost neutral impact on CO₂ emissions, despite the addition of over 2 billion people. The emission-increasing effect of population growth in low-emission lower-income countries is offset by the emission-reducing effect of population decline in higher-income countries. 

References

Lam, David (2023). Has the World Survived the Population Bomb? A 10-year Update. Population and Environment 45(10). https://doi.org/10.1007/s11111-023-00422-7. 

Lam, David (2024). The Next 2 Billion: Can the World Support 10 Billion People? Population and Development Review. https://doi.org/10.1111/padr.12685.

United Nations (2024). World Population Prospects 2024. United Nations Department of Economic and Social Affairs, Population Division.  

The post How will two billion more people affect food, water, energy, and CO2? appeared first on N-IUSSP.

 “I do not dare to bring a child into this world” [Op deze wereld durf ik geen kind te zetten] stated the title of a newspaper article in the Netherlands, at the end of 2021. This sentiment is not an isolated occurrence: titles such as “Given the state of the world, is it irresponsible to have kids”, “The rise and rise of the birth strike”, and “Should I have children? Weighing parenthood amid the climate crisis” have been appearing regularly in news outlets over the past decade. The question of why people choose (not) to have children has never been merely academic. Currently, about half of the world’s population lives in countries with sub-replacement fertility levels. The undeniable impact of declining fertility rates on the sustainability of (Western) welfare states has made this issue a critical area for policymaking. Numerous policy analyses have sought to identify effective strategies for encouraging people to have more children (see Sobotka, Matysiak & Brzozowska 2020). Yet, fertility rates continue to fall, even in countries like Finland and Norway, which have made significant, sustained investments in creating conditions that facilitate the combination of work and family life equitably for both partners. Why?

Societal pessimism and fertility

In a recently published study (Ivanova & Balbo 2024), we pointed to one potential culprit: societal pessimism, or the general feeling that things are not moving in the right direction in society at large. Over recent decades, fear and discontent have increasingly become prominent traits of Western societies. Previous efforts to understand the consequences of societal pessimism have concentrated on its potential to affect attitudes toward out-group members such as migrants, as well as political preferences and voting behaviors, particularly in support of populist parties. We argue that societal pessimism can also influence demographic behaviors, such as fertility. People generally expect their children to fare better than, or at least as well as, themselves. Therefore, especially in contexts where voluntary childlessness is becoming more socially acceptable (for example, the Netherlands; Noordhuizen, de Graaf & Sieben 2010), the perceived conditions that the next generation may face – conditions that individuals feel they cannot directly improve – might significantly shape fertility decisions.

Societal pessimism or pessimism about own future?

To determine whether societal pessimism influences fertility behaviors, we first need to understand if people indeed differentiate between their personal futures and the broader future of the next generation. We utilized the Dutch Longitudinal Internet Studies for the Social Sciences (LISS), a panel created in 2010. When joining the panel, all respondents evaluate various aspects of life (e.g., social relationships, financial prospects, and overall well-being) and, with a 7-point scale, they state whether, in their opinion, the future of the next generation will be significantly better (=1) or worse (=7) than today in each of those areas.

To ascertain by how much people distinguish between the future of the next generation and their own, we carried out additional data collection in 2020. At that point, respondents were asked the original questions plus the same questions but with an explicit focus on their own future. We found that our respondents were more optimistic about their own future than about that of others, specifically the next generation (Figure 1). This aligns with empirical research showing that individuals tend to be more optimistic about their own chances of success and happiness than those of generalized others, a phenomenon known as the ‘optimistic bias’, ‘unrealistic optimism’, and the ‘optimism gap’ (Whitman 1998).

Does societal pessimism matter for the transition to parenthood?

For our study, we focused on adults who were not parents at entry into the LISS panel (when they filled out the societal pessimism measure) but were still relatively young (less than 40 and 45 years old, respectively, for women and men). We “followed” these individuals from their entry into the panel (as early as 2010) until they became parents, dropped out, reached the age of 45 for women and 50 for men, or until the end of observation (2022). Figure 2 displays the predicted probability of becoming a parent, across self-reported societal pessimism scores. What we found is that those who scored higher on societal pessimism were less likely to have a child, even accounting for the level of depression and satisfaction with own income (at the time when they entered LISS).

Why does it matter?

The newspaper headlines mentioned at the start of this article suggest that young adults actively consider the general state of the world when deciding whether to have children. However, these narratives are often captured at a single moment in time. This limitation prevents us from confirming whether these sentiments actually influence behavior. Additionally, it provides detractors with an opportunity to dismiss these narratives, arguing that intentions change as people age. Our findings show that in a society where opting not to have children is increasingly seen as acceptable, the transition to parenthood is indeed influenced by the perception of the world those children would live in. 

We do not imply that the imagined future of the next generation is the sole driver of declining fertility rates across the Global North. Undoubtedly, improving the current living conditions of young adults is paramount if countries are interested in reversing this trend. At the same time, it is essential to recognize that even when individuals feel secure about their own future, they might decide to forego parenthood in a world characterized by climate change, increasing inequality, and social polarization.

References

Ivanova, K. & Balbo, N. (2024). Societal pessimism and the transition to parenthood: A future too bleak to have children? Population and Development Review. https://doi.org/10.1111/padr.12620

Noordhuizen, S., de Graaf, P., & Sieben, I. (2010). The public acceptance of voluntary childlessness in the Netherlands: From 20 to 90 per cent in 30 years. Social Indicators Research, 99, 163-181. https://doi.org/10.1007/s11205-010-9574-y

Sobotka, T., Matysiak, A., & Brzozowska, Z. (2020). Policy responses to low fertility: How effective are they? (Working Paper No. 1). UNFPA, Technical Division, Population & Development Branch.

Whitman, D. (1998). The optimism gap: The I’m OK-They’re Not syndrome and the myth of American decline (1st ed.). Walker & Co.

The post Future fears: societal pessimism and fertility choices appeared first on N-IUSSP.

Households are core units of social organization and reproduction. Their fundamental characteristics, namely size and composition, reflect complex socio-economic and demographic changes (Bongaarts, 2001; Therborn, 2004, 2014). For instance, increases in one-person and two-person households at the expense of larger households might be driven by demographic changes, such as fertility decline or increased longevity. Household characteristics can have far-reaching implications for housing markets, intra and inter-household transfers, or gender relations. Studying household size or composition comparatively and over time allows us to understand how societies change, not only from a demographic standpoint but also from social and economic perspectives.

Comparative research on households on a global scale has remained relatively scarce due to data constraints. The newly developed open-access, open-source CORESIDENCE database aids researchers in filling this gap by providing harmonized household data on the national and subnational scales for 156 countries and 1,511 regions (Galeano et al., 2024), making it possible to study families and households on an international scale. Family researchers, like those studying fertility and mortality, can address questions about family change based on the increasing availability of empirical data, thanks to census microdata available from the Integrated Public Use of Microdata Series (IPUMS) and survey data from Demographic Health Surveys (DHS), Multiple Indicator Cluster Surveys (MICS), or the European Labor Force Surveys (EU-LFS). 

Converging sizes: moving towards smaller households

Despite some definitional differences, households are considered to be suitable units for comparative research (Bongaarts, 2001). Household size refers to the number of people who are living together in a dwelling at the time of data collection. It differs substantially across countries as it reflects their diverse fertility levels (Figure 1). However, it is also determined by social norms and customs governing family-formation processes, such as norms for establishing a new, separate household upon marriage. 

Households are, on average, smallest in some East Asian countries and in Europe. Denmark leads this group, with an average household size of 1.8 people. The largest households, on average, can be found in Sub-Saharan Africa and among countries of the Arabian Peninsula and Central Asia, reaching a maximum of 8.4 people per household in Senegal, for example. The vast majority of the world’s population (around 83 percent) lives in countries with an average household size of between 2.3 and 5 persons. 

Household size has declined quite steadily over the past five decades, by approximately 0.5 people per decade. When average household size in a country falls to around two, the rate of decline slows down: countries with comparatively large households and high fertility levels will likely drive any further decline in average household size. Shrinking households in contexts with relatively small households might suggest an increase in one-person households, driven by young adults who postpone their transition to adulthood or by older adults, such as women who live alone after the death of a partner. However, households do not shrink in all contexts: socio-economic conditions may foster intergenerational coresidence or alternative household configurations that could, at least partially, counteract the reduction in household size due to fertility decline. Moreover, some countries (e.g. Senegal) have maintained large households despite demographic changes. 

Divergent structures: the roles of children and other relatives 

The internal structure of a household is called household composition. This structure is based on a reference person in the household, commonly referred to as the household head or reference person in survey and census data. Household composition varies in response to demographic changes, such as a decline in fertility or changes in family and relationship forms, such as living-apart-together or intergenerational coresidence (Furstenberg, 2019). Household composition further reflects the age structure of a population: the number of children and adults available in a society for distribution across households differs according to the underlying age composition of the population. 

Fertility decline is a core driver of shrinking household size. However, across many contexts, the presence of other relatives of the household head is another crucial component of household composition. If the number of children in the household declines but the presence of other relatives remains stable, their relative weight in the household may increase, with possible implications for care transfers, decision-making processes, and family dynamics. 

Figure 2 highlights the decline in children in the household (upper panel) across all major world regions. While children are the central driver of global heterogeneity in household size and composition, another important group contributing to global variation in household size and composition is that of other relatives (lower panel). In many countries, the presence of other relatives in absolute numbers has remained relatively unchanged. 

Conclusions 

Family patterns, socio-economic and demographic transformations, and cultural systems influence the persistent heterogeneity in household composition (Cherlin, 2012). Our results support the “convergence to divergence” hypothesis, which postulates that future family changes driven by social phenomena are certain but will produce different outcomes in different contexts (Pesando & GFC Team, 2019; Therborn, 2004, 2014). Fertility declines increase the prevalence of smaller households, yet we find persistent internal complexity of households in several countries. Interactions of global trends such as declining fertility and increasing life expectancy will result in different typologies of households, depending on the context in which such changes occur. We further anticipate that a continued decline in fertility rates will lead to smaller households and changes in household composition due to fewer living relatives available for coresidence. 

The timing and occurrence of life-course events, societal structures, and individual factors drive variation in household structures. Shifts in transfer patterns, housing demand, gender relations, the division of labour, and living arrangements are some of the possible implications of changes in household structures. The CORESIDENCE project investigates these issues by examining links between households, living arrangements, and related socio-demographic indicators. The project entails macro- and micro-level analyses across various contexts, data sources, and spatial levels. We also invite and encourage other researchers to use the CORESIDENCE database.  

Funding: CORESIDENCE is supported by an Advanced Grant funded by the European Research Council (ERC). Grant agreement ID: 101052787.

References 

Bongaarts, J. (2001). Household size and composition in the developing world in the 1990s. Population Studies, 55(3), 263–279. https://doi.org/10.1080/00324720127697

Cherlin, A. J. (2012). Goode’s World Revolution and Family Patterns : A Reconsideration at Fifty Years. Population and Development Review, 38(4), 577–607. https://doi.org/10.1111/j.1728-4457.2012.00528.x

Esteve, A., Galeano, J., Turu, A., García-Román, J., Becca, F., Fang, H., Pohl, M., & Trias-Prats, R. (2023). The CORESIDENCE Database: National and Subnational Data on Household and Living Arrangements Around the World, 1964-2021. Zenodo. https://doi.org/10.5281/ZENODO.8142652

Esteve, A., Pohl, M., Becca, F., Fang, H., Galeano, J., García-Román, J., Reher, D., Trias-Prats, R., & Turu, A. (2024). A global perspective on household size and composition, 1970–2020. Genus, 80(1), 2. https://doi.org/10.1186/s41118-024-00211-6

Furstenberg, F. F. (2019). Family Change in Global Perspective: How and Why Family Systems Change. Family Relations, 68(3), 326–341. https://doi.org/10.1111/fare.12361

Galeano, J., Esteve, A., Turu, A., García-Roman, J., Becca, F., Fang, H., Pohl, M., & Trias-Prats, R. (2024). CORESIDENCE: National and subnational data on household size and composition around the world, 1964–2021. Scientific Data, 11(1), 145. https://doi.org/10.1038/s41597-024-02964-3

Pesando, L. M., & GFC team. (2019). Global Family Change: Persistent Diversity with Development. Population and Development Review, 45(1), 133–168. https://doi.org/10.1111/padr.12209

Therborn, G. (2004). Between sex and power: family in the world, 1900 – 2000 (Reprinted). Routledge.

Therborn, G. (2014). Family Systems of the World: Are They Converging? In. J. Treas, J. Scott, & M. Richards (Eds.), The Wiley Blackwell Companion to the Sociology of Families (1st ed., pp. 1–19). Wiley. https://doi.org/10.1002/9781118374085.ch1

The post Global household trends: converging sizes, divergent structures appeared first on N-IUSSP.

Professor Stan Becker, who proposed the debate (UN 2023), asked me to provide background for it, probably because some time ago (Cohen 1995, p. 368) I reviewed 66 published estimates of the human carrying capacity of Earth, as defined in various ways by the publications. The results of 65 of them (omitting one estimate of 10¹⁶ – 10¹⁸ people) are summarized in Figure 1, where the horizontal axis is the year of publication and the vertical axis (billions of people), on a logarithmic scale, is the estimated human carrying capacity. 

The earliest estimate was by Antoni van Leeuwenhoek, inventor of the microscope, in 1679. He estimated that the world could support 13.4 billion people. Estimates in the last half of the 20th century ranged from less than 1 billion to more than 1,000 billion.

Not all these estimates can be right, and some may not correspond to “the population of humans that can be supported sustainably on the planet at a reasonable standard of living.” Moreover, they vary increasingly with time instead of converging to a single best estimate. Some of these figures are more political than scientific: the high estimates are intended to show that there is no present problem with continued population growth, while the low estimates are intended to show that Earth already has more people than it can support sustainably. Half of them lie between 4 billion and 16 billion people.

Van den Bergh and Rietveld (2004) reviewed 94 “limits to world population”: “When taking all studies into account, the best point estimate is [the median estimate,] 7.7 billion people; the lower and upper bounds, given current technology, are 0.65 billion and 98 billion people, respectively.” “Given current technology,” is only one of the factors upon which estimates are conditional.

Choices and constraints in populations, economics, environments, and cultures

Multiple factors influence whether Earth can sustainably support 4 billion or any other number of people. We can describe these factors at three levels of detail.

At the most general level, Earth’s human carrying capacity depends on choices and constraints. Humans do not fully understand Earth’s biological, chemical, and physical responses to past and future perturbations from humans and from nature. Hence our understanding of constraints on future human populations is incomplete. Likewise, future human choices may affect which constraints apply. Unanticipated constraints may affect which choices humans will have.

In more detail, constraints and choices affect populations, economics, environments, and cultures. These four factors interact to influence Earth’s capacity to support people sustainably. For example, culture affects economics (and vice versa) in ways which depend partly on demography (e.g., through urbanization and age structure). The interaction between culture, economics, and demography influences and is influenced by environments (e.g., through biodiversity, infectious viruses, climate, and earthquakes).

At a third, finer level of detail, how many people Earth can support sustainably and well depends on answers to at least 11 questions.

1) What will be the average level of material and cultural well-being, in terms of food, fiber, water, housing, industrial output, health, sanitation, energy, education, and travel, for example?

2) What will be the distribution of material and cultural well-being? What inequalities will we choose or accept in income, wealth, “happiness”, health, and other “goods,” among nations and among individuals within nations? How will we measure inequalities?

3) What technologies will we use to produce goods and services?

4) What will be our domestic and international political arrangements? What will be our individual freedoms, procedures for political change, means of resolving conflicts?

5) What will be our domestic and international economic arrangements? What will be our incentives for economic productivity, modes of production, trade agreements or tariffs, and regulation within and among nations?

6) What will be our domestic and international demographic arrangements regarding, for example, birth, death, migration, marriage, family, households, age structure, and cities?

7) What will be our physical, chemical, and biological environments, including future biological diversity, exposure to infectious agents, and climate?

8) What will be our desired variability or stability of population size? For the last 12,000 years, humans have enjoyed a warm interglacial interval since the last ice age. The previous interglacial occurred 115,000-130,000 years ago, so biologically modern humans have experienced the most recent ice age and could survive a future ice age. Let us assume that Earth can support sustainably at most 1 billion people during an ice age. Do we want a total population size that remains at 1 billion people during ice ages and interglacials, or do we want a total population size that surges to 8 billion or more during interglacials and subsides to 1 billion during ice ages? How much variability or stability do we want?

9) Do we prefer risk or robustness? For example, do we continue to build cities in floodplains and active earthquake zones, or do we limit settlements to safe areas?

10) What is our time horizon? For the next five years we will have plenty of oil. “Given current technology,” it seems very unlikely that we will have plenty of oil for the next 500 years.

11) What will be our values, tastes, and fashions? Will we value jaguars with four legs or Jaguars with four wheels? Parks or parking lots? To settle our political, economic, cultural, and personal differences, will we choose violence or talk?

A better question

Earth’s human carrying capacity is so difficult to define precisely, so sensitive to subtle assumptions, and so dependent on an unknowable future that it has questionable scientific content. I believe we need to change our focus from Earth’s human carrying capacity to a more practical question.
What do we need to do and learn now and in the coming 25 years (roughly one generation) to improve the security, health, prosperity, and well-being of people we have now and will add in the next 25 years, while respecting our obligations to future generations of humans and other life forms?
Three panaceas have been proposed: bigger pie, fewer forks, and better manners.

• The advocates of a bigger pie recommend increasing the productivity of people and machines.
• The advocates of fewer forks recommend slowing or ending population growth through voluntary reductions in fertility.
• The advocates of better manners recommend reducing violence, corruption, inequalities, barriers to economic and technological efficiency, and unwanted material byproducts of consumption and production.

Many advocates insist on one of these panaceas to the exclusion of the other two. I believe we need all three.

References

Cohen, Joel E. 1995 How Many People Can the Earth Support? W. W. Norton, New York. [Japanese translation: Shin-Jinkou-Ron (New Principle of Population) by Nanako Shigesada, Hiromi Seno and Fugo Takasu. Tokyo: Nosan Gyosan Bunka Kyokai. Italian translation: Quante persone possono vivere sulla terra? by Nanni Negro. Bologna: Il Mulino.]

Cohen, Joel E. 2017 How many people can the Earth support? Journal of Population and Sustainability 2(1):37-42. [In French: Cohen, Joel E. 2017 La capacité humaine de la Terre. Fécondité: un enjeu pour la planète? Revue Projet (Society of Jesus, France) 359:78-83.]

IUSSP 2023, Debate: “The population of humans that can be supported sustainably on the planet at a reasonable standard of living is below 4 billion.” Yes or No. 

UN-DESA (2022) World Population Prospects 2022 Revision, New York.

Van Den Bergh, Jeroen C. J. M. and Rietveld, Piet 2004 Reconsidering the limits to world population: meta-analysis and meta-prediction. BioScience 54(3):195-204.

The post Can Earth support 4 billion people sustainably and well? appeared first on N-IUSSP.

Martinique is one of the overseas departments and regions of France (DROM), known for its bananas, its rum and its heavenly beaches. Its population of about 350,000 inhabitants exudes joie de vivre, even though it has the highest average age of all French regions. While this is largely due to low fertility and significant emigration of young people to metropolitan France, it may also have to do with their longevity (Vallin 2020, 2021).

How to ascertain the longevity of Martinique

To determine the level of longevity in Martinique, careful data collection was performed during the years 2020-2022 to ascertain how many people, among the birth cohorts of 1898-1920, lived past the age of one hundred years. Of these, some were already dead at the time of data collection (2022), but these are the “easy cases”, since all deaths are registered in France, including Martinique, and, since the end of 2019, they have been available in an open-access database with data going back to 1898 (http://www.data.gouv.fr). 

Unlike deaths, the centenarians who were still alive in 2022 and the number of births by municipality in the years 1898 to 1922 are more difficult to count or to estimate. A precious source is the census, which counts the number of centenarians born in Martinique, regardless of their (changing) place of residence in France (Table 1). 

Probability of reaching 100 years

The extreme longevity index (ELI), the probability of reaching 100 years of age within a birth cohort, is one of the most widely used indicators of longevity. We calculated it for Martinique (dividing the number of centenarians born there, both alive and dead at the time of observation, by the total number of births of that cohort) and we compared it with the ELI for the whole of France (Figure 1).

The survival advantage of people born in Martinique over the French population as a whole is large and appears to increase over time: for instance, in the 1920 birth cohort, there were 24.5 centenarians per thousand births in Martinique versus 16.7 in France as a whole.

Mortality rates at age 90 years and above

Total and age-specific mortality rates for the birth cohorts of 1898 to 1914 (practically extinct by 2022) can be calculated for Martinique and France with the extinct generation method (Vincent 1951). Figure 2 offers an example of what can be derived from it, and confirms the pattern shown in Figure 1: remarkably lower mortality rates among those born in Martinique and an advantage that increases with age. By age 100 years and over, the mortality of Martinicans is about half that of the French population as a whole.

Conclusion

In the search for areas of very high survival, data quality is essential. In this case, as in others (Poulain, Herm and Pes 2013), all the data were accurately verified, without detecting errors, such as omissions or age exaggeration. And they testify to the exceptional survival of the population of Martinique in general, and not just a few privileged individuals. Martinique displays a longevity comparable to that of the Nicoya peninsula (Costa Rica) and of the islands of Okinawa (Japan), Sardinia (Italy), and Ikaria (Greece): the so-called Longevity Blue Zones thus increase from four to five. More information on these zones and the research project that investigates them can be found on the recently launched website longevitybluezone.com. 

References

Poulain, M., Herm, A. & Pes, G.M. (2013). The blue zones : Areas of exceptional longevity around the world. Vienna Yearbook of Population Research, 11

Vallin, J. (2020). Why are supercentenarians so frequently found in French Overseas Departments? The cases of Guadeloupe and Martinique. Genus, 76, Article number: 26.

Vallin J. (2021). Eight times more supercentenarians in the French West Indies than in metropolitan France. Gérontologie et société, 43166.3, September 2021, 143-165.

Vincent P. (1951). La mortalité des vieillards, Population 6:182-204. doi: http://dx.doi.org/10.2307/1524149

The post A new Longevity Blue Zone in Martinique (France) appeared first on N-IUSSP.

The post 8 billion strong – infinite possibilities for people and planet appeared first on N-IUSSP.

The striking effect of the Covid-19 pandemic on world mortality in 2020 has been highlighted by several studies, among which Aburto et al (2021) and ourselves (Mazzuco and Campostrini 2022). 

Here we use the same data as our 2022 study (the short-term mortality fluctuation data series of the Human Mortality Database) to compare mortality levels in 2019 (before the Covid-19 pandemic) and 2021. Moreover, for the same countries, we calculate the weekly life expectancy, as Trias-Llimos et al (2021) did for 2020, to show the periods of 2021 when excess mortality (if any) occurred. Finally, we study the association of the 2019-2021 difference in weekly life expectancy with the properly lagged vaccination rate in the same countries. Data on vaccination rates are taken from Our World in Data, following Schoeley et al (2022).

Several patterns

Table 1 reports the estimates of life expectancy loss in 2020 and 2021, evidencing particularly high losses in several countries in 2020, as already shown by Aburto et al (2021) and Mazzuco and Campostrini (2022). The figures for 2021 show several clusters: some countries (Denmark, Iceland, Norway, New Zealand, South Korea and Taiwan) had no life expectancy decline either in 2020 or in 2021, or a very moderate one (Israel and Finland). Others (Belgium, Canada, England and Wales, France, Italy, Netherlands, Spain, Sweden, Switzerland) fully or partially recovered from the 2020 decline. However, in several countries, the 2020 survival decline was not followed by a recovery in 2021, a year which, in some cases, proved even worse. This is the case of many eastern European countries (especially Bulgaria), Scotland, North Ireland, Greece and, to a lesser extent, Germany. The U.S.A. suffered a severe decline in life expectancy in 2020, and things did not improve in 2021. For Russia, data for 2021 are not yet available.

These figures result from several waves of the epidemic. At the beginning of 2021, many countries were in the tail of the so-called second wave; a third wave, characterized by the Beta variant, took place in the first half of the year, while a fourth wave started towards the end of 2021. These waves did not hit all countries simultaneously, which explains part of the observed variability, alongside differences in climate conditions and mitigation measures (Figure 1).

In some countries (e.g. UK, Czech Republic, Portugal, Slovak Republic), the decline in life expectancy is attributable to excess mortality in the first part of the year, during the second or third wave. In others (e.g. Austria, Bulgaria, Latvia, Slovenia, USA), substantial excess mortality occurred towards the end of the year, when the fourth wave was at its height.

The role of Covid-19 vaccination

Towards the end of 2021, vaccination coverage was probably crucial for many countries. Earlier, it was probably effective only in countries – such as the UK and Israel – that began their vaccination campaign earlier. However, the vast majority of the countries considered here started to reach effective levels of vaccination coverage between June and September 2021, when the diffusion of Covid was rather low. In Figure 2, we report a month-specific scatterplot of variations in life expectancy (weekly estimates) and the four-week lagged shares of fully vaccinated population.

The blue regression line drawn in the figures was calculated excluding values from Bulgaria and New Zealand, which appear to be outliers. New Zealand is a peculiar case: despite a low vaccination rate, its relative isolation from rest of the world helped to contain the number of Covid-related deaths. Conversely, Bulgaria also has low vaccination coverage, but a particularly high life expectancy decline. Globally, the panels of Figure 2 clearly show that the association between life expectancy drop and vaccination rate changed over time, from slightly negative in August to strongly positive in November and December. 

There may be two concurring explanations for this increasing relationship:

1) vaccination coverage needs to be above a certain threshold (40-50%) to start showing its benefit in terms of mortality;

2) its effect on mortality increases (protecting countries where coverage is higher) when the pandemic is stronger (during the “waves”), and decreases during summer, when the virus is less prevalent. 

The case of USA is emblematic. While vaccination started earlier and was rolled out faster than in several other countries, in the second part of 2021 coverage increased slowly, reaching about 50-60% only at the end of the year and ending with a mere 62.7% of the population fully vaccinated. Conversely, Italy, after experiencing strong excess mortality in 2020, started its vaccination campaign quite late, but eventually covered around 80% of the population. We posit that this explains, at least in part, the different performance of the two countries: between 2019 and 2021, the USA lost more than 2 years of life expectancy, and Italy only 0.88 years.

Public health implications

The study of life expectancy makes it possible to analyse and compare the effect of the Covid-19 pandemic on mortality in different countries, net of different age structures. By 2020, the differences were all solely attributable to how far the pandemic had penetrated the different countries, while the data for 2021, especially the second part of the year, show above all the effect of vaccination campaigns. This longer view helps to better assess the effects of the pandemic and of the countermeasures adopted by governments. Vaccines, for instance, prove effective, but only above a certain threshold level, when some two-thirds of the population are covered. And in some cases, they have been so effective as to bring life expectancies back to their pre-pandemic levels.

References

Aburto J.M., Schöley J., Kashnitsky I., Zhang L., Rahal C., Missov T. I. , Mills M. C , Dowd J. B. , Ridhi Kashyap, 2021 Quantifying impacts of the COVID-19 pandemic through life-expectancy losses: a population-level study of 29 countries, International Journal of Epidemiology, 2021;, dyab207, https://doi.org/10.1093/ije/dyab207

Human Mortality Database, University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany) (2021) Available at www.mortality.org, Last accessed on 2022-06-27.

Mathieu, E., Ritchie, H., Ortiz-Ospina, E. et al. (2021) A global database of COVID-19 vaccinations. Nat Hum Behav 

Mazzuco S., Campostrini S. (2022) Life expectancy drop in 2020. Estimates based on Human Mortality Database. PLoS ONE 17(1): e0262846. https://doi.org/10.1371/journal.pone.0262846 

Schöley J., Aburto JM, Kashnitsky I, Kniffka MS, Zhang L, Jaadla H,  Dowd JB, Kashyap R (2022) Bounce backs amid continued losses: Life expectancy changes since COVID-19 medRxiv 2022.02.23.22271380; doi: https://doi.org/10.1101/2022.02.23.22271380 

Trias-Llimós S, Riffe T, Bilal U (2020) Monitoring life expectancy levels during the COVID-19 pandemic: Example of the unequal impact of the first wave on Spanish regions. PLOS ONE 15(11): e0241952. https://doi.org/10.1371/journal.pone.0241952

Source tab 1: The Human Mortality Database.

https://ourworldindata.org/

The post Life expectancy variation 2019–2021 and the role of Covid-19 vaccination appeared first on N-IUSSP.