How Population Models Shape Demographic Studies and Policies

Demographers don’t just count people, they build models that explain how populations behave over time. These models simplify the messy reality of births, deaths, and migration into tractable mathematical frameworks that planners and policymakers can actually use. Two of the most influential frameworks in this tradition, the stationary population model and the stable population model, continue to shape how governments forecast school enrolments, plan hospitals, and allocate budgets. Understanding them is essential for anyone trying to make sense of demographic data.

What is a population model? 🔗

A population model is a mathematical representation of how a group of people changes in size and structure over time. It takes inputs like fertility rates, mortality rates, and sometimes migration, and produces outputs such as age distribution, growth rate, and projected size. Because real populations rarely behave neatly, demographers often start with simplified “ideal” models and then adjust them to fit observed data.

The two foundational models in this field, stationary and stable, were developed in the early twentieth century but remain in active use. They power the population projections published by the United Nations, the Registrar General of India, and bodies like the Population Reference Bureau, whose tools depend directly on Lotka’s mathematical foundations.

The stationary population model 🔗

A stationary population is one where the total size stays constant over time. The birth rate equals the death rate, there is no net migration, and the age structure remains unchanged from year to year. In other words, for every person who dies, exactly one person is born, and the proportion of children, working-age adults, and elderly stays fixed.

This is a theoretical idealisation. No real country maintains a perfectly stationary population, but the model is extraordinarily useful as a benchmark. It is mathematically equivalent to a life table population, which makes it a natural reference point for comparing actual populations against a zero-growth standard.

Key assumptions 🔗

The stationary model rests on a few strict conditions. The population is closed, meaning no immigration or emigration. Age-specific birth and death rates are constant over time. The crude birth rate equals the crude death rate, so the growth rate is exactly zero. Because these rates do not vary, the proportion of people in each age group stays the same year after year.

Researchers note that a stationary population provides a stable framework for analysing demographic data without the noise introduced by migration or changing fertility, which is why it has been used as the simplest case of stable population theory since the mid-twentieth century.

Applications in demographic studies 🔗

The stationary model is the workhorse behind life tables, which estimate life expectancy at each age. Insurance companies use these tables to price life insurance and annuities. Public health departments use them to compute indicators like under-five mortality and adult survival probability. Historians of demography use stationary models to reconstruct pre-modern populations where records are incomplete, since the assumption of stationarity allows them to infer missing age structures from partial data.

The model also serves as a baseline for comparing countries that are at or near replacement-level fertility. Nations like Japan and Italy, where birth rates have fallen close to death rates, behave in ways that can be meaningfully compared against the stationary ideal, as discussed in recent work on stationary and stable age-structured models.

The stable population model 🔗

The stable population model is a broader concept introduced by the mathematical demographer Alfred J. Lotka in the early twentieth century. A stable population has age-specific fertility and mortality rates that remain constant over time, but unlike a stationary population, it can grow or shrink at a constant rate. Once these rates stay fixed long enough, the population settles into a characteristic age structure that no longer changes, even though the total size keeps moving.

The stationary population is therefore a special case of the stable population, the one where the growth rate happens to be zero. Every stationary population is stable, but not every stable population is stationary.

Lotka’s contribution 🔗

Lotka, working initially as a chemist, began studying the relationship between birth rates, death rates, and population growth in 1907. He showed that a population with unchanging fertility and mortality patterns grows at a constant rate and acquires a fixed age structure. His implicit equation, now called Lotka’s equation, links these quantities mathematically and remains a central tool in formal demography.

Nathan Keyfitz, one of Lotka’s most prominent successors, described his work on stable populations as the greatest single contribution to population theory. Lotka also helped found the Population Association of America in 1931, anchoring stable population theory at the centre of the discipline.

Why the stable model matters 🔗

The stable population model is the foundation of indirect demographic estimation, a set of techniques that infer fertility and mortality levels from incomplete data. This has been crucial for developing countries where birth and death registration has historically been patchy. By assuming a population is approximately stable, demographers can work backwards from observed age structures to estimate the underlying vital rates.

Stable population theory also underpins the concept of population momentum, the tendency of a population with a young age structure to keep growing even after fertility falls to replacement level. India’s continued growth despite falling fertility is largely a story of momentum, and Lotka’s framework is what makes that growth predictable.

Assumptions and limitations 🔗

Like the stationary model, the stable model assumes a closed population with constant age-specific fertility and mortality. These assumptions are clearly unrealistic, real populations experience migration, epidemics, wars, and shifting fertility norms. But the model is valuable precisely because it provides a clean reference against which real-world deviations can be measured.

From theory to practice: the cohort-component method 🔗

Modern population projections do not assume populations are stable or stationary. Instead, they use the cohort-component method, which projects fertility, mortality, and migration separately for each age group and sex. But the intellectual scaffolding for this method comes directly from stable population theory.

The Registrar General of India and the UN Population Division both rely on the cohort-component approach. As an analysis of projection track records explains, the base population from a census is adjusted for age-wise mortality and migration rates to estimate surviving populations for future years. The UN’s short-run projections for India have missed actual census counts by one percentage point or less over the past three censuses, a level of accuracy that would be impossible without the theoretical foundations laid by Lotka and his successors.

Practical uses of population models in India 🔗

Population models are not just academic exercises. They drive decisions worth lakhs of crores of rupees every year.

Population estimation between censuses 🔗

India conducts a census roughly every ten years, but planners cannot wait a decade between updates. The Population Projections for India and States, 2011-2036 report published by the National Commission on Population in July 2020 provides scientific estimates of population size, age structure, and sex composition for every state through 2036. The report projects that the total fertility rate will decline from 2.37 in 2011-15 to 1.73 in 2031-35, and that urban population will rise from 31.8 per cent in 2011 to 38.2 per cent in 2036.

These projections rely on the cohort-component method built on stable population theory. They are used by ministries, the Finance Commission, and state planning departments to allocate resources.

Healthcare planning 🔗

Healthcare systems must anticipate not just the number of people they will serve but the diseases those people will face. A young population needs maternal and child health services; an ageing population needs geriatric care, cardiovascular treatment, and pension support. UNFPA India works with NITI Aayog and the Ministry of Health and Family Welfare on the demographic dividend and ageing, noting that the population aged 60 and above is projected to make up 20 per cent of India’s population by 2050.

Without reliable population projections, the National Health Mission could not plan vaccine supplies, hospital beds, or doctor recruitment. The Economic Survey 2018-19 chapter on India’s demography at 2040 uses projected age structures up to 2041 to estimate future demand for hospitals and schools, recommending that several states consolidate schools rather than build new ones as the school-age population shrinks.

Education planning 🔗

Gross Enrolment Ratio and Net Enrolment Ratio calculations depend on accurate estimates of the school-age population. When the denominator, the projected number of children, is wrong, every education indicator becomes misleading. Recent analysis on Census 2027 and education planning shows that with the next census expected to deliver district-level age-specific data only by 2028-29, education planners have been working with projections that grow more stale every year. This illustrates a deeper truth: population models are only as good as the data and assumptions feeding them.

Policy development and resource allocation 🔗

Population projections feed into formulae used by the Finance Commission to share tax revenues between the central and state governments. They determine the number of Lok Sabha constituencies each state receives. They guide allocations under centrally sponsored schemes for nutrition, sanitation, and rural employment.

Internationally, the United Nations uses stable population theory to project growth and structure for every country, projections that inform global decisions on aid, climate adaptation, and development finance.

Limitations to keep in mind 🔗

Population models simplify reality, and simplifications can mislead. The stationary and stable models assume closed populations, but India experiences significant internal migration that reshapes the demography of states like Maharashtra, Delhi, and Kerala. Both models assume constant vital rates, but fertility in India has dropped sharply over three decades while life expectancy has risen. And both ignore shocks, the COVID-19 pandemic, for instance, disrupted mortality patterns in ways no stable model could anticipate.

The lesson is not that the models are useless, but that they must be applied with judgement. A skilled demographer treats them as a starting point, not a final answer.

What do you think? If population projections directly shape how money is spent on schools and hospitals in your district, how often do you think the underlying models should be revisited, and what would you change about the way they are communicated to the public?

References

1. https://www.prb.org/news/alfred-lotka-mathematical-demographer/
2. https://www.sciencedirect.com/science/chapter/handbook/abs/pii/S0169716118300294
3. https://pmc.ncbi.nlm.nih.gov/articles/PMC10685904/
4. https://link.springer.com/chapter/10.1007/978-0-85729-115-8_10
5. https://www.dataforindia.com/measuring-pop-projections/
6. https://ruralindiaonline.org/en/library/resource/population-projections-for-india-and-states-2011-2036/
7. https://india.unfpa.org/en/topics/population-dynamics-and-research
8. https://www.indiabudget.gov.in/budget2019-20/economicsurvey/doc/vol1chapter/echap07_vol1.pdf
9. https://educationforallinindia.com/wp-content/uploads/2026/03/Census-2027-What-a-16-Year-Population-Data-Gap-Means-for-School-Education.pdf