The demographic transition is the long-term change from a population regime characterized by high death rates and high birth rates to one characterized by low mortality and low fertility. Because mortality usually falls before fertility, the transition commonly produces an interval of rapid population growth, followed by slower growth and a shift toward older age groups. It is a central framework in demography, connecting changes in survival and reproduction with economic development, public health, education, and family organization.
The framework emerged from European and North American historical experience but has since been applied to populations throughout the world. It describes a widespread pattern rather than an invariant sequence: mortality and fertility can change at different speeds, fertility decline can pause or reverse, and migration can substantially alter national outcomes. Its explanatory value therefore depends on distinguishing the demographic pattern itself from the social mechanisms that produce it.
The transition has transformed the scale and composition of the world population. Falling mortality made survival to adulthood increasingly common, while declining fertility reduced average family size and eventually contributed to population ageing. These changes affect schools, employment, housing, intergenerational support, and public finance, but their consequences depend heavily on institutions and on the distribution of resources within societies.
Demographic measures and the stage model
The demographic transition concerns changes in vital rates, not simply changes in population size. Population change over an interval can be expressed through the demographic balancing equation:
where is the initial population, births, deaths, immigration, and emigration. Births minus deaths constitute natural increase. Migration is analytically separate from this component, although it can strongly influence the observed trajectory of a country or region. A country with more deaths than births can still grow through immigration, while substantial emigration can offset natural increase.
The familiar stage model is a simplified representation. In the first stage, high mortality and high fertility produce relatively low long-run growth, interrupted by fluctuations associated with epidemics, harvest failures, and other shocks. In the second, mortality declines while fertility remains comparatively high, widening the gap between births and deaths. In the third, fertility declines and natural increase slows. In the fourth, both mortality and fertility are low. Some accounts add a fifth stage of sustained below-replacement fertility and potential population decline, but this extension is not universally defined or accepted.
Crude birth and death rates measure annual events per 1,000 people. Their interpretation requires attention to age structure. An older population may have a higher crude death rate than a younger population even when people at every comparable age face lower mortality risks. Consequently, a rise in the crude death rate need not indicate worsening health. Age-specific mortality rates and life expectancy provide more informative measures of survival conditions. A life table summarizes the mortality experience that would apply to a hypothetical cohort exposed to a specified set of age-specific death rates.
The total fertility rate is the number of births a hypothetical woman would have if she experienced a given period’s age-specific fertility rates throughout her reproductive life. It is not necessarily the eventual family size of actual women. Replacement-level fertility is the level that allows a generation of women to be replaced by their daughters, in the absence of migration. In low-mortality populations it is approximately 2.1 births per woman, but it is higher where more girls die before or during reproductive ages.
These distinctions prevent several common errors. Low birth rates do not necessarily mean that individual women have very few children, because births also depend on the number and ages of women. Likewise, replacement-level fertility does not guarantee immediate population stability. The age distribution inherited from earlier decades can sustain growth or decline long after fertility changes.
Origins and development of the theory
Demographic transition theory developed through attempts to explain the divergence of birth and death rates during modern population growth. Earlier writers, most notably Thomas Robert Malthus in his 1798 Essay on the Principle of Population, emphasized the relationship between population pressure, subsistence, mortality, and restraints on reproduction. The transition framework shifted attention toward a historical transformation in which mortality could fall substantially and fertility could subsequently adjust, rather than population growth being checked principally by renewed scarcity.
In 1929, the American demographer Warren S. Thompson classified countries according to their patterns of fertility, mortality, and population growth. His categories distinguished populations in which both vital rates were declining from those in which mortality was falling while fertility remained high. In 1934, the French demographer Adolphe Landry developed a related argument in La Révolution démographique. Landry emphasized changes in reproductive behavior and the emergence of deliberate limitation of family size. These contributions preceded the widespread adoption of “demographic transition” as the standard term.
Frank W. Notestein’s 1945 essay “Population—The Long View” provided an especially influential formulation. He connected mortality decline to improvements in sanitation, medical knowledge, food supply, and public order, while associating fertility decline with the changing economic and social conditions of urban-industrial life. Smaller families were understood as an adaptation to the rising costs of children, expanded individual aspirations, and weakening pressures to maintain high fertility. This formulation made demographic transition part of a broader interpretation of modernization.
Subsequent research challenged the idea that a particular level of industrialization automatically triggered fertility decline. The Princeton European Fertility Project, initiated in the 1960s under Ansley J. Coale, assembled extensive historical data on European fertility. Its findings highlighted substantial variation in the economic conditions under which fertility fell and the importance of linguistic, religious, and cultural boundaries. Fertility decline often spread among neighboring or culturally connected populations, suggesting that communication and social acceptability mattered alongside material incentives.
By the late twentieth century, the transition had become both a descriptive generalization and a collection of competing explanations. Economic approaches examined the costs and benefits of children; sociological approaches emphasized institutions and changing family relations; diffusion approaches studied the spread of ideas and practices. Researchers increasingly distinguished the first adoption of deliberate birth limitation from later changes in desired family size. The historical framework remained useful, but its original association with a single European development path became less defensible as transitions occurred in societies with very different political systems, income levels, and family institutions.
Mortality decline and the transformation of survival
Mortality decline usually initiates the period of accelerated growth associated with demographic transition. In many historical populations, infectious diseases, nutritional deficiencies, and high infant and child mortality kept average life expectancy low. Low life expectancy at birth did not mean that adults universally died young: it partly reflected the large number of deaths in infancy and childhood. When those early deaths became less frequent, average life expectancy could rise sharply even before survival at advanced ages improved greatly.
The causes of mortality decline varied across periods and places. In parts of northwestern Europe, eighteenth- and nineteenth-century changes included improvements in food availability, transport, housing, and the management of infectious disease. Smallpox vaccination, following Edward Jenner’s work in 1796, provided an increasingly important means of preventing a major cause of death. During the nineteenth century, clean-water systems, sewerage, food regulation, and urban sanitation became central components of public health. Their effects depended on implementation, access, and the local distribution of disease.
Twentieth-century mortality reductions increasingly involved vaccines, antibiotics, vector control, maternal care, and effective treatment of childhood illness. After the Second World War, medical and public-health technologies could be introduced into countries that had not undergone the long process of industrial development experienced by early European transitions. Death rates consequently fell much faster in many later-transition populations. Oral rehydration treatment, for example, substantially improved survival from diarrhoeal disease when it became widely available, illustrating how a relatively simple intervention could reduce mortality without requiring high national income.
The associated change in the composition of illness and death is often discussed through the epidemiological transition, formulated by Abdel Omran in 1971. As infectious causes of early death become less dominant and more people survive to older ages, cardiovascular disease, cancers, and other chronic conditions account for a larger share of deaths. This does not mean that infections disappear. Many countries experience a double burden of infectious disease and noncommunicable conditions, while epidemics can temporarily interrupt long-term improvements.
Mortality decline also interacts with fertility decisions. When children are more likely to survive, parents may need fewer births to achieve a desired number of surviving children. The effect is not necessarily immediate: families may take time to recognize improved survival, and reproductive preferences are influenced by more than mortality. Conversely, severe mortality shocks do not automatically return a population to an earlier demographic stage. The HIV/AIDS epidemic and the COVID-19 pandemic disrupted survival trends in many places, but their demographic effects depended on age patterns, treatment, institutions, and pre-existing fertility behavior.
Fertility decline and changing reproductive behavior
Fertility decline is the component of demographic transition that most directly involves changes in behavior, although reproductive outcomes are also shaped by health and biological constraints. Fertility can fall through later entry into marriage or partnership, reduced exposure to pregnancy, contraception, longer intervals between births, and the deliberate cessation of childbearing after a target family size. The relative importance of these pathways differs across societies. A reduction in marital fertility, for example, is not equivalent to a decline produced mainly by later marriage.
John Bongaarts’s influential framework, developed in the late 1970s, distinguishes underlying social conditions from the proximate determinants through which they affect fertility. Education, income, and social norms do not directly enter the biological process of reproduction. Their effects operate through factors such as sexual exposure, contraceptive use and effectiveness, induced abortion, and postpartum infecundability, including that associated with breastfeeding. This distinction helps explain why similar social changes can produce different fertility outcomes where reproductive practices and access to services differ.
Economic explanations emphasize changes in the costs and expected benefits of children. In agrarian settings, children may contribute labor and provide support to parents later in life, although their contribution varies considerably by age and household circumstances. With compulsory schooling and restrictions on child labor, children generally spend longer as economic dependants. Urban housing costs and the resources devoted to each child can also increase. Gary Becker’s work formalized the relationship between desired numbers of children and investment per child, while John Caldwell emphasized changes in intergenerational wealth flows.
Education, especially girls’ schooling, is strongly associated with fertility change through multiple pathways. School attendance can delay marriage and first births, improve access to information, and expand employment opportunities. Education may also increase parents’ expectations for their children’s schooling. These relationships are not reducible to a single causal mechanism: educational expansion often occurs alongside better health services, changing labor demand, and shifts in family norms. Gender relations matter because reproductive decisions can depend on bargaining power within couples and on whether women can exercise preferences concerning marriage and childbearing.
Access to family planning can reduce the gap between reproductive intentions and outcomes, but contraception alone does not explain why desired family size changes. The spread of smaller-family ideals through social networks, media, and health services can make birth limitation more acceptable. Conversely, persistent preferences for larger families, unequal decision-making, child mortality, or unreliable services can slow fertility decline. These mechanisms distinguish the motivation to have fewer children from the practical ability to achieve that outcome, a distinction central to explaining variation within countries as well as between them.
Regional pathways and historical diversity
The earliest sustained transitions unfolded in parts of Europe over the eighteenth, nineteenth, and early twentieth centuries, but there was no uniform European sequence. England and Wales experienced substantial mortality decline before widespread limitation of marital fertility. France was an important exception to a simple mortality-first narrative: fertility began declining relatively early, before the degree of industrialization reached in some neighboring countries. Across Europe, marriage systems, landholding, religious affiliation, and cultural connections shaped the timing and form of change.
In North America, population growth was additionally influenced by large-scale immigration and settlement. Fertility decline in the United States began well before the late nineteenth-century development of modern contraceptive technologies, showing that reproductive limitation did not depend exclusively on modern methods. Changes in the timing of marriage, family aspirations, and the perceived costs of raising children all contributed. The post-Second World War baby boom later demonstrated that a long-run transition toward lower fertility could include substantial temporary reversals rather than a smooth downward curve.
Many Asian transitions occurred more rapidly than the early European cases. Japan experienced a sharp postwar fertility decline and subsequently became an early example of sustained low fertility combined with population ageing. South Korea moved from high fertility in the mid-twentieth century to below-replacement fertility within a few decades. China’s fertility fell substantially during the 1970s, before the introduction of the one-child policy around 1980; later administrative restrictions shaped reproductive behavior alongside urbanization, education, economic change, and family preferences. These cases illustrate why observed fertility outcomes cannot be attributed to a single policy or development indicator.
Latin America experienced substantial mortality improvements during the twentieth century, followed by major fertility declines, especially from the 1960s onward. The region’s experience included considerable inequality between urban and rural populations and between socioeconomic groups. Fertility decline also occurred in many Muslim-majority countries, undermining claims that a particular religious tradition necessarily prevents transition. Iran’s rapid decline from high fertility in the 1980s toward near-replacement levels around the turn of the century is a prominent example of change involving health services, education, family planning, and altered reproductive preferences.
Sub-Saharan Africa has generally entered fertility decline later, with substantial variation in timing and pace. Some countries have experienced pronounced reductions, while others have had slower declines or periods of relative stagnation. Mortality improvement has frequently generated rapid population growth because fertility remains comparatively high and populations are young. Treating the region as a single demographic unit obscures differences in child survival, schooling, marriage practices, desired family size, and access to reproductive health services. National averages likewise conceal large differences between cities, rural districts, and social groups.
Population momentum, age structure, and demographic dividends
Demographic transition changes the distribution of people across ages, not merely the number of births and deaths. When mortality falls while fertility remains high, more children survive and the population often becomes more youthful. Once fertility declines, the proportion of children gradually decreases. Large cohorts born during the preceding high-fertility period then move into working ages. Later, as those cohorts grow older and smaller cohorts follow them, the share of older people rises. These changes unfold over decades because age structure preserves the effects of earlier demographic conditions.
Population momentum explains why population growth can continue after fertility reaches replacement level. A youthful population contains many people approaching or entering reproductive ages. Even if each woman has fewer children than women in the preceding generation, the large number of potential parents can sustain a high total number of births. The reverse can occur in an old population: a small generation of potential parents may produce fewer births than the number of deaths even after fertility rises. Replacement fertility is therefore a condition concerning generational reproduction, not an immediate guarantee of zero growth.
A common indicator of age structure is the dependency ratio, often calculated as the population younger than 15 and aged 65 or older divided by the population aged 15–64. This measure is convenient but should not be interpreted literally as the ratio of nonworkers to workers. Many people of working age are outside paid employment, some older people continue working, and children may contribute labor in particular settings. Economic support ratios that incorporate actual employment, consumption, and production provide a different and often more informative picture.
The demographic dividend refers to potential economic advantages arising when the working-age share increases relative to dependent age groups. With fewer children per adult, families and governments may have more resources for education, health, and investment. A larger labor force can increase production if employment expands. This opportunity is temporary and conditional: unemployment, poor health, weak schooling, or barriers to women’s employment can prevent a favorable age structure from producing higher incomes.
A related distinction separates a first dividend, associated primarily with changing age composition and labor supply, from a possible second dividend associated with saving and capital accumulation. Longer expected lives and reduced fertility can encourage accumulation of assets for later life, but the result depends on pension systems, financial institutions, and household resources. Eventually, population ageing increases the relative importance of retirement financing, long-term care, and productivity. The same transition can therefore create an initial expansion of potential labor supply and a later period in which growth depends more heavily on output per worker.
Economic development, institutions, and social consequences
Demographic transition is closely associated with economic development, but the direction of causation runs both ways. Rising incomes can improve nutrition, housing, sanitation, and access to care, reducing mortality. At the same time, better health and longer lives can increase educational investment and labor productivity. Fertility decline can change household saving and expenditure, while improved economic security can alter desired family size. These relationships operate through institutions and individual circumstances rather than through a fixed income threshold at which every society changes demographic regime.
Urbanization is one important context for these interactions. Urban employment generally separates the workplace from the household more sharply than many agricultural activities do, while housing and childcare costs can make large families more expensive. Cities also facilitate access to schools, health facilities, information, and wider social networks. Yet urban residence does not automatically produce low fertility or low mortality. Overcrowding, insecure livelihoods, and inadequate sanitation historically made some cities unusually unhealthy, and large inequalities in access to services persist within contemporary urban populations.
Changes in family size alter the organization of the household and the allocation of resources between generations. Fewer births may allow greater expenditure on each child, but demographic change does not ensure that resources are equally distributed among siblings or between boys and girls. Longer lives also increase the period during which several generations coexist. Families may have fewer young dependants while simultaneously facing prolonged responsibilities for ageing relatives. Where formal care services are limited, these responsibilities can affect employment, migration, and the division of unpaid work.
The implications for the labor market depend on timing. Large youth cohorts create demand for schooling and later for jobs; a decline in births does not immediately reduce the number of labor-market entrants. Expanding women’s paid employment can increase effective labor supply even as population growth slows. International migration can redistribute workers between places with different demographic trajectories, but migrants also age, and migration does not permanently eliminate the arithmetic of ageing. Its effects differ between sending and receiving societies and between national totals and particular occupations or localities.
Demographic change also reshapes public spending. A youthful population generates demand for maternal care, vaccination, and schools; an older population generally requires a different combination of healthcare, pensions, and care services. The fiscal outcome depends on tax systems, employment, productivity, benefit rules, and the health of older adults. Environmental effects are similarly mediated: population size influences demand for land, water, and energy, but consumption per person and production technology are also decisive. Demographic transition alone cannot explain changes in resource use or greenhouse-gas emissions.
Below-replacement fertility and the second demographic transition
Sustained below-replacement fertility raises questions that the original transition model did not fully address. Its early formulations often implied an eventual equilibrium of low mortality and fertility sufficient for population replacement. In practice, many countries have remained below replacement for decades. Population decline can follow once age structure ceases to sustain natural increase, although migration may delay or offset it. The term “fifth stage” is sometimes used for this development, but it does not identify a single agreed mechanism or a universal final destination.
Very low period fertility can reflect both fewer births over a lifetime and postponement of births. When successive cohorts delay childbearing, the annual total fertility rate can fall even if their eventual completed family size changes less dramatically. This is known as a tempo effect. Later recuperation of postponed births may raise period fertility, but recuperation is not necessarily complete. Biological limits, difficulty forming stable partnerships, economic circumstances, and changing preferences can all affect whether delayed births eventually occur.
The second demographic transition, associated particularly with Dirk van de Kaa and Ron Lesthaeghe from the 1980s, is a distinct interpretive framework. It connects persistent low fertility with later marriage, increased cohabitation, greater union instability, and a wider separation between marriage and childbearing. Its explanations emphasize individual autonomy, secularization, and changes in values concerning partnership and family life. It is not simply another name for the later stages of the original mortality–fertility transition.
The framework has generated debate because its components do not move together everywhere. Some societies have very low fertility while marriage remains closely connected to childbearing and cohabitation is comparatively uncommon. Others combine diverse family forms with higher fertility than countries where births occur overwhelmingly within marriage. Economic insecurity, housing conditions, long working hours, and the compatibility of employment with caregiving may be as important as value change. These differences make it difficult to interpret low fertility as the product of a single cultural transformation.
Governments have used parental leave, childcare provision, cash transfers, housing measures, and other policies in response to low fertility. Observed effects differ by policy design, duration, and social context; changes in the timing of births can also be mistaken for lasting increases in completed fertility. Reproductive policy additionally intersects with human rights, including freedom from coercion and the ability to make decisions about family formation. Contemporary analysis therefore separates demographic objectives from individuals’ reproductive intentions and from the institutional conditions affecting whether those intentions are realized.
Evidence, forecasting, and limits of the framework
Evidence for demographic transition comes from several kinds of records, each with particular strengths and weaknesses. Civil registration systems record births and deaths continuously and can support detailed analysis where coverage is complete. The population census provides age structure and population counts, but it is periodic and may omit or misclassify people. Household surveys supply information on birth histories, child mortality, contraception, and reproductive preferences where registration is incomplete. Historical researchers also use parish registers, genealogical records, and family reconstitution to reconstruct earlier demographic behavior.
Measurement problems can substantially affect interpretations. Under-registration of infant deaths may make mortality appear lower than it was, while incomplete reporting of births can distort fertility estimates. Age misstatement can affect both fertility and mortality calculations, especially at advanced ages. Retrospective birth histories are subject to omission and errors in timing. Analysts therefore compare sources, assess internal consistency, and sometimes use indirect estimation methods. Apparent sudden changes may reflect revisions in measurement rather than genuine demographic turning points.
Population projections generally use the cohort-component method. An initial population is divided by age and sex, then advanced through time using assumptions about mortality, fertility, and migration. The United Nations World Population Prospects 2024 estimated the global population at about 8.2 billion in 2024 and projected a peak of roughly 10.3 billion in the mid-2080s. These are conditional demographic projections, not predetermined outcomes. Differences in future fertility, particularly across large populations, can produce substantial differences over long horizons, while migration is especially difficult to anticipate at national scales.
The transition model’s chief limitation is that a recurring sequence is not itself a complete causal explanation. Mortality and fertility are influenced by diverse institutions, technologies, beliefs, and power relations. Countries can reach similar demographic outcomes through different combinations of mechanisms. Aggregate stages also conceal inequalities: wealthy and poor households, urban and rural populations, and different ethnic or religious communities may experience transitions at different times. Assigning an entire country to one stage can obscure these overlapping trajectories.
Researchers also distinguish the transition from the claim that all societies must converge on identical family systems or fertility levels. War, epidemic disease, political disruption, economic crises, and policy changes can interrupt trends without reversing the broader historical transformation. Long-term forecasting consequently combines demographic regularities with uncertainty about behavioral change. Age structure gives relatively strong information about near-term numbers entering school, work, or retirement, whereas births several decades ahead depend increasingly on decisions and circumstances that have not yet occurred.